From 4e65d104129e8be001d561410f6402135f6f8a45 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 13 Jun 2025 20:02:22 -0700
Subject: [PATCH 0001/1455] drm/i915/guc: Enable CT_DEAD output in regular
 debug builds

There is a sporadic failure to enable CTs ocurring in CI on one
specific machine that can't be reproduced locally. The driver already
supports dumping out a whole bunch of GuC related debug info on such a
failure but only when the verbose GuC debug config option is enabled.
It would be preferable to not enable all the verbose debug output. So
just bump the CT_DEAD code to regular I915 debug level rather than GUC
debug level, at least temporarily for CI.

To prevent excessive spam in other parts of CI, also add a check
against doing a CT_DEAD dump during an error injection test. No point
in dumping large amounts of 'why did this fail' info when the fail is
deliberately induced.

v2: Revert accidentally enabling some other verbose debug output.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://lore.kernel.org/r/20250614030222.105601-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c | 11 +++++++----
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h |  2 +-
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 0d5197c0824a..380a11c92d63 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -13,7 +13,7 @@
 #include "intel_guc_ct.h"
 #include "intel_guc_print.h"
 
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG)
 enum {
 	CT_DEAD_ALIVE = 0,
 	CT_DEAD_SETUP,
@@ -144,7 +144,7 @@ void intel_guc_ct_init_early(struct intel_guc_ct *ct)
 	spin_lock_init(&ct->requests.lock);
 	INIT_LIST_HEAD(&ct->requests.pending);
 	INIT_LIST_HEAD(&ct->requests.incoming);
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG)
 	INIT_WORK(&ct->dead_ct_worker, ct_dead_ct_worker_func);
 #endif
 	INIT_WORK(&ct->requests.worker, ct_incoming_request_worker_func);
@@ -373,7 +373,7 @@ int intel_guc_ct_enable(struct intel_guc_ct *ct)
 
 	ct->enabled = true;
 	ct->stall_time = KTIME_MAX;
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG)
 	ct->dead_ct_reported = false;
 	ct->dead_ct_reason = CT_DEAD_ALIVE;
 #endif
@@ -1377,7 +1377,7 @@ void intel_guc_ct_print_info(struct intel_guc_ct *ct,
 		   ct->ctbs.recv.desc->tail);
 }
 
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG)
 static void ct_dead_ct_worker_func(struct work_struct *w)
 {
 	struct intel_guc_ct *ct = container_of(w, struct intel_guc_ct, dead_ct_worker);
@@ -1386,6 +1386,9 @@ static void ct_dead_ct_worker_func(struct work_struct *w)
 	if (ct->dead_ct_reported)
 		return;
 
+	if (i915_error_injected())
+		return;
+
 	ct->dead_ct_reported = true;
 
 	guc_info(guc, "CTB is dead - reason=0x%X\n", ct->dead_ct_reason);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
index 2c4bb9a941be..e9a6ec4e6d38 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.h
@@ -97,7 +97,7 @@ struct intel_guc_ct {
 	/** @stall_time: time of first time a CTB submission is stalled */
 	ktime_t stall_time;
 
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GUC)
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG)
 	int dead_ct_reason;
 	bool dead_ct_reported;
 	struct work_struct dead_ct_worker;

From f7a9dc796567b2f1562f83373a5f134a20db25e9 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Tue, 8 Jul 2025 10:03:27 +0530
Subject: [PATCH 0002/1455] drm/i915/scaler: Use intel_display as argument to
 skl_scaler_max_src_size

skl_scaler_max_src_size has really no use of intel_crtc other than
deriving intel_display. Let's just pass intel_display to it directly.

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250708043328.1086192-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index d77798499c57..a6ba9da03542 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -91,11 +91,9 @@ static void skl_scaler_min_src_size(const struct drm_format_info *format,
 	}
 }
 
-static void skl_scaler_max_src_size(struct intel_crtc *crtc,
+static void skl_scaler_max_src_size(struct intel_display *display,
 				    int *max_w, int *max_h)
 {
-	struct intel_display *display = to_intel_display(crtc);
-
 	if (DISPLAY_VER(display) >= 14) {
 		*max_w = 4096;
 		*max_h = 8192;
@@ -201,7 +199,7 @@ skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 	}
 
 	skl_scaler_min_src_size(format, modifier, &min_src_w, &min_src_h);
-	skl_scaler_max_src_size(crtc, &max_src_w, &max_src_h);
+	skl_scaler_max_src_size(display, &max_src_w, &max_src_h);
 
 	skl_scaler_min_dst_size(&min_dst_w, &min_dst_h);
 	skl_scaler_max_dst_size(crtc, &max_dst_w, &max_dst_h);

From 3347b55f2c6c4bebc7a07343448416e5678b8b5c Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Tue, 8 Jul 2025 10:03:28 +0530
Subject: [PATCH 0003/1455] drm/i915/xe3lpd: Prune modes for YUV420

We only support resolution up to 4k for single pipe when using
YUV420 format so we prune these modes and restrict the plane size
at src. This is because pipe scaling will not support YUV420 scaling
for hwidth > 4096.

--v2
-Use output format to check [Ville]
-Add Bspec references
-Modify commit messge to point to why this is needed

--v3
-Use a function skl_scaler_mode_valid which is routed throug
intel_pfit_mode_valid [Ville]
-Combine the check conditons [Jonathan]

--v4
-mode_valid functions should return drm_mode_status [Jani]

--v5
-Use skl_scaler_max_src_size [Ankit]

Bspec: 49247, 50441
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> #v2
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250708043328.1086192-2-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c   | 11 ++++++++---
 drivers/gpu/drm/i915/display/intel_hdmi.c |  4 ++++
 drivers/gpu/drm/i915/display/intel_pfit.c | 11 +++++++++++
 drivers/gpu/drm/i915/display/intel_pfit.h | 12 +++++++++++-
 drivers/gpu/drm/i915/display/skl_scaler.c | 17 +++++++++++++++++
 drivers/gpu/drm/i915/display/skl_scaler.h | 12 ++++++++++++
 6 files changed, 63 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index f48912f308df..62b812cc0b67 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -1418,6 +1418,7 @@ intel_dp_mode_valid(struct drm_connector *_connector,
 	struct intel_display *display = to_intel_display(_connector->dev);
 	struct intel_connector *connector = to_intel_connector(_connector);
 	struct intel_dp *intel_dp = intel_attached_dp(connector);
+	enum intel_output_format sink_format, output_format;
 	const struct drm_display_mode *fixed_mode;
 	int target_clock = mode->clock;
 	int max_rate, mode_rate, max_lanes, max_link_clock;
@@ -1451,6 +1452,13 @@ intel_dp_mode_valid(struct drm_connector *_connector,
 						     mode->hdisplay, target_clock);
 	max_dotclk *= num_joined_pipes;
 
+	sink_format = intel_dp_sink_format(connector, mode);
+	output_format = intel_dp_output_format(connector, sink_format);
+
+	status = intel_pfit_mode_valid(display, mode, output_format, num_joined_pipes);
+	if (status != MODE_OK)
+		return status;
+
 	if (target_clock > max_dotclk)
 		return MODE_CLOCK_HIGH;
 
@@ -1466,11 +1474,8 @@ intel_dp_mode_valid(struct drm_connector *_connector,
 					   intel_dp_mode_min_output_bpp(connector, mode));
 
 	if (intel_dp_has_dsc(connector)) {
-		enum intel_output_format sink_format, output_format;
 		int pipe_bpp;
 
-		sink_format = intel_dp_sink_format(connector, mode);
-		output_format = intel_dp_output_format(connector, sink_format);
 		/*
 		 * TBD pass the connector BPC,
 		 * for now U8_MAX so that max BPC on that platform would be picked
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 9961ff259298..03045d188817 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -2053,6 +2053,10 @@ intel_hdmi_mode_valid(struct drm_connector *_connector,
 	else
 		sink_format = INTEL_OUTPUT_FORMAT_RGB;
 
+	status = intel_pfit_mode_valid(display, mode, sink_format, 0);
+	if (status != MODE_OK)
+		return status;
+
 	status = intel_hdmi_mode_clock_valid(&connector->base, clock, has_hdmi_sink, sink_format);
 	if (status != MODE_OK) {
 		if (ycbcr_420_only ||
diff --git a/drivers/gpu/drm/i915/display/intel_pfit.c b/drivers/gpu/drm/i915/display/intel_pfit.c
index 13541be4d6df..68539e7c2a24 100644
--- a/drivers/gpu/drm/i915/display/intel_pfit.c
+++ b/drivers/gpu/drm/i915/display/intel_pfit.c
@@ -14,6 +14,7 @@
 #include "intel_lvds_regs.h"
 #include "intel_pfit.h"
 #include "intel_pfit_regs.h"
+#include "skl_scaler.h"
 
 static int intel_pch_pfit_check_dst_window(const struct intel_crtc_state *crtc_state)
 {
@@ -546,6 +547,16 @@ out:
 	return intel_gmch_pfit_check_timings(crtc_state);
 }
 
+enum drm_mode_status
+intel_pfit_mode_valid(struct intel_display *display,
+		      const struct drm_display_mode *mode,
+		      enum intel_output_format output_format,
+		      int num_joined_pipes)
+{
+	return skl_scaler_mode_valid(display, mode, output_format,
+				     num_joined_pipes);
+}
+
 int intel_pfit_compute_config(struct intel_crtc_state *crtc_state,
 			      const struct drm_connector_state *conn_state)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_pfit.h b/drivers/gpu/drm/i915/display/intel_pfit.h
index ef34f9b49d09..2f075748a34e 100644
--- a/drivers/gpu/drm/i915/display/intel_pfit.h
+++ b/drivers/gpu/drm/i915/display/intel_pfit.h
@@ -6,8 +6,14 @@
 #ifndef __INTEL_PFIT_H__
 #define __INTEL_PFIT_H__
 
+#include <linux/types.h>
+
+enum drm_mode_status;
+struct drm_display_mode;
 struct drm_connector_state;
 struct intel_crtc_state;
+struct intel_display;
+enum intel_output_format;
 
 int intel_pfit_compute_config(struct intel_crtc_state *crtc_state,
 			      const struct drm_connector_state *conn_state);
@@ -17,5 +23,9 @@ void ilk_pfit_get_config(struct intel_crtc_state *crtc_state);
 void i9xx_pfit_enable(const struct intel_crtc_state *crtc_state);
 void i9xx_pfit_disable(const struct intel_crtc_state *old_crtc_state);
 void i9xx_pfit_get_config(struct intel_crtc_state *crtc_state);
-
+enum drm_mode_status
+intel_pfit_mode_valid(struct intel_display *display,
+		      const struct drm_display_mode *mode,
+		      enum intel_output_format output_format,
+		      int num_joined_pipes);
 #endif /* __INTEL_PFIT_H__ */
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index a6ba9da03542..4cc55f4e1f9f 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -132,6 +132,23 @@ static void skl_scaler_max_dst_size(struct intel_crtc *crtc,
 	}
 }
 
+enum drm_mode_status
+skl_scaler_mode_valid(struct intel_display *display,
+		      const struct drm_display_mode *mode,
+		      enum intel_output_format output_format,
+		      int num_joined_pipes)
+{
+	int max_h, max_w;
+
+	if (num_joined_pipes < 2 && output_format == INTEL_OUTPUT_FORMAT_YCBCR420) {
+		skl_scaler_max_src_size(display, &max_w, &max_h);
+		if (mode->hdisplay > max_h)
+			return MODE_NO_420;
+	}
+
+	return MODE_OK;
+}
+
 static int
 skl_update_scaler(struct intel_crtc_state *crtc_state, bool force_detach,
 		  unsigned int scaler_user, int *scaler_id,
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h
index 355ea15260ca..692716dd7616 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.h
+++ b/drivers/gpu/drm/i915/display/skl_scaler.h
@@ -5,10 +5,16 @@
 #ifndef INTEL_SCALER_H
 #define INTEL_SCALER_H
 
+#include <linux/types.h>
+
+enum drm_mode_status;
+struct drm_display_mode;
 struct intel_atomic_state;
 struct intel_crtc;
 struct intel_crtc_state;
+struct intel_display;
 struct intel_dsb;
+enum intel_output_format;
 struct intel_plane;
 struct intel_plane_state;
 
@@ -32,4 +38,10 @@ void skl_scaler_disable(const struct intel_crtc_state *old_crtc_state);
 
 void skl_scaler_get_config(struct intel_crtc_state *crtc_state);
 
+enum drm_mode_status
+skl_scaler_mode_valid(struct intel_display *display,
+		      const struct drm_display_mode *mode,
+		      enum intel_output_format output_format,
+		      int num_joined_pipes);
+
 #endif

From 8c9006283e4b767003b2d11182d6e90f8b184c3d Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 10 Jul 2025 10:50:40 +0530
Subject: [PATCH 0004/1455] Revert "drm/i915/dp: Reject HBR3 when sink doesn't
 support TPS4"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit 584cf613c24a4250d9be4819efc841aa2624d5b6.
Commit 584cf613c24a ("drm/i915/dp: Reject HBR3 when sink doesn't support
TPS4") introduced a blanket rejection of HBR3 link rate when the sink does
not support TPS4.

While this was intended to address instability observed on certain eDP
panels [1], there seem to be edp panels that do not follow the
specification. These eDP panels do not advertise TPS4 support, but require
HBR3 to operate at their fixed native resolution [2].

As a result, the change causes blank screens on such panels. Apparently,
Windows driver does not enforce this restriction, and the issue is not seen
there.

Therefore, revert the commit to restore functionality for such panels,
and align behaviour with Windows driver.

[1] https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5969
[2] https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14517

v2: Update the commit message with better justification. (Ville)

Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14517
Acked-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250710052041.1238567-2-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 49 ++++---------------------
 1 file changed, 7 insertions(+), 42 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 62b812cc0b67..1819cf1e07ce 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -173,28 +173,10 @@ int intel_dp_link_symbol_clock(int rate)
 
 static int max_dprx_rate(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-	int max_rate;
-
 	if (intel_dp_tunnel_bw_alloc_is_enabled(intel_dp))
-		max_rate = drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel);
-	else
-		max_rate = drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]);
+		return drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel);
 
-	/*
-	 * Some broken eDP sinks illegally declare support for
-	 * HBR3 without TPS4, and are unable to produce a stable
-	 * output. Reject HBR3 when TPS4 is not available.
-	 */
-	if (max_rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) {
-		drm_dbg_kms(display->drm,
-			    "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n",
-			    encoder->base.base.id, encoder->base.name);
-		max_rate = 540000;
-	}
-
-	return max_rate;
+	return drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]);
 }
 
 static int max_dprx_lane_count(struct intel_dp *intel_dp)
@@ -4278,9 +4260,6 @@ static void intel_edp_mso_init(struct intel_dp *intel_dp)
 static void
 intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
-
 	intel_dp->num_sink_rates = 0;
 
 	if (intel_dp->edp_dpcd[0] >= DP_EDP_14) {
@@ -4291,7 +4270,10 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 				 sink_rates, sizeof(sink_rates));
 
 		for (i = 0; i < ARRAY_SIZE(sink_rates); i++) {
-			int rate;
+			int val = le16_to_cpu(sink_rates[i]);
+
+			if (val == 0)
+				break;
 
 			/* Value read multiplied by 200kHz gives the per-lane
 			 * link rate in kHz. The source rates are, however,
@@ -4299,24 +4281,7 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 			 * back to symbols is
 			 * (val * 200kHz)*(8/10 ch. encoding)*(1/8 bit to Byte)
 			 */
-			rate = le16_to_cpu(sink_rates[i]) * 200 / 10;
-
-			if (rate == 0)
-				break;
-
-			/*
-			 * Some broken eDP sinks illegally declare support for
-			 * HBR3 without TPS4, and are unable to produce a stable
-			 * output. Reject HBR3 when TPS4 is not available.
-			 */
-			if (rate >= 810000 && !drm_dp_tps4_supported(intel_dp->dpcd)) {
-				drm_dbg_kms(display->drm,
-					    "[ENCODER:%d:%s] Rejecting HBR3 due to missing TPS4 support\n",
-					    encoder->base.base.id, encoder->base.name);
-				break;
-			}
-
-			intel_dp->sink_rates[i] = rate;
+			intel_dp->sink_rates[i] = (val * 200) / 10;
 		}
 		intel_dp->num_sink_rates = i;
 	}

From 21c586d9233a1f258e8d437466c441d50885d30f Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Thu, 10 Jul 2025 10:50:41 +0530
Subject: [PATCH 0005/1455] drm/i915/dp: Add device specific quirk to limit eDP
 rate to HBR2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some ICL/TGL platforms with combo PHY ports suffer from signal integrity
issues at HBR3. While certain systems include a Parade PS8461 mux to
mitigate this, its presence cannot be reliably detected. Furthermore,
broken or missing VBT entries make it unsafe to rely on VBT for enforcing
link rate limits.

To address this introduce a device specific quirk to cap the eDP link rate
to HBR2 (540000 kHz). This will override any higher advertised rates from
the sink or DPCD for specific devices.

Currently, the quirk is added for Dell XPS 13 7390 2-in-1 which is reported
in gitlab issue #5969 [1].

[1] https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5969

v2: Align the quirk with the intended quirk name and refactor the
condition to use min(). (Jani)
v3: Use condition `rate > 540000`. Drop extra parentheses. (Ville)

Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5969
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250710052041.1238567-3-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c     | 43 +++++++++++++++++----
 drivers/gpu/drm/i915/display/intel_quirks.c |  9 +++++
 drivers/gpu/drm/i915/display/intel_quirks.h |  1 +
 3 files changed, 45 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 1819cf1e07ce..6d4e0fa0c054 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -173,10 +173,24 @@ int intel_dp_link_symbol_clock(int rate)
 
 static int max_dprx_rate(struct intel_dp *intel_dp)
 {
-	if (intel_dp_tunnel_bw_alloc_is_enabled(intel_dp))
-		return drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel);
+	struct intel_display *display = to_intel_display(intel_dp);
+	int max_rate;
 
-	return drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]);
+	if (intel_dp_tunnel_bw_alloc_is_enabled(intel_dp))
+		max_rate = drm_dp_tunnel_max_dprx_rate(intel_dp->tunnel);
+	else
+		max_rate = drm_dp_bw_code_to_link_rate(intel_dp->dpcd[DP_MAX_LINK_RATE]);
+
+	/*
+	 * Some platforms + eDP panels may not reliably support HBR3
+	 * due to signal integrity limitations, despite advertising it.
+	 * Cap the link rate to HBR2 to avoid unstable configurations for the
+	 * known machines.
+	 */
+	if (intel_dp_is_edp(intel_dp) && intel_has_quirk(display, QUIRK_EDP_LIMIT_RATE_HBR2))
+		max_rate = min(max_rate, 540000);
+
+	return max_rate;
 }
 
 static int max_dprx_lane_count(struct intel_dp *intel_dp)
@@ -4260,6 +4274,8 @@ static void intel_edp_mso_init(struct intel_dp *intel_dp)
 static void
 intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
+
 	intel_dp->num_sink_rates = 0;
 
 	if (intel_dp->edp_dpcd[0] >= DP_EDP_14) {
@@ -4270,10 +4286,7 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 				 sink_rates, sizeof(sink_rates));
 
 		for (i = 0; i < ARRAY_SIZE(sink_rates); i++) {
-			int val = le16_to_cpu(sink_rates[i]);
-
-			if (val == 0)
-				break;
+			int rate;
 
 			/* Value read multiplied by 200kHz gives the per-lane
 			 * link rate in kHz. The source rates are, however,
@@ -4281,7 +4294,21 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 			 * back to symbols is
 			 * (val * 200kHz)*(8/10 ch. encoding)*(1/8 bit to Byte)
 			 */
-			intel_dp->sink_rates[i] = (val * 200) / 10;
+			rate = le16_to_cpu(sink_rates[i]) * 200 / 10;
+
+			if (rate == 0)
+				break;
+
+			/*
+			 * Some platforms cannot reliably drive HBR3 rates due to PHY limitations,
+			 * even if the sink advertises support. Reject any sink rates above HBR2 on
+			 * the known machines for stable output.
+			 */
+			if (rate > 540000 &&
+			    intel_has_quirk(display, QUIRK_EDP_LIMIT_RATE_HBR2))
+				break;
+
+			intel_dp->sink_rates[i] = rate;
 		}
 		intel_dp->num_sink_rates = i;
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_quirks.c b/drivers/gpu/drm/i915/display/intel_quirks.c
index a32fae510ed2..d2e16b79d6be 100644
--- a/drivers/gpu/drm/i915/display/intel_quirks.c
+++ b/drivers/gpu/drm/i915/display/intel_quirks.c
@@ -80,6 +80,12 @@ static void quirk_fw_sync_len(struct intel_dp *intel_dp)
 	drm_info(display->drm, "Applying Fast Wake sync pulse count quirk\n");
 }
 
+static void quirk_edp_limit_rate_hbr2(struct intel_display *display)
+{
+	intel_set_quirk(display, QUIRK_EDP_LIMIT_RATE_HBR2);
+	drm_info(display->drm, "Applying eDP Limit rate to HBR2 quirk\n");
+}
+
 struct intel_quirk {
 	int device;
 	int subsystem_vendor;
@@ -231,6 +237,9 @@ static struct intel_quirk intel_quirks[] = {
 	{ 0x3184, 0x1019, 0xa94d, quirk_increase_ddi_disabled_time },
 	/* HP Notebook - 14-r206nv */
 	{ 0x0f31, 0x103c, 0x220f, quirk_invert_brightness },
+
+	/* Dell XPS 13 7390 2-in-1 */
+	{ 0x8a12, 0x1028, 0x08b0, quirk_edp_limit_rate_hbr2 },
 };
 
 static const struct intel_dpcd_quirk intel_dpcd_quirks[] = {
diff --git a/drivers/gpu/drm/i915/display/intel_quirks.h b/drivers/gpu/drm/i915/display/intel_quirks.h
index cafdebda7535..06da0e286c67 100644
--- a/drivers/gpu/drm/i915/display/intel_quirks.h
+++ b/drivers/gpu/drm/i915/display/intel_quirks.h
@@ -20,6 +20,7 @@ enum intel_quirk_id {
 	QUIRK_LVDS_SSC_DISABLE,
 	QUIRK_NO_PPS_BACKLIGHT_POWER_HOOK,
 	QUIRK_FW_SYNC_LEN,
+	QUIRK_EDP_LIMIT_RATE_HBR2,
 };
 
 void intel_init_quirks(struct intel_display *display);

From bba9aa41654036534d86b198f5647a9ce15ebd7f Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 9 Jul 2025 00:23:31 +0300
Subject: [PATCH 0006/1455] drm/dp: Change AUX DPCD probe address from
 LANE0_1_STATUS to TRAINING_PATTERN_SET
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit a40c5d727b81 ("drm/dp: Change AUX DPCD probe address from
DPCD_REV to LANE0_1_STATUS") stopped using the DPCD_REV register for
DPCD probing, since this results in link training failures at least when
using an Intel Barlow Ridge TBT hub at UHBR link rates (the
DP_INTRA_HOP_AUX_REPLY_INDICATION never getting cleared after the failed
link training). Since accessing DPCD_REV during link training is
prohibited by the DP Standard, LANE0_1_STATUS (0x202) was used instead,
as it falls within the Standard's valid register address range
(0x102-0x106, 0x202-0x207, 0x200c-0x200f, 0x2216) and it fixed the link
training on the above TBT hub.

However, reading the LANE0_1_STATUS register also has a side-effect at
least on a Novatek eDP panel, as reported on the Closes: link below,
resulting in screen flickering on that panel. One clear side-effect when
doing the 1-byte probe reads from LANE0_1_STATUS during link training
before reading out the full 6 byte link status starting at the same
address is that the panel will report the link training as completed
with voltage swing 0. This is different from the normal, flicker-free
scenario when no DPCD probing is done, the panel reporting the link
training complete with voltage swing 2.

Using the TRAINING_PATTERN_SET register for DPCD probing doesn't have
the above side-effect, the panel will link train with voltage swing 2 as
expected and it will stay flicker-free. This register is also in the
above valid register range and is unlikely to have a side-effect as that
of LANE0_1_STATUS: Reading LANE0_1_STATUS is part of the link training
CR/EQ sequences and so it may cause a state change in the sink - even if
inadvertently as I suspect in the case of the above Novatek panel. As
opposed to this, reading TRAINING_PATTERN_SET is not part of the link
training sequence (it must be only written once at the beginning of the
CR/EQ sequences), so it's unlikely to cause any state change in the
sink.

As a side-note, this Novatek panel also lacks support for TPS3, while
claiming support for HBR2, which violates the DP Standard (the Standard
mandating TPS3 for HBR2).

Besides the Novatek panel (PSR 1), which this change fixes, I also
verified the change on a Samsung (PSR 1) and an Analogix (PSR 2) eDP
panel as well as on the Intel Barlow Ridge TBT hub.

Note that in the drm-tip tree (targeting the v6.17 kernel version) the
i915 and xe drivers keep DPCD probing enabled only for the panel known
to require this (HP ZR24w), hence those drivers in drm-tip are not
affected by the above problem.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Fixes: a40c5d727b81 ("drm/dp: Change AUX DPCD probe address from DPCD_REV to LANE0_1_STATUS")
Reported-and-tested-by: Paul Menzel <pmenzel@molgen.mpg.de>
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14558
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250708212331.112898-1-imre.deak@intel.com
---
 drivers/gpu/drm/display/drm_dp_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
index 385a1bfdb272..e631ab81aaf5 100644
--- a/drivers/gpu/drm/display/drm_dp_helper.c
+++ b/drivers/gpu/drm/display/drm_dp_helper.c
@@ -742,7 +742,7 @@ ssize_t drm_dp_dpcd_read(struct drm_dp_aux *aux, unsigned int offset,
 	int ret;
 
 	if (dpcd_access_needs_probe(aux)) {
-		ret = drm_dp_dpcd_probe(aux, DP_LANE0_1_STATUS);
+		ret = drm_dp_dpcd_probe(aux, DP_TRAINING_PATTERN_SET);
 		if (ret < 0)
 			return ret;
 	}

From 8a643df55f3a9c034cf8d1942c25a6eb08d57d9d Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 14 Jul 2025 17:39:48 +0200
Subject: [PATCH 0007/1455] drm/i915: Don't check for atomic context on
 PREEMPT_RT

The !in_atomic() check in _wait_for_atomic() triggers on PREEMPT_RT
because the uncore::lock is a spinlock_t and does not disable
preemption or interrupts.

Changing the uncore:lock to a raw_spinlock_t doubles the worst case
latency on an otherwise idle testbox during testing.

Ignore _WAIT_FOR_ATOMIC_CHECK() on PREEMPT_RT.

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Link: https://lore.kernel.org/all/20211006164628.s2mtsdd2jdbfyf7g@linutronix.de/
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Link: https://lore.kernel.org/r/20250714153954.629393-4-bigeasy@linutronix.de
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
---
 drivers/gpu/drm/i915/i915_utils.h | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index f7fb40cfdb70..9cb40c2c4b12 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -267,8 +267,13 @@ wait_remaining_ms_from_jiffies(unsigned long timestamp_jiffies, int to_wait_ms)
 						   (Wmax))
 #define wait_for(COND, MS)		_wait_for((COND), (MS) * 1000, 10, 1000)
 
-/* If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false. */
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) && IS_ENABLED(CONFIG_PREEMPT_COUNT)
+/*
+ * If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false.
+ * On PREEMPT_RT the context isn't becoming atomic because it is used in an
+ * interrupt handler or because a spinlock_t is acquired. This leads to
+ * warnings which don't occur otherwise and therefore the check is disabled.
+ */
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) && IS_ENABLED(CONFIG_PREEMPT_COUNT) && !defined(CONFIG_PREEMPT_RT)
 # define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) WARN_ON_ONCE((ATOMIC) && !in_atomic())
 #else
 # define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) do { } while (0)

From ed5461daa150b037e36b8202381da1ef85d6b16b Mon Sep 17 00:00:00 2001
From: Raag Jadav <raag.jadav@intel.com>
Date: Tue, 15 Jul 2025 03:25:03 +0530
Subject: [PATCH 0008/1455] drm/xe: Don't fail probe on unsupported mailbox
 command

If the device is running older pcode firmware, it is possible that newer
mailbox commands are not supported by it. The sysfs attributes aren't
useful in that case, but we shouldn't fail driver probe because of it.
As of now, it is unknown if we can distinguish unsupported commands before
attempting them. But until we figure out a way to do that, fix the
regressions.

v2: Add debug message (Lucas)

Fixes: cdc36b66cd41 ("drm/xe: Expose fan control and voltage regulator version")
Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Tested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250714215503.2897748-1-raag.jadav@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_device_sysfs.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index e5fd0cd537bc..bd9015761aa0 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -160,8 +160,13 @@ static int late_bind_create_files(struct device *dev)
 
 	ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0),
 			    &cap, NULL);
-	if (ret)
+	if (ret) {
+		if (ret == -ENXIO) {
+			drm_dbg(&xe->drm, "Late binding not supported by firmware\n");
+			ret = 0;
+		}
 		goto out;
+	}
 
 	if (REG_FIELD_GET(V1_FAN_SUPPORTED, cap)) {
 		ret = sysfs_create_file(&dev->kobj, &dev_attr_lb_fan_control_version.attr);

From 308dc9b27874d0e8a0258869b9e681b0fdd2e579 Mon Sep 17 00:00:00 2001
From: Ashutosh Dixit <ashutosh.dixit@intel.com>
Date: Tue, 15 Jul 2025 11:14:22 -0700
Subject: [PATCH 0009/1455] drm/xe/oa: Fix static checker warning about null gt

There is a static checker warning that gt returned by xe_device_get_gt can
be NULL and that is being dereferenced. Use xe_root_mmio_gt instead, which
is equivalent and cannot return a NULL gt 0.

Fixes: 10d42ef34bce ("drm/xe/oa: Assign hwe for OAM_SAG")
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://lore.kernel.org/r/20250715181422.2807624-1-ashutosh.dixit@intel.com
---
 drivers/gpu/drm/xe/xe_oa.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c
index d991fbd90f20..5729e7d3e335 100644
--- a/drivers/gpu/drm/xe/xe_oa.c
+++ b/drivers/gpu/drm/xe/xe_oa.c
@@ -1941,7 +1941,7 @@ static int xe_oa_assign_hwe(struct xe_oa *oa, struct xe_oa_open_param *param)
 
 	/* If not provided, OA unit defaults to OA unit 0 as per uapi */
 	if (!param->oa_unit)
-		param->oa_unit = &xe_device_get_gt(oa->xe, 0)->oa.oa_unit[0];
+		param->oa_unit = &xe_root_mmio_gt(oa->xe)->oa.oa_unit[0];
 
 	/* When we have an exec_q, get hwe from the exec_q */
 	if (param->exec_q) {

From 2f264d58cc805a3cefc6b98097f90fbc388136ef Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Tue, 15 Jul 2025 17:59:44 -0500
Subject: [PATCH 0010/1455] drm/xe: Fix a NULL vs IS_ERR() bug in
 xe_i2c_register_adapter()

The fwnode_create_software_node() function returns error pointers.  It
never returns NULL.  Update the checks to match.

Fixes: f0e53aadd702 ("drm/xe: Support for I2C attached MCUs")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/65825d00-81ab-4665-af51-4fff6786a250@sabinyo.mountain
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_i2c.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_i2c.c b/drivers/gpu/drm/xe/xe_i2c.c
index db9c0340be5c..1f19718db559 100644
--- a/drivers/gpu/drm/xe/xe_i2c.c
+++ b/drivers/gpu/drm/xe/xe_i2c.c
@@ -96,8 +96,8 @@ static int xe_i2c_register_adapter(struct xe_i2c *i2c)
 	int ret;
 
 	fwnode = fwnode_create_software_node(xe_i2c_adapter_properties, NULL);
-	if (!fwnode)
-		return -ENOMEM;
+	if (IS_ERR(fwnode))
+		return PTR_ERR(fwnode);
 
 	/*
 	 * Not using platform_device_register_full() here because we don't have

From bf81505f7dba467c17097315864790f9f7dca08a Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 14 Jul 2025 21:36:45 +0200
Subject: [PATCH 0011/1455] drm/xe: Move debugfs GT attributes under tile
 directory
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

While in sysfs we are correctly trying to reflect the hardware
architecture and we expose GT attributes in per tile hierarchy,
in debugfs we expose GT attributes at flat level, without tiles.

Create debugfs directories to represent each tile and move GT
attributes under matching parent tile directory. To not break
existing debugfs tools, create symlink under old location:

 /sys/kernel/debug/dri/0000:00:02.0/
 ├── ...
 ├── gt0 -> tile0/gt0
 ├── gt1 -> tile0/gt1
 ├── tile0
 │   ├── gt0
 │   │   ├── ...
 │   ├── gt1
 │   │   ├── ...

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250714193645.763-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_debugfs.c      | 22 ++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_device_types.h |  3 +++
 drivers/gpu/drm/xe/xe_gt_debugfs.c   | 14 +++++++++++++-
 3 files changed, 38 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 26e9d146ccbf..129186f4193b 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -247,14 +247,33 @@ static const struct file_operations atomic_svm_timeslice_ms_fops = {
 	.write = atomic_svm_timeslice_ms_set,
 };
 
+static void create_tile_debugfs(struct xe_tile *tile, struct dentry *root)
+{
+	char name[8];
+
+	snprintf(name, sizeof(name), "tile%u", tile->id);
+	tile->debugfs = debugfs_create_dir(name, root);
+	if (IS_ERR(tile->debugfs))
+		return;
+
+	/*
+	 * Store the xe_tile pointer as private data of the tile/ directory
+	 * node so other tile specific attributes under that directory may
+	 * refer to it by looking at its parent node private data.
+	 */
+	tile->debugfs->d_inode->i_private = tile;
+}
+
 void xe_debugfs_register(struct xe_device *xe)
 {
 	struct ttm_device *bdev = &xe->ttm;
 	struct drm_minor *minor = xe->drm.primary;
 	struct dentry *root = minor->debugfs_root;
 	struct ttm_resource_manager *man;
+	struct xe_tile *tile;
 	struct xe_gt *gt;
 	u32 mem_type;
+	u8 tile_id;
 	u8 id;
 
 	drm_debugfs_create_files(debugfs_list,
@@ -288,6 +307,9 @@ void xe_debugfs_register(struct xe_device *xe)
 	if (man)
 		ttm_resource_manager_create_debugfs(man, root, "stolen_mm");
 
+	for_each_tile(tile, xe, tile_id)
+		create_tile_debugfs(tile, root);
+
 	for_each_gt(gt, xe, id)
 		xe_gt_debugfs_register(gt);
 
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index d4d2c6854790..98ce1255f1d0 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -255,6 +255,9 @@ struct xe_tile {
 
 	/** @sysfs: sysfs' kobj used by xe_tile_sysfs */
 	struct kobject *sysfs;
+
+	/** @debugfs: debugfs directory associated with this tile */
+	struct dentry *debugfs;
 };
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index 848618acdca8..061a81f82e76 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -388,13 +388,18 @@ void xe_gt_debugfs_register(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 	struct drm_minor *minor = gt_to_xe(gt)->drm.primary;
+	struct dentry *parent = gt->tile->debugfs;
 	struct dentry *root;
+	char symlink[16];
 	char name[8];
 
 	xe_gt_assert(gt, minor->debugfs_root);
 
+	if (IS_ERR(parent))
+		return;
+
 	snprintf(name, sizeof(name), "gt%d", gt->info.id);
-	root = debugfs_create_dir(name, minor->debugfs_root);
+	root = debugfs_create_dir(name, parent);
 	if (IS_ERR(root)) {
 		drm_warn(&xe->drm, "Create GT directory failed");
 		return;
@@ -426,4 +431,11 @@ void xe_gt_debugfs_register(struct xe_gt *gt)
 		xe_gt_sriov_pf_debugfs_register(gt, root);
 	else if (IS_SRIOV_VF(xe))
 		xe_gt_sriov_vf_debugfs_register(gt, root);
+
+	/*
+	 * Backwards compatibility only: create a link for the legacy clients
+	 * who may expect gt/ directory at the root level, not the tile level.
+	 */
+	snprintf(symlink, sizeof(symlink), "tile%u/%s", gt->tile->id, name);
+	debugfs_create_symlink(name, minor->debugfs_root, symlink);
 }

From 922ae875230be91c7f05f2aa90d176b6693e2601 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 14 Jul 2025 20:48:14 +0200
Subject: [PATCH 0012/1455] drm/xe: Use devm_ioremap_wc for VRAM mapping and
 drop manual unmap
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Let's replace the manual call to ioremap_wc function with devm_ioremap_wc
function, ensuring that VRAM mappings are automatically released when
the driver is detached.
Since devm_ioremap_wc registers the mapping with the device's managed
resources, the explicit iounmap call in vram_fini is no longer needed,
so let's remove it.

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Acked-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250714184818.89201-2-piotr.piorkowski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_vram.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c
index e421a74fb87c..3a4c84e9efc6 100644
--- a/drivers/gpu/drm/xe/xe_vram.c
+++ b/drivers/gpu/drm/xe/xe_vram.c
@@ -156,7 +156,8 @@ static int determine_lmem_bar_size(struct xe_device *xe)
 	xe->mem.vram.dpa_base = 0;
 
 	/* set up a map to the total memory area. */
-	xe->mem.vram.mapping = ioremap_wc(xe->mem.vram.io_start, xe->mem.vram.io_size);
+	xe->mem.vram.mapping = devm_ioremap_wc(&pdev->dev, xe->mem.vram.io_start,
+					       xe->mem.vram.io_size);
 
 	return 0;
 }
@@ -278,9 +279,6 @@ static void vram_fini(void *arg)
 	struct xe_tile *tile;
 	int id;
 
-	if (xe->mem.vram.mapping)
-		iounmap(xe->mem.vram.mapping);
-
 	xe->mem.vram.mapping = NULL;
 
 	for_each_tile(tile, xe, id)

From f92cfd72d9a650f90260c54accd840c6500c4c3a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 14 Jul 2025 20:48:15 +0200
Subject: [PATCH 0013/1455] drm/xe: Use dynamic allocation for tile and device
 VRAM region structures
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In future platforms, we will need to represent the device and tile
VRAM regions in a more dynamic way, so let's abandon the static
allocation of these structures and start use a dynamic allocation.

v2:
 - Add a helpers for accessing fields of the xe_vram_region structure
v3:
- Add missing EXPORT_SYMBOL_IF_KUNIT for
  xe_vram_region_actual_physical_size

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Acked-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250714184818.89201-3-piotr.piorkowski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/display/xe_fb_pin.c        |   4 +-
 drivers/gpu/drm/xe/display/xe_plane_initial.c |   2 +-
 drivers/gpu/drm/xe/xe_assert.h                |   4 +-
 drivers/gpu/drm/xe/xe_device.c                |  19 +++
 drivers/gpu/drm/xe/xe_device_types.h          |   6 +-
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c    |   2 +-
 drivers/gpu/drm/xe/xe_migrate.c               |  25 ++--
 drivers/gpu/drm/xe/xe_pci.c                   |   6 +
 drivers/gpu/drm/xe/xe_query.c                 |   2 +-
 drivers/gpu/drm/xe/xe_svm.c                   |  24 +---
 drivers/gpu/drm/xe/xe_tile.c                  |  34 ++++-
 drivers/gpu/drm/xe/xe_tile.h                  |   4 +-
 drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c        |  10 +-
 drivers/gpu/drm/xe/xe_ttm_vram_mgr.c          |   7 +-
 drivers/gpu/drm/xe/xe_vram.c                  | 121 +++++++++++++-----
 drivers/gpu/drm/xe/xe_vram.h                  |   9 ++
 16 files changed, 202 insertions(+), 77 deletions(-)

diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index 405967df2c29..a1b203f40390 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -289,7 +289,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 	if (IS_DGFX(to_xe_device(bo->ttm.base.dev)) &&
 	    intel_fb_rc_ccs_cc_plane(&fb->base) >= 0 &&
 	    !(bo->flags & XE_BO_FLAG_NEEDS_CPU_ACCESS)) {
-		struct xe_tile *tile = xe_device_get_root_tile(xe);
+		struct xe_vram_region *vram = xe_device_get_root_tile(xe)->mem.vram;
 
 		/*
 		 * If we need to able to access the clear-color value stored in
@@ -297,7 +297,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 		 * accessible.  This is important on small-bar systems where
 		 * only some subset of VRAM is CPU accessible.
 		 */
-		if (tile->mem.vram.io_size < tile->mem.vram.usable_size) {
+		if (xe_vram_region_io_size(vram) < xe_vram_region_usable_size(vram)) {
 			ret = -EINVAL;
 			goto err;
 		}
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index af47ce34102c..30a341c96e76 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -103,7 +103,7 @@ initial_plane_bo(struct xe_device *xe,
 		 * We don't currently expect this to ever be placed in the
 		 * stolen portion.
 		 */
-		if (phys_base >= tile0->mem.vram.usable_size) {
+		if (phys_base >= xe_vram_region_usable_size(tile0->mem.vram)) {
 			drm_err(&xe->drm,
 				"Initial plane programming using invalid range, phys_base=%pa\n",
 				&phys_base);
diff --git a/drivers/gpu/drm/xe/xe_assert.h b/drivers/gpu/drm/xe/xe_assert.h
index 68fe70ce2be3..a818eaa05b7d 100644
--- a/drivers/gpu/drm/xe/xe_assert.h
+++ b/drivers/gpu/drm/xe/xe_assert.h
@@ -12,6 +12,7 @@
 
 #include "xe_gt_types.h"
 #include "xe_step.h"
+#include "xe_vram.h"
 
 /**
  * DOC: Xe Asserts
@@ -145,7 +146,8 @@
 	const struct xe_tile *__tile = (tile);							\
 	char __buf[10] __maybe_unused;								\
 	xe_assert_msg(tile_to_xe(__tile), condition, "tile: %u VRAM %s\n" msg,			\
-		      __tile->id, ({ string_get_size(__tile->mem.vram.actual_physical_size, 1,	\
+		      __tile->id, ({ string_get_size(						\
+				     xe_vram_region_actual_physical_size(__tile->mem.vram), 1,	\
 				     STRING_UNITS_2, __buf, sizeof(__buf)); __buf; }), ## arg);	\
 })
 
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 6dc84e4ed281..df6f1790d6a6 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -687,6 +687,21 @@ static void sriov_update_device_info(struct xe_device *xe)
 	}
 }
 
+static int xe_device_vram_alloc(struct xe_device *xe)
+{
+	struct xe_vram_region *vram;
+
+	if (!IS_DGFX(xe))
+		return 0;
+
+	vram = drmm_kzalloc(&xe->drm, sizeof(*vram), GFP_KERNEL);
+	if (!vram)
+		return -ENOMEM;
+
+	xe->mem.vram = vram;
+	return 0;
+}
+
 /**
  * xe_device_probe_early: Device early probe
  * @xe: xe device instance
@@ -734,6 +749,10 @@ int xe_device_probe_early(struct xe_device *xe)
 
 	xe->wedged.mode = xe_modparam.wedged_mode;
 
+	err = xe_device_vram_alloc(xe);
+	if (err)
+		return err;
+
 	return 0;
 }
 ALLOW_ERROR_INJECTION(xe_device_probe_early, ERRNO); /* See xe_pci_probe() */
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 98ce1255f1d0..6299f4571c4c 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -77,6 +77,8 @@ struct xe_pxp;
  * device, such as HBM memory or CXL extension memory.
  */
 struct xe_vram_region {
+	/** @tile: Back pointer to tile */
+	struct xe_tile *tile;
 	/** @io_start: IO start address of this VRAM instance */
 	resource_size_t io_start;
 	/**
@@ -216,7 +218,7 @@ struct xe_tile {
 		 * Although VRAM is associated with a specific tile, it can
 		 * still be accessed by all tiles' GTs.
 		 */
-		struct xe_vram_region vram;
+		struct xe_vram_region *vram;
 
 		/** @mem.ggtt: Global graphics translation table */
 		struct xe_ggtt *ggtt;
@@ -415,7 +417,7 @@ struct xe_device {
 	/** @mem: memory info for device */
 	struct {
 		/** @mem.vram: VRAM info for device */
-		struct xe_vram_region vram;
+		struct xe_vram_region *vram;
 		/** @mem.sys_mgr: system TTM manager */
 		struct ttm_resource_manager sys_mgr;
 		/** @mem.sys_mgr: system memory shrinker. */
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index 494909f74eb2..86889df04011 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -1604,7 +1604,7 @@ static u64 pf_query_free_lmem(struct xe_gt *gt)
 {
 	struct xe_tile *tile = gt->tile;
 
-	return xe_ttm_vram_get_avail(&tile->mem.vram.ttm.manager);
+	return xe_ttm_vram_get_avail(&tile->mem.vram->ttm.manager);
 }
 
 static u64 pf_query_max_lmem(struct xe_gt *gt)
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index ba1cff2e4cda..d6186ac530cb 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -34,6 +34,7 @@
 #include "xe_sync.h"
 #include "xe_trace_bo.h"
 #include "xe_vm.h"
+#include "xe_vram.h"
 
 /**
  * struct xe_migrate - migrate context.
@@ -130,34 +131,36 @@ static u64 xe_migrate_vram_ofs(struct xe_device *xe, u64 addr, bool is_comp_pte)
 	u64 identity_offset = IDENTITY_OFFSET;
 
 	if (GRAPHICS_VER(xe) >= 20 && is_comp_pte)
-		identity_offset += DIV_ROUND_UP_ULL(xe->mem.vram.actual_physical_size, SZ_1G);
+		identity_offset += DIV_ROUND_UP_ULL(xe_vram_region_actual_physical_size
+							(xe->mem.vram), SZ_1G);
 
-	addr -= xe->mem.vram.dpa_base;
+	addr -= xe_vram_region_dpa_base(xe->mem.vram);
 	return addr + (identity_offset << xe_pt_shift(2));
 }
 
 static void xe_migrate_program_identity(struct xe_device *xe, struct xe_vm *vm, struct xe_bo *bo,
 					u64 map_ofs, u64 vram_offset, u16 pat_index, u64 pt_2m_ofs)
 {
+	struct xe_vram_region *vram = xe->mem.vram;
+	resource_size_t dpa_base = xe_vram_region_dpa_base(vram);
 	u64 pos, ofs, flags;
 	u64 entry;
 	/* XXX: Unclear if this should be usable_size? */
-	u64 vram_limit =  xe->mem.vram.actual_physical_size +
-		xe->mem.vram.dpa_base;
+	u64 vram_limit = xe_vram_region_actual_physical_size(vram) + dpa_base;
 	u32 level = 2;
 
 	ofs = map_ofs + XE_PAGE_SIZE * level + vram_offset * 8;
 	flags = vm->pt_ops->pte_encode_addr(xe, 0, pat_index, level,
 					    true, 0);
 
-	xe_assert(xe, IS_ALIGNED(xe->mem.vram.usable_size, SZ_2M));
+	xe_assert(xe, IS_ALIGNED(xe_vram_region_usable_size(vram), SZ_2M));
 
 	/*
 	 * Use 1GB pages when possible, last chunk always use 2M
 	 * pages as mixing reserved memory (stolen, WOCPM) with a single
 	 * mapping is not allowed on certain platforms.
 	 */
-	for (pos = xe->mem.vram.dpa_base; pos < vram_limit;
+	for (pos = dpa_base; pos < vram_limit;
 	     pos += SZ_1G, ofs += 8) {
 		if (pos + SZ_1G >= vram_limit) {
 			entry = vm->pt_ops->pde_encode_bo(bo, pt_2m_ofs,
@@ -307,11 +310,11 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 	/* Identity map the entire vram at 256GiB offset */
 	if (IS_DGFX(xe)) {
 		u64 pt30_ofs = xe_bo_size(bo) - 2 * XE_PAGE_SIZE;
+		resource_size_t actual_phy_size = xe_vram_region_actual_physical_size(xe->mem.vram);
 
 		xe_migrate_program_identity(xe, vm, bo, map_ofs, IDENTITY_OFFSET,
 					    pat_index, pt30_ofs);
-		xe_assert(xe, xe->mem.vram.actual_physical_size <=
-					(MAX_NUM_PTE - IDENTITY_OFFSET) * SZ_1G);
+		xe_assert(xe, actual_phy_size <= (MAX_NUM_PTE - IDENTITY_OFFSET) * SZ_1G);
 
 		/*
 		 * Identity map the entire vram for compressed pat_index for xe2+
@@ -320,11 +323,11 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 		if (GRAPHICS_VER(xe) >= 20 && xe_device_has_flat_ccs(xe)) {
 			u16 comp_pat_index = xe->pat.idx[XE_CACHE_NONE_COMPRESSION];
 			u64 vram_offset = IDENTITY_OFFSET +
-				DIV_ROUND_UP_ULL(xe->mem.vram.actual_physical_size, SZ_1G);
+				DIV_ROUND_UP_ULL(actual_phy_size, SZ_1G);
 			u64 pt31_ofs = xe_bo_size(bo) - XE_PAGE_SIZE;
 
-			xe_assert(xe, xe->mem.vram.actual_physical_size <= (MAX_NUM_PTE -
-						IDENTITY_OFFSET - IDENTITY_OFFSET / 2) * SZ_1G);
+			xe_assert(xe, actual_phy_size <= (MAX_NUM_PTE - IDENTITY_OFFSET -
+							  IDENTITY_OFFSET / 2) * SZ_1G);
 			xe_migrate_program_identity(xe, vm, bo, map_ofs, vram_offset,
 						    comp_pat_index, pt31_ofs);
 		}
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 3c40ef426f0c..f64942737a0b 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -687,6 +687,8 @@ static int xe_info_init(struct xe_device *xe,
 	 * All of these together determine the overall GT count.
 	 */
 	for_each_tile(tile, xe, id) {
+		int err;
+
 		gt = tile->primary_gt;
 		gt->info.type = XE_GT_TYPE_MAIN;
 		gt->info.id = tile->id * xe->info.max_gt_per_tile;
@@ -694,6 +696,10 @@ static int xe_info_init(struct xe_device *xe,
 		gt->info.engine_mask = graphics_desc->hw_engine_mask;
 		xe->info.gt_count++;
 
+		err = xe_tile_alloc_vram(tile);
+		if (err)
+			return err;
+
 		if (MEDIA_VER(xe) < 13 && media_desc)
 			gt->info.engine_mask |= media_desc->hw_engine_mask;
 
diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
index d517ec9ddcbf..66ea3403afbe 100644
--- a/drivers/gpu/drm/xe/xe_query.c
+++ b/drivers/gpu/drm/xe/xe_query.c
@@ -337,7 +337,7 @@ static int query_config(struct xe_device *xe, struct drm_xe_device_query *query)
 	config->num_params = num_params;
 	config->info[DRM_XE_QUERY_CONFIG_REV_AND_DEVICE_ID] =
 		xe->info.devid | (xe->info.revid << 16);
-	if (xe_device_get_root_tile(xe)->mem.vram.usable_size)
+	if (xe->mem.vram)
 		config->info[DRM_XE_QUERY_CONFIG_FLAGS] |=
 			DRM_XE_QUERY_CONFIG_FLAG_HAS_VRAM;
 	if (xe->info.has_usm && IS_ENABLED(CONFIG_DRM_XE_GPUSVM))
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index a7ff5975873f..e6871734ffa9 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -306,16 +306,11 @@ static struct xe_vram_region *page_to_vr(struct page *page)
 	return container_of(page_pgmap(page), struct xe_vram_region, pagemap);
 }
 
-static struct xe_tile *vr_to_tile(struct xe_vram_region *vr)
-{
-	return container_of(vr, struct xe_tile, mem.vram);
-}
-
 static u64 xe_vram_region_page_to_dpa(struct xe_vram_region *vr,
 				      struct page *page)
 {
 	u64 dpa;
-	struct xe_tile *tile = vr_to_tile(vr);
+	struct xe_tile *tile = vr->tile;
 	u64 pfn = page_to_pfn(page);
 	u64 offset;
 
@@ -370,7 +365,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 
 		if (!vr && spage) {
 			vr = page_to_vr(spage);
-			tile = vr_to_tile(vr);
+			tile = vr->tile;
 		}
 		XE_WARN_ON(spage && page_to_vr(spage) != vr);
 
@@ -508,7 +503,7 @@ static u64 block_offset_to_pfn(struct xe_vram_region *vr, u64 offset)
 
 static struct drm_buddy *tile_to_buddy(struct xe_tile *tile)
 {
-	return &tile->mem.vram.ttm.mm;
+	return &tile->mem.vram->ttm.mm;
 }
 
 static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocation,
@@ -522,7 +517,7 @@ static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocati
 
 	list_for_each_entry(block, blocks, link) {
 		struct xe_vram_region *vr = block->private;
-		struct xe_tile *tile = vr_to_tile(vr);
+		struct xe_tile *tile = vr->tile;
 		struct drm_buddy *buddy = tile_to_buddy(tile);
 		u64 block_pfn = block_offset_to_pfn(vr, drm_buddy_block_offset(block));
 		int i;
@@ -683,20 +678,15 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 start, u64 end, struct xe_vma *v
 }
 
 #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
-static struct xe_vram_region *tile_to_vr(struct xe_tile *tile)
-{
-	return &tile->mem.vram;
-}
-
 static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 				      unsigned long start, unsigned long end,
 				      struct mm_struct *mm,
 				      unsigned long timeslice_ms)
 {
-	struct xe_tile *tile = container_of(dpagemap, typeof(*tile), mem.vram.dpagemap);
+	struct xe_vram_region *vr = container_of(dpagemap, typeof(*vr), dpagemap);
+	struct xe_tile *tile = vr->tile;
 	struct xe_device *xe = tile_to_xe(tile);
 	struct device *dev = xe->drm.dev;
-	struct xe_vram_region *vr = tile_to_vr(tile);
 	struct drm_buddy_block *block;
 	struct list_head *blocks;
 	struct xe_bo *bo;
@@ -722,7 +712,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 
 	drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm,
 				&dpagemap_devmem_ops,
-				&tile->mem.vram.dpagemap,
+				&tile->mem.vram->dpagemap,
 				end - start);
 
 	blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks;
diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c
index 86e9811e60ba..858ce0183aaa 100644
--- a/drivers/gpu/drm/xe/xe_tile.c
+++ b/drivers/gpu/drm/xe/xe_tile.c
@@ -19,6 +19,7 @@
 #include "xe_tile_sysfs.h"
 #include "xe_ttm_vram_mgr.h"
 #include "xe_wa.h"
+#include "xe_vram.h"
 
 /**
  * DOC: Multi-tile Design
@@ -95,6 +96,33 @@ static int xe_tile_alloc(struct xe_tile *tile)
 	return 0;
 }
 
+/**
+ * xe_tile_alloc_vram - Perform per-tile VRAM structs allocation
+ * @tile: Tile to perform allocations for
+ *
+ * Allocates VRAM per-tile data structures using DRM-managed allocations.
+ * Does not touch the hardware.
+ *
+ * Returns -ENOMEM if allocations fail, otherwise 0.
+ */
+int xe_tile_alloc_vram(struct xe_tile *tile)
+{
+	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_vram_region *vram;
+
+	if (!IS_DGFX(xe))
+		return 0;
+
+	vram = drmm_kzalloc(&xe->drm, sizeof(*vram), GFP_KERNEL);
+	if (!vram)
+		return -ENOMEM;
+
+	vram->tile = tile;
+	tile->mem.vram = vram;
+
+	return 0;
+}
+
 /**
  * xe_tile_init_early - Initialize the tile and primary GT
  * @tile: Tile to initialize
@@ -132,8 +160,8 @@ static int tile_ttm_mgr_init(struct xe_tile *tile)
 	struct xe_device *xe = tile_to_xe(tile);
 	int err;
 
-	if (tile->mem.vram.usable_size) {
-		err = xe_ttm_vram_mgr_init(tile, &tile->mem.vram.ttm);
+	if (tile->mem.vram) {
+		err = xe_ttm_vram_mgr_init(tile, &tile->mem.vram->ttm);
 		if (err)
 			return err;
 		xe->info.mem_region_mask |= BIT(tile->id) << 1;
@@ -168,7 +196,7 @@ int xe_tile_init_noalloc(struct xe_tile *tile)
 	xe_wa_apply_tile_workarounds(tile);
 
 	if (xe->info.has_usm && IS_DGFX(xe))
-		xe_devm_add(tile, &tile->mem.vram);
+		xe_devm_add(tile, tile->mem.vram);
 
 	return xe_tile_sysfs_init(tile);
 }
diff --git a/drivers/gpu/drm/xe/xe_tile.h b/drivers/gpu/drm/xe/xe_tile.h
index cc33e8733983..440bc9e11c8b 100644
--- a/drivers/gpu/drm/xe/xe_tile.h
+++ b/drivers/gpu/drm/xe/xe_tile.h
@@ -14,12 +14,14 @@ int xe_tile_init_early(struct xe_tile *tile, struct xe_device *xe, u8 id);
 int xe_tile_init_noalloc(struct xe_tile *tile);
 int xe_tile_init(struct xe_tile *tile);
 
+int xe_tile_alloc_vram(struct xe_tile *tile);
+
 void xe_tile_migrate_wait(struct xe_tile *tile);
 
 #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
 static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile)
 {
-	return &tile->mem.vram.dpagemap;
+	return &tile->mem.vram->dpagemap;
 }
 #else
 static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile)
diff --git a/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c b/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c
index d9c9d2547aad..9a9733447230 100644
--- a/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c
+++ b/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c
@@ -25,6 +25,7 @@
 #include "xe_ttm_stolen_mgr.h"
 #include "xe_ttm_vram_mgr.h"
 #include "xe_wa.h"
+#include "xe_vram.h"
 
 struct xe_ttm_stolen_mgr {
 	struct xe_ttm_vram_mgr base;
@@ -82,15 +83,16 @@ static u32 get_wopcm_size(struct xe_device *xe)
 
 static s64 detect_bar2_dgfx(struct xe_device *xe, struct xe_ttm_stolen_mgr *mgr)
 {
-	struct xe_tile *tile = xe_device_get_root_tile(xe);
+	struct xe_vram_region *tile_vram = xe_device_get_root_tile(xe)->mem.vram;
+	resource_size_t tile_io_start = xe_vram_region_io_start(tile_vram);
 	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 	u64 stolen_size, wopcm_size;
 	u64 tile_offset;
 	u64 tile_size;
 
-	tile_offset = tile->mem.vram.io_start - xe->mem.vram.io_start;
-	tile_size = tile->mem.vram.actual_physical_size;
+	tile_offset = tile_io_start - xe_vram_region_io_start(xe->mem.vram);
+	tile_size = xe_vram_region_actual_physical_size(tile_vram);
 
 	/* Use DSM base address instead for stolen memory */
 	mgr->stolen_base = (xe_mmio_read64_2x32(mmio, DSMBASE) & BDSM_MASK) - tile_offset;
@@ -107,7 +109,7 @@ static s64 detect_bar2_dgfx(struct xe_device *xe, struct xe_ttm_stolen_mgr *mgr)
 
 	/* Verify usage fits in the actual resource available */
 	if (mgr->stolen_base + stolen_size <= pci_resource_len(pdev, LMEM_BAR))
-		mgr->io_base = tile->mem.vram.io_start + mgr->stolen_base;
+		mgr->io_base = tile_io_start + mgr->stolen_base;
 
 	/*
 	 * There may be few KB of platform dependent reserved memory at the end
diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
index 9e375a40aee9..3de2df47959b 100644
--- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
+++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
@@ -340,10 +340,11 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr,
 int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr)
 {
 	struct xe_device *xe = tile_to_xe(tile);
-	struct xe_vram_region *vram = &tile->mem.vram;
+	struct xe_vram_region *vram = tile->mem.vram;
 
 	return __xe_ttm_vram_mgr_init(xe, mgr, XE_PL_VRAM0 + tile->id,
-				      vram->usable_size, vram->io_size,
+				      xe_vram_region_usable_size(vram),
+				      xe_vram_region_io_size(vram),
 				      PAGE_SIZE);
 }
 
@@ -392,7 +393,7 @@ int xe_ttm_vram_mgr_alloc_sgt(struct xe_device *xe,
 	 */
 	xe_res_first(res, offset, length, &cursor);
 	for_each_sgtable_sg((*sgt), sg, i) {
-		phys_addr_t phys = cursor.start + tile->mem.vram.io_start;
+		phys_addr_t phys = cursor.start + xe_vram_region_io_start(tile->mem.vram);
 		size_t size = min_t(u64, cursor.size, SZ_2G);
 		dma_addr_t addr;
 
diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c
index 3a4c84e9efc6..c93e9b96354b 100644
--- a/drivers/gpu/drm/xe/xe_vram.c
+++ b/drivers/gpu/drm/xe/xe_vram.c
@@ -3,6 +3,7 @@
  * Copyright © 2021-2024 Intel Corporation
  */
 
+#include <kunit/visibility.h>
 #include <linux/pci.h>
 
 #include <drm/drm_managed.h>
@@ -147,17 +148,17 @@ static int determine_lmem_bar_size(struct xe_device *xe)
 
 	resize_vram_bar(xe);
 
-	xe->mem.vram.io_start = pci_resource_start(pdev, LMEM_BAR);
-	xe->mem.vram.io_size = pci_resource_len(pdev, LMEM_BAR);
-	if (!xe->mem.vram.io_size)
+	xe->mem.vram->io_start = pci_resource_start(pdev, LMEM_BAR);
+	xe->mem.vram->io_size = pci_resource_len(pdev, LMEM_BAR);
+	if (!xe->mem.vram->io_size)
 		return -EIO;
 
 	/* XXX: Need to change when xe link code is ready */
-	xe->mem.vram.dpa_base = 0;
+	xe->mem.vram->dpa_base = 0;
 
 	/* set up a map to the total memory area. */
-	xe->mem.vram.mapping = devm_ioremap_wc(&pdev->dev, xe->mem.vram.io_start,
-					       xe->mem.vram.io_size);
+	xe->mem.vram->mapping = devm_ioremap_wc(&pdev->dev, xe->mem.vram->io_start,
+						xe->mem.vram->io_size);
 
 	return 0;
 }
@@ -279,10 +280,10 @@ static void vram_fini(void *arg)
 	struct xe_tile *tile;
 	int id;
 
-	xe->mem.vram.mapping = NULL;
+	xe->mem.vram->mapping = NULL;
 
 	for_each_tile(tile, xe, id)
-		tile->mem.vram.mapping = NULL;
+		tile->mem.vram->mapping = NULL;
 }
 
 /**
@@ -318,10 +319,10 @@ int xe_vram_probe(struct xe_device *xe)
 	if (err)
 		return err;
 
-	drm_info(&xe->drm, "VISIBLE VRAM: %pa, %pa\n", &xe->mem.vram.io_start,
-		 &xe->mem.vram.io_size);
+	drm_info(&xe->drm, "VISIBLE VRAM: %pa, %pa\n", &xe->mem.vram->io_start,
+		 &xe->mem.vram->io_size);
 
-	io_size = xe->mem.vram.io_size;
+	io_size = xe->mem.vram->io_size;
 
 	/* tile specific ranges */
 	for_each_tile(tile, xe, id) {
@@ -329,45 +330,105 @@ int xe_vram_probe(struct xe_device *xe)
 		if (err)
 			return err;
 
-		tile->mem.vram.actual_physical_size = tile_size;
-		tile->mem.vram.io_start = xe->mem.vram.io_start + tile_offset;
-		tile->mem.vram.io_size = min_t(u64, vram_size, io_size);
+		tile->mem.vram->actual_physical_size = tile_size;
+		tile->mem.vram->io_start = xe->mem.vram->io_start + tile_offset;
+		tile->mem.vram->io_size = min_t(u64, vram_size, io_size);
 
-		if (!tile->mem.vram.io_size) {
+		if (!tile->mem.vram->io_size) {
 			drm_err(&xe->drm, "Tile without any CPU visible VRAM. Aborting.\n");
 			return -ENODEV;
 		}
 
-		tile->mem.vram.dpa_base = xe->mem.vram.dpa_base + tile_offset;
-		tile->mem.vram.usable_size = vram_size;
-		tile->mem.vram.mapping = xe->mem.vram.mapping + tile_offset;
+		tile->mem.vram->dpa_base = xe->mem.vram->dpa_base + tile_offset;
+		tile->mem.vram->usable_size = vram_size;
+		tile->mem.vram->mapping = xe->mem.vram->mapping + tile_offset;
 
-		if (tile->mem.vram.io_size < tile->mem.vram.usable_size)
+		if (tile->mem.vram->io_size < tile->mem.vram->usable_size)
 			drm_info(&xe->drm, "Small BAR device\n");
-		drm_info(&xe->drm, "VRAM[%u, %u]: Actual physical size %pa, usable size exclude stolen %pa, CPU accessible size %pa\n", id,
-			 tile->id, &tile->mem.vram.actual_physical_size, &tile->mem.vram.usable_size, &tile->mem.vram.io_size);
-		drm_info(&xe->drm, "VRAM[%u, %u]: DPA range: [%pa-%llx], io range: [%pa-%llx]\n", id, tile->id,
-			 &tile->mem.vram.dpa_base, tile->mem.vram.dpa_base + (u64)tile->mem.vram.actual_physical_size,
-			 &tile->mem.vram.io_start, tile->mem.vram.io_start + (u64)tile->mem.vram.io_size);
+		drm_info(&xe->drm,
+			 "VRAM[%u, %u]: Actual physical size %pa, usable size exclude stolen %pa, CPU accessible size %pa\n",
+			 id, tile->id, &tile->mem.vram->actual_physical_size,
+			 &tile->mem.vram->usable_size, &tile->mem.vram->io_size);
+		drm_info(&xe->drm, "VRAM[%u, %u]: DPA range: [%pa-%llx], io range: [%pa-%llx]\n",
+			 id, tile->id, &tile->mem.vram->dpa_base,
+			 tile->mem.vram->dpa_base + (u64)tile->mem.vram->actual_physical_size,
+			 &tile->mem.vram->io_start,
+			 tile->mem.vram->io_start + (u64)tile->mem.vram->io_size);
 
 		/* calculate total size using tile size to get the correct HW sizing */
 		total_size += tile_size;
 		available_size += vram_size;
 
-		if (total_size > xe->mem.vram.io_size) {
+		if (total_size > xe->mem.vram->io_size) {
 			drm_info(&xe->drm, "VRAM: %pa is larger than resource %pa\n",
-				 &total_size, &xe->mem.vram.io_size);
+				 &total_size, &xe->mem.vram->io_size);
 		}
 
 		io_size -= min_t(u64, tile_size, io_size);
 	}
 
-	xe->mem.vram.actual_physical_size = total_size;
+	xe->mem.vram->actual_physical_size = total_size;
 
-	drm_info(&xe->drm, "Total VRAM: %pa, %pa\n", &xe->mem.vram.io_start,
-		 &xe->mem.vram.actual_physical_size);
-	drm_info(&xe->drm, "Available VRAM: %pa, %pa\n", &xe->mem.vram.io_start,
+	drm_info(&xe->drm, "Total VRAM: %pa, %pa\n", &xe->mem.vram->io_start,
+		 &xe->mem.vram->actual_physical_size);
+	drm_info(&xe->drm, "Available VRAM: %pa, %pa\n", &xe->mem.vram->io_start,
 		 &available_size);
 
 	return devm_add_action_or_reset(xe->drm.dev, vram_fini, xe);
 }
+
+/**
+ * xe_vram_region_io_start - Get the IO start of a VRAM region
+ * @vram: the VRAM region
+ *
+ * Return: the IO start of the VRAM region, or 0 if not valid
+ */
+resource_size_t xe_vram_region_io_start(const struct xe_vram_region *vram)
+{
+	return vram ? vram->io_start : 0;
+}
+
+/**
+ * xe_vram_region_io_size - Get the IO size of a VRAM region
+ * @vram: the VRAM region
+ *
+ * Return: the IO size of the VRAM region, or 0 if not valid
+ */
+resource_size_t xe_vram_region_io_size(const struct xe_vram_region *vram)
+{
+	return vram ? vram->io_size : 0;
+}
+
+/**
+ * xe_vram_region_dpa_base - Get the DPA base of a VRAM region
+ * @vram: the VRAM region
+ *
+ * Return: the DPA base of the VRAM region, or 0 if not valid
+ */
+resource_size_t xe_vram_region_dpa_base(const struct xe_vram_region *vram)
+{
+	return vram ? vram->dpa_base : 0;
+}
+
+/**
+ * xe_vram_region_usable_size - Get the usable size of a VRAM region
+ * @vram: the VRAM region
+ *
+ * Return: the usable size of the VRAM region, or 0 if not valid
+ */
+resource_size_t xe_vram_region_usable_size(const struct xe_vram_region *vram)
+{
+	return vram ? vram->usable_size : 0;
+}
+
+/**
+ * xe_vram_region_actual_physical_size - Get the actual physical size of a VRAM region
+ * @vram: the VRAM region
+ *
+ * Return: the actual physical size of the VRAM region, or 0 if not valid
+ */
+resource_size_t xe_vram_region_actual_physical_size(const struct xe_vram_region *vram)
+{
+	return vram ? vram->actual_physical_size : 0;
+}
+EXPORT_SYMBOL_IF_KUNIT(xe_vram_region_actual_physical_size);
diff --git a/drivers/gpu/drm/xe/xe_vram.h b/drivers/gpu/drm/xe/xe_vram.h
index e31cc04ec0db..d4bf1f9c2a72 100644
--- a/drivers/gpu/drm/xe/xe_vram.h
+++ b/drivers/gpu/drm/xe/xe_vram.h
@@ -6,8 +6,17 @@
 #ifndef _XE_VRAM_H_
 #define _XE_VRAM_H_
 
+#include <linux/types.h>
+
 struct xe_device;
+struct xe_vram_region;
 
 int xe_vram_probe(struct xe_device *xe);
 
+resource_size_t xe_vram_region_io_start(const struct xe_vram_region *vram);
+resource_size_t xe_vram_region_io_size(const struct xe_vram_region *vram);
+resource_size_t xe_vram_region_dpa_base(const struct xe_vram_region *vram);
+resource_size_t xe_vram_region_usable_size(const struct xe_vram_region *vram);
+resource_size_t xe_vram_region_actual_physical_size(const struct xe_vram_region *vram);
+
 #endif

From 7a20b4f558f4291161f71a5b7384262db9ccd6b0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 14 Jul 2025 20:48:16 +0200
Subject: [PATCH 0014/1455] drm/xe: Move struct xe_vram_region to a dedicated
 header
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Let's move the xe_vram_region structure to a new header dedicated to VRAM
to improve modularity and avoid unnecessary dependencies when only
VRAM-related structures are needed.

v2: Fix build if CONFIG_DRM_XE_DEVMEM_MIRROR is enabled
v3: Fix build if CONFIG_DRM_XE_DISPLAY is enabled
v4: Move helper to get tile dpagemap to xe_svm.c

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Suggested-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com> # rev3
Acked-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250714184818.89201-4-piotr.piorkowski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/display/xe_fb_pin.c        |  1 +
 drivers/gpu/drm/xe/display/xe_plane_initial.c |  1 +
 drivers/gpu/drm/xe/xe_bo.c                    |  1 +
 drivers/gpu/drm/xe/xe_bo_types.h              |  1 +
 drivers/gpu/drm/xe/xe_device.c                |  1 +
 drivers/gpu/drm/xe/xe_device_types.h          | 60 +--------------
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c    |  1 +
 drivers/gpu/drm/xe/xe_svm.c                   |  8 +-
 drivers/gpu/drm/xe/xe_tile.c                  |  1 +
 drivers/gpu/drm/xe/xe_tile.h                  | 12 ---
 drivers/gpu/drm/xe/xe_ttm_vram_mgr.c          |  1 +
 drivers/gpu/drm/xe/xe_vram.c                  |  1 +
 drivers/gpu/drm/xe/xe_vram_types.h            | 74 +++++++++++++++++++
 13 files changed, 91 insertions(+), 72 deletions(-)
 create mode 100644 drivers/gpu/drm/xe/xe_vram_types.h

diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index a1b203f40390..3f0362453e73 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -16,6 +16,7 @@
 #include "xe_device.h"
 #include "xe_ggtt.h"
 #include "xe_pm.h"
+#include "xe_vram_types.h"
 
 static void
 write_dpt_rotated(struct xe_bo *bo, struct iosys_map *map, u32 *dpt_ofs, u32 bo_ofs,
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index 30a341c96e76..54ff70ea2d25 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -21,6 +21,7 @@
 #include "intel_frontbuffer.h"
 #include "intel_plane_initial.h"
 #include "xe_bo.h"
+#include "xe_vram_types.h"
 #include "xe_wa.h"
 
 #include <generated/xe_wa_oob.h>
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 18f27da47a36..00ce067d5fd3 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -36,6 +36,7 @@
 #include "xe_trace_bo.h"
 #include "xe_ttm_stolen_mgr.h"
 #include "xe_vm.h"
+#include "xe_vram_types.h"
 
 const char *const xe_mem_type_to_name[TTM_NUM_MEM_TYPES]  = {
 	[XE_PL_SYSTEM] = "system",
diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index ff560d82496f..57d34698139e 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -9,6 +9,7 @@
 #include <linux/iosys-map.h>
 
 #include <drm/drm_gpusvm.h>
+#include <drm/drm_pagemap.h>
 #include <drm/ttm/ttm_bo.h>
 #include <drm/ttm/ttm_device.h>
 #include <drm/ttm/ttm_placement.h>
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index df6f1790d6a6..08f45777f797 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -64,6 +64,7 @@
 #include "xe_ttm_sys_mgr.h"
 #include "xe_vm.h"
 #include "xe_vram.h"
+#include "xe_vram_types.h"
 #include "xe_vsec.h"
 #include "xe_wait_user_fence.h"
 #include "xe_wa.h"
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 6299f4571c4c..eb6105523f23 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -10,7 +10,6 @@
 
 #include <drm/drm_device.h>
 #include <drm/drm_file.h>
-#include <drm/drm_pagemap.h>
 #include <drm/ttm/ttm_device.h>
 
 #include "xe_devcoredump_types.h"
@@ -26,7 +25,6 @@
 #include "xe_sriov_vf_types.h"
 #include "xe_step_types.h"
 #include "xe_survivability_mode_types.h"
-#include "xe_ttm_vram_mgr_types.h"
 
 #if IS_ENABLED(CONFIG_DRM_XE_DEBUG)
 #define TEST_VM_OPS_ERROR
@@ -39,6 +37,7 @@ struct xe_ggtt;
 struct xe_i2c;
 struct xe_pat_ops;
 struct xe_pxp;
+struct xe_vram_region;
 
 #define XE_BO_INVALID_OFFSET	LONG_MAX
 
@@ -71,63 +70,6 @@ struct xe_pxp;
 		 const struct xe_tile * : (const struct xe_device *)((tile__)->xe),	\
 		 struct xe_tile * : (tile__)->xe)
 
-/**
- * struct xe_vram_region - memory region structure
- * This is used to describe a memory region in xe
- * device, such as HBM memory or CXL extension memory.
- */
-struct xe_vram_region {
-	/** @tile: Back pointer to tile */
-	struct xe_tile *tile;
-	/** @io_start: IO start address of this VRAM instance */
-	resource_size_t io_start;
-	/**
-	 * @io_size: IO size of this VRAM instance
-	 *
-	 * This represents how much of this VRAM we can access
-	 * via the CPU through the VRAM BAR. This can be smaller
-	 * than @usable_size, in which case only part of VRAM is CPU
-	 * accessible (typically the first 256M). This
-	 * configuration is known as small-bar.
-	 */
-	resource_size_t io_size;
-	/** @dpa_base: This memory regions's DPA (device physical address) base */
-	resource_size_t dpa_base;
-	/**
-	 * @usable_size: usable size of VRAM
-	 *
-	 * Usable size of VRAM excluding reserved portions
-	 * (e.g stolen mem)
-	 */
-	resource_size_t usable_size;
-	/**
-	 * @actual_physical_size: Actual VRAM size
-	 *
-	 * Actual VRAM size including reserved portions
-	 * (e.g stolen mem)
-	 */
-	resource_size_t actual_physical_size;
-	/** @mapping: pointer to VRAM mappable space */
-	void __iomem *mapping;
-	/** @ttm: VRAM TTM manager */
-	struct xe_ttm_vram_mgr ttm;
-#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
-	/** @pagemap: Used to remap device memory as ZONE_DEVICE */
-	struct dev_pagemap pagemap;
-	/**
-	 * @dpagemap: The struct drm_pagemap of the ZONE_DEVICE memory
-	 * pages of this tile.
-	 */
-	struct drm_pagemap dpagemap;
-	/**
-	 * @hpa_base: base host physical address
-	 *
-	 * This is generated when remap device memory as ZONE_DEVICE
-	 */
-	resource_size_t hpa_base;
-#endif
-};
-
 /**
  * struct xe_mmio - register mmio structure
  *
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index 86889df04011..f00cce81f1ff 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -33,6 +33,7 @@
 #include "xe_migrate.h"
 #include "xe_sriov.h"
 #include "xe_ttm_vram_mgr.h"
+#include "xe_vram_types.h"
 #include "xe_wopcm.h"
 
 #define make_u64_from_u32(hi, lo) ((u64)((u64)(u32)(hi) << 32 | (u32)(lo)))
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index e6871734ffa9..901f9a0268e6 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -17,6 +17,7 @@
 #include "xe_ttm_vram_mgr.h"
 #include "xe_vm.h"
 #include "xe_vm_types.h"
+#include "xe_vram_types.h"
 
 static bool xe_svm_range_in_vram(struct xe_svm_range *range)
 {
@@ -989,6 +990,11 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range,
 
 #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
 
+static struct drm_pagemap *tile_local_pagemap(struct xe_tile *tile)
+{
+	return &tile->mem.vram->dpagemap;
+}
+
 /**
  * xe_svm_alloc_vram()- Allocate device memory pages for range,
  * migrating existing data.
@@ -1006,7 +1012,7 @@ int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range,
 	xe_assert(tile_to_xe(tile), range->base.flags.migrate_devmem);
 	range_debug(range, "ALLOCATE VRAM");
 
-	dpagemap = xe_tile_local_pagemap(tile);
+	dpagemap = tile_local_pagemap(tile);
 	return drm_pagemap_populate_mm(dpagemap, xe_svm_range_start(range),
 				       xe_svm_range_end(range),
 				       range->base.gpusvm->mm,
diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c
index 858ce0183aaa..bd2ff91a7d1c 100644
--- a/drivers/gpu/drm/xe/xe_tile.c
+++ b/drivers/gpu/drm/xe/xe_tile.c
@@ -20,6 +20,7 @@
 #include "xe_ttm_vram_mgr.h"
 #include "xe_wa.h"
 #include "xe_vram.h"
+#include "xe_vram_types.h"
 
 /**
  * DOC: Multi-tile Design
diff --git a/drivers/gpu/drm/xe/xe_tile.h b/drivers/gpu/drm/xe/xe_tile.h
index 440bc9e11c8b..dceb6297aa01 100644
--- a/drivers/gpu/drm/xe/xe_tile.h
+++ b/drivers/gpu/drm/xe/xe_tile.h
@@ -18,18 +18,6 @@ int xe_tile_alloc_vram(struct xe_tile *tile);
 
 void xe_tile_migrate_wait(struct xe_tile *tile);
 
-#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
-static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile)
-{
-	return &tile->mem.vram->dpagemap;
-}
-#else
-static inline struct drm_pagemap *xe_tile_local_pagemap(struct xe_tile *tile)
-{
-	return NULL;
-}
-#endif
-
 static inline bool xe_tile_is_root(struct xe_tile *tile)
 {
 	return tile->id == 0;
diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
index 3de2df47959b..8f9b8a1d2c05 100644
--- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
+++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
@@ -15,6 +15,7 @@
 #include "xe_gt.h"
 #include "xe_res_cursor.h"
 #include "xe_ttm_vram_mgr.h"
+#include "xe_vram_types.h"
 
 static inline struct drm_buddy_block *
 xe_ttm_vram_mgr_first_block(struct list_head *list)
diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c
index c93e9b96354b..366e5d8a85ca 100644
--- a/drivers/gpu/drm/xe/xe_vram.c
+++ b/drivers/gpu/drm/xe/xe_vram.c
@@ -21,6 +21,7 @@
 #include "xe_module.h"
 #include "xe_sriov.h"
 #include "xe_vram.h"
+#include "xe_vram_types.h"
 
 #define BAR_SIZE_SHIFT 20
 
diff --git a/drivers/gpu/drm/xe/xe_vram_types.h b/drivers/gpu/drm/xe/xe_vram_types.h
new file mode 100644
index 000000000000..a01838236036
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_vram_types.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_VRAM_TYPES_H_
+#define _XE_VRAM_TYPES_H_
+
+#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
+#include <drm/drm_pagemap.h>
+#endif
+
+#include "xe_ttm_vram_mgr_types.h"
+
+struct xe_tile;
+
+/**
+ * struct xe_vram_region - memory region structure
+ * This is used to describe a memory region in xe
+ * device, such as HBM memory or CXL extension memory.
+ */
+struct xe_vram_region {
+	/** @tile: Back pointer to tile */
+	struct xe_tile *tile;
+	/** @io_start: IO start address of this VRAM instance */
+	resource_size_t io_start;
+	/**
+	 * @io_size: IO size of this VRAM instance
+	 *
+	 * This represents how much of this VRAM we can access
+	 * via the CPU through the VRAM BAR. This can be smaller
+	 * than @usable_size, in which case only part of VRAM is CPU
+	 * accessible (typically the first 256M). This
+	 * configuration is known as small-bar.
+	 */
+	resource_size_t io_size;
+	/** @dpa_base: This memory regions's DPA (device physical address) base */
+	resource_size_t dpa_base;
+	/**
+	 * @usable_size: usable size of VRAM
+	 *
+	 * Usable size of VRAM excluding reserved portions
+	 * (e.g stolen mem)
+	 */
+	resource_size_t usable_size;
+	/**
+	 * @actual_physical_size: Actual VRAM size
+	 *
+	 * Actual VRAM size including reserved portions
+	 * (e.g stolen mem)
+	 */
+	resource_size_t actual_physical_size;
+	/** @mapping: pointer to VRAM mappable space */
+	void __iomem *mapping;
+	/** @ttm: VRAM TTM manager */
+	struct xe_ttm_vram_mgr ttm;
+#if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
+	/** @pagemap: Used to remap device memory as ZONE_DEVICE */
+	struct dev_pagemap pagemap;
+	/**
+	 * @dpagemap: The struct drm_pagemap of the ZONE_DEVICE memory
+	 * pages of this tile.
+	 */
+	struct drm_pagemap dpagemap;
+	/**
+	 * @hpa_base: base host physical address
+	 *
+	 * This is generated when remap device memory as ZONE_DEVICE
+	 */
+	resource_size_t hpa_base;
+#endif
+};
+
+#endif

From d65ff1ec85355959e4ca452fba458687e4da583a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 14 Jul 2025 20:48:17 +0200
Subject: [PATCH 0015/1455] drm/xe: Split xe_migrate allocation from
 initialization
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, xe_migrate_init handled both allocation and initialization,
Lets moves allocation to xe_tile_alloc via a new xe_migrate_alloc
function, and keep initialization in xe_migrate_init.
This will allow the migration pointers to be passed to other structures
before full initialization.
Also replaces devm_kzalloc with drmm_kzalloc for better
DRM-managed memory.

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250714184818.89201-5-piotr.piorkowski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_migrate.c | 25 ++++++++++++++++++-------
 drivers/gpu/drm/xe/xe_migrate.h |  1 +
 drivers/gpu/drm/xe/xe_tile.c    |  4 ++++
 3 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index d6186ac530cb..6a80ae6104dd 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -389,6 +389,23 @@ static bool xe_migrate_needs_ccs_emit(struct xe_device *xe)
 	return xe_device_has_flat_ccs(xe) && !(GRAPHICS_VER(xe) >= 20 && IS_DGFX(xe));
 }
 
+/**
+ * xe_migrate_alloc - Allocate a migrate struct for a given &xe_tile
+ * @tile: &xe_tile
+ *
+ * Allocates a &xe_migrate for a given tile.
+ *
+ * Return: &xe_migrate on success, or NULL when out of memory.
+ */
+struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile)
+{
+	struct xe_migrate *m = drmm_kzalloc(&tile_to_xe(tile)->drm, sizeof(*m), GFP_KERNEL);
+
+	if (m)
+		m->tile = tile;
+	return m;
+}
+
 /**
  * xe_migrate_init() - Initialize a migrate context
  * @tile: Back-pointer to the tile we're initializing for.
@@ -399,16 +416,10 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 {
 	struct xe_device *xe = tile_to_xe(tile);
 	struct xe_gt *primary_gt = tile->primary_gt;
-	struct xe_migrate *m;
+	struct xe_migrate *m = tile->migrate;
 	struct xe_vm *vm;
 	int err;
 
-	m = devm_kzalloc(xe->drm.dev, sizeof(*m), GFP_KERNEL);
-	if (!m)
-		return ERR_PTR(-ENOMEM);
-
-	m->tile = tile;
-
 	/* Special layout, prepared below.. */
 	vm = xe_vm_create(xe, XE_VM_FLAG_MIGRATION |
 			  XE_VM_FLAG_SET_TILE_ID(tile));
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index fb9839c1bae0..74c60f55004a 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -93,6 +93,7 @@ struct xe_migrate_pt_update {
 	u8 tile_id;
 };
 
+struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile);
 struct xe_migrate *xe_migrate_init(struct xe_tile *tile);
 
 struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c
index bd2ff91a7d1c..171de6a4ba95 100644
--- a/drivers/gpu/drm/xe/xe_tile.c
+++ b/drivers/gpu/drm/xe/xe_tile.c
@@ -94,6 +94,10 @@ static int xe_tile_alloc(struct xe_tile *tile)
 	if (!tile->mem.ggtt)
 		return -ENOMEM;
 
+	tile->migrate = xe_migrate_alloc(tile);
+	if (!tile->migrate)
+		return -ENOMEM;
+
 	return 0;
 }
 

From 4b0a5f5ce7849aab7a67ba9f113ed75626f6de36 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 14 Jul 2025 20:48:18 +0200
Subject: [PATCH 0016/1455] drm/xe: Unify the initialization of VRAM regions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently in the drivers we have defined VRAM regions per device and per
tile. Initialization of these regions is done in two completely different
ways. To simplify the logic of the code and make it easier to add new
regions in the future, let's unify the way we initialize VRAM regions.

v2:
- fix doc comments in struct xe_vram_region
- remove unnecessary includes (Jani)
v3:
- move code from xe_vram_init_regions_managers to xe_tile_init_noalloc
  (Matthew)
- replace ioremap_wc to devm_ioremap_wc for mapping VRAM BAR
  (Matthew)
- Replace the tile id parameter with vram region in the xe_pf_begin
  function.
v4:
- remove tile back pointer from struct xe_vram_region
- add new back pointers: xe and migarte to xe_vram_region

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com> # rev3
Acked-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250714184818.89201-6-piotr.piorkowski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_bo.h           |   4 +-
 drivers/gpu/drm/xe/xe_gt_pagefault.c |  13 ++-
 drivers/gpu/drm/xe/xe_query.c        |   3 +-
 drivers/gpu/drm/xe/xe_svm.c          |  43 ++++----
 drivers/gpu/drm/xe/xe_tile.c         |  37 +++----
 drivers/gpu/drm/xe/xe_ttm_vram_mgr.c |  16 ++-
 drivers/gpu/drm/xe/xe_ttm_vram_mgr.h |   3 +-
 drivers/gpu/drm/xe/xe_vram.c         | 151 ++++++++++++++++-----------
 drivers/gpu/drm/xe/xe_vram.h         |   2 +
 drivers/gpu/drm/xe/xe_vram_types.h   |  17 ++-
 10 files changed, 164 insertions(+), 125 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 02e8cde4c6b2..8cce413b5235 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -12,6 +12,7 @@
 #include "xe_macros.h"
 #include "xe_vm_types.h"
 #include "xe_vm.h"
+#include "xe_vram_types.h"
 
 #define XE_DEFAULT_GTT_SIZE_MB          3072ULL /* 3GB by default */
 
@@ -23,8 +24,9 @@
 #define XE_BO_FLAG_VRAM_MASK		(XE_BO_FLAG_VRAM0 | XE_BO_FLAG_VRAM1)
 /* -- */
 #define XE_BO_FLAG_STOLEN		BIT(4)
+#define XE_BO_FLAG_VRAM(vram)		(XE_BO_FLAG_VRAM0 << ((vram)->id))
 #define XE_BO_FLAG_VRAM_IF_DGFX(tile)	(IS_DGFX(tile_to_xe(tile)) ? \
-					 XE_BO_FLAG_VRAM0 << (tile)->id : \
+					 XE_BO_FLAG_VRAM((tile)->mem.vram) : \
 					 XE_BO_FLAG_SYSTEM)
 #define XE_BO_FLAG_GGTT			BIT(5)
 #define XE_BO_FLAG_IGNORE_MIN_PAGE_SIZE BIT(6)
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index 5a75d56d8558..ab43dec52776 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -23,6 +23,7 @@
 #include "xe_svm.h"
 #include "xe_trace_bo.h"
 #include "xe_vm.h"
+#include "xe_vram_types.h"
 
 struct pagefault {
 	u64 page_addr;
@@ -74,7 +75,7 @@ static bool vma_is_valid(struct xe_tile *tile, struct xe_vma *vma)
 }
 
 static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma,
-		       bool atomic, unsigned int id)
+		       bool atomic, struct xe_vram_region *vram)
 {
 	struct xe_bo *bo = xe_vma_bo(vma);
 	struct xe_vm *vm = xe_vma_vm(vma);
@@ -84,14 +85,16 @@ static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma,
 	if (err)
 		return err;
 
-	if (atomic && IS_DGFX(vm->xe)) {
+	if (atomic && vram) {
+		xe_assert(vm->xe, IS_DGFX(vm->xe));
+
 		if (xe_vma_is_userptr(vma)) {
 			err = -EACCES;
 			return err;
 		}
 
 		/* Migrate to VRAM, move should invalidate the VMA first */
-		err = xe_bo_migrate(bo, XE_PL_VRAM0 + id);
+		err = xe_bo_migrate(bo, vram->placement);
 		if (err)
 			return err;
 	} else if (bo) {
@@ -138,7 +141,7 @@ retry_userptr:
 	/* Lock VM and BOs dma-resv */
 	drm_exec_init(&exec, 0, 0);
 	drm_exec_until_all_locked(&exec) {
-		err = xe_pf_begin(&exec, vma, atomic, tile->id);
+		err = xe_pf_begin(&exec, vma, atomic, tile->mem.vram);
 		drm_exec_retry_on_contention(&exec);
 		if (xe_vm_validate_should_retry(&exec, err, &end))
 			err = -EAGAIN;
@@ -573,7 +576,7 @@ static int handle_acc(struct xe_gt *gt, struct acc *acc)
 	/* Lock VM and BOs dma-resv */
 	drm_exec_init(&exec, 0, 0);
 	drm_exec_until_all_locked(&exec) {
-		ret = xe_pf_begin(&exec, vma, true, tile->id);
+		ret = xe_pf_begin(&exec, vma, true, tile->mem.vram);
 		drm_exec_retry_on_contention(&exec);
 		if (ret)
 			break;
diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
index 66ea3403afbe..56ef3ee2806f 100644
--- a/drivers/gpu/drm/xe/xe_query.c
+++ b/drivers/gpu/drm/xe/xe_query.c
@@ -27,6 +27,7 @@
 #include "xe_oa.h"
 #include "xe_pxp.h"
 #include "xe_ttm_vram_mgr.h"
+#include "xe_vram_types.h"
 #include "xe_wa.h"
 
 static const u16 xe_to_user_engine_class[] = {
@@ -410,7 +411,7 @@ static int query_gt_list(struct xe_device *xe, struct drm_xe_device_query *query
 			gt_list->gt_list[iter].near_mem_regions = 0x1;
 		else
 			gt_list->gt_list[iter].near_mem_regions =
-				BIT(gt_to_tile(gt)->id) << 1;
+				BIT(gt_to_tile(gt)->mem.vram->id) << 1;
 		gt_list->gt_list[iter].far_mem_regions = xe->info.mem_region_mask ^
 			gt_list->gt_list[iter].near_mem_regions;
 
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 901f9a0268e6..10c8a1bcb86e 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -311,12 +311,11 @@ static u64 xe_vram_region_page_to_dpa(struct xe_vram_region *vr,
 				      struct page *page)
 {
 	u64 dpa;
-	struct xe_tile *tile = vr->tile;
 	u64 pfn = page_to_pfn(page);
 	u64 offset;
 
-	xe_tile_assert(tile, is_device_private_page(page));
-	xe_tile_assert(tile, (pfn << PAGE_SHIFT) >= vr->hpa_base);
+	xe_assert(vr->xe, is_device_private_page(page));
+	xe_assert(vr->xe, (pfn << PAGE_SHIFT) >= vr->hpa_base);
 
 	offset = (pfn << PAGE_SHIFT) - vr->hpa_base;
 	dpa = vr->dpa_base + offset;
@@ -333,7 +332,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 		       unsigned long npages, const enum xe_svm_copy_dir dir)
 {
 	struct xe_vram_region *vr = NULL;
-	struct xe_tile *tile;
+	struct xe_device *xe;
 	struct dma_fence *fence = NULL;
 	unsigned long i;
 #define XE_VRAM_ADDR_INVALID	~0x0ull
@@ -366,7 +365,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 
 		if (!vr && spage) {
 			vr = page_to_vr(spage);
-			tile = vr->tile;
+			xe = vr->xe;
 		}
 		XE_WARN_ON(spage && page_to_vr(spage) != vr);
 
@@ -398,18 +397,18 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 
 			if (vram_addr != XE_VRAM_ADDR_INVALID) {
 				if (sram) {
-					vm_dbg(&tile->xe->drm,
+					vm_dbg(&xe->drm,
 					       "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld",
 					       vram_addr, (u64)dma_addr[pos], i - pos + incr);
-					__fence = xe_migrate_from_vram(tile->migrate,
+					__fence = xe_migrate_from_vram(vr->migrate,
 								       i - pos + incr,
 								       vram_addr,
 								       dma_addr + pos);
 				} else {
-					vm_dbg(&tile->xe->drm,
+					vm_dbg(&xe->drm,
 					       "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld",
 					       (u64)dma_addr[pos], vram_addr, i - pos + incr);
-					__fence = xe_migrate_to_vram(tile->migrate,
+					__fence = xe_migrate_to_vram(vr->migrate,
 								     i - pos + incr,
 								     dma_addr + pos,
 								     vram_addr);
@@ -434,17 +433,17 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 			/* Extra mismatched device page, copy it */
 			if (!match && last && vram_addr != XE_VRAM_ADDR_INVALID) {
 				if (sram) {
-					vm_dbg(&tile->xe->drm,
+					vm_dbg(&xe->drm,
 					       "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%d",
 					       vram_addr, (u64)dma_addr[pos], 1);
-					__fence = xe_migrate_from_vram(tile->migrate, 1,
+					__fence = xe_migrate_from_vram(vr->migrate, 1,
 								       vram_addr,
 								       dma_addr + pos);
 				} else {
-					vm_dbg(&tile->xe->drm,
+					vm_dbg(&xe->drm,
 					       "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%d",
 					       (u64)dma_addr[pos], vram_addr, 1);
-					__fence = xe_migrate_to_vram(tile->migrate, 1,
+					__fence = xe_migrate_to_vram(vr->migrate, 1,
 								     dma_addr + pos,
 								     vram_addr);
 				}
@@ -502,9 +501,9 @@ static u64 block_offset_to_pfn(struct xe_vram_region *vr, u64 offset)
 	return PHYS_PFN(offset + vr->hpa_base);
 }
 
-static struct drm_buddy *tile_to_buddy(struct xe_tile *tile)
+static struct drm_buddy *vram_to_buddy(struct xe_vram_region *vram)
 {
-	return &tile->mem.vram->ttm.mm;
+	return &vram->ttm.mm;
 }
 
 static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocation,
@@ -518,8 +517,7 @@ static int xe_svm_populate_devmem_pfn(struct drm_pagemap_devmem *devmem_allocati
 
 	list_for_each_entry(block, blocks, link) {
 		struct xe_vram_region *vr = block->private;
-		struct xe_tile *tile = vr->tile;
-		struct drm_buddy *buddy = tile_to_buddy(tile);
+		struct drm_buddy *buddy = vram_to_buddy(vr);
 		u64 block_pfn = block_offset_to_pfn(vr, drm_buddy_block_offset(block));
 		int i;
 
@@ -685,8 +683,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 				      unsigned long timeslice_ms)
 {
 	struct xe_vram_region *vr = container_of(dpagemap, typeof(*vr), dpagemap);
-	struct xe_tile *tile = vr->tile;
-	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_device *xe = vr->xe;
 	struct device *dev = xe->drm.dev;
 	struct drm_buddy_block *block;
 	struct list_head *blocks;
@@ -700,9 +697,9 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 	xe_pm_runtime_get(xe);
 
  retry:
-	bo = xe_bo_create_locked(tile_to_xe(tile), NULL, NULL, end - start,
+	bo = xe_bo_create_locked(vr->xe, NULL, NULL, end - start,
 				 ttm_bo_type_device,
-				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+				 (IS_DGFX(xe) ? XE_BO_FLAG_VRAM(vr) : XE_BO_FLAG_SYSTEM) |
 				 XE_BO_FLAG_CPU_ADDR_MIRROR);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
@@ -712,9 +709,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 	}
 
 	drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm,
-				&dpagemap_devmem_ops,
-				&tile->mem.vram->dpagemap,
-				end - start);
+				&dpagemap_devmem_ops, dpagemap, end - start);
 
 	blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks;
 	list_for_each_entry(block, blocks, link)
diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c
index 171de6a4ba95..0be0a5c57ef4 100644
--- a/drivers/gpu/drm/xe/xe_tile.c
+++ b/drivers/gpu/drm/xe/xe_tile.c
@@ -7,6 +7,7 @@
 
 #include <drm/drm_managed.h>
 
+#include "xe_bo.h"
 #include "xe_device.h"
 #include "xe_ggtt.h"
 #include "xe_gt.h"
@@ -118,11 +119,9 @@ int xe_tile_alloc_vram(struct xe_tile *tile)
 	if (!IS_DGFX(xe))
 		return 0;
 
-	vram = drmm_kzalloc(&xe->drm, sizeof(*vram), GFP_KERNEL);
-	if (!vram)
-		return -ENOMEM;
-
-	vram->tile = tile;
+	vram = xe_vram_region_alloc(xe, tile->id, XE_PL_VRAM0 + tile->id);
+	if (IS_ERR(vram))
+		return PTR_ERR(vram);
 	tile->mem.vram = vram;
 
 	return 0;
@@ -160,21 +159,6 @@ int xe_tile_init_early(struct xe_tile *tile, struct xe_device *xe, u8 id)
 }
 ALLOW_ERROR_INJECTION(xe_tile_init_early, ERRNO); /* See xe_pci_probe() */
 
-static int tile_ttm_mgr_init(struct xe_tile *tile)
-{
-	struct xe_device *xe = tile_to_xe(tile);
-	int err;
-
-	if (tile->mem.vram) {
-		err = xe_ttm_vram_mgr_init(tile, &tile->mem.vram->ttm);
-		if (err)
-			return err;
-		xe->info.mem_region_mask |= BIT(tile->id) << 1;
-	}
-
-	return 0;
-}
-
 /**
  * xe_tile_init_noalloc - Init tile up to the point where allocations can happen.
  * @tile: The tile to initialize.
@@ -192,17 +176,20 @@ static int tile_ttm_mgr_init(struct xe_tile *tile)
 int xe_tile_init_noalloc(struct xe_tile *tile)
 {
 	struct xe_device *xe = tile_to_xe(tile);
-	int err;
-
-	err = tile_ttm_mgr_init(tile);
-	if (err)
-		return err;
 
 	xe_wa_apply_tile_workarounds(tile);
 
 	if (xe->info.has_usm && IS_DGFX(xe))
 		xe_devm_add(tile, tile->mem.vram);
 
+	if (IS_DGFX(xe) && !ttm_resource_manager_used(&tile->mem.vram->ttm.manager)) {
+		int err = xe_ttm_vram_mgr_init(xe, tile->mem.vram);
+
+		if (err)
+			return err;
+		xe->info.mem_region_mask |= BIT(tile->mem.vram->id) << 1;
+	}
+
 	return xe_tile_sysfs_init(tile);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
index 8f9b8a1d2c05..9175b4a2214b 100644
--- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
+++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.c
@@ -338,12 +338,18 @@ int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr,
 	return drmm_add_action_or_reset(&xe->drm, ttm_vram_mgr_fini, mgr);
 }
 
-int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr)
+/**
+ * xe_ttm_vram_mgr_init - initialize TTM VRAM region
+ * @xe: pointer to Xe device
+ * @vram: pointer to xe_vram_region that contains the memory region attributes
+ *
+ * Initialize the Xe TTM for given @vram region using the given parameters.
+ *
+ * Returns 0 for success, negative error code otherwise.
+ */
+int xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_vram_region *vram)
 {
-	struct xe_device *xe = tile_to_xe(tile);
-	struct xe_vram_region *vram = tile->mem.vram;
-
-	return __xe_ttm_vram_mgr_init(xe, mgr, XE_PL_VRAM0 + tile->id,
+	return __xe_ttm_vram_mgr_init(xe, &vram->ttm, vram->placement,
 				      xe_vram_region_usable_size(vram),
 				      xe_vram_region_io_size(vram),
 				      PAGE_SIZE);
diff --git a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h
index cc76050e376d..87b7fae5edba 100644
--- a/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h
+++ b/drivers/gpu/drm/xe/xe_ttm_vram_mgr.h
@@ -11,11 +11,12 @@
 enum dma_data_direction;
 struct xe_device;
 struct xe_tile;
+struct xe_vram_region;
 
 int __xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_ttm_vram_mgr *mgr,
 			   u32 mem_type, u64 size, u64 io_size,
 			   u64 default_page_size);
-int xe_ttm_vram_mgr_init(struct xe_tile *tile, struct xe_ttm_vram_mgr *mgr);
+int xe_ttm_vram_mgr_init(struct xe_device *xe, struct xe_vram_region *vram);
 int xe_ttm_vram_mgr_alloc_sgt(struct xe_device *xe,
 			      struct ttm_resource *res,
 			      u64 offset, u64 length,
diff --git a/drivers/gpu/drm/xe/xe_vram.c b/drivers/gpu/drm/xe/xe_vram.c
index 366e5d8a85ca..b44ebf50fedb 100644
--- a/drivers/gpu/drm/xe/xe_vram.c
+++ b/drivers/gpu/drm/xe/xe_vram.c
@@ -20,6 +20,7 @@
 #include "xe_mmio.h"
 #include "xe_module.h"
 #include "xe_sriov.h"
+#include "xe_ttm_vram_mgr.h"
 #include "xe_vram.h"
 #include "xe_vram_types.h"
 
@@ -138,7 +139,7 @@ static bool resource_is_valid(struct pci_dev *pdev, int bar)
 	return true;
 }
 
-static int determine_lmem_bar_size(struct xe_device *xe)
+static int determine_lmem_bar_size(struct xe_device *xe, struct xe_vram_region *lmem_bar)
 {
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 
@@ -149,17 +150,16 @@ static int determine_lmem_bar_size(struct xe_device *xe)
 
 	resize_vram_bar(xe);
 
-	xe->mem.vram->io_start = pci_resource_start(pdev, LMEM_BAR);
-	xe->mem.vram->io_size = pci_resource_len(pdev, LMEM_BAR);
-	if (!xe->mem.vram->io_size)
+	lmem_bar->io_start = pci_resource_start(pdev, LMEM_BAR);
+	lmem_bar->io_size = pci_resource_len(pdev, LMEM_BAR);
+	if (!lmem_bar->io_size)
 		return -EIO;
 
 	/* XXX: Need to change when xe link code is ready */
-	xe->mem.vram->dpa_base = 0;
+	lmem_bar->dpa_base = 0;
 
 	/* set up a map to the total memory area. */
-	xe->mem.vram->mapping = devm_ioremap_wc(&pdev->dev, xe->mem.vram->io_start,
-						xe->mem.vram->io_size);
+	lmem_bar->mapping = devm_ioremap_wc(&pdev->dev, lmem_bar->io_start, lmem_bar->io_size);
 
 	return 0;
 }
@@ -287,6 +287,67 @@ static void vram_fini(void *arg)
 		tile->mem.vram->mapping = NULL;
 }
 
+struct xe_vram_region *xe_vram_region_alloc(struct xe_device *xe, u8 id, u32 placement)
+{
+	struct xe_vram_region *vram;
+	struct drm_device *drm = &xe->drm;
+
+	xe_assert(xe, id < xe->info.tile_count);
+
+	vram = drmm_kzalloc(drm, sizeof(*vram), GFP_KERNEL);
+	if (!vram)
+		return NULL;
+
+	vram->xe = xe;
+	vram->id = id;
+	vram->placement = placement;
+#if defined(CONFIG_DRM_XE_PAGEMAP)
+	vram->migrate = xe->tiles[id].migrate;
+#endif
+	return vram;
+}
+
+static void print_vram_region_info(struct xe_device *xe, struct xe_vram_region *vram)
+{
+	struct drm_device *drm = &xe->drm;
+
+	if (vram->io_size < vram->usable_size)
+		drm_info(drm, "Small BAR device\n");
+
+	drm_info(drm,
+		 "VRAM[%u]: Actual physical size %pa, usable size exclude stolen %pa, CPU accessible size %pa\n",
+		 vram->id, &vram->actual_physical_size, &vram->usable_size, &vram->io_size);
+	drm_info(drm, "VRAM[%u]: DPA range: [%pa-%llx], io range: [%pa-%llx]\n",
+		 vram->id, &vram->dpa_base, vram->dpa_base + (u64)vram->actual_physical_size,
+		 &vram->io_start, vram->io_start + (u64)vram->io_size);
+}
+
+static int vram_region_init(struct xe_device *xe, struct xe_vram_region *vram,
+			    struct xe_vram_region *lmem_bar, u64 offset, u64 usable_size,
+			    u64 region_size, resource_size_t remain_io_size)
+{
+	/* Check if VRAM region is already initialized */
+	if (vram->mapping)
+		return 0;
+
+	vram->actual_physical_size = region_size;
+	vram->io_start = lmem_bar->io_start + offset;
+	vram->io_size = min_t(u64, usable_size, remain_io_size);
+
+	if (!vram->io_size) {
+		drm_err(&xe->drm, "Tile without any CPU visible VRAM. Aborting.\n");
+		return -ENODEV;
+	}
+
+	vram->dpa_base = lmem_bar->dpa_base + offset;
+	vram->mapping = lmem_bar->mapping + offset;
+	vram->usable_size = usable_size;
+
+	print_vram_region_info(xe, vram);
+
+	return 0;
+}
+
 /**
  * xe_vram_probe() - Probe VRAM configuration
  * @xe: the &xe_device
@@ -298,82 +359,52 @@ static void vram_fini(void *arg)
 int xe_vram_probe(struct xe_device *xe)
 {
 	struct xe_tile *tile;
-	resource_size_t io_size;
+	struct xe_vram_region lmem_bar;
+	resource_size_t remain_io_size;
 	u64 available_size = 0;
 	u64 total_size = 0;
-	u64 tile_offset;
-	u64 tile_size;
-	u64 vram_size;
 	int err;
 	u8 id;
 
 	if (!IS_DGFX(xe))
 		return 0;
 
-	/* Get the size of the root tile's vram for later accessibility comparison */
-	tile = xe_device_get_root_tile(xe);
-	err = tile_vram_size(tile, &vram_size, &tile_size, &tile_offset);
+	err = determine_lmem_bar_size(xe, &lmem_bar);
 	if (err)
 		return err;
+	drm_info(&xe->drm, "VISIBLE VRAM: %pa, %pa\n", &lmem_bar.io_start, &lmem_bar.io_size);
 
-	err = determine_lmem_bar_size(xe);
-	if (err)
-		return err;
+	remain_io_size = lmem_bar.io_size;
 
-	drm_info(&xe->drm, "VISIBLE VRAM: %pa, %pa\n", &xe->mem.vram->io_start,
-		 &xe->mem.vram->io_size);
-
-	io_size = xe->mem.vram->io_size;
-
-	/* tile specific ranges */
 	for_each_tile(tile, xe, id) {
-		err = tile_vram_size(tile, &vram_size, &tile_size, &tile_offset);
+		u64 region_size;
+		u64 usable_size;
+		u64 tile_offset;
+
+		err = tile_vram_size(tile, &usable_size, &region_size, &tile_offset);
 		if (err)
 			return err;
 
-		tile->mem.vram->actual_physical_size = tile_size;
-		tile->mem.vram->io_start = xe->mem.vram->io_start + tile_offset;
-		tile->mem.vram->io_size = min_t(u64, vram_size, io_size);
+		total_size += region_size;
+		available_size += usable_size;
 
-		if (!tile->mem.vram->io_size) {
-			drm_err(&xe->drm, "Tile without any CPU visible VRAM. Aborting.\n");
-			return -ENODEV;
-		}
+		err = vram_region_init(xe, tile->mem.vram, &lmem_bar, tile_offset, usable_size,
+				       region_size, remain_io_size);
+		if (err)
+			return err;
 
-		tile->mem.vram->dpa_base = xe->mem.vram->dpa_base + tile_offset;
-		tile->mem.vram->usable_size = vram_size;
-		tile->mem.vram->mapping = xe->mem.vram->mapping + tile_offset;
-
-		if (tile->mem.vram->io_size < tile->mem.vram->usable_size)
-			drm_info(&xe->drm, "Small BAR device\n");
-		drm_info(&xe->drm,
-			 "VRAM[%u, %u]: Actual physical size %pa, usable size exclude stolen %pa, CPU accessible size %pa\n",
-			 id, tile->id, &tile->mem.vram->actual_physical_size,
-			 &tile->mem.vram->usable_size, &tile->mem.vram->io_size);
-		drm_info(&xe->drm, "VRAM[%u, %u]: DPA range: [%pa-%llx], io range: [%pa-%llx]\n",
-			 id, tile->id, &tile->mem.vram->dpa_base,
-			 tile->mem.vram->dpa_base + (u64)tile->mem.vram->actual_physical_size,
-			 &tile->mem.vram->io_start,
-			 tile->mem.vram->io_start + (u64)tile->mem.vram->io_size);
-
-		/* calculate total size using tile size to get the correct HW sizing */
-		total_size += tile_size;
-		available_size += vram_size;
-
-		if (total_size > xe->mem.vram->io_size) {
+		if (total_size > lmem_bar.io_size) {
 			drm_info(&xe->drm, "VRAM: %pa is larger than resource %pa\n",
-				 &total_size, &xe->mem.vram->io_size);
+				 &total_size, &lmem_bar.io_size);
 		}
 
-		io_size -= min_t(u64, tile_size, io_size);
+		remain_io_size -= min_t(u64, tile->mem.vram->actual_physical_size, remain_io_size);
 	}
 
-	xe->mem.vram->actual_physical_size = total_size;
-
-	drm_info(&xe->drm, "Total VRAM: %pa, %pa\n", &xe->mem.vram->io_start,
-		 &xe->mem.vram->actual_physical_size);
-	drm_info(&xe->drm, "Available VRAM: %pa, %pa\n", &xe->mem.vram->io_start,
-		 &available_size);
+	err = vram_region_init(xe, xe->mem.vram, &lmem_bar, 0, available_size, total_size,
+			       lmem_bar.io_size);
+	if (err)
+		return err;
 
 	return devm_add_action_or_reset(xe->drm.dev, vram_fini, xe);
 }
diff --git a/drivers/gpu/drm/xe/xe_vram.h b/drivers/gpu/drm/xe/xe_vram.h
index d4bf1f9c2a72..72860f714fc6 100644
--- a/drivers/gpu/drm/xe/xe_vram.h
+++ b/drivers/gpu/drm/xe/xe_vram.h
@@ -13,6 +13,8 @@ struct xe_vram_region;
 
 int xe_vram_probe(struct xe_device *xe);
 
+struct xe_vram_region *xe_vram_region_alloc(struct xe_device *xe, u8 id, u32 placement);
+
 resource_size_t xe_vram_region_io_start(const struct xe_vram_region *vram);
 resource_size_t xe_vram_region_io_size(const struct xe_vram_region *vram);
 resource_size_t xe_vram_region_dpa_base(const struct xe_vram_region *vram);
diff --git a/drivers/gpu/drm/xe/xe_vram_types.h b/drivers/gpu/drm/xe/xe_vram_types.h
index a01838236036..83772dcbf1af 100644
--- a/drivers/gpu/drm/xe/xe_vram_types.h
+++ b/drivers/gpu/drm/xe/xe_vram_types.h
@@ -12,7 +12,8 @@
 
 #include "xe_ttm_vram_mgr_types.h"
 
-struct xe_tile;
+struct xe_device;
+struct xe_migrate;
 
 /**
  * struct xe_vram_region - memory region structure
@@ -20,8 +21,14 @@ struct xe_tile;
  * device, such as HBM memory or CXL extension memory.
  */
 struct xe_vram_region {
-	/** @tile: Back pointer to tile */
-	struct xe_tile *tile;
+	/** @xe: Back pointer to xe device */
+	struct xe_device *xe;
+	/**
+	 * @id: VRAM region instance id
+	 *
+	 * The value should be unique for VRAM region.
+	 */
+	u8 id;
 	/** @io_start: IO start address of this VRAM instance */
 	resource_size_t io_start;
 	/**
@@ -54,7 +61,11 @@ struct xe_vram_region {
 	void __iomem *mapping;
 	/** @ttm: VRAM TTM manager */
 	struct xe_ttm_vram_mgr ttm;
+	/** @placement: TTM placement dedicated for this region */
+	u32 placement;
 #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
+	/** @migrate: Back pointer to migrate */
+	struct xe_migrate *migrate;
 	/** @pagemap: Used to remap device memory as ZONE_DEVICE */
 	struct dev_pagemap pagemap;
 	/**

From 2650bc4007c15e05f995f472b4fc89e793162bc4 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Wed, 16 Jul 2025 09:48:17 +0100
Subject: [PATCH 0017/1455] drm/sched: Fix a race in DRM_GPU_SCHED_STAT_NO_HANG
 test
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The "skip reset" test waits for the timeout handler to run for the
duration of 2 * MOCK_TIMEOUT, and because the mock scheduler opted to
remove the "skip reset" flag once it fires, this gives opportunity for the
timeout handler to run twice. Second time the job will be removed from the
mock scheduler job list and the drm_mock_sched_advance() call in the test
will fail.

Fix it by making the "don't reset" flag persist for the lifetime of the
job and add a new flag to verify that the code path had executed as
expected.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Fixes: 1472e7549f84 ("drm/sched: Add new test for DRM_GPU_SCHED_STAT_NO_HANG")
Cc: Maíra Canal <mcanal@igalia.com>
Cc: Philipp Stanner <phasta@kernel.org>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250716084817.56797-1-tvrtko.ursulin@igalia.com
---
 drivers/gpu/drm/scheduler/tests/mock_scheduler.c | 2 +-
 drivers/gpu/drm/scheduler/tests/sched_tests.h    | 7 ++++---
 drivers/gpu/drm/scheduler/tests/tests_basic.c    | 4 ++--
 3 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c
index 65acffc3fea8..8e9ae7d980eb 100644
--- a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c
+++ b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c
@@ -219,7 +219,7 @@ mock_sched_timedout_job(struct drm_sched_job *sched_job)
 	unsigned long flags;
 
 	if (job->flags & DRM_MOCK_SCHED_JOB_DONT_RESET) {
-		job->flags &= ~DRM_MOCK_SCHED_JOB_DONT_RESET;
+		job->flags |= DRM_MOCK_SCHED_JOB_RESET_SKIPPED;
 		return DRM_GPU_SCHED_STAT_NO_HANG;
 	}
 
diff --git a/drivers/gpu/drm/scheduler/tests/sched_tests.h b/drivers/gpu/drm/scheduler/tests/sched_tests.h
index 63d4f2ac7074..5b262126b776 100644
--- a/drivers/gpu/drm/scheduler/tests/sched_tests.h
+++ b/drivers/gpu/drm/scheduler/tests/sched_tests.h
@@ -95,9 +95,10 @@ struct drm_mock_sched_job {
 
 	struct completion	done;
 
-#define DRM_MOCK_SCHED_JOB_DONE		0x1
-#define DRM_MOCK_SCHED_JOB_TIMEDOUT	0x2
-#define DRM_MOCK_SCHED_JOB_DONT_RESET	0x4
+#define DRM_MOCK_SCHED_JOB_DONE			0x1
+#define DRM_MOCK_SCHED_JOB_TIMEDOUT		0x2
+#define DRM_MOCK_SCHED_JOB_DONT_RESET		0x4
+#define DRM_MOCK_SCHED_JOB_RESET_SKIPPED	0x8
 	unsigned long		flags;
 
 	struct list_head	link;
diff --git a/drivers/gpu/drm/scheduler/tests/tests_basic.c b/drivers/gpu/drm/scheduler/tests/tests_basic.c
index 55eb142bd7c5..82a41a456b0a 100644
--- a/drivers/gpu/drm/scheduler/tests/tests_basic.c
+++ b/drivers/gpu/drm/scheduler/tests/tests_basic.c
@@ -317,8 +317,8 @@ static void drm_sched_skip_reset(struct kunit *test)
 	KUNIT_ASSERT_FALSE(test, done);
 
 	KUNIT_ASSERT_EQ(test,
-			job->flags & DRM_MOCK_SCHED_JOB_DONT_RESET,
-			0);
+			job->flags & DRM_MOCK_SCHED_JOB_RESET_SKIPPED,
+			DRM_MOCK_SCHED_JOB_RESET_SKIPPED);
 
 	i = drm_mock_sched_advance(sched, 1);
 	KUNIT_ASSERT_EQ(test, i, 1);

From 5c30a6e12ce87a1e6ff1d179f1afb9f97e0a1346 Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Fri, 11 Jul 2025 09:48:59 +0530
Subject: [PATCH 0018/1455] drm/i915/display_wa: Add helpers to check wa

Introduce a generic helper to check display workarounds using an enum.

Convert Wa_16023588340 to use the new interface, simplifying WA checks
and making future additions easier.

v2: Use drm_WARN instead of MISSING_CASE and simplify intel_display_wa
macro. (Jani)
v3: Print Missing wa number, instead of enum value. (Gustavo, Jani)

Suggested-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250711041901.1607823-2-ankit.k.nautiyal@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_wa.c | 15 +++++++++++++++
 drivers/gpu/drm/i915/display/intel_display_wa.h |  9 +++++++++
 drivers/gpu/drm/i915/display/intel_fbc.c        |  2 +-
 3 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.c b/drivers/gpu/drm/i915/display/intel_display_wa.c
index f57280e9d041..32719e2c6025 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.c
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.c
@@ -3,6 +3,8 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include <drm/drm_print.h>
+
 #include "i915_reg.h"
 #include "intel_de.h"
 #include "intel_display_core.h"
@@ -39,3 +41,16 @@ void intel_display_wa_apply(struct intel_display *display)
 	else if (DISPLAY_VER(display) == 11)
 		gen11_display_wa_apply(display);
 }
+
+bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name)
+{
+	switch (wa) {
+	case INTEL_DISPLAY_WA_16023588340:
+		return intel_display_needs_wa_16023588340(display);
+	default:
+		drm_WARN(display->drm, 1, "Missing Wa number: %s\n", name);
+		break;
+	}
+
+	return false;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.h b/drivers/gpu/drm/i915/display/intel_display_wa.h
index babd9d16603d..8319e16eb460 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.h
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.h
@@ -21,4 +21,13 @@ static inline bool intel_display_needs_wa_16023588340(struct intel_display *disp
 bool intel_display_needs_wa_16023588340(struct intel_display *display);
 #endif
 
+enum intel_display_wa {
+	INTEL_DISPLAY_WA_16023588340,
+};
+
+bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name);
+
+#define intel_display_wa(__display, __wa) \
+	__intel_display_wa((__display), INTEL_DISPLAY_WA_##__wa, __stringify(__wa))
+
 #endif
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 6e26cb4c5724..e2e03af520b2 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -1464,7 +1464,7 @@ static int intel_fbc_check_plane(struct intel_atomic_state *state,
 		return 0;
 	}
 
-	if (intel_display_needs_wa_16023588340(display)) {
+	if (intel_display_wa(display, 16023588340)) {
 		plane_state->no_fbc_reason = "Wa_16023588340";
 		return 0;
 	}

From 7565fd5dcb0bd6ad843b8d0acbaa9efea5b481a9 Mon Sep 17 00:00:00 2001
From: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Date: Tue, 15 Jul 2025 19:52:11 +0530
Subject: [PATCH 0019/1455] drm/i915/gmbus: Add Wa_16025573575 for PTL/WCL for
 bit-bashing

As per Wa_16025573575 for PTL/WCL, set the GPIO masks bit before starting
bit-bashing and maintain value through the bit-bashing sequence. After the
bit-bashing sequence is done, clear the GPIO masks bits.

v2:
-Use new helper for display workarounds. (Jani)
-Use a separate if-block for the workaround. (Gustavo)

v3:
-Document the workaround details in intel_display_wa.c
-Extend the workaround to WCL too. (Gustavo)

v4:
-Fix the platform check. (Gustavo)
-Avoid read when preserve bits are 0. (Gustavo)

Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250715142211.3145299-1-ankit.k.nautiyal@intel.com
---
 .../gpu/drm/i915/display/intel_display_wa.c   | 12 ++++++
 .../gpu/drm/i915/display/intel_display_wa.h   |  1 +
 drivers/gpu/drm/i915/display/intel_gmbus.c    | 39 ++++++++++++++++---
 3 files changed, 47 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.c b/drivers/gpu/drm/i915/display/intel_display_wa.c
index 32719e2c6025..399c08902413 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.c
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.c
@@ -42,11 +42,23 @@ void intel_display_wa_apply(struct intel_display *display)
 		gen11_display_wa_apply(display);
 }
 
+/*
+ * Wa_16025573575:
+ * Fixes: Issue with bitbashing on Xe3 based platforms.
+ * Workaround: Set masks bits in GPIO CTL and preserve it during bitbashing sequence.
+ */
+static bool intel_display_needs_wa_16025573575(struct intel_display *display)
+{
+	return DISPLAY_VERx100(display) == 3000 || DISPLAY_VERx100(display) == 3002;
+}
+
 bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name)
 {
 	switch (wa) {
 	case INTEL_DISPLAY_WA_16023588340:
 		return intel_display_needs_wa_16023588340(display);
+	case INTEL_DISPLAY_WA_16025573575:
+		return intel_display_needs_wa_16025573575(display);
 	default:
 		drm_WARN(display->drm, 1, "Missing Wa number: %s\n", name);
 		break;
diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.h b/drivers/gpu/drm/i915/display/intel_display_wa.h
index 8319e16eb460..aedea4cfa3ce 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.h
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.h
@@ -23,6 +23,7 @@ bool intel_display_needs_wa_16023588340(struct intel_display *display);
 
 enum intel_display_wa {
 	INTEL_DISPLAY_WA_16023588340,
+	INTEL_DISPLAY_WA_16025573575,
 };
 
 bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name);
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index 0d73f32fe7f1..637f0f23f163 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -39,6 +39,7 @@
 #include "intel_de.h"
 #include "intel_display_regs.h"
 #include "intel_display_types.h"
+#include "intel_display_wa.h"
 #include "intel_gmbus.h"
 #include "intel_gmbus_regs.h"
 
@@ -240,14 +241,20 @@ static void bxt_gmbus_clock_gating(struct intel_display *display,
 static u32 get_reserved(struct intel_gmbus *bus)
 {
 	struct intel_display *display = bus->display;
-	u32 reserved = 0;
+	u32 preserve_bits = 0;
+
+	if (display->platform.i830 || display->platform.i845g)
+		return 0;
 
 	/* On most chips, these bits must be preserved in software. */
-	if (!display->platform.i830 && !display->platform.i845g)
-		reserved = intel_de_read_notrace(display, bus->gpio_reg) &
-			(GPIO_DATA_PULLUP_DISABLE | GPIO_CLOCK_PULLUP_DISABLE);
+	preserve_bits |= GPIO_DATA_PULLUP_DISABLE | GPIO_CLOCK_PULLUP_DISABLE;
 
-	return reserved;
+	/* Wa_16025573575: the masks bits need to be preserved through out */
+	if (intel_display_wa(display, 16025573575))
+		preserve_bits |= GPIO_CLOCK_DIR_MASK | GPIO_CLOCK_VAL_MASK |
+				 GPIO_DATA_DIR_MASK | GPIO_DATA_VAL_MASK;
+
+	return intel_de_read_notrace(display, bus->gpio_reg) & preserve_bits;
 }
 
 static int get_clock(void *data)
@@ -308,6 +315,22 @@ static void set_data(void *data, int state_high)
 	intel_de_posting_read(display, bus->gpio_reg);
 }
 
+static void
+ptl_handle_mask_bits(struct intel_gmbus *bus, bool set)
+{
+	struct intel_display *display = bus->display;
+	u32 reg_val = intel_de_read_notrace(display, bus->gpio_reg);
+	u32 mask_bits = GPIO_CLOCK_DIR_MASK | GPIO_CLOCK_VAL_MASK |
+			GPIO_DATA_DIR_MASK | GPIO_DATA_VAL_MASK;
+	if (set)
+		reg_val |= mask_bits;
+	else
+		reg_val &= ~mask_bits;
+
+	intel_de_write_notrace(display, bus->gpio_reg, reg_val);
+	intel_de_posting_read(display, bus->gpio_reg);
+}
+
 static int
 intel_gpio_pre_xfer(struct i2c_adapter *adapter)
 {
@@ -319,6 +342,9 @@ intel_gpio_pre_xfer(struct i2c_adapter *adapter)
 	if (display->platform.pineview)
 		pnv_gmbus_clock_gating(display, false);
 
+	if (intel_display_wa(display, 16025573575))
+		ptl_handle_mask_bits(bus, true);
+
 	set_data(bus, 1);
 	set_clock(bus, 1);
 	udelay(I2C_RISEFALL_TIME);
@@ -336,6 +362,9 @@ intel_gpio_post_xfer(struct i2c_adapter *adapter)
 
 	if (display->platform.pineview)
 		pnv_gmbus_clock_gating(display, true);
+
+	if (intel_display_wa(display, 16025573575))
+		ptl_handle_mask_bits(bus, false);
 }
 
 static void

From 5774b3cfdedb3624ef0d2c82cccbfd61bcb60fd5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Fri, 11 Jul 2025 12:18:31 -0300
Subject: [PATCH 0020/1455] drm/v3d: Add parameter to retrieve the global
 number of GPU resets
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The GL extension KHR_robustness uses the number of global and per-context
GPU resets to learn about graphics resets that affect a GL context. This
commit introduces a new V3D parameter to retrieve the global number of
GPU resets that have happened since the driver was probed.

To retrieve this information, user-space must use DRM_V3D_PARAM_GLOBAL_RESET_COUNTER.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://lore.kernel.org/r/20250711-v3d-reset-counter-v1-1-1ac73e9fca2d@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_drv.c   | 5 +++++
 drivers/gpu/drm/v3d/v3d_drv.h   | 5 +++++
 drivers/gpu/drm/v3d/v3d_sched.c | 2 ++
 include/uapi/drm/v3d_drm.h      | 1 +
 4 files changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c
index 5e997ae8bc9c..0f53ca746004 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.c
+++ b/drivers/gpu/drm/v3d/v3d_drv.c
@@ -107,6 +107,11 @@ static int v3d_get_param_ioctl(struct drm_device *dev, void *data,
 	case DRM_V3D_PARAM_SUPPORTS_SUPER_PAGES:
 		args->value = !!v3d->gemfs;
 		return 0;
+	case DRM_V3D_PARAM_GLOBAL_RESET_COUNTER:
+		mutex_lock(&v3d->reset_lock);
+		args->value = v3d->reset_counter;
+		mutex_unlock(&v3d->reset_lock);
+		return 0;
 	default:
 		DRM_DEBUG("Unknown parameter %d\n", args->param);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index b51f0b648a08..074cbccf88dd 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -196,6 +196,11 @@ struct v3d_dev {
 	 * all jobs.
 	 */
 	struct v3d_perfmon *global_perfmon;
+
+	/* Global reset counter. The counter must be incremented when
+	 * a GPU reset happens. It must be protected by @reset_lock.
+	 */
+	unsigned int reset_counter;
 };
 
 static inline struct v3d_dev *
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index cc85f1b19ac4..5f8198890e41 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -731,6 +731,8 @@ v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 	/* get the GPU back into the init state */
 	v3d_reset(v3d);
 
+	v3d->reset_counter++;
+
 	for (q = 0; q < V3D_MAX_QUEUES; q++)
 		drm_sched_resubmit_jobs(&v3d->queue[q].sched);
 
diff --git a/include/uapi/drm/v3d_drm.h b/include/uapi/drm/v3d_drm.h
index dbbc404d2b3d..0a7ce2f6be19 100644
--- a/include/uapi/drm/v3d_drm.h
+++ b/include/uapi/drm/v3d_drm.h
@@ -294,6 +294,7 @@ enum drm_v3d_param {
 	DRM_V3D_PARAM_SUPPORTS_CPU_QUEUE,
 	DRM_V3D_PARAM_MAX_PERF_COUNTERS,
 	DRM_V3D_PARAM_SUPPORTS_SUPER_PAGES,
+	DRM_V3D_PARAM_GLOBAL_RESET_COUNTER,
 };
 
 struct drm_v3d_get_param {

From 769c153cfc3c6669c7b318f66c2b21ec3951fb4a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Fri, 11 Jul 2025 12:18:32 -0300
Subject: [PATCH 0021/1455] drm/v3d: Add parameter to retrieve the number of
 GPU resets per-fd
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The GL extension KHR_robustness uses the number of global and per-context
GPU resets to learn about graphics resets that affect a GL context. This
commit introduces a new V3D parameter to retrieve the number of GPU resets
triggered by jobs submitted through a file descriptor.

To retrieve this information, user-space must use DRM_V3D_PARAM_CONTEXT_RESET_COUNTER.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://lore.kernel.org/r/20250711-v3d-reset-counter-v1-2-1ac73e9fca2d@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_drv.c   | 6 ++++++
 drivers/gpu/drm/v3d/v3d_drv.h   | 6 ++++++
 drivers/gpu/drm/v3d/v3d_sched.c | 3 +++
 include/uapi/drm/v3d_drm.h      | 1 +
 4 files changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c
index 0f53ca746004..2def155ce496 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.c
+++ b/drivers/gpu/drm/v3d/v3d_drv.c
@@ -46,6 +46,7 @@ MODULE_PARM_DESC(super_pages, "Enable/Disable Super Pages support.");
 static int v3d_get_param_ioctl(struct drm_device *dev, void *data,
 			       struct drm_file *file_priv)
 {
+	struct v3d_file_priv *v3d_priv = file_priv->driver_priv;
 	struct v3d_dev *v3d = to_v3d_dev(dev);
 	struct drm_v3d_get_param *args = data;
 	static const u32 reg_map[] = {
@@ -112,6 +113,11 @@ static int v3d_get_param_ioctl(struct drm_device *dev, void *data,
 		args->value = v3d->reset_counter;
 		mutex_unlock(&v3d->reset_lock);
 		return 0;
+	case DRM_V3D_PARAM_CONTEXT_RESET_COUNTER:
+		mutex_lock(&v3d->reset_lock);
+		args->value = v3d_priv->reset_counter;
+		mutex_unlock(&v3d->reset_lock);
+		return 0;
 	default:
 		DRM_DEBUG("Unknown parameter %d\n", args->param);
 		return -EINVAL;
diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index 074cbccf88dd..dabda7aaf000 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -230,6 +230,12 @@ struct v3d_file_priv {
 
 	/* Stores the GPU stats for a specific queue for this fd. */
 	struct v3d_stats stats[V3D_MAX_QUEUES];
+
+	/* Per-fd reset counter, must be incremented when a job submitted
+	 * by this fd causes a GPU reset. It must be protected by
+	 * &struct v3d_dev->reset_lock.
+	 */
+	unsigned int reset_counter;
 };
 
 struct v3d_bo {
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 5f8198890e41..7d45664f03c7 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -717,6 +717,8 @@ v3d_cache_clean_job_run(struct drm_sched_job *sched_job)
 static enum drm_gpu_sched_stat
 v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 {
+	struct v3d_job *job = to_v3d_job(sched_job);
+	struct v3d_file_priv *v3d_priv = job->file->driver_priv;
 	enum v3d_queue q;
 
 	mutex_lock(&v3d->reset_lock);
@@ -732,6 +734,7 @@ v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 	v3d_reset(v3d);
 
 	v3d->reset_counter++;
+	v3d_priv->reset_counter++;
 
 	for (q = 0; q < V3D_MAX_QUEUES; q++)
 		drm_sched_resubmit_jobs(&v3d->queue[q].sched);
diff --git a/include/uapi/drm/v3d_drm.h b/include/uapi/drm/v3d_drm.h
index 0a7ce2f6be19..d9b01f4c3a04 100644
--- a/include/uapi/drm/v3d_drm.h
+++ b/include/uapi/drm/v3d_drm.h
@@ -295,6 +295,7 @@ enum drm_v3d_param {
 	DRM_V3D_PARAM_MAX_PERF_COUNTERS,
 	DRM_V3D_PARAM_SUPPORTS_SUPER_PAGES,
 	DRM_V3D_PARAM_GLOBAL_RESET_COUNTER,
+	DRM_V3D_PARAM_CONTEXT_RESET_COUNTER,
 };
 
 struct drm_v3d_get_param {

From 487579fd8524c58dc237904c1b1d9e7c89e53378 Mon Sep 17 00:00:00 2001
From: Soham Purkait <soham.purkait@intel.com>
Date: Wed, 16 Jul 2025 15:44:12 +0530
Subject: [PATCH 0022/1455] drm/xe/xe_debugfs: Exposure of G-State and pcie
 link state residency counters through debugfs

Add debug nodes, "dgfx_pkg_residencies" for G-states (G2, G6, G8, G10,
ModS) and "dgfx_pcie_link_residencies" for PCIe link states(L0, L1, L1.2)
residency counters.

v1:
 - Expose all G-State residency counter values under
   dgfx_pkg_residencies. (Anshuman)
 - Include runtime_get/put. (Riana)
v2:
 - Move offset macros to drm/xe/regs/xe_pmt. (Riana)
v3:
 - Include debugfs node "dgfx_pcie_link_residencies" for pcie link
   residency counter values.  (Anshuman)
v4:
 - Include check for BMG and add helper function for repetitive
   code. (Riana)
 - Add for loop and local struct to avoid repetition. (Riana)
 - Use "drm_debugfs_create_files" to create debugfs. (Karthik)
v5:
 - Reorder commits to reflect the correct dependency hierarchy.  (Jonathan)
 - Simplification of commit message and rectified register offset.(Karthik)
 - Error handling and return before printing.                       (Riana)
v6:
 - Remove check for DGFX as BMG is discrete. (Karthik)
 - Rearrange residency offsets in ascending order. (Riana)
v7:
 - Squash the macros into the patch they are used in. (Lucas)

Signed-off-by: Soham Purkait <soham.purkait@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Karthik Poosa <karthik.poosa@intel.com>
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250716101412.3062780-2-soham.purkait@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_pmt.h | 10 ++++
 drivers/gpu/drm/xe/xe_debugfs.c  | 85 ++++++++++++++++++++++++++++++++
 2 files changed, 95 insertions(+)

diff --git a/drivers/gpu/drm/xe/regs/xe_pmt.h b/drivers/gpu/drm/xe/regs/xe_pmt.h
index 2995d72c3f78..264e9baf949c 100644
--- a/drivers/gpu/drm/xe/regs/xe_pmt.h
+++ b/drivers/gpu/drm/xe/regs/xe_pmt.h
@@ -21,4 +21,14 @@
 #define SG_REMAP_INDEX1			XE_REG(SOC_BASE + 0x08)
 #define   SG_REMAP_BITS			REG_GENMASK(31, 24)
 
+#define BMG_MODS_RESIDENCY_OFFSET		(0x4D0)
+#define BMG_G2_RESIDENCY_OFFSET		(0x530)
+#define BMG_G6_RESIDENCY_OFFSET		(0x538)
+#define BMG_G8_RESIDENCY_OFFSET		(0x540)
+#define BMG_G10_RESIDENCY_OFFSET		(0x548)
+
+#define BMG_PCIE_LINK_L0_RESIDENCY_OFFSET	(0x570)
+#define BMG_PCIE_LINK_L1_RESIDENCY_OFFSET	(0x578)
+#define BMG_PCIE_LINK_L1_2_RESIDENCY_OFFSET	(0x580)
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 129186f4193b..0b4a532f7c45 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -11,18 +11,21 @@
 
 #include <drm/drm_debugfs.h>
 
+#include "regs/xe_pmt.h"
 #include "xe_bo.h"
 #include "xe_device.h"
 #include "xe_force_wake.h"
 #include "xe_gt_debugfs.h"
 #include "xe_gt_printk.h"
 #include "xe_guc_ads.h"
+#include "xe_mmio.h"
 #include "xe_pm.h"
 #include "xe_pxp_debugfs.h"
 #include "xe_sriov.h"
 #include "xe_sriov_pf.h"
 #include "xe_step.h"
 #include "xe_wa.h"
+#include "xe_vsec.h"
 
 #ifdef CONFIG_DRM_XE_DEBUG
 #include "xe_bo_evict.h"
@@ -32,6 +35,23 @@
 
 DECLARE_FAULT_ATTR(gt_reset_failure);
 
+static void read_residency_counter(struct xe_device *xe, struct xe_mmio *mmio,
+				   u32 offset, char *name, struct drm_printer *p)
+{
+	u64 residency = 0;
+	int ret;
+
+	ret = xe_pmt_telem_read(to_pci_dev(xe->drm.dev),
+				xe_mmio_read32(mmio, PUNIT_TELEMETRY_GUID),
+				&residency, offset, sizeof(residency));
+	if (ret != sizeof(residency)) {
+		drm_warn(&xe->drm, "%s counter failed to read, ret %d\n", name, ret);
+		return;
+	}
+
+	drm_printf(p, "%s : %llu\n", name, residency);
+}
+
 static struct xe_device *node_to_xe(struct drm_info_node *node)
 {
 	return to_xe_device(node->minor->dev);
@@ -102,12 +122,72 @@ static int workaround_info(struct seq_file *m, void *data)
 	return 0;
 }
 
+static int dgfx_pkg_residencies_show(struct seq_file *m, void *data)
+{
+	struct xe_device *xe;
+	struct xe_mmio *mmio;
+	struct drm_printer p;
+
+	xe = node_to_xe(m->private);
+	p = drm_seq_file_printer(m);
+	xe_pm_runtime_get(xe);
+	mmio = xe_root_tile_mmio(xe);
+	struct {
+		u32 offset;
+		char *name;
+	} residencies[] = {
+		{BMG_G2_RESIDENCY_OFFSET, "Package G2"},
+		{BMG_G6_RESIDENCY_OFFSET, "Package G6"},
+		{BMG_G8_RESIDENCY_OFFSET, "Package G8"},
+		{BMG_G10_RESIDENCY_OFFSET, "Package G10"},
+		{BMG_MODS_RESIDENCY_OFFSET, "Package ModS"}
+	};
+
+	for (int i = 0; i < ARRAY_SIZE(residencies); i++)
+		read_residency_counter(xe, mmio, residencies[i].offset, residencies[i].name, &p);
+
+	xe_pm_runtime_put(xe);
+	return 0;
+}
+
+static int dgfx_pcie_link_residencies_show(struct seq_file *m, void *data)
+{
+	struct xe_device *xe;
+	struct xe_mmio *mmio;
+	struct drm_printer p;
+
+	xe = node_to_xe(m->private);
+	p = drm_seq_file_printer(m);
+	xe_pm_runtime_get(xe);
+	mmio = xe_root_tile_mmio(xe);
+
+	struct {
+		u32 offset;
+		char *name;
+	} residencies[] = {
+		{BMG_PCIE_LINK_L0_RESIDENCY_OFFSET, "PCIE LINK L0 RESIDENCY"},
+		{BMG_PCIE_LINK_L1_RESIDENCY_OFFSET, "PCIE LINK L1 RESIDENCY"},
+		{BMG_PCIE_LINK_L1_2_RESIDENCY_OFFSET, "PCIE LINK L1.2 RESIDENCY"}
+	};
+
+	for (int i = 0; i < ARRAY_SIZE(residencies); i++)
+		read_residency_counter(xe, mmio, residencies[i].offset, residencies[i].name, &p);
+
+	xe_pm_runtime_put(xe);
+	return 0;
+}
+
 static const struct drm_info_list debugfs_list[] = {
 	{"info", info, 0},
 	{ .name = "sriov_info", .show = sriov_info, },
 	{ .name = "workarounds", .show = workaround_info, },
 };
 
+static const struct drm_info_list debugfs_residencies[] = {
+	{ .name = "dgfx_pkg_residencies", .show = dgfx_pkg_residencies_show, },
+	{ .name = "dgfx_pcie_link_residencies", .show = dgfx_pcie_link_residencies_show, },
+};
+
 static int forcewake_open(struct inode *inode, struct file *file)
 {
 	struct xe_device *xe = inode->i_private;
@@ -280,6 +360,11 @@ void xe_debugfs_register(struct xe_device *xe)
 				 ARRAY_SIZE(debugfs_list),
 				 root, minor);
 
+	if (xe->info.platform == XE_BATTLEMAGE)
+		drm_debugfs_create_files(debugfs_residencies,
+					 ARRAY_SIZE(debugfs_residencies),
+					 root, minor);
+
 	debugfs_create_file("forcewake_all", 0400, root, xe,
 			    &forcewake_all_fops);
 

From b9a572f471993d3e8bf874fcb57f331d66650440 Mon Sep 17 00:00:00 2001
From: Simon Ser <contact@emersion.fr>
Date: Thu, 1 May 2025 11:29:51 +0000
Subject: [PATCH 0023/1455] drm: document DRM_MODE_PAGE_FLIP_EVENT interactions
 with atomic
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It's not obvious off-hand which CRTCs will get a page-flip event
when using this flag in an atomic commit, because it's all
implicitly implied based on the contents of the atomic commit.
Document requirements for using this flag and how to request an
event for a CRTC.

Note, because prepare_signaling() runs right after
drm_atomic_set_property() calls, page-flip events are not delivered
for CRTCs pulled in later by DRM core (e.g. on modeset by
drm_atomic_helper_check_modeset()) or the driver (e.g. other CRTCs
sharing a DP-MST connector).

v2: fix cut off sentence in commit message (Pekka)

Signed-off-by: Simon Ser <contact@emersion.fr>
Reviewed-by: Simona Vetter <simona@ffwll.ch>
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Pekka Paalanen <pekka.paalanen@collabora.com>
Cc: David Turner <david.turner@raspberrypi.com>
Cc: Daniel Stone <daniel@fooishbar.org>
Link: https://lore.kernel.org/r/20250501112945.6448-1-contact@emersion.fr
---
 include/uapi/drm/drm_mode.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/include/uapi/drm/drm_mode.h b/include/uapi/drm/drm_mode.h
index c082810c08a8..a122bea25593 100644
--- a/include/uapi/drm/drm_mode.h
+++ b/include/uapi/drm/drm_mode.h
@@ -962,6 +962,14 @@ struct hdr_output_metadata {
  * Request that the kernel sends back a vblank event (see
  * struct drm_event_vblank) with the &DRM_EVENT_FLIP_COMPLETE type when the
  * page-flip is done.
+ *
+ * When used with atomic uAPI, one event will be delivered per CRTC included in
+ * the atomic commit. A CRTC is included in an atomic commit if one of its
+ * properties is set, or if a property is set on a connector or plane linked
+ * via the CRTC_ID property to the CRTC. At least one CRTC must be included,
+ * and all pulled in CRTCs must be either previously or newly powered on (in
+ * other words, a powered off CRTC which stays off cannot be included in the
+ * atomic commit).
  */
 #define DRM_MODE_PAGE_FLIP_EVENT 0x01
 /**

From a8b874694db5cae7baaf522756f87acd956e6e66 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:12 +0300
Subject: [PATCH 0024/1455] drm/i915/dp: Fix 2.7 Gbps DP_LINK_BW value on g4x
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On g4x we currently use the 96MHz non-SSC refclk, which can't actually
generate an exact 2.7 Gbps link rate. In practice we end up with 2.688
Gbps which seems to be close enough to actually work, but link training
is currently failing due to miscalculating the DP_LINK_BW value (we
calcualte it directly from port_clock which reflects the actual PLL
outpout frequency).

Ideas how to fix this:
- nudge port_clock back up to 270000 during PLL computation/readout
- track port_clock and the nominal link rate separately so they might
  differ a bit
- switch to the 100MHz refclk, but that one should be SSC so perhaps
  not something we want

While we ponder about a better solution apply some band aid to the
immediate issue of miscalculated DP_LINK_BW value. With this
I can again use 2.7 Gbps link rate on g4x.

Cc: stable@vger.kernel.org
Fixes: 665a7b04092c ("drm/i915: Feed the DPLL output freq back into crtc_state")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-2-ville.syrjala@linux.intel.com
Reviewed-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 6d4e0fa0c054..54d88f24b689 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -1607,6 +1607,12 @@ int intel_dp_rate_select(struct intel_dp *intel_dp, int rate)
 void intel_dp_compute_rate(struct intel_dp *intel_dp, int port_clock,
 			   u8 *link_bw, u8 *rate_select)
 {
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	/* FIXME g4x can't generate an exact 2.7GHz with the 96MHz non-SSC refclk */
+	if (display->platform.g4x && port_clock == 268800)
+		port_clock = 270000;
+
 	/* eDP 1.4 rate select method. */
 	if (intel_dp->use_rate_select) {
 		*link_bw = 0;

From f0da19347bccee1c20091e9fd597a2f197792021 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:13 +0300
Subject: [PATCH 0025/1455] drm/i915/dp: Don't switch to idle pattern before
 disable on pre-hsw
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For some reason we are switching over to the idle pattern before
disabling the DP port on pre-hsw. AFAICS this has never been part
of the documented sequence (and on hsw+ the spec explicitly says
not to do this). Get rid of it.

The code goes all the way back to commit 5eb08b69f510 ("drm/i915: enable
DisplayPort support on IGDNG"), and it was accompanied by a 17ms delay
which got changed to vbl wait in commit ab527efc2fea ("drm/i915: use
wait_for_vblank instead of msleep(17)"), and was later completely removed
in  commit 93c9c19b3d25 ("drm/i915: remove unexplained vblank wait in
the DP off code").

Smoke tested on g4x/snb/chv.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-3-ville.syrjala@linux.intel.com
Reviewed-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/g4x_dp.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c
index 87f6b9602b16..b54edf0d1c23 100644
--- a/drivers/gpu/drm/i915/display/g4x_dp.c
+++ b/drivers/gpu/drm/i915/display/g4x_dp.c
@@ -424,17 +424,6 @@ intel_dp_link_down(struct intel_encoder *encoder,
 
 	drm_dbg_kms(display->drm, "\n");
 
-	if ((display->platform.ivybridge && port == PORT_A) ||
-	    (HAS_PCH_CPT(display) && port != PORT_A)) {
-		intel_dp->DP &= ~DP_LINK_TRAIN_MASK_CPT;
-		intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE_CPT;
-	} else {
-		intel_dp->DP &= ~DP_LINK_TRAIN_MASK;
-		intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE;
-	}
-	intel_de_write(display, intel_dp->output_reg, intel_dp->DP);
-	intel_de_posting_read(display, intel_dp->output_reg);
-
 	intel_dp->DP &= ~DP_PORT_EN;
 	intel_de_write(display, intel_dp->output_reg, intel_dp->DP);
 	intel_de_posting_read(display, intel_dp->output_reg);

From 11fab5a2a1ad78d12f23cee209f1705f6d7ca281 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:14 +0300
Subject: [PATCH 0026/1455] drm/i915/dp: Clear DPCD training pattern before
 transmitting the idle pattern
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are supposed to switch off the training pattern in DPCD before
we start transmitting the idle pattern. For LTTPRs we do that
correctly, but for the sink DPRX we only do this correctly
for some platforms.

On pre-HSW (where we don't implement the .set_idle_link_train()
hook), we directly switch from transmitting the training pattern
to normal pixel transmission (the hardware should hopefully
guarantee that the minimum number of required idle patters will
be transmitted during this transition). The DPCD write correctly
precedes the actual switch away from the training pattern.

For HSW+ we start transmitting the idle pattern earlier, and only
switch off the DPCD training pattern after we switch from the idle
pattern to normal pixel transmission. Adjust the code to disable
the DPCD training pattern before we start transmitting the idle
pattern.

v2: Tweak the commit message a bit

Tested-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-4-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_link_training.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index a479b63112ea..322a26c1910c 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -1130,7 +1130,9 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 
 	intel_dp->link.active = true;
 
-	intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
+	if (!intel_dp->set_idle_link_train)
+		intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
+
 	intel_dp_program_link_training_pattern(intel_dp, crtc_state, DP_PHY_DPRX,
 					       DP_TRAINING_PATTERN_DISABLE);
 
@@ -1371,8 +1373,10 @@ intel_dp_link_train_all_phys(struct intel_dp *intel_dp,
 	if (ret)
 		ret = intel_dp_link_train_phy(intel_dp, crtc_state, DP_PHY_DPRX);
 
-	if (intel_dp->set_idle_link_train)
+	if (intel_dp->set_idle_link_train) {
+		intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
 		intel_dp->set_idle_link_train(intel_dp, crtc_state);
+	}
 
 	return ret;
 }

From b840bb0b7ec279660002afc83217c435e6896870 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:15 +0300
Subject: [PATCH 0027/1455] drm/i915/dp: Have intel_dp_get_adjust_train() tell
 us if anything changed
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In order to implement the POST_LT_ADJ_REQ sequence we need to
know whether the sink actually requested a changed to the
vswing/pre-emph values.

Tested-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-5-ville.syrjala@linux.intel.com
---
 .../drm/i915/display/intel_dp_link_training.c  | 18 +++++++++++++-----
 .../drm/i915/display/intel_dp_link_training.h  |  2 +-
 2 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 322a26c1910c..57b9dc52a98d 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -478,12 +478,13 @@ static u8 intel_dp_get_lane_adjust_train(struct intel_dp *intel_dp,
 	_TRAIN_REQ_TX_FFE_ARGS(link_status, 2), \
 	_TRAIN_REQ_TX_FFE_ARGS(link_status, 3)
 
-void
+bool
 intel_dp_get_adjust_train(struct intel_dp *intel_dp,
 			  const struct intel_crtc_state *crtc_state,
 			  enum drm_dp_phy dp_phy,
 			  const u8 link_status[DP_LINK_STATUS_SIZE])
 {
+	bool changed = false;
 	int lane;
 
 	if (intel_dp_is_uhbr(crtc_state)) {
@@ -502,10 +503,17 @@ intel_dp_get_adjust_train(struct intel_dp *intel_dp,
 		       TRAIN_REQ_PREEMPH_ARGS(link_status));
 	}
 
-	for (lane = 0; lane < 4; lane++)
-		intel_dp->train_set[lane] =
-			intel_dp_get_lane_adjust_train(intel_dp, crtc_state,
-						       dp_phy, link_status, lane);
+	for (lane = 0; lane < 4; lane++) {
+		u8 new = intel_dp_get_lane_adjust_train(intel_dp, crtc_state,
+							dp_phy, link_status, lane);
+		if (intel_dp->train_set[lane] == new)
+			continue;
+
+		intel_dp->train_set[lane] = new;
+		changed = true;
+	}
+
+	return changed;
 }
 
 static int intel_dp_training_pattern_set_reg(struct intel_dp *intel_dp,
diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.h b/drivers/gpu/drm/i915/display/intel_dp_link_training.h
index 46614124569f..1ba22ed6db08 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.h
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.h
@@ -23,7 +23,7 @@ void intel_dp_link_training_set_bw(struct intel_dp *intel_dp,
 				   int link_bw, int rate_select, int lane_count,
 				   bool enhanced_framing);
 
-void intel_dp_get_adjust_train(struct intel_dp *intel_dp,
+bool intel_dp_get_adjust_train(struct intel_dp *intel_dp,
 			       const struct intel_crtc_state *crtc_state,
 			       enum drm_dp_phy dp_phy,
 			       const u8 link_status[DP_LINK_STATUS_SIZE]);

From 4cd073be84c1240b077813cf2766992cd78d8d64 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:16 +0300
Subject: [PATCH 0028/1455] drm/i915/dp: Move intel_dp_training_pattern()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move intel_dp_training_pattern() upwards to avoid the forward
declaration for the POST_LT_ADJ_REQ stuff.

Tested-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-6-ville.syrjala@linux.intel.com
---
 .../drm/i915/display/intel_dp_link_training.c | 114 +++++++++---------
 1 file changed, 57 insertions(+), 57 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 57b9dc52a98d..cb3870427f3f 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -766,6 +766,63 @@ void intel_dp_link_training_set_bw(struct intel_dp *intel_dp,
 	}
 }
 
+/*
+ * Pick Training Pattern Sequence (TPS) for channel equalization. 128b/132b TPS2
+ * for UHBR+, TPS4 for HBR3 or for 1.4 devices that support it, TPS3 for HBR2 or
+ * 1.2 devices that support it, TPS2 otherwise.
+ */
+static u32 intel_dp_training_pattern(struct intel_dp *intel_dp,
+				     const struct intel_crtc_state *crtc_state,
+				     enum drm_dp_phy dp_phy)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+	bool source_tps3, sink_tps3, source_tps4, sink_tps4;
+
+	/* UHBR+ use separate 128b/132b TPS2 */
+	if (intel_dp_is_uhbr(crtc_state))
+		return DP_TRAINING_PATTERN_2;
+
+	/*
+	 * TPS4 support is mandatory for all downstream devices that
+	 * support HBR3. There are no known eDP panels that support
+	 * TPS4 as of Feb 2018 as per VESA eDP_v1.4b_E1 specification.
+	 * LTTPRs must support TPS4.
+	 */
+	source_tps4 = intel_dp_source_supports_tps4(display);
+	sink_tps4 = dp_phy != DP_PHY_DPRX ||
+		    drm_dp_tps4_supported(intel_dp->dpcd);
+	if (source_tps4 && sink_tps4) {
+		return DP_TRAINING_PATTERN_4;
+	} else if (crtc_state->port_clock == 810000) {
+		if (!source_tps4)
+			lt_dbg(intel_dp, dp_phy,
+			       "8.1 Gbps link rate without source TPS4 support\n");
+		if (!sink_tps4)
+			lt_dbg(intel_dp, dp_phy,
+			       "8.1 Gbps link rate without sink TPS4 support\n");
+	}
+
+	/*
+	 * TPS3 support is mandatory for downstream devices that
+	 * support HBR2. However, not all sinks follow the spec.
+	 */
+	source_tps3 = intel_dp_source_supports_tps3(display);
+	sink_tps3 = dp_phy != DP_PHY_DPRX ||
+		    drm_dp_tps3_supported(intel_dp->dpcd);
+	if (source_tps3 && sink_tps3) {
+		return  DP_TRAINING_PATTERN_3;
+	} else if (crtc_state->port_clock >= 540000) {
+		if (!source_tps3)
+			lt_dbg(intel_dp, dp_phy,
+			       ">=5.4/6.48 Gbps link rate without source TPS3 support\n");
+		if (!sink_tps3)
+			lt_dbg(intel_dp, dp_phy,
+			       ">=5.4/6.48 Gbps link rate without sink TPS3 support\n");
+	}
+
+	return DP_TRAINING_PATTERN_2;
+}
+
 static void intel_dp_update_link_bw_set(struct intel_dp *intel_dp,
 					const struct intel_crtc_state *crtc_state,
 					u8 link_bw, u8 rate_select)
@@ -957,63 +1014,6 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp,
 	return false;
 }
 
-/*
- * Pick Training Pattern Sequence (TPS) for channel equalization. 128b/132b TPS2
- * for UHBR+, TPS4 for HBR3 or for 1.4 devices that support it, TPS3 for HBR2 or
- * 1.2 devices that support it, TPS2 otherwise.
- */
-static u32 intel_dp_training_pattern(struct intel_dp *intel_dp,
-				     const struct intel_crtc_state *crtc_state,
-				     enum drm_dp_phy dp_phy)
-{
-	struct intel_display *display = to_intel_display(intel_dp);
-	bool source_tps3, sink_tps3, source_tps4, sink_tps4;
-
-	/* UHBR+ use separate 128b/132b TPS2 */
-	if (intel_dp_is_uhbr(crtc_state))
-		return DP_TRAINING_PATTERN_2;
-
-	/*
-	 * TPS4 support is mandatory for all downstream devices that
-	 * support HBR3. There are no known eDP panels that support
-	 * TPS4 as of Feb 2018 as per VESA eDP_v1.4b_E1 specification.
-	 * LTTPRs must support TPS4.
-	 */
-	source_tps4 = intel_dp_source_supports_tps4(display);
-	sink_tps4 = dp_phy != DP_PHY_DPRX ||
-		    drm_dp_tps4_supported(intel_dp->dpcd);
-	if (source_tps4 && sink_tps4) {
-		return DP_TRAINING_PATTERN_4;
-	} else if (crtc_state->port_clock == 810000) {
-		if (!source_tps4)
-			lt_dbg(intel_dp, dp_phy,
-			       "8.1 Gbps link rate without source TPS4 support\n");
-		if (!sink_tps4)
-			lt_dbg(intel_dp, dp_phy,
-			       "8.1 Gbps link rate without sink TPS4 support\n");
-	}
-
-	/*
-	 * TPS3 support is mandatory for downstream devices that
-	 * support HBR2. However, not all sinks follow the spec.
-	 */
-	source_tps3 = intel_dp_source_supports_tps3(display);
-	sink_tps3 = dp_phy != DP_PHY_DPRX ||
-		    drm_dp_tps3_supported(intel_dp->dpcd);
-	if (source_tps3 && sink_tps3) {
-		return  DP_TRAINING_PATTERN_3;
-	} else if (crtc_state->port_clock >= 540000) {
-		if (!source_tps3)
-			lt_dbg(intel_dp, dp_phy,
-			       ">=5.4/6.48 Gbps link rate without source TPS3 support\n");
-		if (!sink_tps3)
-			lt_dbg(intel_dp, dp_phy,
-			       ">=5.4/6.48 Gbps link rate without sink TPS3 support\n");
-	}
-
-	return DP_TRAINING_PATTERN_2;
-}
-
 /*
  * Perform the link training channel equalization phase on the given DP PHY
  * using one of training pattern 2, 3 or 4 depending on the source and

From 071dcf12bca55c8640d8be5389d0e5251887a174 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:17 +0300
Subject: [PATCH 0029/1455] drm/i915/dp: Implement .set_idle_link_train() for
 everyone
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

All platforms are capable of explicitly transmitting the idle
pattern. Implement it for everyone (so far it as implemented
only for HSW+).

The immediate benefit is that we gain the possibility of
implementing the POST_LT_ADJ_REQ sequence for all platforms.

Another potential future use would be a pseudo port sync mode on
pre-BDW where we attempt to sync up multiple ports/pipes by trying
to turn on the transcoders at the same time, and switching the
links to normal pixel transmission at the same time.

I'm not 100% sure the hardware is guaranteed to transmit the
required number of idle patterns (5) when switching away from
training pattern (either via explicit idle pattern, or straight
to the normal pixel output). Would be nice to confirm that at
some point, but for now let's assume it happens correctly in
both cases.

v2: Elaborate a bit more on the min required idle patterns

Tested-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-7-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/g4x_dp.c | 33 +++++++++++++++++++++++++--
 1 file changed, 31 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c
index b54edf0d1c23..846dbd8ae931 100644
--- a/drivers/gpu/drm/i915/display/g4x_dp.c
+++ b/drivers/gpu/drm/i915/display/g4x_dp.c
@@ -600,6 +600,19 @@ cpt_set_link_train(struct intel_dp *intel_dp,
 	intel_de_posting_read(display, intel_dp->output_reg);
 }
 
+static void
+cpt_set_idle_link_train(struct intel_dp *intel_dp,
+			const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	intel_dp->DP &= ~DP_LINK_TRAIN_MASK_CPT;
+	intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE_CPT;
+
+	intel_de_write(display, intel_dp->output_reg, intel_dp->DP);
+	intel_de_posting_read(display, intel_dp->output_reg);
+}
+
 static void
 g4x_set_link_train(struct intel_dp *intel_dp,
 		   const struct intel_crtc_state *crtc_state,
@@ -628,6 +641,19 @@ g4x_set_link_train(struct intel_dp *intel_dp,
 	intel_de_posting_read(display, intel_dp->output_reg);
 }
 
+static void
+g4x_set_idle_link_train(struct intel_dp *intel_dp,
+			const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	intel_dp->DP &= ~DP_LINK_TRAIN_MASK;
+	intel_dp->DP |= DP_LINK_TRAIN_PAT_IDLE;
+
+	intel_de_write(display, intel_dp->output_reg, intel_dp->DP);
+	intel_de_posting_read(display, intel_dp->output_reg);
+}
+
 static void intel_dp_enable_port(struct intel_dp *intel_dp,
 				 const struct intel_crtc_state *crtc_state)
 {
@@ -1331,10 +1357,13 @@ bool g4x_dp_init(struct intel_display *display,
 	intel_encoder->audio_disable = g4x_dp_audio_disable;
 
 	if ((display->platform.ivybridge && port == PORT_A) ||
-	    (HAS_PCH_CPT(display) && port != PORT_A))
+	    (HAS_PCH_CPT(display) && port != PORT_A)) {
 		dig_port->dp.set_link_train = cpt_set_link_train;
-	else
+		dig_port->dp.set_idle_link_train = cpt_set_idle_link_train;
+	} else {
 		dig_port->dp.set_link_train = g4x_set_link_train;
+		dig_port->dp.set_idle_link_train = g4x_set_idle_link_train;
+	}
 
 	if (display->platform.cherryview)
 		intel_encoder->set_signal_levels = chv_set_signal_levels;

From 976d608d6f8404475ac8e133a713395defa14f41 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 10 Jul 2025 23:17:18 +0300
Subject: [PATCH 0030/1455] drm/i915/dp: Make .set_idle_link_train() mandatory
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Everyone implements the .set_idle_link_train() hook now.
Just make it mandatory.

Tested-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250710201718.25310-8-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_link_training.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index cb3870427f3f..344b74109a83 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -1138,9 +1138,6 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 
 	intel_dp->link.active = true;
 
-	if (!intel_dp->set_idle_link_train)
-		intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
-
 	intel_dp_program_link_training_pattern(intel_dp, crtc_state, DP_PHY_DPRX,
 					       DP_TRAINING_PATTERN_DISABLE);
 
@@ -1381,10 +1378,8 @@ intel_dp_link_train_all_phys(struct intel_dp *intel_dp,
 	if (ret)
 		ret = intel_dp_link_train_phy(intel_dp, crtc_state, DP_PHY_DPRX);
 
-	if (intel_dp->set_idle_link_train) {
-		intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
-		intel_dp->set_idle_link_train(intel_dp, crtc_state);
-	}
+	intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
+	intel_dp->set_idle_link_train(intel_dp, crtc_state);
 
 	return ret;
 }

From 53096728b8910c6916ecc6c46a5abc5c678b58d9 Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Thu, 17 Jul 2025 10:35:55 -0400
Subject: [PATCH 0031/1455] drm: Add DRM prime interface to reassign GEM handle
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CRIU restore of drm buffer objects requires the ability to create
or import a buffer object with a specific gem handle.

Add new drm ioctl DRM_IOCTL_GEM_CHANGE_HANDLE, which takes
the gem handle of an object and moves that object to a
specified new gem handle.

This ioctl needs to call drm_prime_remove_buf_handle,
but that function acquires the prime lock, which the ioctl
needs to hold for other purposes.

Make drm_prime_remove_buf_handle not acquire the prime lock,
and change its other caller to reflect this.

The rest of the kernel patches required to enable CRIU can be
found at
https://lore.kernel.org/dri-devel/20250617194536.538681-1-David.Francis@amd.com/

v2 - Move documentation to UAPI headers
v3 - Always return 0 on success

Signed-off-by: David Francis <David.Francis@amd.com>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250717143556.857893-2-David.Francis@amd.com
---
 drivers/gpu/drm/drm_gem.c      | 56 ++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/drm_internal.h |  4 +++
 drivers/gpu/drm/drm_ioctl.c    |  1 +
 drivers/gpu/drm/drm_prime.c    |  6 +---
 include/uapi/drm/drm.h         | 23 ++++++++++++++
 5 files changed, 85 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 0905ef6786e9..480f91db2e15 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -283,7 +283,12 @@ drm_gem_object_release_handle(int id, void *ptr, void *data)
 	if (obj->funcs->close)
 		obj->funcs->close(obj, file_priv);
 
+	mutex_lock(&file_priv->prime.lock);
+
 	drm_prime_remove_buf_handle(&file_priv->prime, id);
+
+	mutex_unlock(&file_priv->prime.lock);
+
 	drm_vma_node_revoke(&obj->vma_node, file_priv);
 
 	drm_gem_object_handle_put_unlocked(obj);
@@ -934,6 +939,57 @@ err:
 	return ret;
 }
 
+int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_priv)
+{
+	struct drm_gem_change_handle *args = data;
+	struct drm_gem_object *obj;
+	int ret;
+
+	if (!drm_core_check_feature(dev, DRIVER_GEM))
+		return -EOPNOTSUPP;
+
+	obj = drm_gem_object_lookup(file_priv, args->handle);
+	if (!obj)
+		return -ENOENT;
+
+	if (args->handle == args->new_handle)
+		return 0;
+
+	mutex_lock(&file_priv->prime.lock);
+
+	spin_lock(&file_priv->table_lock);
+	ret = idr_alloc(&file_priv->object_idr, obj,
+		args->new_handle, args->new_handle + 1, GFP_NOWAIT);
+	spin_unlock(&file_priv->table_lock);
+
+	if (ret < 0)
+		goto out_unlock;
+
+	if (obj->dma_buf) {
+		ret = drm_prime_add_buf_handle(&file_priv->prime, obj->dma_buf, args->new_handle);
+		if (ret < 0) {
+			spin_lock(&file_priv->table_lock);
+			idr_remove(&file_priv->object_idr, args->new_handle);
+			spin_unlock(&file_priv->table_lock);
+			goto out_unlock;
+		}
+
+		drm_prime_remove_buf_handle(&file_priv->prime, args->handle);
+	}
+
+	ret = 0;
+
+	spin_lock(&file_priv->table_lock);
+	idr_remove(&file_priv->object_idr, args->handle);
+	spin_unlock(&file_priv->table_lock);
+
+out_unlock:
+	mutex_unlock(&file_priv->prime.lock);
+
+	return ret;
+}
+
 /**
  * drm_gem_open - initializes GEM file-private structures at devnode open time
  * @dev: drm_device which is being opened by userspace
diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h
index 9078504e789c..5265eac81077 100644
--- a/drivers/gpu/drm/drm_internal.h
+++ b/drivers/gpu/drm/drm_internal.h
@@ -85,6 +85,8 @@ int drm_prime_fd_to_handle_ioctl(struct drm_device *dev, void *data,
 
 void drm_prime_init_file_private(struct drm_prime_file_private *prime_fpriv);
 void drm_prime_destroy_file_private(struct drm_prime_file_private *prime_fpriv);
+int drm_prime_add_buf_handle(struct drm_prime_file_private *prime_fpriv,
+			     struct dma_buf *dma_buf, uint32_t handle);
 void drm_prime_remove_buf_handle(struct drm_prime_file_private *prime_fpriv,
 				 uint32_t handle);
 
@@ -168,6 +170,8 @@ int drm_gem_close_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
 int drm_gem_flink_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
+int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data,
+				struct drm_file *file_priv);
 int drm_gem_open_ioctl(struct drm_device *dev, void *data,
 		       struct drm_file *file_priv);
 void drm_gem_open(struct drm_device *dev, struct drm_file *file_private);
diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c
index f593dc569d31..d8a24875a7ba 100644
--- a/drivers/gpu/drm/drm_ioctl.c
+++ b/drivers/gpu/drm/drm_ioctl.c
@@ -653,6 +653,7 @@ static const struct drm_ioctl_desc drm_ioctls[] = {
 	DRM_IOCTL_DEF(DRM_IOCTL_GEM_CLOSE, drm_gem_close_ioctl, DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF(DRM_IOCTL_GEM_FLINK, drm_gem_flink_ioctl, DRM_AUTH),
 	DRM_IOCTL_DEF(DRM_IOCTL_GEM_OPEN, drm_gem_open_ioctl, DRM_AUTH),
+	DRM_IOCTL_DEF(DRM_IOCTL_GEM_CHANGE_HANDLE, drm_gem_change_handle_ioctl, DRM_RENDER_ALLOW),
 
 	DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETRESOURCES, drm_mode_getresources, 0),
 
diff --git a/drivers/gpu/drm/drm_prime.c b/drivers/gpu/drm/drm_prime.c
index b703f83874e1..a9784074ba41 100644
--- a/drivers/gpu/drm/drm_prime.c
+++ b/drivers/gpu/drm/drm_prime.c
@@ -93,7 +93,7 @@ struct drm_prime_member {
 	struct rb_node handle_rb;
 };
 
-static int drm_prime_add_buf_handle(struct drm_prime_file_private *prime_fpriv,
+int drm_prime_add_buf_handle(struct drm_prime_file_private *prime_fpriv,
 				    struct dma_buf *dma_buf, uint32_t handle)
 {
 	struct drm_prime_member *member;
@@ -190,8 +190,6 @@ void drm_prime_remove_buf_handle(struct drm_prime_file_private *prime_fpriv,
 {
 	struct rb_node *rb;
 
-	mutex_lock(&prime_fpriv->lock);
-
 	rb = prime_fpriv->handles.rb_node;
 	while (rb) {
 		struct drm_prime_member *member;
@@ -210,8 +208,6 @@ void drm_prime_remove_buf_handle(struct drm_prime_file_private *prime_fpriv,
 			rb = rb->rb_left;
 		}
 	}
-
-	mutex_unlock(&prime_fpriv->lock);
 }
 
 void drm_prime_init_file_private(struct drm_prime_file_private *prime_fpriv)
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index e63a71d3c607..7fa123e11c3f 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -625,6 +625,21 @@ struct drm_gem_open {
 	__u64 size;
 };
 
+/**
+ * struct drm_gem_change_handle - Argument for &DRM_IOCTL_GEM_CHANGE_HANDLE ioctl.
+ * @handle: The handle of a gem object.
+ * @new_handle: An available gem handle.
+ *
+ * This ioctl changes the handle of a GEM object to the specified one.
+ * The new handle must be unused. On success the old handle is closed
+ * and all further IOCTL should refer to the new handle only.
+ * Calls to DRM_IOCTL_PRIME_FD_TO_HANDLE will return the new handle.
+ */
+struct drm_gem_change_handle {
+	__u32 handle;
+	__u32 new_handle;
+};
+
 /**
  * DRM_CAP_DUMB_BUFFER
  *
@@ -1309,6 +1324,14 @@ extern "C" {
  */
 #define DRM_IOCTL_SET_CLIENT_NAME	DRM_IOWR(0xD1, struct drm_set_client_name)
 
+/**
+ * DRM_IOCTL_GEM_CHANGE_HANDLE - Move an object to a different handle
+ *
+ * Some applications (notably CRIU) need objects to have specific gem handles.
+ * This ioctl changes the object at one gem handle to use a new gem handle.
+ */
+#define DRM_IOCTL_GEM_CHANGE_HANDLE    DRM_IOWR(0xD2, struct drm_gem_change_handle)
+
 /*
  * Device specific ioctls should only be in their respective headers
  * The device specific ioctl range is from 0x40 to 0x9f.

From 0864197382fa7c8c2641bff0f36355bf4bd76398 Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Thu, 17 Jul 2025 10:35:56 -0400
Subject: [PATCH 0032/1455] drm: Move drm_gem ioctl kerneldoc to uapi file
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The drm_gem ioctls were documented in internal file drm_gem.c
instead of uapi header drm.h. Move them there and change to
appropriate kerneldoc formatting.

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Simona Vetter <simona.vetter@ffwll.ch>
Signed-off-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250717143556.857893-3-David.Francis@amd.com
---
 drivers/gpu/drm/drm_gem.c | 30 -----------------------------
 include/uapi/drm/drm.h    | 40 +++++++++++++++++++++++++++------------
 2 files changed, 28 insertions(+), 42 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 480f91db2e15..9b621a94c2a9 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -821,14 +821,6 @@ long drm_gem_dma_resv_wait(struct drm_file *filep, u32 handle,
 }
 EXPORT_SYMBOL(drm_gem_dma_resv_wait);
 
-/**
- * drm_gem_close_ioctl - implementation of the GEM_CLOSE ioctl
- * @dev: drm_device
- * @data: ioctl data
- * @file_priv: drm file-private structure
- *
- * Releases the handle to an mm object.
- */
 int
 drm_gem_close_ioctl(struct drm_device *dev, void *data,
 		    struct drm_file *file_priv)
@@ -844,17 +836,6 @@ drm_gem_close_ioctl(struct drm_device *dev, void *data,
 	return ret;
 }
 
-/**
- * drm_gem_flink_ioctl - implementation of the GEM_FLINK ioctl
- * @dev: drm_device
- * @data: ioctl data
- * @file_priv: drm file-private structure
- *
- * Create a global name for an object, returning the name.
- *
- * Note that the name does not hold a reference; when the object
- * is freed, the name goes away.
- */
 int
 drm_gem_flink_ioctl(struct drm_device *dev, void *data,
 		    struct drm_file *file_priv)
@@ -894,17 +875,6 @@ err:
 	return ret;
 }
 
-/**
- * drm_gem_open_ioctl - implementation of the GEM_OPEN ioctl
- * @dev: drm_device
- * @data: ioctl data
- * @file_priv: drm file-private structure
- *
- * Open an object using the global name, returning a handle and the size.
- *
- * This handle (of course) holds a reference to the object, so the object
- * will not go away until the handle is deleted.
- */
 int
 drm_gem_open_ioctl(struct drm_device *dev, void *data,
 		   struct drm_file *file_priv)
diff --git a/include/uapi/drm/drm.h b/include/uapi/drm/drm.h
index 7fa123e11c3f..3cd5cf15e3c9 100644
--- a/include/uapi/drm/drm.h
+++ b/include/uapi/drm/drm.h
@@ -597,31 +597,47 @@ struct drm_set_version {
 	int drm_dd_minor;
 };
 
-/* DRM_IOCTL_GEM_CLOSE ioctl argument type */
+/**
+ * struct drm_gem_close - Argument for &DRM_IOCTL_GEM_CLOSE ioctl.
+ * @handle: Handle of the object to be closed.
+ * @pad: Padding.
+ *
+ * Releases the handle to an mm object.
+ */
 struct drm_gem_close {
-	/** Handle of the object to be closed. */
 	__u32 handle;
 	__u32 pad;
 };
 
-/* DRM_IOCTL_GEM_FLINK ioctl argument type */
+/**
+ * struct drm_gem_flink - Argument for &DRM_IOCTL_GEM_FLINK ioctl.
+ * @handle: Handle for the object being named.
+ * @name: Returned global name.
+ *
+ * Create a global name for an object, returning the name.
+ *
+ * Note that the name does not hold a reference; when the object
+ * is freed, the name goes away.
+ */
 struct drm_gem_flink {
-	/** Handle for the object being named */
 	__u32 handle;
-
-	/** Returned global name */
 	__u32 name;
 };
 
-/* DRM_IOCTL_GEM_OPEN ioctl argument type */
+/**
+ * struct drm_gem_open - Argument for &DRM_IOCTL_GEM_OPEN ioctl.
+ * @name: Name of object being opened.
+ * @handle: Returned handle for the object.
+ * @size: Returned size of the object
+ *
+ * Open an object using the global name, returning a handle and the size.
+ *
+ * This handle (of course) holds a reference to the object, so the object
+ * will not go away until the handle is deleted.
+ */
 struct drm_gem_open {
-	/** Name of object being opened */
 	__u32 name;
-
-	/** Returned handle for the object */
 	__u32 handle;
-
-	/** Returned size of the object */
 	__u64 size;
 };
 

From 36caa026b28a49aa91495698350ce46577293e9d Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Wed, 16 Jul 2025 09:51:17 +0100
Subject: [PATCH 0033/1455] drm/sched: Avoid double re-lock on the job free
 path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently the job free work item will lock sched->job_list_lock first time
to see if there are any jobs, free a single job, and then lock again to
decide whether to re-queue itself if there are more finished jobs.

Since drm_sched_get_finished_job() already looks at the second job in the
queue we can simply add the signaled check and have it return the presence
of more jobs to be freed to the caller. That way the work item does not
have to lock the list again and repeat the signaled check.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Maíra Canal <mcanal@igalia.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Reviewed-by: Maíra Canal <mcanal@igalia.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250716085117.56864-1-tvrtko.ursulin@igalia.com
---
 drivers/gpu/drm/scheduler/sched_main.c | 53 ++++++++++----------------
 1 file changed, 21 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index e2cda28a1af4..5a550fd76bf0 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -349,34 +349,13 @@ static void drm_sched_run_job_queue(struct drm_gpu_scheduler *sched)
 }
 
 /**
- * __drm_sched_run_free_queue - enqueue free-job work
- * @sched: scheduler instance
- */
-static void __drm_sched_run_free_queue(struct drm_gpu_scheduler *sched)
-{
-	if (!READ_ONCE(sched->pause_submit))
-		queue_work(sched->submit_wq, &sched->work_free_job);
-}
-
-/**
- * drm_sched_run_free_queue - enqueue free-job work if ready
+ * drm_sched_run_free_queue - enqueue free-job work
  * @sched: scheduler instance
  */
 static void drm_sched_run_free_queue(struct drm_gpu_scheduler *sched)
 {
-	struct drm_sched_job *job;
-
-	job = list_first_entry_or_null(&sched->pending_list,
-				       struct drm_sched_job, list);
-	if (job && dma_fence_is_signaled(&job->s_fence->finished))
-		__drm_sched_run_free_queue(sched);
-}
-
-static void drm_sched_run_free_queue_unlocked(struct drm_gpu_scheduler *sched)
-{
-	spin_lock(&sched->job_list_lock);
-	drm_sched_run_free_queue(sched);
-	spin_unlock(&sched->job_list_lock);
+	if (!READ_ONCE(sched->pause_submit))
+		queue_work(sched->submit_wq, &sched->work_free_job);
 }
 
 /**
@@ -398,7 +377,7 @@ static void drm_sched_job_done(struct drm_sched_job *s_job, int result)
 	dma_fence_get(&s_fence->finished);
 	drm_sched_fence_finished(s_fence, result);
 	dma_fence_put(&s_fence->finished);
-	__drm_sched_run_free_queue(sched);
+	drm_sched_run_free_queue(sched);
 }
 
 /**
@@ -1134,12 +1113,16 @@ drm_sched_select_entity(struct drm_gpu_scheduler *sched)
  * drm_sched_get_finished_job - fetch the next finished job to be destroyed
  *
  * @sched: scheduler instance
+ * @have_more: are there more finished jobs on the list
+ *
+ * Informs the caller through @have_more whether there are more finished jobs
+ * besides the returned one.
  *
  * Returns the next finished job from the pending list (if there is one)
  * ready for it to be destroyed.
  */
 static struct drm_sched_job *
-drm_sched_get_finished_job(struct drm_gpu_scheduler *sched)
+drm_sched_get_finished_job(struct drm_gpu_scheduler *sched, bool *have_more)
 {
 	struct drm_sched_job *job, *next;
 
@@ -1147,22 +1130,25 @@ drm_sched_get_finished_job(struct drm_gpu_scheduler *sched)
 
 	job = list_first_entry_or_null(&sched->pending_list,
 				       struct drm_sched_job, list);
-
 	if (job && dma_fence_is_signaled(&job->s_fence->finished)) {
 		/* remove job from pending_list */
 		list_del_init(&job->list);
 
 		/* cancel this job's TO timer */
 		cancel_delayed_work(&sched->work_tdr);
-		/* make the scheduled timestamp more accurate */
+
+		*have_more = false;
 		next = list_first_entry_or_null(&sched->pending_list,
 						typeof(*next), list);
-
 		if (next) {
+			/* make the scheduled timestamp more accurate */
 			if (test_bit(DMA_FENCE_FLAG_TIMESTAMP_BIT,
 				     &next->s_fence->scheduled.flags))
 				next->s_fence->scheduled.timestamp =
 					dma_fence_timestamp(&job->s_fence->finished);
+
+			*have_more = dma_fence_is_signaled(&next->s_fence->finished);
+
 			/* start TO timer for next job */
 			drm_sched_start_timeout(sched);
 		}
@@ -1221,12 +1207,15 @@ static void drm_sched_free_job_work(struct work_struct *w)
 	struct drm_gpu_scheduler *sched =
 		container_of(w, struct drm_gpu_scheduler, work_free_job);
 	struct drm_sched_job *job;
+	bool have_more;
 
-	job = drm_sched_get_finished_job(sched);
-	if (job)
+	job = drm_sched_get_finished_job(sched, &have_more);
+	if (job) {
 		sched->ops->free_job(job);
+		if (have_more)
+			drm_sched_run_free_queue(sched);
+	}
 
-	drm_sched_run_free_queue_unlocked(sched);
 	drm_sched_run_job_queue(sched);
 }
 

From ca2a6abdaee43808034cdb218428d2ed85fd3db8 Mon Sep 17 00:00:00 2001
From: Jann Horn <jannh@google.com>
Date: Wed, 13 Nov 2024 22:03:39 +0100
Subject: [PATCH 0034/1455] drm/panthor: Fix memory leak in
 panthor_ioctl_group_create()

When bailing out due to group_priority_permit() failure, the queue_args
need to be freed. Fix it by rearranging the function to use the
goto-on-error pattern, such that the success case flows straight without
indentation while error cases jump forward to cleanup.

Cc: stable@vger.kernel.org
Fixes: 5f7762042f8a ("drm/panthor: Restrict high priorities on group_create")
Signed-off-by: Jann Horn <jannh@google.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20241113-panthor-fix-gcq-bailout-v1-1-654307254d68@google.com
---
 drivers/gpu/drm/panthor/panthor_drv.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_drv.c b/drivers/gpu/drm/panthor/panthor_drv.c
index 1116f2d2826e..0304eb83d0b2 100644
--- a/drivers/gpu/drm/panthor/panthor_drv.c
+++ b/drivers/gpu/drm/panthor/panthor_drv.c
@@ -1103,14 +1103,15 @@ static int panthor_ioctl_group_create(struct drm_device *ddev, void *data,
 
 	ret = group_priority_permit(file, args->priority);
 	if (ret)
-		return ret;
+		goto out;
 
 	ret = panthor_group_create(pfile, args, queue_args);
-	if (ret >= 0) {
-		args->group_handle = ret;
-		ret = 0;
-	}
+	if (ret < 0)
+		goto out;
+	args->group_handle = ret;
+	ret = 0;
 
+out:
 	kvfree(queue_args);
 	return ret;
 }

From 4b5514f786dd790ec0d3e2fe66cea7fd2f36bdf3 Mon Sep 17 00:00:00 2001
From: Harish Chegondi <harish.chegondi@intel.com>
Date: Fri, 18 Jul 2025 13:29:47 -0700
Subject: [PATCH 0035/1455] drm/xe: Remove unnecessary EU stall debug message

The EU stall debug message may cause CI to complain on
unsupported platforms. Remove it.

Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Link: https://lore.kernel.org/r/dfb6a080b3442d481c567489aabe47e72f3e784c.1752870172.git.harish.chegondi@intel.com
---
 drivers/gpu/drm/xe/xe_query.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
index 56ef3ee2806f..44d44bbc71dc 100644
--- a/drivers/gpu/drm/xe/xe_query.c
+++ b/drivers/gpu/drm/xe/xe_query.c
@@ -749,10 +749,8 @@ static int query_eu_stall(struct xe_device *xe,
 	u32 num_rates;
 	int ret;
 
-	if (!xe_eu_stall_supported_on_platform(xe)) {
-		drm_dbg(&xe->drm, "EU stall monitoring is not supported on this platform\n");
+	if (!xe_eu_stall_supported_on_platform(xe))
 		return -ENODEV;
-	}
 
 	array_size = xe_eu_stall_get_sampling_rates(&num_rates, &rates);
 	size = sizeof(struct drm_xe_query_eu_stall) + array_size;

From 615c8ec48621f5c157fef3633d6e4be81d7a370f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Fri, 18 Jul 2025 14:29:28 +0300
Subject: [PATCH 0036/1455] drm/i915/dsi: Don't set/read the DSI C clock
 divider on GLK
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

GLK doesn't use the DSI C clock at all, no need to program
the divider for it. Bspec even says: "Do not program this field".

However looks like some firmware versions program this and
some do not. In order to avoid bogus fastset mismatches
we should also filter it out during readout.

v2: Clear all the DSI C clock bits during readout (Jani)
    Adjust platform checks for new style, and add
    has_dsic_clock() while at it.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250718112928.27669-1-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/vlv_dsi_pll.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
index d42b61e6f076..b52463fdec47 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
@@ -262,6 +262,11 @@ void vlv_dsi_pll_disable(struct intel_encoder *encoder)
 	vlv_cck_put(display->drm);
 }
 
+static bool has_dsic_clock(struct intel_display *display)
+{
+	return display->platform.broxton;
+}
+
 bool bxt_dsi_pll_is_enabled(struct intel_display *display)
 {
 	bool enabled;
@@ -284,7 +289,7 @@ bool bxt_dsi_pll_is_enabled(struct intel_display *display)
 	 * causes a system hang.
 	 */
 	val = intel_de_read(display, BXT_DSI_PLL_CTL);
-	if (display->platform.geminilake) {
+	if (!has_dsic_clock(display)) {
 		if (!(val & BXT_DSIA_16X_MASK)) {
 			drm_dbg_kms(display->drm,
 				    "Invalid PLL divider (%08x)\n", val);
@@ -358,6 +363,8 @@ u32 bxt_dsi_get_pclk(struct intel_encoder *encoder,
 	u32 pclk;
 
 	config->dsi_pll.ctrl = intel_de_read(display, BXT_DSI_PLL_CTL);
+	if (!has_dsic_clock(display))
+		config->dsi_pll.ctrl &= ~BXT_DSIC_16X_MASK;
 
 	pclk = bxt_dsi_pclk(encoder, config);
 
@@ -514,7 +521,9 @@ int bxt_dsi_pll_compute(struct intel_encoder *encoder,
 	 * Spec says both have to be programmed, even if one is not getting
 	 * used. Configure MIPI_CLOCK_CTL dividers in modeset
 	 */
-	config->dsi_pll.ctrl = dsi_ratio | BXT_DSIA_16X_BY2 | BXT_DSIC_16X_BY2;
+	config->dsi_pll.ctrl = dsi_ratio | BXT_DSIA_16X_BY2;
+	if (has_dsic_clock(display))
+		config->dsi_pll.ctrl |= BXT_DSIC_16X_BY2;
 
 	/* As per recommendation from hardware team,
 	 * Prog PVD ratio =1 if dsi ratio <= 50

From 971eb92b0460ca2afe581e60e3c8d863563ae38c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 17 Jul 2025 23:32:16 +0300
Subject: [PATCH 0037/1455] drm/i915: Precompute plane SURF address
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we pre-compute the plane surface/base address
partially (only for cursor_needs_physical cases) in
intel_plane_pin_fb() and finish the calculation in the
plane->update_arm(). Let's just precompute the whole thing
instead.

One benefit is that we get rid of all the vma offset stuff
from the low level plane code. Another use I have in mind
is including the surface address in the plane tracepoints,
which should make it easier to analyze display faults.

v2: Deal with xe reuse_vma() hacks
v3: use intel_plane_ggtt_offset() still in reuse_vma()

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250717203216.31258-1-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/i9xx_plane.c     | 53 ++++++++++---------
 drivers/gpu/drm/i915/display/i9xx_plane.h     |  1 +
 drivers/gpu/drm/i915/display/intel_cursor.c   | 18 +++----
 .../drm/i915/display/intel_display_types.h    |  5 +-
 drivers/gpu/drm/i915/display/intel_fb_pin.c   | 27 ++++++----
 .../drm/i915/display/intel_plane_initial.c    |  2 +
 drivers/gpu/drm/i915/display/intel_sprite.c   | 36 ++++++-------
 .../drm/i915/display/skl_universal_plane.c    | 43 +++++++--------
 drivers/gpu/drm/xe/display/xe_fb_pin.c        | 10 ++++
 drivers/gpu/drm/xe/display/xe_plane_initial.c |  3 ++
 10 files changed, 104 insertions(+), 94 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index f291ced989dc..d572e95c3c92 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -360,6 +360,19 @@ i9xx_plane_check(struct intel_crtc_state *crtc_state,
 	return 0;
 }
 
+static u32 i8xx_plane_surf_offset(const struct intel_plane_state *plane_state)
+{
+	int x = plane_state->view.color_plane[0].x;
+	int y = plane_state->view.color_plane[0].y;
+
+	return intel_fb_xy_to_linear(x, y, plane_state, 0);
+}
+
+u32 i965_plane_surf_offset(const struct intel_plane_state *plane_state)
+{
+	return plane_state->view.color_plane[0].offset;
+}
+
 static u32 i9xx_plane_ctl_crtc(const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(crtc_state);
@@ -463,7 +476,7 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb,
 	enum i9xx_plane_id i9xx_plane = plane->i9xx_plane;
 	int x = plane_state->view.color_plane[0].x;
 	int y = plane_state->view.color_plane[0].y;
-	u32 dspcntr, dspaddr_offset, linear_offset;
+	u32 dspcntr;
 
 	dspcntr = plane_state->ctl | i9xx_plane_ctl_crtc(crtc_state);
 
@@ -472,13 +485,6 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb,
 	    crtc_state->async_flip_planes & BIT(plane->id))
 		dspcntr |= DISP_ASYNC_FLIP;
 
-	linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
-
-	if (DISPLAY_VER(display) >= 4)
-		dspaddr_offset = plane_state->view.color_plane[0].offset;
-	else
-		dspaddr_offset = linear_offset;
-
 	if (display->platform.cherryview && i9xx_plane == PLANE_B) {
 		int crtc_x = plane_state->uapi.dst.x1;
 		int crtc_y = plane_state->uapi.dst.y1;
@@ -498,7 +504,7 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb,
 				  DISP_OFFSET_Y(y) | DISP_OFFSET_X(x));
 	} else if (DISPLAY_VER(display) >= 4) {
 		intel_de_write_fw(display, DSPLINOFF(display, i9xx_plane),
-				  linear_offset);
+				  intel_fb_xy_to_linear(x, y, plane_state, 0));
 		intel_de_write_fw(display, DSPTILEOFF(display, i9xx_plane),
 				  DISP_OFFSET_Y(y) | DISP_OFFSET_X(x));
 	}
@@ -511,11 +517,9 @@ static void i9xx_plane_update_arm(struct intel_dsb *dsb,
 	intel_de_write_fw(display, DSPCNTR(display, i9xx_plane), dspcntr);
 
 	if (DISPLAY_VER(display) >= 4)
-		intel_de_write_fw(display, DSPSURF(display, i9xx_plane),
-				  intel_plane_ggtt_offset(plane_state) + dspaddr_offset);
+		intel_de_write_fw(display, DSPSURF(display, i9xx_plane), plane_state->surf);
 	else
-		intel_de_write_fw(display, DSPADDR(display, i9xx_plane),
-				  intel_plane_ggtt_offset(plane_state) + dspaddr_offset);
+		intel_de_write_fw(display, DSPADDR(display, i9xx_plane), plane_state->surf);
 }
 
 static void i830_plane_update_arm(struct intel_dsb *dsb,
@@ -604,16 +608,13 @@ g4x_primary_async_flip(struct intel_dsb *dsb,
 {
 	struct intel_display *display = to_intel_display(plane);
 	u32 dspcntr = plane_state->ctl | i9xx_plane_ctl_crtc(crtc_state);
-	u32 dspaddr_offset = plane_state->view.color_plane[0].offset;
 	enum i9xx_plane_id i9xx_plane = plane->i9xx_plane;
 
 	if (async_flip)
 		dspcntr |= DISP_ASYNC_FLIP;
 
 	intel_de_write_fw(display, DSPCNTR(display, i9xx_plane), dspcntr);
-
-	intel_de_write_fw(display, DSPSURF(display, i9xx_plane),
-			  intel_plane_ggtt_offset(plane_state) + dspaddr_offset);
+	intel_de_write_fw(display, DSPSURF(display, i9xx_plane), plane_state->surf);
 }
 
 static void
@@ -624,11 +625,9 @@ vlv_primary_async_flip(struct intel_dsb *dsb,
 		       bool async_flip)
 {
 	struct intel_display *display = to_intel_display(plane);
-	u32 dspaddr_offset = plane_state->view.color_plane[0].offset;
 	enum i9xx_plane_id i9xx_plane = plane->i9xx_plane;
 
-	intel_de_write_fw(display, DSPADDR_VLV(display, i9xx_plane),
-			  intel_plane_ggtt_offset(plane_state) + dspaddr_offset);
+	intel_de_write_fw(display, DSPADDR_VLV(display, i9xx_plane), plane_state->surf);
 }
 
 static void
@@ -1037,6 +1036,11 @@ intel_primary_plane_create(struct intel_display *display, enum pipe pipe)
 	plane->get_hw_state = i9xx_plane_get_hw_state;
 	plane->check_plane = i9xx_plane_check;
 
+	if (DISPLAY_VER(display) >= 4)
+		plane->surf_offset = i965_plane_surf_offset;
+	else
+		plane->surf_offset = i8xx_plane_surf_offset;
+
 	if (DISPLAY_VER(display) >= 5 || display->platform.g4x)
 		plane->capture_error = g4x_primary_capture_error;
 	else if (DISPLAY_VER(display) >= 4)
@@ -1254,24 +1258,21 @@ bool i9xx_fixup_initial_plane_config(struct intel_crtc *crtc,
 	const struct intel_plane_state *plane_state =
 		to_intel_plane_state(plane->base.state);
 	enum i9xx_plane_id i9xx_plane = plane->i9xx_plane;
-	u32 base;
 
 	if (!plane_state->uapi.visible)
 		return false;
 
-	base = intel_plane_ggtt_offset(plane_state);
-
 	/*
 	 * We may have moved the surface to a different
 	 * part of ggtt, make the plane aware of that.
 	 */
-	if (plane_config->base == base)
+	if (plane_config->base == plane_state->surf)
 		return false;
 
 	if (DISPLAY_VER(display) >= 4)
-		intel_de_write(display, DSPSURF(display, i9xx_plane), base);
+		intel_de_write(display, DSPSURF(display, i9xx_plane), plane_state->surf);
 	else
-		intel_de_write(display, DSPADDR(display, i9xx_plane), base);
+		intel_de_write(display, DSPADDR(display, i9xx_plane), plane_state->surf);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.h b/drivers/gpu/drm/i915/display/i9xx_plane.h
index d90546d60855..565dab751301 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.h
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.h
@@ -24,6 +24,7 @@ unsigned int vlv_plane_min_alignment(struct intel_plane *plane,
 				     const struct drm_framebuffer *fb,
 				     int colot_plane);
 int i9xx_check_plane_surface(struct intel_plane_state *plane_state);
+u32 i965_plane_surf_offset(const struct intel_plane_state *plane_state);
 
 struct intel_plane *
 intel_primary_plane_create(struct intel_display *display, enum pipe pipe);
diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c
index 198e69efe9ac..83718b627bb0 100644
--- a/drivers/gpu/drm/i915/display/intel_cursor.c
+++ b/drivers/gpu/drm/i915/display/intel_cursor.c
@@ -33,17 +33,9 @@ static const u32 intel_cursor_formats[] = {
 	DRM_FORMAT_ARGB8888,
 };
 
-static u32 intel_cursor_base(const struct intel_plane_state *plane_state)
+static u32 intel_cursor_surf_offset(const struct intel_plane_state *plane_state)
 {
-	struct intel_display *display = to_intel_display(plane_state);
-	u32 base;
-
-	if (DISPLAY_INFO(display)->cursor_needs_physical)
-		base = plane_state->phys_dma_addr;
-	else
-		base = intel_plane_ggtt_offset(plane_state);
-
-	return base + plane_state->view.color_plane[0].offset;
+	return plane_state->view.color_plane[0].offset;
 }
 
 static u32 intel_cursor_position(const struct intel_crtc_state *crtc_state,
@@ -297,7 +289,7 @@ static void i845_cursor_update_arm(struct intel_dsb *dsb,
 
 		size = CURSOR_HEIGHT(height) | CURSOR_WIDTH(width);
 
-		base = intel_cursor_base(plane_state);
+		base = plane_state->surf;
 		pos = intel_cursor_position(crtc_state, plane_state, false);
 	}
 
@@ -675,7 +667,7 @@ static void i9xx_cursor_update_arm(struct intel_dsb *dsb,
 		if (width != height)
 			fbc_ctl = CUR_FBC_EN | CUR_FBC_HEIGHT(height - 1);
 
-		base = intel_cursor_base(plane_state);
+		base = plane_state->surf;
 		pos = intel_cursor_position(crtc_state, plane_state, false);
 	}
 
@@ -1051,6 +1043,8 @@ intel_cursor_plane_create(struct intel_display *display,
 		cursor->check_plane = i9xx_check_cursor;
 	}
 
+	cursor->surf_offset = intel_cursor_surf_offset;
+
 	if (DISPLAY_VER(display) >= 5 || display->platform.g4x)
 		cursor->capture_error = g4x_cursor_capture_error;
 	else
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index ce45261c4a8f..8f8019d40d77 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -642,7 +642,6 @@ struct intel_plane_state {
 #define PLANE_HAS_FENCE BIT(0)
 
 	struct intel_fb_view view;
-	u32 phys_dma_addr; /* for cursor_needs_physical */
 
 	/* for legacy cursor fb unpin */
 	struct drm_vblank_work unpin_work;
@@ -665,6 +664,9 @@ struct intel_plane_state {
 	/* chroma upsampler control register */
 	u32 cus_ctl;
 
+	/* surface address register */
+	u32 surf;
+
 	/*
 	 * scaler_id
 	 *    = -1 : not using a scaler
@@ -1534,6 +1536,7 @@ struct intel_plane {
 	bool (*get_hw_state)(struct intel_plane *plane, enum pipe *pipe);
 	int (*check_plane)(struct intel_crtc_state *crtc_state,
 			   struct intel_plane_state *plane_state);
+	u32 (*surf_offset)(const struct intel_plane_state *plane_state);
 	int (*min_cdclk)(const struct intel_crtc_state *crtc_state,
 			 const struct intel_plane_state *plane_state);
 	void (*async_flip)(struct intel_dsb *dsb,
diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index 5a0151775a3a..5401d5b27e30 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -277,17 +277,6 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state,
 
 		plane_state->ggtt_vma = vma;
 
-		/*
-		 * Pre-populate the dma address before we enter the vblank
-		 * evade critical section as i915_gem_object_get_dma_address()
-		 * will trigger might_sleep() even if it won't actually sleep,
-		 * which is the case when the fb has already been pinned.
-		 */
-		if (intel_plane_needs_physical(plane)) {
-			struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base));
-
-			plane_state->phys_dma_addr = i915_gem_object_get_dma_address(obj, 0);
-		}
 	} else {
 		unsigned int alignment = intel_plane_fb_min_alignment(plane_state);
 
@@ -311,6 +300,22 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state,
 		WARN_ON(plane_state->ggtt_vma == plane_state->dpt_vma);
 	}
 
+	/*
+	 * Pre-populate the dma address before we enter the vblank
+	 * evade critical section as i915_gem_object_get_dma_address()
+	 * will trigger might_sleep() even if it won't actually sleep,
+	 * which is the case when the fb has already been pinned.
+	 */
+	if (intel_plane_needs_physical(plane)) {
+		struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base));
+
+		plane_state->surf = i915_gem_object_get_dma_address(obj, 0) +
+			plane->surf_offset(plane_state);
+	} else {
+		plane_state->surf = intel_plane_ggtt_offset(plane_state) +
+			plane->surf_offset(plane_state);
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c
index 2194d39a5c98..c506b5d660c3 100644
--- a/drivers/gpu/drm/i915/display/intel_plane_initial.c
+++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c
@@ -359,6 +359,8 @@ valid_fb:
 	    i915_vma_pin_fence(vma) == 0 && vma->fence)
 		plane_state->flags |= PLANE_HAS_FENCE;
 
+	plane_state->surf = intel_plane_ggtt_offset(plane_state);
+
 	plane_state->uapi.src_x = 0;
 	plane_state->uapi.src_y = 0;
 	plane_state->uapi.src_w = fb->width << 16;
diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c
index e6844df837af..230f90e79d2a 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite.c
@@ -395,15 +395,12 @@ vlv_sprite_update_arm(struct intel_dsb *dsb,
 	enum pipe pipe = plane->pipe;
 	enum plane_id plane_id = plane->id;
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
-	u32 sprsurf_offset = plane_state->view.color_plane[0].offset;
 	u32 x = plane_state->view.color_plane[0].x;
 	u32 y = plane_state->view.color_plane[0].y;
-	u32 sprctl, linear_offset;
+	u32 sprctl;
 
 	sprctl = plane_state->ctl | vlv_sprite_ctl_crtc(crtc_state);
 
-	linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
-
 	if (display->platform.cherryview && pipe == PIPE_B)
 		chv_sprite_update_csc(plane_state);
 
@@ -418,7 +415,8 @@ vlv_sprite_update_arm(struct intel_dsb *dsb,
 
 	intel_de_write_fw(display, SPCONSTALPHA(pipe, plane_id), 0);
 
-	intel_de_write_fw(display, SPLINOFF(pipe, plane_id), linear_offset);
+	intel_de_write_fw(display, SPLINOFF(pipe, plane_id),
+			  intel_fb_xy_to_linear(x, y, plane_state, 0));
 	intel_de_write_fw(display, SPTILEOFF(pipe, plane_id),
 			  SP_OFFSET_Y(y) | SP_OFFSET_X(x));
 
@@ -428,8 +426,7 @@ vlv_sprite_update_arm(struct intel_dsb *dsb,
 	 * the control register just before the surface register.
 	 */
 	intel_de_write_fw(display, SPCNTR(pipe, plane_id), sprctl);
-	intel_de_write_fw(display, SPSURF(pipe, plane_id),
-			  intel_plane_ggtt_offset(plane_state) + sprsurf_offset);
+	intel_de_write_fw(display, SPSURF(pipe, plane_id), plane_state->surf);
 
 	vlv_sprite_update_clrc(plane_state);
 	vlv_sprite_update_gamma(plane_state);
@@ -830,15 +827,12 @@ ivb_sprite_update_arm(struct intel_dsb *dsb,
 	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
-	u32 sprsurf_offset = plane_state->view.color_plane[0].offset;
 	u32 x = plane_state->view.color_plane[0].x;
 	u32 y = plane_state->view.color_plane[0].y;
-	u32 sprctl, linear_offset;
+	u32 sprctl;
 
 	sprctl = plane_state->ctl | ivb_sprite_ctl_crtc(crtc_state);
 
-	linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
-
 	if (key->flags) {
 		intel_de_write_fw(display, SPRKEYVAL(pipe), key->min_value);
 		intel_de_write_fw(display, SPRKEYMSK(pipe),
@@ -852,7 +846,8 @@ ivb_sprite_update_arm(struct intel_dsb *dsb,
 		intel_de_write_fw(display, SPROFFSET(pipe),
 				  SPRITE_OFFSET_Y(y) | SPRITE_OFFSET_X(x));
 	} else {
-		intel_de_write_fw(display, SPRLINOFF(pipe), linear_offset);
+		intel_de_write_fw(display, SPRLINOFF(pipe),
+				  intel_fb_xy_to_linear(x, y, plane_state, 0));
 		intel_de_write_fw(display, SPRTILEOFF(pipe),
 				  SPRITE_OFFSET_Y(y) | SPRITE_OFFSET_X(x));
 	}
@@ -863,8 +858,7 @@ ivb_sprite_update_arm(struct intel_dsb *dsb,
 	 * the control register just before the surface register.
 	 */
 	intel_de_write_fw(display, SPRCTL(pipe), sprctl);
-	intel_de_write_fw(display, SPRSURF(pipe),
-			  intel_plane_ggtt_offset(plane_state) + sprsurf_offset);
+	intel_de_write_fw(display, SPRSURF(pipe), plane_state->surf);
 
 	ivb_sprite_update_gamma(plane_state);
 }
@@ -1181,15 +1175,12 @@ g4x_sprite_update_arm(struct intel_dsb *dsb,
 	struct intel_display *display = to_intel_display(plane);
 	enum pipe pipe = plane->pipe;
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
-	u32 dvssurf_offset = plane_state->view.color_plane[0].offset;
 	u32 x = plane_state->view.color_plane[0].x;
 	u32 y = plane_state->view.color_plane[0].y;
-	u32 dvscntr, linear_offset;
+	u32 dvscntr;
 
 	dvscntr = plane_state->ctl | g4x_sprite_ctl_crtc(crtc_state);
 
-	linear_offset = intel_fb_xy_to_linear(x, y, plane_state, 0);
-
 	if (key->flags) {
 		intel_de_write_fw(display, DVSKEYVAL(pipe), key->min_value);
 		intel_de_write_fw(display, DVSKEYMSK(pipe),
@@ -1197,7 +1188,8 @@ g4x_sprite_update_arm(struct intel_dsb *dsb,
 		intel_de_write_fw(display, DVSKEYMAX(pipe), key->max_value);
 	}
 
-	intel_de_write_fw(display, DVSLINOFF(pipe), linear_offset);
+	intel_de_write_fw(display, DVSLINOFF(pipe),
+			  intel_fb_xy_to_linear(x, y, plane_state, 0));
 	intel_de_write_fw(display, DVSTILEOFF(pipe),
 			  DVS_OFFSET_Y(y) | DVS_OFFSET_X(x));
 
@@ -1207,8 +1199,7 @@ g4x_sprite_update_arm(struct intel_dsb *dsb,
 	 * the control register just before the surface register.
 	 */
 	intel_de_write_fw(display, DVSCNTR(pipe), dvscntr);
-	intel_de_write_fw(display, DVSSURF(pipe),
-			  intel_plane_ggtt_offset(plane_state) + dvssurf_offset);
+	intel_de_write_fw(display, DVSSURF(pipe), plane_state->surf);
 
 	if (display->platform.g4x)
 		g4x_sprite_update_gamma(plane_state);
@@ -1624,6 +1615,7 @@ intel_sprite_plane_create(struct intel_display *display,
 		plane->capture_error = vlv_sprite_capture_error;
 		plane->get_hw_state = vlv_sprite_get_hw_state;
 		plane->check_plane = vlv_sprite_check;
+		plane->surf_offset = i965_plane_surf_offset;
 		plane->max_stride = i965_plane_max_stride;
 		plane->min_alignment = vlv_plane_min_alignment;
 		plane->min_cdclk = vlv_plane_min_cdclk;
@@ -1648,6 +1640,7 @@ intel_sprite_plane_create(struct intel_display *display,
 		plane->capture_error = ivb_sprite_capture_error;
 		plane->get_hw_state = ivb_sprite_get_hw_state;
 		plane->check_plane = g4x_sprite_check;
+		plane->surf_offset = i965_plane_surf_offset;
 
 		if (display->platform.broadwell || display->platform.haswell) {
 			plane->max_stride = hsw_sprite_max_stride;
@@ -1673,6 +1666,7 @@ intel_sprite_plane_create(struct intel_display *display,
 		plane->capture_error = g4x_sprite_capture_error;
 		plane->get_hw_state = g4x_sprite_get_hw_state;
 		plane->check_plane = g4x_sprite_check;
+		plane->surf_offset = i965_plane_surf_offset;
 		plane->max_stride = g4x_sprite_max_stride;
 		plane->min_alignment = g4x_sprite_min_alignment;
 		plane->min_cdclk = g4x_sprite_min_cdclk;
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index e20972ddfa09..a61dc9dea3fb 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -1285,13 +1285,20 @@ static u32 skl_surf_address(const struct intel_plane_state *plane_state,
 	}
 }
 
-static u32 skl_plane_surf(const struct intel_plane_state *plane_state,
-			  int color_plane)
+static int icl_plane_color_plane(const struct intel_plane_state *plane_state)
 {
+	if (plane_state->planar_linked_plane && !plane_state->is_y_plane)
+		return 1;
+	else
+		return 0;
+}
+
+static u32 skl_plane_surf_offset(const struct intel_plane_state *plane_state)
+{
+	int color_plane = icl_plane_color_plane(plane_state);
 	u32 plane_surf;
 
-	plane_surf = intel_plane_ggtt_offset(plane_state) +
-		skl_surf_address(plane_state, color_plane);
+	plane_surf = skl_surf_address(plane_state, color_plane);
 
 	if (plane_state->decrypt)
 		plane_surf |= PLANE_SURF_DECRYPT;
@@ -1373,14 +1380,6 @@ static void icl_plane_csc_load_black(struct intel_dsb *dsb,
 	intel_de_write_dsb(display, dsb, PLANE_CSC_POSTOFF(pipe, plane_id, 2), 0);
 }
 
-static int icl_plane_color_plane(const struct intel_plane_state *plane_state)
-{
-	if (plane_state->planar_linked_plane && !plane_state->is_y_plane)
-		return 1;
-	else
-		return 0;
-}
-
 static void
 skl_plane_update_noarm(struct intel_dsb *dsb,
 		       struct intel_plane *plane,
@@ -1476,7 +1475,7 @@ skl_plane_update_arm(struct intel_dsb *dsb,
 	intel_de_write_dsb(display, dsb, PLANE_CTL(pipe, plane_id),
 			   plane_ctl);
 	intel_de_write_dsb(display, dsb, PLANE_SURF(pipe, plane_id),
-			   skl_plane_surf(plane_state, 0));
+			   plane_state->surf);
 }
 
 static void icl_plane_update_sel_fetch_noarm(struct intel_dsb *dsb,
@@ -1632,7 +1631,6 @@ icl_plane_update_arm(struct intel_dsb *dsb,
 	struct intel_display *display = to_intel_display(plane);
 	enum plane_id plane_id = plane->id;
 	enum pipe pipe = plane->pipe;
-	int color_plane = icl_plane_color_plane(plane_state);
 	u32 plane_ctl;
 
 	plane_ctl = plane_state->ctl |
@@ -1658,7 +1656,7 @@ icl_plane_update_arm(struct intel_dsb *dsb,
 	intel_de_write_dsb(display, dsb, PLANE_CTL(pipe, plane_id),
 			   plane_ctl);
 	intel_de_write_dsb(display, dsb, PLANE_SURF(pipe, plane_id),
-			   skl_plane_surf(plane_state, color_plane));
+			   plane_state->surf);
 }
 
 static void skl_plane_capture_error(struct intel_crtc *crtc,
@@ -1682,10 +1680,10 @@ skl_plane_async_flip(struct intel_dsb *dsb,
 	struct intel_display *display = to_intel_display(plane);
 	enum plane_id plane_id = plane->id;
 	enum pipe pipe = plane->pipe;
-	u32 plane_ctl = plane_state->ctl, plane_surf;
+	u32 plane_ctl = plane_state->ctl;
+	u32 plane_surf = plane_state->surf;
 
 	plane_ctl |= skl_plane_ctl_crtc(crtc_state);
-	plane_surf = skl_plane_surf(plane_state, 0);
 
 	if (async_flip) {
 		if (DISPLAY_VER(display) >= 30)
@@ -2814,7 +2812,7 @@ static void skl_disable_tiling(struct intel_plane *plane)
 	intel_de_write_fw(display, PLANE_CTL(plane->pipe, plane->id), plane_ctl);
 
 	intel_de_write_fw(display, PLANE_SURF(plane->pipe, plane->id),
-			  skl_plane_surf(state, 0));
+			  state->surf);
 }
 
 struct intel_plane *
@@ -2865,6 +2863,8 @@ skl_universal_plane_create(struct intel_display *display,
 	}
 	plane->disable_tiling = skl_disable_tiling;
 
+	plane->surf_offset = skl_plane_surf_offset;
+
 	if (DISPLAY_VER(display) >= 13)
 		plane->max_stride = adl_plane_max_stride;
 	else
@@ -3191,21 +3191,18 @@ bool skl_fixup_initial_plane_config(struct intel_crtc *crtc,
 		to_intel_plane_state(plane->base.state);
 	enum plane_id plane_id = plane->id;
 	enum pipe pipe = crtc->pipe;
-	u32 base;
 
 	if (!plane_state->uapi.visible)
 		return false;
 
-	base = intel_plane_ggtt_offset(plane_state);
-
 	/*
 	 * We may have moved the surface to a different
 	 * part of ggtt, make the plane aware of that.
 	 */
-	if (plane_config->base == base)
+	if (plane_config->base == plane_state->surf)
 		return false;
 
-	intel_de_write(display, PLANE_SURF(pipe, plane_id), base);
+	intel_de_write(display, PLANE_SURF(pipe, plane_id), plane_state->surf);
 
 	return true;
 }
diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index 45122822b051..38861c627476 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -12,6 +12,7 @@
 #include "intel_fb.h"
 #include "intel_fb_pin.h"
 #include "intel_fbdev.h"
+#include "intel_plane.h"
 #include "xe_bo.h"
 #include "xe_device.h"
 #include "xe_ggtt.h"
@@ -381,6 +382,7 @@ static bool reuse_vma(struct intel_plane_state *new_plane_state,
 		      const struct intel_plane_state *old_plane_state)
 {
 	struct intel_framebuffer *fb = to_intel_framebuffer(new_plane_state->hw.fb);
+	struct intel_plane *plane = to_intel_plane(new_plane_state->uapi.plane);
 	struct xe_device *xe = to_xe_device(fb->base.dev);
 	struct intel_display *display = xe->display;
 	struct i915_vma *vma;
@@ -404,6 +406,10 @@ static bool reuse_vma(struct intel_plane_state *new_plane_state,
 found:
 	refcount_inc(&vma->ref);
 	new_plane_state->ggtt_vma = vma;
+
+	new_plane_state->surf = intel_plane_ggtt_offset(new_plane_state) +
+		plane->surf_offset(new_plane_state);
+
 	return true;
 }
 
@@ -430,6 +436,10 @@ int intel_plane_pin_fb(struct intel_plane_state *new_plane_state,
 		return PTR_ERR(vma);
 
 	new_plane_state->ggtt_vma = vma;
+
+	new_plane_state->surf = intel_plane_ggtt_offset(new_plane_state) +
+		plane->surf_offset(new_plane_state);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index b2ede3af9345..2882bbd9b554 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -234,6 +234,9 @@ intel_find_initial_plane_obj(struct intel_crtc *crtc,
 		goto nofb;
 
 	plane_state->ggtt_vma = vma;
+
+	plane_state->surf = intel_plane_ggtt_offset(plane_state);
+
 	plane_state->uapi.src_x = 0;
 	plane_state->uapi.src_y = 0;
 	plane_state->uapi.src_w = fb->width << 16;

From bc0c7fd732a89acac4194ae1f4686de63b32138a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Sat, 19 Jul 2025 20:53:54 +0300
Subject: [PATCH 0038/1455] drm/i915: Nuke intel_plane_ggtt_offset()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We don't really need the extra intel_plane_ggtt_offset() wrapper
anymore. Get rid of it.

v2: Deal with reuse_vma() hacks

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250717171353.23090-3-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_fb_pin.c        | 3 ++-
 drivers/gpu/drm/i915/display/intel_plane.c         | 6 ------
 drivers/gpu/drm/i915/display/intel_plane.h         | 1 -
 drivers/gpu/drm/i915/display/intel_plane_initial.c | 2 +-
 drivers/gpu/drm/xe/display/xe_fb_pin.c             | 5 ++---
 drivers/gpu/drm/xe/display/xe_plane_initial.c      | 3 ++-
 6 files changed, 7 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index 5401d5b27e30..340c765d29fe 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -11,6 +11,7 @@
 #include "gem/i915_gem_object.h"
 
 #include "i915_drv.h"
+#include "i915_vma.h"
 #include "intel_display_core.h"
 #include "intel_display_rpm.h"
 #include "intel_display_types.h"
@@ -312,7 +313,7 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state,
 		plane_state->surf = i915_gem_object_get_dma_address(obj, 0) +
 			plane->surf_offset(plane_state);
 	} else {
-		plane_state->surf = intel_plane_ggtt_offset(plane_state) +
+		plane_state->surf = i915_ggtt_offset(plane_state->ggtt_vma) +
 			plane->surf_offset(plane_state);
 	}
 
diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c
index 36fb07471deb..81f05ee9a21a 100644
--- a/drivers/gpu/drm/i915/display/intel_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_plane.c
@@ -46,7 +46,6 @@
 
 #include "gem/i915_gem_object.h"
 #include "i915_scheduler_types.h"
-#include "i915_vma.h"
 #include "i9xx_plane_regs.h"
 #include "intel_bo.h"
 #include "intel_cdclk.h"
@@ -1749,8 +1748,3 @@ int intel_plane_atomic_check(struct intel_atomic_state *state)
 
 	return 0;
 }
-
-u32 intel_plane_ggtt_offset(const struct intel_plane_state *plane_state)
-{
-	return i915_ggtt_offset(plane_state->ggtt_vma);
-}
diff --git a/drivers/gpu/drm/i915/display/intel_plane.h b/drivers/gpu/drm/i915/display/intel_plane.h
index 4ef012c08fa4..8af41ccc0a69 100644
--- a/drivers/gpu/drm/i915/display/intel_plane.h
+++ b/drivers/gpu/drm/i915/display/intel_plane.h
@@ -87,7 +87,6 @@ int intel_plane_add_affected(struct intel_atomic_state *state,
 			     struct intel_crtc *crtc);
 int intel_plane_atomic_check(struct intel_atomic_state *state);
 
-u32 intel_plane_ggtt_offset(const struct intel_plane_state *plane_state);
 bool intel_plane_format_mod_supported_async(struct drm_plane *plane,
 					    u32 format,
 					    u64 modifier);
diff --git a/drivers/gpu/drm/i915/display/intel_plane_initial.c b/drivers/gpu/drm/i915/display/intel_plane_initial.c
index c506b5d660c3..8901bc1dd8f8 100644
--- a/drivers/gpu/drm/i915/display/intel_plane_initial.c
+++ b/drivers/gpu/drm/i915/display/intel_plane_initial.c
@@ -359,7 +359,7 @@ valid_fb:
 	    i915_vma_pin_fence(vma) == 0 && vma->fence)
 		plane_state->flags |= PLANE_HAS_FENCE;
 
-	plane_state->surf = intel_plane_ggtt_offset(plane_state);
+	plane_state->surf = i915_ggtt_offset(plane_state->ggtt_vma);
 
 	plane_state->uapi.src_x = 0;
 	plane_state->uapi.src_y = 0;
diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index 38861c627476..fe1677cf1a96 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -12,7 +12,6 @@
 #include "intel_fb.h"
 #include "intel_fb_pin.h"
 #include "intel_fbdev.h"
-#include "intel_plane.h"
 #include "xe_bo.h"
 #include "xe_device.h"
 #include "xe_ggtt.h"
@@ -407,7 +406,7 @@ found:
 	refcount_inc(&vma->ref);
 	new_plane_state->ggtt_vma = vma;
 
-	new_plane_state->surf = intel_plane_ggtt_offset(new_plane_state) +
+	new_plane_state->surf = i915_ggtt_offset(new_plane_state->ggtt_vma) +
 		plane->surf_offset(new_plane_state);
 
 	return true;
@@ -437,7 +436,7 @@ int intel_plane_pin_fb(struct intel_plane_state *new_plane_state,
 
 	new_plane_state->ggtt_vma = vma;
 
-	new_plane_state->surf = intel_plane_ggtt_offset(new_plane_state) +
+	new_plane_state->surf = i915_ggtt_offset(new_plane_state->ggtt_vma) +
 		plane->surf_offset(new_plane_state);
 
 	return 0;
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index 2882bbd9b554..edda49daa24e 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -10,6 +10,7 @@
 #include "xe_ggtt.h"
 #include "xe_mmio.h"
 
+#include "i915_vma.h"
 #include "intel_crtc.h"
 #include "intel_display.h"
 #include "intel_display_core.h"
@@ -235,7 +236,7 @@ intel_find_initial_plane_obj(struct intel_crtc *crtc,
 
 	plane_state->ggtt_vma = vma;
 
-	plane_state->surf = intel_plane_ggtt_offset(plane_state);
+	plane_state->surf = i915_ggtt_offset(plane_state->ggtt_vma);
 
 	plane_state->uapi.src_x = 0;
 	plane_state->uapi.src_y = 0;

From 44ebdab4b6d878c435da7385b38d2329f8a411d3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 17 Jul 2025 20:13:49 +0300
Subject: [PATCH 0039/1455] drm/i915: Move the intel_dpt_offset() check into
 intel_plane_pin_fb()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Now that we handle all the other vma offset stuff in
intel_plane_pin_fb() it seems more proper to do the
dpt_vma offset check there as well.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250717171353.23090-4-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_fb_pin.c        | 7 +++++++
 drivers/gpu/drm/i915/display/skl_universal_plane.c | 6 ------
 2 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index 340c765d29fe..b0557976836d 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -261,6 +261,7 @@ intel_plane_fb_vtd_guard(const struct intel_plane_state *plane_state)
 int intel_plane_pin_fb(struct intel_plane_state *plane_state,
 		       const struct intel_plane_state *old_plane_state)
 {
+	struct intel_display *display = to_intel_display(plane_state);
 	struct intel_plane *plane = to_intel_plane(plane_state->uapi.plane);
 	const struct intel_framebuffer *fb =
 		to_intel_framebuffer(plane_state->hw.fb);
@@ -299,6 +300,12 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state,
 		plane_state->dpt_vma = vma;
 
 		WARN_ON(plane_state->ggtt_vma == plane_state->dpt_vma);
+
+		/*
+		 * The DPT object contains only one vma, so
+		 * the VMA's offset within the DPT is always 0.
+		 */
+		drm_WARN_ON(display->drm, intel_dpt_offset(plane_state->dpt_vma));
 	}
 
 	/*
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index a61dc9dea3fb..39c107b10a48 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -1271,12 +1271,6 @@ static u32 skl_surf_address(const struct intel_plane_state *plane_state,
 	u32 offset = plane_state->view.color_plane[color_plane].offset;
 
 	if (intel_fb_uses_dpt(fb)) {
-		/*
-		 * The DPT object contains only one vma, so the VMA's offset
-		 * within the DPT is always 0.
-		 */
-		drm_WARN_ON(display->drm, plane_state->dpt_vma &&
-			    intel_dpt_offset(plane_state->dpt_vma));
 		drm_WARN_ON(display->drm, offset & 0x1fffff);
 		return offset >> 9;
 	} else {

From c6ab589feebf81e11b784fe1b4e81208019e72c7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 17 Jul 2025 20:13:50 +0300
Subject: [PATCH 0040/1455] drm/i915: Use i915_vma_offset() in
 intel_dpt_offset()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace the open coded vma mm node stuff in intel_dpt_offset()
with i915_vma_offset(). This will also include the VT-d guard
in the result. Granted that should always be 0 for DPT, but
it seems prudent to include that in our DPT vma offset check
anyway.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250717171353.23090-5-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_dpt.c    | 2 +-
 drivers/gpu/drm/i915/display/intel_fb_pin.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c
index aea249e2699f..4743d8b0bf40 100644
--- a/drivers/gpu/drm/i915/display/intel_dpt.c
+++ b/drivers/gpu/drm/i915/display/intel_dpt.c
@@ -322,5 +322,5 @@ void intel_dpt_destroy(struct i915_address_space *vm)
 
 u64 intel_dpt_offset(struct i915_vma *dpt_vma)
 {
-	return dpt_vma->node.start;
+	return i915_vma_offset(dpt_vma);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index b0557976836d..d598a005f847 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -302,8 +302,8 @@ int intel_plane_pin_fb(struct intel_plane_state *plane_state,
 		WARN_ON(plane_state->ggtt_vma == plane_state->dpt_vma);
 
 		/*
-		 * The DPT object contains only one vma, so
-		 * the VMA's offset within the DPT is always 0.
+		 * The DPT object contains only one vma, and there is no VT-d
+		 * guard, so the VMA's offset within the DPT is always 0.
 		 */
 		drm_WARN_ON(display->drm, intel_dpt_offset(plane_state->dpt_vma));
 	}

From a9298b3bfebc6cacca3363a884dde14bf05f14d2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 17 Jul 2025 20:13:51 +0300
Subject: [PATCH 0041/1455] drm/i915: Remove unused dpt_total_entries()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

dpt_total_entries() is not used anywhere. Remove it.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250717171353.23090-6-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/intel_dpt.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c
index 4743d8b0bf40..c0a817018d08 100644
--- a/drivers/gpu/drm/i915/display/intel_dpt.c
+++ b/drivers/gpu/drm/i915/display/intel_dpt.c
@@ -33,8 +33,6 @@ i915_vm_to_dpt(struct i915_address_space *vm)
 	return container_of(vm, struct i915_dpt, vm);
 }
 
-#define dpt_total_entries(dpt) ((dpt)->vm.total >> PAGE_SHIFT)
-
 static void gen8_set_pte(void __iomem *addr, gen8_pte_t pte)
 {
 	writeq(pte, addr);

From 01e1575609f100e5a47640491591108e0198e273 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Thu, 17 Jul 2025 20:13:52 +0300
Subject: [PATCH 0042/1455] drm/i915: Don't pass crtc_state to foo_plane_ctl()
 & co.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The *_plane_ctl() functions only consider the state of the
plane (the state of the crtc is handled by the corresponding
*_plane_ctl_crtc()), and thus they don't need the crtc_state
at all. Don't pass it in.

Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250717171353.23090-7-ville.syrjala@linux.intel.com
---
 drivers/gpu/drm/i915/display/i9xx_plane.c         |  5 ++---
 drivers/gpu/drm/i915/display/intel_cursor.c       | 10 ++++------
 drivers/gpu/drm/i915/display/intel_sprite.c       | 15 ++++++---------
 .../gpu/drm/i915/display/skl_universal_plane.c    | 11 ++++-------
 4 files changed, 16 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index d572e95c3c92..3eb96d8abba8 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -155,8 +155,7 @@ static bool i9xx_plane_has_windowing(struct intel_plane *plane)
 			i9xx_plane == PLANE_C;
 }
 
-static u32 i9xx_plane_ctl(const struct intel_crtc_state *crtc_state,
-			  const struct intel_plane_state *plane_state)
+static u32 i9xx_plane_ctl(const struct intel_plane_state *plane_state)
 {
 	struct intel_display *display = to_intel_display(plane_state);
 	const struct drm_framebuffer *fb = plane_state->hw.fb;
@@ -355,7 +354,7 @@ i9xx_plane_check(struct intel_crtc_state *crtc_state,
 	if (ret)
 		return ret;
 
-	plane_state->ctl = i9xx_plane_ctl(crtc_state, plane_state);
+	plane_state->ctl = i9xx_plane_ctl(plane_state);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_cursor.c b/drivers/gpu/drm/i915/display/intel_cursor.c
index 83718b627bb0..d4d181f9dca5 100644
--- a/drivers/gpu/drm/i915/display/intel_cursor.c
+++ b/drivers/gpu/drm/i915/display/intel_cursor.c
@@ -205,8 +205,7 @@ static u32 i845_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	return cntl;
 }
 
-static u32 i845_cursor_ctl(const struct intel_crtc_state *crtc_state,
-			   const struct intel_plane_state *plane_state)
+static u32 i845_cursor_ctl(const struct intel_plane_state *plane_state)
 {
 	return CURSOR_ENABLE |
 		CURSOR_FORMAT_ARGB |
@@ -266,7 +265,7 @@ static int i845_check_cursor(struct intel_crtc_state *crtc_state,
 		return -EINVAL;
 	}
 
-	plane_state->ctl = i845_cursor_ctl(crtc_state, plane_state);
+	plane_state->ctl = i845_cursor_ctl(plane_state);
 
 	return 0;
 }
@@ -398,8 +397,7 @@ static u32 i9xx_cursor_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	return cntl;
 }
 
-static u32 i9xx_cursor_ctl(const struct intel_crtc_state *crtc_state,
-			   const struct intel_plane_state *plane_state)
+static u32 i9xx_cursor_ctl(const struct intel_plane_state *plane_state)
 {
 	struct intel_display *display = to_intel_display(plane_state);
 	u32 cntl = 0;
@@ -526,7 +524,7 @@ static int i9xx_check_cursor(struct intel_crtc_state *crtc_state,
 		return -EINVAL;
 	}
 
-	plane_state->ctl = i9xx_cursor_ctl(crtc_state, plane_state);
+	plane_state->ctl = i9xx_cursor_ctl(plane_state);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_sprite.c b/drivers/gpu/drm/i915/display/intel_sprite.c
index 230f90e79d2a..75bbaa923204 100644
--- a/drivers/gpu/drm/i915/display/intel_sprite.c
+++ b/drivers/gpu/drm/i915/display/intel_sprite.c
@@ -264,8 +264,7 @@ static u32 vlv_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	return sprctl;
 }
 
-static u32 vlv_sprite_ctl(const struct intel_crtc_state *crtc_state,
-			  const struct intel_plane_state *plane_state)
+static u32 vlv_sprite_ctl(const struct intel_plane_state *plane_state)
 {
 	const struct drm_framebuffer *fb = plane_state->hw.fb;
 	unsigned int rotation = plane_state->hw.rotation;
@@ -660,8 +659,7 @@ static bool ivb_need_sprite_gamma(const struct intel_plane_state *plane_state)
 		(display->platform.ivybridge || display->platform.haswell);
 }
 
-static u32 ivb_sprite_ctl(const struct intel_crtc_state *crtc_state,
-			  const struct intel_plane_state *plane_state)
+static u32 ivb_sprite_ctl(const struct intel_plane_state *plane_state)
 {
 	struct intel_display *display = to_intel_display(plane_state);
 	const struct drm_framebuffer *fb = plane_state->hw.fb;
@@ -1010,8 +1008,7 @@ static u32 g4x_sprite_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	return dvscntr;
 }
 
-static u32 g4x_sprite_ctl(const struct intel_crtc_state *crtc_state,
-			  const struct intel_plane_state *plane_state)
+static u32 g4x_sprite_ctl(const struct intel_plane_state *plane_state)
 {
 	struct intel_display *display = to_intel_display(plane_state);
 	const struct drm_framebuffer *fb = plane_state->hw.fb;
@@ -1378,9 +1375,9 @@ g4x_sprite_check(struct intel_crtc_state *crtc_state,
 		return ret;
 
 	if (DISPLAY_VER(display) >= 7)
-		plane_state->ctl = ivb_sprite_ctl(crtc_state, plane_state);
+		plane_state->ctl = ivb_sprite_ctl(plane_state);
 	else
-		plane_state->ctl = g4x_sprite_ctl(crtc_state, plane_state);
+		plane_state->ctl = g4x_sprite_ctl(plane_state);
 
 	return 0;
 }
@@ -1430,7 +1427,7 @@ vlv_sprite_check(struct intel_crtc_state *crtc_state,
 	if (ret)
 		return ret;
 
-	plane_state->ctl = vlv_sprite_ctl(crtc_state, plane_state);
+	plane_state->ctl = vlv_sprite_ctl(plane_state);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 39c107b10a48..ec3fe75f2c6c 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -1166,8 +1166,7 @@ static u32 skl_plane_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	return plane_ctl;
 }
 
-static u32 skl_plane_ctl(const struct intel_crtc_state *crtc_state,
-			 const struct intel_plane_state *plane_state)
+static u32 skl_plane_ctl(const struct intel_plane_state *plane_state)
 {
 	struct intel_display *display = to_intel_display(plane_state);
 	const struct drm_framebuffer *fb = plane_state->hw.fb;
@@ -1225,8 +1224,7 @@ static u32 glk_plane_color_ctl_crtc(const struct intel_crtc_state *crtc_state)
 	return plane_color_ctl;
 }
 
-static u32 glk_plane_color_ctl(const struct intel_crtc_state *crtc_state,
-			       const struct intel_plane_state *plane_state)
+static u32 glk_plane_color_ctl(const struct intel_plane_state *plane_state)
 {
 	struct intel_display *display = to_intel_display(plane_state);
 	const struct drm_framebuffer *fb = plane_state->hw.fb;
@@ -2355,11 +2353,10 @@ static int skl_plane_check(struct intel_crtc_state *crtc_state,
 		plane_state->damage = DRM_RECT_INIT(0, 0, 0, 0);
 	}
 
-	plane_state->ctl = skl_plane_ctl(crtc_state, plane_state);
+	plane_state->ctl = skl_plane_ctl(plane_state);
 
 	if (DISPLAY_VER(display) >= 10)
-		plane_state->color_ctl = glk_plane_color_ctl(crtc_state,
-							     plane_state);
+		plane_state->color_ctl = glk_plane_color_ctl(plane_state);
 
 	if (intel_format_info_is_yuv_semiplanar(fb->format, fb->modifier) &&
 	    icl_is_hdr_plane(display, plane->id))

From d12a58c2f179425093b707696ad00509e03df03c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 15 Jul 2025 13:55:07 +0300
Subject: [PATCH 0043/1455] drm/i915/psr: Do not disable Early Transport when
 enable_psr is set
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Current approach is that Early Transport is disabled in case enable_psr
module parameter is set. Let's ignore enable_psr parameter when choosing if
Early Transport can be used.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250715105509.4146806-2-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index ae9053919211..a2b5688f0c82 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -265,16 +265,6 @@ static bool psr2_global_enabled(struct intel_dp *intel_dp)
 	}
 }
 
-static bool psr2_su_region_et_global_enabled(struct intel_dp *intel_dp)
-{
-	struct intel_display *display = to_intel_display(intel_dp);
-
-	if (display->params.enable_psr != -1)
-		return false;
-
-	return true;
-}
-
 static bool panel_replay_global_enabled(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
@@ -742,8 +732,7 @@ static bool psr2_su_region_et_valid(struct intel_dp *intel_dp, bool panel_replay
 	return panel_replay ?
 		intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_SUPPORT)] &
 		DP_PANEL_REPLAY_EARLY_TRANSPORT_SUPPORT :
-		intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_ET_SUPPORTED &&
-		psr2_su_region_et_global_enabled(intel_dp);
+		intel_dp->psr_dpcd[0] == DP_PSR2_WITH_Y_COORD_ET_SUPPORTED;
 }
 
 static void _panel_replay_enable_sink(struct intel_dp *intel_dp,

From 8b6c828bb7b9c3263223f7cfd33d92f010610bd7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 15 Jul 2025 13:55:08 +0300
Subject: [PATCH 0044/1455] drm/i915/psr: Ignore enable_psr parameter on Panel
 Replay
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we are disabling Panel Replay if enable_psr != -1. Lets ignore
enable_psr completely on Panel Replay.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250715105509.4146806-3-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 27 ++++++------------------
 1 file changed, 7 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index a2b5688f0c82..ee1f9a240283 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -233,16 +233,12 @@ bool intel_psr_needs_aux_io_power(struct intel_encoder *encoder,
 
 static bool psr_global_enabled(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_connector *connector = intel_dp->attached_connector;
 
 	switch (intel_dp->psr.debug & I915_PSR_DEBUG_MODE_MASK) {
 	case I915_PSR_DEBUG_DEFAULT:
-		if (display->params.enable_psr == -1)
-			return intel_dp_is_edp(intel_dp) ?
-				connector->panel.vbt.psr.enable :
-				true;
-		return display->params.enable_psr;
+		return intel_dp_is_edp(intel_dp) ?
+			connector->panel.vbt.psr.enable : true;
 	case I915_PSR_DEBUG_DISABLE:
 		return false;
 	default:
@@ -250,29 +246,20 @@ static bool psr_global_enabled(struct intel_dp *intel_dp)
 	}
 }
 
-static bool psr2_global_enabled(struct intel_dp *intel_dp)
+static bool sel_update_global_enabled(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-
 	switch (intel_dp->psr.debug & I915_PSR_DEBUG_MODE_MASK) {
 	case I915_PSR_DEBUG_DISABLE:
 	case I915_PSR_DEBUG_FORCE_PSR1:
 		return false;
 	default:
-		if (display->params.enable_psr == 1)
-			return false;
 		return true;
 	}
 }
 
 static bool panel_replay_global_enabled(struct intel_dp *intel_dp)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-
-	if ((display->params.enable_psr != -1) ||
-	    (intel_dp->psr.debug & I915_PSR_DEBUG_PANEL_REPLAY_DISABLE))
-		return false;
-	return true;
+	return !(intel_dp->psr.debug & I915_PSR_DEBUG_PANEL_REPLAY_DISABLE);
 }
 
 static u32 psr_irq_psr_error_bit_get(struct intel_dp *intel_dp)
@@ -1412,7 +1399,7 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp,
 	int crtc_vdisplay = crtc_state->hw.adjusted_mode.crtc_vdisplay;
 	int psr_max_h = 0, psr_max_v = 0, max_bpp = 0;
 
-	if (!intel_dp->psr.sink_psr2_support)
+	if (!intel_dp->psr.sink_psr2_support || display->params.enable_psr == 1)
 		return false;
 
 	/* JSL and EHL only supports eDP 1.3 */
@@ -1517,7 +1504,7 @@ static bool intel_sel_update_config_valid(struct intel_dp *intel_dp,
 		goto unsupported;
 	}
 
-	if (!psr2_global_enabled(intel_dp)) {
+	if (!sel_update_global_enabled(intel_dp)) {
 		drm_dbg_kms(display->drm,
 			    "Selective update disabled by flag\n");
 		goto unsupported;
@@ -1565,7 +1552,7 @@ static bool _psr_compute_config(struct intel_dp *intel_dp,
 	const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
 	int entry_setup_frames;
 
-	if (!CAN_PSR(intel_dp))
+	if (!CAN_PSR(intel_dp) || !display->params.enable_psr)
 		return false;
 
 	/*

From f9cc4616ba70681e0f10e6872d7372aaa479bb13 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 15 Jul 2025 13:55:09 +0300
Subject: [PATCH 0045/1455] drm/i915/psr: Add enable_panel_replay module
 parameter
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add new module parameter enable_panel_replay. This can be used to
enable/disable Panel Replay. 0=disabled, 1=enabled. -1=use per-chip default
(default).

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250715105509.4146806-4-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_params.c | 3 +++
 drivers/gpu/drm/i915/display/intel_display_params.h | 1 +
 drivers/gpu/drm/i915/display/intel_psr.c            | 5 ++++-
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_params.c b/drivers/gpu/drm/i915/display/intel_display_params.c
index 75316247ee8a..2aed110c5b09 100644
--- a/drivers/gpu/drm/i915/display/intel_display_params.c
+++ b/drivers/gpu/drm/i915/display/intel_display_params.c
@@ -120,6 +120,9 @@ intel_display_param_named_unsafe(enable_psr, int, 0400,
 	"(0=disabled, 1=enable up to PSR1, 2=enable up to PSR2) "
 	"Default: -1 (use per-chip default)");
 
+intel_display_param_named_unsafe(enable_panel_replay, int, 0400,
+	"Enable Panel Replay (0=disabled, 1=enabled). Default: -1 (use per-chip default)");
+
 intel_display_param_named(psr_safest_params, bool, 0400,
 	"Replace PSR VBT parameters by the safest and not optimal ones. This "
 	"is helpful to detect if PSR issues are related to bad values set in "
diff --git a/drivers/gpu/drm/i915/display/intel_display_params.h b/drivers/gpu/drm/i915/display/intel_display_params.h
index 784e6bae8615..b01bc5700c52 100644
--- a/drivers/gpu/drm/i915/display/intel_display_params.h
+++ b/drivers/gpu/drm/i915/display/intel_display_params.h
@@ -46,6 +46,7 @@ struct drm_printer;
 	param(bool, enable_dp_mst, true, 0600) \
 	param(int, enable_fbc, -1, 0600) \
 	param(int, enable_psr, -1, 0600) \
+	param(int, enable_panel_replay, -1, 0600) \
 	param(bool, psr_safest_params, false, 0400) \
 	param(bool, enable_psr2_sel_fetch, true, 0400) \
 	param(int, enable_dmc_wl, -1, 0400) \
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index ee1f9a240283..6bd3454bb00e 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -259,7 +259,10 @@ static bool sel_update_global_enabled(struct intel_dp *intel_dp)
 
 static bool panel_replay_global_enabled(struct intel_dp *intel_dp)
 {
-	return !(intel_dp->psr.debug & I915_PSR_DEBUG_PANEL_REPLAY_DISABLE);
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	return !(intel_dp->psr.debug & I915_PSR_DEBUG_PANEL_REPLAY_DISABLE) &&
+		display->params.enable_panel_replay;
 }
 
 static u32 psr_irq_psr_error_bit_get(struct intel_dp *intel_dp)

From 6c9e64e83b22405622d1f47417cdb0d20d49ca35 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Fri, 18 Jul 2025 16:23:05 -0500
Subject: [PATCH 0046/1455] drm/xe: Fix an IS_ERR() vs NULL bug in
 xe_tile_alloc_vram()

The xe_vram_region_alloc() function returns NULL on error.  It never
returns error pointers.  Update the error checking to match.

Fixes: 4b0a5f5ce784 ("drm/xe: Unify the initialization of VRAM regions")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/5449065e-9758-4711-b706-78771c0753c4@sabinyo.mountain
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_tile.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_tile.c b/drivers/gpu/drm/xe/xe_tile.c
index 0be0a5c57ef4..d49ba3401963 100644
--- a/drivers/gpu/drm/xe/xe_tile.c
+++ b/drivers/gpu/drm/xe/xe_tile.c
@@ -120,8 +120,8 @@ int xe_tile_alloc_vram(struct xe_tile *tile)
 		return 0;
 
 	vram = xe_vram_region_alloc(xe, tile->id, XE_PL_VRAM0 + tile->id);
-	if (IS_ERR(vram))
-		return PTR_ERR(vram);
+	if (!vram)
+		return -ENOMEM;
 	tile->mem.vram = vram;
 
 	return 0;

From 17133255a32275f0f042f7f432f332ff1cf5e57d Mon Sep 17 00:00:00 2001
From: Ruben Wauters <rubenru09@aol.com>
Date: Tue, 1 Jul 2025 12:54:51 +0100
Subject: [PATCH 0047/1455] drm/i915: replace DRM_DEBUG_SELFTEST with
 DRM_KUNIT_TEST

DRM_DEBUG_SELFTEST was removed in commit fc8d29e298cf (drm: selftest:
convert drm_mm selftest to KUnit) and all functions under it were
converted to KUnit, under the DRM_KUNIT_TEST option

This conversion however did not occur in the Kconfig.debug file in the
i915 directory.

This patch replaces the select for DRM_DEBUG_SELFTEST, an option that no
longer exists, with the correct select, DRM_KUNIT_TEST.

Signed-off-by: Ruben Wauters <rubenru09@aol.com>
Link: https://lore.kernel.org/r/20250701115511.5445-1-rubenru09@aol.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/Kconfig.debug | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/Kconfig.debug b/drivers/gpu/drm/i915/Kconfig.debug
index 1852e0804942..b15b1cecb3aa 100644
--- a/drivers/gpu/drm/i915/Kconfig.debug
+++ b/drivers/gpu/drm/i915/Kconfig.debug
@@ -50,7 +50,7 @@ config DRM_I915_DEBUG
 	select DRM_VGEM # used by igt/prime_vgem (dmabuf interop checks)
 	select DRM_DEBUG_MM if DRM=y
 	select DRM_EXPORT_FOR_TESTS if m
-	select DRM_DEBUG_SELFTEST
+	select DRM_KUNIT_TEST
 	select DMABUF_SELFTESTS
 	select SW_SYNC # signaling validation framework (igt/syncobj*)
 	select DRM_I915_WERROR

From 94febfb5bcfb6ccf02283cc07bf58927c119afca Mon Sep 17 00:00:00 2001
From: Beata Michalska <beata.michalska@arm.com>
Date: Thu, 26 Jun 2025 18:23:13 +0200
Subject: [PATCH 0048/1455] rust: drm: Drop the use of Opaque for ioctl
 arguments

With the Opaque<T>, the expectations are that Rust should not
make any assumptions on the layout or invariants of the wrapped
C types. That runs rather counter to ioctl arguments, which must
adhere to certain data-layout constraints. By using Opaque<T>,
ioctl handlers are forced to use unsafe code where none is actually
needed. This adds needless complexity and maintenance overhead,
brining no safety benefits.
Drop the use of Opaque for ioctl arguments as that is not the best
fit here.

Signed-off-by: Beata Michalska <beata.michalska@arm.com>
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250626162313.2755584-1-beata.michalska@arm.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/nova/file.rs | 23 ++++++--------
 drivers/gpu/drm/nova/nova.rs |  1 -
 drivers/gpu/drm/nova/uapi.rs | 61 ------------------------------------
 rust/kernel/drm/ioctl.rs     | 11 ++++---
 4 files changed, 16 insertions(+), 80 deletions(-)
 delete mode 100644 drivers/gpu/drm/nova/uapi.rs

diff --git a/drivers/gpu/drm/nova/file.rs b/drivers/gpu/drm/nova/file.rs
index 7e59a34b830d..7e7d4e2de2fb 100644
--- a/drivers/gpu/drm/nova/file.rs
+++ b/drivers/gpu/drm/nova/file.rs
@@ -2,13 +2,11 @@
 
 use crate::driver::{NovaDevice, NovaDriver};
 use crate::gem::NovaObject;
-use crate::uapi::{GemCreate, GemInfo, Getparam};
 use kernel::{
     alloc::flags::*,
     drm::{self, gem::BaseObject},
     pci,
     prelude::*,
-    types::Opaque,
     uapi,
 };
 
@@ -26,20 +24,19 @@ impl File {
     /// IOCTL: get_param: Query GPU / driver metadata.
     pub(crate) fn get_param(
         dev: &NovaDevice,
-        getparam: &Opaque<uapi::drm_nova_getparam>,
+        getparam: &mut uapi::drm_nova_getparam,
         _file: &drm::File<File>,
     ) -> Result<u32> {
         let adev = &dev.adev;
         let parent = adev.parent().ok_or(ENOENT)?;
         let pdev: &pci::Device = parent.try_into()?;
-        let getparam: &Getparam = getparam.into();
 
-        let value = match getparam.param() as u32 {
+        let value = match getparam.param as u32 {
             uapi::NOVA_GETPARAM_VRAM_BAR_SIZE => pdev.resource_len(1)?,
             _ => return Err(EINVAL),
         };
 
-        getparam.set_value(value);
+        getparam.value = value;
 
         Ok(0)
     }
@@ -47,13 +44,12 @@ impl File {
     /// IOCTL: gem_create: Create a new DRM GEM object.
     pub(crate) fn gem_create(
         dev: &NovaDevice,
-        req: &Opaque<uapi::drm_nova_gem_create>,
+        req: &mut uapi::drm_nova_gem_create,
         file: &drm::File<File>,
     ) -> Result<u32> {
-        let req: &GemCreate = req.into();
-        let obj = NovaObject::new(dev, req.size().try_into()?)?;
+        let obj = NovaObject::new(dev, req.size.try_into()?)?;
 
-        req.set_handle(obj.create_handle(file)?);
+        req.handle = obj.create_handle(file)?;
 
         Ok(0)
     }
@@ -61,13 +57,12 @@ impl File {
     /// IOCTL: gem_info: Query GEM metadata.
     pub(crate) fn gem_info(
         _dev: &NovaDevice,
-        req: &Opaque<uapi::drm_nova_gem_info>,
+        req: &mut uapi::drm_nova_gem_info,
         file: &drm::File<File>,
     ) -> Result<u32> {
-        let req: &GemInfo = req.into();
-        let bo = NovaObject::lookup_handle(file, req.handle())?;
+        let bo = NovaObject::lookup_handle(file, req.handle)?;
 
-        req.set_size(bo.size().try_into()?);
+        req.size = bo.size().try_into()?;
 
         Ok(0)
     }
diff --git a/drivers/gpu/drm/nova/nova.rs b/drivers/gpu/drm/nova/nova.rs
index 902876aa14d1..730598defe04 100644
--- a/drivers/gpu/drm/nova/nova.rs
+++ b/drivers/gpu/drm/nova/nova.rs
@@ -5,7 +5,6 @@
 mod driver;
 mod file;
 mod gem;
-mod uapi;
 
 use crate::driver::NovaDriver;
 
diff --git a/drivers/gpu/drm/nova/uapi.rs b/drivers/gpu/drm/nova/uapi.rs
deleted file mode 100644
index eb228a58d423..000000000000
--- a/drivers/gpu/drm/nova/uapi.rs
+++ /dev/null
@@ -1,61 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-
-use kernel::uapi;
-
-// TODO Work out some common infrastructure to avoid boilerplate code for uAPI abstractions.
-
-macro_rules! define_uapi_abstraction {
-    ($name:ident <= $inner:ty) => {
-        #[repr(transparent)]
-        pub struct $name(::kernel::types::Opaque<$inner>);
-
-        impl ::core::convert::From<&::kernel::types::Opaque<$inner>> for &$name {
-            fn from(value: &::kernel::types::Opaque<$inner>) -> Self {
-                // SAFETY: `Self` is a transparent wrapper of `$inner`.
-                unsafe { ::core::mem::transmute(value) }
-            }
-        }
-    };
-}
-
-define_uapi_abstraction!(Getparam <= uapi::drm_nova_getparam);
-
-impl Getparam {
-    pub fn param(&self) -> u64 {
-        // SAFETY: `self.get()` is a valid pointer to a `struct drm_nova_getparam`.
-        unsafe { (*self.0.get()).param }
-    }
-
-    pub fn set_value(&self, v: u64) {
-        // SAFETY: `self.get()` is a valid pointer to a `struct drm_nova_getparam`.
-        unsafe { (*self.0.get()).value = v };
-    }
-}
-
-define_uapi_abstraction!(GemCreate <= uapi::drm_nova_gem_create);
-
-impl GemCreate {
-    pub fn size(&self) -> u64 {
-        // SAFETY: `self.get()` is a valid pointer to a `struct drm_nova_gem_create`.
-        unsafe { (*self.0.get()).size }
-    }
-
-    pub fn set_handle(&self, handle: u32) {
-        // SAFETY: `self.get()` is a valid pointer to a `struct drm_nova_gem_create`.
-        unsafe { (*self.0.get()).handle = handle };
-    }
-}
-
-define_uapi_abstraction!(GemInfo <= uapi::drm_nova_gem_info);
-
-impl GemInfo {
-    pub fn handle(&self) -> u32 {
-        // SAFETY: `self.get()` is a valid pointer to a `struct drm_nova_gem_info`.
-        unsafe { (*self.0.get()).handle }
-    }
-
-    pub fn set_size(&self, size: u64) {
-        // SAFETY: `self.get()` is a valid pointer to a `struct drm_nova_gem_info`.
-        unsafe { (*self.0.get()).size = size };
-    }
-}
diff --git a/rust/kernel/drm/ioctl.rs b/rust/kernel/drm/ioctl.rs
index fdec01c37168..af1bb29cf06d 100644
--- a/rust/kernel/drm/ioctl.rs
+++ b/rust/kernel/drm/ioctl.rs
@@ -83,7 +83,7 @@ pub mod internal {
 ///
 /// ```ignore
 /// fn foo(device: &kernel::drm::Device<Self>,
-///        data: &Opaque<uapi::argument_type>,
+///        data: &mut uapi::argument_type,
 ///        file: &kernel::drm::File<Self::File>,
 /// ) -> Result<u32>
 /// ```
@@ -138,9 +138,12 @@ macro_rules! declare_drm_ioctls {
                             // SAFETY: The ioctl argument has size `_IOC_SIZE(cmd)`, which we
                             // asserted above matches the size of this type, and all bit patterns of
                             // UAPI structs must be valid.
-                            let data = unsafe {
-                                &*(raw_data as *const $crate::types::Opaque<$crate::uapi::$struct>)
-                            };
+                            // The `ioctl` argument is exclusively owned by the handler
+                            // and guaranteed by the C implementation (`drm_ioctl()`) to remain
+                            // valid for the entire lifetime of the reference taken here.
+                            // There is no concurrent access or aliasing; no other references
+                            // to this object exist during this call.
+                            let data = unsafe { &mut *(raw_data.cast::<$crate::uapi::$struct>()) };
                             // SAFETY: This is just the DRM file structure
                             let file = unsafe { $crate::drm::File::from_raw(raw_file) };
 

From 9b3700b15cb581d748c3d46e7eb30ffced1642e8 Mon Sep 17 00:00:00 2001
From: Langyan Ye <yelangyan@huaqin.corp-partner.google.com>
Date: Mon, 21 Jul 2025 14:16:27 +0800
Subject: [PATCH 0049/1455] drm/panel-edp: Add disable to 100ms for MNB601LS1-4

For the MNB601LS1-4 panel, the T9+T10 timing does not meet the
requirements of the specification, so disable is set to 100ms.

Fixes: 9d8e91439fc3 ("drm/panel-edp: Add CSW MNB601LS1-4")
Signed-off-by: Langyan Ye <yelangyan@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250721061627.3816612-1-yelangyan@huaqin.corp-partner.google.com
---
 drivers/gpu/drm/panel/panel-edp.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index 9a56e208cbdd..09170470b3ef 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -1828,6 +1828,13 @@ static const struct panel_delay delay_50_500_e200_d200_po2e335 = {
 	.powered_on_to_enable = 335,
 };
 
+static const struct panel_delay delay_200_500_e50_d100 = {
+	.hpd_absent = 200,
+	.unprepare = 500,
+	.enable = 50,
+	.disable = 100,
+};
+
 #define EDP_PANEL_ENTRY(vend_chr_0, vend_chr_1, vend_chr_2, product_id, _delay, _name) \
 { \
 	.ident = { \
@@ -1984,7 +1991,7 @@ static const struct edp_panel_entry edp_panels[] = {
 
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1100, &delay_200_500_e80_d50, "MNB601LS1-1"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1103, &delay_200_500_e80_d50, "MNB601LS1-3"),
-	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50, "MNB601LS1-4"),
+	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50_d100, "MNB601LS1-4"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1457, &delay_80_500_e80_p2e200, "MNE007QS3-8"),
 

From ac6e050e69811c5c2057d519362c7b173a915542 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <dev@lankhorst.se>
Date: Wed, 16 Jul 2025 12:56:36 +0200
Subject: [PATCH 0050/1455] accel/ivpu: Remove lockdep_assert_irqs_disabled()

This breaks on PREEMPT_RT, and should be unneeded since
lockdep can track irq disabled status itself.

Link: https://lore.kernel.org/r/20250715165919.33754-1-dev@lankhorst.se
Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
---
 drivers/accel/ivpu/ivpu_ipc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c
index 39f83225c181..5f00809d448a 100644
--- a/drivers/accel/ivpu/ivpu_ipc.c
+++ b/drivers/accel/ivpu/ivpu_ipc.c
@@ -141,7 +141,6 @@ ivpu_ipc_rx_msg_add(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons,
 	struct ivpu_ipc_rx_msg *rx_msg;
 
 	lockdep_assert_held(&ipc->cons_lock);
-	lockdep_assert_irqs_disabled();
 
 	rx_msg = kzalloc(sizeof(*rx_msg), GFP_ATOMIC);
 	if (!rx_msg) {

From 4d2d28776ae3ad7aa95328d28aff220b0ec6202d Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Tue, 8 Jul 2025 17:48:18 +0200
Subject: [PATCH 0051/1455] drm/bridge: add a cleanup action for scope-based
 drm_bridge_put() invocation

Many functions get a drm_bridge pointer, only use it in the function body
(or a smaller scope such as a loop body), and don't store it. In these
cases they always need to drm_bridge_put() it before returning (or exiting
the scope).

Some of those functions have complex code paths with multiple return points
or loop break/continue. This makes adding drm_bridge_put() in the right
places tricky, ugly and error prone in case of future code changes.

Others use the bridge pointer in the return statement and would need to
split the return line to fit the drm_bridge_put, which is a bit annoying:

  -return some_thing(bridge);
  +ret = some_thing(bridge);
  +drm_bridge_put(bridge);
  +return ret;

To make it easier for all of them to put the bridge reference correctly
without complicating code, define a scope-based cleanup action to be used
with __free().

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250708-drm-bridge-alloc-getput-drm_bridge_chain_get_first_bridge-v9-1-db1ba3df7f58@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 include/drm/drm_bridge.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index 8ed80cad77ec..8290e665554e 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -23,6 +23,7 @@
 #ifndef __DRM_BRIDGE_H__
 #define __DRM_BRIDGE_H__
 
+#include <linux/cleanup.h>
 #include <linux/ctype.h>
 #include <linux/list.h>
 #include <linux/mutex.h>
@@ -1228,6 +1229,9 @@ drm_priv_to_bridge(struct drm_private_obj *priv)
 struct drm_bridge *drm_bridge_get(struct drm_bridge *bridge);
 void drm_bridge_put(struct drm_bridge *bridge);
 
+/* Cleanup action for use with __free() */
+DEFINE_FREE(drm_bridge_put, struct drm_bridge *, if (_T) drm_bridge_put(_T))
+
 void *__devm_drm_bridge_alloc(struct device *dev, size_t size, size_t offset,
 			      const struct drm_bridge_funcs *funcs);
 

From 8fa5909400f377351836419223c33f1131f0f7d3 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Tue, 8 Jul 2025 17:48:19 +0200
Subject: [PATCH 0052/1455] drm/bridge: get the bridge returned by
 drm_bridge_chain_get_first_bridge()

drm_bridge_chain_get_first_bridge() returns a bridge pointer that the
caller could hold for a long time. Increment the refcount of the returned
bridge and document it must be put by the caller.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250708-drm-bridge-alloc-getput-drm_bridge_chain_get_first_bridge-v9-2-db1ba3df7f58@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 include/drm/drm_bridge.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index 8290e665554e..717171d0e587 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -1337,6 +1337,9 @@ drm_bridge_get_prev_bridge(struct drm_bridge *bridge)
  * drm_bridge_chain_get_first_bridge() - Get the first bridge in the chain
  * @encoder: encoder object
  *
+ * The refcount of the returned bridge is incremented. Use drm_bridge_put()
+ * when done with it.
+ *
  * RETURNS:
  * the first bridge in the chain, or NULL if @encoder has no bridge attached
  * to it.
@@ -1344,8 +1347,8 @@ drm_bridge_get_prev_bridge(struct drm_bridge *bridge)
 static inline struct drm_bridge *
 drm_bridge_chain_get_first_bridge(struct drm_encoder *encoder)
 {
-	return list_first_entry_or_null(&encoder->bridge_chain,
-					struct drm_bridge, chain_node);
+	return drm_bridge_get(list_first_entry_or_null(&encoder->bridge_chain,
+						       struct drm_bridge, chain_node));
 }
 
 /**

From a73ddcba33ed8fe8f2bd5be4de88125b06df96b4 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Tue, 8 Jul 2025 17:48:20 +0200
Subject: [PATCH 0053/1455] drm/mxsfb: put the bridge returned by
 drm_bridge_chain_get_first_bridge()

The bridge returned by drm_bridge_chain_get_first_bridge() is
refcounted. Put it when done. Use a scope-based free action to catch all
the code paths.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250708-drm-bridge-alloc-getput-drm_bridge_chain_get_first_bridge-v9-3-db1ba3df7f58@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/mxsfb/lcdif_kms.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/mxsfb/lcdif_kms.c b/drivers/gpu/drm/mxsfb/lcdif_kms.c
index dbd42cc1da87..1c3b33be6c40 100644
--- a/drivers/gpu/drm/mxsfb/lcdif_kms.c
+++ b/drivers/gpu/drm/mxsfb/lcdif_kms.c
@@ -433,7 +433,6 @@ static int lcdif_crtc_atomic_check(struct drm_crtc *crtc,
 	struct drm_connector *connector;
 	struct drm_encoder *encoder;
 	struct drm_bridge_state *bridge_state;
-	struct drm_bridge *bridge;
 	u32 bus_format, bus_flags;
 	bool format_set = false, flags_set = false;
 	int ret, i;
@@ -453,7 +452,8 @@ static int lcdif_crtc_atomic_check(struct drm_crtc *crtc,
 
 		encoder = connector_state->best_encoder;
 
-		bridge = drm_bridge_chain_get_first_bridge(encoder);
+		struct drm_bridge *bridge __free(drm_bridge_put) =
+			drm_bridge_chain_get_first_bridge(encoder);
 		if (!bridge)
 			continue;
 

From c26c844390e1384c49dd9a395c3bab268dcb5bea Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Tue, 8 Jul 2025 17:48:21 +0200
Subject: [PATCH 0054/1455] drm/atomic-helper: put the bridge returned by
 drm_bridge_chain_get_first_bridge()

The bridge returned by drm_bridge_chain_get_first_bridge() is
refcounted. Put it when done.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250708-drm-bridge-alloc-getput-drm_bridge_chain_get_first_bridge-v9-4-db1ba3df7f58@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/drm_atomic_helper.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index ef56b474acf5..d5ebe6ea0acb 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -456,6 +456,7 @@ mode_fixup(struct drm_atomic_state *state)
 		ret = drm_atomic_bridge_chain_check(bridge,
 						    new_crtc_state,
 						    new_conn_state);
+		drm_bridge_put(bridge);
 		if (ret) {
 			drm_dbg_atomic(encoder->dev, "Bridge atomic check failed\n");
 			return ret;
@@ -527,6 +528,7 @@ static enum drm_mode_status mode_valid_path(struct drm_connector *connector,
 	bridge = drm_bridge_chain_get_first_bridge(encoder);
 	ret = drm_bridge_chain_mode_valid(bridge, &connector->display_info,
 					  mode);
+	drm_bridge_put(bridge);
 	if (ret != MODE_OK) {
 		drm_dbg_atomic(encoder->dev, "[BRIDGE] mode_valid() failed\n");
 		return ret;
@@ -1212,6 +1214,7 @@ encoder_bridge_disable(struct drm_device *dev, struct drm_atomic_state *state)
 		 */
 		bridge = drm_bridge_chain_get_first_bridge(encoder);
 		drm_atomic_bridge_chain_disable(bridge, state);
+		drm_bridge_put(bridge);
 
 		/* Right function depends upon target state. */
 		if (funcs) {
@@ -1329,6 +1332,7 @@ encoder_bridge_post_disable(struct drm_device *dev, struct drm_atomic_state *sta
 		 */
 		bridge = drm_bridge_chain_get_first_bridge(encoder);
 		drm_atomic_bridge_chain_post_disable(bridge, state);
+		drm_bridge_put(bridge);
 	}
 }
 
@@ -1501,6 +1505,7 @@ crtc_set_mode(struct drm_device *dev, struct drm_atomic_state *state)
 
 		bridge = drm_bridge_chain_get_first_bridge(encoder);
 		drm_bridge_chain_mode_set(bridge, mode, adjusted_mode);
+		drm_bridge_put(bridge);
 	}
 }
 
@@ -1580,6 +1585,7 @@ encoder_bridge_pre_enable(struct drm_device *dev, struct drm_atomic_state *state
 		 */
 		bridge = drm_bridge_chain_get_first_bridge(encoder);
 		drm_atomic_bridge_chain_pre_enable(bridge, state);
+		drm_bridge_put(bridge);
 	}
 }
 
@@ -1655,6 +1661,7 @@ encoder_bridge_enable(struct drm_device *dev, struct drm_atomic_state *state)
 		}
 
 		drm_atomic_bridge_chain_enable(bridge, state);
+		drm_bridge_put(bridge);
 	}
 }
 

From 956f82e529dd283382f080273ed3b8db1e978699 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Tue, 8 Jul 2025 17:48:22 +0200
Subject: [PATCH 0055/1455] drm/probe-helper: put the bridge returned by
 drm_bridge_chain_get_first_bridge()

The bridge returned by drm_bridge_chain_get_first_bridge() is
refcounted. Put it when done.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250708-drm-bridge-alloc-getput-drm_bridge_chain_get_first_bridge-v9-5-db1ba3df7f58@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/drm_probe_helper.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c
index 6b3541159c0f..09b12c30df69 100644
--- a/drivers/gpu/drm/drm_probe_helper.c
+++ b/drivers/gpu/drm/drm_probe_helper.c
@@ -119,6 +119,7 @@ drm_mode_validate_pipeline(struct drm_display_mode *mode,
 		*status = drm_bridge_chain_mode_valid(bridge,
 						      &connector->display_info,
 						      mode);
+		drm_bridge_put(bridge);
 		if (*status != MODE_OK) {
 			/* There is also no point in continuing for crtc check
 			 * here. */

From bd72d4acda1069579b35123e3cc0b21ec1193a21 Mon Sep 17 00:00:00 2001
From: Lizhi Hou <lizhi.hou@amd.com>
Date: Wed, 16 Jul 2025 09:44:14 -0700
Subject: [PATCH 0056/1455] accel/amdxdna: Support user space allocated buffer

Enhance DRM_IOCTL_AMDXDNA_CREATE_BO to accept user space allocated
buffer pointer. The buffer pages will be pinned in memory. Unless
the CAP_IPC_LOCK is enabled for the application process, the total
pinned memory can not beyond rlimit_memlock.

Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250716164414.112091-1-lizhi.hou@amd.com
---
 drivers/accel/amdxdna/Makefile       |   1 +
 drivers/accel/amdxdna/amdxdna_gem.c  | 139 +++++++++++++---
 drivers/accel/amdxdna/amdxdna_ubuf.c | 232 +++++++++++++++++++++++++++
 drivers/accel/amdxdna/amdxdna_ubuf.h |  19 +++
 include/uapi/drm/amdxdna_accel.h     |  25 +++
 5 files changed, 391 insertions(+), 25 deletions(-)
 create mode 100644 drivers/accel/amdxdna/amdxdna_ubuf.c
 create mode 100644 drivers/accel/amdxdna/amdxdna_ubuf.h

diff --git a/drivers/accel/amdxdna/Makefile b/drivers/accel/amdxdna/Makefile
index 0e9adf6890a0..6797dac65efa 100644
--- a/drivers/accel/amdxdna/Makefile
+++ b/drivers/accel/amdxdna/Makefile
@@ -15,6 +15,7 @@ amdxdna-y := \
 	amdxdna_mailbox_helper.o \
 	amdxdna_pci_drv.o \
 	amdxdna_sysfs.o \
+	amdxdna_ubuf.o \
 	npu1_regs.o \
 	npu2_regs.o \
 	npu4_regs.o \
diff --git a/drivers/accel/amdxdna/amdxdna_gem.c b/drivers/accel/amdxdna/amdxdna_gem.c
index 0f85a0105178..d407a36eb412 100644
--- a/drivers/accel/amdxdna/amdxdna_gem.c
+++ b/drivers/accel/amdxdna/amdxdna_gem.c
@@ -18,6 +18,7 @@
 #include "amdxdna_ctx.h"
 #include "amdxdna_gem.h"
 #include "amdxdna_pci_drv.h"
+#include "amdxdna_ubuf.h"
 
 #define XDNA_MAX_CMD_BO_SIZE	SZ_32K
 
@@ -296,7 +297,7 @@ static int amdxdna_insert_pages(struct amdxdna_gem_obj *abo,
 
 	vma->vm_private_data = NULL;
 	vma->vm_ops = NULL;
-	ret = dma_buf_mmap(to_gobj(abo)->dma_buf, vma, 0);
+	ret = dma_buf_mmap(abo->dma_buf, vma, 0);
 	if (ret) {
 		XDNA_ERR(xdna, "Failed to mmap dma buf %d", ret);
 		return ret;
@@ -391,10 +392,47 @@ static const struct dma_buf_ops amdxdna_dmabuf_ops = {
 	.vunmap = drm_gem_dmabuf_vunmap,
 };
 
+static int amdxdna_gem_obj_vmap(struct drm_gem_object *obj, struct iosys_map *map)
+{
+	struct amdxdna_gem_obj *abo = to_xdna_obj(obj);
+
+	iosys_map_clear(map);
+
+	dma_resv_assert_held(obj->resv);
+
+	if (is_import_bo(abo))
+		dma_buf_vmap(abo->dma_buf, map);
+	else
+		drm_gem_shmem_object_vmap(obj, map);
+
+	if (!map->vaddr)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static void amdxdna_gem_obj_vunmap(struct drm_gem_object *obj, struct iosys_map *map)
+{
+	struct amdxdna_gem_obj *abo = to_xdna_obj(obj);
+
+	dma_resv_assert_held(obj->resv);
+
+	if (is_import_bo(abo))
+		dma_buf_vunmap(abo->dma_buf, map);
+	else
+		drm_gem_shmem_object_vunmap(obj, map);
+}
+
 static struct dma_buf *amdxdna_gem_prime_export(struct drm_gem_object *gobj, int flags)
 {
+	struct amdxdna_gem_obj *abo = to_xdna_obj(gobj);
 	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
 
+	if (abo->dma_buf) {
+		get_dma_buf(abo->dma_buf);
+		return abo->dma_buf;
+	}
+
 	exp_info.ops = &amdxdna_dmabuf_ops;
 	exp_info.size = gobj->size;
 	exp_info.flags = flags;
@@ -451,8 +489,8 @@ static const struct drm_gem_object_funcs amdxdna_gem_shmem_funcs = {
 	.pin = drm_gem_shmem_object_pin,
 	.unpin = drm_gem_shmem_object_unpin,
 	.get_sg_table = drm_gem_shmem_object_get_sg_table,
-	.vmap = drm_gem_shmem_object_vmap,
-	.vunmap = drm_gem_shmem_object_vunmap,
+	.vmap = amdxdna_gem_obj_vmap,
+	.vunmap = amdxdna_gem_obj_vunmap,
 	.mmap = amdxdna_gem_obj_mmap,
 	.vm_ops = &drm_gem_shmem_vm_ops,
 	.export = amdxdna_gem_prime_export,
@@ -494,6 +532,68 @@ amdxdna_gem_create_object_cb(struct drm_device *dev, size_t size)
 	return to_gobj(abo);
 }
 
+static struct amdxdna_gem_obj *
+amdxdna_gem_create_shmem_object(struct drm_device *dev, size_t size)
+{
+	struct drm_gem_shmem_object *shmem = drm_gem_shmem_create(dev, size);
+
+	if (IS_ERR(shmem))
+		return ERR_CAST(shmem);
+
+	shmem->map_wc = false;
+	return to_xdna_obj(&shmem->base);
+}
+
+static struct amdxdna_gem_obj *
+amdxdna_gem_create_ubuf_object(struct drm_device *dev, struct amdxdna_drm_create_bo *args)
+{
+	struct amdxdna_dev *xdna = to_xdna_dev(dev);
+	enum amdxdna_ubuf_flag flags = 0;
+	struct amdxdna_drm_va_tbl va_tbl;
+	struct drm_gem_object *gobj;
+	struct dma_buf *dma_buf;
+
+	if (copy_from_user(&va_tbl, u64_to_user_ptr(args->vaddr), sizeof(va_tbl))) {
+		XDNA_DBG(xdna, "Access va table failed");
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (va_tbl.num_entries) {
+		if (args->type == AMDXDNA_BO_CMD)
+			flags |= AMDXDNA_UBUF_FLAG_MAP_DMA;
+
+		dma_buf = amdxdna_get_ubuf(dev, flags, va_tbl.num_entries,
+					   u64_to_user_ptr(args->vaddr + sizeof(va_tbl)));
+	} else {
+		dma_buf = dma_buf_get(va_tbl.dmabuf_fd);
+	}
+
+	if (IS_ERR(dma_buf))
+		return ERR_CAST(dma_buf);
+
+	gobj = amdxdna_gem_prime_import(dev, dma_buf);
+	if (IS_ERR(gobj)) {
+		dma_buf_put(dma_buf);
+		return ERR_CAST(gobj);
+	}
+
+	dma_buf_put(dma_buf);
+
+	return to_xdna_obj(gobj);
+}
+
+static struct amdxdna_gem_obj *
+amdxdna_gem_create_object(struct drm_device *dev,
+			  struct amdxdna_drm_create_bo *args)
+{
+	size_t aligned_sz = PAGE_ALIGN(args->size);
+
+	if (args->vaddr)
+		return amdxdna_gem_create_ubuf_object(dev, args);
+
+	return amdxdna_gem_create_shmem_object(dev, aligned_sz);
+}
+
 struct drm_gem_object *
 amdxdna_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf)
 {
@@ -545,16 +645,12 @@ amdxdna_drm_alloc_shmem(struct drm_device *dev,
 			struct drm_file *filp)
 {
 	struct amdxdna_client *client = filp->driver_priv;
-	struct drm_gem_shmem_object *shmem;
 	struct amdxdna_gem_obj *abo;
 
-	shmem = drm_gem_shmem_create(dev, args->size);
-	if (IS_ERR(shmem))
-		return ERR_CAST(shmem);
+	abo = amdxdna_gem_create_object(dev, args);
+	if (IS_ERR(abo))
+		return ERR_CAST(abo);
 
-	shmem->map_wc = false;
-
-	abo = to_xdna_obj(&shmem->base);
 	abo->client = client;
 	abo->type = AMDXDNA_BO_SHMEM;
 
@@ -569,7 +665,6 @@ amdxdna_drm_create_dev_heap(struct drm_device *dev,
 	struct amdxdna_client *client = filp->driver_priv;
 	struct iosys_map map = IOSYS_MAP_INIT_VADDR(NULL);
 	struct amdxdna_dev *xdna = to_xdna_dev(dev);
-	struct drm_gem_shmem_object *shmem;
 	struct amdxdna_gem_obj *abo;
 	int ret;
 
@@ -586,14 +681,12 @@ amdxdna_drm_create_dev_heap(struct drm_device *dev,
 		goto mm_unlock;
 	}
 
-	shmem = drm_gem_shmem_create(dev, args->size);
-	if (IS_ERR(shmem)) {
-		ret = PTR_ERR(shmem);
+	abo = amdxdna_gem_create_object(dev, args);
+	if (IS_ERR(abo)) {
+		ret = PTR_ERR(abo);
 		goto mm_unlock;
 	}
 
-	shmem->map_wc = false;
-	abo = to_xdna_obj(&shmem->base);
 	abo->type = AMDXDNA_BO_DEV_HEAP;
 	abo->client = client;
 	abo->mem.dev_addr = client->xdna->dev_info->dev_mem_base;
@@ -657,7 +750,6 @@ amdxdna_drm_create_cmd_bo(struct drm_device *dev,
 {
 	struct iosys_map map = IOSYS_MAP_INIT_VADDR(NULL);
 	struct amdxdna_dev *xdna = to_xdna_dev(dev);
-	struct drm_gem_shmem_object *shmem;
 	struct amdxdna_gem_obj *abo;
 	int ret;
 
@@ -671,12 +763,9 @@ amdxdna_drm_create_cmd_bo(struct drm_device *dev,
 		return ERR_PTR(-EINVAL);
 	}
 
-	shmem = drm_gem_shmem_create(dev, args->size);
-	if (IS_ERR(shmem))
-		return ERR_CAST(shmem);
-
-	shmem->map_wc = false;
-	abo = to_xdna_obj(&shmem->base);
+	abo = amdxdna_gem_create_object(dev, args);
+	if (IS_ERR(abo))
+		return ERR_CAST(abo);
 
 	abo->type = AMDXDNA_BO_CMD;
 	abo->client = filp->driver_priv;
@@ -691,7 +780,7 @@ amdxdna_drm_create_cmd_bo(struct drm_device *dev,
 	return abo;
 
 release_obj:
-	drm_gem_shmem_free(shmem);
+	drm_gem_object_put(to_gobj(abo));
 	return ERR_PTR(ret);
 }
 
@@ -702,7 +791,7 @@ int amdxdna_drm_create_bo_ioctl(struct drm_device *dev, void *data, struct drm_f
 	struct amdxdna_gem_obj *abo;
 	int ret;
 
-	if (args->flags || args->vaddr || !args->size)
+	if (args->flags)
 		return -EINVAL;
 
 	XDNA_DBG(xdna, "BO arg type %d vaddr 0x%llx size 0x%llx flags 0x%llx",
diff --git a/drivers/accel/amdxdna/amdxdna_ubuf.c b/drivers/accel/amdxdna/amdxdna_ubuf.c
new file mode 100644
index 000000000000..077b2261cf2a
--- /dev/null
+++ b/drivers/accel/amdxdna/amdxdna_ubuf.c
@@ -0,0 +1,232 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2025, Advanced Micro Devices, Inc.
+ */
+
+#include <drm/amdxdna_accel.h>
+#include <drm/drm_device.h>
+#include <drm/drm_print.h>
+#include <linux/dma-buf.h>
+#include <linux/pagemap.h>
+#include <linux/vmalloc.h>
+
+#include "amdxdna_pci_drv.h"
+#include "amdxdna_ubuf.h"
+
+struct amdxdna_ubuf_priv {
+	struct page **pages;
+	u64 nr_pages;
+	enum amdxdna_ubuf_flag flags;
+	struct mm_struct *mm;
+};
+
+static struct sg_table *amdxdna_ubuf_map(struct dma_buf_attachment *attach,
+					 enum dma_data_direction direction)
+{
+	struct amdxdna_ubuf_priv *ubuf = attach->dmabuf->priv;
+	struct sg_table *sg;
+	int ret;
+
+	sg = kzalloc(sizeof(*sg), GFP_KERNEL);
+	if (!sg)
+		return ERR_PTR(-ENOMEM);
+
+	ret = sg_alloc_table_from_pages(sg, ubuf->pages, ubuf->nr_pages, 0,
+					ubuf->nr_pages << PAGE_SHIFT, GFP_KERNEL);
+	if (ret)
+		return ERR_PTR(ret);
+
+	if (ubuf->flags & AMDXDNA_UBUF_FLAG_MAP_DMA) {
+		ret = dma_map_sgtable(attach->dev, sg, direction, 0);
+		if (ret)
+			return ERR_PTR(ret);
+	}
+
+	return sg;
+}
+
+static void amdxdna_ubuf_unmap(struct dma_buf_attachment *attach,
+			       struct sg_table *sg,
+			       enum dma_data_direction direction)
+{
+	struct amdxdna_ubuf_priv *ubuf = attach->dmabuf->priv;
+
+	if (ubuf->flags & AMDXDNA_UBUF_FLAG_MAP_DMA)
+		dma_unmap_sgtable(attach->dev, sg, direction, 0);
+
+	sg_free_table(sg);
+	kfree(sg);
+}
+
+static void amdxdna_ubuf_release(struct dma_buf *dbuf)
+{
+	struct amdxdna_ubuf_priv *ubuf = dbuf->priv;
+
+	unpin_user_pages(ubuf->pages, ubuf->nr_pages);
+	kvfree(ubuf->pages);
+	atomic64_sub(ubuf->nr_pages, &ubuf->mm->pinned_vm);
+	mmdrop(ubuf->mm);
+	kfree(ubuf);
+}
+
+static vm_fault_t amdxdna_ubuf_vm_fault(struct vm_fault *vmf)
+{
+	struct vm_area_struct *vma = vmf->vma;
+	struct amdxdna_ubuf_priv *ubuf;
+	unsigned long pfn;
+	pgoff_t pgoff;
+
+	ubuf = vma->vm_private_data;
+	pgoff = (vmf->address - vma->vm_start) >> PAGE_SHIFT;
+
+	pfn = page_to_pfn(ubuf->pages[pgoff]);
+	return vmf_insert_pfn(vma, vmf->address, pfn);
+}
+
+static const struct vm_operations_struct amdxdna_ubuf_vm_ops = {
+	.fault = amdxdna_ubuf_vm_fault,
+};
+
+static int amdxdna_ubuf_mmap(struct dma_buf *dbuf, struct vm_area_struct *vma)
+{
+	struct amdxdna_ubuf_priv *ubuf = dbuf->priv;
+
+	vma->vm_ops = &amdxdna_ubuf_vm_ops;
+	vma->vm_private_data = ubuf;
+	vm_flags_set(vma, VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP);
+
+	return 0;
+}
+
+static int amdxdna_ubuf_vmap(struct dma_buf *dbuf, struct iosys_map *map)
+{
+	struct amdxdna_ubuf_priv *ubuf = dbuf->priv;
+	void *kva;
+
+	kva = vmap(ubuf->pages, ubuf->nr_pages, VM_MAP, PAGE_KERNEL);
+	if (!kva)
+		return -EINVAL;
+
+	iosys_map_set_vaddr(map, kva);
+	return 0;
+}
+
+static void amdxdna_ubuf_vunmap(struct dma_buf *dbuf, struct iosys_map *map)
+{
+	vunmap(map->vaddr);
+}
+
+static const struct dma_buf_ops amdxdna_ubuf_dmabuf_ops = {
+	.map_dma_buf = amdxdna_ubuf_map,
+	.unmap_dma_buf = amdxdna_ubuf_unmap,
+	.release = amdxdna_ubuf_release,
+	.mmap = amdxdna_ubuf_mmap,
+	.vmap = amdxdna_ubuf_vmap,
+	.vunmap = amdxdna_ubuf_vunmap,
+};
+
+struct dma_buf *amdxdna_get_ubuf(struct drm_device *dev,
+				 enum amdxdna_ubuf_flag flags,
+				 u32 num_entries, void __user *va_entries)
+{
+	struct amdxdna_dev *xdna = to_xdna_dev(dev);
+	unsigned long lock_limit, new_pinned;
+	struct amdxdna_drm_va_entry *va_ent;
+	struct amdxdna_ubuf_priv *ubuf;
+	u32 npages, start = 0;
+	struct dma_buf *dbuf;
+	int i, ret;
+	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
+
+	if (!can_do_mlock())
+		return ERR_PTR(-EPERM);
+
+	ubuf = kzalloc(sizeof(*ubuf), GFP_KERNEL);
+	if (!ubuf)
+		return ERR_PTR(-ENOMEM);
+
+	ubuf->flags = flags;
+	ubuf->mm = current->mm;
+	mmgrab(ubuf->mm);
+
+	va_ent = kvcalloc(num_entries, sizeof(*va_ent), GFP_KERNEL);
+	if (!va_ent) {
+		ret = -ENOMEM;
+		goto free_ubuf;
+	}
+
+	if (copy_from_user(va_ent, va_entries, sizeof(*va_ent) * num_entries)) {
+		XDNA_DBG(xdna, "Access va entries failed");
+		ret = -EINVAL;
+		goto free_ent;
+	}
+
+	for (i = 0, exp_info.size = 0; i < num_entries; i++) {
+		if (!IS_ALIGNED(va_ent[i].vaddr, PAGE_SIZE) ||
+		    !IS_ALIGNED(va_ent[i].len, PAGE_SIZE)) {
+			XDNA_ERR(xdna, "Invalid address or len %llx, %llx",
+				 va_ent[i].vaddr, va_ent[i].len);
+			ret = -EINVAL;
+			goto free_ent;
+		}
+
+		exp_info.size += va_ent[i].len;
+	}
+
+	ubuf->nr_pages = exp_info.size >> PAGE_SHIFT;
+	lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
+	new_pinned = atomic64_add_return(ubuf->nr_pages, &ubuf->mm->pinned_vm);
+	if (new_pinned > lock_limit && !capable(CAP_IPC_LOCK)) {
+		XDNA_DBG(xdna, "New pin %ld, limit %ld, cap %d",
+			 new_pinned, lock_limit, capable(CAP_IPC_LOCK));
+		ret = -ENOMEM;
+		goto sub_pin_cnt;
+	}
+
+	ubuf->pages = kvmalloc_array(ubuf->nr_pages, sizeof(*ubuf->pages), GFP_KERNEL);
+	if (!ubuf->pages) {
+		ret = -ENOMEM;
+		goto sub_pin_cnt;
+	}
+
+	for (i = 0; i < num_entries; i++) {
+		npages = va_ent[i].len >> PAGE_SHIFT;
+
+		ret = pin_user_pages_fast(va_ent[i].vaddr, npages,
+					  FOLL_WRITE | FOLL_LONGTERM,
+					  &ubuf->pages[start]);
+		if (ret < 0 || ret != npages) {
+			ret = -ENOMEM;
+			XDNA_ERR(xdna, "Failed to pin pages ret %d", ret);
+			goto destroy_pages;
+		}
+
+		start += ret;
+	}
+
+	exp_info.ops = &amdxdna_ubuf_dmabuf_ops;
+	exp_info.priv = ubuf;
+	exp_info.flags = O_RDWR | O_CLOEXEC;
+
+	dbuf = dma_buf_export(&exp_info);
+	if (IS_ERR(dbuf)) {
+		ret = PTR_ERR(dbuf);
+		goto destroy_pages;
+	}
+	kvfree(va_ent);
+
+	return dbuf;
+
+destroy_pages:
+	if (start)
+		unpin_user_pages(ubuf->pages, start);
+	kvfree(ubuf->pages);
+sub_pin_cnt:
+	atomic64_sub(ubuf->nr_pages, &ubuf->mm->pinned_vm);
+free_ent:
+	kvfree(va_ent);
+free_ubuf:
+	mmdrop(ubuf->mm);
+	kfree(ubuf);
+	return ERR_PTR(ret);
+}
diff --git a/drivers/accel/amdxdna/amdxdna_ubuf.h b/drivers/accel/amdxdna/amdxdna_ubuf.h
new file mode 100644
index 000000000000..e5cb3bdb3ec9
--- /dev/null
+++ b/drivers/accel/amdxdna/amdxdna_ubuf.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2025, Advanced Micro Devices, Inc.
+ */
+#ifndef _AMDXDNA_UBUF_H_
+#define _AMDXDNA_UBUF_H_
+
+#include <drm/drm_device.h>
+#include <linux/dma-buf.h>
+
+enum amdxdna_ubuf_flag {
+	AMDXDNA_UBUF_FLAG_MAP_DMA = 1,
+};
+
+struct dma_buf *amdxdna_get_ubuf(struct drm_device *dev,
+				 enum amdxdna_ubuf_flag flags,
+				 u32 num_entries, void __user *va_entries);
+
+#endif /* _AMDXDNA_UBUF_H_ */
diff --git a/include/uapi/drm/amdxdna_accel.h b/include/uapi/drm/amdxdna_accel.h
index a706ead39082..ce523e9ccc52 100644
--- a/include/uapi/drm/amdxdna_accel.h
+++ b/include/uapi/drm/amdxdna_accel.h
@@ -153,6 +153,31 @@ enum amdxdna_bo_type {
 	AMDXDNA_BO_CMD,
 };
 
+/**
+ * struct amdxdna_drm_va_entry
+ * @vaddr: Virtual address.
+ * @len: Size of entry.
+ */
+struct amdxdna_drm_va_entry {
+	__u64 vaddr;
+	__u64 len;
+};
+
+/**
+ * struct amdxdna_drm_va_tbl
+ * @dmabuf_fd: The fd of dmabuf.
+ * @num_entries: Number of va entries.
+ * @va_entries: Array of va entries.
+ *
+ * The input can be either a dmabuf fd or a virtual address entry table.
+ * When dmabuf_fd is used, num_entries must be zero.
+ */
+struct amdxdna_drm_va_tbl {
+	__s32 dmabuf_fd;
+	__u32 num_entries;
+	struct amdxdna_drm_va_entry va_entries[];
+};
+
 /**
  * struct amdxdna_drm_create_bo - Create a buffer object.
  * @flags: Buffer flags. MBZ.

From 5982a539cdcedcf1a87709c01125e1596eee2c9a Mon Sep 17 00:00:00 2001
From: Salah Triki <salah.triki@gmail.com>
Date: Sat, 19 Jul 2025 07:33:36 +0100
Subject: [PATCH 0057/1455] accel/amdxdna: Delete pci_free_irq_vectors()

The device is managed so pci_free_irq_vectors() is called automatically
no need to do it manually.

Reviewed-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Signed-off-by: Salah Triki <salah.triki@gmail.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/aHs8QAfUlFeNp7qL@pc
---
 drivers/accel/amdxdna/aie2_pci.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_pci.c b/drivers/accel/amdxdna/aie2_pci.c
index c6cf7068d23c..3474a8d4e560 100644
--- a/drivers/accel/amdxdna/aie2_pci.c
+++ b/drivers/accel/amdxdna/aie2_pci.c
@@ -520,14 +520,14 @@ static int aie2_init(struct amdxdna_dev *xdna)
 	if (!ndev->psp_hdl) {
 		XDNA_ERR(xdna, "failed to create psp");
 		ret = -ENOMEM;
-		goto free_irq;
+		goto release_fw;
 	}
 	xdna->dev_handle = ndev;
 
 	ret = aie2_hw_start(xdna);
 	if (ret) {
 		XDNA_ERR(xdna, "start npu failed, ret %d", ret);
-		goto free_irq;
+		goto release_fw;
 	}
 
 	ret = aie2_mgmt_fw_query(ndev);
@@ -578,8 +578,6 @@ async_event_free:
 	aie2_error_async_events_free(ndev);
 stop_hw:
 	aie2_hw_stop(xdna);
-free_irq:
-	pci_free_irq_vectors(pdev);
 release_fw:
 	release_firmware(fw);
 
@@ -588,12 +586,10 @@ release_fw:
 
 static void aie2_fini(struct amdxdna_dev *xdna)
 {
-	struct pci_dev *pdev = to_pci_dev(xdna->ddev.dev);
 	struct amdxdna_dev_hdl *ndev = xdna->dev_handle;
 
 	aie2_hw_stop(xdna);
 	aie2_error_async_events_free(ndev);
-	pci_free_irq_vectors(pdev);
 }
 
 static int aie2_get_aie_status(struct amdxdna_client *client,

From 176f44a5ec0b074aaf44852db77d0c183c36696d Mon Sep 17 00:00:00 2001
From: Zhanjun Dong <zhanjun.dong@intel.com>
Date: Mon, 21 Jul 2025 17:45:20 -0400
Subject: [PATCH 0058/1455] drm/xe/uc: Fix missing unwind goto

Fix missing unwind goto on error handling.

Fixes: b2c4ac219fa4 ("drm/xe/uc: Disable GuC communication on hardware initialization error")
Signed-off-by: Zhanjun Dong <zhanjun.dong@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250721214520.954014-1-zhanjun.dong@intel.com
---
 drivers/gpu/drm/xe/xe_uc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_uc.c b/drivers/gpu/drm/xe/xe_uc.c
index 3e0c3af235f2..465bda355443 100644
--- a/drivers/gpu/drm/xe/xe_uc.c
+++ b/drivers/gpu/drm/xe/xe_uc.c
@@ -164,7 +164,7 @@ static int vf_uc_load_hw(struct xe_uc *uc)
 
 	err = xe_guc_opt_in_features_enable(&uc->guc);
 	if (err)
-		return err;
+		goto err_out;
 
 	err = xe_gt_record_default_lrcs(uc_to_gt(uc));
 	if (err)

From 8265ce0e0e15ba435eb2af72f2b821e203ebcdb9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 22 Jul 2025 15:56:15 +0300
Subject: [PATCH 0059/1455] drm/i915/display: Write PHY_CMN1_CONTROL only when
 using AUXLess ALPM
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are seeing "dmesg-warn/abort - *ERROR* PHY * failed after 3 retries"
since we started configuring LFPS sending. According to Bspec Configuring
LFPS sending is needed only when using AUXLess ALPM. This patch avoids
these failures by configuring LFPS sending only when using AUXLess ALPM.

Bspec: 68849
Fixes: 9dc619680de4 ("drm/i915/display: Add function to configure LFPS sending")
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250722125618.1842615-2-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index ed8e640b96b0..2b0305bb04cd 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -3240,11 +3240,10 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 {
 	struct intel_display *display = to_intel_display(encoder);
 	u8 owned_lane_mask = intel_cx0_get_owned_lane_mask(encoder);
-	bool enable = intel_alpm_is_alpm_aux_less(enc_to_intel_dp(encoder),
-						  crtc_state);
 	int i;
 
-	if (DISPLAY_VER(display) < 20)
+	if (DISPLAY_VER(display) < 20 ||
+	    !intel_alpm_is_alpm_aux_less(enc_to_intel_dp(encoder), crtc_state))
 		return;
 
 	for (i = 0; i < 4; i++) {
@@ -3256,8 +3255,7 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 
 		intel_cx0_rmw(encoder, lane_mask, PHY_CMN1_CONTROL(tx, 0),
 			      CONTROL0_MAC_TRANSMIT_LFPS,
-			      enable ? CONTROL0_MAC_TRANSMIT_LFPS : 0,
-			      MB_WRITE_COMMITTED);
+			      CONTROL0_MAC_TRANSMIT_LFPS, MB_WRITE_COMMITTED);
 	}
 }
 

From d487ed7e2b4ab3126239ab93324405eb1e45ccf5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 22 Jul 2025 15:56:16 +0300
Subject: [PATCH 0060/1455] drm/i915/display: Avoid unnecessarily calling
 intel_cx0_get_owned_lane_mask
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we are always calling intel_cx0_get_owned_lane_mask when
intel_lnl_mac_transmit_lfps is called. Avoid this in cases where it's not
needed.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250722125618.1842615-3-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 2b0305bb04cd..2993dabdbee6 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -3239,13 +3239,14 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
-	u8 owned_lane_mask = intel_cx0_get_owned_lane_mask(encoder);
 	int i;
+	u8 owned_lane_mask;
 
 	if (DISPLAY_VER(display) < 20 ||
 	    !intel_alpm_is_alpm_aux_less(enc_to_intel_dp(encoder), crtc_state))
 		return;
 
+	owned_lane_mask = intel_cx0_get_owned_lane_mask(encoder);
 	for (i = 0; i < 4; i++) {
 		int tx = i % 2 + 1;
 		u8 lane_mask = i < 2 ? INTEL_CX0_LANE0 : INTEL_CX0_LANE1;

From cf433f94f188782166598300c4c05274fd13c5a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 22 Jul 2025 15:56:17 +0300
Subject: [PATCH 0061/1455] drm/i915/display: Ensure phy is accessible on lfps
 configuration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Ensure phy is accessible on lfps configuration by adding
intel_cx0_phy_transaction_begin/end around it.

Fixes: 9dc619680de4 ("drm/i915/display: Add function to configure LFPS sending")
Suggested-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250722125618.1842615-4-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index 2993dabdbee6..a203937d66db 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -3239,6 +3239,7 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 				 const struct intel_crtc_state *crtc_state)
 {
 	struct intel_display *display = to_intel_display(encoder);
+	intel_wakeref_t wakeref;
 	int i;
 	u8 owned_lane_mask;
 
@@ -3247,6 +3248,9 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 		return;
 
 	owned_lane_mask = intel_cx0_get_owned_lane_mask(encoder);
+
+	wakeref = intel_cx0_phy_transaction_begin(encoder);
+
 	for (i = 0; i < 4; i++) {
 		int tx = i % 2 + 1;
 		u8 lane_mask = i < 2 ? INTEL_CX0_LANE0 : INTEL_CX0_LANE1;
@@ -3258,6 +3262,8 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 			      CONTROL0_MAC_TRANSMIT_LFPS,
 			      CONTROL0_MAC_TRANSMIT_LFPS, MB_WRITE_COMMITTED);
 	}
+
+	intel_cx0_phy_transaction_end(encoder, wakeref);
 }
 
 static u8 cx0_power_control_disable_val(struct intel_encoder *encoder)

From 8921dce70d46e3156b5a0b21675f5ac90903d81d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 22 Jul 2025 15:56:18 +0300
Subject: [PATCH 0062/1455] drm/i915/display: Set C10_VDR_CTRL_MSGBUS_ACCESS
 before phy reg read
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

According to C10 VDR Register programming sequence we need set
C10_VDR_CTRL_MSGBUS_ACCESS before accessing PHY internal registers from
MsgBus.

v2: set C10_VDR_CTRL_MSGBUS_ACCESS once for all owned lanes

Bspec: 68962
Fixes: 9dc619680de4 ("drm/i915/display: Add function to configure LFPS sending")
Suggested-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250722125618.1842615-5-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_cx0_phy.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_cx0_phy.c b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
index a203937d66db..801235a5bc0a 100644
--- a/drivers/gpu/drm/i915/display/intel_cx0_phy.c
+++ b/drivers/gpu/drm/i915/display/intel_cx0_phy.c
@@ -3251,6 +3251,10 @@ void intel_lnl_mac_transmit_lfps(struct intel_encoder *encoder,
 
 	wakeref = intel_cx0_phy_transaction_begin(encoder);
 
+	if (intel_encoder_is_c10phy(encoder))
+		intel_cx0_rmw(encoder, owned_lane_mask, PHY_C10_VDR_CONTROL(1), 0,
+			      C10_VDR_CTRL_MSGBUS_ACCESS, MB_WRITE_COMMITTED);
+
 	for (i = 0; i < 4; i++) {
 		int tx = i % 2 + 1;
 		u8 lane_mask = i < 2 ? INTEL_CX0_LANE0 : INTEL_CX0_LANE1;

From 46c366851a0b2159bdb97afcac04e5dec0cf09e8 Mon Sep 17 00:00:00 2001
From: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Date: Tue, 22 Jul 2025 12:04:21 +0200
Subject: [PATCH 0063/1455] MAINTAINERS: Add new intel_vpu maintainer

Add Karol as a new intel_vpu maintainer.

Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://lore.kernel.org/r/20250722100421.500984-1-jacek.lawrynowicz@linux.intel.com
---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index d97e091b1742..5ddf37f0acc9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7386,6 +7386,7 @@ F:	include/linux/power/smartreflex.h
 DRM ACCEL DRIVERS FOR INTEL VPU
 M:	Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
 M:	Maciej Falkowski <maciej.falkowski@linux.intel.com>
+M:	Karol Wachowski <karol.wachowski@linux.intel.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Supported
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git

From cdb16039515a5ac4d2c923f7a651cf19a803a3fe Mon Sep 17 00:00:00 2001
From: Aakash Deep Sarkar <aakash.deep.sarkar@intel.com>
Date: Tue, 8 Jul 2025 07:45:40 +0000
Subject: [PATCH 0064/1455] drm/i915/display: Fix dma_fence_wait_timeout()
 return value handling

dma_fence_wait_timeout returns a long type but the driver is
only using the lower 32 bits of the retval and discarding the
upper 32 bits.

This is particularly problematic if there are already signalled
or stub fences on some of the hw planes. In this case the
dma_fence_wait_timeout function will immediately return with
timeout value MAX_SCHEDULE_TIMEOUT (0x7fffffffffffffff) since
the fence is already signalled. If the driver only uses the lower
32 bits of this return value then it'll interpret it as an error
code (0xFFFFFFFF or (-1)) and skip the wait on the remaining fences.

This issue was first observed in the xe driver with the Android
compositor where the GPU composited layer was not properly waited
on when there were stub fences in other overlay planes resulting in
visual artifacts.

Fixes: d59cf7bb73f3c ("drm/i915/display: Use dma_fence interfaces instead of i915_sw_fence")
Signed-off-by: Aakash Deep Sarkar <aakash.deep.sarkar@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250708074540.1948068-1-aakash.deep.sarkar@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 456fc4b04cda..7035c1fc9033 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -7092,7 +7092,8 @@ static void intel_atomic_commit_fence_wait(struct intel_atomic_state *intel_stat
 	struct drm_i915_private *i915 = to_i915(intel_state->base.dev);
 	struct drm_plane *plane;
 	struct drm_plane_state *new_plane_state;
-	int ret, i;
+	long ret;
+	int i;
 
 	for_each_new_plane_in_state(&intel_state->base, plane, new_plane_state, i) {
 		if (new_plane_state->fence) {

From 9b75346e3c2b8ecb5b90b132c2fc185ddd30ecf3 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Wed, 9 Jul 2025 17:59:37 +0200
Subject: [PATCH 0065/1455] drm/bridge: get the bridge returned by
 drm_bridge_get_prev_bridge()

drm_bridge_get_prev_bridge() returns a bridge pointer that the
caller could hold for a long time. Increment the refcount of the returned
bridge and document it must be put by the caller.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250709-drm-bridge-alloc-getput-drm_bridge_get_prev_bridge-v1-1-34ba6f395aaa@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 include/drm/drm_bridge.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index 717171d0e587..620e119cc24c 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -1321,6 +1321,13 @@ drm_bridge_get_next_bridge(struct drm_bridge *bridge)
  * drm_bridge_get_prev_bridge() - Get the previous bridge in the chain
  * @bridge: bridge object
  *
+ * The caller is responsible of having a reference to @bridge via
+ * drm_bridge_get() or equivalent. This function leaves the refcount of
+ * @bridge unmodified.
+ *
+ * The refcount of the returned bridge is incremented. Use drm_bridge_put()
+ * when done with it.
+ *
  * RETURNS:
  * the previous bridge in the chain, or NULL if @bridge is the first.
  */
@@ -1330,7 +1337,7 @@ drm_bridge_get_prev_bridge(struct drm_bridge *bridge)
 	if (list_is_first(&bridge->chain_node, &bridge->encoder->bridge_chain))
 		return NULL;
 
-	return list_prev_entry(bridge, chain_node);
+	return drm_bridge_get(list_prev_entry(bridge, chain_node));
 }
 
 /**

From d4eecb4c24dc160f4a003c804602c746fb8fec58 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Wed, 9 Jul 2025 17:59:38 +0200
Subject: [PATCH 0066/1455] drm/bridge: select_bus_fmt_recursive(): put the
 bridge obtained by drm_bridge_get_prev_bridge()

The bridge returned by drm_bridge_get_prev_bridge() is refcounted. Put it
when done.

select_bus_fmt_recursive() has several return points, and ensuring
drm_bridge_put() is always called in the right place would be error-prone
(especially with future changes to the select_bus_fmt_recursive() code) and
make code uglier. Instead use a scope-based free, which is future-proof and
a lot cleaner.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250709-drm-bridge-alloc-getput-drm_bridge_get_prev_bridge-v1-2-34ba6f395aaa@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/drm_bridge.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c
index dd45d9b504d8..c3bfcd735a3c 100644
--- a/drivers/gpu/drm/drm_bridge.c
+++ b/drivers/gpu/drm/drm_bridge.c
@@ -941,11 +941,11 @@ static int select_bus_fmt_recursive(struct drm_bridge *first_bridge,
 {
 	unsigned int i, num_in_bus_fmts = 0;
 	struct drm_bridge_state *cur_state;
-	struct drm_bridge *prev_bridge;
+	struct drm_bridge *prev_bridge __free(drm_bridge_put) =
+		drm_bridge_get_prev_bridge(cur_bridge);
 	u32 *in_bus_fmts;
 	int ret;
 
-	prev_bridge = drm_bridge_get_prev_bridge(cur_bridge);
 	cur_state = drm_atomic_get_new_bridge_state(crtc_state->state,
 						    cur_bridge);
 

From c571cb70e1ed43ee543c70151e61a001ab2eefa2 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Wed, 9 Jul 2025 17:59:39 +0200
Subject: [PATCH 0067/1455] drm/bridge: display-connector: put the bridge
 obtained by drm_bridge_get_prev_bridge()

The bridge returned by drm_bridge_get_prev_bridge() is refcounted. Put it
when done.

To keep the code clean and future-proof use a scope-based free.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250709-drm-bridge-alloc-getput-drm_bridge_get_prev_bridge-v1-3-34ba6f395aaa@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/bridge/display-connector.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c
index 52b7b5889e6f..d7e1c2f8f53c 100644
--- a/drivers/gpu/drm/bridge/display-connector.c
+++ b/drivers/gpu/drm/bridge/display-connector.c
@@ -108,7 +108,7 @@ static u32 *display_connector_get_output_bus_fmts(struct drm_bridge *bridge,
 					struct drm_connector_state *conn_state,
 					unsigned int *num_output_fmts)
 {
-	struct drm_bridge *prev_bridge = drm_bridge_get_prev_bridge(bridge);
+	struct drm_bridge *prev_bridge __free(drm_bridge_put) = drm_bridge_get_prev_bridge(bridge);
 	struct drm_bridge_state *prev_bridge_state;
 
 	if (!prev_bridge || !prev_bridge->funcs->atomic_get_output_bus_fmts) {
@@ -151,7 +151,7 @@ static u32 *display_connector_get_input_bus_fmts(struct drm_bridge *bridge,
 					u32 output_fmt,
 					unsigned int *num_input_fmts)
 {
-	struct drm_bridge *prev_bridge = drm_bridge_get_prev_bridge(bridge);
+	struct drm_bridge *prev_bridge __free(drm_bridge_put) = drm_bridge_get_prev_bridge(bridge);
 	struct drm_bridge_state *prev_bridge_state;
 
 	if (!prev_bridge || !prev_bridge->funcs->atomic_get_input_bus_fmts) {

From 9a220e065914b67b55d3d0ab91c3e215742fdd73 Mon Sep 17 00:00:00 2001
From: Lukasz Laguna <lukasz.laguna@intel.com>
Date: Thu, 17 Jul 2025 17:54:20 +0200
Subject: [PATCH 0068/1455] drm/xe/vf: Don't register I2C devices if VF

VF drivers can't access I2C devices, so skip their registration when
running as VF.

Signed-off-by: Lukasz Laguna <lukasz.laguna@intel.com>
Fixes: f0e53aadd702 ("drm/xe: Support for I2C attached MCUs")
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250717155420.25298-1-lukasz.laguna@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_i2c.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_i2c.c b/drivers/gpu/drm/xe/xe_i2c.c
index 1f19718db559..bc7dc2099470 100644
--- a/drivers/gpu/drm/xe/xe_i2c.c
+++ b/drivers/gpu/drm/xe/xe_i2c.c
@@ -283,6 +283,9 @@ int xe_i2c_probe(struct xe_device *xe)
 	if (xe->info.platform != XE_BATTLEMAGE)
 		return 0;
 
+	if (IS_SRIOV_VF(xe))
+		return 0;
+
 	xe_i2c_read_endpoint(xe_root_tile_mmio(xe), &ep);
 	if (ep.cookie != XE_I2C_EP_COOKIE_DEVICE)
 		return 0;

From 92e34a5241ddf4b084df20e6953275d16f156aa8 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Sat, 19 Jul 2025 13:58:13 +0300
Subject: [PATCH 0069/1455] drm/display: bridge-connector: correct CEC bridge
 pointers in drm_bridge_connector_init

The bridge used in drm_bridge_connector_init() for CEC init does not
correctly point to the required HDMI CEC bridge, which can lead to
errors during CEC initialization.

Fixes: 65a2575a68e4 ("drm/display: bridge-connector: hook in CEC notifier support")
Fixes: a74288c8ded7 ("drm/display: bridge-connector: handle CEC adapters")
Reported-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Closes: http://lore.kernel.org/r/20250718164156.194702d9@booty/
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Link: https://lore.kernel.org/r/20250719-fix-cec-bridges-v1-1-a60b1333c87d@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/display/drm_bridge_connector.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c
index 3e4d561a3a32..51f30307d445 100644
--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -813,6 +813,8 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 
 	if (bridge_connector->bridge_hdmi_cec &&
 	    bridge_connector->bridge_hdmi_cec->ops & DRM_BRIDGE_OP_HDMI_CEC_NOTIFIER) {
+		bridge = bridge_connector->bridge_hdmi_cec;
+
 		ret = drmm_connector_hdmi_cec_notifier_register(connector,
 								NULL,
 								bridge->hdmi_cec_dev);
@@ -822,6 +824,8 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 
 	if (bridge_connector->bridge_hdmi_cec &&
 	    bridge_connector->bridge_hdmi_cec->ops & DRM_BRIDGE_OP_HDMI_CEC_ADAPTER) {
+		bridge = bridge_connector->bridge_hdmi_cec;
+
 		ret = drmm_connector_hdmi_cec_register(connector,
 						       &drm_bridge_connector_hdmi_cec_funcs,
 						       bridge->hdmi_cec_adapter_name,

From f3009272ff2ec0498426001e4dd2a12c50234e08 Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Tue, 22 Jul 2025 17:35:04 +0530
Subject: [PATCH 0070/1455] drm/xe/vf: Create contexts for CCS read write
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Create two LRCs to handle CCS meta data read / write from CCS pool in the
VM. Read context is used to hold GPU instructions to be executed at save
time and write context is used to hold GPU instructions to be executed at
the restore time.

Allocate batch buffer pool using suballocator for both read and write
contexts.

Migration framework is reused to create LRCAs for read and write.

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250722120506.6483-2-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/Makefile                |   1 +
 drivers/gpu/drm/xe/xe_device.c             |   4 +
 drivers/gpu/drm/xe/xe_device_types.h       |   4 +
 drivers/gpu/drm/xe/xe_gt_debugfs.c         |  39 ++++
 drivers/gpu/drm/xe/xe_sriov.c              |  19 ++
 drivers/gpu/drm/xe/xe_sriov.h              |   1 +
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       | 208 +++++++++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.h       |  13 ++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h |  45 +++++
 drivers/gpu/drm/xe/xe_sriov_vf_types.h     |   6 +
 10 files changed, 340 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
 create mode 100644 drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
 create mode 100644 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 74231ca3fb9a..b046e1b8d726 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -149,6 +149,7 @@ xe-y += \
 	xe_memirq.o \
 	xe_sriov.o \
 	xe_sriov_vf.o \
+	xe_sriov_vf_ccs.o \
 	xe_tile_sriov_vf.o
 
 xe-$(CONFIG_PCI_IOV) += \
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 08f45777f797..d04a0ae018e6 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -940,6 +940,10 @@ int xe_device_probe(struct xe_device *xe)
 
 	xe_vsec_init(xe);
 
+	err = xe_sriov_late_init(xe);
+	if (err)
+		goto err_unregister_display;
+
 	return devm_add_action_or_reset(xe->drm.dev, xe_device_sanitize, xe);
 
 err_unregister_display:
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index eb6105523f23..38c8329b4d2c 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -23,6 +23,7 @@
 #include "xe_sriov_pf_types.h"
 #include "xe_sriov_types.h"
 #include "xe_sriov_vf_types.h"
+#include "xe_sriov_vf_ccs_types.h"
 #include "xe_step_types.h"
 #include "xe_survivability_mode_types.h"
 
@@ -182,6 +183,9 @@ struct xe_tile {
 		struct {
 			/** @sriov.vf.ggtt_balloon: GGTT regions excluded from use. */
 			struct xe_ggtt_node *ggtt_balloon[2];
+
+			/** @sriov.vf.ccs: CCS read and write contexts for VF. */
+			struct xe_tile_vf_ccs ccs[XE_SRIOV_VF_CCS_CTX_COUNT];
 		} vf;
 	} sriov;
 
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index 061a81f82e76..8d4b66ca38b7 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -134,6 +134,33 @@ static int sa_info(struct xe_gt *gt, struct drm_printer *p)
 	return 0;
 }
 
+static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p)
+{
+	struct xe_tile *tile = gt_to_tile(gt);
+	struct xe_sa_manager *bb_pool;
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+
+	if (!IS_VF_CCS_READY(gt_to_xe(gt)))
+		return 0;
+
+	xe_pm_runtime_get(gt_to_xe(gt));
+
+	for_each_ccs_rw_ctx(ctx_id) {
+		bb_pool = tile->sriov.vf.ccs[ctx_id].mem.ccs_bb_pool;
+		if (!bb_pool)
+			break;
+
+		drm_printf(p, "ccs %s bb suballoc info\n", ctx_id ? "write" : "read");
+		drm_printf(p, "-------------------------\n");
+		drm_suballoc_dump_debug_info(&bb_pool->base, p, bb_pool->gpu_addr);
+		drm_puts(p, "\n");
+	}
+
+	xe_pm_runtime_put(gt_to_xe(gt));
+
+	return 0;
+}
+
 static int topology(struct xe_gt *gt, struct drm_printer *p)
 {
 	xe_pm_runtime_get(gt_to_xe(gt));
@@ -303,6 +330,13 @@ static const struct drm_info_list vf_safe_debugfs_list[] = {
 	{"hwconfig", .show = xe_gt_debugfs_simple_show, .data = hwconfig},
 };
 
+/*
+ * only for GT debugfs files which are valid on VF. Not valid on PF.
+ */
+static const struct drm_info_list vf_only_debugfs_list[] = {
+	{"sa_info_vf_ccs", .show = xe_gt_debugfs_simple_show, .data = sa_info_vf_ccs},
+};
+
 /* everything else should be added here */
 static const struct drm_info_list pf_only_debugfs_list[] = {
 	{"hw_engines", .show = xe_gt_debugfs_simple_show, .data = hw_engines},
@@ -424,6 +458,11 @@ void xe_gt_debugfs_register(struct xe_gt *gt)
 		drm_debugfs_create_files(pf_only_debugfs_list,
 					 ARRAY_SIZE(pf_only_debugfs_list),
 					 root, minor);
+	else
+		drm_debugfs_create_files(vf_only_debugfs_list,
+					 ARRAY_SIZE(vf_only_debugfs_list),
+					 root, minor);
+
 
 	xe_uc_debugfs_register(&gt->uc, root);
 
diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c
index a0eab44c0e76..87911fb4eea7 100644
--- a/drivers/gpu/drm/xe/xe_sriov.c
+++ b/drivers/gpu/drm/xe/xe_sriov.c
@@ -15,6 +15,7 @@
 #include "xe_sriov.h"
 #include "xe_sriov_pf.h"
 #include "xe_sriov_vf.h"
+#include "xe_sriov_vf_ccs.h"
 
 /**
  * xe_sriov_mode_to_string - Convert enum value to string.
@@ -157,3 +158,21 @@ const char *xe_sriov_function_name(unsigned int n, char *buf, size_t size)
 		strscpy(buf, "PF", size);
 	return buf;
 }
+
+/**
+ * xe_sriov_late_init() - SR-IOV late initialization functions.
+ * @xe: the &xe_device to initialize
+ *
+ * On VF this function will initialize code for CCS migration.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int xe_sriov_late_init(struct xe_device *xe)
+{
+	int err = 0;
+
+	if (IS_VF_CCS_INIT_NEEDED(xe))
+		err = xe_sriov_vf_ccs_init(xe);
+
+	return err;
+}
diff --git a/drivers/gpu/drm/xe/xe_sriov.h b/drivers/gpu/drm/xe/xe_sriov.h
index 688fbabf08f1..0e0c1abf2d14 100644
--- a/drivers/gpu/drm/xe/xe_sriov.h
+++ b/drivers/gpu/drm/xe/xe_sriov.h
@@ -18,6 +18,7 @@ const char *xe_sriov_function_name(unsigned int n, char *buf, size_t len);
 void xe_sriov_probe_early(struct xe_device *xe);
 void xe_sriov_print_info(struct xe_device *xe, struct drm_printer *p);
 int xe_sriov_init(struct xe_device *xe);
+int xe_sriov_late_init(struct xe_device *xe);
 
 static inline enum xe_sriov_mode xe_device_sriov_mode(const struct xe_device *xe)
 {
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
new file mode 100644
index 000000000000..697eea46659a
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -0,0 +1,208 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "instructions/xe_mi_commands.h"
+#include "instructions/xe_gpu_commands.h"
+#include "xe_bo.h"
+#include "xe_device.h"
+#include "xe_migrate.h"
+#include "xe_sa.h"
+#include "xe_sriov_printk.h"
+#include "xe_sriov_vf_ccs.h"
+#include "xe_sriov_vf_ccs_types.h"
+
+/**
+ * DOC: VF save/restore of compression Meta Data
+ *
+ * VF KMD registers two special contexts/LRCAs.
+ *
+ * Save Context/LRCA: contain necessary cmds+page table to trigger Meta data /
+ * compression control surface (Aka CCS) save in regular System memory in VM.
+ *
+ * Restore Context/LRCA: contain necessary cmds+page table to trigger Meta data /
+ * compression control surface (Aka CCS) Restore from regular System memory in
+ * VM to corresponding CCS pool.
+ *
+ * Below diagram explain steps needed for VF save/Restore of compression Meta Data::
+ *
+ *    CCS Save    CCS Restore          VF KMD                          Guc       BCS
+ *     LRCA        LRCA
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |     Create Save LRCA            |                              |         |
+ *     [ ]<----------------------------- [ ]                             |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |       Register save LRCA     |         |
+ *      |           |                     |           with Guc           |         |
+ *      |           |                    [ ]--------------------------->[ ]        |
+ *      |           |                     |                              |         |
+ *      |           | Create restore LRCA |                              |         |
+ *      |          [ ]<------------------[ ]                             |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |       Register restore LRCA  |         |
+ *      |           |                     |           with Guc           |         |
+ *      |           |                    [ ]--------------------------->[ ]        |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                    [ ]-------------------------    |         |
+ *      |           |                    [ ]  Allocate main memory.  |   |         |
+ *      |           |                    [ ]  Allocate CCS memory.   |   |         |
+ *      |           |                    [ ]  Update Main memory &   |   |         |
+ *     [ ]<------------------------------[ ]  CCS pages PPGTT + BB   |   |         |
+ *      |          [ ]<------------------[ ]  cmds to save & restore.|   |         |
+ *      |           |                    [ ]<------------------------    |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      :           :                     :                              :         :
+ *      ---------------------------- VF Paused -------------------------------------
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |Schedule |
+ *      |           |                     |                              |CCS Save |
+ *      |           |                     |                              | LRCA    |
+ *      |           |                     |                             [ ]------>[ ]
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |CCS save |
+ *      |           |                     |                              |completed|
+ *      |           |                     |                             [ ]<------[ ]
+ *      |           |                     |                              |         |
+ *      :           :                     :                              :         :
+ *      ---------------------------- VM Migrated -----------------------------------
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      :           :                     :                              :         :
+ *      ---------------------------- VF Resumed ------------------------------------
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                    [ ]--------------               |         |
+ *      |           |                    [ ] Fix up GGTT  |              |         |
+ *      |           |                    [ ]<-------------               |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |  Notify VF_RESFIX_DONE       |         |
+ *      |           |                    [ ]--------------------------->[ ]        |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |Schedule |
+ *      |           |                     |                              |CCS      |
+ *      |           |                     |                              |Restore  |
+ *      |           |                     |                              |LRCA     |
+ *      |           |                     |                             [ ]------>[ ]
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |CCS      |
+ *      |           |                     |                              |restore  |
+ *      |           |                     |                              |completed|
+ *      |           |                     |                             [ ]<------[ ]
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      |           |                     |  VF_RESFIX_DONE complete     |         |
+ *      |           |                     |       notification           |         |
+ *      |           |                    [ ]<---------------------------[ ]        |
+ *      |           |                     |                              |         |
+ *      |           |                     |                              |         |
+ *      :           :                     :                              :         :
+ *      ------------------------- Continue VM restore ------------------------------
+ */
+
+static u64 get_ccs_bb_pool_size(struct xe_device *xe)
+{
+	u64 sys_mem_size, ccs_mem_size, ptes, bb_pool_size;
+	struct sysinfo si;
+
+	si_meminfo(&si);
+	sys_mem_size = si.totalram * si.mem_unit;
+	ccs_mem_size = div64_u64(sys_mem_size, NUM_BYTES_PER_CCS_BYTE(xe));
+	ptes = DIV_ROUND_UP_ULL(sys_mem_size + ccs_mem_size, XE_PAGE_SIZE);
+
+	/**
+	 * We need below BB size to hold PTE mappings and some DWs for copy
+	 * command. In reality, we need space for many copy commands. So, let
+	 * us allocate double the calculated size which is enough to holds GPU
+	 * instructions for the whole region.
+	 */
+	bb_pool_size = ptes * sizeof(u32);
+
+	return round_up(bb_pool_size * 2, SZ_1M);
+}
+
+static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx)
+{
+	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_sa_manager *sa_manager;
+	u64 bb_pool_size;
+	int offset, err;
+
+	bb_pool_size = get_ccs_bb_pool_size(xe);
+	xe_sriov_info(xe, "Allocating %s CCS BB pool size = %lldMB\n",
+		      ctx->ctx_id ? "Restore" : "Save", bb_pool_size / SZ_1M);
+
+	sa_manager = xe_sa_bo_manager_init(tile, bb_pool_size, SZ_16);
+
+	if (IS_ERR(sa_manager)) {
+		xe_sriov_err(xe, "Suballocator init failed with error: %pe\n",
+			     sa_manager);
+		err = PTR_ERR(sa_manager);
+		return err;
+	}
+
+	offset = 0;
+	xe_map_memset(xe, &sa_manager->bo->vmap, offset, MI_NOOP,
+		      bb_pool_size);
+
+	offset = bb_pool_size - sizeof(u32);
+	xe_map_wr(xe, &sa_manager->bo->vmap, offset, u32, MI_BATCH_BUFFER_END);
+
+	ctx->mem.ccs_bb_pool = sa_manager;
+
+	return 0;
+}
+
+/**
+ * xe_sriov_vf_ccs_init - Setup LRCA for save & restore.
+ * @xe: the &xe_device to start recovery on
+ *
+ * This function shall be called only by VF. It initializes
+ * LRCA and suballocator needed for CCS save & restore.
+ *
+ * Return: 0 on success. Negative error code on failure.
+ */
+int xe_sriov_vf_ccs_init(struct xe_device *xe)
+{
+	struct xe_tile *tile = xe_device_get_root_tile(xe);
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+	struct xe_migrate *migrate;
+	struct xe_tile_vf_ccs *ctx;
+	int err;
+
+	xe_assert(xe, IS_SRIOV_VF(xe));
+	xe_assert(xe, !IS_DGFX(xe));
+	xe_assert(xe, xe_device_has_flat_ccs(xe));
+
+	for_each_ccs_rw_ctx(ctx_id) {
+		ctx = &tile->sriov.vf.ccs[ctx_id];
+		ctx->ctx_id = ctx_id;
+
+		migrate = xe_migrate_init(tile);
+		if (IS_ERR(migrate)) {
+			err = PTR_ERR(migrate);
+			goto err_ret;
+		}
+		ctx->migrate = migrate;
+
+		err = alloc_bb_pool(tile, ctx);
+		if (err)
+			goto err_ret;
+	}
+
+	xe->sriov.vf.ccs.initialized = 1;
+
+	return 0;
+
+err_ret:
+	return err;
+}
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
new file mode 100644
index 000000000000..5df9ba028d14
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_SRIOV_VF_CCS_H_
+#define _XE_SRIOV_VF_CCS_H_
+
+struct xe_device;
+
+int xe_sriov_vf_ccs_init(struct xe_device *xe);
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
new file mode 100644
index 000000000000..6dc279d206ec
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_SRIOV_VF_CCS_TYPES_H_
+#define _XE_SRIOV_VF_CCS_TYPES_H_
+
+#define for_each_ccs_rw_ctx(id__) \
+	for ((id__) = 0; (id__) < XE_SRIOV_VF_CCS_CTX_COUNT; (id__)++)
+
+#define IS_VF_CCS_READY(xe) ({ \
+		struct xe_device *___xe = (xe); \
+		xe_assert(___xe, IS_SRIOV_VF(___xe)); \
+		___xe->sriov.vf.ccs.initialized; \
+		})
+
+#define IS_VF_CCS_INIT_NEEDED(xe) ({\
+		struct xe_device *___xe = (xe); \
+		IS_SRIOV_VF(___xe) && !IS_DGFX(___xe) && \
+		xe_device_has_flat_ccs(___xe) && GRAPHICS_VER(___xe) >= 20; \
+		})
+
+enum xe_sriov_vf_ccs_rw_ctxs {
+	XE_SRIOV_VF_CCS_READ_CTX,
+	XE_SRIOV_VF_CCS_WRITE_CTX,
+	XE_SRIOV_VF_CCS_CTX_COUNT
+};
+
+struct xe_migrate;
+struct xe_sa_manager;
+
+struct xe_tile_vf_ccs {
+	/** @id: Id to which context it belongs to */
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+	/** @migrate: Migration helper for save/restore of CCS data */
+	struct xe_migrate *migrate;
+
+	struct {
+		/** @ccs_rw_bb_pool: Pool from which batch buffers are allocated. */
+		struct xe_sa_manager *ccs_bb_pool;
+	} mem;
+};
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
index 8300416a6226..24a873c50c49 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
@@ -36,6 +36,12 @@ struct xe_device_vf {
 		/** @migration.gt_flags: Per-GT request flags for VF migration recovery */
 		unsigned long gt_flags;
 	} migration;
+
+	/** @ccs: VF CCS state data */
+	struct {
+		/** @ccs.initialized: Initilalization of VF CCS is completed or not */
+		bool initialized;
+	} ccs;
 };
 
 #endif

From 864690cf4dd62482b6dd049d82c509886c904303 Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Tue, 22 Jul 2025 17:35:05 +0530
Subject: [PATCH 0071/1455] drm/xe/vf: Attach and detach CCS copy commands with
 BO
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Attach CCS read/write copy commands to BO for old and new mem types as
NULL -> tt or system -> tt.
Detach the CCS read/write copy commands from BO while deleting ttm bo
from xe_ttm_bo_delete_mem_notify().

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250722120506.6483-3-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/xe_bb.c                 |  35 +++++
 drivers/gpu/drm/xe/xe_bb.h                 |   3 +
 drivers/gpu/drm/xe/xe_bo.c                 |  23 ++++
 drivers/gpu/drm/xe/xe_bo_types.h           |   3 +
 drivers/gpu/drm/xe/xe_migrate.c            | 148 +++++++++++++++++++++
 drivers/gpu/drm/xe/xe_migrate.h            |   7 +
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       |  67 ++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.h       |   3 +
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h |   8 ++
 9 files changed, 297 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_bb.c b/drivers/gpu/drm/xe/xe_bb.c
index 5ce0e26822f2..feb6e013dc38 100644
--- a/drivers/gpu/drm/xe/xe_bb.c
+++ b/drivers/gpu/drm/xe/xe_bb.c
@@ -60,6 +60,41 @@ err:
 	return ERR_PTR(err);
 }
 
+struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
+			    enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
+{
+	struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
+	struct xe_tile *tile = gt_to_tile(gt);
+	struct xe_sa_manager *bb_pool;
+	int err;
+
+	if (!bb)
+		return ERR_PTR(-ENOMEM);
+	/*
+	 * We need to allocate space for the requested number of dwords &
+	 * one additional MI_BATCH_BUFFER_END dword. Since the whole SA
+	 * is submitted to HW, we need to make sure that the last instruction
+	 * is not over written when the last chunk of SA is allocated for BB.
+	 * So, this extra DW acts as a guard here.
+	 */
+
+	bb_pool = tile->sriov.vf.ccs[ctx_id].mem.ccs_bb_pool;
+	bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
+
+	if (IS_ERR(bb->bo)) {
+		err = PTR_ERR(bb->bo);
+		goto err;
+	}
+
+	bb->cs = xe_sa_bo_cpu_addr(bb->bo);
+	bb->len = 0;
+
+	return bb;
+err:
+	kfree(bb);
+	return ERR_PTR(err);
+}
+
 static struct xe_sched_job *
 __xe_bb_create_job(struct xe_exec_queue *q, struct xe_bb *bb, u64 *addr)
 {
diff --git a/drivers/gpu/drm/xe/xe_bb.h b/drivers/gpu/drm/xe/xe_bb.h
index b5cc65506696..2a8adc9a6dee 100644
--- a/drivers/gpu/drm/xe/xe_bb.h
+++ b/drivers/gpu/drm/xe/xe_bb.h
@@ -13,8 +13,11 @@ struct dma_fence;
 struct xe_gt;
 struct xe_exec_queue;
 struct xe_sched_job;
+enum xe_sriov_vf_ccs_rw_ctxs;
 
 struct xe_bb *xe_bb_new(struct xe_gt *gt, u32 dwords, bool usm);
+struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
+			    enum xe_sriov_vf_ccs_rw_ctxs ctx_id);
 struct xe_sched_job *xe_bb_create_job(struct xe_exec_queue *q,
 				      struct xe_bb *bb);
 struct xe_sched_job *xe_bb_create_migration_job(struct xe_exec_queue *q,
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 00ce067d5fd3..ffca1cea5585 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -33,6 +33,7 @@
 #include "xe_pxp.h"
 #include "xe_res_cursor.h"
 #include "xe_shrinker.h"
+#include "xe_sriov_vf_ccs.h"
 #include "xe_trace_bo.h"
 #include "xe_ttm_stolen_mgr.h"
 #include "xe_vm.h"
@@ -965,6 +966,20 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,
 	dma_fence_put(fence);
 	xe_pm_runtime_put(xe);
 
+	/*
+	 * CCS meta data is migrated from TT -> SMEM. So, let us detach the
+	 * BBs from BO as it is no longer needed.
+	 */
+	if (IS_VF_CCS_BB_VALID(xe, bo) && old_mem_type == XE_PL_TT &&
+	    new_mem->mem_type == XE_PL_SYSTEM)
+		xe_sriov_vf_ccs_detach_bo(bo);
+
+	if (IS_SRIOV_VF(xe) &&
+	    ((move_lacks_source && new_mem->mem_type == XE_PL_TT) ||
+	     (old_mem_type == XE_PL_SYSTEM && new_mem->mem_type == XE_PL_TT)) &&
+	    handle_system_ccs)
+		ret = xe_sriov_vf_ccs_attach_bo(bo);
+
 out:
 	if ((!ttm_bo->resource || ttm_bo->resource->mem_type == XE_PL_SYSTEM) &&
 	    ttm_bo->ttm) {
@@ -975,6 +990,9 @@ out:
 		if (timeout < 0)
 			ret = timeout;
 
+		if (IS_VF_CCS_BB_VALID(xe, bo))
+			xe_sriov_vf_ccs_detach_bo(bo);
+
 		xe_tt_unmap_sg(xe, ttm_bo->ttm);
 	}
 
@@ -1502,9 +1520,14 @@ static void xe_ttm_bo_release_notify(struct ttm_buffer_object *ttm_bo)
 
 static void xe_ttm_bo_delete_mem_notify(struct ttm_buffer_object *ttm_bo)
 {
+	struct xe_bo *bo = ttm_to_xe_bo(ttm_bo);
+
 	if (!xe_bo_is_xe_bo(ttm_bo))
 		return;
 
+	if (IS_VF_CCS_BB_VALID(ttm_to_xe_device(ttm_bo->bdev), bo))
+		xe_sriov_vf_ccs_detach_bo(bo);
+
 	/*
 	 * Object is idle and about to be destroyed. Release the
 	 * dma-buf attachment.
diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index 57d34698139e..cf604adc13a3 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -77,6 +77,9 @@ struct xe_bo {
 	/** @ccs_cleared */
 	bool ccs_cleared;
 
+	/** @bb_ccs_rw: BB instructions of CCS read/write. Valid only for VF */
+	struct xe_bb *bb_ccs[XE_SRIOV_VF_CCS_CTX_COUNT];
+
 	/**
 	 * @cpu_caching: CPU caching mode. Currently only used for userspace
 	 * objects. Exceptions are system memory on DGFX, which is always
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 6a80ae6104dd..5a108e34c554 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -30,6 +30,7 @@
 #include "xe_mocs.h"
 #include "xe_pt.h"
 #include "xe_res_cursor.h"
+#include "xe_sa.h"
 #include "xe_sched_job.h"
 #include "xe_sync.h"
 #include "xe_trace_bo.h"
@@ -954,6 +955,153 @@ err_sync:
 	return fence;
 }
 
+/**
+ * xe_get_migrate_lrc() - Get the LRC from migrate context.
+ * @migrate: Migrate context.
+ *
+ * Return: Pointer to LRC on success, error on failure
+ */
+struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate)
+{
+	return migrate->q->lrc[0];
+}
+
+static int emit_flush_invalidate(struct xe_migrate *m, u32 *dw, int i,
+				 u32 flags)
+{
+	dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW |
+		  MI_FLUSH_IMM_DW | flags;
+	dw[i++] = lower_32_bits(xe_lrc_start_seqno_ggtt_addr(xe_migrate_lrc(m))) |
+		  MI_FLUSH_DW_USE_GTT;
+	dw[i++] = upper_32_bits(xe_lrc_start_seqno_ggtt_addr(xe_migrate_lrc(m)));
+	dw[i++] = MI_NOOP;
+	dw[i++] = MI_NOOP;
+
+	return i;
+}
+
+/**
+ * xe_migrate_ccs_rw_copy() - Copy content of TTM resources.
+ * @m: The migration context.
+ * @src_bo: The buffer object @src is currently bound to.
+ * @read_write : Creates BB commands for CCS read/write.
+ *
+ * Creates batch buffer instructions to copy CCS metadata from CCS pool to
+ * memory and vice versa.
+ *
+ * This function should only be called for IGPU.
+ *
+ * Return: 0 if successful, negative error code on failure.
+ */
+int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
+			   struct xe_bo *src_bo,
+			   enum xe_sriov_vf_ccs_rw_ctxs read_write)
+
+{
+	bool src_is_pltt = read_write == XE_SRIOV_VF_CCS_READ_CTX;
+	bool dst_is_pltt = read_write == XE_SRIOV_VF_CCS_WRITE_CTX;
+	struct ttm_resource *src = src_bo->ttm.resource;
+	struct xe_gt *gt = m->tile->primary_gt;
+	u32 batch_size, batch_size_allocated;
+	struct xe_device *xe = gt_to_xe(gt);
+	struct xe_res_cursor src_it, ccs_it;
+	u64 size = xe_bo_size(src_bo);
+	struct xe_bb *bb = NULL;
+	u64 src_L0, src_L0_ofs;
+	u32 src_L0_pt;
+	int err;
+
+	xe_res_first_sg(xe_bo_sg(src_bo), 0, size, &src_it);
+
+	xe_res_first_sg(xe_bo_sg(src_bo), xe_bo_ccs_pages_start(src_bo),
+			PAGE_ALIGN(xe_device_ccs_bytes(xe, size)),
+			&ccs_it);
+
+	/* Calculate Batch buffer size */
+	batch_size = 0;
+	while (size) {
+		batch_size += 10; /* Flush + ggtt addr + 2 NOP */
+		u64 ccs_ofs, ccs_size;
+		u32 ccs_pt;
+
+		u32 avail_pts = max_mem_transfer_per_pass(xe) / LEVEL0_PAGE_TABLE_ENCODE_SIZE;
+
+		src_L0 = min_t(u64, max_mem_transfer_per_pass(xe), size);
+
+		batch_size += pte_update_size(m, false, src, &src_it, &src_L0,
+					      &src_L0_ofs, &src_L0_pt, 0, 0,
+					      avail_pts);
+
+		ccs_size = xe_device_ccs_bytes(xe, src_L0);
+		batch_size += pte_update_size(m, 0, NULL, &ccs_it, &ccs_size, &ccs_ofs,
+					      &ccs_pt, 0, avail_pts, avail_pts);
+		xe_assert(xe, IS_ALIGNED(ccs_it.start, PAGE_SIZE));
+
+		/* Add copy commands size here */
+		batch_size += EMIT_COPY_CCS_DW;
+
+		size -= src_L0;
+	}
+
+	bb = xe_bb_ccs_new(gt, batch_size, read_write);
+	if (IS_ERR(bb)) {
+		drm_err(&xe->drm, "BB allocation failed.\n");
+		err = PTR_ERR(bb);
+		goto err_ret;
+	}
+
+	batch_size_allocated = batch_size;
+	size = xe_bo_size(src_bo);
+	batch_size = 0;
+
+	/*
+	 * Emit PTE and copy commands here.
+	 * The CCS copy command can only support limited size. If the size to be
+	 * copied is more than the limit, divide copy into chunks. So, calculate
+	 * sizes here again before copy command is emitted.
+	 */
+	while (size) {
+		batch_size += 10; /* Flush + ggtt addr + 2 NOP */
+		u32 flush_flags = 0;
+		u64 ccs_ofs, ccs_size;
+		u32 ccs_pt;
+
+		u32 avail_pts = max_mem_transfer_per_pass(xe) / LEVEL0_PAGE_TABLE_ENCODE_SIZE;
+
+		src_L0 = xe_migrate_res_sizes(m, &src_it);
+
+		batch_size += pte_update_size(m, false, src, &src_it, &src_L0,
+					      &src_L0_ofs, &src_L0_pt, 0, 0,
+					      avail_pts);
+
+		ccs_size = xe_device_ccs_bytes(xe, src_L0);
+		batch_size += pte_update_size(m, 0, NULL, &ccs_it, &ccs_size, &ccs_ofs,
+					      &ccs_pt, 0, avail_pts, avail_pts);
+		xe_assert(xe, IS_ALIGNED(ccs_it.start, PAGE_SIZE));
+		batch_size += EMIT_COPY_CCS_DW;
+
+		emit_pte(m, bb, src_L0_pt, false, true, &src_it, src_L0, src);
+
+		emit_pte(m, bb, ccs_pt, false, false, &ccs_it, ccs_size, src);
+
+		bb->len = emit_flush_invalidate(m, bb->cs, bb->len, flush_flags);
+		flush_flags = xe_migrate_ccs_copy(m, bb, src_L0_ofs, src_is_pltt,
+						  src_L0_ofs, dst_is_pltt,
+						  src_L0, ccs_ofs, true);
+		bb->len = emit_flush_invalidate(m, bb->cs, bb->len, flush_flags);
+
+		size -= src_L0;
+	}
+
+	xe_assert(xe, (batch_size_allocated == bb->len));
+	src_bo->bb_ccs[read_write] = bb;
+
+	return 0;
+
+err_ret:
+	return err;
+}
+
 static void emit_clear_link_copy(struct xe_gt *gt, struct xe_bb *bb, u64 src_ofs,
 				 u32 size, u32 pitch)
 {
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 74c60f55004a..cd948c7b8dea 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -24,6 +24,8 @@ struct xe_vm;
 struct xe_vm_pgtable_update;
 struct xe_vma;
 
+enum xe_sriov_vf_ccs_rw_ctxs;
+
 /**
  * struct xe_migrate_pt_update_ops - Callbacks for the
  * xe_migrate_update_pgtables() function.
@@ -113,6 +115,11 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 				  struct ttm_resource *dst,
 				  bool copy_only_ccs);
 
+int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
+			   struct xe_bo *src_bo,
+			   enum xe_sriov_vf_ccs_rw_ctxs read_write);
+
+struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate);
 int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 			     unsigned long offset, void *buf, int len,
 			     int write);
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 697eea46659a..6ddeb016b892 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -5,6 +5,7 @@
 
 #include "instructions/xe_mi_commands.h"
 #include "instructions/xe_gpu_commands.h"
+#include "xe_bb.h"
 #include "xe_bo.h"
 #include "xe_device.h"
 #include "xe_migrate.h"
@@ -206,3 +207,69 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 err_ret:
 	return err;
 }
+
+/**
+ * xe_sriov_vf_ccs_attach_bo - Insert CCS read write commands in the BO.
+ * @bo: the &buffer object to which batch buffer commands will be added.
+ *
+ * This function shall be called only by VF. It inserts the PTEs and copy
+ * command instructions in the BO by calling xe_migrate_ccs_rw_copy()
+ * function.
+ *
+ * Returns: 0 if successful, negative error code on failure.
+ */
+int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo)
+{
+	struct xe_device *xe = xe_bo_device(bo);
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+	struct xe_migrate *migrate;
+	struct xe_tile *tile;
+	struct xe_bb *bb;
+	int err = 0;
+
+	if (!IS_VF_CCS_READY(xe))
+		return 0;
+
+	tile = xe_device_get_root_tile(xe);
+
+	for_each_ccs_rw_ctx(ctx_id) {
+		bb = bo->bb_ccs[ctx_id];
+		/* bb should be NULL here. Assert if not NULL */
+		xe_assert(xe, !bb);
+
+		migrate = tile->sriov.vf.ccs[ctx_id].migrate;
+		err = xe_migrate_ccs_rw_copy(migrate, bo, ctx_id);
+	}
+	return err;
+}
+
+/**
+ * xe_sriov_vf_ccs_detach_bo - Remove CCS read write commands from the BO.
+ * @bo: the &buffer object from which batch buffer commands will be removed.
+ *
+ * This function shall be called only by VF. It removes the PTEs and copy
+ * command instructions from the BO. Make sure to update the BB with MI_NOOP
+ * before freeing.
+ *
+ * Returns: 0 if successful.
+ */
+int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo)
+{
+	struct xe_device *xe = xe_bo_device(bo);
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+	struct xe_bb *bb;
+
+	if (!IS_VF_CCS_READY(xe))
+		return 0;
+
+	for_each_ccs_rw_ctx(ctx_id) {
+		bb = bo->bb_ccs[ctx_id];
+		if (!bb)
+			continue;
+
+		memset(bb->cs, MI_NOOP, bb->len * sizeof(u32));
+		xe_bb_free(bb, NULL);
+		bo->bb_ccs[ctx_id] = NULL;
+	}
+	return 0;
+}
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
index 5df9ba028d14..5d5e4bd25904 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
@@ -7,7 +7,10 @@
 #define _XE_SRIOV_VF_CCS_H_
 
 struct xe_device;
+struct xe_bo;
 
 int xe_sriov_vf_ccs_init(struct xe_device *xe);
+int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
+int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
index 6dc279d206ec..e240f3fd18af 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -27,6 +27,14 @@ enum xe_sriov_vf_ccs_rw_ctxs {
 	XE_SRIOV_VF_CCS_CTX_COUNT
 };
 
+#define IS_VF_CCS_BB_VALID(xe, bo) ({ \
+		struct xe_device *___xe = (xe); \
+		struct xe_bo *___bo = (bo); \
+		IS_SRIOV_VF(___xe) && \
+		___bo->bb_ccs[XE_SRIOV_VF_CCS_READ_CTX] && \
+		___bo->bb_ccs[XE_SRIOV_VF_CCS_WRITE_CTX]; \
+		})
+
 struct xe_migrate;
 struct xe_sa_manager;
 

From 916ee4704a8653910f10c65c2b5d6699dfac5df8 Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Tue, 22 Jul 2025 17:35:06 +0530
Subject: [PATCH 0072/1455] drm/xe/vf: Register CCS read/write contexts with
 Guc

Register read write contexts with newly added flags with GUC and
enable the context immediately after registration.
Re-register the context with Guc when resuming from runtime suspend as
soft reset is applied to Guc during xe_pm_runtime_resume().
Make Ring head=tail while unbinding device to avoid issues with VF pause
after device is unbinded.

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250722120506.6483-4-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/xe_guc_fwif.h     |  5 ++
 drivers/gpu/drm/xe/xe_guc_submit.c   | 34 +++++++++-
 drivers/gpu/drm/xe/xe_guc_submit.h   |  1 +
 drivers/gpu/drm/xe/xe_migrate.c      | 24 ++++----
 drivers/gpu/drm/xe/xe_migrate.h      |  3 +-
 drivers/gpu/drm/xe/xe_pm.c           |  4 ++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 92 ++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.h |  1 +
 drivers/gpu/drm/xe/xe_vm.c           |  6 +-
 9 files changed, 150 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index 6f57578b07cb..ca9f999d38d1 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -45,6 +45,11 @@
 #define GUC_MAX_ENGINE_CLASSES		16
 #define GUC_MAX_INSTANCES_PER_CLASS	32
 
+#define GUC_CONTEXT_NORMAL			0
+#define GUC_CONTEXT_COMPRESSION_SAVE		1
+#define GUC_CONTEXT_COMPRESSION_RESTORE	2
+#define GUC_CONTEXT_COUNT			(GUC_CONTEXT_COMPRESSION_RESTORE + 1)
+
 /* Helper for context registration H2G */
 struct guc_ctxt_registration_info {
 	u32 flags;
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 5f91b9a6ab7d..679ae229cc82 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -542,7 +542,7 @@ static void __register_exec_queue(struct xe_guc *guc,
 	xe_guc_ct_send(&guc->ct, action, ARRAY_SIZE(action), 0, 0);
 }
 
-static void register_exec_queue(struct xe_exec_queue *q)
+static void register_exec_queue(struct xe_exec_queue *q, int ctx_type)
 {
 	struct xe_guc *guc = exec_queue_to_guc(q);
 	struct xe_device *xe = guc_to_xe(guc);
@@ -550,6 +550,7 @@ static void register_exec_queue(struct xe_exec_queue *q)
 	struct guc_ctxt_registration_info info;
 
 	xe_gt_assert(guc_to_gt(guc), !exec_queue_registered(q));
+	xe_gt_assert(guc_to_gt(guc), ctx_type < GUC_CONTEXT_COUNT);
 
 	memset(&info, 0, sizeof(info));
 	info.context_idx = q->guc->id;
@@ -559,6 +560,9 @@ static void register_exec_queue(struct xe_exec_queue *q)
 	info.hwlrca_hi = upper_32_bits(xe_lrc_descriptor(lrc));
 	info.flags = CONTEXT_REGISTRATION_FLAG_KMD;
 
+	if (ctx_type != GUC_CONTEXT_NORMAL)
+		info.flags |= BIT(ctx_type);
+
 	if (xe_exec_queue_is_parallel(q)) {
 		u64 ggtt_addr = xe_lrc_parallel_ggtt_addr(lrc);
 		struct iosys_map map = xe_lrc_parallel_map(lrc);
@@ -761,7 +765,7 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
 
 	if (!exec_queue_killed_or_banned_or_wedged(q) && !xe_sched_job_is_error(job)) {
 		if (!exec_queue_registered(q))
-			register_exec_queue(q);
+			register_exec_queue(q, GUC_CONTEXT_NORMAL);
 		if (!lr)	/* LR jobs are emitted in the exec IOCTL */
 			q->ring_ops->emit_job(job);
 		submit_exec_queue(q);
@@ -2383,6 +2387,32 @@ static void guc_exec_queue_print(struct xe_exec_queue *q, struct drm_printer *p)
 	xe_guc_exec_queue_snapshot_free(snapshot);
 }
 
+/**
+ * xe_guc_register_exec_queue - Register exec queue for a given context type.
+ * @q: Execution queue
+ * @ctx_type: Type of the context
+ *
+ * This function registers the execution queue with the guc. Special context
+ * types like GUC_CONTEXT_COMPRESSION_SAVE and GUC_CONTEXT_COMPRESSION_RESTORE
+ * are only applicable for IGPU and in the VF.
+ * Submits the execution queue to GUC after registering it.
+ *
+ * Returns - None.
+ */
+void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type)
+{
+	struct xe_guc *guc = exec_queue_to_guc(q);
+	struct xe_device *xe = guc_to_xe(guc);
+
+	xe_assert(xe, IS_SRIOV_VF(xe));
+	xe_assert(xe, !IS_DGFX(xe));
+	xe_assert(xe, (ctx_type > GUC_CONTEXT_NORMAL &&
+		       ctx_type < GUC_CONTEXT_COUNT));
+
+	register_exec_queue(q, ctx_type);
+	enable_scheduling(q);
+}
+
 /**
  * xe_guc_submit_print - GuC Submit Print.
  * @guc: GuC.
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 9b71a986c6ca..8f64e799283b 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -39,5 +39,6 @@ xe_guc_exec_queue_snapshot_print(struct xe_guc_submit_exec_queue_snapshot *snaps
 void
 xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapshot);
 void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p);
+void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 5a108e34c554..0a61fb8b123a 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -86,19 +86,6 @@ struct xe_migrate {
  */
 #define MAX_PTE_PER_SDI 0x1FEU
 
-/**
- * xe_tile_migrate_exec_queue() - Get this tile's migrate exec queue.
- * @tile: The tile.
- *
- * Returns the default migrate exec queue of this tile.
- *
- * Return: The default migrate exec queue
- */
-struct xe_exec_queue *xe_tile_migrate_exec_queue(struct xe_tile *tile)
-{
-	return tile->migrate->q;
-}
-
 static void xe_migrate_fini(void *arg)
 {
 	struct xe_migrate *m = arg;
@@ -1102,6 +1089,17 @@ err_ret:
 	return err;
 }
 
+/**
+ * xe_get_migrate_exec_queue() - Get the execution queue from migrate context.
+ * @migrate: Migrate context.
+ *
+ * Return: Pointer to execution queue on success, error on failure
+ */
+struct xe_exec_queue *xe_migrate_exec_queue(struct xe_migrate *migrate)
+{
+	return migrate->q;
+}
+
 static void emit_clear_link_copy(struct xe_gt *gt, struct xe_bb *bb, u64 src_ofs,
 				 u32 size, u32 pitch)
 {
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index cd948c7b8dea..f4dac270dba3 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -120,6 +120,7 @@ int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
 			   enum xe_sriov_vf_ccs_rw_ctxs read_write);
 
 struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate);
+struct xe_exec_queue *xe_migrate_exec_queue(struct xe_migrate *migrate);
 int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 			     unsigned long offset, void *buf, int len,
 			     int write);
@@ -140,6 +141,4 @@ xe_migrate_update_pgtables(struct xe_migrate *m,
 			   struct xe_migrate_pt_update *pt_update);
 
 void xe_migrate_wait(struct xe_migrate *m);
-
-struct xe_exec_queue *xe_tile_migrate_exec_queue(struct xe_tile *tile);
 #endif
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index e279b47ba03b..ed29791beb7c 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -23,6 +23,7 @@
 #include "xe_irq.h"
 #include "xe_pcode.h"
 #include "xe_pxp.h"
+#include "xe_sriov_vf_ccs.h"
 #include "xe_trace.h"
 #include "xe_wa.h"
 
@@ -558,6 +559,9 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 
 	xe_pxp_pm_resume(xe->pxp);
 
+	if (IS_SRIOV_VF(xe))
+		xe_sriov_vf_ccs_register_context(xe);
+
 out:
 	xe_rpm_lockmap_release(xe);
 	xe_pm_write_callback_task(xe, NULL);
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 6ddeb016b892..af43e04179aa 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -8,6 +8,9 @@
 #include "xe_bb.h"
 #include "xe_bo.h"
 #include "xe_device.h"
+#include "xe_exec_queue_types.h"
+#include "xe_guc_submit.h"
+#include "xe_lrc.h"
 #include "xe_migrate.h"
 #include "xe_sa.h"
 #include "xe_sriov_printk.h"
@@ -163,6 +166,85 @@ static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx)
 	return 0;
 }
 
+static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
+{
+	struct xe_lrc *lrc = xe_migrate_lrc(ctx->migrate);
+	u64 addr = ctx->mem.ccs_bb_pool->gpu_addr;
+	u32 dw[10], i = 0;
+
+	dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;
+	dw[i++] = MI_BATCH_BUFFER_START | XE_INSTR_NUM_DW(3);
+	dw[i++] = lower_32_bits(addr);
+	dw[i++] = upper_32_bits(addr);
+	dw[i++] = MI_NOOP;
+	dw[i++] = MI_NOOP;
+
+	xe_lrc_write_ring(lrc, dw, i * sizeof(u32));
+	xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
+}
+
+static int register_save_restore_context(struct xe_migrate *m,
+					 enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
+{
+	int err = -EINVAL;
+	int ctx_type;
+
+	switch (ctx_id) {
+	case XE_SRIOV_VF_CCS_READ_CTX:
+		ctx_type = GUC_CONTEXT_COMPRESSION_SAVE;
+		break;
+	case XE_SRIOV_VF_CCS_WRITE_CTX:
+		ctx_type = GUC_CONTEXT_COMPRESSION_RESTORE;
+		break;
+	default:
+		return err;
+	}
+
+	xe_guc_register_exec_queue(xe_migrate_exec_queue(m), ctx_type);
+	return 0;
+}
+
+/**
+ * xe_sriov_vf_ccs_register_context - Register read/write contexts with guc.
+ * @xe: the &xe_device to register contexts on.
+ *
+ * This function registers read and write contexts with Guc. Re-registration
+ * is needed whenever resuming from pm runtime suspend.
+ *
+ * Return: 0 on success. Negative error code on failure.
+ */
+int xe_sriov_vf_ccs_register_context(struct xe_device *xe)
+{
+	struct xe_tile *tile = xe_device_get_root_tile(xe);
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+	struct xe_tile_vf_ccs *ctx;
+	int err;
+
+	if (!IS_VF_CCS_READY(xe))
+		return 0;
+
+	for_each_ccs_rw_ctx(ctx_id) {
+		ctx = &tile->sriov.vf.ccs[ctx_id];
+		err = register_save_restore_context(ctx->migrate, ctx_id);
+		if (err)
+			return err;
+	}
+
+	return err;
+}
+
+static void xe_sriov_vf_ccs_fini(void *arg)
+{
+	struct xe_tile_vf_ccs *ctx = arg;
+	struct xe_lrc *lrc = xe_migrate_lrc(ctx->migrate);
+
+	/*
+	 * Make TAIL = HEAD in the ring so that no issues are seen if Guc
+	 * submits this context to HW on VF pause after unbinding device.
+	 */
+	xe_lrc_set_ring_tail(lrc, xe_lrc_ring_head(lrc));
+}
+
 /**
  * xe_sriov_vf_ccs_init - Setup LRCA for save & restore.
  * @xe: the &xe_device to start recovery on
@@ -198,6 +280,16 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 		err = alloc_bb_pool(tile, ctx);
 		if (err)
 			goto err_ret;
+
+		ccs_rw_update_ring(ctx);
+
+		err = register_save_restore_context(ctx->migrate, ctx_id);
+		if (err)
+			goto err_ret;
+
+		err = devm_add_action_or_reset(xe->drm.dev,
+					       xe_sriov_vf_ccs_fini,
+					       ctx);
 	}
 
 	xe->sriov.vf.ccs.initialized = 1;
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
index 5d5e4bd25904..1f1baf685fec 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
@@ -12,5 +12,6 @@ struct xe_bo;
 int xe_sriov_vf_ccs_init(struct xe_device *xe);
 int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
 int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
+int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 2035604121e6..432ea325677d 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -953,7 +953,7 @@ struct dma_fence *xe_vma_rebind(struct xe_vm *vm, struct xe_vma *vma, u8 tile_ma
 	for_each_tile(tile, vm->xe, id) {
 		vops.pt_update_ops[id].wait_vm_bookkeep = true;
 		vops.pt_update_ops[tile->id].q =
-			xe_tile_migrate_exec_queue(tile);
+			xe_migrate_exec_queue(tile->migrate);
 	}
 
 	err = xe_vm_ops_add_rebind(&vops, vma, tile_mask);
@@ -1043,7 +1043,7 @@ struct dma_fence *xe_vm_range_rebind(struct xe_vm *vm,
 	for_each_tile(tile, vm->xe, id) {
 		vops.pt_update_ops[id].wait_vm_bookkeep = true;
 		vops.pt_update_ops[tile->id].q =
-			xe_tile_migrate_exec_queue(tile);
+			xe_migrate_exec_queue(tile->migrate);
 	}
 
 	err = xe_vm_ops_add_range_rebind(&vops, vma, range, tile_mask);
@@ -1126,7 +1126,7 @@ struct dma_fence *xe_vm_range_unbind(struct xe_vm *vm,
 	for_each_tile(tile, vm->xe, id) {
 		vops.pt_update_ops[id].wait_vm_bookkeep = true;
 		vops.pt_update_ops[tile->id].q =
-			xe_tile_migrate_exec_queue(tile);
+			xe_migrate_exec_queue(tile->migrate);
 	}
 
 	err = xe_vm_ops_add_range_unbind(&vops, range);

From 4d3bbe9dd28c0a4ca119e4b8823c5f5e9cb3ff90 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 22 Jul 2025 12:52:08 -0700
Subject: [PATCH 0073/1455] drm/xe: Fix build without debugfs

When CONFIG_DEBUG_FS is off, drivers/gpu/drm/xe/xe_gt_debugfs.o
is not built and build fails on some setups with:

	ld: drivers/gpu/drm/xe/xe_gt.o: in function `xe_fault_inject_gt_reset':
	drivers/gpu/drm/xe/xe_gt.h:27:(.text+0x1659): undefined reference to `gt_reset_failure'
	ld: drivers/gpu/drm/xe/xe_gt.h:27:(.text+0x1c16): undefined reference to `gt_reset_failure'
	collect2: error: ld returned 1 exit status

Do not use the gt_reset_failure attribute if debugfs is not enabled.

Fixes: 8f3013e0b222 ("drm/xe: Introduce fault injection for gt reset")
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Tested-by: Randy Dunlap <rdunlap@infradead.org>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250722-xe-fix-build-fault-v1-1-157384d50987@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_gt.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_gt.h b/drivers/gpu/drm/xe/xe_gt.h
index e9ccab8aedbe..41880979f4de 100644
--- a/drivers/gpu/drm/xe/xe_gt.h
+++ b/drivers/gpu/drm/xe/xe_gt.h
@@ -24,7 +24,7 @@
 extern struct fault_attr gt_reset_failure;
 static inline bool xe_fault_inject_gt_reset(void)
 {
-	return should_fail(&gt_reset_failure, 1);
+	return IS_ENABLED(CONFIG_DEBUG_FS) && should_fail(&gt_reset_failure, 1);
 }
 
 struct xe_gt *xe_gt_alloc(struct xe_tile *tile);

From 1511d3c4d2bb30f784924a877f3cef518bb73077 Mon Sep 17 00:00:00 2001
From: Langyan Ye <yelangyan@huaqin.corp-partner.google.com>
Date: Wed, 23 Jul 2025 15:25:13 +0800
Subject: [PATCH 0074/1455] drm/panel-edp: Add 50ms disable delay for four
 panels

Add 50ms disable delay for NV116WHM-N49, NV122WUM-N41, and MNC207QS1-1
to satisfy T9+T10 timing. Add 50ms disable delay for MNE007JA1-2
as well, since MNE007JA1-2 copies the timing of MNC207QS1-1.

Specifically, it should be noted that the MNE007JA1-2 panel was added
by someone who did not have the panel documentation, so they simply
copied the timing from the MNC207QS1-1 panel. Adding an extra 50 ms
of delay should be safe.

Fixes: 0547692ac146 ("drm/panel-edp: Add several generic edp panels")
Fixes: 50625eab3972 ("drm/edp-panel: Add panel used by T14s Gen6 Snapdragon")
Signed-off-by: Langyan Ye <yelangyan@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250723072513.2880369-1-yelangyan@huaqin.corp-partner.google.com
---
 drivers/gpu/drm/panel/panel-edp.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index 09170470b3ef..d0aa602ecc9d 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -1736,10 +1736,11 @@ static const struct panel_delay delay_200_500_e50 = {
 	.enable = 50,
 };
 
-static const struct panel_delay delay_200_500_e50_p2e200 = {
+static const struct panel_delay delay_200_500_e50_d50_p2e200 = {
 	.hpd_absent = 200,
 	.unprepare = 500,
 	.enable = 50,
+	.disable = 50,
 	.prepare_to_enable = 200,
 };
 
@@ -1941,13 +1942,13 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x09dd, &delay_200_500_e50, "NT116WHM-N21"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a1b, &delay_200_500_e50, "NV133WUM-N63"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a36, &delay_200_500_e200, "Unknown"),
-	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a3e, &delay_200_500_e80, "NV116WHM-N49"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a3e, &delay_200_500_e80_d50, "NV116WHM-N49"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a5d, &delay_200_500_e50, "NV116WHM-N45"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ac5, &delay_200_500_e50, "NV116WHM-N4C"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ae8, &delay_200_500_e50_p2e80, "NV140WUM-N41"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b09, &delay_200_500_e50_po2e200, "NV140FHM-NZ"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b1e, &delay_200_500_e80, "NE140QDM-N6A"),
-	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b34, &delay_200_500_e80, "NV122WUM-N41"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b34, &delay_200_500_e80_d50, "NV122WUM-N41"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b43, &delay_200_500_e200, "NV140FHM-T09"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b56, &delay_200_500_e80, "NT140FHM-N47"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b66, &delay_200_500_e80, "NE140WUM-N6G"),
@@ -1986,8 +1987,8 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x14e5, &delay_200_500_e80_d50, "N140HGA-EA1"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x162b, &delay_200_500_e80_d50, "N160JCE-ELL"),
 
-	EDP_PANEL_ENTRY('C', 'S', 'O', 0x1200, &delay_200_500_e50_p2e200, "MNC207QS1-1"),
-	EDP_PANEL_ENTRY('C', 'S', 'O', 0x1413, &delay_200_500_e50_p2e200, "MNE007JA1-2"),
+	EDP_PANEL_ENTRY('C', 'S', 'O', 0x1200, &delay_200_500_e50_d50_p2e200, "MNC207QS1-1"),
+	EDP_PANEL_ENTRY('C', 'S', 'O', 0x1413, &delay_200_500_e50_d50_p2e200, "MNE007JA1-2"),
 
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1100, &delay_200_500_e80_d50, "MNB601LS1-1"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1103, &delay_200_500_e80_d50, "MNB601LS1-3"),

From a2b461bd6f3b36bded0a74178dec0e58e4714d3d Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 22 Jul 2025 20:26:16 +0200
Subject: [PATCH 0075/1455] drm/xe/pf: Enable SR-IOV PF mode by default

We already claim official support for SR-IOV PF/VF modes on PTL
and BMG platforms, but by default we start the Xe driver on those
platforms in non-virtualized mode (native) since we still have
max_vfs modparam set to disable creation of the VFs.

It's time to let the Xe driver support SR-IOV PF mode by default.
We were already testing this on our CI, which was relying on the
patch that was enabling it for CONFIG_DRM_XE_DEBUG used by our CI.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250722182618.30811-3-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_module.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c
index 107ffe87808c..d9391bd08194 100644
--- a/drivers/gpu/drm/xe/xe_module.c
+++ b/drivers/gpu/drm/xe/xe_module.c
@@ -27,6 +27,8 @@
 #define DEFAULT_PROBE_DISPLAY		true
 #define DEFAULT_VRAM_BAR_SIZE		0
 #define DEFAULT_FORCE_PROBE		CONFIG_DRM_XE_FORCE_PROBE
+#define DEFAULT_MAX_VFS			~0
+#define DEFAULT_MAX_VFS_STR		"unlimited"
 #define DEFAULT_WEDGED_MODE		1
 #define DEFAULT_SVM_NOTIFIER_SIZE	512
 
@@ -34,6 +36,9 @@ struct xe_modparam xe_modparam = {
 	.probe_display =	DEFAULT_PROBE_DISPLAY,
 	.guc_log_level =	DEFAULT_GUC_LOG_LEVEL,
 	.force_probe =		DEFAULT_FORCE_PROBE,
+#ifdef CONFIG_PCI_IOV
+	.max_vfs =		DEFAULT_MAX_VFS,
+#endif
 	.wedged_mode =		DEFAULT_WEDGED_MODE,
 	.svm_notifier_size =	DEFAULT_SVM_NOTIFIER_SIZE,
 	/* the rest are 0 by default */
@@ -79,7 +84,8 @@ MODULE_PARM_DESC(force_probe,
 module_param_named(max_vfs, xe_modparam.max_vfs, uint, 0400);
 MODULE_PARM_DESC(max_vfs,
 		 "Limit number of Virtual Functions (VFs) that could be managed. "
-		 "(0 = no VFs [default]; N = allow up to N VFs)");
+		 "(0=no VFs; N=allow up to N VFs "
+		 "[default=" DEFAULT_MAX_VFS_STR "])");
 #endif
 
 module_param_named_unsafe(wedged_mode, xe_modparam.wedged_mode, int, 0600);

From 2e761039985271a69bf6eada8f236701f8594638 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 22 Jul 2025 20:26:17 +0200
Subject: [PATCH 0076/1455] drm/xe: Enable SR-IOV for ADL/ATSM

We were already testing those two platforms for a while on our CI,
but enabling flag (has_sriov) was only available on the topic branch
and only for builds with CONFIG_DRM_XE_DEBUG config.

Since those two platforms are guarded by the another enabling flag
(require_force_probe) and we believe our SR-IOV support for them is
at sufficient level to start enjoying the feature, turn on the
SR-IOV enabling flag unconditionally.

Suggested-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250722182618.30811-4-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_pci.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index f64942737a0b..c672af43d361 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -193,6 +193,7 @@ static const struct xe_device_desc adl_s_desc = {
 	.dma_mask_size = 39,
 	.has_display = true,
 	.has_llc = true,
+	.has_sriov = true,
 	.max_gt_per_tile = 1,
 	.require_force_probe = true,
 	.subplatforms = (const struct xe_subplatform_desc[]) {
@@ -210,6 +211,7 @@ static const struct xe_device_desc adl_p_desc = {
 	.dma_mask_size = 39,
 	.has_display = true,
 	.has_llc = true,
+	.has_sriov = true,
 	.max_gt_per_tile = 1,
 	.require_force_probe = true,
 	.subplatforms = (const struct xe_subplatform_desc[]) {
@@ -225,6 +227,7 @@ static const struct xe_device_desc adl_n_desc = {
 	.dma_mask_size = 39,
 	.has_display = true,
 	.has_llc = true,
+	.has_sriov = true,
 	.max_gt_per_tile = 1,
 	.require_force_probe = true,
 };
@@ -270,6 +273,7 @@ static const struct xe_device_desc ats_m_desc = {
 
 	DG2_FEATURES,
 	.has_display = false,
+	.has_sriov = true,
 };
 
 static const struct xe_device_desc dg2_desc = {

From 6983ea9cd720fdd409b4944caf9605731323bb8d Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 22 Jul 2025 20:26:18 +0200
Subject: [PATCH 0077/1455] drm/xe: Enable SR-IOV for TGL

While we don't have official CI SR-IOV coverage for the Tigerlake
platforms, we were using this platform for the feature enabling
and Xe driver already has all required changes to support it.

Since TGL platforms are guarded by the xe.require_force_probe flag
enable SR-IOV feature on them, like we recently did for ADL/ATSM.

Suggested-by: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250722182618.30811-5-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_pci.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index c672af43d361..053faa7f9b01 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -169,6 +169,7 @@ static const struct xe_device_desc tgl_desc = {
 	.dma_mask_size = 39,
 	.has_display = true,
 	.has_llc = true,
+	.has_sriov = true,
 	.max_gt_per_tile = 1,
 	.require_force_probe = true,
 };

From 78ededb610138ede03c35232f1d5cd0251a6a469 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Thu, 17 Jul 2025 23:38:14 -0700
Subject: [PATCH 0078/1455] panthor: set owner field for driver fops

It allows us to get rid of manual try_module_get / module_put.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250718063816.1452123-2-olvaffe@gmail.com
---
 drivers/gpu/drm/panthor/panthor_drv.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_drv.c b/drivers/gpu/drm/panthor/panthor_drv.c
index 0304eb83d0b2..9256806eb662 100644
--- a/drivers/gpu/drm/panthor/panthor_drv.c
+++ b/drivers/gpu/drm/panthor/panthor_drv.c
@@ -1401,14 +1401,9 @@ panthor_open(struct drm_device *ddev, struct drm_file *file)
 	struct panthor_file *pfile;
 	int ret;
 
-	if (!try_module_get(THIS_MODULE))
-		return -EINVAL;
-
 	pfile = kzalloc(sizeof(*pfile), GFP_KERNEL);
-	if (!pfile) {
-		ret = -ENOMEM;
-		goto err_put_mod;
-	}
+	if (!pfile)
+		return -ENOMEM;
 
 	pfile->ptdev = ptdev;
 	pfile->user_mmio.offset = DRM_PANTHOR_USER_MMIO_OFFSET;
@@ -1440,9 +1435,6 @@ err_destroy_vm_pool:
 
 err_free_file:
 	kfree(pfile);
-
-err_put_mod:
-	module_put(THIS_MODULE);
 	return ret;
 }
 
@@ -1455,7 +1447,6 @@ panthor_postclose(struct drm_device *ddev, struct drm_file *file)
 	panthor_vm_pool_destroy(pfile);
 
 	kfree(pfile);
-	module_put(THIS_MODULE);
 }
 
 static const struct drm_ioctl_desc panthor_drm_driver_ioctls[] = {
@@ -1556,6 +1547,7 @@ static void panthor_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 }
 
 static const struct file_operations panthor_drm_driver_fops = {
+	.owner = THIS_MODULE,
 	.open = drm_open,
 	.release = drm_release,
 	.unlocked_ioctl = drm_ioctl,

From 33b9cb6dcda2520600ac4fec725946af32c2e586 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Thu, 17 Jul 2025 23:38:15 -0700
Subject: [PATCH 0079/1455] panthor: save task pid and comm in panthor_group

We would like to report them on gpu errors.

We choose to save the info on panthor_group_create rather than on
panthor_open because, when the two differ, we are more interested in the
task that created the group.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250718063816.1452123-3-olvaffe@gmail.com
---
 drivers/gpu/drm/panthor/panthor_sched.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
index 8f17394cc82a..9f4d04214053 100644
--- a/drivers/gpu/drm/panthor/panthor_sched.c
+++ b/drivers/gpu/drm/panthor/panthor_sched.c
@@ -641,6 +641,15 @@ struct panthor_group {
 		size_t kbo_sizes;
 	} fdinfo;
 
+	/** @task_info: Info of current->group_leader that created the group. */
+	struct {
+		/** @task_info.pid: pid of current->group_leader */
+		pid_t pid;
+
+		/** @task_info.comm: comm of current->group_leader */
+		char comm[TASK_COMM_LEN];
+	} task_info;
+
 	/** @state: Group state. */
 	enum panthor_group_state state;
 
@@ -3389,6 +3398,14 @@ err_free_queue:
 	return ERR_PTR(ret);
 }
 
+static void group_init_task_info(struct panthor_group *group)
+{
+	struct task_struct *task = current->group_leader;
+
+	group->task_info.pid = task->pid;
+	get_task_comm(group->task_info.comm, task);
+}
+
 static void add_group_kbo_sizes(struct panthor_device *ptdev,
 				struct panthor_group *group)
 {
@@ -3540,6 +3557,8 @@ int panthor_group_create(struct panthor_file *pfile,
 	add_group_kbo_sizes(group->ptdev, group);
 	spin_lock_init(&group->fdinfo.lock);
 
+	group_init_task_info(group);
+
 	return gid;
 
 err_put_group:

From e48123c607a0db8b9ad02f83c8c3d39918dbda06 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Thu, 17 Jul 2025 23:38:16 -0700
Subject: [PATCH 0080/1455] panthor: dump task pid and comm on gpu errors

It is useful to know which tasks cause gpu errors.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250718063816.1452123-4-olvaffe@gmail.com
---
 drivers/gpu/drm/panthor/panthor_sched.c | 24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_sched.c b/drivers/gpu/drm/panthor/panthor_sched.c
index 9f4d04214053..ba5dc3e443d9 100644
--- a/drivers/gpu/drm/panthor/panthor_sched.c
+++ b/drivers/gpu/drm/panthor/panthor_sched.c
@@ -1364,8 +1364,12 @@ cs_slot_process_fatal_event_locked(struct panthor_device *ptdev,
 	fatal = cs_iface->output->fatal;
 	info = cs_iface->output->fatal_info;
 
-	if (group)
+	if (group) {
+		drm_warn(&ptdev->base, "CS_FATAL: pid=%d, comm=%s\n",
+			 group->task_info.pid, group->task_info.comm);
+
 		group->fatal_queues |= BIT(cs_id);
+	}
 
 	if (CS_EXCEPTION_TYPE(fatal) == DRM_PANTHOR_EXCEPTION_CS_UNRECOVERABLE) {
 		/* If this exception is unrecoverable, queue a reset, and make
@@ -1425,6 +1429,11 @@ cs_slot_process_fault_event_locked(struct panthor_device *ptdev,
 		spin_unlock(&queue->fence_ctx.lock);
 	}
 
+	if (group) {
+		drm_warn(&ptdev->base, "CS_FAULT: pid=%d, comm=%s\n",
+			 group->task_info.pid, group->task_info.comm);
+	}
+
 	drm_warn(&ptdev->base,
 		 "CSG slot %d CS slot: %d\n"
 		 "CS_FAULT.EXCEPTION_TYPE: 0x%x (%s)\n"
@@ -1641,11 +1650,15 @@ csg_slot_process_progress_timer_event_locked(struct panthor_device *ptdev, u32 c
 
 	lockdep_assert_held(&sched->lock);
 
-	drm_warn(&ptdev->base, "CSG slot %d progress timeout\n", csg_id);
-
 	group = csg_slot->group;
-	if (!drm_WARN_ON(&ptdev->base, !group))
+	if (!drm_WARN_ON(&ptdev->base, !group)) {
+		drm_warn(&ptdev->base, "CSG_PROGRESS_TIMER_EVENT: pid=%d, comm=%s\n",
+			 group->task_info.pid, group->task_info.comm);
+
 		group->timedout = true;
+	}
+
+	drm_warn(&ptdev->base, "CSG slot %d progress timeout\n", csg_id);
 
 	sched_queue_delayed_work(sched, tick, 0);
 }
@@ -3227,7 +3240,8 @@ queue_timedout_job(struct drm_sched_job *sched_job)
 	struct panthor_scheduler *sched = ptdev->scheduler;
 	struct panthor_queue *queue = group->queues[job->queue_idx];
 
-	drm_warn(&ptdev->base, "job timeout\n");
+	drm_warn(&ptdev->base, "job timeout: pid=%d, comm=%s, seqno=%llu\n",
+		 group->task_info.pid, group->task_info.comm, job->done_fence->seqno);
 
 	drm_WARN_ON(&ptdev->base, atomic_read(&sched->reset.in_progress));
 

From 538b27a09af974f2c5e7d358a24af7a722fa4bfc Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Wed, 23 Jul 2025 15:30:15 +0200
Subject: [PATCH 0081/1455] drm/xe: Make GGTT TLB invalidation failure message
 GT oriented

GGTT TLB invalidation is performed on the specific GT, thus any
failure message shall be also GT specific. And to help investigate
any unexpected failures, promote message from warn level to WARN
to get full call stack of this unlikely case.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250723133015.206601-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_ggtt.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index 29d4d3f51da1..249713139f69 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -439,8 +439,7 @@ static void ggtt_invalidate_gt_tlb(struct xe_gt *gt)
 		return;
 
 	err = xe_gt_tlb_invalidation_ggtt(gt);
-	if (err)
-		drm_warn(&gt_to_xe(gt)->drm, "xe_gt_tlb_invalidation_ggtt error=%d", err);
+	xe_gt_WARN(gt, err, "Failed to invalidate GGTT (%pe)", ERR_PTR(err));
 }
 
 static void xe_ggtt_invalidate(struct xe_ggtt *ggtt)

From 159afd92bae8153bdd8d8b34aea0d463fe19c978 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Wed, 23 Jul 2025 19:56:39 +0200
Subject: [PATCH 0082/1455] drm/xe/guc: Clear whole g2h_fence during
 initialization

The struct g2h_fence must be explicitly initializated using the
g2h_fence_init() function to avoid trash values in its members,
but we missed to update this helper function with the new member.

To fix that and avoid any future mistakes, memset the whole struct
first, then update remaining non-zero members.

Fixes: 94de94d24ea8 ("drm/xe/guc: Cancel ongoing H2G requests when stopping CT")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Lukasz Laguna <lukasz.laguna@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250723175639.206875-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_guc_ct.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index b6acccfcd351..3f4e6a46ff16 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -95,12 +95,8 @@ struct g2h_fence {
 
 static void g2h_fence_init(struct g2h_fence *g2h_fence, u32 *response_buffer)
 {
+	memset(g2h_fence, 0, sizeof(*g2h_fence));
 	g2h_fence->response_buffer = response_buffer;
-	g2h_fence->response_data = 0;
-	g2h_fence->response_len = 0;
-	g2h_fence->fail = false;
-	g2h_fence->retry = false;
-	g2h_fence->done = false;
 	g2h_fence->seqno = ~0x0;
 }
 

From bf91bac3ef027852b3e7698af6a00e4df124072c Mon Sep 17 00:00:00 2001
From: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Date: Fri, 18 Jul 2025 10:28:04 +0000
Subject: [PATCH 0083/1455] drm/i915: Add braces around the else block in
 clflush_write32()

According to the kernel coding style, if only one branch of a
conditional statement is a single statement, braces should
still be used in both branches.

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/20250718102752.684975-2-sebastian.brzezinka@intel.com
---
 drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index ca7e9216934a..0801d4a140e3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -1382,8 +1382,9 @@ static void clflush_write32(u32 *addr, u32 value, unsigned int flushes)
 		 */
 		if (flushes & CLFLUSH_AFTER)
 			drm_clflush_virt_range(addr, sizeof(*addr));
-	} else
+	} else {
 		*addr = value;
+	}
 }
 
 static u64

From c8bdf3165fccadac12d670e94ea4a9a893a5a813 Mon Sep 17 00:00:00 2001
From: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Date: Fri, 18 Jul 2025 10:28:19 +0000
Subject: [PATCH 0084/1455] drm/i915: Replace empty conditional with continue
 in eb_relocate_vma()

Simplifies the control flow by replacing an empty
`if (likely(offset == 0))` block with a `continue` statement. This
improves readability and avoids unnecessary nesting.

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/20250718102752.684975-4-sebastian.brzezinka@intel.com
---
 .../gpu/drm/i915/gem/i915_gem_execbuffer.c    | 54 +++++++++----------
 1 file changed, 27 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index 0801d4a140e3..f243f8a5215d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -1568,36 +1568,36 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct eb_vma *ev)
 		do {
 			u64 offset = eb_relocate_entry(eb, ev, r);
 
-			if (likely(offset == 0)) {
-			} else if ((s64)offset < 0) {
+			if (likely(offset == 0))
+				continue;
+
+			if ((s64)offset < 0) {
 				remain = (int)offset;
 				goto out;
-			} else {
-				/*
-				 * Note that reporting an error now
-				 * leaves everything in an inconsistent
-				 * state as we have *already* changed
-				 * the relocation value inside the
-				 * object. As we have not changed the
-				 * reloc.presumed_offset or will not
-				 * change the execobject.offset, on the
-				 * call we may not rewrite the value
-				 * inside the object, leaving it
-				 * dangling and causing a GPU hang. Unless
-				 * userspace dynamically rebuilds the
-				 * relocations on each execbuf rather than
-				 * presume a static tree.
-				 *
-				 * We did previously check if the relocations
-				 * were writable (access_ok), an error now
-				 * would be a strange race with mprotect,
-				 * having already demonstrated that we
-				 * can read from this userspace address.
-				 */
-				offset = gen8_canonical_addr(offset & ~UPDATE);
-				__put_user(offset,
-					   &urelocs[r - stack].presumed_offset);
 			}
+			/*
+			 * Note that reporting an error now
+			 * leaves everything in an inconsistent
+			 * state as we have *already* changed
+			 * the relocation value inside the
+			 * object. As we have not changed the
+			 * reloc.presumed_offset or will not
+			 * change the execobject.offset, on the
+			 * call we may not rewrite the value
+			 * inside the object, leaving it
+			 * dangling and causing a GPU hang. Unless
+			 * userspace dynamically rebuilds the
+			 * relocations on each execbuf rather than
+			 * presume a static tree.
+			 *
+			 * We did previously check if the relocations
+			 * were writable (access_ok), an error now
+			 * would be a strange race with mprotect,
+			 * having already demonstrated that we
+			 * can read from this userspace address.
+			 */
+			offset = gen8_canonical_addr(offset & ~UPDATE);
+			__put_user(offset, &urelocs[r - stack].presumed_offset);
 		} while (r++, --count);
 		urelocs += ARRAY_SIZE(stack);
 	} while (remain);

From a313d9059f00adefdadc6c5612c104bbd78a4808 Mon Sep 17 00:00:00 2001
From: Nitin Gote <nitin.r.gote@intel.com>
Date: Wed, 23 Jul 2025 19:40:39 +0530
Subject: [PATCH 0085/1455] drm/xe: Rename MCFG_MCR_SELECTOR to STEER_SEMAPHORE

The register at offset 0xfd0 was incorrectly named MCFG_MCR_SELECTOR,
likely copied from i915. According to the hardware specification (Bspec),
this register is actually called STEER_SEMAPHORE.

Rename the register definition and update its usage in xe_gt_mcr.c to
match the official hardware documentation.

No functional changes.

v2: Add Bspec reference (Tejas)

Bspec: 67113
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://lore.kernel.org/r/20250723141039.3848390-1-nitin.r.gote@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_gt_regs.h | 2 +-
 drivers/gpu/drm/xe/xe_gt_mcr.c       | 4 +---
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
index 5cd5ab8529c5..f96b2e2b3064 100644
--- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
@@ -42,7 +42,7 @@
 #define FORCEWAKE_ACK_GSC			XE_REG(0xdf8)
 #define FORCEWAKE_ACK_GT_MTL			XE_REG(0xdfc)
 
-#define MCFG_MCR_SELECTOR			XE_REG(0xfd0)
+#define STEER_SEMAPHORE				XE_REG(0xfd0)
 #define MTL_MCR_SELECTOR			XE_REG(0xfd4)
 #define SF_MCR_SELECTOR				XE_REG(0xfd8)
 #define MCR_SELECTOR				XE_REG(0xfdc)
diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c
index 64a2f0d6aaf9..683ac021a06d 100644
--- a/drivers/gpu/drm/xe/xe_gt_mcr.c
+++ b/drivers/gpu/drm/xe/xe_gt_mcr.c
@@ -46,8 +46,6 @@
  * MCR registers are not available on Virtual Function (VF).
  */
 
-#define STEER_SEMAPHORE		XE_REG(0xFD0)
-
 static inline struct xe_reg to_xe_reg(struct xe_reg_mcr reg_mcr)
 {
 	return reg_mcr.__reg;
@@ -533,7 +531,7 @@ void xe_gt_mcr_set_implicit_defaults(struct xe_gt *gt)
 		u32 steer_val = REG_FIELD_PREP(MCR_SLICE_MASK, 0) |
 			REG_FIELD_PREP(MCR_SUBSLICE_MASK, 2);
 
-		xe_mmio_write32(&gt->mmio, MCFG_MCR_SELECTOR, steer_val);
+		xe_mmio_write32(&gt->mmio, STEER_SEMAPHORE, steer_val);
 		xe_mmio_write32(&gt->mmio, SF_MCR_SELECTOR, steer_val);
 		/*
 		 * For GAM registers, all reads should be directed to instance 1

From a98cdd979c2b529363be502cb057aea79e65fbf7 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Thu, 24 Jul 2025 14:16:58 +0100
Subject: [PATCH 0086/1455] drm/xe: Use emit_flush_imm_ggtt helper instead of
 open coding

Helper is already there so lets just use it.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250724131711.74291-2-tvrtko.ursulin@igalia.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_ring_ops.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
index 7b50c7c1ee21..e8f22ec5f9af 100644
--- a/drivers/gpu/drm/xe/xe_ring_ops.c
+++ b/drivers/gpu/drm/xe/xe_ring_ops.c
@@ -415,11 +415,9 @@ static void emit_migration_job_gen12(struct xe_sched_job *job,
 
 	i = emit_bb_start(job->ptrs[1].batch_addr, BIT(8), dw, i);
 
-	dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | job->migrate_flush_flags |
-		MI_FLUSH_DW_OP_STOREDW | MI_FLUSH_IMM_DW;
-	dw[i++] = xe_lrc_seqno_ggtt_addr(lrc) | MI_FLUSH_DW_USE_GTT;
-	dw[i++] = 0;
-	dw[i++] = seqno; /* value */
+	i = emit_flush_imm_ggtt(xe_lrc_seqno_ggtt_addr(lrc), seqno,
+				MI_INVALIDATE_TLB | job->migrate_flush_flags,
+				dw, i);
 
 	i = emit_user_interrupt(dw, i);
 

From d72779c29d82c6e371cea8b427550bd6923c2577 Mon Sep 17 00:00:00 2001
From: Sk Anirban <sk.anirban@intel.com>
Date: Wed, 16 Jul 2025 15:46:23 +0530
Subject: [PATCH 0087/1455] drm/xe/ptl: Apply Wa_16026007364

As part of this WA GuC will save and restore value of two XE3_Media
control registers that were not included in the HW power context.

Signed-off-by: Sk Anirban <sk.anirban@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://lore.kernel.org/r/20250716101622.3421480-2-sk.anirban@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/abi/guc_klvs_abi.h |  1 +
 drivers/gpu/drm/xe/xe_guc_ads.c       | 35 +++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_wa_oob.rules    |  1 +
 3 files changed, 37 insertions(+)

diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
index 0366a9da5977..31dbfeee289e 100644
--- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
@@ -396,6 +396,7 @@ enum xe_guc_klv_ids {
 	GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET				= 0x9009,
 	GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO					= 0x900a,
 	GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH					= 0x900b,
+	GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG					= 0x900c,
 };
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
index 131cfc56be00..8ff8626227ae 100644
--- a/drivers/gpu/drm/xe/xe_guc_ads.c
+++ b/drivers/gpu/drm/xe/xe_guc_ads.c
@@ -284,6 +284,35 @@ static size_t calculate_golden_lrc_size(struct xe_guc_ads *ads)
 	return total_size;
 }
 
+static void guc_waklv_enable_two_word(struct xe_guc_ads *ads,
+				      enum xe_guc_klv_ids klv_id,
+				      u32 value1,
+				      u32 value2,
+				      u32 *offset, u32 *remain)
+{
+	u32 size;
+	u32 klv_entry[] = {
+			/* 16:16 key/length */
+			FIELD_PREP(GUC_KLV_0_KEY, klv_id) |
+			FIELD_PREP(GUC_KLV_0_LEN, 2),
+			value1,
+			value2,
+			/* 2 dword data */
+	};
+
+	size = sizeof(klv_entry);
+
+	if (*remain < size) {
+		drm_warn(&ads_to_xe(ads)->drm,
+			 "w/a klv buffer too small to add klv id %d\n", klv_id);
+	} else {
+		xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), *offset,
+				 klv_entry, size);
+		*offset += size;
+		*remain -= size;
+	}
+}
+
 static void guc_waklv_enable_one_word(struct xe_guc_ads *ads,
 				      enum xe_guc_klv_ids klv_id,
 				      u32 value,
@@ -381,6 +410,12 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 		guc_waklv_enable_simple(ads,
 					GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH,
 					&offset, &remain);
+	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 47, 0) && XE_WA(gt, 16026007364))
+		guc_waklv_enable_two_word(ads,
+					  GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG,
+					  0x0,
+					  0xF,
+					  &offset, &remain);
 
 	size = guc_ads_waklv_size(ads) - remain;
 	if (!size)
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index e990f20eccfe..46ad72bb1f8d 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -72,3 +72,4 @@ no_media_l3	MEDIA_VERSION(3000)
 14022085890	GRAPHICS_VERSION(2001)
 
 15015404425_disable	PLATFORM(PANTHERLAKE), MEDIA_STEP(B0, FOREVER)
+16026007364    MEDIA_VERSION(3000)

From c3ead4ecfc4c02806aa1c202a595ee3ecb2c039a Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:10 -0700
Subject: [PATCH 0088/1455] drm/xe: Explicitly mark migration queues with flag

Rather than inferring if an exec queue is a migration queue for a flag,
explicitly mark migration queues with a flag.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Francois Dugast <francois.dugast@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-2-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_exec_queue_types.h | 2 ++
 drivers/gpu/drm/xe/xe_migrate.c          | 6 ++++--
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h
index cc1cffb5c87f..abdf4a57e6e2 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue_types.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h
@@ -87,6 +87,8 @@ struct xe_exec_queue {
 #define EXEC_QUEUE_FLAG_HIGH_PRIORITY		BIT(4)
 /* flag to indicate low latency hint to guc */
 #define EXEC_QUEUE_FLAG_LOW_LATENCY		BIT(5)
+/* for migration (kernel copy, clear, bind) jobs */
+#define EXEC_QUEUE_FLAG_MIGRATE			BIT(6)
 
 	/**
 	 * @flags: flags for this exec queue, should statically setup aside from ban
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 0a61fb8b123a..ed04c2b368dc 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -439,12 +439,14 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 		m->q = xe_exec_queue_create(xe, vm, logical_mask, 1, hwe,
 					    EXEC_QUEUE_FLAG_KERNEL |
 					    EXEC_QUEUE_FLAG_PERMANENT |
-					    EXEC_QUEUE_FLAG_HIGH_PRIORITY, 0);
+					    EXEC_QUEUE_FLAG_HIGH_PRIORITY |
+					    EXEC_QUEUE_FLAG_MIGRATE, 0);
 	} else {
 		m->q = xe_exec_queue_create_class(xe, primary_gt, vm,
 						  XE_ENGINE_CLASS_COPY,
 						  EXEC_QUEUE_FLAG_KERNEL |
-						  EXEC_QUEUE_FLAG_PERMANENT, 0);
+						  EXEC_QUEUE_FLAG_PERMANENT |
+						  EXEC_QUEUE_FLAG_MIGRATE, 0);
 	}
 	if (IS_ERR(m->q)) {
 		xe_vm_close_and_put(vm);

From 69f187d446c94f66d413819f16e2725b7c31f378 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:11 -0700
Subject: [PATCH 0089/1455] drm/xe: Add generic dependecy jobs / scheduler

Add generic dependecy jobs / scheduler which serves as wrapper for DRM
scheduler. Useful when we want delay a generic operation until a
dma-fence signals.

Existing use cases could be destroying of resources based fences /
dma-resv, the preempt rebind worker, and pipelined GT TLB invalidations.

Written in such a way it could be moved to DRM subsystem if needed.

v3:
 - Remove unnecessary cast (Staurt)

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-3-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/Makefile           |   1 +
 drivers/gpu/drm/xe/xe_dep_job_types.h |  29 ++++++
 drivers/gpu/drm/xe/xe_dep_scheduler.c | 143 ++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_dep_scheduler.h |  21 ++++
 4 files changed, 194 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_dep_job_types.h
 create mode 100644 drivers/gpu/drm/xe/xe_dep_scheduler.c
 create mode 100644 drivers/gpu/drm/xe/xe_dep_scheduler.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index b046e1b8d726..6e1fd6560557 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -35,6 +35,7 @@ $(obj)/generated/%_device_wa_oob.c $(obj)/generated/%_device_wa_oob.h: $(obj)/xe
 xe-y += xe_bb.o \
 	xe_bo.o \
 	xe_bo_evict.o \
+	xe_dep_scheduler.o \
 	xe_devcoredump.o \
 	xe_device.o \
 	xe_device_sysfs.o \
diff --git a/drivers/gpu/drm/xe/xe_dep_job_types.h b/drivers/gpu/drm/xe/xe_dep_job_types.h
new file mode 100644
index 000000000000..c6a484f24c8c
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_dep_job_types.h
@@ -0,0 +1,29 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_DEP_JOB_TYPES_H_
+#define _XE_DEP_JOB_TYPES_H_
+
+#include <drm/gpu_scheduler.h>
+
+struct xe_dep_job;
+
+/** struct xe_dep_job_ops - Generic Xe dependency job operations */
+struct xe_dep_job_ops {
+	/** @run_job: Run generic Xe dependency job */
+	struct dma_fence *(*run_job)(struct xe_dep_job *job);
+	/** @free_job: Free generic Xe dependency job */
+	void (*free_job)(struct xe_dep_job *job);
+};
+
+/** struct xe_dep_job - Generic dependency Xe job */
+struct xe_dep_job {
+	/** @drm: base DRM scheduler job */
+	struct drm_sched_job drm;
+	/** @ops: dependency job operations */
+	const struct xe_dep_job_ops *ops;
+};
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_dep_scheduler.c b/drivers/gpu/drm/xe/xe_dep_scheduler.c
new file mode 100644
index 000000000000..9bd3bfd2e526
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_dep_scheduler.c
@@ -0,0 +1,143 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/slab.h>
+
+#include <drm/gpu_scheduler.h>
+
+#include "xe_dep_job_types.h"
+#include "xe_dep_scheduler.h"
+#include "xe_device_types.h"
+
+/**
+ * DOC: Xe Dependency Scheduler
+ *
+ * The Xe dependency scheduler is a simple wrapper built around the DRM
+ * scheduler to execute jobs once their dependencies are resolved (i.e., all
+ * input fences specified as dependencies are signaled). The jobs that are
+ * executed contain virtual functions to run (execute) and free the job,
+ * allowing a single dependency scheduler to handle jobs performing different
+ * operations.
+ *
+ * Example use cases include deferred resource freeing, TLB invalidations after
+ * bind jobs, etc.
+ */
+
+/** struct xe_dep_scheduler - Generic Xe dependency scheduler */
+struct xe_dep_scheduler {
+	/** @sched: DRM GPU scheduler */
+	struct drm_gpu_scheduler sched;
+	/** @entity: DRM scheduler entity  */
+	struct drm_sched_entity entity;
+	/** @rcu: For safe freeing of exported dma fences */
+	struct rcu_head rcu;
+};
+
+static struct dma_fence *xe_dep_scheduler_run_job(struct drm_sched_job *drm_job)
+{
+	struct xe_dep_job *dep_job =
+		container_of(drm_job, typeof(*dep_job), drm);
+
+	return dep_job->ops->run_job(dep_job);
+}
+
+static void xe_dep_scheduler_free_job(struct drm_sched_job *drm_job)
+{
+	struct xe_dep_job *dep_job =
+		container_of(drm_job, typeof(*dep_job), drm);
+
+	dep_job->ops->free_job(dep_job);
+}
+
+static const struct drm_sched_backend_ops sched_ops = {
+	.run_job = xe_dep_scheduler_run_job,
+	.free_job = xe_dep_scheduler_free_job,
+};
+
+/**
+ * xe_dep_scheduler_create() - Generic Xe dependency scheduler create
+ * @xe: Xe device
+ * @submit_wq: Submit workqueue struct (can be NULL)
+ * @name: Name of dependency scheduler
+ * @job_limit: Max dependency jobs that can be scheduled
+ *
+ * Create a generic Xe dependency scheduler and initialize internal DRM
+ * scheduler objects.
+ *
+ * Return: Generic Xe dependency scheduler object on success, ERR_PTR failure
+ */
+struct xe_dep_scheduler *
+xe_dep_scheduler_create(struct xe_device *xe,
+			struct workqueue_struct *submit_wq,
+			const char *name, u32 job_limit)
+{
+	struct xe_dep_scheduler *dep_scheduler;
+	struct drm_gpu_scheduler *sched;
+	const struct drm_sched_init_args args = {
+		.ops = &sched_ops,
+		.submit_wq = submit_wq,
+		.num_rqs = 1,
+		.credit_limit = job_limit,
+		.timeout = MAX_SCHEDULE_TIMEOUT,
+		.name = name,
+		.dev = xe->drm.dev,
+	};
+	int err;
+
+	dep_scheduler = kzalloc(sizeof(*dep_scheduler), GFP_KERNEL);
+	if (!dep_scheduler)
+		return ERR_PTR(-ENOMEM);
+
+	err = drm_sched_init(&dep_scheduler->sched, &args);
+	if (err)
+		goto err_free;
+
+	sched = &dep_scheduler->sched;
+	err = drm_sched_entity_init(&dep_scheduler->entity, 0, &sched, 1, NULL);
+	if (err)
+		goto err_sched;
+
+	init_rcu_head(&dep_scheduler->rcu);
+
+	return dep_scheduler;
+
+err_sched:
+	drm_sched_fini(&dep_scheduler->sched);
+err_free:
+	kfree(dep_scheduler);
+
+	return ERR_PTR(err);
+}
+
+/**
+ * xe_dep_scheduler_fini() - Generic Xe dependency scheduler finalize
+ * @dep_scheduler: Generic Xe dependency scheduler object
+ *
+ * Finalize internal DRM scheduler objects and free generic Xe dependency
+ * scheduler object
+ */
+void xe_dep_scheduler_fini(struct xe_dep_scheduler *dep_scheduler)
+{
+	drm_sched_entity_fini(&dep_scheduler->entity);
+	drm_sched_fini(&dep_scheduler->sched);
+	/*
+	 * RCU free due sched being exported via DRM scheduler fences
+	 * (timeline name).
+	 */
+	kfree_rcu(dep_scheduler, rcu);
+}
+
+/**
+ * xe_dep_scheduler_entity() - Retrieve a generic Xe dependency scheduler
+ *                             DRM scheduler entity
+ * @dep_scheduler: Generic Xe dependency scheduler object
+ *
+ * Return: The generic Xe dependency scheduler's DRM scheduler entity
+ */
+struct drm_sched_entity *
+xe_dep_scheduler_entity(struct xe_dep_scheduler *dep_scheduler)
+{
+	return &dep_scheduler->entity;
+}
diff --git a/drivers/gpu/drm/xe/xe_dep_scheduler.h b/drivers/gpu/drm/xe/xe_dep_scheduler.h
new file mode 100644
index 000000000000..853961eec64b
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_dep_scheduler.h
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/types.h>
+
+struct drm_sched_entity;
+struct workqueue_struct;
+struct xe_dep_scheduler;
+struct xe_device;
+
+struct xe_dep_scheduler *
+xe_dep_scheduler_create(struct xe_device *xe,
+			struct workqueue_struct *submit_wq,
+			const char *name, u32 job_limit);
+
+void xe_dep_scheduler_fini(struct xe_dep_scheduler *dep_scheduler);
+
+struct drm_sched_entity *
+xe_dep_scheduler_entity(struct xe_dep_scheduler *dep_scheduler);

From ada51219489f53f9f35833c3d0d8a4fe3f4c5705 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:12 -0700
Subject: [PATCH 0090/1455] drm/xe: Create ordered workqueue for GT TLB
 invalidation jobs

No sense to schedule GT TLB invalidation jobs in parallel which target
the same GT given these all contend on the same lock, create ordered
workqueue for GT TLB invalidation jobs.

v3:
 - Fix type in commmit message (Stuart)

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-4-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 8 ++++++++
 drivers/gpu/drm/xe/xe_gt_types.h            | 2 ++
 2 files changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
index 086c12ee3d9d..02f0bb92d6e0 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
@@ -3,6 +3,8 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include <drm/drm_managed.h>
+
 #include "xe_gt_tlb_invalidation.h"
 
 #include "abi/guc_actions_abi.h"
@@ -123,6 +125,12 @@ int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt)
 	INIT_DELAYED_WORK(&gt->tlb_invalidation.fence_tdr,
 			  xe_gt_tlb_fence_timeout);
 
+	gt->tlb_invalidation.job_wq =
+		drmm_alloc_ordered_workqueue(&gt_to_xe(gt)->drm, "gt-tbl-inval-job-wq",
+					     WQ_MEM_RECLAIM);
+	if (IS_ERR(gt->tlb_invalidation.job_wq))
+		return PTR_ERR(gt->tlb_invalidation.job_wq);
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
index 96344c604726..dfd4a16da5f0 100644
--- a/drivers/gpu/drm/xe/xe_gt_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_types.h
@@ -210,6 +210,8 @@ struct xe_gt {
 		 * xe_gt_tlb_fence_timeout after the timeut interval is over.
 		 */
 		struct delayed_work fence_tdr;
+		/** @wtlb_invalidation.wq: schedules GT TLB invalidation jobs */
+		struct workqueue_struct *job_wq;
 		/** @tlb_invalidation.lock: protects TLB invalidation fences */
 		spinlock_t lock;
 	} tlb_invalidation;

From 535c445eb94c9abce3350aa61424fc34b8138ee8 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:13 -0700
Subject: [PATCH 0091/1455] drm/xe: Add dependency scheduler for GT TLB
 invalidations to bind queues
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a generic dependency scheduler for GT TLB invalidations, used to
schedule jobs that issue GT TLB invalidations to bind queues.

v2:
 - Use shared GT TLB invalidation queue for dep scheduler
 - Break allocation of dep scheduler into its own function
 - Add define for max number tlb invalidations
 - Skip media if not present

Suggested-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-5-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_exec_queue.c       | 48 ++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_exec_queue_types.h | 13 +++++++
 2 files changed, 61 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 8991b4aed440..d5958518ef00 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -12,6 +12,7 @@
 #include <drm/drm_file.h>
 #include <uapi/drm/xe_drm.h>
 
+#include "xe_dep_scheduler.h"
 #include "xe_device.h"
 #include "xe_gt.h"
 #include "xe_hw_engine_class_sysfs.h"
@@ -39,6 +40,12 @@ static int exec_queue_user_extensions(struct xe_device *xe, struct xe_exec_queue
 
 static void __xe_exec_queue_free(struct xe_exec_queue *q)
 {
+	int i;
+
+	for (i = 0; i < XE_EXEC_QUEUE_TLB_INVAL_COUNT; ++i)
+		if (q->tlb_inval[i].dep_scheduler)
+			xe_dep_scheduler_fini(q->tlb_inval[i].dep_scheduler);
+
 	if (xe_exec_queue_uses_pxp(q))
 		xe_pxp_exec_queue_remove(gt_to_xe(q->gt)->pxp, q);
 	if (q->vm)
@@ -50,6 +57,39 @@ static void __xe_exec_queue_free(struct xe_exec_queue *q)
 	kfree(q);
 }
 
+static int alloc_dep_schedulers(struct xe_device *xe, struct xe_exec_queue *q)
+{
+	struct xe_tile *tile = gt_to_tile(q->gt);
+	int i;
+
+	for (i = 0; i < XE_EXEC_QUEUE_TLB_INVAL_COUNT; ++i) {
+		struct xe_dep_scheduler *dep_scheduler;
+		struct xe_gt *gt;
+		struct workqueue_struct *wq;
+
+		if (i == XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT)
+			gt = tile->primary_gt;
+		else
+			gt = tile->media_gt;
+
+		if (!gt)
+			continue;
+
+		wq = gt->tlb_invalidation.job_wq;
+
+#define MAX_TLB_INVAL_JOBS	16	/* Picking a reasonable value */
+		dep_scheduler = xe_dep_scheduler_create(xe, wq, q->name,
+							MAX_TLB_INVAL_JOBS);
+		if (IS_ERR(dep_scheduler))
+			return PTR_ERR(dep_scheduler);
+
+		q->tlb_inval[i].dep_scheduler = dep_scheduler;
+	}
+#undef MAX_TLB_INVAL_JOBS
+
+	return 0;
+}
+
 static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe,
 						   struct xe_vm *vm,
 						   u32 logical_mask,
@@ -94,6 +134,14 @@ static struct xe_exec_queue *__xe_exec_queue_alloc(struct xe_device *xe,
 	else
 		q->sched_props.priority = XE_EXEC_QUEUE_PRIORITY_NORMAL;
 
+	if (q->flags & (EXEC_QUEUE_FLAG_MIGRATE | EXEC_QUEUE_FLAG_VM)) {
+		err = alloc_dep_schedulers(xe, q);
+		if (err) {
+			__xe_exec_queue_free(q);
+			return ERR_PTR(err);
+		}
+	}
+
 	if (vm)
 		q->vm = xe_vm_get(vm);
 
diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h
index abdf4a57e6e2..ba443a497b38 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue_types.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h
@@ -134,6 +134,19 @@ struct xe_exec_queue {
 		struct list_head link;
 	} lr;
 
+#define XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT	0
+#define XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT	1
+#define XE_EXEC_QUEUE_TLB_INVAL_COUNT		(XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT  + 1)
+
+	/** @tlb_inval: TLB invalidations exec queue state */
+	struct {
+		/**
+		 * @tlb_inval.dep_scheduler: The TLB invalidation
+		 * dependency scheduler
+		 */
+		struct xe_dep_scheduler *dep_scheduler;
+	} tlb_inval[XE_EXEC_QUEUE_TLB_INVAL_COUNT];
+
 	/** @pxp: PXP info tracking */
 	struct {
 		/** @pxp.type: PXP session type used by this queue */

From dba89840a920ffcab7a722299510f89d7800d79b Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:14 -0700
Subject: [PATCH 0092/1455] drm/xe: Add GT TLB invalidation jobs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add GT TLB invalidation jobs which issue GT TLB invalidations. Built on
top of Xe generic dependency scheduler.

v2:
 - Fix checkpatch
v3:
 - Fix kernel doc in xe_gt_tlb_inval_job_alloc_dep,
   xe_gt_tlb_inval_job_push
 - Use IS_ERR_OR_NULL in xe_gt_tlb_inval_job_put
 - Squash migrate lock / unlock helpers into this patch (Stuart)

Suggested-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-6-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/Makefile              |   1 +
 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c | 274 +++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h |  34 +++
 drivers/gpu/drm/xe/xe_migrate.c          |  36 +++
 drivers/gpu/drm/xe/xe_migrate.h          |   4 +
 5 files changed, 349 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
 create mode 100644 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 6e1fd6560557..7eaf4054c99e 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -62,6 +62,7 @@ xe-y += xe_bb.o \
 	xe_gt_sysfs.o \
 	xe_gt_throttle.o \
 	xe_gt_tlb_invalidation.o \
+	xe_gt_tlb_inval_job.o \
 	xe_gt_topology.o \
 	xe_guc.o \
 	xe_guc_ads.o \
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
new file mode 100644
index 000000000000..e9255be26467
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
@@ -0,0 +1,274 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "xe_dep_job_types.h"
+#include "xe_dep_scheduler.h"
+#include "xe_exec_queue.h"
+#include "xe_gt.h"
+#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval_job.h"
+#include "xe_migrate.h"
+#include "xe_pm.h"
+
+/** struct xe_gt_tlb_inval_job - GT TLB invalidation job */
+struct xe_gt_tlb_inval_job {
+	/** @dep: base generic dependency Xe job */
+	struct xe_dep_job dep;
+	/** @gt: GT to invalidate */
+	struct xe_gt *gt;
+	/** @q: exec queue issuing the invalidate */
+	struct xe_exec_queue *q;
+	/** @refcount: ref count of this job */
+	struct kref refcount;
+	/**
+	 * @fence: dma fence to indicate completion. 1 way relationship - job
+	 * can safely reference fence, fence cannot safely reference job.
+	 */
+	struct dma_fence *fence;
+	/** @start: Start address to invalidate */
+	u64 start;
+	/** @end: End address to invalidate */
+	u64 end;
+	/** @asid: Address space ID to invalidate */
+	u32 asid;
+	/** @fence_armed: Fence has been armed */
+	bool fence_armed;
+};
+
+static struct dma_fence *xe_gt_tlb_inval_job_run(struct xe_dep_job *dep_job)
+{
+	struct xe_gt_tlb_inval_job *job =
+		container_of(dep_job, typeof(*job), dep);
+	struct xe_gt_tlb_invalidation_fence *ifence =
+		container_of(job->fence, typeof(*ifence), base);
+
+	xe_gt_tlb_invalidation_range(job->gt, ifence, job->start,
+				     job->end, job->asid);
+
+	return job->fence;
+}
+
+static void xe_gt_tlb_inval_job_free(struct xe_dep_job *dep_job)
+{
+	struct xe_gt_tlb_inval_job *job =
+		container_of(dep_job, typeof(*job), dep);
+
+	/* Pairs with get in xe_gt_tlb_inval_job_push */
+	xe_gt_tlb_inval_job_put(job);
+}
+
+static const struct xe_dep_job_ops dep_job_ops = {
+	.run_job = xe_gt_tlb_inval_job_run,
+	.free_job = xe_gt_tlb_inval_job_free,
+};
+
+static int xe_gt_tlb_inval_context(struct xe_gt *gt)
+{
+	return xe_gt_is_media_type(gt) ? XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT :
+		XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT;
+}
+
+/**
+ * xe_gt_tlb_inval_job_create() - GT TLB invalidation job create
+ * @gt: GT to invalidate
+ * @q: exec queue issuing the invalidate
+ * @start: Start address to invalidate
+ * @end: End address to invalidate
+ * @asid: Address space ID to invalidate
+ *
+ * Create a GT TLB invalidation job and initialize internal fields. The caller is
+ * responsible for releasing the creation reference.
+ *
+ * Return: GT TLB invalidation job object on success, ERR_PTR failure
+ */
+struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
+						       struct xe_gt *gt,
+						       u64 start, u64 end,
+						       u32 asid)
+{
+	struct xe_gt_tlb_inval_job *job;
+	struct xe_dep_scheduler *dep_scheduler =
+		q->tlb_inval[xe_gt_tlb_inval_context(gt)].dep_scheduler;
+	struct drm_sched_entity *entity =
+		xe_dep_scheduler_entity(dep_scheduler);
+	struct xe_gt_tlb_invalidation_fence *ifence;
+	int err;
+
+	job = kmalloc(sizeof(*job), GFP_KERNEL);
+	if (!job)
+		return ERR_PTR(-ENOMEM);
+
+	job->q = q;
+	job->gt = gt;
+	job->start = start;
+	job->end = end;
+	job->asid = asid;
+	job->fence_armed = false;
+	job->dep.ops = &dep_job_ops;
+	kref_init(&job->refcount);
+	xe_exec_queue_get(q);	/* Pairs with put in xe_gt_tlb_inval_job_destroy */
+
+	ifence = kmalloc(sizeof(*ifence), GFP_KERNEL);
+	if (!ifence) {
+		err = -ENOMEM;
+		goto err_job;
+	}
+	job->fence = &ifence->base;
+
+	err = drm_sched_job_init(&job->dep.drm, entity, 1, NULL,
+				 q->xef ? q->xef->drm->client_id : 0);
+	if (err)
+		goto err_fence;
+
+	/* Pairs with put in xe_gt_tlb_inval_job_destroy */
+	xe_pm_runtime_get_noresume(gt_to_xe(job->gt));
+
+	return job;
+
+err_fence:
+	kfree(ifence);
+err_job:
+	xe_exec_queue_put(q);
+	kfree(job);
+
+	return ERR_PTR(err);
+}
+
+static void xe_gt_tlb_inval_job_destroy(struct kref *ref)
+{
+	struct xe_gt_tlb_inval_job *job = container_of(ref, typeof(*job),
+						       refcount);
+	struct xe_gt_tlb_invalidation_fence *ifence =
+		container_of(job->fence, typeof(*ifence), base);
+	struct xe_device *xe = gt_to_xe(job->gt);
+	struct xe_exec_queue *q = job->q;
+
+	if (!job->fence_armed)
+		kfree(ifence);
+	else
+		/* Ref from xe_gt_tlb_invalidation_fence_init */
+		dma_fence_put(job->fence);
+
+	drm_sched_job_cleanup(&job->dep.drm);
+	kfree(job);
+	xe_exec_queue_put(q);	/* Pairs with get from xe_gt_tlb_inval_job_create */
+	xe_pm_runtime_put(xe);	/* Pairs with get from xe_gt_tlb_inval_job_create */
+}
+
+/**
+ * xe_gt_tlb_inval_alloc_dep() - GT TLB invalidation job alloc dependency
+ * @job: GT TLB invalidation job to alloc dependency for
+ *
+ * Allocate storage for a dependency in the GT TLB invalidation fence. This
+ * function should be called at most once per job and must be paired with
+ * xe_gt_tlb_inval_job_push being called with a real fence.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int xe_gt_tlb_inval_job_alloc_dep(struct xe_gt_tlb_inval_job *job)
+{
+	xe_assert(gt_to_xe(job->gt), !xa_load(&job->dep.drm.dependencies, 0));
+	might_alloc(GFP_KERNEL);
+
+	return drm_sched_job_add_dependency(&job->dep.drm,
+					    dma_fence_get_stub());
+}
+
+/**
+ * xe_gt_tlb_inval_job_push() - GT TLB invalidation job push
+ * @job: GT TLB invalidation job to push
+ * @m: The migration object being used
+ * @fence: Dependency for GT TLB invalidation job
+ *
+ * Pushes a GT TLB invalidation job for execution, using @fence as a dependency.
+ * Storage for @fence must be preallocated with xe_gt_tlb_inval_job_alloc_dep
+ * prior to this call if @fence is not signaled. Takes a reference to the job’s
+ * finished fence, which the caller is responsible for releasing, and return it
+ * to the caller. This function is safe to be called in the path of reclaim.
+ *
+ * Return: Job's finished fence on success, cannot fail
+ */
+struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
+					   struct xe_migrate *m,
+					   struct dma_fence *fence)
+{
+	struct xe_gt_tlb_invalidation_fence *ifence =
+		container_of(job->fence, typeof(*ifence), base);
+
+	if (!dma_fence_is_signaled(fence)) {
+		void *ptr;
+
+		/*
+		 * Can be in path of reclaim, hence the preallocation of fence
+		 * storage in xe_gt_tlb_inval_job_alloc_dep. Verify caller did
+		 * this correctly.
+		 */
+		xe_assert(gt_to_xe(job->gt),
+			  xa_load(&job->dep.drm.dependencies, 0) ==
+			  dma_fence_get_stub());
+
+		dma_fence_get(fence);	/* ref released once dependency processed by scheduler */
+		ptr = xa_store(&job->dep.drm.dependencies, 0, fence,
+			       GFP_ATOMIC);
+		xe_assert(gt_to_xe(job->gt), !xa_is_err(ptr));
+	}
+
+	xe_gt_tlb_inval_job_get(job);	/* Pairs with put in free_job */
+	job->fence_armed = true;
+
+	/*
+	 * We need the migration lock to protect the job's seqno and the spsc
+	 * queue, only taken on migration queue, user queues protected dma-resv
+	 * VM lock.
+	 */
+	xe_migrate_job_lock(m, job->q);
+
+	/* Creation ref pairs with put in xe_gt_tlb_inval_job_destroy */
+	xe_gt_tlb_invalidation_fence_init(job->gt, ifence, false);
+	dma_fence_get(job->fence);	/* Pairs with put in DRM scheduler */
+
+	drm_sched_job_arm(&job->dep.drm);
+	/*
+	 * caller ref, get must be done before job push as it could immediately
+	 * signal and free.
+	 */
+	dma_fence_get(&job->dep.drm.s_fence->finished);
+	drm_sched_entity_push_job(&job->dep.drm);
+
+	xe_migrate_job_unlock(m, job->q);
+
+	/*
+	 * Not using job->fence, as it has its own dma-fence context, which does
+	 * not allow GT TLB invalidation fences on the same queue, GT tuple to
+	 * be squashed in dma-resv/DRM scheduler. Instead, we use the DRM scheduler
+	 * context and job's finished fence, which enables squashing.
+	 */
+	return &job->dep.drm.s_fence->finished;
+}
+
+/**
+ * xe_gt_tlb_inval_job_get() - Get a reference to GT TLB invalidation job
+ * @job: GT TLB invalidation job object
+ *
+ * Increment the GT TLB invalidation job's reference count
+ */
+void xe_gt_tlb_inval_job_get(struct xe_gt_tlb_inval_job *job)
+{
+	kref_get(&job->refcount);
+}
+
+/**
+ * xe_gt_tlb_inval_job_put() - Put a reference to GT TLB invalidation job
+ * @job: GT TLB invalidation job object
+ *
+ * Decrement the GT TLB invalidation job's reference count, call
+ * xe_gt_tlb_inval_job_destroy when reference count == 0. Skips decrement if
+ * input @job is NULL or IS_ERR.
+ */
+void xe_gt_tlb_inval_job_put(struct xe_gt_tlb_inval_job *job)
+{
+	if (!IS_ERR_OR_NULL(job))
+		kref_put(&job->refcount, xe_gt_tlb_inval_job_destroy);
+}
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h
new file mode 100644
index 000000000000..883896194a34
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h
@@ -0,0 +1,34 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_GT_TLB_INVAL_JOB_H_
+#define _XE_GT_TLB_INVAL_JOB_H_
+
+#include <linux/types.h>
+
+struct dma_fence;
+struct drm_sched_job;
+struct kref;
+struct xe_exec_queue;
+struct xe_gt;
+struct xe_gt_tlb_inval_job;
+struct xe_migrate;
+
+struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
+						       struct xe_gt *gt,
+						       u64 start, u64 end,
+						       u32 asid);
+
+int xe_gt_tlb_inval_job_alloc_dep(struct xe_gt_tlb_inval_job *job);
+
+struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
+					   struct xe_migrate *m,
+					   struct dma_fence *fence);
+
+void xe_gt_tlb_inval_job_get(struct xe_gt_tlb_inval_job *job);
+
+void xe_gt_tlb_inval_job_put(struct xe_gt_tlb_inval_job *job);
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index ed04c2b368dc..90065d7d29ff 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -2077,6 +2077,42 @@ out_err:
 	return IS_ERR(fence) ? PTR_ERR(fence) : 0;
 }
 
+/**
+ * xe_migrate_job_lock() - Lock migrate job lock
+ * @m: The migration context.
+ * @q: Queue associated with the operation which requires a lock
+ *
+ * Lock the migrate job lock if the queue is a migration queue, otherwise
+ * assert the VM's dma-resv is held (user queue's have own locking).
+ */
+void xe_migrate_job_lock(struct xe_migrate *m, struct xe_exec_queue *q)
+{
+	bool is_migrate = q == m->q;
+
+	if (is_migrate)
+		mutex_lock(&m->job_mutex);
+	else
+		xe_vm_assert_held(q->vm);	/* User queues VM's should be locked */
+}
+
+/**
+ * xe_migrate_job_unlock() - Unlock migrate job lock
+ * @m: The migration context.
+ * @q: Queue associated with the operation which requires a lock
+ *
+ * Unlock the migrate job lock if the queue is a migration queue, otherwise
+ * assert the VM's dma-resv is held (user queue's have own locking).
+ */
+void xe_migrate_job_unlock(struct xe_migrate *m, struct xe_exec_queue *q)
+{
+	bool is_migrate = q == m->q;
+
+	if (is_migrate)
+		mutex_unlock(&m->job_mutex);
+	else
+		xe_vm_assert_held(q->vm);	/* User queues VM's should be locked */
+}
+
 #if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST)
 #include "tests/xe_migrate.c"
 #endif
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index f4dac270dba3..5af2f5aa645f 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -141,4 +141,8 @@ xe_migrate_update_pgtables(struct xe_migrate *m,
 			   struct xe_migrate_pt_update *pt_update);
 
 void xe_migrate_wait(struct xe_migrate *m);
+
+void xe_migrate_job_lock(struct xe_migrate *m, struct xe_exec_queue *q);
+void xe_migrate_job_unlock(struct xe_migrate *m, struct xe_exec_queue *q);
+
 #endif

From b8d5779eee382003b26b1a147071adf08c5d294c Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:15 -0700
Subject: [PATCH 0093/1455] drm/xe: Use GT TLB invalidation jobs in PT layer
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rather than open-coding GT TLB invalidations in the PT layer, use GT TLB
invalidation jobs. The real benefit is that GT TLB invalidation jobs use
a single dma-fence context, allowing the generated fences to be squashed
in dma-resv/DRM scheduler.

v2:
 - s/;;/; (checkpatch)
 - Move ijob/mjob job push after range fence install
v3:
 - Remove extra newline (Stuart)
 - Set ijob/mjob near creation (Stuart)
 - Add comment back in (Stuart)

Suggested-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-7-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.h |   9 ++
 drivers/gpu/drm/xe/xe_pt.c      | 178 +++++++++++++-------------------
 2 files changed, 80 insertions(+), 107 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 5af2f5aa645f..3758f9615484 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -14,6 +14,7 @@ struct ttm_resource;
 
 struct xe_bo;
 struct xe_gt;
+struct xe_gt_tlb_inval_job;
 struct xe_exec_queue;
 struct xe_migrate;
 struct xe_migrate_pt_update;
@@ -91,6 +92,14 @@ struct xe_migrate_pt_update {
 	struct xe_vma_ops *vops;
 	/** @job: The job if a GPU page-table update. NULL otherwise */
 	struct xe_sched_job *job;
+	/**
+	 * @ijob: The GT TLB invalidation job for primary tile. NULL otherwise
+	 */
+	struct xe_gt_tlb_inval_job *ijob;
+	/**
+	 * @mjob: The GT TLB invalidation job for media tile. NULL otherwise
+	 */
+	struct xe_gt_tlb_inval_job *mjob;
 	/** @tile_id: Tile ID of the update */
 	u8 tile_id;
 };
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index c8e63bd23300..330cc0f54a3f 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -13,7 +13,7 @@
 #include "xe_drm_client.h"
 #include "xe_exec_queue.h"
 #include "xe_gt.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval_job.h"
 #include "xe_migrate.h"
 #include "xe_pt_types.h"
 #include "xe_pt_walk.h"
@@ -1261,6 +1261,8 @@ static int op_add_deps(struct xe_vm *vm, struct xe_vma_op *op,
 }
 
 static int xe_pt_vm_dependencies(struct xe_sched_job *job,
+				 struct xe_gt_tlb_inval_job *ijob,
+				 struct xe_gt_tlb_inval_job *mjob,
 				 struct xe_vm *vm,
 				 struct xe_vma_ops *vops,
 				 struct xe_vm_pgtable_update_ops *pt_update_ops,
@@ -1328,6 +1330,20 @@ static int xe_pt_vm_dependencies(struct xe_sched_job *job,
 	for (i = 0; job && !err && i < vops->num_syncs; i++)
 		err = xe_sync_entry_add_deps(&vops->syncs[i], job);
 
+	if (job) {
+		if (ijob) {
+			err = xe_gt_tlb_inval_job_alloc_dep(ijob);
+			if (err)
+				return err;
+		}
+
+		if (mjob) {
+			err = xe_gt_tlb_inval_job_alloc_dep(mjob);
+			if (err)
+				return err;
+		}
+	}
+
 	return err;
 }
 
@@ -1339,7 +1355,8 @@ static int xe_pt_pre_commit(struct xe_migrate_pt_update *pt_update)
 	struct xe_vm_pgtable_update_ops *pt_update_ops =
 		&vops->pt_update_ops[pt_update->tile_id];
 
-	return xe_pt_vm_dependencies(pt_update->job, vm, pt_update->vops,
+	return xe_pt_vm_dependencies(pt_update->job, pt_update->ijob,
+				     pt_update->mjob, vm, pt_update->vops,
 				     pt_update_ops, rftree);
 }
 
@@ -1509,75 +1526,6 @@ static int xe_pt_svm_pre_commit(struct xe_migrate_pt_update *pt_update)
 }
 #endif
 
-struct invalidation_fence {
-	struct xe_gt_tlb_invalidation_fence base;
-	struct xe_gt *gt;
-	struct dma_fence *fence;
-	struct dma_fence_cb cb;
-	struct work_struct work;
-	u64 start;
-	u64 end;
-	u32 asid;
-};
-
-static void invalidation_fence_cb(struct dma_fence *fence,
-				  struct dma_fence_cb *cb)
-{
-	struct invalidation_fence *ifence =
-		container_of(cb, struct invalidation_fence, cb);
-	struct xe_device *xe = gt_to_xe(ifence->gt);
-
-	trace_xe_gt_tlb_invalidation_fence_cb(xe, &ifence->base);
-	if (!ifence->fence->error) {
-		queue_work(system_wq, &ifence->work);
-	} else {
-		ifence->base.base.error = ifence->fence->error;
-		xe_gt_tlb_invalidation_fence_signal(&ifence->base);
-	}
-	dma_fence_put(ifence->fence);
-}
-
-static void invalidation_fence_work_func(struct work_struct *w)
-{
-	struct invalidation_fence *ifence =
-		container_of(w, struct invalidation_fence, work);
-	struct xe_device *xe = gt_to_xe(ifence->gt);
-
-	trace_xe_gt_tlb_invalidation_fence_work_func(xe, &ifence->base);
-	xe_gt_tlb_invalidation_range(ifence->gt, &ifence->base, ifence->start,
-				     ifence->end, ifence->asid);
-}
-
-static void invalidation_fence_init(struct xe_gt *gt,
-				    struct invalidation_fence *ifence,
-				    struct dma_fence *fence,
-				    u64 start, u64 end, u32 asid)
-{
-	int ret;
-
-	trace_xe_gt_tlb_invalidation_fence_create(gt_to_xe(gt), &ifence->base);
-
-	xe_gt_tlb_invalidation_fence_init(gt, &ifence->base, false);
-
-	ifence->fence = fence;
-	ifence->gt = gt;
-	ifence->start = start;
-	ifence->end = end;
-	ifence->asid = asid;
-
-	INIT_WORK(&ifence->work, invalidation_fence_work_func);
-	ret = dma_fence_add_callback(fence, &ifence->cb, invalidation_fence_cb);
-	if (ret == -ENOENT) {
-		dma_fence_put(ifence->fence);	/* Usually dropped in CB */
-		invalidation_fence_work_func(&ifence->work);
-	} else if (ret) {
-		dma_fence_put(&ifence->base.base);	/* Caller ref */
-		dma_fence_put(&ifence->base.base);	/* Creation ref */
-	}
-
-	xe_gt_assert(gt, !ret || ret == -ENOENT);
-}
-
 struct xe_pt_stage_unbind_walk {
 	/** @base: The pagewalk base-class. */
 	struct xe_pt_walk base;
@@ -2407,8 +2355,8 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	struct xe_vm *vm = vops->vm;
 	struct xe_vm_pgtable_update_ops *pt_update_ops =
 		&vops->pt_update_ops[tile->id];
-	struct dma_fence *fence;
-	struct invalidation_fence *ifence = NULL, *mfence = NULL;
+	struct dma_fence *fence, *ifence, *mfence;
+	struct xe_gt_tlb_inval_job *ijob = NULL, *mjob = NULL;
 	struct dma_fence **fences = NULL;
 	struct dma_fence_array *cf = NULL;
 	struct xe_range_fence *rfence;
@@ -2440,26 +2388,38 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 #endif
 
 	if (pt_update_ops->needs_invalidation) {
-		ifence = kzalloc(sizeof(*ifence), GFP_KERNEL);
-		if (!ifence) {
-			err = -ENOMEM;
+		ijob = xe_gt_tlb_inval_job_create(pt_update_ops->q,
+						  tile->primary_gt,
+						  pt_update_ops->start,
+						  pt_update_ops->last,
+						  vm->usm.asid);
+		if (IS_ERR(ijob)) {
+			err = PTR_ERR(ijob);
 			goto kill_vm_tile1;
 		}
+		update.ijob = ijob;
+
 		if (tile->media_gt) {
-			mfence = kzalloc(sizeof(*ifence), GFP_KERNEL);
-			if (!mfence) {
-				err = -ENOMEM;
-				goto free_ifence;
+			mjob = xe_gt_tlb_inval_job_create(pt_update_ops->q,
+							  tile->media_gt,
+							  pt_update_ops->start,
+							  pt_update_ops->last,
+							  vm->usm.asid);
+			if (IS_ERR(mjob)) {
+				err = PTR_ERR(mjob);
+				goto free_ijob;
 			}
+			update.mjob = mjob;
+
 			fences = kmalloc_array(2, sizeof(*fences), GFP_KERNEL);
 			if (!fences) {
 				err = -ENOMEM;
-				goto free_ifence;
+				goto free_ijob;
 			}
 			cf = dma_fence_array_alloc(2);
 			if (!cf) {
 				err = -ENOMEM;
-				goto free_ifence;
+				goto free_ijob;
 			}
 		}
 	}
@@ -2467,7 +2427,7 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	rfence = kzalloc(sizeof(*rfence), GFP_KERNEL);
 	if (!rfence) {
 		err = -ENOMEM;
-		goto free_ifence;
+		goto free_ijob;
 	}
 
 	fence = xe_migrate_update_pgtables(tile->migrate, &update);
@@ -2491,30 +2451,31 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 				  pt_update_ops->last, fence))
 		dma_fence_wait(fence, false);
 
-	/* tlb invalidation must be done before signaling rebind */
-	if (ifence) {
-		if (mfence)
-			dma_fence_get(fence);
-		invalidation_fence_init(tile->primary_gt, ifence, fence,
-					pt_update_ops->start,
-					pt_update_ops->last, vm->usm.asid);
-		if (mfence) {
-			invalidation_fence_init(tile->media_gt, mfence, fence,
-						pt_update_ops->start,
-						pt_update_ops->last, vm->usm.asid);
-			fences[0] = &ifence->base.base;
-			fences[1] = &mfence->base.base;
+	/* tlb invalidation must be done before signaling unbind/rebind */
+	if (ijob) {
+		struct dma_fence *__fence;
+
+		ifence = xe_gt_tlb_inval_job_push(ijob, tile->migrate, fence);
+		__fence = ifence;
+
+		if (mjob) {
+			fences[0] = ifence;
+			mfence = xe_gt_tlb_inval_job_push(mjob, tile->migrate,
+							  fence);
+			fences[1] = mfence;
+
 			dma_fence_array_init(cf, 2, fences,
 					     vm->composite_fence_ctx,
 					     vm->composite_fence_seqno++,
 					     false);
-			fence = &cf->base;
-		} else {
-			fence = &ifence->base.base;
+			__fence = &cf->base;
 		}
+
+		dma_fence_put(fence);
+		fence = __fence;
 	}
 
-	if (!mfence) {
+	if (!mjob) {
 		dma_resv_add_fence(xe_vm_resv(vm), fence,
 				   pt_update_ops->wait_vm_bookkeep ?
 				   DMA_RESV_USAGE_KERNEL :
@@ -2523,19 +2484,19 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 		list_for_each_entry(op, &vops->list, link)
 			op_commit(vops->vm, tile, pt_update_ops, op, fence, NULL);
 	} else {
-		dma_resv_add_fence(xe_vm_resv(vm), &ifence->base.base,
+		dma_resv_add_fence(xe_vm_resv(vm), ifence,
 				   pt_update_ops->wait_vm_bookkeep ?
 				   DMA_RESV_USAGE_KERNEL :
 				   DMA_RESV_USAGE_BOOKKEEP);
 
-		dma_resv_add_fence(xe_vm_resv(vm), &mfence->base.base,
+		dma_resv_add_fence(xe_vm_resv(vm), mfence,
 				   pt_update_ops->wait_vm_bookkeep ?
 				   DMA_RESV_USAGE_KERNEL :
 				   DMA_RESV_USAGE_BOOKKEEP);
 
 		list_for_each_entry(op, &vops->list, link)
-			op_commit(vops->vm, tile, pt_update_ops, op,
-				  &ifence->base.base, &mfence->base.base);
+			op_commit(vops->vm, tile, pt_update_ops, op, ifence,
+				  mfence);
 	}
 
 	if (pt_update_ops->needs_svm_lock)
@@ -2543,15 +2504,18 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	if (pt_update_ops->needs_userptr_lock)
 		up_read(&vm->userptr.notifier_lock);
 
+	xe_gt_tlb_inval_job_put(mjob);
+	xe_gt_tlb_inval_job_put(ijob);
+
 	return fence;
 
 free_rfence:
 	kfree(rfence);
-free_ifence:
+free_ijob:
 	kfree(cf);
 	kfree(fences);
-	kfree(mfence);
-	kfree(ifence);
+	xe_gt_tlb_inval_job_put(mjob);
+	xe_gt_tlb_inval_job_put(ijob);
 kill_vm_tile1:
 	if (err != -EAGAIN && err != -ENODATA && tile->id)
 		xe_vm_kill(vops->vm, false);

From 51330ba66caa29c1a73408fada7f824b36b3238f Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Thu, 24 Jul 2025 12:12:16 -0700
Subject: [PATCH 0094/1455] drm/xe: Remove unused GT TLB invalidation trace
 points

Remove unused GT TLB invalidation trace points after converting to used
GT TLB invalidation jobs. The trace points removed were used during
early bring up of unstable driver, with a stable driver no need to
replace with new tracepoints.

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250724191216.4076566-8-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_trace.h | 16 ----------------
 1 file changed, 16 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_trace.h b/drivers/gpu/drm/xe/xe_trace.h
index b4a3577df70c..21486a6f693a 100644
--- a/drivers/gpu/drm/xe/xe_trace.h
+++ b/drivers/gpu/drm/xe/xe_trace.h
@@ -45,22 +45,6 @@ DECLARE_EVENT_CLASS(xe_gt_tlb_invalidation_fence,
 			      __get_str(dev), __entry->fence, __entry->seqno)
 );
 
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_create,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
-	     TP_ARGS(xe, fence)
-);
-
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence,
-	     xe_gt_tlb_invalidation_fence_work_func,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
-	     TP_ARGS(xe, fence)
-);
-
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_cb,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
-	     TP_ARGS(xe, fence)
-);
-
 DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_send,
 	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
 	     TP_ARGS(xe, fence)

From f98de826b418885a21ece67f0f5b921ae759b7bf Mon Sep 17 00:00:00 2001
From: Shuicheng Lin <shuicheng.lin@intel.com>
Date: Thu, 24 Jul 2025 19:38:55 +0000
Subject: [PATCH 0095/1455] drm/xe/hw_engine_group: Avoid call kfree() for
 drmm_kzalloc()

Memory allocated with drmm_kzalloc() should not be freed using
kfree(), as it is managed by the DRM subsystem. The memory will
be automatically freed when the associated drm_device is released.
These 3 group pointers are allocated using drmm_kzalloc() in
hw_engine_group_alloc(), so they don't require manual deallocation.

Fixes: 67979060740f ("drm/xe/hw_engine_group: Fix potential leak")
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250724193854.1124510-2-shuicheng.lin@intel.com
---
 drivers/gpu/drm/xe/xe_hw_engine_group.c | 28 ++++++-------------------
 1 file changed, 6 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_hw_engine_group.c b/drivers/gpu/drm/xe/xe_hw_engine_group.c
index 87a6dcb1b4b5..c926f840c87b 100644
--- a/drivers/gpu/drm/xe/xe_hw_engine_group.c
+++ b/drivers/gpu/drm/xe/xe_hw_engine_group.c
@@ -75,25 +75,18 @@ int xe_hw_engine_setup_groups(struct xe_gt *gt)
 	enum xe_hw_engine_id id;
 	struct xe_hw_engine_group *group_rcs_ccs, *group_bcs, *group_vcs_vecs;
 	struct xe_device *xe = gt_to_xe(gt);
-	int err;
 
 	group_rcs_ccs = hw_engine_group_alloc(xe);
-	if (IS_ERR(group_rcs_ccs)) {
-		err = PTR_ERR(group_rcs_ccs);
-		goto err_group_rcs_ccs;
-	}
+	if (IS_ERR(group_rcs_ccs))
+		return PTR_ERR(group_rcs_ccs);
 
 	group_bcs = hw_engine_group_alloc(xe);
-	if (IS_ERR(group_bcs)) {
-		err = PTR_ERR(group_bcs);
-		goto err_group_bcs;
-	}
+	if (IS_ERR(group_bcs))
+		return PTR_ERR(group_bcs);
 
 	group_vcs_vecs = hw_engine_group_alloc(xe);
-	if (IS_ERR(group_vcs_vecs)) {
-		err = PTR_ERR(group_vcs_vecs);
-		goto err_group_vcs_vecs;
-	}
+	if (IS_ERR(group_vcs_vecs))
+		return PTR_ERR(group_vcs_vecs);
 
 	for_each_hw_engine(hwe, gt, id) {
 		switch (hwe->class) {
@@ -116,15 +109,6 @@ int xe_hw_engine_setup_groups(struct xe_gt *gt)
 	}
 
 	return 0;
-
-err_group_vcs_vecs:
-	kfree(group_vcs_vecs);
-err_group_bcs:
-	kfree(group_bcs);
-err_group_rcs_ccs:
-	kfree(group_rcs_ccs);
-
-	return err;
 }
 
 /**

From 16d2a92e7ec8c7828069f86a0aa7974acafb7e57 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Thu, 24 Jul 2025 12:02:37 +0300
Subject: [PATCH 0096/1455] drm/i915: Fix selecting CONFIG_DRM_KUNIT_TEST in
 debug builds
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Selecting an option which depends on other options only works if the
dependencies are guaranteed to be selected (as these dependencies will
not be automatically selected). CONFIG_DRM_KUNIT_TEST depends on DRM,
MMU and KUNIT the first two of which are guaranteed to be selected for
i915, but the last one is not. Hence, selecting CONFIG_DRM_KUNIT_TEST in
i915 debug builds may result in CONFIG_DRM_KUNIT_TEST being selected
without the CONFIG_KUNIT dependency being selected. This causes at least
the following compile error:

drivers/gpu/drm/tests/drm_bridge_test.c: In function ‘drm_test_bridge_alloc_init’:
drivers/gpu/drm/tests/drm_bridge_test.c:449:21: error: implicit declaration of function ‘kunit_device_register’; did you mean ‘root_device_register’? [-Werror=implicit-function-declaration]
  449 |         priv->dev = kunit_device_register(test, "drm-bridge-dev");

Fix the above by selecting CONFIG_DRM_KUNIT_TEST only if CONFIG_KUNIT is
also selected.

Fixes: 17133255a322 ("drm/i915: replace DRM_DEBUG_SELFTEST with DRM_KUNIT_TEST")
Cc: Ruben Wauters <rubenru09@aol.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250724090237.92040-1-imre.deak@intel.com
---
 drivers/gpu/drm/i915/Kconfig.debug | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/Kconfig.debug b/drivers/gpu/drm/i915/Kconfig.debug
index b15b1cecb3aa..3562a02ef7ad 100644
--- a/drivers/gpu/drm/i915/Kconfig.debug
+++ b/drivers/gpu/drm/i915/Kconfig.debug
@@ -50,7 +50,7 @@ config DRM_I915_DEBUG
 	select DRM_VGEM # used by igt/prime_vgem (dmabuf interop checks)
 	select DRM_DEBUG_MM if DRM=y
 	select DRM_EXPORT_FOR_TESTS if m
-	select DRM_KUNIT_TEST
+	select DRM_KUNIT_TEST if KUNIT
 	select DMABUF_SELFTESTS
 	select SW_SYNC # signaling validation framework (igt/syncobj*)
 	select DRM_I915_WERROR

From 0bdd05c2a82bbf2419415d012fd4f5faeca7f1af Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 22 Jul 2025 16:10:54 +0200
Subject: [PATCH 0097/1455] drm/xe/configfs: Fix pci_dev reference leak

We are using pci_get_domain_bus_and_slot() function to verify if
the given config directory name matches any existing PCI device,
but we missed to call matching pci_dev_put() to release reference.

While around, also change error code in case of no device match,
to make it more specific than generic formatting error.

Fixes: 16280ded45fb ("drm/xe: Add configfs to enable survivability mode")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250722141059.30707-2-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 8ec1ff1e4e80..e9b46a2d0019 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -267,7 +267,8 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 
 	pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function));
 	if (!pdev)
-		return ERR_PTR(-EINVAL);
+		return ERR_PTR(-ENODEV);
+	pci_dev_put(pdev);
 
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
 	if (!dev)

From 400a6da1e967c4f117e4757412df06dcfaea0e6a Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 22 Jul 2025 16:10:55 +0200
Subject: [PATCH 0098/1455] drm/xe/configfs: Enforce canonical device names
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

While we expect config directory names to match PCI device name,
currently we are only scanning provided names for domain, bus,
device and function numbers, without checking their format.
This would pass slightly broken entries like:

  /sys/kernel/config/xe/
  ├── 0000:00:02.0000000000000
  │   └── ...
  ├── 0000:00:02.0x
  │   └── ...
  ├──  0: 0: 2. 0
  │   └── ...
  └── 0:0:2.0
      └── ...

To avoid such mistakes, check if the name provided exactly matches
the canonical PCI device address format, which we recreated from
the parsed BDF data. Also simplify scanf format as it can't really
catch all formatting errors.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250722141059.30707-3-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index e9b46a2d0019..f5ce48a0adc8 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -259,12 +259,19 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 	unsigned int domain, bus, slot, function;
 	struct xe_config_device *dev;
 	struct pci_dev *pdev;
+	char canonical[16];
 	int ret;
 
-	ret = sscanf(name, "%04x:%02x:%02x.%x", &domain, &bus, &slot, &function);
+	ret = sscanf(name, "%x:%x:%x.%x", &domain, &bus, &slot, &function);
 	if (ret != 4)
 		return ERR_PTR(-EINVAL);
 
+	ret = scnprintf(canonical, sizeof(canonical), "%04x:%02x:%02x.%d", domain, bus,
+			PCI_SLOT(PCI_DEVFN(slot, function)),
+			PCI_FUNC(PCI_DEVFN(slot, function)));
+	if (ret != 12 || strcmp(name, canonical))
+		return ERR_PTR(-EINVAL);
+
 	pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function));
 	if (!pdev)
 		return ERR_PTR(-ENODEV);

From 9b807f0bb0d0f4b029e56dd2daa4a5e6d87fab61 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 22 Jul 2025 16:10:56 +0200
Subject: [PATCH 0099/1455] drm/xe/configfs: Use pci_name() for lookup

There is no need to manually build PCI device name from BDF data,
since it was already prepared and assigned and can be accessed by
calling pci_name() function.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250722141059.30707-4-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index f5ce48a0adc8..36e2b45b305f 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -312,13 +312,9 @@ static struct configfs_subsystem xe_configfs = {
 static struct xe_config_device *configfs_find_group(struct pci_dev *pdev)
 {
 	struct config_item *item;
-	char name[64];
-
-	snprintf(name, sizeof(name), "%04x:%02x:%02x.%x", pci_domain_nr(pdev->bus),
-		 pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn));
 
 	mutex_lock(&xe_configfs.su_mutex);
-	item = config_group_find_item(&xe_configfs.su_group, name);
+	item = config_group_find_item(&xe_configfs.su_group, pci_name(pdev));
 	mutex_unlock(&xe_configfs.su_mutex);
 
 	if (!item)

From c143db7dba1d3a4edce36c89c50e54dd856d7c41 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Thu, 24 Jul 2025 21:29:00 +0300
Subject: [PATCH 0100/1455] drm/i915/dp: Fix disabling training pattern at end
 of UHBR link training
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The Fixed: commit below overlooked the fact that
intel_dp_link_train_all_phys() is only used for non-UHBR link rates, but
intel_dp_stop_link_train() is used for both non-UHBR and UHBR link
rates. Hence, after removing the disabling of the training pattern from
intel_dp_stop_link_train(), the commit missed adding this back to the
end of UHBR link training in intel_dp_128b132b_link_train(). This left
the sink in link training mode at the end of an UHBR rate link training.

Fix things by disabling the training pattern at the end of UHBR link
training as well.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Fixes: 11fab5a2a1ad ("drm/i915/dp: Clear DPCD training pattern before transmitting the idle pattern")
Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250724182900.160891-1-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_link_training.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 344b74109a83..9d14b4e59545 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -1609,6 +1609,8 @@ out:
 		intel_dp_program_link_training_pattern(intel_dp, crtc_state,
 						       DP_PHY_DPRX, DP_TRAINING_PATTERN_2);
 
+	intel_dp_disable_dpcd_training_pattern(intel_dp, DP_PHY_DPRX);
+
 	return passed;
 }
 

From 4af5de27d3a9ba19b7f61ac2fa0d6d35ee876c95 Mon Sep 17 00:00:00 2001
From: Gustavo Sousa <gustavo.sousa@intel.com>
Date: Thu, 17 Jul 2025 17:59:15 -0300
Subject: [PATCH 0101/1455] drm/i915/display: Remove unused declarations of
 intel_io_*
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Declarations for both intel_io_mmio_fw_write and intel_io_reg_write
were added with commit 01389846f7d6 ("drm/i915: Plumb 'dsb' all way to
the plane hooks"), but they never got used. Let's remove them.

Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250717-drop-unused-intel_io-declarations-v1-1-bdea2c749571@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_types.h | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 8f8019d40d77..4d9df803ad47 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -943,10 +943,6 @@ struct intel_csc_matrix {
 	u16 postoff[3];
 };
 
-void intel_io_mmio_fw_write(void *ctx, i915_reg_t reg, u32 val);
-
-typedef void (*intel_io_reg_write)(void *ctx, i915_reg_t reg, u32 val);
-
 struct intel_crtc_state {
 	/*
 	 * uapi (drm) state. This is the software state shown to userspace.

From c79291f7332465e0651afd07020ebc19eda78ed4 Mon Sep 17 00:00:00 2001
From: Carl Vanderlip <carl.vanderlip@oss.qualcomm.com>
Date: Mon, 21 Jul 2025 15:08:10 -0700
Subject: [PATCH 0102/1455] MAINTAINERS: Update email address for Carl
 Vanderlip

Qualcomm is changing open source email address policy.
LKML and other busy mailing lists use the oss.qualcomm.com domain.

Signed-off-by: Carl Vanderlip <carl.vanderlip@oss.qualcomm.com>
Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721220811.962509-1-carl.vanderlip@oss.qualcomm.com
---
 .mailmap    | 2 ++
 MAINTAINERS | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/.mailmap b/.mailmap
index b77cd34cf852..d30f31ea01a5 100644
--- a/.mailmap
+++ b/.mailmap
@@ -163,6 +163,8 @@ Casey Connolly <casey.connolly@linaro.org> <caleb@connolly.tech>
 Casey Connolly <casey.connolly@linaro.org> <caleb@postmarketos.org>
 Can Guo <quic_cang@quicinc.com> <cang@codeaurora.org>
 Carl Huang <quic_cjhuang@quicinc.com> <cjhuang@codeaurora.org>
+Carl Vanderlip <carl.vanderlip@oss.qualcomm.com> <carlv@codeaurora.org>
+Carl Vanderlip <carl.vanderlip@oss.qualcomm.com> <quic_carlv@quicinc.com>
 Carlos Bilbao <carlos.bilbao@kernel.org> <carlos.bilbao@amd.com>
 Carlos Bilbao <carlos.bilbao@kernel.org> <carlos.bilbao.osdev@gmail.com>
 Carlos Bilbao <carlos.bilbao@kernel.org> <bilbao@vt.edu>
diff --git a/MAINTAINERS b/MAINTAINERS
index 5ddf37f0acc9..62fa528c0884 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -20401,7 +20401,7 @@ F:	include/dt-bindings/clock/qcom,*
 
 QUALCOMM CLOUD AI (QAIC) DRIVER
 M:	Jeff Hugo <jeff.hugo@oss.qualcomm.com>
-R:	Carl Vanderlip <quic_carlv@quicinc.com>
+R:	Carl Vanderlip <carl.vanderlip@oss.qualcomm.com>
 L:	linux-arm-msm@vger.kernel.org
 L:	dri-devel@lists.freedesktop.org
 S:	Supported

From e8372edec948318934c71542e939d1a8fb8fabcf Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Fri, 11 Jul 2025 17:01:52 +0100
Subject: [PATCH 0103/1455] drm/xe/xelp: Implement Wa_16010904313

Add XeLP workaround 16010904313.

The description calls for it to be emitted as the indirect context buffer
workaround for render and compute, and from the workaround batch buffer
for the other engines. Therefore we plug into the previously added
respective top level emission functions.

The actual command streamer programming sequence differs from what is
described in the PRM, in that it assumes the listed LRCA offset was
supposed to actually refer to the location of the CTX_TIMESTAMP register
instead of LRCA + 0x180c (which is in GPR space). Latter appears to make
more sense under the assumption that multiple writes are helping with
restoring the CTX_TIMESTAMP register content from the saved context state.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20250711160153.49833-8-tvrtko.ursulin@igalia.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 .../gpu/drm/xe/instructions/xe_mi_commands.h  |  1 +
 drivers/gpu/drm/xe/xe_lrc.c                   | 44 +++++++++++++++++++
 drivers/gpu/drm/xe/xe_wa_oob.rules            |  1 +
 3 files changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
index e3f5e8bb3ebc..c47b290e0e9f 100644
--- a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
+++ b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
@@ -65,6 +65,7 @@
 
 #define MI_LOAD_REGISTER_MEM		(__MI_INSTR(0x29) | XE_INSTR_NUM_DW(4))
 #define   MI_LRM_USE_GGTT		REG_BIT(22)
+#define   MI_LRM_ASYNC			REG_BIT(21)
 
 #define MI_LOAD_REGISTER_REG		(__MI_INSTR(0x2a) | XE_INSTR_NUM_DW(3))
 #define   MI_LRR_DST_CS_MMIO		REG_BIT(19)
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 6d38411bdeba..110731a1e547 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -76,6 +76,11 @@ lrc_to_xe(struct xe_lrc *lrc)
 static bool
 gt_engine_needs_indirect_ctx(struct xe_gt *gt, enum xe_engine_class class)
 {
+	if (XE_WA(gt, 16010904313) &&
+	    (class == XE_ENGINE_CLASS_RENDER ||
+	     class == XE_ENGINE_CLASS_COMPUTE))
+		return true;
+
 	return false;
 }
 
@@ -1014,6 +1019,43 @@ static ssize_t setup_utilization_wa(struct xe_lrc *lrc,
 	return cmd - batch;
 }
 
+static ssize_t setup_timestamp_wa(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
+				  u32 *batch, size_t max_len)
+{
+	const u32 ts_addr = __xe_lrc_ctx_timestamp_ggtt_addr(lrc);
+	u32 *cmd = batch;
+
+	if (!XE_WA(lrc->gt, 16010904313) ||
+	    !(hwe->class == XE_ENGINE_CLASS_RENDER ||
+	      hwe->class == XE_ENGINE_CLASS_COMPUTE ||
+	      hwe->class == XE_ENGINE_CLASS_COPY ||
+	      hwe->class == XE_ENGINE_CLASS_VIDEO_DECODE ||
+	      hwe->class == XE_ENGINE_CLASS_VIDEO_ENHANCE))
+		return 0;
+
+	if (xe_gt_WARN_ON(lrc->gt, max_len < 12))
+		return -ENOSPC;
+
+	*cmd++ = MI_LOAD_REGISTER_MEM | MI_LRM_USE_GGTT | MI_LRI_LRM_CS_MMIO |
+		 MI_LRM_ASYNC;
+	*cmd++ = RING_CTX_TIMESTAMP(0).addr;
+	*cmd++ = ts_addr;
+	*cmd++ = 0;
+
+	*cmd++ = MI_LOAD_REGISTER_MEM | MI_LRM_USE_GGTT | MI_LRI_LRM_CS_MMIO |
+		 MI_LRM_ASYNC;
+	*cmd++ = RING_CTX_TIMESTAMP(0).addr;
+	*cmd++ = ts_addr;
+	*cmd++ = 0;
+
+	*cmd++ = MI_LOAD_REGISTER_MEM | MI_LRM_USE_GGTT | MI_LRI_LRM_CS_MMIO;
+	*cmd++ = RING_CTX_TIMESTAMP(0).addr;
+	*cmd++ = ts_addr;
+	*cmd++ = 0;
+
+	return cmd - batch;
+}
+
 struct bo_setup {
 	ssize_t (*setup)(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 			 u32 *batch, size_t max_size);
@@ -1089,6 +1131,7 @@ static void finish_bo(struct bo_setup_state *state)
 static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 {
 	static const struct bo_setup funcs[] = {
+		{ .setup = setup_timestamp_wa },
 		{ .setup = setup_utilization_wa },
 	};
 	struct bo_setup_state state = {
@@ -1121,6 +1164,7 @@ static int
 setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 {
 	static struct bo_setup rcs_funcs[] = {
+		{ .setup = setup_timestamp_wa },
 	};
 	struct bo_setup_state state = {
 		.lrc = lrc,
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 46ad72bb1f8d..e74256944208 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -1,4 +1,5 @@
 1607983814	GRAPHICS_VERSION_RANGE(1200, 1210)
+16010904313	GRAPHICS_VERSION_RANGE(1200, 1210)
 22012773006	GRAPHICS_VERSION_RANGE(1200, 1250)
 14014475959	GRAPHICS_VERSION_RANGE(1270, 1271), GRAPHICS_STEP(A0, B0)
 		PLATFORM(DG2)

From ca33cd271ef9b27acddf96742eef136d0f4caaa5 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Fri, 11 Jul 2025 17:01:53 +0100
Subject: [PATCH 0104/1455] drm/xe/xelp: Add Wa_18022495364

Add Wa_18022495364 as a context workaround batch buffer workaround.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20250711160153.49833-9-tvrtko.ursulin@igalia.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_engine_regs.h |  3 +++
 drivers/gpu/drm/xe/xe_lrc.c              | 21 +++++++++++++++++++++
 drivers/gpu/drm/xe/xe_wa_oob.rules       |  1 +
 3 files changed, 25 insertions(+)

diff --git a/drivers/gpu/drm/xe/regs/xe_engine_regs.h b/drivers/gpu/drm/xe/regs/xe_engine_regs.h
index 7ade41e2b7b3..f4c3e1187a00 100644
--- a/drivers/gpu/drm/xe/regs/xe_engine_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_engine_regs.h
@@ -111,6 +111,9 @@
 #define   PPHWSP_CSB_AND_TIMESTAMP_REPORT_DIS	REG_BIT(14)
 #define   CS_PRIORITY_MEM_READ			REG_BIT(7)
 
+#define CS_DEBUG_MODE2(base)			XE_REG((base) + 0xd8, XE_REG_OPTION_MASKED)
+#define   INSTRUCTION_STATE_CACHE_INVALIDATE	REG_BIT(6)
+
 #define FF_SLICE_CS_CHICKEN1(base)		XE_REG((base) + 0xe0, XE_REG_OPTION_MASKED)
 #define   FFSC_PERCTX_PREEMPT_CTRL		REG_BIT(14)
 
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 110731a1e547..540f044bf425 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -1056,6 +1056,26 @@ static ssize_t setup_timestamp_wa(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 	return cmd - batch;
 }
 
+static ssize_t setup_invalidate_state_cache_wa(struct xe_lrc *lrc,
+					       struct xe_hw_engine *hwe,
+					       u32 *batch, size_t max_len)
+{
+	u32 *cmd = batch;
+
+	if (!XE_WA(lrc->gt, 18022495364) ||
+	    hwe->class != XE_ENGINE_CLASS_RENDER)
+		return 0;
+
+	if (xe_gt_WARN_ON(lrc->gt, max_len < 3))
+		return -ENOSPC;
+
+	*cmd++ = MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(1);
+	*cmd++ = CS_DEBUG_MODE1(0).addr;
+	*cmd++ = _MASKED_BIT_ENABLE(INSTRUCTION_STATE_CACHE_INVALIDATE);
+
+	return cmd - batch;
+}
+
 struct bo_setup {
 	ssize_t (*setup)(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 			 u32 *batch, size_t max_size);
@@ -1132,6 +1152,7 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 {
 	static const struct bo_setup funcs[] = {
 		{ .setup = setup_timestamp_wa },
+		{ .setup = setup_invalidate_state_cache_wa },
 		{ .setup = setup_utilization_wa },
 	};
 	struct bo_setup_state state = {
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index e74256944208..8d0aabab6777 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -1,5 +1,6 @@
 1607983814	GRAPHICS_VERSION_RANGE(1200, 1210)
 16010904313	GRAPHICS_VERSION_RANGE(1200, 1210)
+18022495364	GRAPHICS_VERSION_RANGE(1200, 1210)
 22012773006	GRAPHICS_VERSION_RANGE(1200, 1250)
 14014475959	GRAPHICS_VERSION_RANGE(1270, 1271), GRAPHICS_STEP(A0, B0)
 		PLATFORM(DG2)

From 5fc2bfddb0ae6c6ebd5e690014abf551c6b29bfd Mon Sep 17 00:00:00 2001
From: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Date: Mon, 21 Jul 2025 11:17:28 +0200
Subject: [PATCH 0105/1455] accel/rocket: Add registers header

A XML file was generated with the data from the TRM, and then this
header was generated from it.

The canonical location for the XML file is the Mesa3D repository.

v3:
- Make use of GPL-2.0-only for the copyright notice (Jeff Hugo)

v8:
- Remove full MIT license blob, to match other files with the same
  licensing arrangement in the kernel

Reviewed-by: Robert Foss <rfoss@kernel.org>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721-6-10-rocket-v9-1-77ebd484941e@tomeuvizoso.net
---
 drivers/accel/rocket/rocket_registers.h | 4404 +++++++++++++++++++++++
 1 file changed, 4404 insertions(+)
 create mode 100644 drivers/accel/rocket/rocket_registers.h

diff --git a/drivers/accel/rocket/rocket_registers.h b/drivers/accel/rocket/rocket_registers.h
new file mode 100644
index 000000000000..9aef614c3470
--- /dev/null
+++ b/drivers/accel/rocket/rocket_registers.h
@@ -0,0 +1,4404 @@
+/* SPDX-License-Identifier: GPL-2.0-only OR MIT */
+
+#ifndef __ROCKET_REGISTERS_XML__
+#define __ROCKET_REGISTERS_XML__
+
+/* Autogenerated file, DO NOT EDIT manually!
+
+This file was generated by the rules-ng-ng gen_header.py tool in this git repository:
+http://gitlab.freedesktop.org/mesa/mesa/
+git clone https://gitlab.freedesktop.org/mesa/mesa.git
+
+The rules-ng-ng source files this header was generated from are:
+
+- /home/tomeu/src/mesa/src/gallium/drivers/rocket/registers.xml (  60076 bytes, from Wed Jun 12 10:02:25 2024)
+
+Copyright (C) 2024-2025 by the following authors:
+- Tomeu Vizoso <tomeu@tomeuvizoso.net>
+*/
+
+#define REG_PC_VERSION						0x00000000
+#define PC_VERSION_VERSION__MASK				0xffffffff
+#define PC_VERSION_VERSION__SHIFT				0
+static inline uint32_t PC_VERSION_VERSION(uint32_t val)
+{
+	return ((val) << PC_VERSION_VERSION__SHIFT) & PC_VERSION_VERSION__MASK;
+}
+
+#define REG_PC_VERSION_NUM					0x00000004
+#define PC_VERSION_NUM_VERSION_NUM__MASK			0xffffffff
+#define PC_VERSION_NUM_VERSION_NUM__SHIFT			0
+static inline uint32_t PC_VERSION_NUM_VERSION_NUM(uint32_t val)
+{
+	return ((val) << PC_VERSION_NUM_VERSION_NUM__SHIFT) & PC_VERSION_NUM_VERSION_NUM__MASK;
+}
+
+#define REG_PC_OPERATION_ENABLE					0x00000008
+#define PC_OPERATION_ENABLE_RESERVED_0__MASK			0xfffffffe
+#define PC_OPERATION_ENABLE_RESERVED_0__SHIFT			1
+static inline uint32_t PC_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_OPERATION_ENABLE_RESERVED_0__SHIFT) & PC_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define PC_OPERATION_ENABLE_OP_EN__MASK				0x00000001
+#define PC_OPERATION_ENABLE_OP_EN__SHIFT			0
+static inline uint32_t PC_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << PC_OPERATION_ENABLE_OP_EN__SHIFT) & PC_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_PC_BASE_ADDRESS					0x00000010
+#define PC_BASE_ADDRESS_PC_SOURCE_ADDR__MASK			0xfffffff0
+#define PC_BASE_ADDRESS_PC_SOURCE_ADDR__SHIFT			4
+static inline uint32_t PC_BASE_ADDRESS_PC_SOURCE_ADDR(uint32_t val)
+{
+	return ((val) << PC_BASE_ADDRESS_PC_SOURCE_ADDR__SHIFT) & PC_BASE_ADDRESS_PC_SOURCE_ADDR__MASK;
+}
+#define PC_BASE_ADDRESS_RESERVED_0__MASK			0x0000000e
+#define PC_BASE_ADDRESS_RESERVED_0__SHIFT			1
+static inline uint32_t PC_BASE_ADDRESS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_BASE_ADDRESS_RESERVED_0__SHIFT) & PC_BASE_ADDRESS_RESERVED_0__MASK;
+}
+#define PC_BASE_ADDRESS_PC_SEL__MASK				0x00000001
+#define PC_BASE_ADDRESS_PC_SEL__SHIFT				0
+static inline uint32_t PC_BASE_ADDRESS_PC_SEL(uint32_t val)
+{
+	return ((val) << PC_BASE_ADDRESS_PC_SEL__SHIFT) & PC_BASE_ADDRESS_PC_SEL__MASK;
+}
+
+#define REG_PC_REGISTER_AMOUNTS					0x00000014
+#define PC_REGISTER_AMOUNTS_RESERVED_0__MASK			0xffff0000
+#define PC_REGISTER_AMOUNTS_RESERVED_0__SHIFT			16
+static inline uint32_t PC_REGISTER_AMOUNTS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_REGISTER_AMOUNTS_RESERVED_0__SHIFT) & PC_REGISTER_AMOUNTS_RESERVED_0__MASK;
+}
+#define PC_REGISTER_AMOUNTS_PC_DATA_AMOUNT__MASK		0x0000ffff
+#define PC_REGISTER_AMOUNTS_PC_DATA_AMOUNT__SHIFT		0
+static inline uint32_t PC_REGISTER_AMOUNTS_PC_DATA_AMOUNT(uint32_t val)
+{
+	return ((val) << PC_REGISTER_AMOUNTS_PC_DATA_AMOUNT__SHIFT) & PC_REGISTER_AMOUNTS_PC_DATA_AMOUNT__MASK;
+}
+
+#define REG_PC_INTERRUPT_MASK					0x00000020
+#define PC_INTERRUPT_MASK_RESERVED_0__MASK			0xffffc000
+#define PC_INTERRUPT_MASK_RESERVED_0__SHIFT			14
+static inline uint32_t PC_INTERRUPT_MASK_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_INTERRUPT_MASK_RESERVED_0__SHIFT) & PC_INTERRUPT_MASK_RESERVED_0__MASK;
+}
+#define PC_INTERRUPT_MASK_DMA_WRITE_ERROR			0x00002000
+#define PC_INTERRUPT_MASK_DMA_READ_ERROR			0x00001000
+#define PC_INTERRUPT_MASK_PPU_1					0x00000800
+#define PC_INTERRUPT_MASK_PPU_0					0x00000400
+#define PC_INTERRUPT_MASK_DPU_1					0x00000200
+#define PC_INTERRUPT_MASK_DPU_0					0x00000100
+#define PC_INTERRUPT_MASK_CORE_1				0x00000080
+#define PC_INTERRUPT_MASK_CORE_0				0x00000040
+#define PC_INTERRUPT_MASK_CNA_CSC_1				0x00000020
+#define PC_INTERRUPT_MASK_CNA_CSC_0				0x00000010
+#define PC_INTERRUPT_MASK_CNA_WEIGHT_1				0x00000008
+#define PC_INTERRUPT_MASK_CNA_WEIGHT_0				0x00000004
+#define PC_INTERRUPT_MASK_CNA_FEATURE_1				0x00000002
+#define PC_INTERRUPT_MASK_CNA_FEATURE_0				0x00000001
+
+#define REG_PC_INTERRUPT_CLEAR					0x00000024
+#define PC_INTERRUPT_CLEAR_RESERVED_0__MASK			0xffffc000
+#define PC_INTERRUPT_CLEAR_RESERVED_0__SHIFT			14
+static inline uint32_t PC_INTERRUPT_CLEAR_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_INTERRUPT_CLEAR_RESERVED_0__SHIFT) & PC_INTERRUPT_CLEAR_RESERVED_0__MASK;
+}
+#define PC_INTERRUPT_CLEAR_DMA_WRITE_ERROR			0x00002000
+#define PC_INTERRUPT_CLEAR_DMA_READ_ERROR			0x00001000
+#define PC_INTERRUPT_CLEAR_PPU_1				0x00000800
+#define PC_INTERRUPT_CLEAR_PPU_0				0x00000400
+#define PC_INTERRUPT_CLEAR_DPU_1				0x00000200
+#define PC_INTERRUPT_CLEAR_DPU_0				0x00000100
+#define PC_INTERRUPT_CLEAR_CORE_1				0x00000080
+#define PC_INTERRUPT_CLEAR_CORE_0				0x00000040
+#define PC_INTERRUPT_CLEAR_CNA_CSC_1				0x00000020
+#define PC_INTERRUPT_CLEAR_CNA_CSC_0				0x00000010
+#define PC_INTERRUPT_CLEAR_CNA_WEIGHT_1				0x00000008
+#define PC_INTERRUPT_CLEAR_CNA_WEIGHT_0				0x00000004
+#define PC_INTERRUPT_CLEAR_CNA_FEATURE_1			0x00000002
+#define PC_INTERRUPT_CLEAR_CNA_FEATURE_0			0x00000001
+
+#define REG_PC_INTERRUPT_STATUS					0x00000028
+#define PC_INTERRUPT_STATUS_RESERVED_0__MASK			0xffffc000
+#define PC_INTERRUPT_STATUS_RESERVED_0__SHIFT			14
+static inline uint32_t PC_INTERRUPT_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_INTERRUPT_STATUS_RESERVED_0__SHIFT) & PC_INTERRUPT_STATUS_RESERVED_0__MASK;
+}
+#define PC_INTERRUPT_STATUS_DMA_WRITE_ERROR			0x00002000
+#define PC_INTERRUPT_STATUS_DMA_READ_ERROR			0x00001000
+#define PC_INTERRUPT_STATUS_PPU_1				0x00000800
+#define PC_INTERRUPT_STATUS_PPU_0				0x00000400
+#define PC_INTERRUPT_STATUS_DPU_1				0x00000200
+#define PC_INTERRUPT_STATUS_DPU_0				0x00000100
+#define PC_INTERRUPT_STATUS_CORE_1				0x00000080
+#define PC_INTERRUPT_STATUS_CORE_0				0x00000040
+#define PC_INTERRUPT_STATUS_CNA_CSC_1				0x00000020
+#define PC_INTERRUPT_STATUS_CNA_CSC_0				0x00000010
+#define PC_INTERRUPT_STATUS_CNA_WEIGHT_1			0x00000008
+#define PC_INTERRUPT_STATUS_CNA_WEIGHT_0			0x00000004
+#define PC_INTERRUPT_STATUS_CNA_FEATURE_1			0x00000002
+#define PC_INTERRUPT_STATUS_CNA_FEATURE_0			0x00000001
+
+#define REG_PC_INTERRUPT_RAW_STATUS				0x0000002c
+#define PC_INTERRUPT_RAW_STATUS_RESERVED_0__MASK		0xffffc000
+#define PC_INTERRUPT_RAW_STATUS_RESERVED_0__SHIFT		14
+static inline uint32_t PC_INTERRUPT_RAW_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_INTERRUPT_RAW_STATUS_RESERVED_0__SHIFT) & PC_INTERRUPT_RAW_STATUS_RESERVED_0__MASK;
+}
+#define PC_INTERRUPT_RAW_STATUS_DMA_WRITE_ERROR			0x00002000
+#define PC_INTERRUPT_RAW_STATUS_DMA_READ_ERROR			0x00001000
+#define PC_INTERRUPT_RAW_STATUS_PPU_1				0x00000800
+#define PC_INTERRUPT_RAW_STATUS_PPU_0				0x00000400
+#define PC_INTERRUPT_RAW_STATUS_DPU_1				0x00000200
+#define PC_INTERRUPT_RAW_STATUS_DPU_0				0x00000100
+#define PC_INTERRUPT_RAW_STATUS_CORE_1				0x00000080
+#define PC_INTERRUPT_RAW_STATUS_CORE_0				0x00000040
+#define PC_INTERRUPT_RAW_STATUS_CNA_CSC_1			0x00000020
+#define PC_INTERRUPT_RAW_STATUS_CNA_CSC_0			0x00000010
+#define PC_INTERRUPT_RAW_STATUS_CNA_WEIGHT_1			0x00000008
+#define PC_INTERRUPT_RAW_STATUS_CNA_WEIGHT_0			0x00000004
+#define PC_INTERRUPT_RAW_STATUS_CNA_FEATURE_1			0x00000002
+#define PC_INTERRUPT_RAW_STATUS_CNA_FEATURE_0			0x00000001
+
+#define REG_PC_TASK_CON						0x00000030
+#define PC_TASK_CON_RESERVED_0__MASK				0xffffc000
+#define PC_TASK_CON_RESERVED_0__SHIFT				14
+static inline uint32_t PC_TASK_CON_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_TASK_CON_RESERVED_0__SHIFT) & PC_TASK_CON_RESERVED_0__MASK;
+}
+#define PC_TASK_CON_TASK_COUNT_CLEAR__MASK			0x00002000
+#define PC_TASK_CON_TASK_COUNT_CLEAR__SHIFT			13
+static inline uint32_t PC_TASK_CON_TASK_COUNT_CLEAR(uint32_t val)
+{
+	return ((val) << PC_TASK_CON_TASK_COUNT_CLEAR__SHIFT) & PC_TASK_CON_TASK_COUNT_CLEAR__MASK;
+}
+#define PC_TASK_CON_TASK_PP_EN__MASK				0x00001000
+#define PC_TASK_CON_TASK_PP_EN__SHIFT				12
+static inline uint32_t PC_TASK_CON_TASK_PP_EN(uint32_t val)
+{
+	return ((val) << PC_TASK_CON_TASK_PP_EN__SHIFT) & PC_TASK_CON_TASK_PP_EN__MASK;
+}
+#define PC_TASK_CON_TASK_NUMBER__MASK				0x00000fff
+#define PC_TASK_CON_TASK_NUMBER__SHIFT				0
+static inline uint32_t PC_TASK_CON_TASK_NUMBER(uint32_t val)
+{
+	return ((val) << PC_TASK_CON_TASK_NUMBER__SHIFT) & PC_TASK_CON_TASK_NUMBER__MASK;
+}
+
+#define REG_PC_TASK_DMA_BASE_ADDR				0x00000034
+#define PC_TASK_DMA_BASE_ADDR_DMA_BASE_ADDR__MASK		0xfffffff0
+#define PC_TASK_DMA_BASE_ADDR_DMA_BASE_ADDR__SHIFT		4
+static inline uint32_t PC_TASK_DMA_BASE_ADDR_DMA_BASE_ADDR(uint32_t val)
+{
+	return ((val) << PC_TASK_DMA_BASE_ADDR_DMA_BASE_ADDR__SHIFT) & PC_TASK_DMA_BASE_ADDR_DMA_BASE_ADDR__MASK;
+}
+#define PC_TASK_DMA_BASE_ADDR_RESERVED_0__MASK			0x0000000f
+#define PC_TASK_DMA_BASE_ADDR_RESERVED_0__SHIFT			0
+static inline uint32_t PC_TASK_DMA_BASE_ADDR_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_TASK_DMA_BASE_ADDR_RESERVED_0__SHIFT) & PC_TASK_DMA_BASE_ADDR_RESERVED_0__MASK;
+}
+
+#define REG_PC_TASK_STATUS					0x0000003c
+#define PC_TASK_STATUS_RESERVED_0__MASK				0xf0000000
+#define PC_TASK_STATUS_RESERVED_0__SHIFT			28
+static inline uint32_t PC_TASK_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PC_TASK_STATUS_RESERVED_0__SHIFT) & PC_TASK_STATUS_RESERVED_0__MASK;
+}
+#define PC_TASK_STATUS_TASK_STATUS__MASK			0x0fffffff
+#define PC_TASK_STATUS_TASK_STATUS__SHIFT			0
+static inline uint32_t PC_TASK_STATUS_TASK_STATUS(uint32_t val)
+{
+	return ((val) << PC_TASK_STATUS_TASK_STATUS__SHIFT) & PC_TASK_STATUS_TASK_STATUS__MASK;
+}
+
+#define REG_CNA_S_STATUS					0x00001000
+#define CNA_S_STATUS_RESERVED_0__MASK				0xfffc0000
+#define CNA_S_STATUS_RESERVED_0__SHIFT				18
+static inline uint32_t CNA_S_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_S_STATUS_RESERVED_0__SHIFT) & CNA_S_STATUS_RESERVED_0__MASK;
+}
+#define CNA_S_STATUS_STATUS_1__MASK				0x00030000
+#define CNA_S_STATUS_STATUS_1__SHIFT				16
+static inline uint32_t CNA_S_STATUS_STATUS_1(uint32_t val)
+{
+	return ((val) << CNA_S_STATUS_STATUS_1__SHIFT) & CNA_S_STATUS_STATUS_1__MASK;
+}
+#define CNA_S_STATUS_RESERVED_1__MASK				0x0000fffc
+#define CNA_S_STATUS_RESERVED_1__SHIFT				2
+static inline uint32_t CNA_S_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_S_STATUS_RESERVED_1__SHIFT) & CNA_S_STATUS_RESERVED_1__MASK;
+}
+#define CNA_S_STATUS_STATUS_0__MASK				0x00000003
+#define CNA_S_STATUS_STATUS_0__SHIFT				0
+static inline uint32_t CNA_S_STATUS_STATUS_0(uint32_t val)
+{
+	return ((val) << CNA_S_STATUS_STATUS_0__SHIFT) & CNA_S_STATUS_STATUS_0__MASK;
+}
+
+#define REG_CNA_S_POINTER					0x00001004
+#define CNA_S_POINTER_RESERVED_0__MASK				0xfffe0000
+#define CNA_S_POINTER_RESERVED_0__SHIFT				17
+static inline uint32_t CNA_S_POINTER_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_RESERVED_0__SHIFT) & CNA_S_POINTER_RESERVED_0__MASK;
+}
+#define CNA_S_POINTER_EXECUTER__MASK				0x00010000
+#define CNA_S_POINTER_EXECUTER__SHIFT				16
+static inline uint32_t CNA_S_POINTER_EXECUTER(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_EXECUTER__SHIFT) & CNA_S_POINTER_EXECUTER__MASK;
+}
+#define CNA_S_POINTER_RESERVED_1__MASK				0x0000ffc0
+#define CNA_S_POINTER_RESERVED_1__SHIFT				6
+static inline uint32_t CNA_S_POINTER_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_RESERVED_1__SHIFT) & CNA_S_POINTER_RESERVED_1__MASK;
+}
+#define CNA_S_POINTER_EXECUTER_PP_CLEAR__MASK			0x00000020
+#define CNA_S_POINTER_EXECUTER_PP_CLEAR__SHIFT			5
+static inline uint32_t CNA_S_POINTER_EXECUTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_EXECUTER_PP_CLEAR__SHIFT) & CNA_S_POINTER_EXECUTER_PP_CLEAR__MASK;
+}
+#define CNA_S_POINTER_POINTER_PP_CLEAR__MASK			0x00000010
+#define CNA_S_POINTER_POINTER_PP_CLEAR__SHIFT			4
+static inline uint32_t CNA_S_POINTER_POINTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_POINTER_PP_CLEAR__SHIFT) & CNA_S_POINTER_POINTER_PP_CLEAR__MASK;
+}
+#define CNA_S_POINTER_POINTER_PP_MODE__MASK			0x00000008
+#define CNA_S_POINTER_POINTER_PP_MODE__SHIFT			3
+static inline uint32_t CNA_S_POINTER_POINTER_PP_MODE(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_POINTER_PP_MODE__SHIFT) & CNA_S_POINTER_POINTER_PP_MODE__MASK;
+}
+#define CNA_S_POINTER_EXECUTER_PP_EN__MASK			0x00000004
+#define CNA_S_POINTER_EXECUTER_PP_EN__SHIFT			2
+static inline uint32_t CNA_S_POINTER_EXECUTER_PP_EN(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_EXECUTER_PP_EN__SHIFT) & CNA_S_POINTER_EXECUTER_PP_EN__MASK;
+}
+#define CNA_S_POINTER_POINTER_PP_EN__MASK			0x00000002
+#define CNA_S_POINTER_POINTER_PP_EN__SHIFT			1
+static inline uint32_t CNA_S_POINTER_POINTER_PP_EN(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_POINTER_PP_EN__SHIFT) & CNA_S_POINTER_POINTER_PP_EN__MASK;
+}
+#define CNA_S_POINTER_POINTER__MASK				0x00000001
+#define CNA_S_POINTER_POINTER__SHIFT				0
+static inline uint32_t CNA_S_POINTER_POINTER(uint32_t val)
+{
+	return ((val) << CNA_S_POINTER_POINTER__SHIFT) & CNA_S_POINTER_POINTER__MASK;
+}
+
+#define REG_CNA_OPERATION_ENABLE				0x00001008
+#define CNA_OPERATION_ENABLE_RESERVED_0__MASK			0xfffffffe
+#define CNA_OPERATION_ENABLE_RESERVED_0__SHIFT			1
+static inline uint32_t CNA_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_OPERATION_ENABLE_RESERVED_0__SHIFT) & CNA_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define CNA_OPERATION_ENABLE_OP_EN__MASK			0x00000001
+#define CNA_OPERATION_ENABLE_OP_EN__SHIFT			0
+static inline uint32_t CNA_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << CNA_OPERATION_ENABLE_OP_EN__SHIFT) & CNA_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_CNA_CONV_CON1					0x0000100c
+#define CNA_CONV_CON1_RESERVED_0__MASK				0x80000000
+#define CNA_CONV_CON1_RESERVED_0__SHIFT				31
+static inline uint32_t CNA_CONV_CON1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_RESERVED_0__SHIFT) & CNA_CONV_CON1_RESERVED_0__MASK;
+}
+#define CNA_CONV_CON1_NONALIGN_DMA__MASK			0x40000000
+#define CNA_CONV_CON1_NONALIGN_DMA__SHIFT			30
+static inline uint32_t CNA_CONV_CON1_NONALIGN_DMA(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_NONALIGN_DMA__SHIFT) & CNA_CONV_CON1_NONALIGN_DMA__MASK;
+}
+#define CNA_CONV_CON1_GROUP_LINE_OFF__MASK			0x20000000
+#define CNA_CONV_CON1_GROUP_LINE_OFF__SHIFT			29
+static inline uint32_t CNA_CONV_CON1_GROUP_LINE_OFF(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_GROUP_LINE_OFF__SHIFT) & CNA_CONV_CON1_GROUP_LINE_OFF__MASK;
+}
+#define CNA_CONV_CON1_RESERVED_1__MASK				0x1ffe0000
+#define CNA_CONV_CON1_RESERVED_1__SHIFT				17
+static inline uint32_t CNA_CONV_CON1_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_RESERVED_1__SHIFT) & CNA_CONV_CON1_RESERVED_1__MASK;
+}
+#define CNA_CONV_CON1_DECONV__MASK				0x00010000
+#define CNA_CONV_CON1_DECONV__SHIFT				16
+static inline uint32_t CNA_CONV_CON1_DECONV(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_DECONV__SHIFT) & CNA_CONV_CON1_DECONV__MASK;
+}
+#define CNA_CONV_CON1_ARGB_IN__MASK				0x0000f000
+#define CNA_CONV_CON1_ARGB_IN__SHIFT				12
+static inline uint32_t CNA_CONV_CON1_ARGB_IN(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_ARGB_IN__SHIFT) & CNA_CONV_CON1_ARGB_IN__MASK;
+}
+#define CNA_CONV_CON1_RESERVED_2__MASK				0x00000c00
+#define CNA_CONV_CON1_RESERVED_2__SHIFT				10
+static inline uint32_t CNA_CONV_CON1_RESERVED_2(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_RESERVED_2__SHIFT) & CNA_CONV_CON1_RESERVED_2__MASK;
+}
+#define CNA_CONV_CON1_PROC_PRECISION__MASK			0x00000380
+#define CNA_CONV_CON1_PROC_PRECISION__SHIFT			7
+static inline uint32_t CNA_CONV_CON1_PROC_PRECISION(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_PROC_PRECISION__SHIFT) & CNA_CONV_CON1_PROC_PRECISION__MASK;
+}
+#define CNA_CONV_CON1_IN_PRECISION__MASK			0x00000070
+#define CNA_CONV_CON1_IN_PRECISION__SHIFT			4
+static inline uint32_t CNA_CONV_CON1_IN_PRECISION(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_IN_PRECISION__SHIFT) & CNA_CONV_CON1_IN_PRECISION__MASK;
+}
+#define CNA_CONV_CON1_CONV_MODE__MASK				0x0000000f
+#define CNA_CONV_CON1_CONV_MODE__SHIFT				0
+static inline uint32_t CNA_CONV_CON1_CONV_MODE(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON1_CONV_MODE__SHIFT) & CNA_CONV_CON1_CONV_MODE__MASK;
+}
+
+#define REG_CNA_CONV_CON2					0x00001010
+#define CNA_CONV_CON2_RESERVED_0__MASK				0xff000000
+#define CNA_CONV_CON2_RESERVED_0__SHIFT				24
+static inline uint32_t CNA_CONV_CON2_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_RESERVED_0__SHIFT) & CNA_CONV_CON2_RESERVED_0__MASK;
+}
+#define CNA_CONV_CON2_KERNEL_GROUP__MASK			0x00ff0000
+#define CNA_CONV_CON2_KERNEL_GROUP__SHIFT			16
+static inline uint32_t CNA_CONV_CON2_KERNEL_GROUP(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_KERNEL_GROUP__SHIFT) & CNA_CONV_CON2_KERNEL_GROUP__MASK;
+}
+#define CNA_CONV_CON2_RESERVED_1__MASK				0x0000c000
+#define CNA_CONV_CON2_RESERVED_1__SHIFT				14
+static inline uint32_t CNA_CONV_CON2_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_RESERVED_1__SHIFT) & CNA_CONV_CON2_RESERVED_1__MASK;
+}
+#define CNA_CONV_CON2_FEATURE_GRAINS__MASK			0x00003ff0
+#define CNA_CONV_CON2_FEATURE_GRAINS__SHIFT			4
+static inline uint32_t CNA_CONV_CON2_FEATURE_GRAINS(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_FEATURE_GRAINS__SHIFT) & CNA_CONV_CON2_FEATURE_GRAINS__MASK;
+}
+#define CNA_CONV_CON2_RESERVED_2__MASK				0x00000008
+#define CNA_CONV_CON2_RESERVED_2__SHIFT				3
+static inline uint32_t CNA_CONV_CON2_RESERVED_2(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_RESERVED_2__SHIFT) & CNA_CONV_CON2_RESERVED_2__MASK;
+}
+#define CNA_CONV_CON2_CSC_WO_EN__MASK				0x00000004
+#define CNA_CONV_CON2_CSC_WO_EN__SHIFT				2
+static inline uint32_t CNA_CONV_CON2_CSC_WO_EN(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_CSC_WO_EN__SHIFT) & CNA_CONV_CON2_CSC_WO_EN__MASK;
+}
+#define CNA_CONV_CON2_CSC_DO_EN__MASK				0x00000002
+#define CNA_CONV_CON2_CSC_DO_EN__SHIFT				1
+static inline uint32_t CNA_CONV_CON2_CSC_DO_EN(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_CSC_DO_EN__SHIFT) & CNA_CONV_CON2_CSC_DO_EN__MASK;
+}
+#define CNA_CONV_CON2_CMD_FIFO_SRST__MASK			0x00000001
+#define CNA_CONV_CON2_CMD_FIFO_SRST__SHIFT			0
+static inline uint32_t CNA_CONV_CON2_CMD_FIFO_SRST(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON2_CMD_FIFO_SRST__SHIFT) & CNA_CONV_CON2_CMD_FIFO_SRST__MASK;
+}
+
+#define REG_CNA_CONV_CON3					0x00001014
+#define CNA_CONV_CON3_RESERVED_0__MASK				0x80000000
+#define CNA_CONV_CON3_RESERVED_0__SHIFT				31
+static inline uint32_t CNA_CONV_CON3_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_RESERVED_0__SHIFT) & CNA_CONV_CON3_RESERVED_0__MASK;
+}
+#define CNA_CONV_CON3_NN_MODE__MASK				0x70000000
+#define CNA_CONV_CON3_NN_MODE__SHIFT				28
+static inline uint32_t CNA_CONV_CON3_NN_MODE(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_NN_MODE__SHIFT) & CNA_CONV_CON3_NN_MODE__MASK;
+}
+#define CNA_CONV_CON3_RESERVED_1__MASK				0x0c000000
+#define CNA_CONV_CON3_RESERVED_1__SHIFT				26
+static inline uint32_t CNA_CONV_CON3_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_RESERVED_1__SHIFT) & CNA_CONV_CON3_RESERVED_1__MASK;
+}
+#define CNA_CONV_CON3_ATROUS_Y_DILATION__MASK			0x03e00000
+#define CNA_CONV_CON3_ATROUS_Y_DILATION__SHIFT			21
+static inline uint32_t CNA_CONV_CON3_ATROUS_Y_DILATION(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_ATROUS_Y_DILATION__SHIFT) & CNA_CONV_CON3_ATROUS_Y_DILATION__MASK;
+}
+#define CNA_CONV_CON3_ATROUS_X_DILATION__MASK			0x001f0000
+#define CNA_CONV_CON3_ATROUS_X_DILATION__SHIFT			16
+static inline uint32_t CNA_CONV_CON3_ATROUS_X_DILATION(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_ATROUS_X_DILATION__SHIFT) & CNA_CONV_CON3_ATROUS_X_DILATION__MASK;
+}
+#define CNA_CONV_CON3_RESERVED_2__MASK				0x0000c000
+#define CNA_CONV_CON3_RESERVED_2__SHIFT				14
+static inline uint32_t CNA_CONV_CON3_RESERVED_2(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_RESERVED_2__SHIFT) & CNA_CONV_CON3_RESERVED_2__MASK;
+}
+#define CNA_CONV_CON3_DECONV_Y_STRIDE__MASK			0x00003800
+#define CNA_CONV_CON3_DECONV_Y_STRIDE__SHIFT			11
+static inline uint32_t CNA_CONV_CON3_DECONV_Y_STRIDE(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_DECONV_Y_STRIDE__SHIFT) & CNA_CONV_CON3_DECONV_Y_STRIDE__MASK;
+}
+#define CNA_CONV_CON3_DECONV_X_STRIDE__MASK			0x00000700
+#define CNA_CONV_CON3_DECONV_X_STRIDE__SHIFT			8
+static inline uint32_t CNA_CONV_CON3_DECONV_X_STRIDE(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_DECONV_X_STRIDE__SHIFT) & CNA_CONV_CON3_DECONV_X_STRIDE__MASK;
+}
+#define CNA_CONV_CON3_RESERVED_3__MASK				0x000000c0
+#define CNA_CONV_CON3_RESERVED_3__SHIFT				6
+static inline uint32_t CNA_CONV_CON3_RESERVED_3(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_RESERVED_3__SHIFT) & CNA_CONV_CON3_RESERVED_3__MASK;
+}
+#define CNA_CONV_CON3_CONV_Y_STRIDE__MASK			0x00000038
+#define CNA_CONV_CON3_CONV_Y_STRIDE__SHIFT			3
+static inline uint32_t CNA_CONV_CON3_CONV_Y_STRIDE(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_CONV_Y_STRIDE__SHIFT) & CNA_CONV_CON3_CONV_Y_STRIDE__MASK;
+}
+#define CNA_CONV_CON3_CONV_X_STRIDE__MASK			0x00000007
+#define CNA_CONV_CON3_CONV_X_STRIDE__SHIFT			0
+static inline uint32_t CNA_CONV_CON3_CONV_X_STRIDE(uint32_t val)
+{
+	return ((val) << CNA_CONV_CON3_CONV_X_STRIDE__SHIFT) & CNA_CONV_CON3_CONV_X_STRIDE__MASK;
+}
+
+#define REG_CNA_DATA_SIZE0					0x00001020
+#define CNA_DATA_SIZE0_RESERVED_0__MASK				0xf8000000
+#define CNA_DATA_SIZE0_RESERVED_0__SHIFT			27
+static inline uint32_t CNA_DATA_SIZE0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE0_RESERVED_0__SHIFT) & CNA_DATA_SIZE0_RESERVED_0__MASK;
+}
+#define CNA_DATA_SIZE0_DATAIN_WIDTH__MASK			0x07ff0000
+#define CNA_DATA_SIZE0_DATAIN_WIDTH__SHIFT			16
+static inline uint32_t CNA_DATA_SIZE0_DATAIN_WIDTH(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE0_DATAIN_WIDTH__SHIFT) & CNA_DATA_SIZE0_DATAIN_WIDTH__MASK;
+}
+#define CNA_DATA_SIZE0_RESERVED_1__MASK				0x0000f800
+#define CNA_DATA_SIZE0_RESERVED_1__SHIFT			11
+static inline uint32_t CNA_DATA_SIZE0_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE0_RESERVED_1__SHIFT) & CNA_DATA_SIZE0_RESERVED_1__MASK;
+}
+#define CNA_DATA_SIZE0_DATAIN_HEIGHT__MASK			0x000007ff
+#define CNA_DATA_SIZE0_DATAIN_HEIGHT__SHIFT			0
+static inline uint32_t CNA_DATA_SIZE0_DATAIN_HEIGHT(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE0_DATAIN_HEIGHT__SHIFT) & CNA_DATA_SIZE0_DATAIN_HEIGHT__MASK;
+}
+
+#define REG_CNA_DATA_SIZE1					0x00001024
+#define CNA_DATA_SIZE1_RESERVED_0__MASK				0xc0000000
+#define CNA_DATA_SIZE1_RESERVED_0__SHIFT			30
+static inline uint32_t CNA_DATA_SIZE1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE1_RESERVED_0__SHIFT) & CNA_DATA_SIZE1_RESERVED_0__MASK;
+}
+#define CNA_DATA_SIZE1_DATAIN_CHANNEL_REAL__MASK		0x3fff0000
+#define CNA_DATA_SIZE1_DATAIN_CHANNEL_REAL__SHIFT		16
+static inline uint32_t CNA_DATA_SIZE1_DATAIN_CHANNEL_REAL(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE1_DATAIN_CHANNEL_REAL__SHIFT) & CNA_DATA_SIZE1_DATAIN_CHANNEL_REAL__MASK;
+}
+#define CNA_DATA_SIZE1_DATAIN_CHANNEL__MASK			0x0000ffff
+#define CNA_DATA_SIZE1_DATAIN_CHANNEL__SHIFT			0
+static inline uint32_t CNA_DATA_SIZE1_DATAIN_CHANNEL(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE1_DATAIN_CHANNEL__SHIFT) & CNA_DATA_SIZE1_DATAIN_CHANNEL__MASK;
+}
+
+#define REG_CNA_DATA_SIZE2					0x00001028
+#define CNA_DATA_SIZE2_RESERVED_0__MASK				0xfffff800
+#define CNA_DATA_SIZE2_RESERVED_0__SHIFT			11
+static inline uint32_t CNA_DATA_SIZE2_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE2_RESERVED_0__SHIFT) & CNA_DATA_SIZE2_RESERVED_0__MASK;
+}
+#define CNA_DATA_SIZE2_DATAOUT_WIDTH__MASK			0x000007ff
+#define CNA_DATA_SIZE2_DATAOUT_WIDTH__SHIFT			0
+static inline uint32_t CNA_DATA_SIZE2_DATAOUT_WIDTH(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE2_DATAOUT_WIDTH__SHIFT) & CNA_DATA_SIZE2_DATAOUT_WIDTH__MASK;
+}
+
+#define REG_CNA_DATA_SIZE3					0x0000102c
+#define CNA_DATA_SIZE3_RESERVED_0__MASK				0xff000000
+#define CNA_DATA_SIZE3_RESERVED_0__SHIFT			24
+static inline uint32_t CNA_DATA_SIZE3_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE3_RESERVED_0__SHIFT) & CNA_DATA_SIZE3_RESERVED_0__MASK;
+}
+#define CNA_DATA_SIZE3_SURF_MODE__MASK				0x00c00000
+#define CNA_DATA_SIZE3_SURF_MODE__SHIFT				22
+static inline uint32_t CNA_DATA_SIZE3_SURF_MODE(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE3_SURF_MODE__SHIFT) & CNA_DATA_SIZE3_SURF_MODE__MASK;
+}
+#define CNA_DATA_SIZE3_DATAOUT_ATOMICS__MASK			0x003fffff
+#define CNA_DATA_SIZE3_DATAOUT_ATOMICS__SHIFT			0
+static inline uint32_t CNA_DATA_SIZE3_DATAOUT_ATOMICS(uint32_t val)
+{
+	return ((val) << CNA_DATA_SIZE3_DATAOUT_ATOMICS__SHIFT) & CNA_DATA_SIZE3_DATAOUT_ATOMICS__MASK;
+}
+
+#define REG_CNA_WEIGHT_SIZE0					0x00001030
+#define CNA_WEIGHT_SIZE0_WEIGHT_BYTES__MASK			0xffffffff
+#define CNA_WEIGHT_SIZE0_WEIGHT_BYTES__SHIFT			0
+static inline uint32_t CNA_WEIGHT_SIZE0_WEIGHT_BYTES(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE0_WEIGHT_BYTES__SHIFT) & CNA_WEIGHT_SIZE0_WEIGHT_BYTES__MASK;
+}
+
+#define REG_CNA_WEIGHT_SIZE1					0x00001034
+#define CNA_WEIGHT_SIZE1_RESERVED_0__MASK			0xfff80000
+#define CNA_WEIGHT_SIZE1_RESERVED_0__SHIFT			19
+static inline uint32_t CNA_WEIGHT_SIZE1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE1_RESERVED_0__SHIFT) & CNA_WEIGHT_SIZE1_RESERVED_0__MASK;
+}
+#define CNA_WEIGHT_SIZE1_WEIGHT_BYTES_PER_KERNEL__MASK		0x0007ffff
+#define CNA_WEIGHT_SIZE1_WEIGHT_BYTES_PER_KERNEL__SHIFT		0
+static inline uint32_t CNA_WEIGHT_SIZE1_WEIGHT_BYTES_PER_KERNEL(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE1_WEIGHT_BYTES_PER_KERNEL__SHIFT) & CNA_WEIGHT_SIZE1_WEIGHT_BYTES_PER_KERNEL__MASK;
+}
+
+#define REG_CNA_WEIGHT_SIZE2					0x00001038
+#define CNA_WEIGHT_SIZE2_RESERVED_0__MASK			0xe0000000
+#define CNA_WEIGHT_SIZE2_RESERVED_0__SHIFT			29
+static inline uint32_t CNA_WEIGHT_SIZE2_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE2_RESERVED_0__SHIFT) & CNA_WEIGHT_SIZE2_RESERVED_0__MASK;
+}
+#define CNA_WEIGHT_SIZE2_WEIGHT_WIDTH__MASK			0x1f000000
+#define CNA_WEIGHT_SIZE2_WEIGHT_WIDTH__SHIFT			24
+static inline uint32_t CNA_WEIGHT_SIZE2_WEIGHT_WIDTH(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE2_WEIGHT_WIDTH__SHIFT) & CNA_WEIGHT_SIZE2_WEIGHT_WIDTH__MASK;
+}
+#define CNA_WEIGHT_SIZE2_RESERVED_1__MASK			0x00e00000
+#define CNA_WEIGHT_SIZE2_RESERVED_1__SHIFT			21
+static inline uint32_t CNA_WEIGHT_SIZE2_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE2_RESERVED_1__SHIFT) & CNA_WEIGHT_SIZE2_RESERVED_1__MASK;
+}
+#define CNA_WEIGHT_SIZE2_WEIGHT_HEIGHT__MASK			0x001f0000
+#define CNA_WEIGHT_SIZE2_WEIGHT_HEIGHT__SHIFT			16
+static inline uint32_t CNA_WEIGHT_SIZE2_WEIGHT_HEIGHT(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE2_WEIGHT_HEIGHT__SHIFT) & CNA_WEIGHT_SIZE2_WEIGHT_HEIGHT__MASK;
+}
+#define CNA_WEIGHT_SIZE2_RESERVED_2__MASK			0x0000c000
+#define CNA_WEIGHT_SIZE2_RESERVED_2__SHIFT			14
+static inline uint32_t CNA_WEIGHT_SIZE2_RESERVED_2(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE2_RESERVED_2__SHIFT) & CNA_WEIGHT_SIZE2_RESERVED_2__MASK;
+}
+#define CNA_WEIGHT_SIZE2_WEIGHT_KERNELS__MASK			0x00003fff
+#define CNA_WEIGHT_SIZE2_WEIGHT_KERNELS__SHIFT			0
+static inline uint32_t CNA_WEIGHT_SIZE2_WEIGHT_KERNELS(uint32_t val)
+{
+	return ((val) << CNA_WEIGHT_SIZE2_WEIGHT_KERNELS__SHIFT) & CNA_WEIGHT_SIZE2_WEIGHT_KERNELS__MASK;
+}
+
+#define REG_CNA_CBUF_CON0					0x00001040
+#define CNA_CBUF_CON0_RESERVED_0__MASK				0xffffc000
+#define CNA_CBUF_CON0_RESERVED_0__SHIFT				14
+static inline uint32_t CNA_CBUF_CON0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_RESERVED_0__SHIFT) & CNA_CBUF_CON0_RESERVED_0__MASK;
+}
+#define CNA_CBUF_CON0_WEIGHT_REUSE__MASK			0x00002000
+#define CNA_CBUF_CON0_WEIGHT_REUSE__SHIFT			13
+static inline uint32_t CNA_CBUF_CON0_WEIGHT_REUSE(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_WEIGHT_REUSE__SHIFT) & CNA_CBUF_CON0_WEIGHT_REUSE__MASK;
+}
+#define CNA_CBUF_CON0_DATA_REUSE__MASK				0x00001000
+#define CNA_CBUF_CON0_DATA_REUSE__SHIFT				12
+static inline uint32_t CNA_CBUF_CON0_DATA_REUSE(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_DATA_REUSE__SHIFT) & CNA_CBUF_CON0_DATA_REUSE__MASK;
+}
+#define CNA_CBUF_CON0_RESERVED_1__MASK				0x00000800
+#define CNA_CBUF_CON0_RESERVED_1__SHIFT				11
+static inline uint32_t CNA_CBUF_CON0_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_RESERVED_1__SHIFT) & CNA_CBUF_CON0_RESERVED_1__MASK;
+}
+#define CNA_CBUF_CON0_FC_DATA_BANK__MASK			0x00000700
+#define CNA_CBUF_CON0_FC_DATA_BANK__SHIFT			8
+static inline uint32_t CNA_CBUF_CON0_FC_DATA_BANK(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_FC_DATA_BANK__SHIFT) & CNA_CBUF_CON0_FC_DATA_BANK__MASK;
+}
+#define CNA_CBUF_CON0_WEIGHT_BANK__MASK				0x000000f0
+#define CNA_CBUF_CON0_WEIGHT_BANK__SHIFT			4
+static inline uint32_t CNA_CBUF_CON0_WEIGHT_BANK(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_WEIGHT_BANK__SHIFT) & CNA_CBUF_CON0_WEIGHT_BANK__MASK;
+}
+#define CNA_CBUF_CON0_DATA_BANK__MASK				0x0000000f
+#define CNA_CBUF_CON0_DATA_BANK__SHIFT				0
+static inline uint32_t CNA_CBUF_CON0_DATA_BANK(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON0_DATA_BANK__SHIFT) & CNA_CBUF_CON0_DATA_BANK__MASK;
+}
+
+#define REG_CNA_CBUF_CON1					0x00001044
+#define CNA_CBUF_CON1_RESERVED_0__MASK				0xffffc000
+#define CNA_CBUF_CON1_RESERVED_0__SHIFT				14
+static inline uint32_t CNA_CBUF_CON1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON1_RESERVED_0__SHIFT) & CNA_CBUF_CON1_RESERVED_0__MASK;
+}
+#define CNA_CBUF_CON1_DATA_ENTRIES__MASK			0x00003fff
+#define CNA_CBUF_CON1_DATA_ENTRIES__SHIFT			0
+static inline uint32_t CNA_CBUF_CON1_DATA_ENTRIES(uint32_t val)
+{
+	return ((val) << CNA_CBUF_CON1_DATA_ENTRIES__SHIFT) & CNA_CBUF_CON1_DATA_ENTRIES__MASK;
+}
+
+#define REG_CNA_CVT_CON0					0x0000104c
+#define CNA_CVT_CON0_RESERVED_0__MASK				0xf0000000
+#define CNA_CVT_CON0_RESERVED_0__SHIFT				28
+static inline uint32_t CNA_CVT_CON0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_RESERVED_0__SHIFT) & CNA_CVT_CON0_RESERVED_0__MASK;
+}
+#define CNA_CVT_CON0_CVT_TRUNCATE_3__MASK			0x0fc00000
+#define CNA_CVT_CON0_CVT_TRUNCATE_3__SHIFT			22
+static inline uint32_t CNA_CVT_CON0_CVT_TRUNCATE_3(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_CVT_TRUNCATE_3__SHIFT) & CNA_CVT_CON0_CVT_TRUNCATE_3__MASK;
+}
+#define CNA_CVT_CON0_CVT_TRUNCATE_2__MASK			0x003f0000
+#define CNA_CVT_CON0_CVT_TRUNCATE_2__SHIFT			16
+static inline uint32_t CNA_CVT_CON0_CVT_TRUNCATE_2(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_CVT_TRUNCATE_2__SHIFT) & CNA_CVT_CON0_CVT_TRUNCATE_2__MASK;
+}
+#define CNA_CVT_CON0_CVT_TRUNCATE_1__MASK			0x0000fc00
+#define CNA_CVT_CON0_CVT_TRUNCATE_1__SHIFT			10
+static inline uint32_t CNA_CVT_CON0_CVT_TRUNCATE_1(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_CVT_TRUNCATE_1__SHIFT) & CNA_CVT_CON0_CVT_TRUNCATE_1__MASK;
+}
+#define CNA_CVT_CON0_CVT_TRUNCATE_0__MASK			0x000003f0
+#define CNA_CVT_CON0_CVT_TRUNCATE_0__SHIFT			4
+static inline uint32_t CNA_CVT_CON0_CVT_TRUNCATE_0(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_CVT_TRUNCATE_0__SHIFT) & CNA_CVT_CON0_CVT_TRUNCATE_0__MASK;
+}
+#define CNA_CVT_CON0_DATA_SIGN__MASK				0x00000008
+#define CNA_CVT_CON0_DATA_SIGN__SHIFT				3
+static inline uint32_t CNA_CVT_CON0_DATA_SIGN(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_DATA_SIGN__SHIFT) & CNA_CVT_CON0_DATA_SIGN__MASK;
+}
+#define CNA_CVT_CON0_ROUND_TYPE__MASK				0x00000004
+#define CNA_CVT_CON0_ROUND_TYPE__SHIFT				2
+static inline uint32_t CNA_CVT_CON0_ROUND_TYPE(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_ROUND_TYPE__SHIFT) & CNA_CVT_CON0_ROUND_TYPE__MASK;
+}
+#define CNA_CVT_CON0_CVT_TYPE__MASK				0x00000002
+#define CNA_CVT_CON0_CVT_TYPE__SHIFT				1
+static inline uint32_t CNA_CVT_CON0_CVT_TYPE(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_CVT_TYPE__SHIFT) & CNA_CVT_CON0_CVT_TYPE__MASK;
+}
+#define CNA_CVT_CON0_CVT_BYPASS__MASK				0x00000001
+#define CNA_CVT_CON0_CVT_BYPASS__SHIFT				0
+static inline uint32_t CNA_CVT_CON0_CVT_BYPASS(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON0_CVT_BYPASS__SHIFT) & CNA_CVT_CON0_CVT_BYPASS__MASK;
+}
+
+#define REG_CNA_CVT_CON1					0x00001050
+#define CNA_CVT_CON1_CVT_SCALE0__MASK				0xffff0000
+#define CNA_CVT_CON1_CVT_SCALE0__SHIFT				16
+static inline uint32_t CNA_CVT_CON1_CVT_SCALE0(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON1_CVT_SCALE0__SHIFT) & CNA_CVT_CON1_CVT_SCALE0__MASK;
+}
+#define CNA_CVT_CON1_CVT_OFFSET0__MASK				0x0000ffff
+#define CNA_CVT_CON1_CVT_OFFSET0__SHIFT				0
+static inline uint32_t CNA_CVT_CON1_CVT_OFFSET0(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON1_CVT_OFFSET0__SHIFT) & CNA_CVT_CON1_CVT_OFFSET0__MASK;
+}
+
+#define REG_CNA_CVT_CON2					0x00001054
+#define CNA_CVT_CON2_CVT_SCALE1__MASK				0xffff0000
+#define CNA_CVT_CON2_CVT_SCALE1__SHIFT				16
+static inline uint32_t CNA_CVT_CON2_CVT_SCALE1(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON2_CVT_SCALE1__SHIFT) & CNA_CVT_CON2_CVT_SCALE1__MASK;
+}
+#define CNA_CVT_CON2_CVT_OFFSET1__MASK				0x0000ffff
+#define CNA_CVT_CON2_CVT_OFFSET1__SHIFT				0
+static inline uint32_t CNA_CVT_CON2_CVT_OFFSET1(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON2_CVT_OFFSET1__SHIFT) & CNA_CVT_CON2_CVT_OFFSET1__MASK;
+}
+
+#define REG_CNA_CVT_CON3					0x00001058
+#define CNA_CVT_CON3_CVT_SCALE2__MASK				0xffff0000
+#define CNA_CVT_CON3_CVT_SCALE2__SHIFT				16
+static inline uint32_t CNA_CVT_CON3_CVT_SCALE2(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON3_CVT_SCALE2__SHIFT) & CNA_CVT_CON3_CVT_SCALE2__MASK;
+}
+#define CNA_CVT_CON3_CVT_OFFSET2__MASK				0x0000ffff
+#define CNA_CVT_CON3_CVT_OFFSET2__SHIFT				0
+static inline uint32_t CNA_CVT_CON3_CVT_OFFSET2(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON3_CVT_OFFSET2__SHIFT) & CNA_CVT_CON3_CVT_OFFSET2__MASK;
+}
+
+#define REG_CNA_CVT_CON4					0x0000105c
+#define CNA_CVT_CON4_CVT_SCALE3__MASK				0xffff0000
+#define CNA_CVT_CON4_CVT_SCALE3__SHIFT				16
+static inline uint32_t CNA_CVT_CON4_CVT_SCALE3(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON4_CVT_SCALE3__SHIFT) & CNA_CVT_CON4_CVT_SCALE3__MASK;
+}
+#define CNA_CVT_CON4_CVT_OFFSET3__MASK				0x0000ffff
+#define CNA_CVT_CON4_CVT_OFFSET3__SHIFT				0
+static inline uint32_t CNA_CVT_CON4_CVT_OFFSET3(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON4_CVT_OFFSET3__SHIFT) & CNA_CVT_CON4_CVT_OFFSET3__MASK;
+}
+
+#define REG_CNA_FC_CON0						0x00001060
+#define CNA_FC_CON0_FC_SKIP_DATA__MASK				0xffff0000
+#define CNA_FC_CON0_FC_SKIP_DATA__SHIFT				16
+static inline uint32_t CNA_FC_CON0_FC_SKIP_DATA(uint32_t val)
+{
+	return ((val) << CNA_FC_CON0_FC_SKIP_DATA__SHIFT) & CNA_FC_CON0_FC_SKIP_DATA__MASK;
+}
+#define CNA_FC_CON0_RESERVED_0__MASK				0x0000fffe
+#define CNA_FC_CON0_RESERVED_0__SHIFT				1
+static inline uint32_t CNA_FC_CON0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_FC_CON0_RESERVED_0__SHIFT) & CNA_FC_CON0_RESERVED_0__MASK;
+}
+#define CNA_FC_CON0_FC_SKIP_EN__MASK				0x00000001
+#define CNA_FC_CON0_FC_SKIP_EN__SHIFT				0
+static inline uint32_t CNA_FC_CON0_FC_SKIP_EN(uint32_t val)
+{
+	return ((val) << CNA_FC_CON0_FC_SKIP_EN__SHIFT) & CNA_FC_CON0_FC_SKIP_EN__MASK;
+}
+
+#define REG_CNA_FC_CON1						0x00001064
+#define CNA_FC_CON1_RESERVED_0__MASK				0xfffe0000
+#define CNA_FC_CON1_RESERVED_0__SHIFT				17
+static inline uint32_t CNA_FC_CON1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_FC_CON1_RESERVED_0__SHIFT) & CNA_FC_CON1_RESERVED_0__MASK;
+}
+#define CNA_FC_CON1_DATA_OFFSET__MASK				0x0001ffff
+#define CNA_FC_CON1_DATA_OFFSET__SHIFT				0
+static inline uint32_t CNA_FC_CON1_DATA_OFFSET(uint32_t val)
+{
+	return ((val) << CNA_FC_CON1_DATA_OFFSET__SHIFT) & CNA_FC_CON1_DATA_OFFSET__MASK;
+}
+
+#define REG_CNA_PAD_CON0					0x00001068
+#define CNA_PAD_CON0_RESERVED_0__MASK				0xffffff00
+#define CNA_PAD_CON0_RESERVED_0__SHIFT				8
+static inline uint32_t CNA_PAD_CON0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_PAD_CON0_RESERVED_0__SHIFT) & CNA_PAD_CON0_RESERVED_0__MASK;
+}
+#define CNA_PAD_CON0_PAD_LEFT__MASK				0x000000f0
+#define CNA_PAD_CON0_PAD_LEFT__SHIFT				4
+static inline uint32_t CNA_PAD_CON0_PAD_LEFT(uint32_t val)
+{
+	return ((val) << CNA_PAD_CON0_PAD_LEFT__SHIFT) & CNA_PAD_CON0_PAD_LEFT__MASK;
+}
+#define CNA_PAD_CON0_PAD_TOP__MASK				0x0000000f
+#define CNA_PAD_CON0_PAD_TOP__SHIFT				0
+static inline uint32_t CNA_PAD_CON0_PAD_TOP(uint32_t val)
+{
+	return ((val) << CNA_PAD_CON0_PAD_TOP__SHIFT) & CNA_PAD_CON0_PAD_TOP__MASK;
+}
+
+#define REG_CNA_FEATURE_DATA_ADDR				0x00001070
+#define CNA_FEATURE_DATA_ADDR_FEATURE_BASE_ADDR__MASK		0xffffffff
+#define CNA_FEATURE_DATA_ADDR_FEATURE_BASE_ADDR__SHIFT		0
+static inline uint32_t CNA_FEATURE_DATA_ADDR_FEATURE_BASE_ADDR(uint32_t val)
+{
+	return ((val) << CNA_FEATURE_DATA_ADDR_FEATURE_BASE_ADDR__SHIFT) & CNA_FEATURE_DATA_ADDR_FEATURE_BASE_ADDR__MASK;
+}
+
+#define REG_CNA_FC_CON2						0x00001074
+#define CNA_FC_CON2_RESERVED_0__MASK				0xfffe0000
+#define CNA_FC_CON2_RESERVED_0__SHIFT				17
+static inline uint32_t CNA_FC_CON2_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_FC_CON2_RESERVED_0__SHIFT) & CNA_FC_CON2_RESERVED_0__MASK;
+}
+#define CNA_FC_CON2_WEIGHT_OFFSET__MASK				0x0001ffff
+#define CNA_FC_CON2_WEIGHT_OFFSET__SHIFT			0
+static inline uint32_t CNA_FC_CON2_WEIGHT_OFFSET(uint32_t val)
+{
+	return ((val) << CNA_FC_CON2_WEIGHT_OFFSET__SHIFT) & CNA_FC_CON2_WEIGHT_OFFSET__MASK;
+}
+
+#define REG_CNA_DMA_CON0					0x00001078
+#define CNA_DMA_CON0_OV4K_BYPASS__MASK				0x80000000
+#define CNA_DMA_CON0_OV4K_BYPASS__SHIFT				31
+static inline uint32_t CNA_DMA_CON0_OV4K_BYPASS(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON0_OV4K_BYPASS__SHIFT) & CNA_DMA_CON0_OV4K_BYPASS__MASK;
+}
+#define CNA_DMA_CON0_RESERVED_0__MASK				0x7ff00000
+#define CNA_DMA_CON0_RESERVED_0__SHIFT				20
+static inline uint32_t CNA_DMA_CON0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON0_RESERVED_0__SHIFT) & CNA_DMA_CON0_RESERVED_0__MASK;
+}
+#define CNA_DMA_CON0_WEIGHT_BURST_LEN__MASK			0x000f0000
+#define CNA_DMA_CON0_WEIGHT_BURST_LEN__SHIFT			16
+static inline uint32_t CNA_DMA_CON0_WEIGHT_BURST_LEN(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON0_WEIGHT_BURST_LEN__SHIFT) & CNA_DMA_CON0_WEIGHT_BURST_LEN__MASK;
+}
+#define CNA_DMA_CON0_RESERVED_1__MASK				0x0000fff0
+#define CNA_DMA_CON0_RESERVED_1__SHIFT				4
+static inline uint32_t CNA_DMA_CON0_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON0_RESERVED_1__SHIFT) & CNA_DMA_CON0_RESERVED_1__MASK;
+}
+#define CNA_DMA_CON0_DATA_BURST_LEN__MASK			0x0000000f
+#define CNA_DMA_CON0_DATA_BURST_LEN__SHIFT			0
+static inline uint32_t CNA_DMA_CON0_DATA_BURST_LEN(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON0_DATA_BURST_LEN__SHIFT) & CNA_DMA_CON0_DATA_BURST_LEN__MASK;
+}
+
+#define REG_CNA_DMA_CON1					0x0000107c
+#define CNA_DMA_CON1_RESERVED_0__MASK				0xf0000000
+#define CNA_DMA_CON1_RESERVED_0__SHIFT				28
+static inline uint32_t CNA_DMA_CON1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON1_RESERVED_0__SHIFT) & CNA_DMA_CON1_RESERVED_0__MASK;
+}
+#define CNA_DMA_CON1_LINE_STRIDE__MASK				0x0fffffff
+#define CNA_DMA_CON1_LINE_STRIDE__SHIFT				0
+static inline uint32_t CNA_DMA_CON1_LINE_STRIDE(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON1_LINE_STRIDE__SHIFT) & CNA_DMA_CON1_LINE_STRIDE__MASK;
+}
+
+#define REG_CNA_DMA_CON2					0x00001080
+#define CNA_DMA_CON2_RESERVED_0__MASK				0xf0000000
+#define CNA_DMA_CON2_RESERVED_0__SHIFT				28
+static inline uint32_t CNA_DMA_CON2_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON2_RESERVED_0__SHIFT) & CNA_DMA_CON2_RESERVED_0__MASK;
+}
+#define CNA_DMA_CON2_SURF_STRIDE__MASK				0x0fffffff
+#define CNA_DMA_CON2_SURF_STRIDE__SHIFT				0
+static inline uint32_t CNA_DMA_CON2_SURF_STRIDE(uint32_t val)
+{
+	return ((val) << CNA_DMA_CON2_SURF_STRIDE__SHIFT) & CNA_DMA_CON2_SURF_STRIDE__MASK;
+}
+
+#define REG_CNA_FC_DATA_SIZE0					0x00001084
+#define CNA_FC_DATA_SIZE0_RESERVED_0__MASK			0xc0000000
+#define CNA_FC_DATA_SIZE0_RESERVED_0__SHIFT			30
+static inline uint32_t CNA_FC_DATA_SIZE0_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_FC_DATA_SIZE0_RESERVED_0__SHIFT) & CNA_FC_DATA_SIZE0_RESERVED_0__MASK;
+}
+#define CNA_FC_DATA_SIZE0_DMA_WIDTH__MASK			0x3fff0000
+#define CNA_FC_DATA_SIZE0_DMA_WIDTH__SHIFT			16
+static inline uint32_t CNA_FC_DATA_SIZE0_DMA_WIDTH(uint32_t val)
+{
+	return ((val) << CNA_FC_DATA_SIZE0_DMA_WIDTH__SHIFT) & CNA_FC_DATA_SIZE0_DMA_WIDTH__MASK;
+}
+#define CNA_FC_DATA_SIZE0_RESERVED_1__MASK			0x0000f800
+#define CNA_FC_DATA_SIZE0_RESERVED_1__SHIFT			11
+static inline uint32_t CNA_FC_DATA_SIZE0_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_FC_DATA_SIZE0_RESERVED_1__SHIFT) & CNA_FC_DATA_SIZE0_RESERVED_1__MASK;
+}
+#define CNA_FC_DATA_SIZE0_DMA_HEIGHT__MASK			0x000007ff
+#define CNA_FC_DATA_SIZE0_DMA_HEIGHT__SHIFT			0
+static inline uint32_t CNA_FC_DATA_SIZE0_DMA_HEIGHT(uint32_t val)
+{
+	return ((val) << CNA_FC_DATA_SIZE0_DMA_HEIGHT__SHIFT) & CNA_FC_DATA_SIZE0_DMA_HEIGHT__MASK;
+}
+
+#define REG_CNA_FC_DATA_SIZE1					0x00001088
+#define CNA_FC_DATA_SIZE1_RESERVED_0__MASK			0xffff0000
+#define CNA_FC_DATA_SIZE1_RESERVED_0__SHIFT			16
+static inline uint32_t CNA_FC_DATA_SIZE1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_FC_DATA_SIZE1_RESERVED_0__SHIFT) & CNA_FC_DATA_SIZE1_RESERVED_0__MASK;
+}
+#define CNA_FC_DATA_SIZE1_DMA_CHANNEL__MASK			0x0000ffff
+#define CNA_FC_DATA_SIZE1_DMA_CHANNEL__SHIFT			0
+static inline uint32_t CNA_FC_DATA_SIZE1_DMA_CHANNEL(uint32_t val)
+{
+	return ((val) << CNA_FC_DATA_SIZE1_DMA_CHANNEL__SHIFT) & CNA_FC_DATA_SIZE1_DMA_CHANNEL__MASK;
+}
+
+#define REG_CNA_CLK_GATE					0x00001090
+#define CNA_CLK_GATE_RESERVED_0__MASK				0xffffffe0
+#define CNA_CLK_GATE_RESERVED_0__SHIFT				5
+static inline uint32_t CNA_CLK_GATE_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_CLK_GATE_RESERVED_0__SHIFT) & CNA_CLK_GATE_RESERVED_0__MASK;
+}
+#define CNA_CLK_GATE_CBUF_CS_DISABLE_CLKGATE__MASK		0x00000010
+#define CNA_CLK_GATE_CBUF_CS_DISABLE_CLKGATE__SHIFT		4
+static inline uint32_t CNA_CLK_GATE_CBUF_CS_DISABLE_CLKGATE(uint32_t val)
+{
+	return ((val) << CNA_CLK_GATE_CBUF_CS_DISABLE_CLKGATE__SHIFT) & CNA_CLK_GATE_CBUF_CS_DISABLE_CLKGATE__MASK;
+}
+#define CNA_CLK_GATE_RESERVED_1__MASK				0x00000008
+#define CNA_CLK_GATE_RESERVED_1__SHIFT				3
+static inline uint32_t CNA_CLK_GATE_RESERVED_1(uint32_t val)
+{
+	return ((val) << CNA_CLK_GATE_RESERVED_1__SHIFT) & CNA_CLK_GATE_RESERVED_1__MASK;
+}
+#define CNA_CLK_GATE_CSC_DISABLE_CLKGATE__MASK			0x00000004
+#define CNA_CLK_GATE_CSC_DISABLE_CLKGATE__SHIFT			2
+static inline uint32_t CNA_CLK_GATE_CSC_DISABLE_CLKGATE(uint32_t val)
+{
+	return ((val) << CNA_CLK_GATE_CSC_DISABLE_CLKGATE__SHIFT) & CNA_CLK_GATE_CSC_DISABLE_CLKGATE__MASK;
+}
+#define CNA_CLK_GATE_CNA_WEIGHT_DISABLE_CLKGATE__MASK		0x00000002
+#define CNA_CLK_GATE_CNA_WEIGHT_DISABLE_CLKGATE__SHIFT		1
+static inline uint32_t CNA_CLK_GATE_CNA_WEIGHT_DISABLE_CLKGATE(uint32_t val)
+{
+	return ((val) << CNA_CLK_GATE_CNA_WEIGHT_DISABLE_CLKGATE__SHIFT) & CNA_CLK_GATE_CNA_WEIGHT_DISABLE_CLKGATE__MASK;
+}
+#define CNA_CLK_GATE_CNA_FEATURE_DISABLE_CLKGATE__MASK		0x00000001
+#define CNA_CLK_GATE_CNA_FEATURE_DISABLE_CLKGATE__SHIFT		0
+static inline uint32_t CNA_CLK_GATE_CNA_FEATURE_DISABLE_CLKGATE(uint32_t val)
+{
+	return ((val) << CNA_CLK_GATE_CNA_FEATURE_DISABLE_CLKGATE__SHIFT) & CNA_CLK_GATE_CNA_FEATURE_DISABLE_CLKGATE__MASK;
+}
+
+#define REG_CNA_DCOMP_CTRL					0x00001100
+#define CNA_DCOMP_CTRL_RESERVED_0__MASK				0xfffffff0
+#define CNA_DCOMP_CTRL_RESERVED_0__SHIFT			4
+static inline uint32_t CNA_DCOMP_CTRL_RESERVED_0(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_CTRL_RESERVED_0__SHIFT) & CNA_DCOMP_CTRL_RESERVED_0__MASK;
+}
+#define CNA_DCOMP_CTRL_WT_DEC_BYPASS__MASK			0x00000008
+#define CNA_DCOMP_CTRL_WT_DEC_BYPASS__SHIFT			3
+static inline uint32_t CNA_DCOMP_CTRL_WT_DEC_BYPASS(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_CTRL_WT_DEC_BYPASS__SHIFT) & CNA_DCOMP_CTRL_WT_DEC_BYPASS__MASK;
+}
+#define CNA_DCOMP_CTRL_DECOMP_CONTROL__MASK			0x00000007
+#define CNA_DCOMP_CTRL_DECOMP_CONTROL__SHIFT			0
+static inline uint32_t CNA_DCOMP_CTRL_DECOMP_CONTROL(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_CTRL_DECOMP_CONTROL__SHIFT) & CNA_DCOMP_CTRL_DECOMP_CONTROL__MASK;
+}
+
+#define REG_CNA_DCOMP_REGNUM					0x00001104
+#define CNA_DCOMP_REGNUM_DCOMP_REGNUM__MASK			0xffffffff
+#define CNA_DCOMP_REGNUM_DCOMP_REGNUM__SHIFT			0
+static inline uint32_t CNA_DCOMP_REGNUM_DCOMP_REGNUM(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_REGNUM_DCOMP_REGNUM__SHIFT) & CNA_DCOMP_REGNUM_DCOMP_REGNUM__MASK;
+}
+
+#define REG_CNA_DCOMP_ADDR0					0x00001110
+#define CNA_DCOMP_ADDR0_DECOMPRESS_ADDR0__MASK			0xffffffff
+#define CNA_DCOMP_ADDR0_DECOMPRESS_ADDR0__SHIFT			0
+static inline uint32_t CNA_DCOMP_ADDR0_DECOMPRESS_ADDR0(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_ADDR0_DECOMPRESS_ADDR0__SHIFT) & CNA_DCOMP_ADDR0_DECOMPRESS_ADDR0__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT0					0x00001140
+#define CNA_DCOMP_AMOUNT0_DCOMP_AMOUNT0__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT0_DCOMP_AMOUNT0__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT0_DCOMP_AMOUNT0(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT0_DCOMP_AMOUNT0__SHIFT) & CNA_DCOMP_AMOUNT0_DCOMP_AMOUNT0__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT1					0x00001144
+#define CNA_DCOMP_AMOUNT1_DCOMP_AMOUNT1__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT1_DCOMP_AMOUNT1__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT1_DCOMP_AMOUNT1(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT1_DCOMP_AMOUNT1__SHIFT) & CNA_DCOMP_AMOUNT1_DCOMP_AMOUNT1__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT2					0x00001148
+#define CNA_DCOMP_AMOUNT2_DCOMP_AMOUNT2__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT2_DCOMP_AMOUNT2__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT2_DCOMP_AMOUNT2(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT2_DCOMP_AMOUNT2__SHIFT) & CNA_DCOMP_AMOUNT2_DCOMP_AMOUNT2__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT3					0x0000114c
+#define CNA_DCOMP_AMOUNT3_DCOMP_AMOUNT3__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT3_DCOMP_AMOUNT3__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT3_DCOMP_AMOUNT3(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT3_DCOMP_AMOUNT3__SHIFT) & CNA_DCOMP_AMOUNT3_DCOMP_AMOUNT3__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT4					0x00001150
+#define CNA_DCOMP_AMOUNT4_DCOMP_AMOUNT4__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT4_DCOMP_AMOUNT4__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT4_DCOMP_AMOUNT4(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT4_DCOMP_AMOUNT4__SHIFT) & CNA_DCOMP_AMOUNT4_DCOMP_AMOUNT4__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT5					0x00001154
+#define CNA_DCOMP_AMOUNT5_DCOMP_AMOUNT5__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT5_DCOMP_AMOUNT5__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT5_DCOMP_AMOUNT5(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT5_DCOMP_AMOUNT5__SHIFT) & CNA_DCOMP_AMOUNT5_DCOMP_AMOUNT5__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT6					0x00001158
+#define CNA_DCOMP_AMOUNT6_DCOMP_AMOUNT6__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT6_DCOMP_AMOUNT6__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT6_DCOMP_AMOUNT6(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT6_DCOMP_AMOUNT6__SHIFT) & CNA_DCOMP_AMOUNT6_DCOMP_AMOUNT6__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT7					0x0000115c
+#define CNA_DCOMP_AMOUNT7_DCOMP_AMOUNT7__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT7_DCOMP_AMOUNT7__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT7_DCOMP_AMOUNT7(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT7_DCOMP_AMOUNT7__SHIFT) & CNA_DCOMP_AMOUNT7_DCOMP_AMOUNT7__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT8					0x00001160
+#define CNA_DCOMP_AMOUNT8_DCOMP_AMOUNT8__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT8_DCOMP_AMOUNT8__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT8_DCOMP_AMOUNT8(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT8_DCOMP_AMOUNT8__SHIFT) & CNA_DCOMP_AMOUNT8_DCOMP_AMOUNT8__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT9					0x00001164
+#define CNA_DCOMP_AMOUNT9_DCOMP_AMOUNT9__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT9_DCOMP_AMOUNT9__SHIFT			0
+static inline uint32_t CNA_DCOMP_AMOUNT9_DCOMP_AMOUNT9(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT9_DCOMP_AMOUNT9__SHIFT) & CNA_DCOMP_AMOUNT9_DCOMP_AMOUNT9__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT10					0x00001168
+#define CNA_DCOMP_AMOUNT10_DCOMP_AMOUNT10__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT10_DCOMP_AMOUNT10__SHIFT		0
+static inline uint32_t CNA_DCOMP_AMOUNT10_DCOMP_AMOUNT10(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT10_DCOMP_AMOUNT10__SHIFT) & CNA_DCOMP_AMOUNT10_DCOMP_AMOUNT10__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT11					0x0000116c
+#define CNA_DCOMP_AMOUNT11_DCOMP_AMOUNT11__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT11_DCOMP_AMOUNT11__SHIFT		0
+static inline uint32_t CNA_DCOMP_AMOUNT11_DCOMP_AMOUNT11(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT11_DCOMP_AMOUNT11__SHIFT) & CNA_DCOMP_AMOUNT11_DCOMP_AMOUNT11__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT12					0x00001170
+#define CNA_DCOMP_AMOUNT12_DCOMP_AMOUNT12__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT12_DCOMP_AMOUNT12__SHIFT		0
+static inline uint32_t CNA_DCOMP_AMOUNT12_DCOMP_AMOUNT12(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT12_DCOMP_AMOUNT12__SHIFT) & CNA_DCOMP_AMOUNT12_DCOMP_AMOUNT12__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT13					0x00001174
+#define CNA_DCOMP_AMOUNT13_DCOMP_AMOUNT13__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT13_DCOMP_AMOUNT13__SHIFT		0
+static inline uint32_t CNA_DCOMP_AMOUNT13_DCOMP_AMOUNT13(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT13_DCOMP_AMOUNT13__SHIFT) & CNA_DCOMP_AMOUNT13_DCOMP_AMOUNT13__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT14					0x00001178
+#define CNA_DCOMP_AMOUNT14_DCOMP_AMOUNT14__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT14_DCOMP_AMOUNT14__SHIFT		0
+static inline uint32_t CNA_DCOMP_AMOUNT14_DCOMP_AMOUNT14(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT14_DCOMP_AMOUNT14__SHIFT) & CNA_DCOMP_AMOUNT14_DCOMP_AMOUNT14__MASK;
+}
+
+#define REG_CNA_DCOMP_AMOUNT15					0x0000117c
+#define CNA_DCOMP_AMOUNT15_DCOMP_AMOUNT15__MASK			0xffffffff
+#define CNA_DCOMP_AMOUNT15_DCOMP_AMOUNT15__SHIFT		0
+static inline uint32_t CNA_DCOMP_AMOUNT15_DCOMP_AMOUNT15(uint32_t val)
+{
+	return ((val) << CNA_DCOMP_AMOUNT15_DCOMP_AMOUNT15__SHIFT) & CNA_DCOMP_AMOUNT15_DCOMP_AMOUNT15__MASK;
+}
+
+#define REG_CNA_CVT_CON5					0x00001180
+#define CNA_CVT_CON5_PER_CHANNEL_CVT_EN__MASK			0xffffffff
+#define CNA_CVT_CON5_PER_CHANNEL_CVT_EN__SHIFT			0
+static inline uint32_t CNA_CVT_CON5_PER_CHANNEL_CVT_EN(uint32_t val)
+{
+	return ((val) << CNA_CVT_CON5_PER_CHANNEL_CVT_EN__SHIFT) & CNA_CVT_CON5_PER_CHANNEL_CVT_EN__MASK;
+}
+
+#define REG_CNA_PAD_CON1					0x00001184
+#define CNA_PAD_CON1_PAD_VALUE__MASK				0xffffffff
+#define CNA_PAD_CON1_PAD_VALUE__SHIFT				0
+static inline uint32_t CNA_PAD_CON1_PAD_VALUE(uint32_t val)
+{
+	return ((val) << CNA_PAD_CON1_PAD_VALUE__SHIFT) & CNA_PAD_CON1_PAD_VALUE__MASK;
+}
+
+#define REG_CORE_S_STATUS					0x00003000
+#define CORE_S_STATUS_RESERVED_0__MASK				0xfffc0000
+#define CORE_S_STATUS_RESERVED_0__SHIFT				18
+static inline uint32_t CORE_S_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_S_STATUS_RESERVED_0__SHIFT) & CORE_S_STATUS_RESERVED_0__MASK;
+}
+#define CORE_S_STATUS_STATUS_1__MASK				0x00030000
+#define CORE_S_STATUS_STATUS_1__SHIFT				16
+static inline uint32_t CORE_S_STATUS_STATUS_1(uint32_t val)
+{
+	return ((val) << CORE_S_STATUS_STATUS_1__SHIFT) & CORE_S_STATUS_STATUS_1__MASK;
+}
+#define CORE_S_STATUS_RESERVED_1__MASK				0x0000fffc
+#define CORE_S_STATUS_RESERVED_1__SHIFT				2
+static inline uint32_t CORE_S_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << CORE_S_STATUS_RESERVED_1__SHIFT) & CORE_S_STATUS_RESERVED_1__MASK;
+}
+#define CORE_S_STATUS_STATUS_0__MASK				0x00000003
+#define CORE_S_STATUS_STATUS_0__SHIFT				0
+static inline uint32_t CORE_S_STATUS_STATUS_0(uint32_t val)
+{
+	return ((val) << CORE_S_STATUS_STATUS_0__SHIFT) & CORE_S_STATUS_STATUS_0__MASK;
+}
+
+#define REG_CORE_S_POINTER					0x00003004
+#define CORE_S_POINTER_RESERVED_0__MASK				0xfffe0000
+#define CORE_S_POINTER_RESERVED_0__SHIFT			17
+static inline uint32_t CORE_S_POINTER_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_RESERVED_0__SHIFT) & CORE_S_POINTER_RESERVED_0__MASK;
+}
+#define CORE_S_POINTER_EXECUTER__MASK				0x00010000
+#define CORE_S_POINTER_EXECUTER__SHIFT				16
+static inline uint32_t CORE_S_POINTER_EXECUTER(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_EXECUTER__SHIFT) & CORE_S_POINTER_EXECUTER__MASK;
+}
+#define CORE_S_POINTER_RESERVED_1__MASK				0x0000ffc0
+#define CORE_S_POINTER_RESERVED_1__SHIFT			6
+static inline uint32_t CORE_S_POINTER_RESERVED_1(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_RESERVED_1__SHIFT) & CORE_S_POINTER_RESERVED_1__MASK;
+}
+#define CORE_S_POINTER_EXECUTER_PP_CLEAR__MASK			0x00000020
+#define CORE_S_POINTER_EXECUTER_PP_CLEAR__SHIFT			5
+static inline uint32_t CORE_S_POINTER_EXECUTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_EXECUTER_PP_CLEAR__SHIFT) & CORE_S_POINTER_EXECUTER_PP_CLEAR__MASK;
+}
+#define CORE_S_POINTER_POINTER_PP_CLEAR__MASK			0x00000010
+#define CORE_S_POINTER_POINTER_PP_CLEAR__SHIFT			4
+static inline uint32_t CORE_S_POINTER_POINTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_POINTER_PP_CLEAR__SHIFT) & CORE_S_POINTER_POINTER_PP_CLEAR__MASK;
+}
+#define CORE_S_POINTER_POINTER_PP_MODE__MASK			0x00000008
+#define CORE_S_POINTER_POINTER_PP_MODE__SHIFT			3
+static inline uint32_t CORE_S_POINTER_POINTER_PP_MODE(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_POINTER_PP_MODE__SHIFT) & CORE_S_POINTER_POINTER_PP_MODE__MASK;
+}
+#define CORE_S_POINTER_EXECUTER_PP_EN__MASK			0x00000004
+#define CORE_S_POINTER_EXECUTER_PP_EN__SHIFT			2
+static inline uint32_t CORE_S_POINTER_EXECUTER_PP_EN(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_EXECUTER_PP_EN__SHIFT) & CORE_S_POINTER_EXECUTER_PP_EN__MASK;
+}
+#define CORE_S_POINTER_POINTER_PP_EN__MASK			0x00000002
+#define CORE_S_POINTER_POINTER_PP_EN__SHIFT			1
+static inline uint32_t CORE_S_POINTER_POINTER_PP_EN(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_POINTER_PP_EN__SHIFT) & CORE_S_POINTER_POINTER_PP_EN__MASK;
+}
+#define CORE_S_POINTER_POINTER__MASK				0x00000001
+#define CORE_S_POINTER_POINTER__SHIFT				0
+static inline uint32_t CORE_S_POINTER_POINTER(uint32_t val)
+{
+	return ((val) << CORE_S_POINTER_POINTER__SHIFT) & CORE_S_POINTER_POINTER__MASK;
+}
+
+#define REG_CORE_OPERATION_ENABLE				0x00003008
+#define CORE_OPERATION_ENABLE_RESERVED_0__MASK			0xfffffffe
+#define CORE_OPERATION_ENABLE_RESERVED_0__SHIFT			1
+static inline uint32_t CORE_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_OPERATION_ENABLE_RESERVED_0__SHIFT) & CORE_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define CORE_OPERATION_ENABLE_OP_EN__MASK			0x00000001
+#define CORE_OPERATION_ENABLE_OP_EN__SHIFT			0
+static inline uint32_t CORE_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << CORE_OPERATION_ENABLE_OP_EN__SHIFT) & CORE_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_CORE_MAC_GATING					0x0000300c
+#define CORE_MAC_GATING_RESERVED_0__MASK			0xf8000000
+#define CORE_MAC_GATING_RESERVED_0__SHIFT			27
+static inline uint32_t CORE_MAC_GATING_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_MAC_GATING_RESERVED_0__SHIFT) & CORE_MAC_GATING_RESERVED_0__MASK;
+}
+#define CORE_MAC_GATING_SLCG_OP_EN__MASK			0x07ffffff
+#define CORE_MAC_GATING_SLCG_OP_EN__SHIFT			0
+static inline uint32_t CORE_MAC_GATING_SLCG_OP_EN(uint32_t val)
+{
+	return ((val) << CORE_MAC_GATING_SLCG_OP_EN__SHIFT) & CORE_MAC_GATING_SLCG_OP_EN__MASK;
+}
+
+#define REG_CORE_MISC_CFG					0x00003010
+#define CORE_MISC_CFG_RESERVED_0__MASK				0xfff00000
+#define CORE_MISC_CFG_RESERVED_0__SHIFT				20
+static inline uint32_t CORE_MISC_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_RESERVED_0__SHIFT) & CORE_MISC_CFG_RESERVED_0__MASK;
+}
+#define CORE_MISC_CFG_SOFT_GATING__MASK				0x000fc000
+#define CORE_MISC_CFG_SOFT_GATING__SHIFT			14
+static inline uint32_t CORE_MISC_CFG_SOFT_GATING(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_SOFT_GATING__SHIFT) & CORE_MISC_CFG_SOFT_GATING__MASK;
+}
+#define CORE_MISC_CFG_RESERVED_1__MASK				0x00003800
+#define CORE_MISC_CFG_RESERVED_1__SHIFT				11
+static inline uint32_t CORE_MISC_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_RESERVED_1__SHIFT) & CORE_MISC_CFG_RESERVED_1__MASK;
+}
+#define CORE_MISC_CFG_PROC_PRECISION__MASK			0x00000700
+#define CORE_MISC_CFG_PROC_PRECISION__SHIFT			8
+static inline uint32_t CORE_MISC_CFG_PROC_PRECISION(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_PROC_PRECISION__SHIFT) & CORE_MISC_CFG_PROC_PRECISION__MASK;
+}
+#define CORE_MISC_CFG_RESERVED_2__MASK				0x000000fc
+#define CORE_MISC_CFG_RESERVED_2__SHIFT				2
+static inline uint32_t CORE_MISC_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_RESERVED_2__SHIFT) & CORE_MISC_CFG_RESERVED_2__MASK;
+}
+#define CORE_MISC_CFG_DW_EN__MASK				0x00000002
+#define CORE_MISC_CFG_DW_EN__SHIFT				1
+static inline uint32_t CORE_MISC_CFG_DW_EN(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_DW_EN__SHIFT) & CORE_MISC_CFG_DW_EN__MASK;
+}
+#define CORE_MISC_CFG_QD_EN__MASK				0x00000001
+#define CORE_MISC_CFG_QD_EN__SHIFT				0
+static inline uint32_t CORE_MISC_CFG_QD_EN(uint32_t val)
+{
+	return ((val) << CORE_MISC_CFG_QD_EN__SHIFT) & CORE_MISC_CFG_QD_EN__MASK;
+}
+
+#define REG_CORE_DATAOUT_SIZE_0					0x00003014
+#define CORE_DATAOUT_SIZE_0_DATAOUT_HEIGHT__MASK		0xffff0000
+#define CORE_DATAOUT_SIZE_0_DATAOUT_HEIGHT__SHIFT		16
+static inline uint32_t CORE_DATAOUT_SIZE_0_DATAOUT_HEIGHT(uint32_t val)
+{
+	return ((val) << CORE_DATAOUT_SIZE_0_DATAOUT_HEIGHT__SHIFT) & CORE_DATAOUT_SIZE_0_DATAOUT_HEIGHT__MASK;
+}
+#define CORE_DATAOUT_SIZE_0_DATAOUT_WIDTH__MASK			0x0000ffff
+#define CORE_DATAOUT_SIZE_0_DATAOUT_WIDTH__SHIFT		0
+static inline uint32_t CORE_DATAOUT_SIZE_0_DATAOUT_WIDTH(uint32_t val)
+{
+	return ((val) << CORE_DATAOUT_SIZE_0_DATAOUT_WIDTH__SHIFT) & CORE_DATAOUT_SIZE_0_DATAOUT_WIDTH__MASK;
+}
+
+#define REG_CORE_DATAOUT_SIZE_1					0x00003018
+#define CORE_DATAOUT_SIZE_1_RESERVED_0__MASK			0xffff0000
+#define CORE_DATAOUT_SIZE_1_RESERVED_0__SHIFT			16
+static inline uint32_t CORE_DATAOUT_SIZE_1_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_DATAOUT_SIZE_1_RESERVED_0__SHIFT) & CORE_DATAOUT_SIZE_1_RESERVED_0__MASK;
+}
+#define CORE_DATAOUT_SIZE_1_DATAOUT_CHANNEL__MASK		0x0000ffff
+#define CORE_DATAOUT_SIZE_1_DATAOUT_CHANNEL__SHIFT		0
+static inline uint32_t CORE_DATAOUT_SIZE_1_DATAOUT_CHANNEL(uint32_t val)
+{
+	return ((val) << CORE_DATAOUT_SIZE_1_DATAOUT_CHANNEL__SHIFT) & CORE_DATAOUT_SIZE_1_DATAOUT_CHANNEL__MASK;
+}
+
+#define REG_CORE_CLIP_TRUNCATE					0x0000301c
+#define CORE_CLIP_TRUNCATE_RESERVED_0__MASK			0xffffff80
+#define CORE_CLIP_TRUNCATE_RESERVED_0__SHIFT			7
+static inline uint32_t CORE_CLIP_TRUNCATE_RESERVED_0(uint32_t val)
+{
+	return ((val) << CORE_CLIP_TRUNCATE_RESERVED_0__SHIFT) & CORE_CLIP_TRUNCATE_RESERVED_0__MASK;
+}
+#define CORE_CLIP_TRUNCATE_ROUND_TYPE__MASK			0x00000040
+#define CORE_CLIP_TRUNCATE_ROUND_TYPE__SHIFT			6
+static inline uint32_t CORE_CLIP_TRUNCATE_ROUND_TYPE(uint32_t val)
+{
+	return ((val) << CORE_CLIP_TRUNCATE_ROUND_TYPE__SHIFT) & CORE_CLIP_TRUNCATE_ROUND_TYPE__MASK;
+}
+#define CORE_CLIP_TRUNCATE_RESERVED_1__MASK			0x00000020
+#define CORE_CLIP_TRUNCATE_RESERVED_1__SHIFT			5
+static inline uint32_t CORE_CLIP_TRUNCATE_RESERVED_1(uint32_t val)
+{
+	return ((val) << CORE_CLIP_TRUNCATE_RESERVED_1__SHIFT) & CORE_CLIP_TRUNCATE_RESERVED_1__MASK;
+}
+#define CORE_CLIP_TRUNCATE_CLIP_TRUNCATE__MASK			0x0000001f
+#define CORE_CLIP_TRUNCATE_CLIP_TRUNCATE__SHIFT			0
+static inline uint32_t CORE_CLIP_TRUNCATE_CLIP_TRUNCATE(uint32_t val)
+{
+	return ((val) << CORE_CLIP_TRUNCATE_CLIP_TRUNCATE__SHIFT) & CORE_CLIP_TRUNCATE_CLIP_TRUNCATE__MASK;
+}
+
+#define REG_DPU_S_STATUS					0x00004000
+#define DPU_S_STATUS_RESERVED_0__MASK				0xfffc0000
+#define DPU_S_STATUS_RESERVED_0__SHIFT				18
+static inline uint32_t DPU_S_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_S_STATUS_RESERVED_0__SHIFT) & DPU_S_STATUS_RESERVED_0__MASK;
+}
+#define DPU_S_STATUS_STATUS_1__MASK				0x00030000
+#define DPU_S_STATUS_STATUS_1__SHIFT				16
+static inline uint32_t DPU_S_STATUS_STATUS_1(uint32_t val)
+{
+	return ((val) << DPU_S_STATUS_STATUS_1__SHIFT) & DPU_S_STATUS_STATUS_1__MASK;
+}
+#define DPU_S_STATUS_RESERVED_1__MASK				0x0000fffc
+#define DPU_S_STATUS_RESERVED_1__SHIFT				2
+static inline uint32_t DPU_S_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_S_STATUS_RESERVED_1__SHIFT) & DPU_S_STATUS_RESERVED_1__MASK;
+}
+#define DPU_S_STATUS_STATUS_0__MASK				0x00000003
+#define DPU_S_STATUS_STATUS_0__SHIFT				0
+static inline uint32_t DPU_S_STATUS_STATUS_0(uint32_t val)
+{
+	return ((val) << DPU_S_STATUS_STATUS_0__SHIFT) & DPU_S_STATUS_STATUS_0__MASK;
+}
+
+#define REG_DPU_S_POINTER					0x00004004
+#define DPU_S_POINTER_RESERVED_0__MASK				0xfffe0000
+#define DPU_S_POINTER_RESERVED_0__SHIFT				17
+static inline uint32_t DPU_S_POINTER_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_RESERVED_0__SHIFT) & DPU_S_POINTER_RESERVED_0__MASK;
+}
+#define DPU_S_POINTER_EXECUTER__MASK				0x00010000
+#define DPU_S_POINTER_EXECUTER__SHIFT				16
+static inline uint32_t DPU_S_POINTER_EXECUTER(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_EXECUTER__SHIFT) & DPU_S_POINTER_EXECUTER__MASK;
+}
+#define DPU_S_POINTER_RESERVED_1__MASK				0x0000ffc0
+#define DPU_S_POINTER_RESERVED_1__SHIFT				6
+static inline uint32_t DPU_S_POINTER_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_RESERVED_1__SHIFT) & DPU_S_POINTER_RESERVED_1__MASK;
+}
+#define DPU_S_POINTER_EXECUTER_PP_CLEAR__MASK			0x00000020
+#define DPU_S_POINTER_EXECUTER_PP_CLEAR__SHIFT			5
+static inline uint32_t DPU_S_POINTER_EXECUTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_EXECUTER_PP_CLEAR__SHIFT) & DPU_S_POINTER_EXECUTER_PP_CLEAR__MASK;
+}
+#define DPU_S_POINTER_POINTER_PP_CLEAR__MASK			0x00000010
+#define DPU_S_POINTER_POINTER_PP_CLEAR__SHIFT			4
+static inline uint32_t DPU_S_POINTER_POINTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_POINTER_PP_CLEAR__SHIFT) & DPU_S_POINTER_POINTER_PP_CLEAR__MASK;
+}
+#define DPU_S_POINTER_POINTER_PP_MODE__MASK			0x00000008
+#define DPU_S_POINTER_POINTER_PP_MODE__SHIFT			3
+static inline uint32_t DPU_S_POINTER_POINTER_PP_MODE(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_POINTER_PP_MODE__SHIFT) & DPU_S_POINTER_POINTER_PP_MODE__MASK;
+}
+#define DPU_S_POINTER_EXECUTER_PP_EN__MASK			0x00000004
+#define DPU_S_POINTER_EXECUTER_PP_EN__SHIFT			2
+static inline uint32_t DPU_S_POINTER_EXECUTER_PP_EN(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_EXECUTER_PP_EN__SHIFT) & DPU_S_POINTER_EXECUTER_PP_EN__MASK;
+}
+#define DPU_S_POINTER_POINTER_PP_EN__MASK			0x00000002
+#define DPU_S_POINTER_POINTER_PP_EN__SHIFT			1
+static inline uint32_t DPU_S_POINTER_POINTER_PP_EN(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_POINTER_PP_EN__SHIFT) & DPU_S_POINTER_POINTER_PP_EN__MASK;
+}
+#define DPU_S_POINTER_POINTER__MASK				0x00000001
+#define DPU_S_POINTER_POINTER__SHIFT				0
+static inline uint32_t DPU_S_POINTER_POINTER(uint32_t val)
+{
+	return ((val) << DPU_S_POINTER_POINTER__SHIFT) & DPU_S_POINTER_POINTER__MASK;
+}
+
+#define REG_DPU_OPERATION_ENABLE				0x00004008
+#define DPU_OPERATION_ENABLE_RESERVED_0__MASK			0xfffffffe
+#define DPU_OPERATION_ENABLE_RESERVED_0__SHIFT			1
+static inline uint32_t DPU_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_OPERATION_ENABLE_RESERVED_0__SHIFT) & DPU_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define DPU_OPERATION_ENABLE_OP_EN__MASK			0x00000001
+#define DPU_OPERATION_ENABLE_OP_EN__SHIFT			0
+static inline uint32_t DPU_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << DPU_OPERATION_ENABLE_OP_EN__SHIFT) & DPU_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_DPU_FEATURE_MODE_CFG				0x0000400c
+#define DPU_FEATURE_MODE_CFG_COMB_USE__MASK			0x80000000
+#define DPU_FEATURE_MODE_CFG_COMB_USE__SHIFT			31
+static inline uint32_t DPU_FEATURE_MODE_CFG_COMB_USE(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_COMB_USE__SHIFT) & DPU_FEATURE_MODE_CFG_COMB_USE__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_TP_EN__MASK			0x40000000
+#define DPU_FEATURE_MODE_CFG_TP_EN__SHIFT			30
+static inline uint32_t DPU_FEATURE_MODE_CFG_TP_EN(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_TP_EN__SHIFT) & DPU_FEATURE_MODE_CFG_TP_EN__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_RGP_TYPE__MASK			0x3c000000
+#define DPU_FEATURE_MODE_CFG_RGP_TYPE__SHIFT			26
+static inline uint32_t DPU_FEATURE_MODE_CFG_RGP_TYPE(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_RGP_TYPE__SHIFT) & DPU_FEATURE_MODE_CFG_RGP_TYPE__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_NONALIGN__MASK			0x02000000
+#define DPU_FEATURE_MODE_CFG_NONALIGN__SHIFT			25
+static inline uint32_t DPU_FEATURE_MODE_CFG_NONALIGN(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_NONALIGN__SHIFT) & DPU_FEATURE_MODE_CFG_NONALIGN__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_SURF_LEN__MASK			0x01fffe00
+#define DPU_FEATURE_MODE_CFG_SURF_LEN__SHIFT			9
+static inline uint32_t DPU_FEATURE_MODE_CFG_SURF_LEN(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_SURF_LEN__SHIFT) & DPU_FEATURE_MODE_CFG_SURF_LEN__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_BURST_LEN__MASK			0x000001e0
+#define DPU_FEATURE_MODE_CFG_BURST_LEN__SHIFT			5
+static inline uint32_t DPU_FEATURE_MODE_CFG_BURST_LEN(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_BURST_LEN__SHIFT) & DPU_FEATURE_MODE_CFG_BURST_LEN__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_CONV_MODE__MASK			0x00000018
+#define DPU_FEATURE_MODE_CFG_CONV_MODE__SHIFT			3
+static inline uint32_t DPU_FEATURE_MODE_CFG_CONV_MODE(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_CONV_MODE__SHIFT) & DPU_FEATURE_MODE_CFG_CONV_MODE__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_OUTPUT_MODE__MASK			0x00000006
+#define DPU_FEATURE_MODE_CFG_OUTPUT_MODE__SHIFT			1
+static inline uint32_t DPU_FEATURE_MODE_CFG_OUTPUT_MODE(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_OUTPUT_MODE__SHIFT) & DPU_FEATURE_MODE_CFG_OUTPUT_MODE__MASK;
+}
+#define DPU_FEATURE_MODE_CFG_FLYING_MODE__MASK			0x00000001
+#define DPU_FEATURE_MODE_CFG_FLYING_MODE__SHIFT			0
+static inline uint32_t DPU_FEATURE_MODE_CFG_FLYING_MODE(uint32_t val)
+{
+	return ((val) << DPU_FEATURE_MODE_CFG_FLYING_MODE__SHIFT) & DPU_FEATURE_MODE_CFG_FLYING_MODE__MASK;
+}
+
+#define REG_DPU_DATA_FORMAT					0x00004010
+#define DPU_DATA_FORMAT_OUT_PRECISION__MASK			0xe0000000
+#define DPU_DATA_FORMAT_OUT_PRECISION__SHIFT			29
+static inline uint32_t DPU_DATA_FORMAT_OUT_PRECISION(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_OUT_PRECISION__SHIFT) & DPU_DATA_FORMAT_OUT_PRECISION__MASK;
+}
+#define DPU_DATA_FORMAT_IN_PRECISION__MASK			0x1c000000
+#define DPU_DATA_FORMAT_IN_PRECISION__SHIFT			26
+static inline uint32_t DPU_DATA_FORMAT_IN_PRECISION(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_IN_PRECISION__SHIFT) & DPU_DATA_FORMAT_IN_PRECISION__MASK;
+}
+#define DPU_DATA_FORMAT_EW_TRUNCATE_NEG__MASK			0x03ff0000
+#define DPU_DATA_FORMAT_EW_TRUNCATE_NEG__SHIFT			16
+static inline uint32_t DPU_DATA_FORMAT_EW_TRUNCATE_NEG(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_EW_TRUNCATE_NEG__SHIFT) & DPU_DATA_FORMAT_EW_TRUNCATE_NEG__MASK;
+}
+#define DPU_DATA_FORMAT_BN_MUL_SHIFT_VALUE_NEG__MASK		0x0000fc00
+#define DPU_DATA_FORMAT_BN_MUL_SHIFT_VALUE_NEG__SHIFT		10
+static inline uint32_t DPU_DATA_FORMAT_BN_MUL_SHIFT_VALUE_NEG(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_BN_MUL_SHIFT_VALUE_NEG__SHIFT) & DPU_DATA_FORMAT_BN_MUL_SHIFT_VALUE_NEG__MASK;
+}
+#define DPU_DATA_FORMAT_BS_MUL_SHIFT_VALUE_NEG__MASK		0x000003f0
+#define DPU_DATA_FORMAT_BS_MUL_SHIFT_VALUE_NEG__SHIFT		4
+static inline uint32_t DPU_DATA_FORMAT_BS_MUL_SHIFT_VALUE_NEG(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_BS_MUL_SHIFT_VALUE_NEG__SHIFT) & DPU_DATA_FORMAT_BS_MUL_SHIFT_VALUE_NEG__MASK;
+}
+#define DPU_DATA_FORMAT_MC_SURF_OUT__MASK			0x00000008
+#define DPU_DATA_FORMAT_MC_SURF_OUT__SHIFT			3
+static inline uint32_t DPU_DATA_FORMAT_MC_SURF_OUT(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_MC_SURF_OUT__SHIFT) & DPU_DATA_FORMAT_MC_SURF_OUT__MASK;
+}
+#define DPU_DATA_FORMAT_PROC_PRECISION__MASK			0x00000007
+#define DPU_DATA_FORMAT_PROC_PRECISION__SHIFT			0
+static inline uint32_t DPU_DATA_FORMAT_PROC_PRECISION(uint32_t val)
+{
+	return ((val) << DPU_DATA_FORMAT_PROC_PRECISION__SHIFT) & DPU_DATA_FORMAT_PROC_PRECISION__MASK;
+}
+
+#define REG_DPU_OFFSET_PEND					0x00004014
+#define DPU_OFFSET_PEND_RESERVED_0__MASK			0xffff0000
+#define DPU_OFFSET_PEND_RESERVED_0__SHIFT			16
+static inline uint32_t DPU_OFFSET_PEND_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_OFFSET_PEND_RESERVED_0__SHIFT) & DPU_OFFSET_PEND_RESERVED_0__MASK;
+}
+#define DPU_OFFSET_PEND_OFFSET_PEND__MASK			0x0000ffff
+#define DPU_OFFSET_PEND_OFFSET_PEND__SHIFT			0
+static inline uint32_t DPU_OFFSET_PEND_OFFSET_PEND(uint32_t val)
+{
+	return ((val) << DPU_OFFSET_PEND_OFFSET_PEND__SHIFT) & DPU_OFFSET_PEND_OFFSET_PEND__MASK;
+}
+
+#define REG_DPU_DST_BASE_ADDR					0x00004020
+#define DPU_DST_BASE_ADDR_DST_BASE_ADDR__MASK			0xffffffff
+#define DPU_DST_BASE_ADDR_DST_BASE_ADDR__SHIFT			0
+static inline uint32_t DPU_DST_BASE_ADDR_DST_BASE_ADDR(uint32_t val)
+{
+	return ((val) << DPU_DST_BASE_ADDR_DST_BASE_ADDR__SHIFT) & DPU_DST_BASE_ADDR_DST_BASE_ADDR__MASK;
+}
+
+#define REG_DPU_DST_SURF_STRIDE					0x00004024
+#define DPU_DST_SURF_STRIDE_DST_SURF_STRIDE__MASK		0xfffffff0
+#define DPU_DST_SURF_STRIDE_DST_SURF_STRIDE__SHIFT		4
+static inline uint32_t DPU_DST_SURF_STRIDE_DST_SURF_STRIDE(uint32_t val)
+{
+	return ((val) << DPU_DST_SURF_STRIDE_DST_SURF_STRIDE__SHIFT) & DPU_DST_SURF_STRIDE_DST_SURF_STRIDE__MASK;
+}
+#define DPU_DST_SURF_STRIDE_RESERVED_0__MASK			0x0000000f
+#define DPU_DST_SURF_STRIDE_RESERVED_0__SHIFT			0
+static inline uint32_t DPU_DST_SURF_STRIDE_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_DST_SURF_STRIDE_RESERVED_0__SHIFT) & DPU_DST_SURF_STRIDE_RESERVED_0__MASK;
+}
+
+#define REG_DPU_DATA_CUBE_WIDTH					0x00004030
+#define DPU_DATA_CUBE_WIDTH_RESERVED_0__MASK			0xffffe000
+#define DPU_DATA_CUBE_WIDTH_RESERVED_0__SHIFT			13
+static inline uint32_t DPU_DATA_CUBE_WIDTH_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_WIDTH_RESERVED_0__SHIFT) & DPU_DATA_CUBE_WIDTH_RESERVED_0__MASK;
+}
+#define DPU_DATA_CUBE_WIDTH_WIDTH__MASK				0x00001fff
+#define DPU_DATA_CUBE_WIDTH_WIDTH__SHIFT			0
+static inline uint32_t DPU_DATA_CUBE_WIDTH_WIDTH(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_WIDTH_WIDTH__SHIFT) & DPU_DATA_CUBE_WIDTH_WIDTH__MASK;
+}
+
+#define REG_DPU_DATA_CUBE_HEIGHT				0x00004034
+#define DPU_DATA_CUBE_HEIGHT_RESERVED_0__MASK			0xfe000000
+#define DPU_DATA_CUBE_HEIGHT_RESERVED_0__SHIFT			25
+static inline uint32_t DPU_DATA_CUBE_HEIGHT_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_HEIGHT_RESERVED_0__SHIFT) & DPU_DATA_CUBE_HEIGHT_RESERVED_0__MASK;
+}
+#define DPU_DATA_CUBE_HEIGHT_MINMAX_CTL__MASK			0x01c00000
+#define DPU_DATA_CUBE_HEIGHT_MINMAX_CTL__SHIFT			22
+static inline uint32_t DPU_DATA_CUBE_HEIGHT_MINMAX_CTL(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_HEIGHT_MINMAX_CTL__SHIFT) & DPU_DATA_CUBE_HEIGHT_MINMAX_CTL__MASK;
+}
+#define DPU_DATA_CUBE_HEIGHT_RESERVED_1__MASK			0x003fe000
+#define DPU_DATA_CUBE_HEIGHT_RESERVED_1__SHIFT			13
+static inline uint32_t DPU_DATA_CUBE_HEIGHT_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_HEIGHT_RESERVED_1__SHIFT) & DPU_DATA_CUBE_HEIGHT_RESERVED_1__MASK;
+}
+#define DPU_DATA_CUBE_HEIGHT_HEIGHT__MASK			0x00001fff
+#define DPU_DATA_CUBE_HEIGHT_HEIGHT__SHIFT			0
+static inline uint32_t DPU_DATA_CUBE_HEIGHT_HEIGHT(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_HEIGHT_HEIGHT__SHIFT) & DPU_DATA_CUBE_HEIGHT_HEIGHT__MASK;
+}
+
+#define REG_DPU_DATA_CUBE_NOTCH_ADDR				0x00004038
+#define DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_0__MASK		0xe0000000
+#define DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_0__SHIFT		29
+static inline uint32_t DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_0__SHIFT) & DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_0__MASK;
+}
+#define DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_1__MASK		0x1fff0000
+#define DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_1__SHIFT		16
+static inline uint32_t DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_1(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_1__SHIFT) & DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_1__MASK;
+}
+#define DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_1__MASK		0x0000e000
+#define DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_1__SHIFT		13
+static inline uint32_t DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_1__SHIFT) & DPU_DATA_CUBE_NOTCH_ADDR_RESERVED_1__MASK;
+}
+#define DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_0__MASK		0x00001fff
+#define DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_0__SHIFT		0
+static inline uint32_t DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_0(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_0__SHIFT) & DPU_DATA_CUBE_NOTCH_ADDR_NOTCH_ADDR_0__MASK;
+}
+
+#define REG_DPU_DATA_CUBE_CHANNEL				0x0000403c
+#define DPU_DATA_CUBE_CHANNEL_RESERVED_0__MASK			0xe0000000
+#define DPU_DATA_CUBE_CHANNEL_RESERVED_0__SHIFT			29
+static inline uint32_t DPU_DATA_CUBE_CHANNEL_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_CHANNEL_RESERVED_0__SHIFT) & DPU_DATA_CUBE_CHANNEL_RESERVED_0__MASK;
+}
+#define DPU_DATA_CUBE_CHANNEL_ORIG_CHANNEL__MASK		0x1fff0000
+#define DPU_DATA_CUBE_CHANNEL_ORIG_CHANNEL__SHIFT		16
+static inline uint32_t DPU_DATA_CUBE_CHANNEL_ORIG_CHANNEL(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_CHANNEL_ORIG_CHANNEL__SHIFT) & DPU_DATA_CUBE_CHANNEL_ORIG_CHANNEL__MASK;
+}
+#define DPU_DATA_CUBE_CHANNEL_RESERVED_1__MASK			0x0000e000
+#define DPU_DATA_CUBE_CHANNEL_RESERVED_1__SHIFT			13
+static inline uint32_t DPU_DATA_CUBE_CHANNEL_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_CHANNEL_RESERVED_1__SHIFT) & DPU_DATA_CUBE_CHANNEL_RESERVED_1__MASK;
+}
+#define DPU_DATA_CUBE_CHANNEL_CHANNEL__MASK			0x00001fff
+#define DPU_DATA_CUBE_CHANNEL_CHANNEL__SHIFT			0
+static inline uint32_t DPU_DATA_CUBE_CHANNEL_CHANNEL(uint32_t val)
+{
+	return ((val) << DPU_DATA_CUBE_CHANNEL_CHANNEL__SHIFT) & DPU_DATA_CUBE_CHANNEL_CHANNEL__MASK;
+}
+
+#define REG_DPU_BS_CFG						0x00004040
+#define DPU_BS_CFG_RESERVED_0__MASK				0xfff00000
+#define DPU_BS_CFG_RESERVED_0__SHIFT				20
+static inline uint32_t DPU_BS_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_RESERVED_0__SHIFT) & DPU_BS_CFG_RESERVED_0__MASK;
+}
+#define DPU_BS_CFG_BS_ALU_ALGO__MASK				0x000f0000
+#define DPU_BS_CFG_BS_ALU_ALGO__SHIFT				16
+static inline uint32_t DPU_BS_CFG_BS_ALU_ALGO(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_ALU_ALGO__SHIFT) & DPU_BS_CFG_BS_ALU_ALGO__MASK;
+}
+#define DPU_BS_CFG_RESERVED_1__MASK				0x0000fe00
+#define DPU_BS_CFG_RESERVED_1__SHIFT				9
+static inline uint32_t DPU_BS_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_RESERVED_1__SHIFT) & DPU_BS_CFG_RESERVED_1__MASK;
+}
+#define DPU_BS_CFG_BS_ALU_SRC__MASK				0x00000100
+#define DPU_BS_CFG_BS_ALU_SRC__SHIFT				8
+static inline uint32_t DPU_BS_CFG_BS_ALU_SRC(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_ALU_SRC__SHIFT) & DPU_BS_CFG_BS_ALU_SRC__MASK;
+}
+#define DPU_BS_CFG_BS_RELUX_EN__MASK				0x00000080
+#define DPU_BS_CFG_BS_RELUX_EN__SHIFT				7
+static inline uint32_t DPU_BS_CFG_BS_RELUX_EN(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_RELUX_EN__SHIFT) & DPU_BS_CFG_BS_RELUX_EN__MASK;
+}
+#define DPU_BS_CFG_BS_RELU_BYPASS__MASK				0x00000040
+#define DPU_BS_CFG_BS_RELU_BYPASS__SHIFT			6
+static inline uint32_t DPU_BS_CFG_BS_RELU_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_RELU_BYPASS__SHIFT) & DPU_BS_CFG_BS_RELU_BYPASS__MASK;
+}
+#define DPU_BS_CFG_BS_MUL_PRELU__MASK				0x00000020
+#define DPU_BS_CFG_BS_MUL_PRELU__SHIFT				5
+static inline uint32_t DPU_BS_CFG_BS_MUL_PRELU(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_MUL_PRELU__SHIFT) & DPU_BS_CFG_BS_MUL_PRELU__MASK;
+}
+#define DPU_BS_CFG_BS_MUL_BYPASS__MASK				0x00000010
+#define DPU_BS_CFG_BS_MUL_BYPASS__SHIFT				4
+static inline uint32_t DPU_BS_CFG_BS_MUL_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_MUL_BYPASS__SHIFT) & DPU_BS_CFG_BS_MUL_BYPASS__MASK;
+}
+#define DPU_BS_CFG_RESERVED_2__MASK				0x0000000c
+#define DPU_BS_CFG_RESERVED_2__SHIFT				2
+static inline uint32_t DPU_BS_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_RESERVED_2__SHIFT) & DPU_BS_CFG_RESERVED_2__MASK;
+}
+#define DPU_BS_CFG_BS_ALU_BYPASS__MASK				0x00000002
+#define DPU_BS_CFG_BS_ALU_BYPASS__SHIFT				1
+static inline uint32_t DPU_BS_CFG_BS_ALU_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_ALU_BYPASS__SHIFT) & DPU_BS_CFG_BS_ALU_BYPASS__MASK;
+}
+#define DPU_BS_CFG_BS_BYPASS__MASK				0x00000001
+#define DPU_BS_CFG_BS_BYPASS__SHIFT				0
+static inline uint32_t DPU_BS_CFG_BS_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BS_CFG_BS_BYPASS__SHIFT) & DPU_BS_CFG_BS_BYPASS__MASK;
+}
+
+#define REG_DPU_BS_ALU_CFG					0x00004044
+#define DPU_BS_ALU_CFG_BS_ALU_OPERAND__MASK			0xffffffff
+#define DPU_BS_ALU_CFG_BS_ALU_OPERAND__SHIFT			0
+static inline uint32_t DPU_BS_ALU_CFG_BS_ALU_OPERAND(uint32_t val)
+{
+	return ((val) << DPU_BS_ALU_CFG_BS_ALU_OPERAND__SHIFT) & DPU_BS_ALU_CFG_BS_ALU_OPERAND__MASK;
+}
+
+#define REG_DPU_BS_MUL_CFG					0x00004048
+#define DPU_BS_MUL_CFG_BS_MUL_OPERAND__MASK			0xffff0000
+#define DPU_BS_MUL_CFG_BS_MUL_OPERAND__SHIFT			16
+static inline uint32_t DPU_BS_MUL_CFG_BS_MUL_OPERAND(uint32_t val)
+{
+	return ((val) << DPU_BS_MUL_CFG_BS_MUL_OPERAND__SHIFT) & DPU_BS_MUL_CFG_BS_MUL_OPERAND__MASK;
+}
+#define DPU_BS_MUL_CFG_RESERVED_0__MASK				0x0000c000
+#define DPU_BS_MUL_CFG_RESERVED_0__SHIFT			14
+static inline uint32_t DPU_BS_MUL_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_BS_MUL_CFG_RESERVED_0__SHIFT) & DPU_BS_MUL_CFG_RESERVED_0__MASK;
+}
+#define DPU_BS_MUL_CFG_BS_MUL_SHIFT_VALUE__MASK			0x00003f00
+#define DPU_BS_MUL_CFG_BS_MUL_SHIFT_VALUE__SHIFT		8
+static inline uint32_t DPU_BS_MUL_CFG_BS_MUL_SHIFT_VALUE(uint32_t val)
+{
+	return ((val) << DPU_BS_MUL_CFG_BS_MUL_SHIFT_VALUE__SHIFT) & DPU_BS_MUL_CFG_BS_MUL_SHIFT_VALUE__MASK;
+}
+#define DPU_BS_MUL_CFG_RESERVED_1__MASK				0x000000fc
+#define DPU_BS_MUL_CFG_RESERVED_1__SHIFT			2
+static inline uint32_t DPU_BS_MUL_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_BS_MUL_CFG_RESERVED_1__SHIFT) & DPU_BS_MUL_CFG_RESERVED_1__MASK;
+}
+#define DPU_BS_MUL_CFG_BS_TRUNCATE_SRC__MASK			0x00000002
+#define DPU_BS_MUL_CFG_BS_TRUNCATE_SRC__SHIFT			1
+static inline uint32_t DPU_BS_MUL_CFG_BS_TRUNCATE_SRC(uint32_t val)
+{
+	return ((val) << DPU_BS_MUL_CFG_BS_TRUNCATE_SRC__SHIFT) & DPU_BS_MUL_CFG_BS_TRUNCATE_SRC__MASK;
+}
+#define DPU_BS_MUL_CFG_BS_MUL_SRC__MASK				0x00000001
+#define DPU_BS_MUL_CFG_BS_MUL_SRC__SHIFT			0
+static inline uint32_t DPU_BS_MUL_CFG_BS_MUL_SRC(uint32_t val)
+{
+	return ((val) << DPU_BS_MUL_CFG_BS_MUL_SRC__SHIFT) & DPU_BS_MUL_CFG_BS_MUL_SRC__MASK;
+}
+
+#define REG_DPU_BS_RELUX_CMP_VALUE				0x0000404c
+#define DPU_BS_RELUX_CMP_VALUE_BS_RELUX_CMP_DAT__MASK		0xffffffff
+#define DPU_BS_RELUX_CMP_VALUE_BS_RELUX_CMP_DAT__SHIFT		0
+static inline uint32_t DPU_BS_RELUX_CMP_VALUE_BS_RELUX_CMP_DAT(uint32_t val)
+{
+	return ((val) << DPU_BS_RELUX_CMP_VALUE_BS_RELUX_CMP_DAT__SHIFT) & DPU_BS_RELUX_CMP_VALUE_BS_RELUX_CMP_DAT__MASK;
+}
+
+#define REG_DPU_BS_OW_CFG					0x00004050
+#define DPU_BS_OW_CFG_RGP_CNTER__MASK				0xf0000000
+#define DPU_BS_OW_CFG_RGP_CNTER__SHIFT				28
+static inline uint32_t DPU_BS_OW_CFG_RGP_CNTER(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_RGP_CNTER__SHIFT) & DPU_BS_OW_CFG_RGP_CNTER__MASK;
+}
+#define DPU_BS_OW_CFG_TP_ORG_EN__MASK				0x08000000
+#define DPU_BS_OW_CFG_TP_ORG_EN__SHIFT				27
+static inline uint32_t DPU_BS_OW_CFG_TP_ORG_EN(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_TP_ORG_EN__SHIFT) & DPU_BS_OW_CFG_TP_ORG_EN__MASK;
+}
+#define DPU_BS_OW_CFG_RESERVED_0__MASK				0x07fff800
+#define DPU_BS_OW_CFG_RESERVED_0__SHIFT				11
+static inline uint32_t DPU_BS_OW_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_RESERVED_0__SHIFT) & DPU_BS_OW_CFG_RESERVED_0__MASK;
+}
+#define DPU_BS_OW_CFG_SIZE_E_2__MASK				0x00000700
+#define DPU_BS_OW_CFG_SIZE_E_2__SHIFT				8
+static inline uint32_t DPU_BS_OW_CFG_SIZE_E_2(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_SIZE_E_2__SHIFT) & DPU_BS_OW_CFG_SIZE_E_2__MASK;
+}
+#define DPU_BS_OW_CFG_SIZE_E_1__MASK				0x000000e0
+#define DPU_BS_OW_CFG_SIZE_E_1__SHIFT				5
+static inline uint32_t DPU_BS_OW_CFG_SIZE_E_1(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_SIZE_E_1__SHIFT) & DPU_BS_OW_CFG_SIZE_E_1__MASK;
+}
+#define DPU_BS_OW_CFG_SIZE_E_0__MASK				0x0000001c
+#define DPU_BS_OW_CFG_SIZE_E_0__SHIFT				2
+static inline uint32_t DPU_BS_OW_CFG_SIZE_E_0(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_SIZE_E_0__SHIFT) & DPU_BS_OW_CFG_SIZE_E_0__MASK;
+}
+#define DPU_BS_OW_CFG_OD_BYPASS__MASK				0x00000002
+#define DPU_BS_OW_CFG_OD_BYPASS__SHIFT				1
+static inline uint32_t DPU_BS_OW_CFG_OD_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_OD_BYPASS__SHIFT) & DPU_BS_OW_CFG_OD_BYPASS__MASK;
+}
+#define DPU_BS_OW_CFG_OW_SRC__MASK				0x00000001
+#define DPU_BS_OW_CFG_OW_SRC__SHIFT				0
+static inline uint32_t DPU_BS_OW_CFG_OW_SRC(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_CFG_OW_SRC__SHIFT) & DPU_BS_OW_CFG_OW_SRC__MASK;
+}
+
+#define REG_DPU_BS_OW_OP					0x00004054
+#define DPU_BS_OW_OP_RESERVED_0__MASK				0xffff0000
+#define DPU_BS_OW_OP_RESERVED_0__SHIFT				16
+static inline uint32_t DPU_BS_OW_OP_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_OP_RESERVED_0__SHIFT) & DPU_BS_OW_OP_RESERVED_0__MASK;
+}
+#define DPU_BS_OW_OP_OW_OP__MASK				0x0000ffff
+#define DPU_BS_OW_OP_OW_OP__SHIFT				0
+static inline uint32_t DPU_BS_OW_OP_OW_OP(uint32_t val)
+{
+	return ((val) << DPU_BS_OW_OP_OW_OP__SHIFT) & DPU_BS_OW_OP_OW_OP__MASK;
+}
+
+#define REG_DPU_WDMA_SIZE_0					0x00004058
+#define DPU_WDMA_SIZE_0_RESERVED_0__MASK			0xf0000000
+#define DPU_WDMA_SIZE_0_RESERVED_0__SHIFT			28
+static inline uint32_t DPU_WDMA_SIZE_0_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_0_RESERVED_0__SHIFT) & DPU_WDMA_SIZE_0_RESERVED_0__MASK;
+}
+#define DPU_WDMA_SIZE_0_TP_PRECISION__MASK			0x08000000
+#define DPU_WDMA_SIZE_0_TP_PRECISION__SHIFT			27
+static inline uint32_t DPU_WDMA_SIZE_0_TP_PRECISION(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_0_TP_PRECISION__SHIFT) & DPU_WDMA_SIZE_0_TP_PRECISION__MASK;
+}
+#define DPU_WDMA_SIZE_0_SIZE_C_WDMA__MASK			0x07ff0000
+#define DPU_WDMA_SIZE_0_SIZE_C_WDMA__SHIFT			16
+static inline uint32_t DPU_WDMA_SIZE_0_SIZE_C_WDMA(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_0_SIZE_C_WDMA__SHIFT) & DPU_WDMA_SIZE_0_SIZE_C_WDMA__MASK;
+}
+#define DPU_WDMA_SIZE_0_RESERVED_1__MASK			0x0000e000
+#define DPU_WDMA_SIZE_0_RESERVED_1__SHIFT			13
+static inline uint32_t DPU_WDMA_SIZE_0_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_0_RESERVED_1__SHIFT) & DPU_WDMA_SIZE_0_RESERVED_1__MASK;
+}
+#define DPU_WDMA_SIZE_0_CHANNEL_WDMA__MASK			0x00001fff
+#define DPU_WDMA_SIZE_0_CHANNEL_WDMA__SHIFT			0
+static inline uint32_t DPU_WDMA_SIZE_0_CHANNEL_WDMA(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_0_CHANNEL_WDMA__SHIFT) & DPU_WDMA_SIZE_0_CHANNEL_WDMA__MASK;
+}
+
+#define REG_DPU_WDMA_SIZE_1					0x0000405c
+#define DPU_WDMA_SIZE_1_RESERVED_0__MASK			0xe0000000
+#define DPU_WDMA_SIZE_1_RESERVED_0__SHIFT			29
+static inline uint32_t DPU_WDMA_SIZE_1_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_1_RESERVED_0__SHIFT) & DPU_WDMA_SIZE_1_RESERVED_0__MASK;
+}
+#define DPU_WDMA_SIZE_1_HEIGHT_WDMA__MASK			0x1fff0000
+#define DPU_WDMA_SIZE_1_HEIGHT_WDMA__SHIFT			16
+static inline uint32_t DPU_WDMA_SIZE_1_HEIGHT_WDMA(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_1_HEIGHT_WDMA__SHIFT) & DPU_WDMA_SIZE_1_HEIGHT_WDMA__MASK;
+}
+#define DPU_WDMA_SIZE_1_RESERVED_1__MASK			0x0000e000
+#define DPU_WDMA_SIZE_1_RESERVED_1__SHIFT			13
+static inline uint32_t DPU_WDMA_SIZE_1_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_1_RESERVED_1__SHIFT) & DPU_WDMA_SIZE_1_RESERVED_1__MASK;
+}
+#define DPU_WDMA_SIZE_1_WIDTH_WDMA__MASK			0x00001fff
+#define DPU_WDMA_SIZE_1_WIDTH_WDMA__SHIFT			0
+static inline uint32_t DPU_WDMA_SIZE_1_WIDTH_WDMA(uint32_t val)
+{
+	return ((val) << DPU_WDMA_SIZE_1_WIDTH_WDMA__SHIFT) & DPU_WDMA_SIZE_1_WIDTH_WDMA__MASK;
+}
+
+#define REG_DPU_BN_CFG						0x00004060
+#define DPU_BN_CFG_RESERVED_0__MASK				0xfff00000
+#define DPU_BN_CFG_RESERVED_0__SHIFT				20
+static inline uint32_t DPU_BN_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_RESERVED_0__SHIFT) & DPU_BN_CFG_RESERVED_0__MASK;
+}
+#define DPU_BN_CFG_BN_ALU_ALGO__MASK				0x000f0000
+#define DPU_BN_CFG_BN_ALU_ALGO__SHIFT				16
+static inline uint32_t DPU_BN_CFG_BN_ALU_ALGO(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_ALU_ALGO__SHIFT) & DPU_BN_CFG_BN_ALU_ALGO__MASK;
+}
+#define DPU_BN_CFG_RESERVED_1__MASK				0x0000fe00
+#define DPU_BN_CFG_RESERVED_1__SHIFT				9
+static inline uint32_t DPU_BN_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_RESERVED_1__SHIFT) & DPU_BN_CFG_RESERVED_1__MASK;
+}
+#define DPU_BN_CFG_BN_ALU_SRC__MASK				0x00000100
+#define DPU_BN_CFG_BN_ALU_SRC__SHIFT				8
+static inline uint32_t DPU_BN_CFG_BN_ALU_SRC(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_ALU_SRC__SHIFT) & DPU_BN_CFG_BN_ALU_SRC__MASK;
+}
+#define DPU_BN_CFG_BN_RELUX_EN__MASK				0x00000080
+#define DPU_BN_CFG_BN_RELUX_EN__SHIFT				7
+static inline uint32_t DPU_BN_CFG_BN_RELUX_EN(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_RELUX_EN__SHIFT) & DPU_BN_CFG_BN_RELUX_EN__MASK;
+}
+#define DPU_BN_CFG_BN_RELU_BYPASS__MASK				0x00000040
+#define DPU_BN_CFG_BN_RELU_BYPASS__SHIFT			6
+static inline uint32_t DPU_BN_CFG_BN_RELU_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_RELU_BYPASS__SHIFT) & DPU_BN_CFG_BN_RELU_BYPASS__MASK;
+}
+#define DPU_BN_CFG_BN_MUL_PRELU__MASK				0x00000020
+#define DPU_BN_CFG_BN_MUL_PRELU__SHIFT				5
+static inline uint32_t DPU_BN_CFG_BN_MUL_PRELU(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_MUL_PRELU__SHIFT) & DPU_BN_CFG_BN_MUL_PRELU__MASK;
+}
+#define DPU_BN_CFG_BN_MUL_BYPASS__MASK				0x00000010
+#define DPU_BN_CFG_BN_MUL_BYPASS__SHIFT				4
+static inline uint32_t DPU_BN_CFG_BN_MUL_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_MUL_BYPASS__SHIFT) & DPU_BN_CFG_BN_MUL_BYPASS__MASK;
+}
+#define DPU_BN_CFG_RESERVED_2__MASK				0x0000000c
+#define DPU_BN_CFG_RESERVED_2__SHIFT				2
+static inline uint32_t DPU_BN_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_RESERVED_2__SHIFT) & DPU_BN_CFG_RESERVED_2__MASK;
+}
+#define DPU_BN_CFG_BN_ALU_BYPASS__MASK				0x00000002
+#define DPU_BN_CFG_BN_ALU_BYPASS__SHIFT				1
+static inline uint32_t DPU_BN_CFG_BN_ALU_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_ALU_BYPASS__SHIFT) & DPU_BN_CFG_BN_ALU_BYPASS__MASK;
+}
+#define DPU_BN_CFG_BN_BYPASS__MASK				0x00000001
+#define DPU_BN_CFG_BN_BYPASS__SHIFT				0
+static inline uint32_t DPU_BN_CFG_BN_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_BN_CFG_BN_BYPASS__SHIFT) & DPU_BN_CFG_BN_BYPASS__MASK;
+}
+
+#define REG_DPU_BN_ALU_CFG					0x00004064
+#define DPU_BN_ALU_CFG_BN_ALU_OPERAND__MASK			0xffffffff
+#define DPU_BN_ALU_CFG_BN_ALU_OPERAND__SHIFT			0
+static inline uint32_t DPU_BN_ALU_CFG_BN_ALU_OPERAND(uint32_t val)
+{
+	return ((val) << DPU_BN_ALU_CFG_BN_ALU_OPERAND__SHIFT) & DPU_BN_ALU_CFG_BN_ALU_OPERAND__MASK;
+}
+
+#define REG_DPU_BN_MUL_CFG					0x00004068
+#define DPU_BN_MUL_CFG_BN_MUL_OPERAND__MASK			0xffff0000
+#define DPU_BN_MUL_CFG_BN_MUL_OPERAND__SHIFT			16
+static inline uint32_t DPU_BN_MUL_CFG_BN_MUL_OPERAND(uint32_t val)
+{
+	return ((val) << DPU_BN_MUL_CFG_BN_MUL_OPERAND__SHIFT) & DPU_BN_MUL_CFG_BN_MUL_OPERAND__MASK;
+}
+#define DPU_BN_MUL_CFG_RESERVED_0__MASK				0x0000c000
+#define DPU_BN_MUL_CFG_RESERVED_0__SHIFT			14
+static inline uint32_t DPU_BN_MUL_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_BN_MUL_CFG_RESERVED_0__SHIFT) & DPU_BN_MUL_CFG_RESERVED_0__MASK;
+}
+#define DPU_BN_MUL_CFG_BN_MUL_SHIFT_VALUE__MASK			0x00003f00
+#define DPU_BN_MUL_CFG_BN_MUL_SHIFT_VALUE__SHIFT		8
+static inline uint32_t DPU_BN_MUL_CFG_BN_MUL_SHIFT_VALUE(uint32_t val)
+{
+	return ((val) << DPU_BN_MUL_CFG_BN_MUL_SHIFT_VALUE__SHIFT) & DPU_BN_MUL_CFG_BN_MUL_SHIFT_VALUE__MASK;
+}
+#define DPU_BN_MUL_CFG_RESERVED_1__MASK				0x000000fc
+#define DPU_BN_MUL_CFG_RESERVED_1__SHIFT			2
+static inline uint32_t DPU_BN_MUL_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_BN_MUL_CFG_RESERVED_1__SHIFT) & DPU_BN_MUL_CFG_RESERVED_1__MASK;
+}
+#define DPU_BN_MUL_CFG_BN_TRUNCATE_SRC__MASK			0x00000002
+#define DPU_BN_MUL_CFG_BN_TRUNCATE_SRC__SHIFT			1
+static inline uint32_t DPU_BN_MUL_CFG_BN_TRUNCATE_SRC(uint32_t val)
+{
+	return ((val) << DPU_BN_MUL_CFG_BN_TRUNCATE_SRC__SHIFT) & DPU_BN_MUL_CFG_BN_TRUNCATE_SRC__MASK;
+}
+#define DPU_BN_MUL_CFG_BN_MUL_SRC__MASK				0x00000001
+#define DPU_BN_MUL_CFG_BN_MUL_SRC__SHIFT			0
+static inline uint32_t DPU_BN_MUL_CFG_BN_MUL_SRC(uint32_t val)
+{
+	return ((val) << DPU_BN_MUL_CFG_BN_MUL_SRC__SHIFT) & DPU_BN_MUL_CFG_BN_MUL_SRC__MASK;
+}
+
+#define REG_DPU_BN_RELUX_CMP_VALUE				0x0000406c
+#define DPU_BN_RELUX_CMP_VALUE_BN_RELUX_CMP_DAT__MASK		0xffffffff
+#define DPU_BN_RELUX_CMP_VALUE_BN_RELUX_CMP_DAT__SHIFT		0
+static inline uint32_t DPU_BN_RELUX_CMP_VALUE_BN_RELUX_CMP_DAT(uint32_t val)
+{
+	return ((val) << DPU_BN_RELUX_CMP_VALUE_BN_RELUX_CMP_DAT__SHIFT) & DPU_BN_RELUX_CMP_VALUE_BN_RELUX_CMP_DAT__MASK;
+}
+
+#define REG_DPU_EW_CFG						0x00004070
+#define DPU_EW_CFG_EW_CVT_TYPE__MASK				0x80000000
+#define DPU_EW_CFG_EW_CVT_TYPE__SHIFT				31
+static inline uint32_t DPU_EW_CFG_EW_CVT_TYPE(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_CVT_TYPE__SHIFT) & DPU_EW_CFG_EW_CVT_TYPE__MASK;
+}
+#define DPU_EW_CFG_EW_CVT_ROUND__MASK				0x40000000
+#define DPU_EW_CFG_EW_CVT_ROUND__SHIFT				30
+static inline uint32_t DPU_EW_CFG_EW_CVT_ROUND(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_CVT_ROUND__SHIFT) & DPU_EW_CFG_EW_CVT_ROUND__MASK;
+}
+#define DPU_EW_CFG_EW_DATA_MODE__MASK				0x30000000
+#define DPU_EW_CFG_EW_DATA_MODE__SHIFT				28
+static inline uint32_t DPU_EW_CFG_EW_DATA_MODE(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_DATA_MODE__SHIFT) & DPU_EW_CFG_EW_DATA_MODE__MASK;
+}
+#define DPU_EW_CFG_RESERVED_0__MASK				0x0f000000
+#define DPU_EW_CFG_RESERVED_0__SHIFT				24
+static inline uint32_t DPU_EW_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_RESERVED_0__SHIFT) & DPU_EW_CFG_RESERVED_0__MASK;
+}
+#define DPU_EW_CFG_EDATA_SIZE__MASK				0x00c00000
+#define DPU_EW_CFG_EDATA_SIZE__SHIFT				22
+static inline uint32_t DPU_EW_CFG_EDATA_SIZE(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EDATA_SIZE__SHIFT) & DPU_EW_CFG_EDATA_SIZE__MASK;
+}
+#define DPU_EW_CFG_EW_EQUAL_EN__MASK				0x00200000
+#define DPU_EW_CFG_EW_EQUAL_EN__SHIFT				21
+static inline uint32_t DPU_EW_CFG_EW_EQUAL_EN(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_EQUAL_EN__SHIFT) & DPU_EW_CFG_EW_EQUAL_EN__MASK;
+}
+#define DPU_EW_CFG_EW_BINARY_EN__MASK				0x00100000
+#define DPU_EW_CFG_EW_BINARY_EN__SHIFT				20
+static inline uint32_t DPU_EW_CFG_EW_BINARY_EN(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_BINARY_EN__SHIFT) & DPU_EW_CFG_EW_BINARY_EN__MASK;
+}
+#define DPU_EW_CFG_EW_ALU_ALGO__MASK				0x000f0000
+#define DPU_EW_CFG_EW_ALU_ALGO__SHIFT				16
+static inline uint32_t DPU_EW_CFG_EW_ALU_ALGO(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_ALU_ALGO__SHIFT) & DPU_EW_CFG_EW_ALU_ALGO__MASK;
+}
+#define DPU_EW_CFG_RESERVED_1__MASK				0x0000f800
+#define DPU_EW_CFG_RESERVED_1__SHIFT				11
+static inline uint32_t DPU_EW_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_RESERVED_1__SHIFT) & DPU_EW_CFG_RESERVED_1__MASK;
+}
+#define DPU_EW_CFG_EW_RELUX_EN__MASK				0x00000400
+#define DPU_EW_CFG_EW_RELUX_EN__SHIFT				10
+static inline uint32_t DPU_EW_CFG_EW_RELUX_EN(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_RELUX_EN__SHIFT) & DPU_EW_CFG_EW_RELUX_EN__MASK;
+}
+#define DPU_EW_CFG_EW_RELU_BYPASS__MASK				0x00000200
+#define DPU_EW_CFG_EW_RELU_BYPASS__SHIFT			9
+static inline uint32_t DPU_EW_CFG_EW_RELU_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_RELU_BYPASS__SHIFT) & DPU_EW_CFG_EW_RELU_BYPASS__MASK;
+}
+#define DPU_EW_CFG_EW_OP_CVT_BYPASS__MASK			0x00000100
+#define DPU_EW_CFG_EW_OP_CVT_BYPASS__SHIFT			8
+static inline uint32_t DPU_EW_CFG_EW_OP_CVT_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_OP_CVT_BYPASS__SHIFT) & DPU_EW_CFG_EW_OP_CVT_BYPASS__MASK;
+}
+#define DPU_EW_CFG_EW_LUT_BYPASS__MASK				0x00000080
+#define DPU_EW_CFG_EW_LUT_BYPASS__SHIFT				7
+static inline uint32_t DPU_EW_CFG_EW_LUT_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_LUT_BYPASS__SHIFT) & DPU_EW_CFG_EW_LUT_BYPASS__MASK;
+}
+#define DPU_EW_CFG_EW_OP_SRC__MASK				0x00000040
+#define DPU_EW_CFG_EW_OP_SRC__SHIFT				6
+static inline uint32_t DPU_EW_CFG_EW_OP_SRC(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_OP_SRC__SHIFT) & DPU_EW_CFG_EW_OP_SRC__MASK;
+}
+#define DPU_EW_CFG_EW_MUL_PRELU__MASK				0x00000020
+#define DPU_EW_CFG_EW_MUL_PRELU__SHIFT				5
+static inline uint32_t DPU_EW_CFG_EW_MUL_PRELU(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_MUL_PRELU__SHIFT) & DPU_EW_CFG_EW_MUL_PRELU__MASK;
+}
+#define DPU_EW_CFG_RESERVED_2__MASK				0x00000018
+#define DPU_EW_CFG_RESERVED_2__SHIFT				3
+static inline uint32_t DPU_EW_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_RESERVED_2__SHIFT) & DPU_EW_CFG_RESERVED_2__MASK;
+}
+#define DPU_EW_CFG_EW_OP_TYPE__MASK				0x00000004
+#define DPU_EW_CFG_EW_OP_TYPE__SHIFT				2
+static inline uint32_t DPU_EW_CFG_EW_OP_TYPE(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_OP_TYPE__SHIFT) & DPU_EW_CFG_EW_OP_TYPE__MASK;
+}
+#define DPU_EW_CFG_EW_OP_BYPASS__MASK				0x00000002
+#define DPU_EW_CFG_EW_OP_BYPASS__SHIFT				1
+static inline uint32_t DPU_EW_CFG_EW_OP_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_OP_BYPASS__SHIFT) & DPU_EW_CFG_EW_OP_BYPASS__MASK;
+}
+#define DPU_EW_CFG_EW_BYPASS__MASK				0x00000001
+#define DPU_EW_CFG_EW_BYPASS__SHIFT				0
+static inline uint32_t DPU_EW_CFG_EW_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_EW_CFG_EW_BYPASS__SHIFT) & DPU_EW_CFG_EW_BYPASS__MASK;
+}
+
+#define REG_DPU_EW_CVT_OFFSET_VALUE				0x00004074
+#define DPU_EW_CVT_OFFSET_VALUE_EW_OP_CVT_OFFSET__MASK		0xffffffff
+#define DPU_EW_CVT_OFFSET_VALUE_EW_OP_CVT_OFFSET__SHIFT		0
+static inline uint32_t DPU_EW_CVT_OFFSET_VALUE_EW_OP_CVT_OFFSET(uint32_t val)
+{
+	return ((val) << DPU_EW_CVT_OFFSET_VALUE_EW_OP_CVT_OFFSET__SHIFT) & DPU_EW_CVT_OFFSET_VALUE_EW_OP_CVT_OFFSET__MASK;
+}
+
+#define REG_DPU_EW_CVT_SCALE_VALUE				0x00004078
+#define DPU_EW_CVT_SCALE_VALUE_EW_TRUNCATE__MASK		0xffc00000
+#define DPU_EW_CVT_SCALE_VALUE_EW_TRUNCATE__SHIFT		22
+static inline uint32_t DPU_EW_CVT_SCALE_VALUE_EW_TRUNCATE(uint32_t val)
+{
+	return ((val) << DPU_EW_CVT_SCALE_VALUE_EW_TRUNCATE__SHIFT) & DPU_EW_CVT_SCALE_VALUE_EW_TRUNCATE__MASK;
+}
+#define DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SHIFT__MASK		0x003f0000
+#define DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SHIFT__SHIFT		16
+static inline uint32_t DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SHIFT(uint32_t val)
+{
+	return ((val) << DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SHIFT__SHIFT) & DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SHIFT__MASK;
+}
+#define DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SCALE__MASK		0x0000ffff
+#define DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SCALE__SHIFT		0
+static inline uint32_t DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SCALE(uint32_t val)
+{
+	return ((val) << DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SCALE__SHIFT) & DPU_EW_CVT_SCALE_VALUE_EW_OP_CVT_SCALE__MASK;
+}
+
+#define REG_DPU_EW_RELUX_CMP_VALUE				0x0000407c
+#define DPU_EW_RELUX_CMP_VALUE_EW_RELUX_CMP_DAT__MASK		0xffffffff
+#define DPU_EW_RELUX_CMP_VALUE_EW_RELUX_CMP_DAT__SHIFT		0
+static inline uint32_t DPU_EW_RELUX_CMP_VALUE_EW_RELUX_CMP_DAT(uint32_t val)
+{
+	return ((val) << DPU_EW_RELUX_CMP_VALUE_EW_RELUX_CMP_DAT__SHIFT) & DPU_EW_RELUX_CMP_VALUE_EW_RELUX_CMP_DAT__MASK;
+}
+
+#define REG_DPU_OUT_CVT_OFFSET					0x00004080
+#define DPU_OUT_CVT_OFFSET_OUT_CVT_OFFSET__MASK			0xffffffff
+#define DPU_OUT_CVT_OFFSET_OUT_CVT_OFFSET__SHIFT		0
+static inline uint32_t DPU_OUT_CVT_OFFSET_OUT_CVT_OFFSET(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_OFFSET_OUT_CVT_OFFSET__SHIFT) & DPU_OUT_CVT_OFFSET_OUT_CVT_OFFSET__MASK;
+}
+
+#define REG_DPU_OUT_CVT_SCALE					0x00004084
+#define DPU_OUT_CVT_SCALE_RESERVED_0__MASK			0xfffe0000
+#define DPU_OUT_CVT_SCALE_RESERVED_0__SHIFT			17
+static inline uint32_t DPU_OUT_CVT_SCALE_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SCALE_RESERVED_0__SHIFT) & DPU_OUT_CVT_SCALE_RESERVED_0__MASK;
+}
+#define DPU_OUT_CVT_SCALE_FP32TOFP16_EN__MASK			0x00010000
+#define DPU_OUT_CVT_SCALE_FP32TOFP16_EN__SHIFT			16
+static inline uint32_t DPU_OUT_CVT_SCALE_FP32TOFP16_EN(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SCALE_FP32TOFP16_EN__SHIFT) & DPU_OUT_CVT_SCALE_FP32TOFP16_EN__MASK;
+}
+#define DPU_OUT_CVT_SCALE_OUT_CVT_SCALE__MASK			0x0000ffff
+#define DPU_OUT_CVT_SCALE_OUT_CVT_SCALE__SHIFT			0
+static inline uint32_t DPU_OUT_CVT_SCALE_OUT_CVT_SCALE(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SCALE_OUT_CVT_SCALE__SHIFT) & DPU_OUT_CVT_SCALE_OUT_CVT_SCALE__MASK;
+}
+
+#define REG_DPU_OUT_CVT_SHIFT					0x00004088
+#define DPU_OUT_CVT_SHIFT_CVT_TYPE__MASK			0x80000000
+#define DPU_OUT_CVT_SHIFT_CVT_TYPE__SHIFT			31
+static inline uint32_t DPU_OUT_CVT_SHIFT_CVT_TYPE(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SHIFT_CVT_TYPE__SHIFT) & DPU_OUT_CVT_SHIFT_CVT_TYPE__MASK;
+}
+#define DPU_OUT_CVT_SHIFT_CVT_ROUND__MASK			0x40000000
+#define DPU_OUT_CVT_SHIFT_CVT_ROUND__SHIFT			30
+static inline uint32_t DPU_OUT_CVT_SHIFT_CVT_ROUND(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SHIFT_CVT_ROUND__SHIFT) & DPU_OUT_CVT_SHIFT_CVT_ROUND__MASK;
+}
+#define DPU_OUT_CVT_SHIFT_RESERVED_0__MASK			0x3ff00000
+#define DPU_OUT_CVT_SHIFT_RESERVED_0__SHIFT			20
+static inline uint32_t DPU_OUT_CVT_SHIFT_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SHIFT_RESERVED_0__SHIFT) & DPU_OUT_CVT_SHIFT_RESERVED_0__MASK;
+}
+#define DPU_OUT_CVT_SHIFT_MINUS_EXP__MASK			0x000ff000
+#define DPU_OUT_CVT_SHIFT_MINUS_EXP__SHIFT			12
+static inline uint32_t DPU_OUT_CVT_SHIFT_MINUS_EXP(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SHIFT_MINUS_EXP__SHIFT) & DPU_OUT_CVT_SHIFT_MINUS_EXP__MASK;
+}
+#define DPU_OUT_CVT_SHIFT_OUT_CVT_SHIFT__MASK			0x00000fff
+#define DPU_OUT_CVT_SHIFT_OUT_CVT_SHIFT__SHIFT			0
+static inline uint32_t DPU_OUT_CVT_SHIFT_OUT_CVT_SHIFT(uint32_t val)
+{
+	return ((val) << DPU_OUT_CVT_SHIFT_OUT_CVT_SHIFT__SHIFT) & DPU_OUT_CVT_SHIFT_OUT_CVT_SHIFT__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_0					0x00004090
+#define DPU_EW_OP_VALUE_0_EW_OPERAND_0__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_0_EW_OPERAND_0__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_0_EW_OPERAND_0(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_0_EW_OPERAND_0__SHIFT) & DPU_EW_OP_VALUE_0_EW_OPERAND_0__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_1					0x00004094
+#define DPU_EW_OP_VALUE_1_EW_OPERAND_1__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_1_EW_OPERAND_1__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_1_EW_OPERAND_1(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_1_EW_OPERAND_1__SHIFT) & DPU_EW_OP_VALUE_1_EW_OPERAND_1__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_2					0x00004098
+#define DPU_EW_OP_VALUE_2_EW_OPERAND_2__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_2_EW_OPERAND_2__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_2_EW_OPERAND_2(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_2_EW_OPERAND_2__SHIFT) & DPU_EW_OP_VALUE_2_EW_OPERAND_2__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_3					0x0000409c
+#define DPU_EW_OP_VALUE_3_EW_OPERAND_3__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_3_EW_OPERAND_3__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_3_EW_OPERAND_3(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_3_EW_OPERAND_3__SHIFT) & DPU_EW_OP_VALUE_3_EW_OPERAND_3__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_4					0x000040a0
+#define DPU_EW_OP_VALUE_4_EW_OPERAND_4__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_4_EW_OPERAND_4__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_4_EW_OPERAND_4(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_4_EW_OPERAND_4__SHIFT) & DPU_EW_OP_VALUE_4_EW_OPERAND_4__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_5					0x000040a4
+#define DPU_EW_OP_VALUE_5_EW_OPERAND_5__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_5_EW_OPERAND_5__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_5_EW_OPERAND_5(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_5_EW_OPERAND_5__SHIFT) & DPU_EW_OP_VALUE_5_EW_OPERAND_5__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_6					0x000040a8
+#define DPU_EW_OP_VALUE_6_EW_OPERAND_6__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_6_EW_OPERAND_6__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_6_EW_OPERAND_6(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_6_EW_OPERAND_6__SHIFT) & DPU_EW_OP_VALUE_6_EW_OPERAND_6__MASK;
+}
+
+#define REG_DPU_EW_OP_VALUE_7					0x000040ac
+#define DPU_EW_OP_VALUE_7_EW_OPERAND_7__MASK			0xffffffff
+#define DPU_EW_OP_VALUE_7_EW_OPERAND_7__SHIFT			0
+static inline uint32_t DPU_EW_OP_VALUE_7_EW_OPERAND_7(uint32_t val)
+{
+	return ((val) << DPU_EW_OP_VALUE_7_EW_OPERAND_7__SHIFT) & DPU_EW_OP_VALUE_7_EW_OPERAND_7__MASK;
+}
+
+#define REG_DPU_SURFACE_ADD					0x000040c0
+#define DPU_SURFACE_ADD_SURF_ADD__MASK				0xfffffff0
+#define DPU_SURFACE_ADD_SURF_ADD__SHIFT				4
+static inline uint32_t DPU_SURFACE_ADD_SURF_ADD(uint32_t val)
+{
+	return ((val) << DPU_SURFACE_ADD_SURF_ADD__SHIFT) & DPU_SURFACE_ADD_SURF_ADD__MASK;
+}
+#define DPU_SURFACE_ADD_RESERVED_0__MASK			0x0000000f
+#define DPU_SURFACE_ADD_RESERVED_0__SHIFT			0
+static inline uint32_t DPU_SURFACE_ADD_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_SURFACE_ADD_RESERVED_0__SHIFT) & DPU_SURFACE_ADD_RESERVED_0__MASK;
+}
+
+#define REG_DPU_LUT_ACCESS_CFG					0x00004100
+#define DPU_LUT_ACCESS_CFG_RESERVED_0__MASK			0xfffc0000
+#define DPU_LUT_ACCESS_CFG_RESERVED_0__SHIFT			18
+static inline uint32_t DPU_LUT_ACCESS_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_CFG_RESERVED_0__SHIFT) & DPU_LUT_ACCESS_CFG_RESERVED_0__MASK;
+}
+#define DPU_LUT_ACCESS_CFG_LUT_ACCESS_TYPE__MASK		0x00020000
+#define DPU_LUT_ACCESS_CFG_LUT_ACCESS_TYPE__SHIFT		17
+static inline uint32_t DPU_LUT_ACCESS_CFG_LUT_ACCESS_TYPE(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_CFG_LUT_ACCESS_TYPE__SHIFT) & DPU_LUT_ACCESS_CFG_LUT_ACCESS_TYPE__MASK;
+}
+#define DPU_LUT_ACCESS_CFG_LUT_TABLE_ID__MASK			0x00010000
+#define DPU_LUT_ACCESS_CFG_LUT_TABLE_ID__SHIFT			16
+static inline uint32_t DPU_LUT_ACCESS_CFG_LUT_TABLE_ID(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_CFG_LUT_TABLE_ID__SHIFT) & DPU_LUT_ACCESS_CFG_LUT_TABLE_ID__MASK;
+}
+#define DPU_LUT_ACCESS_CFG_RESERVED_1__MASK			0x0000fc00
+#define DPU_LUT_ACCESS_CFG_RESERVED_1__SHIFT			10
+static inline uint32_t DPU_LUT_ACCESS_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_CFG_RESERVED_1__SHIFT) & DPU_LUT_ACCESS_CFG_RESERVED_1__MASK;
+}
+#define DPU_LUT_ACCESS_CFG_LUT_ADDR__MASK			0x000003ff
+#define DPU_LUT_ACCESS_CFG_LUT_ADDR__SHIFT			0
+static inline uint32_t DPU_LUT_ACCESS_CFG_LUT_ADDR(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_CFG_LUT_ADDR__SHIFT) & DPU_LUT_ACCESS_CFG_LUT_ADDR__MASK;
+}
+
+#define REG_DPU_LUT_ACCESS_DATA					0x00004104
+#define DPU_LUT_ACCESS_DATA_RESERVED_0__MASK			0xffff0000
+#define DPU_LUT_ACCESS_DATA_RESERVED_0__SHIFT			16
+static inline uint32_t DPU_LUT_ACCESS_DATA_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_DATA_RESERVED_0__SHIFT) & DPU_LUT_ACCESS_DATA_RESERVED_0__MASK;
+}
+#define DPU_LUT_ACCESS_DATA_LUT_ACCESS_DATA__MASK		0x0000ffff
+#define DPU_LUT_ACCESS_DATA_LUT_ACCESS_DATA__SHIFT		0
+static inline uint32_t DPU_LUT_ACCESS_DATA_LUT_ACCESS_DATA(uint32_t val)
+{
+	return ((val) << DPU_LUT_ACCESS_DATA_LUT_ACCESS_DATA__SHIFT) & DPU_LUT_ACCESS_DATA_LUT_ACCESS_DATA__MASK;
+}
+
+#define REG_DPU_LUT_CFG						0x00004108
+#define DPU_LUT_CFG_RESERVED_0__MASK				0xffffff00
+#define DPU_LUT_CFG_RESERVED_0__SHIFT				8
+static inline uint32_t DPU_LUT_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_RESERVED_0__SHIFT) & DPU_LUT_CFG_RESERVED_0__MASK;
+}
+#define DPU_LUT_CFG_LUT_CAL_SEL__MASK				0x00000080
+#define DPU_LUT_CFG_LUT_CAL_SEL__SHIFT				7
+static inline uint32_t DPU_LUT_CFG_LUT_CAL_SEL(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_CAL_SEL__SHIFT) & DPU_LUT_CFG_LUT_CAL_SEL__MASK;
+}
+#define DPU_LUT_CFG_LUT_HYBRID_PRIORITY__MASK			0x00000040
+#define DPU_LUT_CFG_LUT_HYBRID_PRIORITY__SHIFT			6
+static inline uint32_t DPU_LUT_CFG_LUT_HYBRID_PRIORITY(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_HYBRID_PRIORITY__SHIFT) & DPU_LUT_CFG_LUT_HYBRID_PRIORITY__MASK;
+}
+#define DPU_LUT_CFG_LUT_OFLOW_PRIORITY__MASK			0x00000020
+#define DPU_LUT_CFG_LUT_OFLOW_PRIORITY__SHIFT			5
+static inline uint32_t DPU_LUT_CFG_LUT_OFLOW_PRIORITY(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_OFLOW_PRIORITY__SHIFT) & DPU_LUT_CFG_LUT_OFLOW_PRIORITY__MASK;
+}
+#define DPU_LUT_CFG_LUT_UFLOW_PRIORITY__MASK			0x00000010
+#define DPU_LUT_CFG_LUT_UFLOW_PRIORITY__SHIFT			4
+static inline uint32_t DPU_LUT_CFG_LUT_UFLOW_PRIORITY(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_UFLOW_PRIORITY__SHIFT) & DPU_LUT_CFG_LUT_UFLOW_PRIORITY__MASK;
+}
+#define DPU_LUT_CFG_LUT_LO_LE_MUX__MASK				0x0000000c
+#define DPU_LUT_CFG_LUT_LO_LE_MUX__SHIFT			2
+static inline uint32_t DPU_LUT_CFG_LUT_LO_LE_MUX(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_LO_LE_MUX__SHIFT) & DPU_LUT_CFG_LUT_LO_LE_MUX__MASK;
+}
+#define DPU_LUT_CFG_LUT_EXPAND_EN__MASK				0x00000002
+#define DPU_LUT_CFG_LUT_EXPAND_EN__SHIFT			1
+static inline uint32_t DPU_LUT_CFG_LUT_EXPAND_EN(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_EXPAND_EN__SHIFT) & DPU_LUT_CFG_LUT_EXPAND_EN__MASK;
+}
+#define DPU_LUT_CFG_LUT_ROAD_SEL__MASK				0x00000001
+#define DPU_LUT_CFG_LUT_ROAD_SEL__SHIFT				0
+static inline uint32_t DPU_LUT_CFG_LUT_ROAD_SEL(uint32_t val)
+{
+	return ((val) << DPU_LUT_CFG_LUT_ROAD_SEL__SHIFT) & DPU_LUT_CFG_LUT_ROAD_SEL__MASK;
+}
+
+#define REG_DPU_LUT_INFO					0x0000410c
+#define DPU_LUT_INFO_RESERVED_0__MASK				0xff000000
+#define DPU_LUT_INFO_RESERVED_0__SHIFT				24
+static inline uint32_t DPU_LUT_INFO_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_LUT_INFO_RESERVED_0__SHIFT) & DPU_LUT_INFO_RESERVED_0__MASK;
+}
+#define DPU_LUT_INFO_LUT_LO_INDEX_SELECT__MASK			0x00ff0000
+#define DPU_LUT_INFO_LUT_LO_INDEX_SELECT__SHIFT			16
+static inline uint32_t DPU_LUT_INFO_LUT_LO_INDEX_SELECT(uint32_t val)
+{
+	return ((val) << DPU_LUT_INFO_LUT_LO_INDEX_SELECT__SHIFT) & DPU_LUT_INFO_LUT_LO_INDEX_SELECT__MASK;
+}
+#define DPU_LUT_INFO_LUT_LE_INDEX_SELECT__MASK			0x0000ff00
+#define DPU_LUT_INFO_LUT_LE_INDEX_SELECT__SHIFT			8
+static inline uint32_t DPU_LUT_INFO_LUT_LE_INDEX_SELECT(uint32_t val)
+{
+	return ((val) << DPU_LUT_INFO_LUT_LE_INDEX_SELECT__SHIFT) & DPU_LUT_INFO_LUT_LE_INDEX_SELECT__MASK;
+}
+#define DPU_LUT_INFO_RESERVED_1__MASK				0x000000ff
+#define DPU_LUT_INFO_RESERVED_1__SHIFT				0
+static inline uint32_t DPU_LUT_INFO_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_LUT_INFO_RESERVED_1__SHIFT) & DPU_LUT_INFO_RESERVED_1__MASK;
+}
+
+#define REG_DPU_LUT_LE_START					0x00004110
+#define DPU_LUT_LE_START_LUT_LE_START__MASK			0xffffffff
+#define DPU_LUT_LE_START_LUT_LE_START__SHIFT			0
+static inline uint32_t DPU_LUT_LE_START_LUT_LE_START(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_START_LUT_LE_START__SHIFT) & DPU_LUT_LE_START_LUT_LE_START__MASK;
+}
+
+#define REG_DPU_LUT_LE_END					0x00004114
+#define DPU_LUT_LE_END_LUT_LE_END__MASK				0xffffffff
+#define DPU_LUT_LE_END_LUT_LE_END__SHIFT			0
+static inline uint32_t DPU_LUT_LE_END_LUT_LE_END(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_END_LUT_LE_END__SHIFT) & DPU_LUT_LE_END_LUT_LE_END__MASK;
+}
+
+#define REG_DPU_LUT_LO_START					0x00004118
+#define DPU_LUT_LO_START_LUT_LO_START__MASK			0xffffffff
+#define DPU_LUT_LO_START_LUT_LO_START__SHIFT			0
+static inline uint32_t DPU_LUT_LO_START_LUT_LO_START(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_START_LUT_LO_START__SHIFT) & DPU_LUT_LO_START_LUT_LO_START__MASK;
+}
+
+#define REG_DPU_LUT_LO_END					0x0000411c
+#define DPU_LUT_LO_END_LUT_LO_END__MASK				0xffffffff
+#define DPU_LUT_LO_END_LUT_LO_END__SHIFT			0
+static inline uint32_t DPU_LUT_LO_END_LUT_LO_END(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_END_LUT_LO_END__SHIFT) & DPU_LUT_LO_END_LUT_LO_END__MASK;
+}
+
+#define REG_DPU_LUT_LE_SLOPE_SCALE				0x00004120
+#define DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_OFLOW_SCALE__MASK	0xffff0000
+#define DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_OFLOW_SCALE__SHIFT	16
+static inline uint32_t DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_OFLOW_SCALE(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_OFLOW_SCALE__SHIFT) & DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_OFLOW_SCALE__MASK;
+}
+#define DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_UFLOW_SCALE__MASK	0x0000ffff
+#define DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_UFLOW_SCALE__SHIFT	0
+static inline uint32_t DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_UFLOW_SCALE(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_UFLOW_SCALE__SHIFT) & DPU_LUT_LE_SLOPE_SCALE_LUT_LE_SLOPE_UFLOW_SCALE__MASK;
+}
+
+#define REG_DPU_LUT_LE_SLOPE_SHIFT				0x00004124
+#define DPU_LUT_LE_SLOPE_SHIFT_RESERVED_0__MASK			0xfffffc00
+#define DPU_LUT_LE_SLOPE_SHIFT_RESERVED_0__SHIFT		10
+static inline uint32_t DPU_LUT_LE_SLOPE_SHIFT_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_SLOPE_SHIFT_RESERVED_0__SHIFT) & DPU_LUT_LE_SLOPE_SHIFT_RESERVED_0__MASK;
+}
+#define DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_OFLOW_SHIFT__MASK	0x000003e0
+#define DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_OFLOW_SHIFT__SHIFT	5
+static inline uint32_t DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_OFLOW_SHIFT(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_OFLOW_SHIFT__SHIFT) & DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_OFLOW_SHIFT__MASK;
+}
+#define DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_UFLOW_SHIFT__MASK	0x0000001f
+#define DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_UFLOW_SHIFT__SHIFT	0
+static inline uint32_t DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_UFLOW_SHIFT(uint32_t val)
+{
+	return ((val) << DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_UFLOW_SHIFT__SHIFT) & DPU_LUT_LE_SLOPE_SHIFT_LUT_LE_SLOPE_UFLOW_SHIFT__MASK;
+}
+
+#define REG_DPU_LUT_LO_SLOPE_SCALE				0x00004128
+#define DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_OFLOW_SCALE__MASK	0xffff0000
+#define DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_OFLOW_SCALE__SHIFT	16
+static inline uint32_t DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_OFLOW_SCALE(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_OFLOW_SCALE__SHIFT) & DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_OFLOW_SCALE__MASK;
+}
+#define DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_UFLOW_SCALE__MASK	0x0000ffff
+#define DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_UFLOW_SCALE__SHIFT	0
+static inline uint32_t DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_UFLOW_SCALE(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_UFLOW_SCALE__SHIFT) & DPU_LUT_LO_SLOPE_SCALE_LUT_LO_SLOPE_UFLOW_SCALE__MASK;
+}
+
+#define REG_DPU_LUT_LO_SLOPE_SHIFT				0x0000412c
+#define DPU_LUT_LO_SLOPE_SHIFT_RESERVED_0__MASK			0xfffffc00
+#define DPU_LUT_LO_SLOPE_SHIFT_RESERVED_0__SHIFT		10
+static inline uint32_t DPU_LUT_LO_SLOPE_SHIFT_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_SLOPE_SHIFT_RESERVED_0__SHIFT) & DPU_LUT_LO_SLOPE_SHIFT_RESERVED_0__MASK;
+}
+#define DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_OFLOW_SHIFT__MASK	0x000003e0
+#define DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_OFLOW_SHIFT__SHIFT	5
+static inline uint32_t DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_OFLOW_SHIFT(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_OFLOW_SHIFT__SHIFT) & DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_OFLOW_SHIFT__MASK;
+}
+#define DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_UFLOW_SHIFT__MASK	0x0000001f
+#define DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_UFLOW_SHIFT__SHIFT	0
+static inline uint32_t DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_UFLOW_SHIFT(uint32_t val)
+{
+	return ((val) << DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_UFLOW_SHIFT__SHIFT) & DPU_LUT_LO_SLOPE_SHIFT_LUT_LO_SLOPE_UFLOW_SHIFT__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_S_STATUS				0x00005000
+#define DPU_RDMA_RDMA_S_STATUS_RESERVED_0__MASK			0xfffc0000
+#define DPU_RDMA_RDMA_S_STATUS_RESERVED_0__SHIFT		18
+static inline uint32_t DPU_RDMA_RDMA_S_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_STATUS_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_S_STATUS_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_S_STATUS_STATUS_1__MASK			0x00030000
+#define DPU_RDMA_RDMA_S_STATUS_STATUS_1__SHIFT			16
+static inline uint32_t DPU_RDMA_RDMA_S_STATUS_STATUS_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_STATUS_STATUS_1__SHIFT) & DPU_RDMA_RDMA_S_STATUS_STATUS_1__MASK;
+}
+#define DPU_RDMA_RDMA_S_STATUS_RESERVED_1__MASK			0x0000fffc
+#define DPU_RDMA_RDMA_S_STATUS_RESERVED_1__SHIFT		2
+static inline uint32_t DPU_RDMA_RDMA_S_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_STATUS_RESERVED_1__SHIFT) & DPU_RDMA_RDMA_S_STATUS_RESERVED_1__MASK;
+}
+#define DPU_RDMA_RDMA_S_STATUS_STATUS_0__MASK			0x00000003
+#define DPU_RDMA_RDMA_S_STATUS_STATUS_0__SHIFT			0
+static inline uint32_t DPU_RDMA_RDMA_S_STATUS_STATUS_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_STATUS_STATUS_0__SHIFT) & DPU_RDMA_RDMA_S_STATUS_STATUS_0__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_S_POINTER				0x00005004
+#define DPU_RDMA_RDMA_S_POINTER_RESERVED_0__MASK		0xfffe0000
+#define DPU_RDMA_RDMA_S_POINTER_RESERVED_0__SHIFT		17
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_S_POINTER_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_EXECUTER__MASK			0x00010000
+#define DPU_RDMA_RDMA_S_POINTER_EXECUTER__SHIFT			16
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_EXECUTER(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_EXECUTER__SHIFT) & DPU_RDMA_RDMA_S_POINTER_EXECUTER__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_RESERVED_1__MASK		0x0000ffc0
+#define DPU_RDMA_RDMA_S_POINTER_RESERVED_1__SHIFT		6
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_RESERVED_1__SHIFT) & DPU_RDMA_RDMA_S_POINTER_RESERVED_1__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__MASK		0x00000020
+#define DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__SHIFT	5
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__SHIFT) & DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__MASK		0x00000010
+#define DPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__SHIFT		4
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__SHIFT) & DPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__MASK		0x00000008
+#define DPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__SHIFT		3
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__SHIFT) & DPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__MASK		0x00000004
+#define DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__SHIFT		2
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__SHIFT) & DPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__MASK		0x00000002
+#define DPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__SHIFT		1
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__SHIFT) & DPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__MASK;
+}
+#define DPU_RDMA_RDMA_S_POINTER_POINTER__MASK			0x00000001
+#define DPU_RDMA_RDMA_S_POINTER_POINTER__SHIFT			0
+static inline uint32_t DPU_RDMA_RDMA_S_POINTER_POINTER(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_S_POINTER_POINTER__SHIFT) & DPU_RDMA_RDMA_S_POINTER_POINTER__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_OPERATION_ENABLE			0x00005008
+#define DPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__MASK		0xfffffffe
+#define DPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__SHIFT	1
+static inline uint32_t DPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__MASK		0x00000001
+#define DPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__SHIFT) & DPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_DATA_CUBE_WIDTH			0x0000500c
+#define DPU_RDMA_RDMA_DATA_CUBE_WIDTH_RESERVED_0__MASK		0xffffe000
+#define DPU_RDMA_RDMA_DATA_CUBE_WIDTH_RESERVED_0__SHIFT		13
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_WIDTH_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_WIDTH_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_WIDTH_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_DATA_CUBE_WIDTH_WIDTH__MASK		0x00001fff
+#define DPU_RDMA_RDMA_DATA_CUBE_WIDTH_WIDTH__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_WIDTH_WIDTH(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_WIDTH_WIDTH__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_WIDTH_WIDTH__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_DATA_CUBE_HEIGHT			0x00005010
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_0__MASK		0xe0000000
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_0__SHIFT	29
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_EW_LINE_NOTCH_ADDR__MASK	0x1fff0000
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_EW_LINE_NOTCH_ADDR__SHIFT	16
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_EW_LINE_NOTCH_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_EW_LINE_NOTCH_ADDR__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_EW_LINE_NOTCH_ADDR__MASK;
+}
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_1__MASK		0x0000e000
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_1__SHIFT	13
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_1__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_RESERVED_1__MASK;
+}
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_HEIGHT__MASK		0x00001fff
+#define DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_HEIGHT__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_HEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_HEIGHT__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_HEIGHT_HEIGHT__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_DATA_CUBE_CHANNEL			0x00005014
+#define DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_RESERVED_0__MASK	0xffffe000
+#define DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_RESERVED_0__SHIFT	13
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_CHANNEL__MASK		0x00001fff
+#define DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_CHANNEL__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_CHANNEL(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_CHANNEL__SHIFT) & DPU_RDMA_RDMA_DATA_CUBE_CHANNEL_CHANNEL__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_SRC_BASE_ADDR				0x00005018
+#define DPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__MASK		0xffffffff
+#define DPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__SHIFT	0
+static inline uint32_t DPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__SHIFT) & DPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_BRDMA_CFG				0x0000501c
+#define DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_0__MASK		0xffffffe0
+#define DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_0__SHIFT		5
+static inline uint32_t DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_BRDMA_CFG_BRDMA_DATA_USE__MASK		0x0000001e
+#define DPU_RDMA_RDMA_BRDMA_CFG_BRDMA_DATA_USE__SHIFT		1
+static inline uint32_t DPU_RDMA_RDMA_BRDMA_CFG_BRDMA_DATA_USE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_BRDMA_CFG_BRDMA_DATA_USE__SHIFT) & DPU_RDMA_RDMA_BRDMA_CFG_BRDMA_DATA_USE__MASK;
+}
+#define DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_1__MASK		0x00000001
+#define DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_1__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_1__SHIFT) & DPU_RDMA_RDMA_BRDMA_CFG_RESERVED_1__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_BS_BASE_ADDR				0x00005020
+#define DPU_RDMA_RDMA_BS_BASE_ADDR_BS_BASE_ADDR__MASK		0xffffffff
+#define DPU_RDMA_RDMA_BS_BASE_ADDR_BS_BASE_ADDR__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_BS_BASE_ADDR_BS_BASE_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_BS_BASE_ADDR_BS_BASE_ADDR__SHIFT) & DPU_RDMA_RDMA_BS_BASE_ADDR_BS_BASE_ADDR__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_NRDMA_CFG				0x00005028
+#define DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_0__MASK		0xffffffe0
+#define DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_0__SHIFT		5
+static inline uint32_t DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_NRDMA_CFG_NRDMA_DATA_USE__MASK		0x0000001e
+#define DPU_RDMA_RDMA_NRDMA_CFG_NRDMA_DATA_USE__SHIFT		1
+static inline uint32_t DPU_RDMA_RDMA_NRDMA_CFG_NRDMA_DATA_USE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_NRDMA_CFG_NRDMA_DATA_USE__SHIFT) & DPU_RDMA_RDMA_NRDMA_CFG_NRDMA_DATA_USE__MASK;
+}
+#define DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_1__MASK		0x00000001
+#define DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_1__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_1__SHIFT) & DPU_RDMA_RDMA_NRDMA_CFG_RESERVED_1__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_BN_BASE_ADDR				0x0000502c
+#define DPU_RDMA_RDMA_BN_BASE_ADDR_BN_BASE_ADDR__MASK		0xffffffff
+#define DPU_RDMA_RDMA_BN_BASE_ADDR_BN_BASE_ADDR__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_BN_BASE_ADDR_BN_BASE_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_BN_BASE_ADDR_BN_BASE_ADDR__SHIFT) & DPU_RDMA_RDMA_BN_BASE_ADDR_BN_BASE_ADDR__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_ERDMA_CFG				0x00005034
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_MODE__MASK		0xc0000000
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_MODE__SHIFT		30
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_MODE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_MODE__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_MODE__MASK;
+}
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_SURF_MODE__MASK		0x20000000
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_SURF_MODE__SHIFT		29
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_SURF_MODE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_SURF_MODE__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_SURF_MODE__MASK;
+}
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_NONALIGN__MASK		0x10000000
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_NONALIGN__SHIFT		28
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_NONALIGN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_NONALIGN__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_NONALIGN__MASK;
+}
+#define DPU_RDMA_RDMA_ERDMA_CFG_RESERVED_0__MASK		0x0ffffff0
+#define DPU_RDMA_RDMA_ERDMA_CFG_RESERVED_0__SHIFT		4
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_SIZE__MASK		0x0000000c
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_SIZE__SHIFT		2
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_SIZE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_SIZE__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DATA_SIZE__MASK;
+}
+#define DPU_RDMA_RDMA_ERDMA_CFG_OV4K_BYPASS__MASK		0x00000002
+#define DPU_RDMA_RDMA_ERDMA_CFG_OV4K_BYPASS__SHIFT		1
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_OV4K_BYPASS(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_OV4K_BYPASS__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_OV4K_BYPASS__MASK;
+}
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DISABLE__MASK		0x00000001
+#define DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DISABLE__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DISABLE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DISABLE__SHIFT) & DPU_RDMA_RDMA_ERDMA_CFG_ERDMA_DISABLE__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_EW_BASE_ADDR				0x00005038
+#define DPU_RDMA_RDMA_EW_BASE_ADDR_EW_BASE_ADDR__MASK		0xffffffff
+#define DPU_RDMA_RDMA_EW_BASE_ADDR_EW_BASE_ADDR__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_EW_BASE_ADDR_EW_BASE_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_EW_BASE_ADDR_EW_BASE_ADDR__SHIFT) & DPU_RDMA_RDMA_EW_BASE_ADDR_EW_BASE_ADDR__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_EW_SURF_STRIDE			0x00005040
+#define DPU_RDMA_RDMA_EW_SURF_STRIDE_EW_SURF_STRIDE__MASK	0xfffffff0
+#define DPU_RDMA_RDMA_EW_SURF_STRIDE_EW_SURF_STRIDE__SHIFT	4
+static inline uint32_t DPU_RDMA_RDMA_EW_SURF_STRIDE_EW_SURF_STRIDE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_EW_SURF_STRIDE_EW_SURF_STRIDE__SHIFT) & DPU_RDMA_RDMA_EW_SURF_STRIDE_EW_SURF_STRIDE__MASK;
+}
+#define DPU_RDMA_RDMA_EW_SURF_STRIDE_RESERVED_0__MASK		0x0000000f
+#define DPU_RDMA_RDMA_EW_SURF_STRIDE_RESERVED_0__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_EW_SURF_STRIDE_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_EW_SURF_STRIDE_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_EW_SURF_STRIDE_RESERVED_0__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_FEATURE_MODE_CFG			0x00005044
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_RESERVED_0__MASK		0xfffc0000
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_RESERVED_0__SHIFT	18
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_IN_PRECISION__MASK	0x00038000
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_IN_PRECISION__SHIFT	15
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_IN_PRECISION(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_IN_PRECISION__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_IN_PRECISION__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_BURST_LEN__MASK		0x00007800
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_BURST_LEN__SHIFT		11
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_BURST_LEN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_BURST_LEN__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_BURST_LEN__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_COMB_USE__MASK		0x00000700
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_COMB_USE__SHIFT		8
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_COMB_USE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_COMB_USE__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_COMB_USE__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_PROC_PRECISION__MASK	0x000000e0
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_PROC_PRECISION__SHIFT	5
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_PROC_PRECISION(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_PROC_PRECISION__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_PROC_PRECISION__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_DISABLE__MASK	0x00000010
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_DISABLE__SHIFT	4
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_DISABLE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_DISABLE__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_DISABLE__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_FP16TOFP32_EN__MASK	0x00000008
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_FP16TOFP32_EN__SHIFT	3
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_FP16TOFP32_EN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_FP16TOFP32_EN__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_MRDMA_FP16TOFP32_EN__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_CONV_MODE__MASK		0x00000006
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_CONV_MODE__SHIFT		1
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_CONV_MODE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_CONV_MODE__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_CONV_MODE__MASK;
+}
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_FLYING_MODE__MASK	0x00000001
+#define DPU_RDMA_RDMA_FEATURE_MODE_CFG_FLYING_MODE__SHIFT	0
+static inline uint32_t DPU_RDMA_RDMA_FEATURE_MODE_CFG_FLYING_MODE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_FEATURE_MODE_CFG_FLYING_MODE__SHIFT) & DPU_RDMA_RDMA_FEATURE_MODE_CFG_FLYING_MODE__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_SRC_DMA_CFG				0x00005048
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_LINE_NOTCH_ADDR__MASK		0xfff80000
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_LINE_NOTCH_ADDR__SHIFT	19
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_LINE_NOTCH_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_LINE_NOTCH_ADDR__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_LINE_NOTCH_ADDR__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_RESERVED_0__MASK		0x0007c000
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_RESERVED_0__SHIFT		14
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_POOLING_METHOD__MASK		0x00002000
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_POOLING_METHOD__SHIFT		13
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_POOLING_METHOD(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_POOLING_METHOD__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_POOLING_METHOD__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_UNPOOLING_EN__MASK		0x00001000
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_UNPOOLING_EN__SHIFT		12
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_UNPOOLING_EN(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_UNPOOLING_EN__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_UNPOOLING_EN__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_HEIGHT__MASK	0x00000e00
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_HEIGHT__SHIFT	9
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_HEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_HEIGHT__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_HEIGHT__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_WIDTH__MASK	0x000001c0
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_WIDTH__SHIFT	6
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_WIDTH(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_WIDTH__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_STRIDE_WIDTH__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_HEIGHT__MASK		0x00000038
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_HEIGHT__SHIFT		3
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_HEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_HEIGHT__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_HEIGHT__MASK;
+}
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_WIDTH__MASK		0x00000007
+#define DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_WIDTH__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_WIDTH(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_WIDTH__SHIFT) & DPU_RDMA_RDMA_SRC_DMA_CFG_KERNEL_WIDTH__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_SURF_NOTCH				0x0000504c
+#define DPU_RDMA_RDMA_SURF_NOTCH_SURF_NOTCH_ADDR__MASK		0xfffffff0
+#define DPU_RDMA_RDMA_SURF_NOTCH_SURF_NOTCH_ADDR__SHIFT		4
+static inline uint32_t DPU_RDMA_RDMA_SURF_NOTCH_SURF_NOTCH_ADDR(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SURF_NOTCH_SURF_NOTCH_ADDR__SHIFT) & DPU_RDMA_RDMA_SURF_NOTCH_SURF_NOTCH_ADDR__MASK;
+}
+#define DPU_RDMA_RDMA_SURF_NOTCH_RESERVED_0__MASK		0x0000000f
+#define DPU_RDMA_RDMA_SURF_NOTCH_RESERVED_0__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_SURF_NOTCH_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_SURF_NOTCH_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_SURF_NOTCH_RESERVED_0__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_PAD_CFG				0x00005064
+#define DPU_RDMA_RDMA_PAD_CFG_PAD_VALUE__MASK			0xffff0000
+#define DPU_RDMA_RDMA_PAD_CFG_PAD_VALUE__SHIFT			16
+static inline uint32_t DPU_RDMA_RDMA_PAD_CFG_PAD_VALUE(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_PAD_CFG_PAD_VALUE__SHIFT) & DPU_RDMA_RDMA_PAD_CFG_PAD_VALUE__MASK;
+}
+#define DPU_RDMA_RDMA_PAD_CFG_RESERVED_0__MASK			0x0000ff80
+#define DPU_RDMA_RDMA_PAD_CFG_RESERVED_0__SHIFT			7
+static inline uint32_t DPU_RDMA_RDMA_PAD_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_PAD_CFG_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_PAD_CFG_RESERVED_0__MASK;
+}
+#define DPU_RDMA_RDMA_PAD_CFG_PAD_TOP__MASK			0x00000070
+#define DPU_RDMA_RDMA_PAD_CFG_PAD_TOP__SHIFT			4
+static inline uint32_t DPU_RDMA_RDMA_PAD_CFG_PAD_TOP(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_PAD_CFG_PAD_TOP__SHIFT) & DPU_RDMA_RDMA_PAD_CFG_PAD_TOP__MASK;
+}
+#define DPU_RDMA_RDMA_PAD_CFG_RESERVED_1__MASK			0x00000008
+#define DPU_RDMA_RDMA_PAD_CFG_RESERVED_1__SHIFT			3
+static inline uint32_t DPU_RDMA_RDMA_PAD_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_PAD_CFG_RESERVED_1__SHIFT) & DPU_RDMA_RDMA_PAD_CFG_RESERVED_1__MASK;
+}
+#define DPU_RDMA_RDMA_PAD_CFG_PAD_LEFT__MASK			0x00000007
+#define DPU_RDMA_RDMA_PAD_CFG_PAD_LEFT__SHIFT			0
+static inline uint32_t DPU_RDMA_RDMA_PAD_CFG_PAD_LEFT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_PAD_CFG_PAD_LEFT__SHIFT) & DPU_RDMA_RDMA_PAD_CFG_PAD_LEFT__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_WEIGHT				0x00005068
+#define DPU_RDMA_RDMA_WEIGHT_E_WEIGHT__MASK			0xff000000
+#define DPU_RDMA_RDMA_WEIGHT_E_WEIGHT__SHIFT			24
+static inline uint32_t DPU_RDMA_RDMA_WEIGHT_E_WEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_WEIGHT_E_WEIGHT__SHIFT) & DPU_RDMA_RDMA_WEIGHT_E_WEIGHT__MASK;
+}
+#define DPU_RDMA_RDMA_WEIGHT_N_WEIGHT__MASK			0x00ff0000
+#define DPU_RDMA_RDMA_WEIGHT_N_WEIGHT__SHIFT			16
+static inline uint32_t DPU_RDMA_RDMA_WEIGHT_N_WEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_WEIGHT_N_WEIGHT__SHIFT) & DPU_RDMA_RDMA_WEIGHT_N_WEIGHT__MASK;
+}
+#define DPU_RDMA_RDMA_WEIGHT_B_WEIGHT__MASK			0x0000ff00
+#define DPU_RDMA_RDMA_WEIGHT_B_WEIGHT__SHIFT			8
+static inline uint32_t DPU_RDMA_RDMA_WEIGHT_B_WEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_WEIGHT_B_WEIGHT__SHIFT) & DPU_RDMA_RDMA_WEIGHT_B_WEIGHT__MASK;
+}
+#define DPU_RDMA_RDMA_WEIGHT_M_WEIGHT__MASK			0x000000ff
+#define DPU_RDMA_RDMA_WEIGHT_M_WEIGHT__SHIFT			0
+static inline uint32_t DPU_RDMA_RDMA_WEIGHT_M_WEIGHT(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_WEIGHT_M_WEIGHT__SHIFT) & DPU_RDMA_RDMA_WEIGHT_M_WEIGHT__MASK;
+}
+
+#define REG_DPU_RDMA_RDMA_EW_SURF_NOTCH				0x0000506c
+#define DPU_RDMA_RDMA_EW_SURF_NOTCH_EW_SURF_NOTCH__MASK		0xfffffff0
+#define DPU_RDMA_RDMA_EW_SURF_NOTCH_EW_SURF_NOTCH__SHIFT	4
+static inline uint32_t DPU_RDMA_RDMA_EW_SURF_NOTCH_EW_SURF_NOTCH(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_EW_SURF_NOTCH_EW_SURF_NOTCH__SHIFT) & DPU_RDMA_RDMA_EW_SURF_NOTCH_EW_SURF_NOTCH__MASK;
+}
+#define DPU_RDMA_RDMA_EW_SURF_NOTCH_RESERVED_0__MASK		0x0000000f
+#define DPU_RDMA_RDMA_EW_SURF_NOTCH_RESERVED_0__SHIFT		0
+static inline uint32_t DPU_RDMA_RDMA_EW_SURF_NOTCH_RESERVED_0(uint32_t val)
+{
+	return ((val) << DPU_RDMA_RDMA_EW_SURF_NOTCH_RESERVED_0__SHIFT) & DPU_RDMA_RDMA_EW_SURF_NOTCH_RESERVED_0__MASK;
+}
+
+#define REG_PPU_S_STATUS					0x00006000
+#define PPU_S_STATUS_RESERVED_0__MASK				0xfffc0000
+#define PPU_S_STATUS_RESERVED_0__SHIFT				18
+static inline uint32_t PPU_S_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_S_STATUS_RESERVED_0__SHIFT) & PPU_S_STATUS_RESERVED_0__MASK;
+}
+#define PPU_S_STATUS_STATUS_1__MASK				0x00030000
+#define PPU_S_STATUS_STATUS_1__SHIFT				16
+static inline uint32_t PPU_S_STATUS_STATUS_1(uint32_t val)
+{
+	return ((val) << PPU_S_STATUS_STATUS_1__SHIFT) & PPU_S_STATUS_STATUS_1__MASK;
+}
+#define PPU_S_STATUS_RESERVED_1__MASK				0x0000fffc
+#define PPU_S_STATUS_RESERVED_1__SHIFT				2
+static inline uint32_t PPU_S_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_S_STATUS_RESERVED_1__SHIFT) & PPU_S_STATUS_RESERVED_1__MASK;
+}
+#define PPU_S_STATUS_STATUS_0__MASK				0x00000003
+#define PPU_S_STATUS_STATUS_0__SHIFT				0
+static inline uint32_t PPU_S_STATUS_STATUS_0(uint32_t val)
+{
+	return ((val) << PPU_S_STATUS_STATUS_0__SHIFT) & PPU_S_STATUS_STATUS_0__MASK;
+}
+
+#define REG_PPU_S_POINTER					0x00006004
+#define PPU_S_POINTER_RESERVED_0__MASK				0xfffe0000
+#define PPU_S_POINTER_RESERVED_0__SHIFT				17
+static inline uint32_t PPU_S_POINTER_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_RESERVED_0__SHIFT) & PPU_S_POINTER_RESERVED_0__MASK;
+}
+#define PPU_S_POINTER_EXECUTER__MASK				0x00010000
+#define PPU_S_POINTER_EXECUTER__SHIFT				16
+static inline uint32_t PPU_S_POINTER_EXECUTER(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_EXECUTER__SHIFT) & PPU_S_POINTER_EXECUTER__MASK;
+}
+#define PPU_S_POINTER_RESERVED_1__MASK				0x0000ffc0
+#define PPU_S_POINTER_RESERVED_1__SHIFT				6
+static inline uint32_t PPU_S_POINTER_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_RESERVED_1__SHIFT) & PPU_S_POINTER_RESERVED_1__MASK;
+}
+#define PPU_S_POINTER_EXECUTER_PP_CLEAR__MASK			0x00000020
+#define PPU_S_POINTER_EXECUTER_PP_CLEAR__SHIFT			5
+static inline uint32_t PPU_S_POINTER_EXECUTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_EXECUTER_PP_CLEAR__SHIFT) & PPU_S_POINTER_EXECUTER_PP_CLEAR__MASK;
+}
+#define PPU_S_POINTER_POINTER_PP_CLEAR__MASK			0x00000010
+#define PPU_S_POINTER_POINTER_PP_CLEAR__SHIFT			4
+static inline uint32_t PPU_S_POINTER_POINTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_POINTER_PP_CLEAR__SHIFT) & PPU_S_POINTER_POINTER_PP_CLEAR__MASK;
+}
+#define PPU_S_POINTER_POINTER_PP_MODE__MASK			0x00000008
+#define PPU_S_POINTER_POINTER_PP_MODE__SHIFT			3
+static inline uint32_t PPU_S_POINTER_POINTER_PP_MODE(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_POINTER_PP_MODE__SHIFT) & PPU_S_POINTER_POINTER_PP_MODE__MASK;
+}
+#define PPU_S_POINTER_EXECUTER_PP_EN__MASK			0x00000004
+#define PPU_S_POINTER_EXECUTER_PP_EN__SHIFT			2
+static inline uint32_t PPU_S_POINTER_EXECUTER_PP_EN(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_EXECUTER_PP_EN__SHIFT) & PPU_S_POINTER_EXECUTER_PP_EN__MASK;
+}
+#define PPU_S_POINTER_POINTER_PP_EN__MASK			0x00000002
+#define PPU_S_POINTER_POINTER_PP_EN__SHIFT			1
+static inline uint32_t PPU_S_POINTER_POINTER_PP_EN(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_POINTER_PP_EN__SHIFT) & PPU_S_POINTER_POINTER_PP_EN__MASK;
+}
+#define PPU_S_POINTER_POINTER__MASK				0x00000001
+#define PPU_S_POINTER_POINTER__SHIFT				0
+static inline uint32_t PPU_S_POINTER_POINTER(uint32_t val)
+{
+	return ((val) << PPU_S_POINTER_POINTER__SHIFT) & PPU_S_POINTER_POINTER__MASK;
+}
+
+#define REG_PPU_OPERATION_ENABLE				0x00006008
+#define PPU_OPERATION_ENABLE_RESERVED_0__MASK			0xfffffffe
+#define PPU_OPERATION_ENABLE_RESERVED_0__SHIFT			1
+static inline uint32_t PPU_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_ENABLE_RESERVED_0__SHIFT) & PPU_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define PPU_OPERATION_ENABLE_OP_EN__MASK			0x00000001
+#define PPU_OPERATION_ENABLE_OP_EN__SHIFT			0
+static inline uint32_t PPU_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_ENABLE_OP_EN__SHIFT) & PPU_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_PPU_DATA_CUBE_IN_WIDTH				0x0000600c
+#define PPU_DATA_CUBE_IN_WIDTH_RESERVED_0__MASK			0xffffe000
+#define PPU_DATA_CUBE_IN_WIDTH_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_DATA_CUBE_IN_WIDTH_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_IN_WIDTH_RESERVED_0__SHIFT) & PPU_DATA_CUBE_IN_WIDTH_RESERVED_0__MASK;
+}
+#define PPU_DATA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__MASK		0x00001fff
+#define PPU_DATA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__SHIFT		0
+static inline uint32_t PPU_DATA_CUBE_IN_WIDTH_CUBE_IN_WIDTH(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__SHIFT) & PPU_DATA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__MASK;
+}
+
+#define REG_PPU_DATA_CUBE_IN_HEIGHT				0x00006010
+#define PPU_DATA_CUBE_IN_HEIGHT_RESERVED_0__MASK		0xffffe000
+#define PPU_DATA_CUBE_IN_HEIGHT_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_DATA_CUBE_IN_HEIGHT_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_IN_HEIGHT_RESERVED_0__SHIFT) & PPU_DATA_CUBE_IN_HEIGHT_RESERVED_0__MASK;
+}
+#define PPU_DATA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__MASK		0x00001fff
+#define PPU_DATA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__SHIFT		0
+static inline uint32_t PPU_DATA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__SHIFT) & PPU_DATA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__MASK;
+}
+
+#define REG_PPU_DATA_CUBE_IN_CHANNEL				0x00006014
+#define PPU_DATA_CUBE_IN_CHANNEL_RESERVED_0__MASK		0xffffe000
+#define PPU_DATA_CUBE_IN_CHANNEL_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_DATA_CUBE_IN_CHANNEL_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_IN_CHANNEL_RESERVED_0__SHIFT) & PPU_DATA_CUBE_IN_CHANNEL_RESERVED_0__MASK;
+}
+#define PPU_DATA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__MASK		0x00001fff
+#define PPU_DATA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__SHIFT		0
+static inline uint32_t PPU_DATA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__SHIFT) & PPU_DATA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__MASK;
+}
+
+#define REG_PPU_DATA_CUBE_OUT_WIDTH				0x00006018
+#define PPU_DATA_CUBE_OUT_WIDTH_RESERVED_0__MASK		0xffffe000
+#define PPU_DATA_CUBE_OUT_WIDTH_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_DATA_CUBE_OUT_WIDTH_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_OUT_WIDTH_RESERVED_0__SHIFT) & PPU_DATA_CUBE_OUT_WIDTH_RESERVED_0__MASK;
+}
+#define PPU_DATA_CUBE_OUT_WIDTH_CUBE_OUT_WIDTH__MASK		0x00001fff
+#define PPU_DATA_CUBE_OUT_WIDTH_CUBE_OUT_WIDTH__SHIFT		0
+static inline uint32_t PPU_DATA_CUBE_OUT_WIDTH_CUBE_OUT_WIDTH(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_OUT_WIDTH_CUBE_OUT_WIDTH__SHIFT) & PPU_DATA_CUBE_OUT_WIDTH_CUBE_OUT_WIDTH__MASK;
+}
+
+#define REG_PPU_DATA_CUBE_OUT_HEIGHT				0x0000601c
+#define PPU_DATA_CUBE_OUT_HEIGHT_RESERVED_0__MASK		0xffffe000
+#define PPU_DATA_CUBE_OUT_HEIGHT_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_DATA_CUBE_OUT_HEIGHT_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_OUT_HEIGHT_RESERVED_0__SHIFT) & PPU_DATA_CUBE_OUT_HEIGHT_RESERVED_0__MASK;
+}
+#define PPU_DATA_CUBE_OUT_HEIGHT_CUBE_OUT_HEIGHT__MASK		0x00001fff
+#define PPU_DATA_CUBE_OUT_HEIGHT_CUBE_OUT_HEIGHT__SHIFT		0
+static inline uint32_t PPU_DATA_CUBE_OUT_HEIGHT_CUBE_OUT_HEIGHT(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_OUT_HEIGHT_CUBE_OUT_HEIGHT__SHIFT) & PPU_DATA_CUBE_OUT_HEIGHT_CUBE_OUT_HEIGHT__MASK;
+}
+
+#define REG_PPU_DATA_CUBE_OUT_CHANNEL				0x00006020
+#define PPU_DATA_CUBE_OUT_CHANNEL_RESERVED_0__MASK		0xffffe000
+#define PPU_DATA_CUBE_OUT_CHANNEL_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_DATA_CUBE_OUT_CHANNEL_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_OUT_CHANNEL_RESERVED_0__SHIFT) & PPU_DATA_CUBE_OUT_CHANNEL_RESERVED_0__MASK;
+}
+#define PPU_DATA_CUBE_OUT_CHANNEL_CUBE_OUT_CHANNEL__MASK	0x00001fff
+#define PPU_DATA_CUBE_OUT_CHANNEL_CUBE_OUT_CHANNEL__SHIFT	0
+static inline uint32_t PPU_DATA_CUBE_OUT_CHANNEL_CUBE_OUT_CHANNEL(uint32_t val)
+{
+	return ((val) << PPU_DATA_CUBE_OUT_CHANNEL_CUBE_OUT_CHANNEL__SHIFT) & PPU_DATA_CUBE_OUT_CHANNEL_CUBE_OUT_CHANNEL__MASK;
+}
+
+#define REG_PPU_OPERATION_MODE_CFG				0x00006024
+#define PPU_OPERATION_MODE_CFG_RESERVED_0__MASK			0x80000000
+#define PPU_OPERATION_MODE_CFG_RESERVED_0__SHIFT		31
+static inline uint32_t PPU_OPERATION_MODE_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_RESERVED_0__SHIFT) & PPU_OPERATION_MODE_CFG_RESERVED_0__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_INDEX_EN__MASK			0x40000000
+#define PPU_OPERATION_MODE_CFG_INDEX_EN__SHIFT			30
+static inline uint32_t PPU_OPERATION_MODE_CFG_INDEX_EN(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_INDEX_EN__SHIFT) & PPU_OPERATION_MODE_CFG_INDEX_EN__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_RESERVED_1__MASK			0x20000000
+#define PPU_OPERATION_MODE_CFG_RESERVED_1__SHIFT		29
+static inline uint32_t PPU_OPERATION_MODE_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_RESERVED_1__SHIFT) & PPU_OPERATION_MODE_CFG_RESERVED_1__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_NOTCH_ADDR__MASK			0x1fff0000
+#define PPU_OPERATION_MODE_CFG_NOTCH_ADDR__SHIFT		16
+static inline uint32_t PPU_OPERATION_MODE_CFG_NOTCH_ADDR(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_NOTCH_ADDR__SHIFT) & PPU_OPERATION_MODE_CFG_NOTCH_ADDR__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_RESERVED_2__MASK			0x0000ff00
+#define PPU_OPERATION_MODE_CFG_RESERVED_2__SHIFT		8
+static inline uint32_t PPU_OPERATION_MODE_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_RESERVED_2__SHIFT) & PPU_OPERATION_MODE_CFG_RESERVED_2__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_USE_CNT__MASK			0x000000e0
+#define PPU_OPERATION_MODE_CFG_USE_CNT__SHIFT			5
+static inline uint32_t PPU_OPERATION_MODE_CFG_USE_CNT(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_USE_CNT__SHIFT) & PPU_OPERATION_MODE_CFG_USE_CNT__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_FLYING_MODE__MASK		0x00000010
+#define PPU_OPERATION_MODE_CFG_FLYING_MODE__SHIFT		4
+static inline uint32_t PPU_OPERATION_MODE_CFG_FLYING_MODE(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_FLYING_MODE__SHIFT) & PPU_OPERATION_MODE_CFG_FLYING_MODE__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_RESERVED_3__MASK			0x0000000c
+#define PPU_OPERATION_MODE_CFG_RESERVED_3__SHIFT		2
+static inline uint32_t PPU_OPERATION_MODE_CFG_RESERVED_3(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_RESERVED_3__SHIFT) & PPU_OPERATION_MODE_CFG_RESERVED_3__MASK;
+}
+#define PPU_OPERATION_MODE_CFG_POOLING_METHOD__MASK		0x00000003
+#define PPU_OPERATION_MODE_CFG_POOLING_METHOD__SHIFT		0
+static inline uint32_t PPU_OPERATION_MODE_CFG_POOLING_METHOD(uint32_t val)
+{
+	return ((val) << PPU_OPERATION_MODE_CFG_POOLING_METHOD__SHIFT) & PPU_OPERATION_MODE_CFG_POOLING_METHOD__MASK;
+}
+
+#define REG_PPU_POOLING_KERNEL_CFG				0x00006034
+#define PPU_POOLING_KERNEL_CFG_RESERVED_0__MASK			0xff000000
+#define PPU_POOLING_KERNEL_CFG_RESERVED_0__SHIFT		24
+static inline uint32_t PPU_POOLING_KERNEL_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_RESERVED_0__SHIFT) & PPU_POOLING_KERNEL_CFG_RESERVED_0__MASK;
+}
+#define PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_HEIGHT__MASK	0x00f00000
+#define PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_HEIGHT__SHIFT	20
+static inline uint32_t PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_HEIGHT(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_HEIGHT__SHIFT) & PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_HEIGHT__MASK;
+}
+#define PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_WIDTH__MASK	0x000f0000
+#define PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_WIDTH__SHIFT	16
+static inline uint32_t PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_WIDTH(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_WIDTH__SHIFT) & PPU_POOLING_KERNEL_CFG_KERNEL_STRIDE_WIDTH__MASK;
+}
+#define PPU_POOLING_KERNEL_CFG_RESERVED_1__MASK			0x0000f000
+#define PPU_POOLING_KERNEL_CFG_RESERVED_1__SHIFT		12
+static inline uint32_t PPU_POOLING_KERNEL_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_RESERVED_1__SHIFT) & PPU_POOLING_KERNEL_CFG_RESERVED_1__MASK;
+}
+#define PPU_POOLING_KERNEL_CFG_KERNEL_HEIGHT__MASK		0x00000f00
+#define PPU_POOLING_KERNEL_CFG_KERNEL_HEIGHT__SHIFT		8
+static inline uint32_t PPU_POOLING_KERNEL_CFG_KERNEL_HEIGHT(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_KERNEL_HEIGHT__SHIFT) & PPU_POOLING_KERNEL_CFG_KERNEL_HEIGHT__MASK;
+}
+#define PPU_POOLING_KERNEL_CFG_RESERVED_2__MASK			0x000000f0
+#define PPU_POOLING_KERNEL_CFG_RESERVED_2__SHIFT		4
+static inline uint32_t PPU_POOLING_KERNEL_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_RESERVED_2__SHIFT) & PPU_POOLING_KERNEL_CFG_RESERVED_2__MASK;
+}
+#define PPU_POOLING_KERNEL_CFG_KERNEL_WIDTH__MASK		0x0000000f
+#define PPU_POOLING_KERNEL_CFG_KERNEL_WIDTH__SHIFT		0
+static inline uint32_t PPU_POOLING_KERNEL_CFG_KERNEL_WIDTH(uint32_t val)
+{
+	return ((val) << PPU_POOLING_KERNEL_CFG_KERNEL_WIDTH__SHIFT) & PPU_POOLING_KERNEL_CFG_KERNEL_WIDTH__MASK;
+}
+
+#define REG_PPU_RECIP_KERNEL_WIDTH				0x00006038
+#define PPU_RECIP_KERNEL_WIDTH_RESERVED_0__MASK			0xfffe0000
+#define PPU_RECIP_KERNEL_WIDTH_RESERVED_0__SHIFT		17
+static inline uint32_t PPU_RECIP_KERNEL_WIDTH_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RECIP_KERNEL_WIDTH_RESERVED_0__SHIFT) & PPU_RECIP_KERNEL_WIDTH_RESERVED_0__MASK;
+}
+#define PPU_RECIP_KERNEL_WIDTH_RECIP_KERNEL_WIDTH__MASK		0x0001ffff
+#define PPU_RECIP_KERNEL_WIDTH_RECIP_KERNEL_WIDTH__SHIFT	0
+static inline uint32_t PPU_RECIP_KERNEL_WIDTH_RECIP_KERNEL_WIDTH(uint32_t val)
+{
+	return ((val) << PPU_RECIP_KERNEL_WIDTH_RECIP_KERNEL_WIDTH__SHIFT) & PPU_RECIP_KERNEL_WIDTH_RECIP_KERNEL_WIDTH__MASK;
+}
+
+#define REG_PPU_RECIP_KERNEL_HEIGHT				0x0000603c
+#define PPU_RECIP_KERNEL_HEIGHT_RESERVED_0__MASK		0xfffe0000
+#define PPU_RECIP_KERNEL_HEIGHT_RESERVED_0__SHIFT		17
+static inline uint32_t PPU_RECIP_KERNEL_HEIGHT_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RECIP_KERNEL_HEIGHT_RESERVED_0__SHIFT) & PPU_RECIP_KERNEL_HEIGHT_RESERVED_0__MASK;
+}
+#define PPU_RECIP_KERNEL_HEIGHT_RECIP_KERNEL_HEIGHT__MASK	0x0001ffff
+#define PPU_RECIP_KERNEL_HEIGHT_RECIP_KERNEL_HEIGHT__SHIFT	0
+static inline uint32_t PPU_RECIP_KERNEL_HEIGHT_RECIP_KERNEL_HEIGHT(uint32_t val)
+{
+	return ((val) << PPU_RECIP_KERNEL_HEIGHT_RECIP_KERNEL_HEIGHT__SHIFT) & PPU_RECIP_KERNEL_HEIGHT_RECIP_KERNEL_HEIGHT__MASK;
+}
+
+#define REG_PPU_POOLING_PADDING_CFG				0x00006040
+#define PPU_POOLING_PADDING_CFG_RESERVED_0__MASK		0xffff8000
+#define PPU_POOLING_PADDING_CFG_RESERVED_0__SHIFT		15
+static inline uint32_t PPU_POOLING_PADDING_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_RESERVED_0__SHIFT) & PPU_POOLING_PADDING_CFG_RESERVED_0__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_PAD_BOTTOM__MASK		0x00007000
+#define PPU_POOLING_PADDING_CFG_PAD_BOTTOM__SHIFT		12
+static inline uint32_t PPU_POOLING_PADDING_CFG_PAD_BOTTOM(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_PAD_BOTTOM__SHIFT) & PPU_POOLING_PADDING_CFG_PAD_BOTTOM__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_RESERVED_1__MASK		0x00000800
+#define PPU_POOLING_PADDING_CFG_RESERVED_1__SHIFT		11
+static inline uint32_t PPU_POOLING_PADDING_CFG_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_RESERVED_1__SHIFT) & PPU_POOLING_PADDING_CFG_RESERVED_1__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_PAD_RIGHT__MASK			0x00000700
+#define PPU_POOLING_PADDING_CFG_PAD_RIGHT__SHIFT		8
+static inline uint32_t PPU_POOLING_PADDING_CFG_PAD_RIGHT(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_PAD_RIGHT__SHIFT) & PPU_POOLING_PADDING_CFG_PAD_RIGHT__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_RESERVED_2__MASK		0x00000080
+#define PPU_POOLING_PADDING_CFG_RESERVED_2__SHIFT		7
+static inline uint32_t PPU_POOLING_PADDING_CFG_RESERVED_2(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_RESERVED_2__SHIFT) & PPU_POOLING_PADDING_CFG_RESERVED_2__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_PAD_TOP__MASK			0x00000070
+#define PPU_POOLING_PADDING_CFG_PAD_TOP__SHIFT			4
+static inline uint32_t PPU_POOLING_PADDING_CFG_PAD_TOP(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_PAD_TOP__SHIFT) & PPU_POOLING_PADDING_CFG_PAD_TOP__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_RESERVED_3__MASK		0x00000008
+#define PPU_POOLING_PADDING_CFG_RESERVED_3__SHIFT		3
+static inline uint32_t PPU_POOLING_PADDING_CFG_RESERVED_3(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_RESERVED_3__SHIFT) & PPU_POOLING_PADDING_CFG_RESERVED_3__MASK;
+}
+#define PPU_POOLING_PADDING_CFG_PAD_LEFT__MASK			0x00000007
+#define PPU_POOLING_PADDING_CFG_PAD_LEFT__SHIFT			0
+static inline uint32_t PPU_POOLING_PADDING_CFG_PAD_LEFT(uint32_t val)
+{
+	return ((val) << PPU_POOLING_PADDING_CFG_PAD_LEFT__SHIFT) & PPU_POOLING_PADDING_CFG_PAD_LEFT__MASK;
+}
+
+#define REG_PPU_PADDING_VALUE_1_CFG				0x00006044
+#define PPU_PADDING_VALUE_1_CFG_PAD_VALUE_0__MASK		0xffffffff
+#define PPU_PADDING_VALUE_1_CFG_PAD_VALUE_0__SHIFT		0
+static inline uint32_t PPU_PADDING_VALUE_1_CFG_PAD_VALUE_0(uint32_t val)
+{
+	return ((val) << PPU_PADDING_VALUE_1_CFG_PAD_VALUE_0__SHIFT) & PPU_PADDING_VALUE_1_CFG_PAD_VALUE_0__MASK;
+}
+
+#define REG_PPU_PADDING_VALUE_2_CFG				0x00006048
+#define PPU_PADDING_VALUE_2_CFG_RESERVED_0__MASK		0xfffffff8
+#define PPU_PADDING_VALUE_2_CFG_RESERVED_0__SHIFT		3
+static inline uint32_t PPU_PADDING_VALUE_2_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_PADDING_VALUE_2_CFG_RESERVED_0__SHIFT) & PPU_PADDING_VALUE_2_CFG_RESERVED_0__MASK;
+}
+#define PPU_PADDING_VALUE_2_CFG_PAD_VALUE_1__MASK		0x00000007
+#define PPU_PADDING_VALUE_2_CFG_PAD_VALUE_1__SHIFT		0
+static inline uint32_t PPU_PADDING_VALUE_2_CFG_PAD_VALUE_1(uint32_t val)
+{
+	return ((val) << PPU_PADDING_VALUE_2_CFG_PAD_VALUE_1__SHIFT) & PPU_PADDING_VALUE_2_CFG_PAD_VALUE_1__MASK;
+}
+
+#define REG_PPU_DST_BASE_ADDR					0x00006070
+#define PPU_DST_BASE_ADDR_DST_BASE_ADDR__MASK			0xfffffff0
+#define PPU_DST_BASE_ADDR_DST_BASE_ADDR__SHIFT			4
+static inline uint32_t PPU_DST_BASE_ADDR_DST_BASE_ADDR(uint32_t val)
+{
+	return ((val) << PPU_DST_BASE_ADDR_DST_BASE_ADDR__SHIFT) & PPU_DST_BASE_ADDR_DST_BASE_ADDR__MASK;
+}
+#define PPU_DST_BASE_ADDR_RESERVED_0__MASK			0x0000000f
+#define PPU_DST_BASE_ADDR_RESERVED_0__SHIFT			0
+static inline uint32_t PPU_DST_BASE_ADDR_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DST_BASE_ADDR_RESERVED_0__SHIFT) & PPU_DST_BASE_ADDR_RESERVED_0__MASK;
+}
+
+#define REG_PPU_DST_SURF_STRIDE					0x0000607c
+#define PPU_DST_SURF_STRIDE_DST_SURF_STRIDE__MASK		0xfffffff0
+#define PPU_DST_SURF_STRIDE_DST_SURF_STRIDE__SHIFT		4
+static inline uint32_t PPU_DST_SURF_STRIDE_DST_SURF_STRIDE(uint32_t val)
+{
+	return ((val) << PPU_DST_SURF_STRIDE_DST_SURF_STRIDE__SHIFT) & PPU_DST_SURF_STRIDE_DST_SURF_STRIDE__MASK;
+}
+#define PPU_DST_SURF_STRIDE_RESERVED_0__MASK			0x0000000f
+#define PPU_DST_SURF_STRIDE_RESERVED_0__SHIFT			0
+static inline uint32_t PPU_DST_SURF_STRIDE_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_DST_SURF_STRIDE_RESERVED_0__SHIFT) & PPU_DST_SURF_STRIDE_RESERVED_0__MASK;
+}
+
+#define REG_PPU_DATA_FORMAT					0x00006084
+#define PPU_DATA_FORMAT_INDEX_ADD__MASK				0xfffffff0
+#define PPU_DATA_FORMAT_INDEX_ADD__SHIFT			4
+static inline uint32_t PPU_DATA_FORMAT_INDEX_ADD(uint32_t val)
+{
+	return ((val) << PPU_DATA_FORMAT_INDEX_ADD__SHIFT) & PPU_DATA_FORMAT_INDEX_ADD__MASK;
+}
+#define PPU_DATA_FORMAT_DPU_FLYIN__MASK				0x00000008
+#define PPU_DATA_FORMAT_DPU_FLYIN__SHIFT			3
+static inline uint32_t PPU_DATA_FORMAT_DPU_FLYIN(uint32_t val)
+{
+	return ((val) << PPU_DATA_FORMAT_DPU_FLYIN__SHIFT) & PPU_DATA_FORMAT_DPU_FLYIN__MASK;
+}
+#define PPU_DATA_FORMAT_PROC_PRECISION__MASK			0x00000007
+#define PPU_DATA_FORMAT_PROC_PRECISION__SHIFT			0
+static inline uint32_t PPU_DATA_FORMAT_PROC_PRECISION(uint32_t val)
+{
+	return ((val) << PPU_DATA_FORMAT_PROC_PRECISION__SHIFT) & PPU_DATA_FORMAT_PROC_PRECISION__MASK;
+}
+
+#define REG_PPU_MISC_CTRL					0x000060dc
+#define PPU_MISC_CTRL_SURF_LEN__MASK				0xffff0000
+#define PPU_MISC_CTRL_SURF_LEN__SHIFT				16
+static inline uint32_t PPU_MISC_CTRL_SURF_LEN(uint32_t val)
+{
+	return ((val) << PPU_MISC_CTRL_SURF_LEN__SHIFT) & PPU_MISC_CTRL_SURF_LEN__MASK;
+}
+#define PPU_MISC_CTRL_RESERVED_0__MASK				0x0000fe00
+#define PPU_MISC_CTRL_RESERVED_0__SHIFT				9
+static inline uint32_t PPU_MISC_CTRL_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_MISC_CTRL_RESERVED_0__SHIFT) & PPU_MISC_CTRL_RESERVED_0__MASK;
+}
+#define PPU_MISC_CTRL_MC_SURF_OUT__MASK				0x00000100
+#define PPU_MISC_CTRL_MC_SURF_OUT__SHIFT			8
+static inline uint32_t PPU_MISC_CTRL_MC_SURF_OUT(uint32_t val)
+{
+	return ((val) << PPU_MISC_CTRL_MC_SURF_OUT__SHIFT) & PPU_MISC_CTRL_MC_SURF_OUT__MASK;
+}
+#define PPU_MISC_CTRL_NONALIGN__MASK				0x00000080
+#define PPU_MISC_CTRL_NONALIGN__SHIFT				7
+static inline uint32_t PPU_MISC_CTRL_NONALIGN(uint32_t val)
+{
+	return ((val) << PPU_MISC_CTRL_NONALIGN__SHIFT) & PPU_MISC_CTRL_NONALIGN__MASK;
+}
+#define PPU_MISC_CTRL_RESERVED_1__MASK				0x00000070
+#define PPU_MISC_CTRL_RESERVED_1__SHIFT				4
+static inline uint32_t PPU_MISC_CTRL_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_MISC_CTRL_RESERVED_1__SHIFT) & PPU_MISC_CTRL_RESERVED_1__MASK;
+}
+#define PPU_MISC_CTRL_BURST_LEN__MASK				0x0000000f
+#define PPU_MISC_CTRL_BURST_LEN__SHIFT				0
+static inline uint32_t PPU_MISC_CTRL_BURST_LEN(uint32_t val)
+{
+	return ((val) << PPU_MISC_CTRL_BURST_LEN__SHIFT) & PPU_MISC_CTRL_BURST_LEN__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_S_STATUS				0x00007000
+#define PPU_RDMA_RDMA_S_STATUS_RESERVED_0__MASK			0xfffc0000
+#define PPU_RDMA_RDMA_S_STATUS_RESERVED_0__SHIFT		18
+static inline uint32_t PPU_RDMA_RDMA_S_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_STATUS_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_S_STATUS_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_S_STATUS_STATUS_1__MASK			0x00030000
+#define PPU_RDMA_RDMA_S_STATUS_STATUS_1__SHIFT			16
+static inline uint32_t PPU_RDMA_RDMA_S_STATUS_STATUS_1(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_STATUS_STATUS_1__SHIFT) & PPU_RDMA_RDMA_S_STATUS_STATUS_1__MASK;
+}
+#define PPU_RDMA_RDMA_S_STATUS_RESERVED_1__MASK			0x0000fffc
+#define PPU_RDMA_RDMA_S_STATUS_RESERVED_1__SHIFT		2
+static inline uint32_t PPU_RDMA_RDMA_S_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_STATUS_RESERVED_1__SHIFT) & PPU_RDMA_RDMA_S_STATUS_RESERVED_1__MASK;
+}
+#define PPU_RDMA_RDMA_S_STATUS_STATUS_0__MASK			0x00000003
+#define PPU_RDMA_RDMA_S_STATUS_STATUS_0__SHIFT			0
+static inline uint32_t PPU_RDMA_RDMA_S_STATUS_STATUS_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_STATUS_STATUS_0__SHIFT) & PPU_RDMA_RDMA_S_STATUS_STATUS_0__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_S_POINTER				0x00007004
+#define PPU_RDMA_RDMA_S_POINTER_RESERVED_0__MASK		0xfffe0000
+#define PPU_RDMA_RDMA_S_POINTER_RESERVED_0__SHIFT		17
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_S_POINTER_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_EXECUTER__MASK			0x00010000
+#define PPU_RDMA_RDMA_S_POINTER_EXECUTER__SHIFT			16
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_EXECUTER(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_EXECUTER__SHIFT) & PPU_RDMA_RDMA_S_POINTER_EXECUTER__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_RESERVED_1__MASK		0x0000ffc0
+#define PPU_RDMA_RDMA_S_POINTER_RESERVED_1__SHIFT		6
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_RESERVED_1(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_RESERVED_1__SHIFT) & PPU_RDMA_RDMA_S_POINTER_RESERVED_1__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__MASK		0x00000020
+#define PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__SHIFT	5
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__SHIFT) & PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_CLEAR__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__MASK		0x00000010
+#define PPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__SHIFT		4
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__SHIFT) & PPU_RDMA_RDMA_S_POINTER_POINTER_PP_CLEAR__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__MASK		0x00000008
+#define PPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__SHIFT		3
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__SHIFT) & PPU_RDMA_RDMA_S_POINTER_POINTER_PP_MODE__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__MASK		0x00000004
+#define PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__SHIFT		2
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__SHIFT) & PPU_RDMA_RDMA_S_POINTER_EXECUTER_PP_EN__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__MASK		0x00000002
+#define PPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__SHIFT		1
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__SHIFT) & PPU_RDMA_RDMA_S_POINTER_POINTER_PP_EN__MASK;
+}
+#define PPU_RDMA_RDMA_S_POINTER_POINTER__MASK			0x00000001
+#define PPU_RDMA_RDMA_S_POINTER_POINTER__SHIFT			0
+static inline uint32_t PPU_RDMA_RDMA_S_POINTER_POINTER(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_S_POINTER_POINTER__SHIFT) & PPU_RDMA_RDMA_S_POINTER_POINTER__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_OPERATION_ENABLE			0x00007008
+#define PPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__MASK		0xfffffffe
+#define PPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__SHIFT	1
+static inline uint32_t PPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__MASK		0x00000001
+#define PPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__SHIFT		0
+static inline uint32_t PPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__SHIFT) & PPU_RDMA_RDMA_OPERATION_ENABLE_OP_EN__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_CUBE_IN_WIDTH				0x0000700c
+#define PPU_RDMA_RDMA_CUBE_IN_WIDTH_RESERVED_0__MASK		0xffffe000
+#define PPU_RDMA_RDMA_CUBE_IN_WIDTH_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_RDMA_RDMA_CUBE_IN_WIDTH_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_CUBE_IN_WIDTH_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_CUBE_IN_WIDTH_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__MASK		0x00001fff
+#define PPU_RDMA_RDMA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__SHIFT	0
+static inline uint32_t PPU_RDMA_RDMA_CUBE_IN_WIDTH_CUBE_IN_WIDTH(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__SHIFT) & PPU_RDMA_RDMA_CUBE_IN_WIDTH_CUBE_IN_WIDTH__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_CUBE_IN_HEIGHT			0x00007010
+#define PPU_RDMA_RDMA_CUBE_IN_HEIGHT_RESERVED_0__MASK		0xffffe000
+#define PPU_RDMA_RDMA_CUBE_IN_HEIGHT_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_RDMA_RDMA_CUBE_IN_HEIGHT_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_CUBE_IN_HEIGHT_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_CUBE_IN_HEIGHT_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__MASK	0x00001fff
+#define PPU_RDMA_RDMA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__SHIFT	0
+static inline uint32_t PPU_RDMA_RDMA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__SHIFT) & PPU_RDMA_RDMA_CUBE_IN_HEIGHT_CUBE_IN_HEIGHT__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_CUBE_IN_CHANNEL			0x00007014
+#define PPU_RDMA_RDMA_CUBE_IN_CHANNEL_RESERVED_0__MASK		0xffffe000
+#define PPU_RDMA_RDMA_CUBE_IN_CHANNEL_RESERVED_0__SHIFT		13
+static inline uint32_t PPU_RDMA_RDMA_CUBE_IN_CHANNEL_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_CUBE_IN_CHANNEL_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_CUBE_IN_CHANNEL_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__MASK	0x00001fff
+#define PPU_RDMA_RDMA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__SHIFT	0
+static inline uint32_t PPU_RDMA_RDMA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__SHIFT) & PPU_RDMA_RDMA_CUBE_IN_CHANNEL_CUBE_IN_CHANNEL__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_SRC_BASE_ADDR				0x0000701c
+#define PPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__MASK		0xffffffff
+#define PPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__SHIFT	0
+static inline uint32_t PPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__SHIFT) & PPU_RDMA_RDMA_SRC_BASE_ADDR_SRC_BASE_ADDR__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_SRC_LINE_STRIDE			0x00007024
+#define PPU_RDMA_RDMA_SRC_LINE_STRIDE_SRC_LINE_STRIDE__MASK	0xfffffff0
+#define PPU_RDMA_RDMA_SRC_LINE_STRIDE_SRC_LINE_STRIDE__SHIFT	4
+static inline uint32_t PPU_RDMA_RDMA_SRC_LINE_STRIDE_SRC_LINE_STRIDE(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_SRC_LINE_STRIDE_SRC_LINE_STRIDE__SHIFT) & PPU_RDMA_RDMA_SRC_LINE_STRIDE_SRC_LINE_STRIDE__MASK;
+}
+#define PPU_RDMA_RDMA_SRC_LINE_STRIDE_RESERVED_0__MASK		0x0000000f
+#define PPU_RDMA_RDMA_SRC_LINE_STRIDE_RESERVED_0__SHIFT		0
+static inline uint32_t PPU_RDMA_RDMA_SRC_LINE_STRIDE_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_SRC_LINE_STRIDE_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_SRC_LINE_STRIDE_RESERVED_0__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_SRC_SURF_STRIDE			0x00007028
+#define PPU_RDMA_RDMA_SRC_SURF_STRIDE_SRC_SURF_STRIDE__MASK	0xfffffff0
+#define PPU_RDMA_RDMA_SRC_SURF_STRIDE_SRC_SURF_STRIDE__SHIFT	4
+static inline uint32_t PPU_RDMA_RDMA_SRC_SURF_STRIDE_SRC_SURF_STRIDE(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_SRC_SURF_STRIDE_SRC_SURF_STRIDE__SHIFT) & PPU_RDMA_RDMA_SRC_SURF_STRIDE_SRC_SURF_STRIDE__MASK;
+}
+#define PPU_RDMA_RDMA_SRC_SURF_STRIDE_RESERVED_0__MASK		0x0000000f
+#define PPU_RDMA_RDMA_SRC_SURF_STRIDE_RESERVED_0__SHIFT		0
+static inline uint32_t PPU_RDMA_RDMA_SRC_SURF_STRIDE_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_SRC_SURF_STRIDE_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_SRC_SURF_STRIDE_RESERVED_0__MASK;
+}
+
+#define REG_PPU_RDMA_RDMA_DATA_FORMAT				0x00007030
+#define PPU_RDMA_RDMA_DATA_FORMAT_RESERVED_0__MASK		0xfffffffc
+#define PPU_RDMA_RDMA_DATA_FORMAT_RESERVED_0__SHIFT		2
+static inline uint32_t PPU_RDMA_RDMA_DATA_FORMAT_RESERVED_0(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_DATA_FORMAT_RESERVED_0__SHIFT) & PPU_RDMA_RDMA_DATA_FORMAT_RESERVED_0__MASK;
+}
+#define PPU_RDMA_RDMA_DATA_FORMAT_IN_PRECISION__MASK		0x00000003
+#define PPU_RDMA_RDMA_DATA_FORMAT_IN_PRECISION__SHIFT		0
+static inline uint32_t PPU_RDMA_RDMA_DATA_FORMAT_IN_PRECISION(uint32_t val)
+{
+	return ((val) << PPU_RDMA_RDMA_DATA_FORMAT_IN_PRECISION__SHIFT) & PPU_RDMA_RDMA_DATA_FORMAT_IN_PRECISION__MASK;
+}
+
+#define REG_DDMA_CFG_OUTSTANDING				0x00008000
+#define DDMA_CFG_OUTSTANDING_RESERVED_0__MASK			0xffff0000
+#define DDMA_CFG_OUTSTANDING_RESERVED_0__SHIFT			16
+static inline uint32_t DDMA_CFG_OUTSTANDING_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_OUTSTANDING_RESERVED_0__SHIFT) & DDMA_CFG_OUTSTANDING_RESERVED_0__MASK;
+}
+#define DDMA_CFG_OUTSTANDING_WR_OS_CNT__MASK			0x0000ff00
+#define DDMA_CFG_OUTSTANDING_WR_OS_CNT__SHIFT			8
+static inline uint32_t DDMA_CFG_OUTSTANDING_WR_OS_CNT(uint32_t val)
+{
+	return ((val) << DDMA_CFG_OUTSTANDING_WR_OS_CNT__SHIFT) & DDMA_CFG_OUTSTANDING_WR_OS_CNT__MASK;
+}
+#define DDMA_CFG_OUTSTANDING_RD_OS_CNT__MASK			0x000000ff
+#define DDMA_CFG_OUTSTANDING_RD_OS_CNT__SHIFT			0
+static inline uint32_t DDMA_CFG_OUTSTANDING_RD_OS_CNT(uint32_t val)
+{
+	return ((val) << DDMA_CFG_OUTSTANDING_RD_OS_CNT__SHIFT) & DDMA_CFG_OUTSTANDING_RD_OS_CNT__MASK;
+}
+
+#define REG_DDMA_RD_WEIGHT_0					0x00008004
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__MASK			0xff000000
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__SHIFT			24
+static inline uint32_t DDMA_RD_WEIGHT_0_RD_WEIGHT_PDP(uint32_t val)
+{
+	return ((val) << DDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__SHIFT) & DDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__MASK;
+}
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__MASK			0x00ff0000
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__SHIFT			16
+static inline uint32_t DDMA_RD_WEIGHT_0_RD_WEIGHT_DPU(uint32_t val)
+{
+	return ((val) << DDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__SHIFT) & DDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__MASK;
+}
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__MASK			0x0000ff00
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__SHIFT		8
+static inline uint32_t DDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL(uint32_t val)
+{
+	return ((val) << DDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__SHIFT) & DDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__MASK;
+}
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__MASK		0x000000ff
+#define DDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__SHIFT		0
+static inline uint32_t DDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE(uint32_t val)
+{
+	return ((val) << DDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__SHIFT) & DDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__MASK;
+}
+
+#define REG_DDMA_WR_WEIGHT_0					0x00008008
+#define DDMA_WR_WEIGHT_0_RESERVED_0__MASK			0xffff0000
+#define DDMA_WR_WEIGHT_0_RESERVED_0__SHIFT			16
+static inline uint32_t DDMA_WR_WEIGHT_0_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_WR_WEIGHT_0_RESERVED_0__SHIFT) & DDMA_WR_WEIGHT_0_RESERVED_0__MASK;
+}
+#define DDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__MASK			0x0000ff00
+#define DDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__SHIFT			8
+static inline uint32_t DDMA_WR_WEIGHT_0_WR_WEIGHT_PDP(uint32_t val)
+{
+	return ((val) << DDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__SHIFT) & DDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__MASK;
+}
+#define DDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__MASK			0x000000ff
+#define DDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__SHIFT			0
+static inline uint32_t DDMA_WR_WEIGHT_0_WR_WEIGHT_DPU(uint32_t val)
+{
+	return ((val) << DDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__SHIFT) & DDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__MASK;
+}
+
+#define REG_DDMA_CFG_ID_ERROR					0x0000800c
+#define DDMA_CFG_ID_ERROR_RESERVED_0__MASK			0xfffffc00
+#define DDMA_CFG_ID_ERROR_RESERVED_0__SHIFT			10
+static inline uint32_t DDMA_CFG_ID_ERROR_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_ID_ERROR_RESERVED_0__SHIFT) & DDMA_CFG_ID_ERROR_RESERVED_0__MASK;
+}
+#define DDMA_CFG_ID_ERROR_WR_RESP_ID__MASK			0x000003c0
+#define DDMA_CFG_ID_ERROR_WR_RESP_ID__SHIFT			6
+static inline uint32_t DDMA_CFG_ID_ERROR_WR_RESP_ID(uint32_t val)
+{
+	return ((val) << DDMA_CFG_ID_ERROR_WR_RESP_ID__SHIFT) & DDMA_CFG_ID_ERROR_WR_RESP_ID__MASK;
+}
+#define DDMA_CFG_ID_ERROR_RESERVED_1__MASK			0x00000020
+#define DDMA_CFG_ID_ERROR_RESERVED_1__SHIFT			5
+static inline uint32_t DDMA_CFG_ID_ERROR_RESERVED_1(uint32_t val)
+{
+	return ((val) << DDMA_CFG_ID_ERROR_RESERVED_1__SHIFT) & DDMA_CFG_ID_ERROR_RESERVED_1__MASK;
+}
+#define DDMA_CFG_ID_ERROR_RD_RESP_ID__MASK			0x0000001f
+#define DDMA_CFG_ID_ERROR_RD_RESP_ID__SHIFT			0
+static inline uint32_t DDMA_CFG_ID_ERROR_RD_RESP_ID(uint32_t val)
+{
+	return ((val) << DDMA_CFG_ID_ERROR_RD_RESP_ID__SHIFT) & DDMA_CFG_ID_ERROR_RD_RESP_ID__MASK;
+}
+
+#define REG_DDMA_RD_WEIGHT_1					0x00008010
+#define DDMA_RD_WEIGHT_1_RESERVED_0__MASK			0xffffff00
+#define DDMA_RD_WEIGHT_1_RESERVED_0__SHIFT			8
+static inline uint32_t DDMA_RD_WEIGHT_1_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_RD_WEIGHT_1_RESERVED_0__SHIFT) & DDMA_RD_WEIGHT_1_RESERVED_0__MASK;
+}
+#define DDMA_RD_WEIGHT_1_RD_WEIGHT_PC__MASK			0x000000ff
+#define DDMA_RD_WEIGHT_1_RD_WEIGHT_PC__SHIFT			0
+static inline uint32_t DDMA_RD_WEIGHT_1_RD_WEIGHT_PC(uint32_t val)
+{
+	return ((val) << DDMA_RD_WEIGHT_1_RD_WEIGHT_PC__SHIFT) & DDMA_RD_WEIGHT_1_RD_WEIGHT_PC__MASK;
+}
+
+#define REG_DDMA_CFG_DMA_FIFO_CLR				0x00008014
+#define DDMA_CFG_DMA_FIFO_CLR_RESERVED_0__MASK			0xfffffffe
+#define DDMA_CFG_DMA_FIFO_CLR_RESERVED_0__SHIFT			1
+static inline uint32_t DDMA_CFG_DMA_FIFO_CLR_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_FIFO_CLR_RESERVED_0__SHIFT) & DDMA_CFG_DMA_FIFO_CLR_RESERVED_0__MASK;
+}
+#define DDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__MASK		0x00000001
+#define DDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__SHIFT		0
+static inline uint32_t DDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__SHIFT) & DDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__MASK;
+}
+
+#define REG_DDMA_CFG_DMA_ARB					0x00008018
+#define DDMA_CFG_DMA_ARB_RESERVED_0__MASK			0xfffffc00
+#define DDMA_CFG_DMA_ARB_RESERVED_0__SHIFT			10
+static inline uint32_t DDMA_CFG_DMA_ARB_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_RESERVED_0__SHIFT) & DDMA_CFG_DMA_ARB_RESERVED_0__MASK;
+}
+#define DDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__MASK			0x00000200
+#define DDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__SHIFT			9
+static inline uint32_t DDMA_CFG_DMA_ARB_WR_ARBIT_MODEL(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__SHIFT) & DDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__MASK;
+}
+#define DDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__MASK			0x00000100
+#define DDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__SHIFT			8
+static inline uint32_t DDMA_CFG_DMA_ARB_RD_ARBIT_MODEL(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__SHIFT) & DDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__MASK;
+}
+#define DDMA_CFG_DMA_ARB_RESERVED_1__MASK			0x00000080
+#define DDMA_CFG_DMA_ARB_RESERVED_1__SHIFT			7
+static inline uint32_t DDMA_CFG_DMA_ARB_RESERVED_1(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_RESERVED_1__SHIFT) & DDMA_CFG_DMA_ARB_RESERVED_1__MASK;
+}
+#define DDMA_CFG_DMA_ARB_WR_FIX_ARB__MASK			0x00000070
+#define DDMA_CFG_DMA_ARB_WR_FIX_ARB__SHIFT			4
+static inline uint32_t DDMA_CFG_DMA_ARB_WR_FIX_ARB(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_WR_FIX_ARB__SHIFT) & DDMA_CFG_DMA_ARB_WR_FIX_ARB__MASK;
+}
+#define DDMA_CFG_DMA_ARB_RESERVED_2__MASK			0x00000008
+#define DDMA_CFG_DMA_ARB_RESERVED_2__SHIFT			3
+static inline uint32_t DDMA_CFG_DMA_ARB_RESERVED_2(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_RESERVED_2__SHIFT) & DDMA_CFG_DMA_ARB_RESERVED_2__MASK;
+}
+#define DDMA_CFG_DMA_ARB_RD_FIX_ARB__MASK			0x00000007
+#define DDMA_CFG_DMA_ARB_RD_FIX_ARB__SHIFT			0
+static inline uint32_t DDMA_CFG_DMA_ARB_RD_FIX_ARB(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_ARB_RD_FIX_ARB__SHIFT) & DDMA_CFG_DMA_ARB_RD_FIX_ARB__MASK;
+}
+
+#define REG_DDMA_CFG_DMA_RD_QOS					0x00008020
+#define DDMA_CFG_DMA_RD_QOS_RESERVED_0__MASK			0xfffffc00
+#define DDMA_CFG_DMA_RD_QOS_RESERVED_0__SHIFT			10
+static inline uint32_t DDMA_CFG_DMA_RD_QOS_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_QOS_RESERVED_0__SHIFT) & DDMA_CFG_DMA_RD_QOS_RESERVED_0__MASK;
+}
+#define DDMA_CFG_DMA_RD_QOS_RD_PC_QOS__MASK			0x00000300
+#define DDMA_CFG_DMA_RD_QOS_RD_PC_QOS__SHIFT			8
+static inline uint32_t DDMA_CFG_DMA_RD_QOS_RD_PC_QOS(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_QOS_RD_PC_QOS__SHIFT) & DDMA_CFG_DMA_RD_QOS_RD_PC_QOS__MASK;
+}
+#define DDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__MASK			0x000000c0
+#define DDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__SHIFT			6
+static inline uint32_t DDMA_CFG_DMA_RD_QOS_RD_PPU_QOS(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__SHIFT) & DDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__MASK;
+}
+#define DDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__MASK			0x00000030
+#define DDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__SHIFT			4
+static inline uint32_t DDMA_CFG_DMA_RD_QOS_RD_DPU_QOS(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__SHIFT) & DDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__MASK;
+}
+#define DDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__MASK			0x0000000c
+#define DDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__SHIFT		2
+static inline uint32_t DDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__SHIFT) & DDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__MASK;
+}
+#define DDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__MASK		0x00000003
+#define DDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__SHIFT		0
+static inline uint32_t DDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__SHIFT) & DDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__MASK;
+}
+
+#define REG_DDMA_CFG_DMA_RD_CFG					0x00008024
+#define DDMA_CFG_DMA_RD_CFG_RESERVED_0__MASK			0xffffe000
+#define DDMA_CFG_DMA_RD_CFG_RESERVED_0__SHIFT			13
+static inline uint32_t DDMA_CFG_DMA_RD_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_CFG_RESERVED_0__SHIFT) & DDMA_CFG_DMA_RD_CFG_RESERVED_0__MASK;
+}
+#define DDMA_CFG_DMA_RD_CFG_RD_ARLOCK__MASK			0x00001000
+#define DDMA_CFG_DMA_RD_CFG_RD_ARLOCK__SHIFT			12
+static inline uint32_t DDMA_CFG_DMA_RD_CFG_RD_ARLOCK(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_CFG_RD_ARLOCK__SHIFT) & DDMA_CFG_DMA_RD_CFG_RD_ARLOCK__MASK;
+}
+#define DDMA_CFG_DMA_RD_CFG_RD_ARCACHE__MASK			0x00000f00
+#define DDMA_CFG_DMA_RD_CFG_RD_ARCACHE__SHIFT			8
+static inline uint32_t DDMA_CFG_DMA_RD_CFG_RD_ARCACHE(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_CFG_RD_ARCACHE__SHIFT) & DDMA_CFG_DMA_RD_CFG_RD_ARCACHE__MASK;
+}
+#define DDMA_CFG_DMA_RD_CFG_RD_ARPROT__MASK			0x000000e0
+#define DDMA_CFG_DMA_RD_CFG_RD_ARPROT__SHIFT			5
+static inline uint32_t DDMA_CFG_DMA_RD_CFG_RD_ARPROT(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_CFG_RD_ARPROT__SHIFT) & DDMA_CFG_DMA_RD_CFG_RD_ARPROT__MASK;
+}
+#define DDMA_CFG_DMA_RD_CFG_RD_ARBURST__MASK			0x00000018
+#define DDMA_CFG_DMA_RD_CFG_RD_ARBURST__SHIFT			3
+static inline uint32_t DDMA_CFG_DMA_RD_CFG_RD_ARBURST(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_CFG_RD_ARBURST__SHIFT) & DDMA_CFG_DMA_RD_CFG_RD_ARBURST__MASK;
+}
+#define DDMA_CFG_DMA_RD_CFG_RD_ARSIZE__MASK			0x00000007
+#define DDMA_CFG_DMA_RD_CFG_RD_ARSIZE__SHIFT			0
+static inline uint32_t DDMA_CFG_DMA_RD_CFG_RD_ARSIZE(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_RD_CFG_RD_ARSIZE__SHIFT) & DDMA_CFG_DMA_RD_CFG_RD_ARSIZE__MASK;
+}
+
+#define REG_DDMA_CFG_DMA_WR_CFG					0x00008028
+#define DDMA_CFG_DMA_WR_CFG_RESERVED_0__MASK			0xffffe000
+#define DDMA_CFG_DMA_WR_CFG_RESERVED_0__SHIFT			13
+static inline uint32_t DDMA_CFG_DMA_WR_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WR_CFG_RESERVED_0__SHIFT) & DDMA_CFG_DMA_WR_CFG_RESERVED_0__MASK;
+}
+#define DDMA_CFG_DMA_WR_CFG_WR_AWLOCK__MASK			0x00001000
+#define DDMA_CFG_DMA_WR_CFG_WR_AWLOCK__SHIFT			12
+static inline uint32_t DDMA_CFG_DMA_WR_CFG_WR_AWLOCK(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WR_CFG_WR_AWLOCK__SHIFT) & DDMA_CFG_DMA_WR_CFG_WR_AWLOCK__MASK;
+}
+#define DDMA_CFG_DMA_WR_CFG_WR_AWCACHE__MASK			0x00000f00
+#define DDMA_CFG_DMA_WR_CFG_WR_AWCACHE__SHIFT			8
+static inline uint32_t DDMA_CFG_DMA_WR_CFG_WR_AWCACHE(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WR_CFG_WR_AWCACHE__SHIFT) & DDMA_CFG_DMA_WR_CFG_WR_AWCACHE__MASK;
+}
+#define DDMA_CFG_DMA_WR_CFG_WR_AWPROT__MASK			0x000000e0
+#define DDMA_CFG_DMA_WR_CFG_WR_AWPROT__SHIFT			5
+static inline uint32_t DDMA_CFG_DMA_WR_CFG_WR_AWPROT(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WR_CFG_WR_AWPROT__SHIFT) & DDMA_CFG_DMA_WR_CFG_WR_AWPROT__MASK;
+}
+#define DDMA_CFG_DMA_WR_CFG_WR_AWBURST__MASK			0x00000018
+#define DDMA_CFG_DMA_WR_CFG_WR_AWBURST__SHIFT			3
+static inline uint32_t DDMA_CFG_DMA_WR_CFG_WR_AWBURST(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WR_CFG_WR_AWBURST__SHIFT) & DDMA_CFG_DMA_WR_CFG_WR_AWBURST__MASK;
+}
+#define DDMA_CFG_DMA_WR_CFG_WR_AWSIZE__MASK			0x00000007
+#define DDMA_CFG_DMA_WR_CFG_WR_AWSIZE__SHIFT			0
+static inline uint32_t DDMA_CFG_DMA_WR_CFG_WR_AWSIZE(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WR_CFG_WR_AWSIZE__SHIFT) & DDMA_CFG_DMA_WR_CFG_WR_AWSIZE__MASK;
+}
+
+#define REG_DDMA_CFG_DMA_WSTRB					0x0000802c
+#define DDMA_CFG_DMA_WSTRB_WR_WSTRB__MASK			0xffffffff
+#define DDMA_CFG_DMA_WSTRB_WR_WSTRB__SHIFT			0
+static inline uint32_t DDMA_CFG_DMA_WSTRB_WR_WSTRB(uint32_t val)
+{
+	return ((val) << DDMA_CFG_DMA_WSTRB_WR_WSTRB__SHIFT) & DDMA_CFG_DMA_WSTRB_WR_WSTRB__MASK;
+}
+
+#define REG_DDMA_CFG_STATUS					0x00008030
+#define DDMA_CFG_STATUS_RESERVED_0__MASK			0xfffffe00
+#define DDMA_CFG_STATUS_RESERVED_0__SHIFT			9
+static inline uint32_t DDMA_CFG_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << DDMA_CFG_STATUS_RESERVED_0__SHIFT) & DDMA_CFG_STATUS_RESERVED_0__MASK;
+}
+#define DDMA_CFG_STATUS_IDEL__MASK				0x00000100
+#define DDMA_CFG_STATUS_IDEL__SHIFT				8
+static inline uint32_t DDMA_CFG_STATUS_IDEL(uint32_t val)
+{
+	return ((val) << DDMA_CFG_STATUS_IDEL__SHIFT) & DDMA_CFG_STATUS_IDEL__MASK;
+}
+#define DDMA_CFG_STATUS_RESERVED_1__MASK			0x000000ff
+#define DDMA_CFG_STATUS_RESERVED_1__SHIFT			0
+static inline uint32_t DDMA_CFG_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << DDMA_CFG_STATUS_RESERVED_1__SHIFT) & DDMA_CFG_STATUS_RESERVED_1__MASK;
+}
+
+#define REG_SDMA_CFG_OUTSTANDING				0x00009000
+#define SDMA_CFG_OUTSTANDING_RESERVED_0__MASK			0xffff0000
+#define SDMA_CFG_OUTSTANDING_RESERVED_0__SHIFT			16
+static inline uint32_t SDMA_CFG_OUTSTANDING_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_OUTSTANDING_RESERVED_0__SHIFT) & SDMA_CFG_OUTSTANDING_RESERVED_0__MASK;
+}
+#define SDMA_CFG_OUTSTANDING_WR_OS_CNT__MASK			0x0000ff00
+#define SDMA_CFG_OUTSTANDING_WR_OS_CNT__SHIFT			8
+static inline uint32_t SDMA_CFG_OUTSTANDING_WR_OS_CNT(uint32_t val)
+{
+	return ((val) << SDMA_CFG_OUTSTANDING_WR_OS_CNT__SHIFT) & SDMA_CFG_OUTSTANDING_WR_OS_CNT__MASK;
+}
+#define SDMA_CFG_OUTSTANDING_RD_OS_CNT__MASK			0x000000ff
+#define SDMA_CFG_OUTSTANDING_RD_OS_CNT__SHIFT			0
+static inline uint32_t SDMA_CFG_OUTSTANDING_RD_OS_CNT(uint32_t val)
+{
+	return ((val) << SDMA_CFG_OUTSTANDING_RD_OS_CNT__SHIFT) & SDMA_CFG_OUTSTANDING_RD_OS_CNT__MASK;
+}
+
+#define REG_SDMA_RD_WEIGHT_0					0x00009004
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__MASK			0xff000000
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__SHIFT			24
+static inline uint32_t SDMA_RD_WEIGHT_0_RD_WEIGHT_PDP(uint32_t val)
+{
+	return ((val) << SDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__SHIFT) & SDMA_RD_WEIGHT_0_RD_WEIGHT_PDP__MASK;
+}
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__MASK			0x00ff0000
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__SHIFT			16
+static inline uint32_t SDMA_RD_WEIGHT_0_RD_WEIGHT_DPU(uint32_t val)
+{
+	return ((val) << SDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__SHIFT) & SDMA_RD_WEIGHT_0_RD_WEIGHT_DPU__MASK;
+}
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__MASK			0x0000ff00
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__SHIFT		8
+static inline uint32_t SDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL(uint32_t val)
+{
+	return ((val) << SDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__SHIFT) & SDMA_RD_WEIGHT_0_RD_WEIGHT_KERNEL__MASK;
+}
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__MASK		0x000000ff
+#define SDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__SHIFT		0
+static inline uint32_t SDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE(uint32_t val)
+{
+	return ((val) << SDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__SHIFT) & SDMA_RD_WEIGHT_0_RD_WEIGHT_FEATURE__MASK;
+}
+
+#define REG_SDMA_WR_WEIGHT_0					0x00009008
+#define SDMA_WR_WEIGHT_0_RESERVED_0__MASK			0xffff0000
+#define SDMA_WR_WEIGHT_0_RESERVED_0__SHIFT			16
+static inline uint32_t SDMA_WR_WEIGHT_0_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_WR_WEIGHT_0_RESERVED_0__SHIFT) & SDMA_WR_WEIGHT_0_RESERVED_0__MASK;
+}
+#define SDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__MASK			0x0000ff00
+#define SDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__SHIFT			8
+static inline uint32_t SDMA_WR_WEIGHT_0_WR_WEIGHT_PDP(uint32_t val)
+{
+	return ((val) << SDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__SHIFT) & SDMA_WR_WEIGHT_0_WR_WEIGHT_PDP__MASK;
+}
+#define SDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__MASK			0x000000ff
+#define SDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__SHIFT			0
+static inline uint32_t SDMA_WR_WEIGHT_0_WR_WEIGHT_DPU(uint32_t val)
+{
+	return ((val) << SDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__SHIFT) & SDMA_WR_WEIGHT_0_WR_WEIGHT_DPU__MASK;
+}
+
+#define REG_SDMA_CFG_ID_ERROR					0x0000900c
+#define SDMA_CFG_ID_ERROR_RESERVED_0__MASK			0xfffffc00
+#define SDMA_CFG_ID_ERROR_RESERVED_0__SHIFT			10
+static inline uint32_t SDMA_CFG_ID_ERROR_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_ID_ERROR_RESERVED_0__SHIFT) & SDMA_CFG_ID_ERROR_RESERVED_0__MASK;
+}
+#define SDMA_CFG_ID_ERROR_WR_RESP_ID__MASK			0x000003c0
+#define SDMA_CFG_ID_ERROR_WR_RESP_ID__SHIFT			6
+static inline uint32_t SDMA_CFG_ID_ERROR_WR_RESP_ID(uint32_t val)
+{
+	return ((val) << SDMA_CFG_ID_ERROR_WR_RESP_ID__SHIFT) & SDMA_CFG_ID_ERROR_WR_RESP_ID__MASK;
+}
+#define SDMA_CFG_ID_ERROR_RESERVED_1__MASK			0x00000020
+#define SDMA_CFG_ID_ERROR_RESERVED_1__SHIFT			5
+static inline uint32_t SDMA_CFG_ID_ERROR_RESERVED_1(uint32_t val)
+{
+	return ((val) << SDMA_CFG_ID_ERROR_RESERVED_1__SHIFT) & SDMA_CFG_ID_ERROR_RESERVED_1__MASK;
+}
+#define SDMA_CFG_ID_ERROR_RD_RESP_ID__MASK			0x0000001f
+#define SDMA_CFG_ID_ERROR_RD_RESP_ID__SHIFT			0
+static inline uint32_t SDMA_CFG_ID_ERROR_RD_RESP_ID(uint32_t val)
+{
+	return ((val) << SDMA_CFG_ID_ERROR_RD_RESP_ID__SHIFT) & SDMA_CFG_ID_ERROR_RD_RESP_ID__MASK;
+}
+
+#define REG_SDMA_RD_WEIGHT_1					0x00009010
+#define SDMA_RD_WEIGHT_1_RESERVED_0__MASK			0xffffff00
+#define SDMA_RD_WEIGHT_1_RESERVED_0__SHIFT			8
+static inline uint32_t SDMA_RD_WEIGHT_1_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_RD_WEIGHT_1_RESERVED_0__SHIFT) & SDMA_RD_WEIGHT_1_RESERVED_0__MASK;
+}
+#define SDMA_RD_WEIGHT_1_RD_WEIGHT_PC__MASK			0x000000ff
+#define SDMA_RD_WEIGHT_1_RD_WEIGHT_PC__SHIFT			0
+static inline uint32_t SDMA_RD_WEIGHT_1_RD_WEIGHT_PC(uint32_t val)
+{
+	return ((val) << SDMA_RD_WEIGHT_1_RD_WEIGHT_PC__SHIFT) & SDMA_RD_WEIGHT_1_RD_WEIGHT_PC__MASK;
+}
+
+#define REG_SDMA_CFG_DMA_FIFO_CLR				0x00009014
+#define SDMA_CFG_DMA_FIFO_CLR_RESERVED_0__MASK			0xfffffffe
+#define SDMA_CFG_DMA_FIFO_CLR_RESERVED_0__SHIFT			1
+static inline uint32_t SDMA_CFG_DMA_FIFO_CLR_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_FIFO_CLR_RESERVED_0__SHIFT) & SDMA_CFG_DMA_FIFO_CLR_RESERVED_0__MASK;
+}
+#define SDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__MASK		0x00000001
+#define SDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__SHIFT		0
+static inline uint32_t SDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__SHIFT) & SDMA_CFG_DMA_FIFO_CLR_DMA_FIFO_CLR__MASK;
+}
+
+#define REG_SDMA_CFG_DMA_ARB					0x00009018
+#define SDMA_CFG_DMA_ARB_RESERVED_0__MASK			0xfffffc00
+#define SDMA_CFG_DMA_ARB_RESERVED_0__SHIFT			10
+static inline uint32_t SDMA_CFG_DMA_ARB_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_RESERVED_0__SHIFT) & SDMA_CFG_DMA_ARB_RESERVED_0__MASK;
+}
+#define SDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__MASK			0x00000200
+#define SDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__SHIFT			9
+static inline uint32_t SDMA_CFG_DMA_ARB_WR_ARBIT_MODEL(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__SHIFT) & SDMA_CFG_DMA_ARB_WR_ARBIT_MODEL__MASK;
+}
+#define SDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__MASK			0x00000100
+#define SDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__SHIFT			8
+static inline uint32_t SDMA_CFG_DMA_ARB_RD_ARBIT_MODEL(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__SHIFT) & SDMA_CFG_DMA_ARB_RD_ARBIT_MODEL__MASK;
+}
+#define SDMA_CFG_DMA_ARB_RESERVED_1__MASK			0x00000080
+#define SDMA_CFG_DMA_ARB_RESERVED_1__SHIFT			7
+static inline uint32_t SDMA_CFG_DMA_ARB_RESERVED_1(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_RESERVED_1__SHIFT) & SDMA_CFG_DMA_ARB_RESERVED_1__MASK;
+}
+#define SDMA_CFG_DMA_ARB_WR_FIX_ARB__MASK			0x00000070
+#define SDMA_CFG_DMA_ARB_WR_FIX_ARB__SHIFT			4
+static inline uint32_t SDMA_CFG_DMA_ARB_WR_FIX_ARB(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_WR_FIX_ARB__SHIFT) & SDMA_CFG_DMA_ARB_WR_FIX_ARB__MASK;
+}
+#define SDMA_CFG_DMA_ARB_RESERVED_2__MASK			0x00000008
+#define SDMA_CFG_DMA_ARB_RESERVED_2__SHIFT			3
+static inline uint32_t SDMA_CFG_DMA_ARB_RESERVED_2(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_RESERVED_2__SHIFT) & SDMA_CFG_DMA_ARB_RESERVED_2__MASK;
+}
+#define SDMA_CFG_DMA_ARB_RD_FIX_ARB__MASK			0x00000007
+#define SDMA_CFG_DMA_ARB_RD_FIX_ARB__SHIFT			0
+static inline uint32_t SDMA_CFG_DMA_ARB_RD_FIX_ARB(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_ARB_RD_FIX_ARB__SHIFT) & SDMA_CFG_DMA_ARB_RD_FIX_ARB__MASK;
+}
+
+#define REG_SDMA_CFG_DMA_RD_QOS					0x00009020
+#define SDMA_CFG_DMA_RD_QOS_RESERVED_0__MASK			0xfffffc00
+#define SDMA_CFG_DMA_RD_QOS_RESERVED_0__SHIFT			10
+static inline uint32_t SDMA_CFG_DMA_RD_QOS_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_QOS_RESERVED_0__SHIFT) & SDMA_CFG_DMA_RD_QOS_RESERVED_0__MASK;
+}
+#define SDMA_CFG_DMA_RD_QOS_RD_PC_QOS__MASK			0x00000300
+#define SDMA_CFG_DMA_RD_QOS_RD_PC_QOS__SHIFT			8
+static inline uint32_t SDMA_CFG_DMA_RD_QOS_RD_PC_QOS(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_QOS_RD_PC_QOS__SHIFT) & SDMA_CFG_DMA_RD_QOS_RD_PC_QOS__MASK;
+}
+#define SDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__MASK			0x000000c0
+#define SDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__SHIFT			6
+static inline uint32_t SDMA_CFG_DMA_RD_QOS_RD_PPU_QOS(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__SHIFT) & SDMA_CFG_DMA_RD_QOS_RD_PPU_QOS__MASK;
+}
+#define SDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__MASK			0x00000030
+#define SDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__SHIFT			4
+static inline uint32_t SDMA_CFG_DMA_RD_QOS_RD_DPU_QOS(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__SHIFT) & SDMA_CFG_DMA_RD_QOS_RD_DPU_QOS__MASK;
+}
+#define SDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__MASK			0x0000000c
+#define SDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__SHIFT		2
+static inline uint32_t SDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__SHIFT) & SDMA_CFG_DMA_RD_QOS_RD_KERNEL_QOS__MASK;
+}
+#define SDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__MASK		0x00000003
+#define SDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__SHIFT		0
+static inline uint32_t SDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__SHIFT) & SDMA_CFG_DMA_RD_QOS_RD_FEATURE_QOS__MASK;
+}
+
+#define REG_SDMA_CFG_DMA_RD_CFG					0x00009024
+#define SDMA_CFG_DMA_RD_CFG_RESERVED_0__MASK			0xffffe000
+#define SDMA_CFG_DMA_RD_CFG_RESERVED_0__SHIFT			13
+static inline uint32_t SDMA_CFG_DMA_RD_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_CFG_RESERVED_0__SHIFT) & SDMA_CFG_DMA_RD_CFG_RESERVED_0__MASK;
+}
+#define SDMA_CFG_DMA_RD_CFG_RD_ARLOCK__MASK			0x00001000
+#define SDMA_CFG_DMA_RD_CFG_RD_ARLOCK__SHIFT			12
+static inline uint32_t SDMA_CFG_DMA_RD_CFG_RD_ARLOCK(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_CFG_RD_ARLOCK__SHIFT) & SDMA_CFG_DMA_RD_CFG_RD_ARLOCK__MASK;
+}
+#define SDMA_CFG_DMA_RD_CFG_RD_ARCACHE__MASK			0x00000f00
+#define SDMA_CFG_DMA_RD_CFG_RD_ARCACHE__SHIFT			8
+static inline uint32_t SDMA_CFG_DMA_RD_CFG_RD_ARCACHE(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_CFG_RD_ARCACHE__SHIFT) & SDMA_CFG_DMA_RD_CFG_RD_ARCACHE__MASK;
+}
+#define SDMA_CFG_DMA_RD_CFG_RD_ARPROT__MASK			0x000000e0
+#define SDMA_CFG_DMA_RD_CFG_RD_ARPROT__SHIFT			5
+static inline uint32_t SDMA_CFG_DMA_RD_CFG_RD_ARPROT(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_CFG_RD_ARPROT__SHIFT) & SDMA_CFG_DMA_RD_CFG_RD_ARPROT__MASK;
+}
+#define SDMA_CFG_DMA_RD_CFG_RD_ARBURST__MASK			0x00000018
+#define SDMA_CFG_DMA_RD_CFG_RD_ARBURST__SHIFT			3
+static inline uint32_t SDMA_CFG_DMA_RD_CFG_RD_ARBURST(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_CFG_RD_ARBURST__SHIFT) & SDMA_CFG_DMA_RD_CFG_RD_ARBURST__MASK;
+}
+#define SDMA_CFG_DMA_RD_CFG_RD_ARSIZE__MASK			0x00000007
+#define SDMA_CFG_DMA_RD_CFG_RD_ARSIZE__SHIFT			0
+static inline uint32_t SDMA_CFG_DMA_RD_CFG_RD_ARSIZE(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_RD_CFG_RD_ARSIZE__SHIFT) & SDMA_CFG_DMA_RD_CFG_RD_ARSIZE__MASK;
+}
+
+#define REG_SDMA_CFG_DMA_WR_CFG					0x00009028
+#define SDMA_CFG_DMA_WR_CFG_RESERVED_0__MASK			0xffffe000
+#define SDMA_CFG_DMA_WR_CFG_RESERVED_0__SHIFT			13
+static inline uint32_t SDMA_CFG_DMA_WR_CFG_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WR_CFG_RESERVED_0__SHIFT) & SDMA_CFG_DMA_WR_CFG_RESERVED_0__MASK;
+}
+#define SDMA_CFG_DMA_WR_CFG_WR_AWLOCK__MASK			0x00001000
+#define SDMA_CFG_DMA_WR_CFG_WR_AWLOCK__SHIFT			12
+static inline uint32_t SDMA_CFG_DMA_WR_CFG_WR_AWLOCK(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WR_CFG_WR_AWLOCK__SHIFT) & SDMA_CFG_DMA_WR_CFG_WR_AWLOCK__MASK;
+}
+#define SDMA_CFG_DMA_WR_CFG_WR_AWCACHE__MASK			0x00000f00
+#define SDMA_CFG_DMA_WR_CFG_WR_AWCACHE__SHIFT			8
+static inline uint32_t SDMA_CFG_DMA_WR_CFG_WR_AWCACHE(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WR_CFG_WR_AWCACHE__SHIFT) & SDMA_CFG_DMA_WR_CFG_WR_AWCACHE__MASK;
+}
+#define SDMA_CFG_DMA_WR_CFG_WR_AWPROT__MASK			0x000000e0
+#define SDMA_CFG_DMA_WR_CFG_WR_AWPROT__SHIFT			5
+static inline uint32_t SDMA_CFG_DMA_WR_CFG_WR_AWPROT(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WR_CFG_WR_AWPROT__SHIFT) & SDMA_CFG_DMA_WR_CFG_WR_AWPROT__MASK;
+}
+#define SDMA_CFG_DMA_WR_CFG_WR_AWBURST__MASK			0x00000018
+#define SDMA_CFG_DMA_WR_CFG_WR_AWBURST__SHIFT			3
+static inline uint32_t SDMA_CFG_DMA_WR_CFG_WR_AWBURST(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WR_CFG_WR_AWBURST__SHIFT) & SDMA_CFG_DMA_WR_CFG_WR_AWBURST__MASK;
+}
+#define SDMA_CFG_DMA_WR_CFG_WR_AWSIZE__MASK			0x00000007
+#define SDMA_CFG_DMA_WR_CFG_WR_AWSIZE__SHIFT			0
+static inline uint32_t SDMA_CFG_DMA_WR_CFG_WR_AWSIZE(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WR_CFG_WR_AWSIZE__SHIFT) & SDMA_CFG_DMA_WR_CFG_WR_AWSIZE__MASK;
+}
+
+#define REG_SDMA_CFG_DMA_WSTRB					0x0000902c
+#define SDMA_CFG_DMA_WSTRB_WR_WSTRB__MASK			0xffffffff
+#define SDMA_CFG_DMA_WSTRB_WR_WSTRB__SHIFT			0
+static inline uint32_t SDMA_CFG_DMA_WSTRB_WR_WSTRB(uint32_t val)
+{
+	return ((val) << SDMA_CFG_DMA_WSTRB_WR_WSTRB__SHIFT) & SDMA_CFG_DMA_WSTRB_WR_WSTRB__MASK;
+}
+
+#define REG_SDMA_CFG_STATUS					0x00009030
+#define SDMA_CFG_STATUS_RESERVED_0__MASK			0xfffffe00
+#define SDMA_CFG_STATUS_RESERVED_0__SHIFT			9
+static inline uint32_t SDMA_CFG_STATUS_RESERVED_0(uint32_t val)
+{
+	return ((val) << SDMA_CFG_STATUS_RESERVED_0__SHIFT) & SDMA_CFG_STATUS_RESERVED_0__MASK;
+}
+#define SDMA_CFG_STATUS_IDEL__MASK				0x00000100
+#define SDMA_CFG_STATUS_IDEL__SHIFT				8
+static inline uint32_t SDMA_CFG_STATUS_IDEL(uint32_t val)
+{
+	return ((val) << SDMA_CFG_STATUS_IDEL__SHIFT) & SDMA_CFG_STATUS_IDEL__MASK;
+}
+#define SDMA_CFG_STATUS_RESERVED_1__MASK			0x000000ff
+#define SDMA_CFG_STATUS_RESERVED_1__SHIFT			0
+static inline uint32_t SDMA_CFG_STATUS_RESERVED_1(uint32_t val)
+{
+	return ((val) << SDMA_CFG_STATUS_RESERVED_1__SHIFT) & SDMA_CFG_STATUS_RESERVED_1__MASK;
+}
+
+#define REG_GLOBAL_OPERATION_ENABLE				0x0000f008
+#define GLOBAL_OPERATION_ENABLE_RESERVED_0__MASK		0xffffff80
+#define GLOBAL_OPERATION_ENABLE_RESERVED_0__SHIFT		7
+static inline uint32_t GLOBAL_OPERATION_ENABLE_RESERVED_0(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_RESERVED_0__SHIFT) & GLOBAL_OPERATION_ENABLE_RESERVED_0__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_PPU_RDMA_OP_EN__MASK		0x00000040
+#define GLOBAL_OPERATION_ENABLE_PPU_RDMA_OP_EN__SHIFT		6
+static inline uint32_t GLOBAL_OPERATION_ENABLE_PPU_RDMA_OP_EN(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_PPU_RDMA_OP_EN__SHIFT) & GLOBAL_OPERATION_ENABLE_PPU_RDMA_OP_EN__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_PPU_OP_EN__MASK			0x00000020
+#define GLOBAL_OPERATION_ENABLE_PPU_OP_EN__SHIFT		5
+static inline uint32_t GLOBAL_OPERATION_ENABLE_PPU_OP_EN(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_PPU_OP_EN__SHIFT) & GLOBAL_OPERATION_ENABLE_PPU_OP_EN__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_DPU_RDMA_OP_EN__MASK		0x00000010
+#define GLOBAL_OPERATION_ENABLE_DPU_RDMA_OP_EN__SHIFT		4
+static inline uint32_t GLOBAL_OPERATION_ENABLE_DPU_RDMA_OP_EN(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_DPU_RDMA_OP_EN__SHIFT) & GLOBAL_OPERATION_ENABLE_DPU_RDMA_OP_EN__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_DPU_OP_EN__MASK			0x00000008
+#define GLOBAL_OPERATION_ENABLE_DPU_OP_EN__SHIFT		3
+static inline uint32_t GLOBAL_OPERATION_ENABLE_DPU_OP_EN(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_DPU_OP_EN__SHIFT) & GLOBAL_OPERATION_ENABLE_DPU_OP_EN__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_CORE_OP_EN__MASK		0x00000004
+#define GLOBAL_OPERATION_ENABLE_CORE_OP_EN__SHIFT		2
+static inline uint32_t GLOBAL_OPERATION_ENABLE_CORE_OP_EN(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_CORE_OP_EN__SHIFT) & GLOBAL_OPERATION_ENABLE_CORE_OP_EN__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_RESERVED_1__MASK		0x00000002
+#define GLOBAL_OPERATION_ENABLE_RESERVED_1__SHIFT		1
+static inline uint32_t GLOBAL_OPERATION_ENABLE_RESERVED_1(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_RESERVED_1__SHIFT) & GLOBAL_OPERATION_ENABLE_RESERVED_1__MASK;
+}
+#define GLOBAL_OPERATION_ENABLE_CNA_OP_EN__MASK			0x00000001
+#define GLOBAL_OPERATION_ENABLE_CNA_OP_EN__SHIFT		0
+static inline uint32_t GLOBAL_OPERATION_ENABLE_CNA_OP_EN(uint32_t val)
+{
+	return ((val) << GLOBAL_OPERATION_ENABLE_CNA_OP_EN__SHIFT) & GLOBAL_OPERATION_ENABLE_CNA_OP_EN__MASK;
+}
+
+#endif /* __ROCKET_REGISTERS_XML__ */

From ed98261b41687323ffa02ca20fef1e60b38fd1aa Mon Sep 17 00:00:00 2001
From: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Date: Mon, 21 Jul 2025 11:17:29 +0200
Subject: [PATCH 0106/1455] accel/rocket: Add a new driver for Rockchip's NPU

This initial version supports the NPU as shipped in the RK3588 SoC and
described in the first part of its TRM, in Chapter 36.

This NPU contains 3 independent cores that the driver can submit jobs
to.

This commit adds just hardware initialization and power management.

v2:
- Split cores and IOMMUs as independent devices (Sebastian Reichel)
- Add some documentation (Jeffrey Hugo)
- Be more explicit in the Kconfig documentation (Jeffrey Hugo)
- Remove resets, as these haven't been found useful so far (Zenghui Yu)
- Repack structs (Jeffrey Hugo)
- Use DEFINE_DRM_ACCEL_FOPS (Jeffrey Hugo)
- Use devm_drm_dev_alloc (Jeffrey Hugo)
- Use probe log helper (Jeffrey Hugo)
- Introduce UABI header in a later patch (Jeffrey Hugo)

v3:
- Adapt to a split of the register block in the DT bindings (Nicolas
  Frattaroli)
- Move registers header to its own commit (Thomas Zimmermann)
- Misc. cleanups (Thomas Zimmermann and Jeff Hugo)
- Make use of GPL-2.0-only for the copyright notice (Jeff Hugo)
- PM improvements (Nicolas Frattaroli)

v4:
- Use bulk clk API (Krzysztof Kozlowski)

v6:
- Remove mention to NVDLA, as the hardware is only incidentally related
  (Kever Yang)
- Use calloc instead of GFP_ZERO (Jeff Hugo)
- Explicitly include linux/container_of.h (Jeff Hugo)
- pclk and npu clocks are now needed by all cores (Rob Herring)

v7:
- Assign its own IOMMU domain to each client, for isolation (Daniel
  Stone and Robin Murphy)

v8:
- Kconfig: fix depends to be more explicit about Rockchip, and remove
  superfluous selects (Robin Murphy)
- Use reset lines to reset the cores (Robin Murphy)
- Reference count the module
- Set dma_set_max_seg_size
- Correctly acquire a reference to the IOMMU (Robin Murphy)
- Remove notion of top core (Robin Murphy)

Reviewed-by: Robert Foss <rfoss@kernel.org>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721-6-10-rocket-v9-2-77ebd484941e@tomeuvizoso.net
---
 Documentation/accel/index.rst        |   1 +
 Documentation/accel/rocket/index.rst |  19 ++
 MAINTAINERS                          |  10 +
 drivers/accel/Kconfig                |   1 +
 drivers/accel/Makefile               |   1 +
 drivers/accel/rocket/Kconfig         |  24 +++
 drivers/accel/rocket/Makefile        |   8 +
 drivers/accel/rocket/rocket_core.c   | 100 ++++++++++
 drivers/accel/rocket/rocket_core.h   |  49 +++++
 drivers/accel/rocket/rocket_device.c |  56 ++++++
 drivers/accel/rocket/rocket_device.h |  28 +++
 drivers/accel/rocket/rocket_drv.c    | 261 +++++++++++++++++++++++++++
 drivers/accel/rocket/rocket_drv.h    |  23 +++
 13 files changed, 581 insertions(+)
 create mode 100644 Documentation/accel/rocket/index.rst
 create mode 100644 drivers/accel/rocket/Kconfig
 create mode 100644 drivers/accel/rocket/Makefile
 create mode 100644 drivers/accel/rocket/rocket_core.c
 create mode 100644 drivers/accel/rocket/rocket_core.h
 create mode 100644 drivers/accel/rocket/rocket_device.c
 create mode 100644 drivers/accel/rocket/rocket_device.h
 create mode 100644 drivers/accel/rocket/rocket_drv.c
 create mode 100644 drivers/accel/rocket/rocket_drv.h

diff --git a/Documentation/accel/index.rst b/Documentation/accel/index.rst
index bc85f26533d8..d8fa332d60a8 100644
--- a/Documentation/accel/index.rst
+++ b/Documentation/accel/index.rst
@@ -10,6 +10,7 @@ Compute Accelerators
    introduction
    amdxdna/index
    qaic/index
+   rocket/index
 
 .. only::  subproject and html
 
diff --git a/Documentation/accel/rocket/index.rst b/Documentation/accel/rocket/index.rst
new file mode 100644
index 000000000000..70f97bccf100
--- /dev/null
+++ b/Documentation/accel/rocket/index.rst
@@ -0,0 +1,19 @@
+.. SPDX-License-Identifier: GPL-2.0-only
+
+=====================================
+ accel/rocket Rockchip NPU driver
+=====================================
+
+The accel/rocket driver supports the Neural Processing Units (NPUs) inside some
+Rockchip SoCs such as the RK3588. Rockchip calls it RKNN and sometimes RKNPU.
+
+The hardware is described in chapter 36 in the RK3588 TRM.
+
+This driver just powers the hardware on and off, allocates and maps buffers to
+the device and submits jobs to the frontend unit. Everything else is done in
+userspace, as a Gallium driver (also called rocket) that is part of the Mesa3D
+project.
+
+Hardware currently supported:
+
+* RK3588
diff --git a/MAINTAINERS b/MAINTAINERS
index 62fa528c0884..b5746d13e30b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7393,6 +7393,16 @@ T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	drivers/accel/ivpu/
 F:	include/uapi/drm/ivpu_accel.h
 
+DRM ACCEL DRIVER FOR ROCKCHIP NPU
+M:	Tomeu Vizoso <tomeu@tomeuvizoso.net>
+L:	dri-devel@lists.freedesktop.org
+S:	Supported
+T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
+F:	Documentation/accel/rocket/
+F:	Documentation/devicetree/bindings/npu/rockchip,rknn-core.yaml
+F:	drivers/accel/rocket/
+F:	include/uapi/drm/rocket_accel.h
+
 DRM COMPUTE ACCELERATORS DRIVERS AND FRAMEWORK
 M:	Oded Gabbay <ogabbay@kernel.org>
 L:	dri-devel@lists.freedesktop.org
diff --git a/drivers/accel/Kconfig b/drivers/accel/Kconfig
index 5b9490367a39..bb01cebc42bf 100644
--- a/drivers/accel/Kconfig
+++ b/drivers/accel/Kconfig
@@ -28,5 +28,6 @@ source "drivers/accel/amdxdna/Kconfig"
 source "drivers/accel/habanalabs/Kconfig"
 source "drivers/accel/ivpu/Kconfig"
 source "drivers/accel/qaic/Kconfig"
+source "drivers/accel/rocket/Kconfig"
 
 endif
diff --git a/drivers/accel/Makefile b/drivers/accel/Makefile
index a301fb6089d4..ffc3fa588666 100644
--- a/drivers/accel/Makefile
+++ b/drivers/accel/Makefile
@@ -4,3 +4,4 @@ obj-$(CONFIG_DRM_ACCEL_AMDXDNA)		+= amdxdna/
 obj-$(CONFIG_DRM_ACCEL_HABANALABS)	+= habanalabs/
 obj-$(CONFIG_DRM_ACCEL_IVPU)		+= ivpu/
 obj-$(CONFIG_DRM_ACCEL_QAIC)		+= qaic/
+obj-$(CONFIG_DRM_ACCEL_ROCKET)		+= rocket/
\ No newline at end of file
diff --git a/drivers/accel/rocket/Kconfig b/drivers/accel/rocket/Kconfig
new file mode 100644
index 000000000000..43d6cd98ec8e
--- /dev/null
+++ b/drivers/accel/rocket/Kconfig
@@ -0,0 +1,24 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+config DRM_ACCEL_ROCKET
+       tristate "Rocket (support for Rockchip NPUs)"
+       depends on DRM
+       depends on (ARCH_ROCKCHIP && ARM64) || COMPILE_TEST
+       depends on ROCKCHIP_IOMMU || COMPILE_TEST
+       depends on MMU
+       select DRM_SCHED
+       select DRM_GEM_SHMEM_HELPER
+       help
+	  Choose this option if you have a Rockchip SoC that contains a
+	  compatible Neural Processing Unit (NPU), such as the RK3588. Called by
+	  Rockchip either RKNN or RKNPU, it accelerates inference of neural
+	  networks.
+
+	  The interface exposed to userspace is described in
+	  include/uapi/drm/rocket_accel.h and is used by the Rocket userspace
+	  driver in Mesa3D.
+
+	  If unsure, say N.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called rocket.
diff --git a/drivers/accel/rocket/Makefile b/drivers/accel/rocket/Makefile
new file mode 100644
index 000000000000..abdd75f2492e
--- /dev/null
+++ b/drivers/accel/rocket/Makefile
@@ -0,0 +1,8 @@
+# SPDX-License-Identifier: GPL-2.0-only
+
+obj-$(CONFIG_DRM_ACCEL_ROCKET) := rocket.o
+
+rocket-y := \
+	rocket_core.o \
+	rocket_device.o \
+	rocket_drv.o
diff --git a/drivers/accel/rocket/rocket_core.c b/drivers/accel/rocket/rocket_core.c
new file mode 100644
index 000000000000..9be964b5fbae
--- /dev/null
+++ b/drivers/accel/rocket/rocket_core.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/dev_printk.h>
+#include <linux/dma-mapping.h>
+#include <linux/err.h>
+#include <linux/iommu.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/reset.h>
+
+#include "rocket_core.h"
+
+int rocket_core_init(struct rocket_core *core)
+{
+	struct device *dev = core->dev;
+	struct platform_device *pdev = to_platform_device(dev);
+	u32 version;
+	int err = 0;
+
+	core->resets[0].id = "srst_a";
+	core->resets[1].id = "srst_h";
+	err = devm_reset_control_bulk_get_exclusive(&pdev->dev, ARRAY_SIZE(core->resets),
+						    core->resets);
+	if (err)
+		return dev_err_probe(dev, err, "failed to get resets for core %d\n", core->index);
+
+	err = devm_clk_bulk_get(dev, ARRAY_SIZE(core->clks), core->clks);
+	if (err)
+		return dev_err_probe(dev, err, "failed to get clocks for core %d\n", core->index);
+
+	core->pc_iomem = devm_platform_ioremap_resource_byname(pdev, "pc");
+	if (IS_ERR(core->pc_iomem)) {
+		dev_err(dev, "couldn't find PC registers %ld\n", PTR_ERR(core->pc_iomem));
+		return PTR_ERR(core->pc_iomem);
+	}
+
+	core->cna_iomem = devm_platform_ioremap_resource_byname(pdev, "cna");
+	if (IS_ERR(core->cna_iomem)) {
+		dev_err(dev, "couldn't find CNA registers %ld\n", PTR_ERR(core->cna_iomem));
+		return PTR_ERR(core->cna_iomem);
+	}
+
+	core->core_iomem = devm_platform_ioremap_resource_byname(pdev, "core");
+	if (IS_ERR(core->core_iomem)) {
+		dev_err(dev, "couldn't find CORE registers %ld\n", PTR_ERR(core->core_iomem));
+		return PTR_ERR(core->core_iomem);
+	}
+
+	dma_set_max_seg_size(dev, UINT_MAX);
+
+	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(40));
+	if (err)
+		return err;
+
+	core->iommu_group = iommu_group_get(dev);
+
+	pm_runtime_use_autosuspend(dev);
+
+	/*
+	 * As this NPU will be most often used as part of a media pipeline that
+	 * ends presenting in a display, choose 50 ms (~3 frames at 60Hz) as an
+	 * autosuspend delay as that will keep the device powered up while the
+	 * pipeline is running.
+	 */
+	pm_runtime_set_autosuspend_delay(dev, 50);
+
+	pm_runtime_enable(dev);
+
+	err = pm_runtime_get_sync(dev);
+
+	version = rocket_pc_readl(core, VERSION);
+	version += rocket_pc_readl(core, VERSION_NUM) & 0xffff;
+
+	pm_runtime_mark_last_busy(dev);
+	pm_runtime_put_autosuspend(dev);
+
+	dev_info(dev, "Rockchip NPU core %d version: %d\n", core->index, version);
+
+	return 0;
+}
+
+void rocket_core_fini(struct rocket_core *core)
+{
+	pm_runtime_dont_use_autosuspend(core->dev);
+	pm_runtime_disable(core->dev);
+	iommu_group_put(core->iommu_group);
+	core->iommu_group = NULL;
+}
+
+void rocket_core_reset(struct rocket_core *core)
+{
+	reset_control_bulk_assert(ARRAY_SIZE(core->resets), core->resets);
+
+	udelay(10);
+
+	reset_control_bulk_deassert(ARRAY_SIZE(core->resets), core->resets);
+}
diff --git a/drivers/accel/rocket/rocket_core.h b/drivers/accel/rocket/rocket_core.h
new file mode 100644
index 000000000000..660de2d70f7d
--- /dev/null
+++ b/drivers/accel/rocket/rocket_core.h
@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#ifndef __ROCKET_CORE_H__
+#define __ROCKET_CORE_H__
+
+#include <drm/gpu_scheduler.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+#include <linux/mutex_types.h>
+#include <linux/reset.h>
+
+#include "rocket_registers.h"
+
+#define rocket_pc_readl(core, reg) \
+	readl((core)->pc_iomem + (REG_PC_##reg))
+#define rocket_pc_writel(core, reg, value) \
+	writel(value, (core)->pc_iomem + (REG_PC_##reg))
+
+#define rocket_cna_readl(core, reg) \
+	readl((core)->cna_iomem + (REG_CNA_##reg) - REG_CNA_S_STATUS)
+#define rocket_cna_writel(core, reg, value) \
+	writel(value, (core)->cna_iomem + (REG_CNA_##reg) - REG_CNA_S_STATUS)
+
+#define rocket_core_readl(core, reg) \
+	readl((core)->core_iomem + (REG_CORE_##reg) - REG_CORE_S_STATUS)
+#define rocket_core_writel(core, reg, value) \
+	writel(value, (core)->core_iomem + (REG_CORE_##reg) - REG_CORE_S_STATUS)
+
+struct rocket_core {
+	struct device *dev;
+	struct rocket_device *rdev;
+	unsigned int index;
+
+	int irq;
+	void __iomem *pc_iomem;
+	void __iomem *cna_iomem;
+	void __iomem *core_iomem;
+	struct clk_bulk_data clks[4];
+	struct reset_control_bulk_data resets[2];
+
+	struct iommu_group *iommu_group;
+};
+
+int rocket_core_init(struct rocket_core *core);
+void rocket_core_fini(struct rocket_core *core);
+void rocket_core_reset(struct rocket_core *core);
+
+#endif
diff --git a/drivers/accel/rocket/rocket_device.c b/drivers/accel/rocket/rocket_device.c
new file mode 100644
index 000000000000..b05a0df91d48
--- /dev/null
+++ b/drivers/accel/rocket/rocket_device.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#include <drm/drm_drv.h>
+#include <linux/array_size.h>
+#include <linux/clk.h>
+#include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+
+#include "rocket_device.h"
+
+struct rocket_device *rocket_device_init(struct platform_device *pdev,
+					 const struct drm_driver *rocket_drm_driver)
+{
+	struct device *dev = &pdev->dev;
+	struct device_node *core_node;
+	struct rocket_device *rdev;
+	struct drm_device *ddev;
+	unsigned int num_cores = 0;
+	int err;
+
+	rdev = devm_drm_dev_alloc(dev, rocket_drm_driver, struct rocket_device, ddev);
+	if (IS_ERR(rdev))
+		return rdev;
+
+	ddev = &rdev->ddev;
+	dev_set_drvdata(dev, rdev);
+
+	for_each_compatible_node(core_node, NULL, "rockchip,rk3588-rknn-core")
+		if (of_device_is_available(core_node))
+			num_cores++;
+
+	rdev->cores = devm_kcalloc(dev, num_cores, sizeof(*rdev->cores), GFP_KERNEL);
+	if (!rdev->cores)
+		return ERR_PTR(-ENOMEM);
+
+	dma_set_max_seg_size(dev, UINT_MAX);
+
+	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(40));
+	if (err)
+		return ERR_PTR(err);
+
+	err = drm_dev_register(ddev, 0);
+	if (err)
+		return ERR_PTR(err);
+
+	return rdev;
+}
+
+void rocket_device_fini(struct rocket_device *rdev)
+{
+	WARN_ON(rdev->num_cores > 0);
+
+	drm_dev_unregister(&rdev->ddev);
+}
diff --git a/drivers/accel/rocket/rocket_device.h b/drivers/accel/rocket/rocket_device.h
new file mode 100644
index 000000000000..a5a5857bb199
--- /dev/null
+++ b/drivers/accel/rocket/rocket_device.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#ifndef __ROCKET_DEVICE_H__
+#define __ROCKET_DEVICE_H__
+
+#include <drm/drm_device.h>
+#include <linux/clk.h>
+#include <linux/container_of.h>
+#include <linux/iommu.h>
+#include <linux/platform_device.h>
+
+#include "rocket_core.h"
+
+struct rocket_device {
+	struct drm_device ddev;
+
+	struct rocket_core *cores;
+	unsigned int num_cores;
+};
+
+struct rocket_device *rocket_device_init(struct platform_device *pdev,
+					 const struct drm_driver *rocket_drm_driver);
+void rocket_device_fini(struct rocket_device *rdev);
+#define to_rocket_device(drm_dev) \
+	((struct rocket_device *)(container_of((drm_dev), struct rocket_device, ddev)))
+
+#endif /* __ROCKET_DEVICE_H__ */
diff --git a/drivers/accel/rocket/rocket_drv.c b/drivers/accel/rocket/rocket_drv.c
new file mode 100644
index 000000000000..a5df94f6b125
--- /dev/null
+++ b/drivers/accel/rocket/rocket_drv.c
@@ -0,0 +1,261 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#include <drm/drm_accel.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_gem.h>
+#include <drm/drm_ioctl.h>
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/iommu.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+
+#include "rocket_drv.h"
+
+/*
+ * Facade device, used to expose a single DRM device to userspace, that
+ * schedules jobs to any RKNN cores in the system.
+ */
+static struct platform_device *drm_dev;
+static struct rocket_device *rdev;
+
+static void
+rocket_iommu_domain_destroy(struct kref *kref)
+{
+	struct rocket_iommu_domain *domain = container_of(kref, struct rocket_iommu_domain, kref);
+
+	iommu_domain_free(domain->domain);
+	domain->domain = NULL;
+	kfree(domain);
+}
+
+static struct rocket_iommu_domain*
+rocket_iommu_domain_create(struct device *dev)
+{
+	struct rocket_iommu_domain *domain = kmalloc(sizeof(*domain), GFP_KERNEL);
+	void *err;
+
+	if (!domain)
+		return ERR_PTR(-ENOMEM);
+
+	domain->domain = iommu_paging_domain_alloc(dev);
+	if (IS_ERR(domain->domain)) {
+		err = ERR_CAST(domain->domain);
+		kfree(domain);
+		return err;
+	}
+	kref_init(&domain->kref);
+
+	return domain;
+}
+
+struct rocket_iommu_domain *
+rocket_iommu_domain_get(struct rocket_file_priv *rocket_priv)
+{
+	kref_get(&rocket_priv->domain->kref);
+	return rocket_priv->domain;
+}
+
+void
+rocket_iommu_domain_put(struct rocket_iommu_domain *domain)
+{
+	kref_put(&domain->kref, rocket_iommu_domain_destroy);
+}
+
+static int
+rocket_open(struct drm_device *dev, struct drm_file *file)
+{
+	struct rocket_device *rdev = to_rocket_device(dev);
+	struct rocket_file_priv *rocket_priv;
+	int ret;
+
+	if (!try_module_get(THIS_MODULE))
+		return -EINVAL;
+
+	rocket_priv = kzalloc(sizeof(*rocket_priv), GFP_KERNEL);
+	if (!rocket_priv) {
+		ret = -ENOMEM;
+		goto err_put_mod;
+	}
+
+	rocket_priv->rdev = rdev;
+	rocket_priv->domain = rocket_iommu_domain_create(rdev->cores[0].dev);
+	if (IS_ERR(rocket_priv->domain)) {
+		ret = PTR_ERR(rocket_priv->domain);
+		goto err_free;
+	}
+
+	file->driver_priv = rocket_priv;
+
+	return 0;
+
+err_free:
+	kfree(rocket_priv);
+err_put_mod:
+	module_put(THIS_MODULE);
+	return ret;
+}
+
+static void
+rocket_postclose(struct drm_device *dev, struct drm_file *file)
+{
+	struct rocket_file_priv *rocket_priv = file->driver_priv;
+
+	rocket_iommu_domain_put(rocket_priv->domain);
+	kfree(rocket_priv);
+	module_put(THIS_MODULE);
+}
+
+static const struct drm_ioctl_desc rocket_drm_driver_ioctls[] = {
+#define ROCKET_IOCTL(n, func) \
+	DRM_IOCTL_DEF_DRV(ROCKET_##n, rocket_ioctl_##func, 0)
+};
+
+DEFINE_DRM_ACCEL_FOPS(rocket_accel_driver_fops);
+
+/*
+ * Rocket driver version:
+ * - 1.0 - initial interface
+ */
+static const struct drm_driver rocket_drm_driver = {
+	.driver_features	= DRIVER_COMPUTE_ACCEL,
+	.open			= rocket_open,
+	.postclose		= rocket_postclose,
+	.ioctls			= rocket_drm_driver_ioctls,
+	.num_ioctls		= ARRAY_SIZE(rocket_drm_driver_ioctls),
+	.fops			= &rocket_accel_driver_fops,
+	.name			= "rocket",
+	.desc			= "rocket DRM",
+};
+
+static int rocket_probe(struct platform_device *pdev)
+{
+	if (rdev == NULL) {
+		/* First core probing, initialize DRM device. */
+		rdev = rocket_device_init(drm_dev, &rocket_drm_driver);
+		if (IS_ERR(rdev)) {
+			dev_err(&pdev->dev, "failed to initialize rocket device\n");
+			return PTR_ERR(rdev);
+		}
+	}
+
+	unsigned int core = rdev->num_cores;
+
+	dev_set_drvdata(&pdev->dev, rdev);
+
+	rdev->cores[core].rdev = rdev;
+	rdev->cores[core].dev = &pdev->dev;
+	rdev->cores[core].index = core;
+
+	rdev->num_cores++;
+
+	return rocket_core_init(&rdev->cores[core]);
+}
+
+static void rocket_remove(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+
+	for (unsigned int core = 0; core < rdev->num_cores; core++) {
+		if (rdev->cores[core].dev == dev) {
+			rocket_core_fini(&rdev->cores[core]);
+			rdev->num_cores--;
+			break;
+		}
+	}
+
+	if (rdev->num_cores == 0) {
+		/* Last core removed, deinitialize DRM device. */
+		rocket_device_fini(rdev);
+		rdev = NULL;
+	}
+}
+
+static const struct of_device_id dt_match[] = {
+	{ .compatible = "rockchip,rk3588-rknn-core" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, dt_match);
+
+static int find_core_for_dev(struct device *dev)
+{
+	struct rocket_device *rdev = dev_get_drvdata(dev);
+
+	for (unsigned int core = 0; core < rdev->num_cores; core++) {
+		if (dev == rdev->cores[core].dev)
+			return core;
+	}
+
+	return -1;
+}
+
+static int rocket_device_runtime_resume(struct device *dev)
+{
+	struct rocket_device *rdev = dev_get_drvdata(dev);
+	int core = find_core_for_dev(dev);
+	int err = 0;
+
+	if (core < 0)
+		return -ENODEV;
+
+	err = clk_bulk_prepare_enable(ARRAY_SIZE(rdev->cores[core].clks), rdev->cores[core].clks);
+	if (err) {
+		dev_err(dev, "failed to enable (%d) clocks for core %d\n", err, core);
+		return err;
+	}
+
+	return 0;
+}
+
+static int rocket_device_runtime_suspend(struct device *dev)
+{
+	struct rocket_device *rdev = dev_get_drvdata(dev);
+	int core = find_core_for_dev(dev);
+
+	if (core < 0)
+		return -ENODEV;
+
+	clk_bulk_disable_unprepare(ARRAY_SIZE(rdev->cores[core].clks), rdev->cores[core].clks);
+
+	return 0;
+}
+
+EXPORT_GPL_DEV_PM_OPS(rocket_pm_ops) = {
+	RUNTIME_PM_OPS(rocket_device_runtime_suspend, rocket_device_runtime_resume, NULL)
+	SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume)
+};
+
+static struct platform_driver rocket_driver = {
+	.probe = rocket_probe,
+	.remove = rocket_remove,
+	.driver	 = {
+		.name = "rocket",
+		.pm = pm_ptr(&rocket_pm_ops),
+		.of_match_table = dt_match,
+	},
+};
+
+static int __init rocket_register(void)
+{
+	drm_dev = platform_device_register_simple("rknn", -1, NULL, 0);
+	if (IS_ERR(drm_dev))
+		return PTR_ERR(drm_dev);
+
+	return platform_driver_register(&rocket_driver);
+}
+
+static void __exit rocket_unregister(void)
+{
+	platform_driver_unregister(&rocket_driver);
+
+	platform_device_unregister(drm_dev);
+}
+
+module_init(rocket_register);
+module_exit(rocket_unregister);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("DRM driver for the Rockchip NPU IP");
+MODULE_AUTHOR("Tomeu Vizoso");
diff --git a/drivers/accel/rocket/rocket_drv.h b/drivers/accel/rocket/rocket_drv.h
new file mode 100644
index 000000000000..36b1291b0ead
--- /dev/null
+++ b/drivers/accel/rocket/rocket_drv.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#ifndef __ROCKET_DRV_H__
+#define __ROCKET_DRV_H__
+
+#include "rocket_device.h"
+
+struct rocket_iommu_domain {
+	struct iommu_domain *domain;
+	struct kref kref;
+};
+
+struct rocket_file_priv {
+	struct rocket_device *rdev;
+
+	struct rocket_iommu_domain *domain;
+};
+
+struct rocket_iommu_domain *rocket_iommu_domain_get(struct rocket_file_priv *rocket_priv);
+void rocket_iommu_domain_put(struct rocket_iommu_domain *domain);
+
+#endif

From 658ebeac33517bd3169d4b65ed801e9065d0211a Mon Sep 17 00:00:00 2001
From: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Date: Mon, 21 Jul 2025 11:17:30 +0200
Subject: [PATCH 0107/1455] accel/rocket: Add IOCTL for BO creation

This uses the SHMEM DRM helpers and we map right away to the CPU and NPU
sides, as all buffers are expected to be accessed from both.

v2:
- Sync the IOMMUs for the other cores when mapping and unmapping.

v3:
- Make use of GPL-2.0-only for the copyright notice (Jeff Hugo)

v6:
- Use mutexes guard (Markus Elfring)

v7:
- Assign its own IOMMU domain to each client, for isolation (Daniel
  Stone and Robin Murphy)

v8:
- Correctly acquire a reference to the IOMMU (Robin Murphy)
- Allocate DMA address ourselves with drm_mm (Robin Murphy)
- Use refcount_read (Heiko Stuebner)
- Remove superfluous dma_sync_sgtable_for_device (Robin Murphy)

Reviewed-by: Jeffrey Hugo <quic_jhugo@quicinc.com>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721-6-10-rocket-v9-3-77ebd484941e@tomeuvizoso.net
---
 drivers/accel/rocket/Makefile     |   3 +-
 drivers/accel/rocket/rocket_drv.c |  15 +++-
 drivers/accel/rocket/rocket_drv.h |   4 +
 drivers/accel/rocket/rocket_gem.c | 125 ++++++++++++++++++++++++++++++
 drivers/accel/rocket/rocket_gem.h |  30 +++++++
 include/uapi/drm/rocket_accel.h   |  44 +++++++++++
 6 files changed, 219 insertions(+), 2 deletions(-)
 create mode 100644 drivers/accel/rocket/rocket_gem.c
 create mode 100644 drivers/accel/rocket/rocket_gem.h
 create mode 100644 include/uapi/drm/rocket_accel.h

diff --git a/drivers/accel/rocket/Makefile b/drivers/accel/rocket/Makefile
index abdd75f2492e..4deef267f9e1 100644
--- a/drivers/accel/rocket/Makefile
+++ b/drivers/accel/rocket/Makefile
@@ -5,4 +5,5 @@ obj-$(CONFIG_DRM_ACCEL_ROCKET) := rocket.o
 rocket-y := \
 	rocket_core.o \
 	rocket_device.o \
-	rocket_drv.o
+	rocket_drv.o \
+	rocket_gem.o
diff --git a/drivers/accel/rocket/rocket_drv.c b/drivers/accel/rocket/rocket_drv.c
index a5df94f6b125..8b7fbe9226f4 100644
--- a/drivers/accel/rocket/rocket_drv.c
+++ b/drivers/accel/rocket/rocket_drv.c
@@ -5,6 +5,7 @@
 #include <drm/drm_drv.h>
 #include <drm/drm_gem.h>
 #include <drm/drm_ioctl.h>
+#include <drm/rocket_accel.h>
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/iommu.h>
@@ -13,6 +14,7 @@
 #include <linux/pm_runtime.h>
 
 #include "rocket_drv.h"
+#include "rocket_gem.h"
 
 /*
  * Facade device, used to expose a single DRM device to userspace, that
@@ -69,6 +71,7 @@ rocket_open(struct drm_device *dev, struct drm_file *file)
 {
 	struct rocket_device *rdev = to_rocket_device(dev);
 	struct rocket_file_priv *rocket_priv;
+	u64 start, end;
 	int ret;
 
 	if (!try_module_get(THIS_MODULE))
@@ -89,6 +92,11 @@ rocket_open(struct drm_device *dev, struct drm_file *file)
 
 	file->driver_priv = rocket_priv;
 
+	start = rocket_priv->domain->domain->geometry.aperture_start;
+	end = rocket_priv->domain->domain->geometry.aperture_end;
+	drm_mm_init(&rocket_priv->mm, start, end - start + 1);
+	mutex_init(&rocket_priv->mm_lock);
+
 	return 0;
 
 err_free:
@@ -103,6 +111,8 @@ rocket_postclose(struct drm_device *dev, struct drm_file *file)
 {
 	struct rocket_file_priv *rocket_priv = file->driver_priv;
 
+	mutex_destroy(&rocket_priv->mm_lock);
+	drm_mm_takedown(&rocket_priv->mm);
 	rocket_iommu_domain_put(rocket_priv->domain);
 	kfree(rocket_priv);
 	module_put(THIS_MODULE);
@@ -111,6 +121,8 @@ rocket_postclose(struct drm_device *dev, struct drm_file *file)
 static const struct drm_ioctl_desc rocket_drm_driver_ioctls[] = {
 #define ROCKET_IOCTL(n, func) \
 	DRM_IOCTL_DEF_DRV(ROCKET_##n, rocket_ioctl_##func, 0)
+
+	ROCKET_IOCTL(CREATE_BO, create_bo),
 };
 
 DEFINE_DRM_ACCEL_FOPS(rocket_accel_driver_fops);
@@ -120,9 +132,10 @@ DEFINE_DRM_ACCEL_FOPS(rocket_accel_driver_fops);
  * - 1.0 - initial interface
  */
 static const struct drm_driver rocket_drm_driver = {
-	.driver_features	= DRIVER_COMPUTE_ACCEL,
+	.driver_features	= DRIVER_COMPUTE_ACCEL | DRIVER_GEM,
 	.open			= rocket_open,
 	.postclose		= rocket_postclose,
+	.gem_create_object	= rocket_gem_create_object,
 	.ioctls			= rocket_drm_driver_ioctls,
 	.num_ioctls		= ARRAY_SIZE(rocket_drm_driver_ioctls),
 	.fops			= &rocket_accel_driver_fops,
diff --git a/drivers/accel/rocket/rocket_drv.h b/drivers/accel/rocket/rocket_drv.h
index 36b1291b0ead..2944e0136ab9 100644
--- a/drivers/accel/rocket/rocket_drv.h
+++ b/drivers/accel/rocket/rocket_drv.h
@@ -4,6 +4,8 @@
 #ifndef __ROCKET_DRV_H__
 #define __ROCKET_DRV_H__
 
+#include <drm/drm_mm.h>
+
 #include "rocket_device.h"
 
 struct rocket_iommu_domain {
@@ -15,6 +17,8 @@ struct rocket_file_priv {
 	struct rocket_device *rdev;
 
 	struct rocket_iommu_domain *domain;
+	struct drm_mm mm;
+	struct mutex mm_lock;
 };
 
 struct rocket_iommu_domain *rocket_iommu_domain_get(struct rocket_file_priv *rocket_priv);
diff --git a/drivers/accel/rocket/rocket_gem.c b/drivers/accel/rocket/rocket_gem.c
new file mode 100644
index 000000000000..05cf46040865
--- /dev/null
+++ b/drivers/accel/rocket/rocket_gem.c
@@ -0,0 +1,125 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#include <drm/drm_device.h>
+#include <drm/drm_utils.h>
+#include <drm/rocket_accel.h>
+#include <linux/dma-mapping.h>
+#include <linux/iommu.h>
+
+#include "rocket_drv.h"
+#include "rocket_gem.h"
+
+static void rocket_gem_bo_free(struct drm_gem_object *obj)
+{
+	struct rocket_gem_object *bo = to_rocket_bo(obj);
+	struct rocket_file_priv *rocket_priv = bo->driver_priv;
+	size_t unmapped;
+
+	drm_WARN_ON(obj->dev, refcount_read(&bo->base.pages_use_count) > 1);
+
+	unmapped = iommu_unmap(bo->domain->domain, bo->mm.start, bo->size);
+	drm_WARN_ON(obj->dev, unmapped != bo->size);
+
+	mutex_lock(&rocket_priv->mm_lock);
+	drm_mm_remove_node(&bo->mm);
+	mutex_unlock(&rocket_priv->mm_lock);
+
+	rocket_iommu_domain_put(bo->domain);
+	bo->domain = NULL;
+
+	drm_gem_shmem_free(&bo->base);
+}
+
+static const struct drm_gem_object_funcs rocket_gem_funcs = {
+	.free = rocket_gem_bo_free,
+	.print_info = drm_gem_shmem_object_print_info,
+	.pin = drm_gem_shmem_object_pin,
+	.unpin = drm_gem_shmem_object_unpin,
+	.get_sg_table = drm_gem_shmem_object_get_sg_table,
+	.vmap = drm_gem_shmem_object_vmap,
+	.vunmap = drm_gem_shmem_object_vunmap,
+	.mmap = drm_gem_shmem_object_mmap,
+	.vm_ops = &drm_gem_shmem_vm_ops,
+};
+
+struct drm_gem_object *rocket_gem_create_object(struct drm_device *dev, size_t size)
+{
+	struct rocket_gem_object *obj;
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return ERR_PTR(-ENOMEM);
+
+	obj->base.base.funcs = &rocket_gem_funcs;
+
+	return &obj->base.base;
+}
+
+int rocket_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct rocket_file_priv *rocket_priv = file->driver_priv;
+	struct drm_rocket_create_bo *args = data;
+	struct drm_gem_shmem_object *shmem_obj;
+	struct rocket_gem_object *rkt_obj;
+	struct drm_gem_object *gem_obj;
+	struct sg_table *sgt;
+	int ret;
+
+	shmem_obj = drm_gem_shmem_create(dev, args->size);
+	if (IS_ERR(shmem_obj))
+		return PTR_ERR(shmem_obj);
+
+	gem_obj = &shmem_obj->base;
+	rkt_obj = to_rocket_bo(gem_obj);
+
+	rkt_obj->driver_priv = rocket_priv;
+	rkt_obj->domain = rocket_iommu_domain_get(rocket_priv);
+	rkt_obj->size = args->size;
+	rkt_obj->offset = 0;
+
+	ret = drm_gem_handle_create(file, gem_obj, &args->handle);
+	drm_gem_object_put(gem_obj);
+	if (ret)
+		goto err;
+
+	sgt = drm_gem_shmem_get_pages_sgt(shmem_obj);
+	if (IS_ERR(sgt)) {
+		ret = PTR_ERR(sgt);
+		goto err;
+	}
+
+	mutex_lock(&rocket_priv->mm_lock);
+	ret = drm_mm_insert_node_generic(&rocket_priv->mm, &rkt_obj->mm,
+					 rkt_obj->size, PAGE_SIZE,
+					 0, 0);
+	mutex_unlock(&rocket_priv->mm_lock);
+
+	ret = iommu_map_sgtable(rocket_priv->domain->domain,
+				rkt_obj->mm.start,
+				shmem_obj->sgt,
+				IOMMU_READ | IOMMU_WRITE);
+	if (ret < 0 || ret < args->size) {
+		drm_err(dev, "failed to map buffer: size=%d request_size=%u\n",
+			ret, args->size);
+		ret = -ENOMEM;
+		goto err_remove_node;
+	}
+
+	/* iommu_map_sgtable might have aligned the size */
+	rkt_obj->size = ret;
+	args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node);
+	args->dma_address = rkt_obj->mm.start;
+
+	return 0;
+
+err_remove_node:
+	mutex_lock(&rocket_priv->mm_lock);
+	drm_mm_remove_node(&rkt_obj->mm);
+	mutex_unlock(&rocket_priv->mm_lock);
+
+err:
+	drm_gem_shmem_object_free(gem_obj);
+
+	return ret;
+}
diff --git a/drivers/accel/rocket/rocket_gem.h b/drivers/accel/rocket/rocket_gem.h
new file mode 100644
index 000000000000..91a1fc09c56c
--- /dev/null
+++ b/drivers/accel/rocket/rocket_gem.h
@@ -0,0 +1,30 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#ifndef __ROCKET_GEM_H__
+#define __ROCKET_GEM_H__
+
+#include <drm/drm_gem_shmem_helper.h>
+
+struct rocket_gem_object {
+	struct drm_gem_shmem_object base;
+
+	struct rocket_file_priv *driver_priv;
+
+	struct rocket_iommu_domain *domain;
+	struct drm_mm_node mm;
+	size_t size;
+	u32 offset;
+};
+
+struct drm_gem_object *rocket_gem_create_object(struct drm_device *dev, size_t size);
+
+int rocket_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file *file);
+
+static inline
+struct  rocket_gem_object *to_rocket_bo(struct drm_gem_object *obj)
+{
+	return container_of(to_drm_gem_shmem_obj(obj), struct rocket_gem_object, base);
+}
+
+#endif
diff --git a/include/uapi/drm/rocket_accel.h b/include/uapi/drm/rocket_accel.h
new file mode 100644
index 000000000000..95720702b7c4
--- /dev/null
+++ b/include/uapi/drm/rocket_accel.h
@@ -0,0 +1,44 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2024 Tomeu Vizoso
+ */
+#ifndef __DRM_UAPI_ROCKET_ACCEL_H__
+#define __DRM_UAPI_ROCKET_ACCEL_H__
+
+#include "drm.h"
+
+#if defined(__cplusplus)
+extern "C" {
+#endif
+
+#define DRM_ROCKET_CREATE_BO			0x00
+
+#define DRM_IOCTL_ROCKET_CREATE_BO		DRM_IOWR(DRM_COMMAND_BASE + DRM_ROCKET_CREATE_BO, struct drm_rocket_create_bo)
+
+/**
+ * struct drm_rocket_create_bo - ioctl argument for creating Rocket BOs.
+ *
+ */
+struct drm_rocket_create_bo {
+	/** Input: Size of the requested BO. */
+	__u32 size;
+
+	/** Output: GEM handle for the BO. */
+	__u32 handle;
+
+	/**
+	 * Output: DMA address for the BO in the NPU address space.  This address
+	 * is private to the DRM fd and is valid for the lifetime of the GEM
+	 * handle.
+	 */
+	__u64 dma_address;
+
+	/** Output: Offset into the drm node to use for subsequent mmap call. */
+	__u64 offset;
+};
+
+#if defined(__cplusplus)
+}
+#endif
+
+#endif /* __DRM_UAPI_ROCKET_ACCEL_H__ */

From 0810d5ad88a18f1e6d549853a388ad0316f74e36 Mon Sep 17 00:00:00 2001
From: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Date: Mon, 21 Jul 2025 11:17:31 +0200
Subject: [PATCH 0108/1455] accel/rocket: Add job submission IOCTL

Using the DRM GPU scheduler infrastructure, with a scheduler for each
core.

Userspace can decide for a series of tasks to be executed sequentially
in the same core, so SRAM locality can be taken advantage of.

The job submission code was initially based on Panfrost.

v2:
- Remove hardcoded number of cores
- Misc. style fixes (Jeffrey Hugo)
- Repack IOCTL struct (Jeffrey Hugo)

v3:
- Adapt to a split of the register block in the DT bindings (Nicolas
  Frattaroli)
- Make use of GPL-2.0-only for the copyright notice (Jeff Hugo)
- Use drm_* logging functions (Thomas Zimmermann)
- Rename reg i/o macros (Thomas Zimmermann)
- Add padding to ioctls and check for zero (Jeff Hugo)
- Improve error handling (Nicolas Frattaroli)

v6:
- Use mutexes guard (Markus Elfring)
- Use u64_to_user_ptr (Jeff Hugo)
- Drop rocket_fence (Rob Herring)

v7:
- Assign its own IOMMU domain to each client, for isolation (Daniel
  Stone and Robin Murphy)

v8:
- Use reset lines to reset the cores (Robin Murphy)
- Use the macros to compute the values for the bitfields (Robin Murphy)
- More descriptive name for the IRQ (Robin Murphy)
- Simplify job interrupt handing (Robin Murphy)
- Correctly acquire a reference to the IOMMU (Robin Murphy)
- Specify the size of the embedded structs in the IOCTLs for future
  extensibility (Rob Herring)
- Expose only 32 bits for the address of the regcmd BO (Robin Murphy)

Tested-by: Heiko Stuebner <heiko@sntech.de>
Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721-6-10-rocket-v9-4-77ebd484941e@tomeuvizoso.net
---
 drivers/accel/rocket/Makefile        |   3 +-
 drivers/accel/rocket/rocket_core.c   |  10 +
 drivers/accel/rocket/rocket_core.h   |  15 +
 drivers/accel/rocket/rocket_device.c |   4 +
 drivers/accel/rocket/rocket_device.h |   2 +
 drivers/accel/rocket/rocket_drv.c    |  14 +
 drivers/accel/rocket/rocket_drv.h    |   3 +
 drivers/accel/rocket/rocket_job.c    | 636 +++++++++++++++++++++++++++
 drivers/accel/rocket/rocket_job.h    |  52 +++
 include/uapi/drm/rocket_accel.h      |  64 +++
 10 files changed, 802 insertions(+), 1 deletion(-)
 create mode 100644 drivers/accel/rocket/rocket_job.c
 create mode 100644 drivers/accel/rocket/rocket_job.h

diff --git a/drivers/accel/rocket/Makefile b/drivers/accel/rocket/Makefile
index 4deef267f9e1..3713dfe223d6 100644
--- a/drivers/accel/rocket/Makefile
+++ b/drivers/accel/rocket/Makefile
@@ -6,4 +6,5 @@ rocket-y := \
 	rocket_core.o \
 	rocket_device.o \
 	rocket_drv.o \
-	rocket_gem.o
+	rocket_gem.o \
+	rocket_job.o
diff --git a/drivers/accel/rocket/rocket_core.c b/drivers/accel/rocket/rocket_core.c
index 9be964b5fbae..72fb5e5798fa 100644
--- a/drivers/accel/rocket/rocket_core.c
+++ b/drivers/accel/rocket/rocket_core.c
@@ -12,6 +12,7 @@
 #include <linux/reset.h>
 
 #include "rocket_core.h"
+#include "rocket_job.h"
 
 int rocket_core_init(struct rocket_core *core)
 {
@@ -57,6 +58,10 @@ int rocket_core_init(struct rocket_core *core)
 
 	core->iommu_group = iommu_group_get(dev);
 
+	err = rocket_job_init(core);
+	if (err)
+		return err;
+
 	pm_runtime_use_autosuspend(dev);
 
 	/*
@@ -70,6 +75,10 @@ int rocket_core_init(struct rocket_core *core)
 	pm_runtime_enable(dev);
 
 	err = pm_runtime_get_sync(dev);
+	if (err) {
+		rocket_job_fini(core);
+		return err;
+	}
 
 	version = rocket_pc_readl(core, VERSION);
 	version += rocket_pc_readl(core, VERSION_NUM) & 0xffff;
@@ -88,6 +97,7 @@ void rocket_core_fini(struct rocket_core *core)
 	pm_runtime_disable(core->dev);
 	iommu_group_put(core->iommu_group);
 	core->iommu_group = NULL;
+	rocket_job_fini(core);
 }
 
 void rocket_core_reset(struct rocket_core *core)
diff --git a/drivers/accel/rocket/rocket_core.h b/drivers/accel/rocket/rocket_core.h
index 660de2d70f7d..f6d7382854ca 100644
--- a/drivers/accel/rocket/rocket_core.h
+++ b/drivers/accel/rocket/rocket_core.h
@@ -40,6 +40,21 @@ struct rocket_core {
 	struct reset_control_bulk_data resets[2];
 
 	struct iommu_group *iommu_group;
+
+	struct mutex job_lock;
+	struct rocket_job *in_flight_job;
+
+	spinlock_t fence_lock;
+
+	struct {
+		struct workqueue_struct *wq;
+		struct work_struct work;
+		atomic_t pending;
+	} reset;
+
+	struct drm_gpu_scheduler sched;
+	u64 fence_context;
+	u64 emit_seqno;
 };
 
 int rocket_core_init(struct rocket_core *core);
diff --git a/drivers/accel/rocket/rocket_device.c b/drivers/accel/rocket/rocket_device.c
index b05a0df91d48..46e6ee1e72c5 100644
--- a/drivers/accel/rocket/rocket_device.c
+++ b/drivers/accel/rocket/rocket_device.c
@@ -41,6 +41,10 @@ struct rocket_device *rocket_device_init(struct platform_device *pdev,
 	if (err)
 		return ERR_PTR(err);
 
+	err = devm_mutex_init(dev, &rdev->sched_lock);
+	if (err)
+		return ERR_PTR(-ENOMEM);
+
 	err = drm_dev_register(ddev, 0);
 	if (err)
 		return ERR_PTR(err);
diff --git a/drivers/accel/rocket/rocket_device.h b/drivers/accel/rocket/rocket_device.h
index a5a5857bb199..ce662abc01d3 100644
--- a/drivers/accel/rocket/rocket_device.h
+++ b/drivers/accel/rocket/rocket_device.h
@@ -15,6 +15,8 @@
 struct rocket_device {
 	struct drm_device ddev;
 
+	struct mutex sched_lock;
+
 	struct rocket_core *cores;
 	unsigned int num_cores;
 };
diff --git a/drivers/accel/rocket/rocket_drv.c b/drivers/accel/rocket/rocket_drv.c
index 8b7fbe9226f4..a21aa9aa189b 100644
--- a/drivers/accel/rocket/rocket_drv.c
+++ b/drivers/accel/rocket/rocket_drv.c
@@ -15,6 +15,7 @@
 
 #include "rocket_drv.h"
 #include "rocket_gem.h"
+#include "rocket_job.h"
 
 /*
  * Facade device, used to expose a single DRM device to userspace, that
@@ -97,8 +98,16 @@ rocket_open(struct drm_device *dev, struct drm_file *file)
 	drm_mm_init(&rocket_priv->mm, start, end - start + 1);
 	mutex_init(&rocket_priv->mm_lock);
 
+	ret = rocket_job_open(rocket_priv);
+	if (ret)
+		goto err_mm_takedown;
+
 	return 0;
 
+err_mm_takedown:
+	mutex_destroy(&rocket_priv->mm_lock);
+	drm_mm_takedown(&rocket_priv->mm);
+	rocket_iommu_domain_put(rocket_priv->domain);
 err_free:
 	kfree(rocket_priv);
 err_put_mod:
@@ -111,6 +120,7 @@ rocket_postclose(struct drm_device *dev, struct drm_file *file)
 {
 	struct rocket_file_priv *rocket_priv = file->driver_priv;
 
+	rocket_job_close(rocket_priv);
 	mutex_destroy(&rocket_priv->mm_lock);
 	drm_mm_takedown(&rocket_priv->mm);
 	rocket_iommu_domain_put(rocket_priv->domain);
@@ -123,6 +133,7 @@ static const struct drm_ioctl_desc rocket_drm_driver_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(ROCKET_##n, rocket_ioctl_##func, 0)
 
 	ROCKET_IOCTL(CREATE_BO, create_bo),
+	ROCKET_IOCTL(SUBMIT, submit),
 };
 
 DEFINE_DRM_ACCEL_FOPS(rocket_accel_driver_fops);
@@ -230,6 +241,9 @@ static int rocket_device_runtime_suspend(struct device *dev)
 	if (core < 0)
 		return -ENODEV;
 
+	if (!rocket_job_is_idle(&rdev->cores[core]))
+		return -EBUSY;
+
 	clk_bulk_disable_unprepare(ARRAY_SIZE(rdev->cores[core].clks), rdev->cores[core].clks);
 
 	return 0;
diff --git a/drivers/accel/rocket/rocket_drv.h b/drivers/accel/rocket/rocket_drv.h
index 2944e0136ab9..f50634935b60 100644
--- a/drivers/accel/rocket/rocket_drv.h
+++ b/drivers/accel/rocket/rocket_drv.h
@@ -5,6 +5,7 @@
 #define __ROCKET_DRV_H__
 
 #include <drm/drm_mm.h>
+#include <drm/gpu_scheduler.h>
 
 #include "rocket_device.h"
 
@@ -19,6 +20,8 @@ struct rocket_file_priv {
 	struct rocket_iommu_domain *domain;
 	struct drm_mm mm;
 	struct mutex mm_lock;
+
+	struct drm_sched_entity sched_entity;
 };
 
 struct rocket_iommu_domain *rocket_iommu_domain_get(struct rocket_file_priv *rocket_priv);
diff --git a/drivers/accel/rocket/rocket_job.c b/drivers/accel/rocket/rocket_job.c
new file mode 100644
index 000000000000..e731da15ebff
--- /dev/null
+++ b/drivers/accel/rocket/rocket_job.c
@@ -0,0 +1,636 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright 2019 Linaro, Ltd, Rob Herring <robh@kernel.org> */
+/* Copyright 2019 Collabora ltd. */
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#include <drm/drm_print.h>
+#include <drm/drm_file.h>
+#include <drm/drm_gem.h>
+#include <drm/rocket_accel.h>
+#include <linux/interrupt.h>
+#include <linux/iommu.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+
+#include "rocket_core.h"
+#include "rocket_device.h"
+#include "rocket_drv.h"
+#include "rocket_job.h"
+#include "rocket_registers.h"
+
+#define JOB_TIMEOUT_MS 500
+
+static struct rocket_job *
+to_rocket_job(struct drm_sched_job *sched_job)
+{
+	return container_of(sched_job, struct rocket_job, base);
+}
+
+static const char *rocket_fence_get_driver_name(struct dma_fence *fence)
+{
+	return "rocket";
+}
+
+static const char *rocket_fence_get_timeline_name(struct dma_fence *fence)
+{
+	return "rockchip-npu";
+}
+
+static const struct dma_fence_ops rocket_fence_ops = {
+	.get_driver_name = rocket_fence_get_driver_name,
+	.get_timeline_name = rocket_fence_get_timeline_name,
+};
+
+static struct dma_fence *rocket_fence_create(struct rocket_core *core)
+{
+	struct dma_fence *fence;
+
+	fence = kzalloc(sizeof(*fence), GFP_KERNEL);
+	if (!fence)
+		return ERR_PTR(-ENOMEM);
+
+	dma_fence_init(fence, &rocket_fence_ops, &core->fence_lock,
+		       core->fence_context, ++core->emit_seqno);
+
+	return fence;
+}
+
+static int
+rocket_copy_tasks(struct drm_device *dev,
+		  struct drm_file *file_priv,
+		  struct drm_rocket_job *job,
+		  struct rocket_job *rjob)
+{
+	int ret = 0;
+
+	if (job->task_struct_size < sizeof(struct drm_rocket_task))
+		return -EINVAL;
+
+	rjob->task_count = job->task_count;
+
+	if (!rjob->task_count)
+		return 0;
+
+	rjob->tasks = kvmalloc_array(job->task_count, sizeof(*rjob->tasks), GFP_KERNEL);
+	if (!rjob->tasks) {
+		drm_dbg(dev, "Failed to allocate task array\n");
+		return -ENOMEM;
+	}
+
+	for (int i = 0; i < rjob->task_count; i++) {
+		struct drm_rocket_task task = {0};
+
+		if (copy_from_user(&task,
+				   u64_to_user_ptr(job->tasks) + i * job->task_struct_size,
+				   sizeof(task))) {
+			drm_dbg(dev, "Failed to copy incoming tasks\n");
+			ret = -EFAULT;
+			goto fail;
+		}
+
+		if (task.regcmd_count == 0) {
+			drm_dbg(dev, "regcmd_count field in drm_rocket_task should be > 0.\n");
+			ret = -EINVAL;
+			goto fail;
+		}
+
+		rjob->tasks[i].regcmd = task.regcmd;
+		rjob->tasks[i].regcmd_count = task.regcmd_count;
+	}
+
+	return 0;
+
+fail:
+	kvfree(rjob->tasks);
+	return ret;
+}
+
+static void rocket_job_hw_submit(struct rocket_core *core, struct rocket_job *job)
+{
+	struct rocket_task *task;
+	unsigned int extra_bit;
+
+	/* Don't queue the job if a reset is in progress */
+	if (atomic_read(&core->reset.pending))
+		return;
+
+	/* GO ! */
+
+	task = &job->tasks[job->next_task_idx];
+	job->next_task_idx++;
+
+	rocket_pc_writel(core, BASE_ADDRESS, 0x1);
+
+	 /* From rknpu, in the TRM this bit is marked as reserved */
+	extra_bit = 0x10000000 * core->index;
+	rocket_cna_writel(core, S_POINTER, CNA_S_POINTER_POINTER_PP_EN(1) |
+					   CNA_S_POINTER_EXECUTER_PP_EN(1) |
+					   CNA_S_POINTER_POINTER_PP_MODE(1) |
+					   extra_bit);
+
+	rocket_core_writel(core, S_POINTER, CORE_S_POINTER_POINTER_PP_EN(1) |
+					    CORE_S_POINTER_EXECUTER_PP_EN(1) |
+					    CORE_S_POINTER_POINTER_PP_MODE(1) |
+					    extra_bit);
+
+	rocket_pc_writel(core, BASE_ADDRESS, task->regcmd);
+	rocket_pc_writel(core, REGISTER_AMOUNTS,
+			 PC_REGISTER_AMOUNTS_PC_DATA_AMOUNT((task->regcmd_count + 1) / 2 - 1));
+
+	rocket_pc_writel(core, INTERRUPT_MASK, PC_INTERRUPT_MASK_DPU_0 | PC_INTERRUPT_MASK_DPU_1);
+	rocket_pc_writel(core, INTERRUPT_CLEAR, PC_INTERRUPT_CLEAR_DPU_0 | PC_INTERRUPT_CLEAR_DPU_1);
+
+	rocket_pc_writel(core, TASK_CON, PC_TASK_CON_RESERVED_0(1) |
+					 PC_TASK_CON_TASK_COUNT_CLEAR(1) |
+					 PC_TASK_CON_TASK_NUMBER(1) |
+					 PC_TASK_CON_TASK_PP_EN(1));
+
+	rocket_pc_writel(core, TASK_DMA_BASE_ADDR, PC_TASK_DMA_BASE_ADDR_DMA_BASE_ADDR(0x0));
+
+	rocket_pc_writel(core, OPERATION_ENABLE, PC_OPERATION_ENABLE_OP_EN(1));
+
+	dev_dbg(core->dev, "Submitted regcmd at 0x%llx to core %d", task->regcmd, core->index);
+}
+
+static int rocket_acquire_object_fences(struct drm_gem_object **bos,
+					int bo_count,
+					struct drm_sched_job *job,
+					bool is_write)
+{
+	int i, ret;
+
+	for (i = 0; i < bo_count; i++) {
+		ret = dma_resv_reserve_fences(bos[i]->resv, 1);
+		if (ret)
+			return ret;
+
+		ret = drm_sched_job_add_implicit_dependencies(job, bos[i],
+							      is_write);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static void rocket_attach_object_fences(struct drm_gem_object **bos,
+					int bo_count,
+					struct dma_fence *fence)
+{
+	int i;
+
+	for (i = 0; i < bo_count; i++)
+		dma_resv_add_fence(bos[i]->resv, fence, DMA_RESV_USAGE_WRITE);
+}
+
+static int rocket_job_push(struct rocket_job *job)
+{
+	struct rocket_device *rdev = job->rdev;
+	struct drm_gem_object **bos;
+	struct ww_acquire_ctx acquire_ctx;
+	int ret = 0;
+
+	bos = kvmalloc_array(job->in_bo_count + job->out_bo_count, sizeof(void *),
+			     GFP_KERNEL);
+	memcpy(bos, job->in_bos, job->in_bo_count * sizeof(void *));
+	memcpy(&bos[job->in_bo_count], job->out_bos, job->out_bo_count * sizeof(void *));
+
+	ret = drm_gem_lock_reservations(bos, job->in_bo_count + job->out_bo_count, &acquire_ctx);
+	if (ret)
+		goto err;
+
+	scoped_guard(mutex, &rdev->sched_lock) {
+		drm_sched_job_arm(&job->base);
+
+		job->inference_done_fence = dma_fence_get(&job->base.s_fence->finished);
+
+		ret = rocket_acquire_object_fences(job->in_bos, job->in_bo_count, &job->base, false);
+		if (ret)
+			goto err_unlock;
+
+		ret = rocket_acquire_object_fences(job->out_bos, job->out_bo_count, &job->base, true);
+		if (ret)
+			goto err_unlock;
+
+		kref_get(&job->refcount); /* put by scheduler job completion */
+
+		drm_sched_entity_push_job(&job->base);
+	}
+
+	rocket_attach_object_fences(job->out_bos, job->out_bo_count, job->inference_done_fence);
+
+err_unlock:
+	drm_gem_unlock_reservations(bos, job->in_bo_count + job->out_bo_count, &acquire_ctx);
+err:
+	kfree(bos);
+
+	return ret;
+}
+
+static void rocket_job_cleanup(struct kref *ref)
+{
+	struct rocket_job *job = container_of(ref, struct rocket_job,
+						refcount);
+	unsigned int i;
+
+	rocket_iommu_domain_put(job->domain);
+
+	dma_fence_put(job->done_fence);
+	dma_fence_put(job->inference_done_fence);
+
+	if (job->in_bos) {
+		for (i = 0; i < job->in_bo_count; i++)
+			drm_gem_object_put(job->in_bos[i]);
+
+		kvfree(job->in_bos);
+	}
+
+	if (job->out_bos) {
+		for (i = 0; i < job->out_bo_count; i++)
+			drm_gem_object_put(job->out_bos[i]);
+
+		kvfree(job->out_bos);
+	}
+
+	kvfree(job->tasks);
+
+	kfree(job);
+}
+
+static void rocket_job_put(struct rocket_job *job)
+{
+	kref_put(&job->refcount, rocket_job_cleanup);
+}
+
+static void rocket_job_free(struct drm_sched_job *sched_job)
+{
+	struct rocket_job *job = to_rocket_job(sched_job);
+
+	drm_sched_job_cleanup(sched_job);
+
+	rocket_job_put(job);
+}
+
+static struct rocket_core *sched_to_core(struct rocket_device *rdev,
+					 struct drm_gpu_scheduler *sched)
+{
+	unsigned int core;
+
+	for (core = 0; core < rdev->num_cores; core++) {
+		if (&rdev->cores[core].sched == sched)
+			return &rdev->cores[core];
+	}
+
+	return NULL;
+}
+
+static struct dma_fence *rocket_job_run(struct drm_sched_job *sched_job)
+{
+	struct rocket_job *job = to_rocket_job(sched_job);
+	struct rocket_device *rdev = job->rdev;
+	struct rocket_core *core = sched_to_core(rdev, sched_job->sched);
+	struct dma_fence *fence = NULL;
+	int ret;
+
+	if (unlikely(job->base.s_fence->finished.error))
+		return NULL;
+
+	/*
+	 * Nothing to execute: can happen if the job has finished while
+	 * we were resetting the NPU.
+	 */
+	if (job->next_task_idx == job->task_count)
+		return NULL;
+
+	fence = rocket_fence_create(core);
+	if (IS_ERR(fence))
+		return fence;
+
+	if (job->done_fence)
+		dma_fence_put(job->done_fence);
+	job->done_fence = dma_fence_get(fence);
+
+	ret = pm_runtime_get_sync(core->dev);
+	if (ret < 0)
+		return fence;
+
+	ret = iommu_attach_group(job->domain->domain, core->iommu_group);
+	if (ret < 0)
+		return fence;
+
+	scoped_guard(mutex, &core->job_lock) {
+		core->in_flight_job = job;
+		rocket_job_hw_submit(core, job);
+	}
+
+	return fence;
+}
+
+static void rocket_job_handle_irq(struct rocket_core *core)
+{
+	pm_runtime_mark_last_busy(core->dev);
+
+	rocket_pc_writel(core, OPERATION_ENABLE, 0x0);
+	rocket_pc_writel(core, INTERRUPT_CLEAR, 0x1ffff);
+
+	scoped_guard(mutex, &core->job_lock)
+		if (core->in_flight_job) {
+			if (core->in_flight_job->next_task_idx < core->in_flight_job->task_count) {
+				rocket_job_hw_submit(core, core->in_flight_job);
+				return;
+			}
+
+			iommu_detach_group(NULL, iommu_group_get(core->dev));
+			dma_fence_signal(core->in_flight_job->done_fence);
+			pm_runtime_put_autosuspend(core->dev);
+			core->in_flight_job = NULL;
+		}
+}
+
+static void
+rocket_reset(struct rocket_core *core, struct drm_sched_job *bad)
+{
+	if (!atomic_read(&core->reset.pending))
+		return;
+
+	drm_sched_stop(&core->sched, bad);
+
+	/*
+	 * Remaining interrupts have been handled, but we might still have
+	 * stuck jobs. Let's make sure the PM counters stay balanced by
+	 * manually calling pm_runtime_put_noidle().
+	 */
+	scoped_guard(mutex, &core->job_lock) {
+		if (core->in_flight_job)
+			pm_runtime_put_noidle(core->dev);
+
+		iommu_detach_group(NULL, core->iommu_group);
+
+		core->in_flight_job = NULL;
+	}
+
+	/* Proceed with reset now. */
+	rocket_core_reset(core);
+
+	/* NPU has been reset, we can clear the reset pending bit. */
+	atomic_set(&core->reset.pending, 0);
+
+	/* Restart the scheduler */
+	drm_sched_start(&core->sched, 0);
+}
+
+static enum drm_gpu_sched_stat rocket_job_timedout(struct drm_sched_job *sched_job)
+{
+	struct rocket_job *job = to_rocket_job(sched_job);
+	struct rocket_device *rdev = job->rdev;
+	struct rocket_core *core = sched_to_core(rdev, sched_job->sched);
+
+	dev_err(core->dev, "NPU job timed out");
+
+	atomic_set(&core->reset.pending, 1);
+	rocket_reset(core, sched_job);
+
+	return DRM_GPU_SCHED_STAT_NOMINAL;
+}
+
+static void rocket_reset_work(struct work_struct *work)
+{
+	struct rocket_core *core;
+
+	core = container_of(work, struct rocket_core, reset.work);
+	rocket_reset(core, NULL);
+}
+
+static const struct drm_sched_backend_ops rocket_sched_ops = {
+	.run_job = rocket_job_run,
+	.timedout_job = rocket_job_timedout,
+	.free_job = rocket_job_free
+};
+
+static irqreturn_t rocket_job_irq_handler_thread(int irq, void *data)
+{
+	struct rocket_core *core = data;
+
+	rocket_job_handle_irq(core);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t rocket_job_irq_handler(int irq, void *data)
+{
+	struct rocket_core *core = data;
+	u32 raw_status = rocket_pc_readl(core, INTERRUPT_RAW_STATUS);
+
+	WARN_ON(raw_status & PC_INTERRUPT_RAW_STATUS_DMA_READ_ERROR);
+	WARN_ON(raw_status & PC_INTERRUPT_RAW_STATUS_DMA_READ_ERROR);
+
+	if (!(raw_status & PC_INTERRUPT_RAW_STATUS_DPU_0 ||
+	      raw_status & PC_INTERRUPT_RAW_STATUS_DPU_1))
+		return IRQ_NONE;
+
+	rocket_pc_writel(core, INTERRUPT_MASK, 0x0);
+
+	return IRQ_WAKE_THREAD;
+}
+
+int rocket_job_init(struct rocket_core *core)
+{
+	struct drm_sched_init_args args = {
+		.ops = &rocket_sched_ops,
+		.num_rqs = DRM_SCHED_PRIORITY_COUNT,
+		.credit_limit = 1,
+		.timeout = msecs_to_jiffies(JOB_TIMEOUT_MS),
+		.name = dev_name(core->dev),
+		.dev = core->dev,
+	};
+	int ret;
+
+	INIT_WORK(&core->reset.work, rocket_reset_work);
+	spin_lock_init(&core->fence_lock);
+	mutex_init(&core->job_lock);
+
+	core->irq = platform_get_irq(to_platform_device(core->dev), 0);
+	if (core->irq < 0)
+		return core->irq;
+
+	ret = devm_request_threaded_irq(core->dev, core->irq,
+					rocket_job_irq_handler,
+					rocket_job_irq_handler_thread,
+					IRQF_SHARED, dev_name(core->dev),
+					core);
+	if (ret) {
+		dev_err(core->dev, "failed to request job irq");
+		return ret;
+	}
+
+	core->reset.wq = alloc_ordered_workqueue("rocket-reset-%d", 0, core->index);
+	if (!core->reset.wq)
+		return -ENOMEM;
+
+	core->fence_context = dma_fence_context_alloc(1);
+
+	args.timeout_wq = core->reset.wq;
+	ret = drm_sched_init(&core->sched, &args);
+	if (ret) {
+		dev_err(core->dev, "Failed to create scheduler: %d.", ret);
+		goto err_sched;
+	}
+
+	return 0;
+
+err_sched:
+	drm_sched_fini(&core->sched);
+
+	destroy_workqueue(core->reset.wq);
+	return ret;
+}
+
+void rocket_job_fini(struct rocket_core *core)
+{
+	drm_sched_fini(&core->sched);
+
+	cancel_work_sync(&core->reset.work);
+	destroy_workqueue(core->reset.wq);
+}
+
+int rocket_job_open(struct rocket_file_priv *rocket_priv)
+{
+	struct rocket_device *rdev = rocket_priv->rdev;
+	struct drm_gpu_scheduler **scheds = kmalloc_array(rdev->num_cores, sizeof(scheds),
+							  GFP_KERNEL);
+	unsigned int core;
+	int ret;
+
+	for (core = 0; core < rdev->num_cores; core++)
+		scheds[core] = &rdev->cores[core].sched;
+
+	ret = drm_sched_entity_init(&rocket_priv->sched_entity,
+				    DRM_SCHED_PRIORITY_NORMAL,
+				    scheds,
+				    rdev->num_cores, NULL);
+	if (WARN_ON(ret))
+		return ret;
+
+	return 0;
+}
+
+void rocket_job_close(struct rocket_file_priv *rocket_priv)
+{
+	struct drm_sched_entity *entity = &rocket_priv->sched_entity;
+
+	kfree(entity->sched_list);
+	drm_sched_entity_destroy(entity);
+}
+
+int rocket_job_is_idle(struct rocket_core *core)
+{
+	/* If there are any jobs in this HW queue, we're not idle */
+	if (atomic_read(&core->sched.credit_count))
+		return false;
+
+	return true;
+}
+
+static int rocket_ioctl_submit_job(struct drm_device *dev, struct drm_file *file,
+				   struct drm_rocket_job *job)
+{
+	struct rocket_device *rdev = to_rocket_device(dev);
+	struct rocket_file_priv *file_priv = file->driver_priv;
+	struct rocket_job *rjob = NULL;
+	int ret = 0;
+
+	if (job->task_count == 0)
+		return -EINVAL;
+
+	rjob = kzalloc(sizeof(*rjob), GFP_KERNEL);
+	if (!rjob)
+		return -ENOMEM;
+
+	kref_init(&rjob->refcount);
+
+	rjob->rdev = rdev;
+
+	ret = drm_sched_job_init(&rjob->base,
+				 &file_priv->sched_entity,
+				 1, NULL);
+	if (ret)
+		goto out_put_job;
+
+	ret = rocket_copy_tasks(dev, file, job, rjob);
+	if (ret)
+		goto out_cleanup_job;
+
+	ret = drm_gem_objects_lookup(file, u64_to_user_ptr(job->in_bo_handles),
+				     job->in_bo_handle_count, &rjob->in_bos);
+	if (ret)
+		goto out_cleanup_job;
+
+	rjob->in_bo_count = job->in_bo_handle_count;
+
+	ret = drm_gem_objects_lookup(file, u64_to_user_ptr(job->out_bo_handles),
+				     job->out_bo_handle_count, &rjob->out_bos);
+	if (ret)
+		goto out_cleanup_job;
+
+	rjob->out_bo_count = job->out_bo_handle_count;
+
+	rjob->domain = rocket_iommu_domain_get(file_priv);
+
+	ret = rocket_job_push(rjob);
+	if (ret)
+		goto out_cleanup_job;
+
+out_cleanup_job:
+	if (ret)
+		drm_sched_job_cleanup(&rjob->base);
+out_put_job:
+	rocket_job_put(rjob);
+
+	return ret;
+}
+
+int rocket_ioctl_submit(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_rocket_submit *args = data;
+	struct drm_rocket_job *jobs;
+	int ret = 0;
+	unsigned int i = 0;
+
+	if (args->job_count == 0)
+		return 0;
+
+	if (args->job_struct_size < sizeof(struct drm_rocket_job)) {
+		drm_dbg(dev, "job_struct_size field in drm_rocket_submit struct is too small.\n");
+		return -EINVAL;
+	}
+
+	if (args->reserved != 0) {
+		drm_dbg(dev, "Reserved field in drm_rocket_submit struct should be 0.\n");
+		return -EINVAL;
+	}
+
+	jobs = kvmalloc_array(args->job_count, sizeof(*jobs), GFP_KERNEL);
+	if (!jobs) {
+		drm_dbg(dev, "Failed to allocate incoming job array\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < args->job_count; i++) {
+		if (copy_from_user(&jobs[i],
+				   u64_to_user_ptr(args->jobs) + i * args->job_struct_size,
+				   sizeof(*jobs))) {
+			ret = -EFAULT;
+			drm_dbg(dev, "Failed to copy incoming job array\n");
+			goto exit;
+		}
+	}
+
+
+	for (i = 0; i < args->job_count; i++)
+		rocket_ioctl_submit_job(dev, file, &jobs[i]);
+
+exit:
+	kfree(jobs);
+
+	return ret;
+}
diff --git a/drivers/accel/rocket/rocket_job.h b/drivers/accel/rocket/rocket_job.h
new file mode 100644
index 000000000000..4ae00feec3b9
--- /dev/null
+++ b/drivers/accel/rocket/rocket_job.h
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright 2024-2025 Tomeu Vizoso <tomeu@tomeuvizoso.net> */
+
+#ifndef __ROCKET_JOB_H__
+#define __ROCKET_JOB_H__
+
+#include <drm/drm_drv.h>
+#include <drm/gpu_scheduler.h>
+
+#include "rocket_core.h"
+#include "rocket_drv.h"
+
+struct rocket_task {
+	u64 regcmd;
+	u32 regcmd_count;
+};
+
+struct rocket_job {
+	struct drm_sched_job base;
+
+	struct rocket_device *rdev;
+
+	struct drm_gem_object **in_bos;
+	struct drm_gem_object **out_bos;
+
+	u32 in_bo_count;
+	u32 out_bo_count;
+
+	struct rocket_task *tasks;
+	u32 task_count;
+	u32 next_task_idx;
+
+	/* Fence to be signaled by drm-sched once its done with the job */
+	struct dma_fence *inference_done_fence;
+
+	/* Fence to be signaled by IRQ handler when the job is complete. */
+	struct dma_fence *done_fence;
+
+	struct rocket_iommu_domain *domain;
+
+	struct kref refcount;
+};
+
+int rocket_ioctl_submit(struct drm_device *dev, void *data, struct drm_file *file);
+
+int rocket_job_init(struct rocket_core *core);
+void rocket_job_fini(struct rocket_core *core);
+int rocket_job_open(struct rocket_file_priv *rocket_priv);
+void rocket_job_close(struct rocket_file_priv *rocket_priv);
+int rocket_job_is_idle(struct rocket_core *core);
+
+#endif
diff --git a/include/uapi/drm/rocket_accel.h b/include/uapi/drm/rocket_accel.h
index 95720702b7c4..374f8370ac9d 100644
--- a/include/uapi/drm/rocket_accel.h
+++ b/include/uapi/drm/rocket_accel.h
@@ -12,8 +12,10 @@ extern "C" {
 #endif
 
 #define DRM_ROCKET_CREATE_BO			0x00
+#define DRM_ROCKET_SUBMIT			0x01
 
 #define DRM_IOCTL_ROCKET_CREATE_BO		DRM_IOWR(DRM_COMMAND_BASE + DRM_ROCKET_CREATE_BO, struct drm_rocket_create_bo)
+#define DRM_IOCTL_ROCKET_SUBMIT			DRM_IOW(DRM_COMMAND_BASE + DRM_ROCKET_SUBMIT, struct drm_rocket_submit)
 
 /**
  * struct drm_rocket_create_bo - ioctl argument for creating Rocket BOs.
@@ -37,6 +39,68 @@ struct drm_rocket_create_bo {
 	__u64 offset;
 };
 
+/**
+ * struct drm_rocket_task - A task to be run on the NPU
+ *
+ * A task is the smallest unit of work that can be run on the NPU.
+ */
+struct drm_rocket_task {
+	/** Input: DMA address to NPU mapping of register command buffer */
+	__u32 regcmd;
+
+	/** Input: Number of commands in the register command buffer */
+	__u32 regcmd_count;
+};
+
+/**
+ * struct drm_rocket_job - A job to be run on the NPU
+ *
+ * The kernel will schedule the execution of this job taking into account its
+ * dependencies with other jobs. All tasks in the same job will be executed
+ * sequentially on the same core, to benefit from memory residency in SRAM.
+ */
+struct drm_rocket_job {
+	/** Input: Pointer to an array of struct drm_rocket_task. */
+	__u64 tasks;
+
+	/** Input: Pointer to a u32 array of the BOs that are read by the job. */
+	__u64 in_bo_handles;
+
+	/** Input: Pointer to a u32 array of the BOs that are written to by the job. */
+	__u64 out_bo_handles;
+
+	/** Input: Number of tasks passed in. */
+	__u32 task_count;
+
+	/** Input: Size in bytes of the structs in the @tasks field. */
+	__u32 task_struct_size;
+
+	/** Input: Number of input BO handles passed in (size is that times 4). */
+	__u32 in_bo_handle_count;
+
+	/** Input: Number of output BO handles passed in (size is that times 4). */
+	__u32 out_bo_handle_count;
+};
+
+/**
+ * struct drm_rocket_submit - ioctl argument for submitting commands to the NPU.
+ *
+ * The kernel will schedule the execution of these jobs in dependency order.
+ */
+struct drm_rocket_submit {
+	/** Input: Pointer to an array of struct drm_rocket_job. */
+	__u64 jobs;
+
+	/** Input: Number of jobs passed in. */
+	__u32 job_count;
+
+	/** Input: Size in bytes of the structs in the @jobs field. */
+	__u32 job_struct_size;
+
+	/** Reserved, must be zero. */
+	__u64 reserved;
+};
+
 #if defined(__cplusplus)
 }
 #endif

From 525ad89dd90434d529b76a87b1c653a69fedc416 Mon Sep 17 00:00:00 2001
From: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Date: Mon, 21 Jul 2025 11:17:32 +0200
Subject: [PATCH 0109/1455] accel/rocket: Add IOCTLs for synchronizing memory
 accesses

The NPU cores have their own access to the memory bus, and this isn't
cache coherent with the CPUs.

Add IOCTLs so userspace can mark when the caches need to be flushed, and
also when a writer job needs to be waited for before the buffer can be
accessed from the CPU.

Initially based on the same IOCTLs from the Etnaviv driver.

v2:
- Don't break UABI by reordering the IOCTL IDs (Jeff Hugo)

v3:
- Check that padding fields in IOCTLs are zero (Jeff Hugo)

v6:
- Fix conversion logic to make sure we use DMA_BIDIRECTIONAL when needed
  (Lucas Stach)

v8:
- Always sync BOs in both directions (Robin Murphy)

Reviewed-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721-6-10-rocket-v9-5-77ebd484941e@tomeuvizoso.net
---
 drivers/accel/rocket/rocket_drv.c |  2 ++
 drivers/accel/rocket/rocket_gem.c | 56 +++++++++++++++++++++++++++++++
 drivers/accel/rocket/rocket_gem.h |  4 +++
 include/uapi/drm/rocket_accel.h   | 34 +++++++++++++++++++
 4 files changed, 96 insertions(+)

diff --git a/drivers/accel/rocket/rocket_drv.c b/drivers/accel/rocket/rocket_drv.c
index a21aa9aa189b..5c0b63f0a8f0 100644
--- a/drivers/accel/rocket/rocket_drv.c
+++ b/drivers/accel/rocket/rocket_drv.c
@@ -134,6 +134,8 @@ static const struct drm_ioctl_desc rocket_drm_driver_ioctls[] = {
 
 	ROCKET_IOCTL(CREATE_BO, create_bo),
 	ROCKET_IOCTL(SUBMIT, submit),
+	ROCKET_IOCTL(PREP_BO, prep_bo),
+	ROCKET_IOCTL(FINI_BO, fini_bo),
 };
 
 DEFINE_DRM_ACCEL_FOPS(rocket_accel_driver_fops);
diff --git a/drivers/accel/rocket/rocket_gem.c b/drivers/accel/rocket/rocket_gem.c
index 05cf46040865..0551e11cc184 100644
--- a/drivers/accel/rocket/rocket_gem.c
+++ b/drivers/accel/rocket/rocket_gem.c
@@ -123,3 +123,59 @@ err:
 
 	return ret;
 }
+
+int rocket_ioctl_prep_bo(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_rocket_prep_bo *args = data;
+	unsigned long timeout = drm_timeout_abs_to_jiffies(args->timeout_ns);
+	struct drm_gem_object *gem_obj;
+	struct drm_gem_shmem_object *shmem_obj;
+	long ret = 0;
+
+	if (args->reserved != 0) {
+		drm_dbg(dev, "Reserved field in drm_rocket_prep_bo struct should be 0.\n");
+		return -EINVAL;
+	}
+
+	gem_obj = drm_gem_object_lookup(file, args->handle);
+	if (!gem_obj)
+		return -ENOENT;
+
+	ret = dma_resv_wait_timeout(gem_obj->resv, DMA_RESV_USAGE_WRITE, true, timeout);
+	if (!ret)
+		ret = timeout ? -ETIMEDOUT : -EBUSY;
+
+	shmem_obj = &to_rocket_bo(gem_obj)->base;
+
+	dma_sync_sgtable_for_cpu(dev->dev, shmem_obj->sgt, DMA_BIDIRECTIONAL);
+
+	drm_gem_object_put(gem_obj);
+
+	return ret;
+}
+
+int rocket_ioctl_fini_bo(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct drm_rocket_fini_bo *args = data;
+	struct drm_gem_shmem_object *shmem_obj;
+	struct rocket_gem_object *rkt_obj;
+	struct drm_gem_object *gem_obj;
+
+	if (args->reserved != 0) {
+		drm_dbg(dev, "Reserved field in drm_rocket_fini_bo struct should be 0.\n");
+		return -EINVAL;
+	}
+
+	gem_obj = drm_gem_object_lookup(file, args->handle);
+	if (!gem_obj)
+		return -ENOENT;
+
+	rkt_obj = to_rocket_bo(gem_obj);
+	shmem_obj = &rkt_obj->base;
+
+	dma_sync_sgtable_for_device(dev->dev, shmem_obj->sgt, DMA_BIDIRECTIONAL);
+
+	drm_gem_object_put(gem_obj);
+
+	return 0;
+}
diff --git a/drivers/accel/rocket/rocket_gem.h b/drivers/accel/rocket/rocket_gem.h
index 91a1fc09c56c..240430334509 100644
--- a/drivers/accel/rocket/rocket_gem.h
+++ b/drivers/accel/rocket/rocket_gem.h
@@ -21,6 +21,10 @@ struct drm_gem_object *rocket_gem_create_object(struct drm_device *dev, size_t s
 
 int rocket_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file *file);
 
+int rocket_ioctl_prep_bo(struct drm_device *dev, void *data, struct drm_file *file);
+
+int rocket_ioctl_fini_bo(struct drm_device *dev, void *data, struct drm_file *file);
+
 static inline
 struct  rocket_gem_object *to_rocket_bo(struct drm_gem_object *obj)
 {
diff --git a/include/uapi/drm/rocket_accel.h b/include/uapi/drm/rocket_accel.h
index 374f8370ac9d..14b2e12b7c49 100644
--- a/include/uapi/drm/rocket_accel.h
+++ b/include/uapi/drm/rocket_accel.h
@@ -13,9 +13,13 @@ extern "C" {
 
 #define DRM_ROCKET_CREATE_BO			0x00
 #define DRM_ROCKET_SUBMIT			0x01
+#define DRM_ROCKET_PREP_BO			0x02
+#define DRM_ROCKET_FINI_BO			0x03
 
 #define DRM_IOCTL_ROCKET_CREATE_BO		DRM_IOWR(DRM_COMMAND_BASE + DRM_ROCKET_CREATE_BO, struct drm_rocket_create_bo)
 #define DRM_IOCTL_ROCKET_SUBMIT			DRM_IOW(DRM_COMMAND_BASE + DRM_ROCKET_SUBMIT, struct drm_rocket_submit)
+#define DRM_IOCTL_ROCKET_PREP_BO		DRM_IOW(DRM_COMMAND_BASE + DRM_ROCKET_PREP_BO, struct drm_rocket_prep_bo)
+#define DRM_IOCTL_ROCKET_FINI_BO		DRM_IOW(DRM_COMMAND_BASE + DRM_ROCKET_FINI_BO, struct drm_rocket_fini_bo)
 
 /**
  * struct drm_rocket_create_bo - ioctl argument for creating Rocket BOs.
@@ -39,6 +43,36 @@ struct drm_rocket_create_bo {
 	__u64 offset;
 };
 
+/**
+ * struct drm_rocket_prep_bo - ioctl argument for starting CPU ownership of the BO.
+ *
+ * Takes care of waiting for any NPU jobs that might still use the NPU and performs cache
+ * synchronization.
+ */
+struct drm_rocket_prep_bo {
+	/** Input: GEM handle of the buffer object. */
+	__u32 handle;
+
+	/** Reserved, must be zero. */
+	__u32 reserved;
+
+	/** Input: Amount of time to wait for NPU jobs. */
+	__s64 timeout_ns;
+};
+
+/**
+ * struct drm_rocket_fini_bo - ioctl argument for finishing CPU ownership of the BO.
+ *
+ * Synchronize caches for NPU access.
+ */
+struct drm_rocket_fini_bo {
+	/** Input: GEM handle of the buffer object. */
+	__u32 handle;
+
+	/** Reserved, must be zero. */
+	__u32 reserved;
+};
+
 /**
  * struct drm_rocket_task - A task to be run on the NPU
  *

From a7352c849492a30b5d8491fcb9314ab376a3942f Mon Sep 17 00:00:00 2001
From: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Date: Mon, 21 Jul 2025 11:17:33 +0200
Subject: [PATCH 0110/1455] dt-bindings: npu: rockchip,rknn: Add bindings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add the bindings for the Neural Processing Unit IP from Rockchip.

v2:
- Adapt to new node structure (one node per core, each with its own
  IOMMU)
- Several misc. fixes from Sebastian Reichel

v3:
- Split register block in its constituent subblocks, and only require
  the ones that the kernel would ever use (Nicolas Frattaroli)
- Group supplies (Rob Herring)
- Explain the way in which the top core is special (Rob Herring)

v4:
- Change required node name to npu@ (Rob Herring and Krzysztof Kozlowski)
- Remove unneeded items: (Krzysztof Kozlowski)
- Fix use of minItems/maxItems (Krzysztof Kozlowski)
- Add reg-names to list of required properties (Krzysztof Kozlowski)
- Fix example (Krzysztof Kozlowski)

v5:
- Rename file to rockchip,rk3588-rknn-core.yaml (Krzysztof Kozlowski)
- Streamline compatible property (Krzysztof Kozlowski)

v6:
- Remove mention to NVDLA, as the hardware is only incidentally related
  (Kever Yang)
- Mark pclk and npu clocks as required by all clocks (Rob Herring)

v7:
- Remove allOf section, not needed now that all nodes require 4 clocks
  (Heiko Stübner)

v8:
- Remove notion of top core (Robin Murphy)

Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Tested-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Jeff Hugo <jeff.hugo@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250721-6-10-rocket-v9-6-77ebd484941e@tomeuvizoso.net
---
 .../npu/rockchip,rk3588-rknn-core.yaml        | 112 ++++++++++++++++++
 1 file changed, 112 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/npu/rockchip,rk3588-rknn-core.yaml

diff --git a/Documentation/devicetree/bindings/npu/rockchip,rk3588-rknn-core.yaml b/Documentation/devicetree/bindings/npu/rockchip,rk3588-rknn-core.yaml
new file mode 100644
index 000000000000..caca2a4903cd
--- /dev/null
+++ b/Documentation/devicetree/bindings/npu/rockchip,rk3588-rknn-core.yaml
@@ -0,0 +1,112 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/npu/rockchip,rk3588-rknn-core.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Neural Processing Unit IP from Rockchip
+
+maintainers:
+  - Tomeu Vizoso <tomeu@tomeuvizoso.net>
+
+description:
+  Rockchip IP for accelerating inference of neural networks.
+
+  There is to be a node per each NPU core in the SoC, and each core should reference all the
+  resources that it needs to function, such as clocks, power domains, and resets.
+
+properties:
+  $nodename:
+    pattern: '^npu@[a-f0-9]+$'
+
+  compatible:
+    enum:
+      - rockchip,rk3588-rknn-core
+
+  reg:
+    maxItems: 3
+
+  reg-names:
+    items:
+      - const: pc # Program Control-related registers
+      - const: cna # Convolution Neural Network Accelerator registers
+      - const: core # Main NPU core processing unit registers
+
+  clocks:
+    maxItems: 4
+
+  clock-names:
+    items:
+      - const: aclk
+      - const: hclk
+      - const: npu
+      - const: pclk
+
+  interrupts:
+    maxItems: 1
+
+  iommus:
+    maxItems: 1
+
+  npu-supply: true
+
+  power-domains:
+    maxItems: 1
+
+  resets:
+    maxItems: 2
+
+  reset-names:
+    items:
+      - const: srst_a
+      - const: srst_h
+
+  sram-supply: true
+
+required:
+  - compatible
+  - reg
+  - reg-names
+  - clocks
+  - clock-names
+  - interrupts
+  - iommus
+  - power-domains
+  - resets
+  - reset-names
+  - npu-supply
+  - sram-supply
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/rockchip,rk3588-cru.h>
+    #include <dt-bindings/interrupt-controller/irq.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/power/rk3588-power.h>
+    #include <dt-bindings/reset/rockchip,rk3588-cru.h>
+
+    bus {
+      #address-cells = <2>;
+      #size-cells = <2>;
+
+      npu@fdab0000 {
+        compatible = "rockchip,rk3588-rknn-core";
+        reg = <0x0 0xfdab0000 0x0 0x1000>,
+              <0x0 0xfdab1000 0x0 0x1000>,
+              <0x0 0xfdab3000 0x0 0x1000>;
+        reg-names = "pc", "cna", "core";
+        clocks = <&cru ACLK_NPU0>, <&cru HCLK_NPU0>,
+                 <&scmi_clk SCMI_CLK_NPU>, <&cru PCLK_NPU_ROOT>;
+        clock-names = "aclk", "hclk", "npu", "pclk";
+        interrupts = <GIC_SPI 110 IRQ_TYPE_LEVEL_HIGH 0>;
+        iommus = <&rknn_mmu_0>;
+        npu-supply = <&vdd_npu_s0>;
+        power-domains = <&power RK3588_PD_NPUTOP>;
+        resets = <&cru SRST_A_RKNN0>, <&cru SRST_H_RKNN0>;
+        reset-names = "srst_a", "srst_h";
+        sram-supply = <&vdd_npu_mem_s0>;
+      };
+    };
+...

From 1ffcf8b8ae8a786b61350ee082b6c82846a31e53 Mon Sep 17 00:00:00 2001
From: Ilia Levi <ilia.levi@intel.com>
Date: Mon, 14 Jul 2025 15:26:58 +0300
Subject: [PATCH 0111/1455] drm/xe: Support for mmap-ing mmio regions

Allow the driver to expose hardware register spaces to userspace
through GEM objects with fake mmap offsets. This can be useful
for userspace-firmware communication, debugging, etc.

v2: Minor doc fix (CI)
v3: Enforce MAP_SHARED (Tejas)
    Add fault handler with dummy page (Tejas, Matt Auld)
    Store physical address instead of xe_mmio in the GEM object (MattB)
v4: Separate xe_mmio_gem from xe_mmio and make it private (MattB)

Signed-off-by: Ilia Levi <ilia.levi@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://lore.kernel.org/r/20250714122658.1803-1-ilia.levi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/Makefile      |   1 +
 drivers/gpu/drm/xe/xe_mmio_gem.c | 226 +++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_mmio_gem.h |  20 +++
 3 files changed, 247 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_mmio_gem.c
 create mode 100644 drivers/gpu/drm/xe/xe_mmio_gem.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 7eaf4054c99e..42c6ca5b3f76 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -87,6 +87,7 @@ xe-y += xe_bb.o \
 	xe_lrc.o \
 	xe_migrate.o \
 	xe_mmio.o \
+	xe_mmio_gem.o \
 	xe_mocs.o \
 	xe_module.o \
 	xe_nvm.o \
diff --git a/drivers/gpu/drm/xe/xe_mmio_gem.c b/drivers/gpu/drm/xe/xe_mmio_gem.c
new file mode 100644
index 000000000000..9a97c4387e4f
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_mmio_gem.c
@@ -0,0 +1,226 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "xe_mmio_gem.h"
+
+#include <drm/drm_drv.h>
+#include <drm/drm_gem.h>
+#include <drm/drm_managed.h>
+
+#include "xe_device_types.h"
+
+/**
+ * DOC: Exposing MMIO regions to userspace
+ *
+ * In certain cases, the driver may allow userspace to mmap a portion of the hardware registers.
+ *
+ * This can be done as follows:
+ * 1. Call xe_mmio_gem_create() to create a GEM object with an mmap-able fake offset.
+ * 2. Use xe_mmio_gem_mmap_offset() on the created GEM object to retrieve the fake offset.
+ * 3. Provide the fake offset to userspace.
+ * 4. Userspace can call mmap with the fake offset. The length provided to mmap
+ *    must match the size of the GEM object.
+ * 5. When the region is no longer needed, call xe_mmio_gem_destroy() to release the GEM object.
+ *
+ * NOTE: The exposed MMIO region must be page-aligned with regards to its BAR offset and size.
+ *
+ * WARNING: Exposing MMIO regions to userspace can have security and stability implications.
+ * Make sure not to expose any sensitive registers.
+ */
+
+static void xe_mmio_gem_free(struct drm_gem_object *);
+static int xe_mmio_gem_mmap(struct drm_gem_object *, struct vm_area_struct *);
+static vm_fault_t xe_mmio_gem_vm_fault(struct vm_fault *);
+
+struct xe_mmio_gem {
+	struct drm_gem_object base;
+	phys_addr_t phys_addr;
+};
+
+static const struct vm_operations_struct vm_ops = {
+	.open = drm_gem_vm_open,
+	.close = drm_gem_vm_close,
+	.fault = xe_mmio_gem_vm_fault,
+};
+
+static const struct drm_gem_object_funcs xe_mmio_gem_funcs = {
+	.free = xe_mmio_gem_free,
+	.mmap = xe_mmio_gem_mmap,
+	.vm_ops = &vm_ops,
+};
+
+static inline struct xe_mmio_gem *to_xe_mmio_gem(struct drm_gem_object *obj)
+{
+	return container_of(obj, struct xe_mmio_gem, base);
+}
+
+/**
+ * xe_mmio_gem_create - Expose an MMIO region to userspace
+ * @xe: The xe device
+ * @file: DRM file descriptor
+ * @phys_addr: Start of the exposed MMIO region
+ * @size: The size of the exposed MMIO region
+ *
+ * This function creates a GEM object that exposes an MMIO region with an mmap-able
+ * fake offset.
+ *
+ * See: "Exposing MMIO regions to userspace"
+ */
+struct xe_mmio_gem *xe_mmio_gem_create(struct xe_device *xe, struct drm_file *file,
+				       phys_addr_t phys_addr, size_t size)
+{
+	struct xe_mmio_gem *obj;
+	struct drm_gem_object *base;
+	int err;
+
+	if ((phys_addr % PAGE_SIZE != 0) || (size % PAGE_SIZE != 0))
+		return ERR_PTR(-EINVAL);
+
+	obj = kzalloc(sizeof(*obj), GFP_KERNEL);
+	if (!obj)
+		return ERR_PTR(-ENOMEM);
+
+	base = &obj->base;
+	base->funcs = &xe_mmio_gem_funcs;
+	obj->phys_addr = phys_addr;
+
+	drm_gem_private_object_init(&xe->drm, base, size);
+
+	err = drm_gem_create_mmap_offset(base);
+	if (err)
+		goto free_gem;
+
+	err = drm_vma_node_allow(&base->vma_node, file);
+	if (err)
+		goto free_gem;
+
+	return obj;
+
+free_gem:
+	xe_mmio_gem_free(base);
+	return ERR_PTR(err);
+}
+
+/**
+ * xe_mmio_gem_mmap_offset - Return the mmap-able fake offset
+ * @gem: the GEM object created with xe_mmio_gem_create()
+ *
+ * This function returns the mmap-able fake offset allocated during
+ * xe_mmio_gem_create().
+ *
+ * See: "Exposing MMIO regions to userspace"
+ */
+u64 xe_mmio_gem_mmap_offset(struct xe_mmio_gem *gem)
+{
+	return drm_vma_node_offset_addr(&gem->base.vma_node);
+}
+
+static void xe_mmio_gem_free(struct drm_gem_object *base)
+{
+	struct xe_mmio_gem *obj = to_xe_mmio_gem(base);
+
+	drm_gem_object_release(base);
+	kfree(obj);
+}
+
+/**
+ * xe_mmio_gem_destroy - Destroy the GEM object that exposes an MMIO region
+ * @gem: the GEM object to destroy
+ *
+ * This function releases resources associated with the GEM object created by
+ * xe_mmio_gem_create().
+ *
+ * See: "Exposing MMIO regions to userspace"
+ */
+void xe_mmio_gem_destroy(struct xe_mmio_gem *gem)
+{
+	xe_mmio_gem_free(&gem->base);
+}
+
+static int xe_mmio_gem_mmap(struct drm_gem_object *base, struct vm_area_struct *vma)
+{
+	if (vma->vm_end - vma->vm_start != base->size)
+		return -EINVAL;
+
+	if ((vma->vm_flags & VM_SHARED) == 0)
+		return -EINVAL;
+
+	/* Set vm_pgoff (used as a fake buffer offset by DRM) to 0 */
+	vma->vm_pgoff = 0;
+	vma->vm_page_prot = pgprot_noncached(vm_get_page_prot(vma->vm_flags));
+	vm_flags_set(vma, VM_IO | VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP |
+		     VM_DONTCOPY | VM_NORESERVE);
+
+	/* Defer actual mapping to the fault handler. */
+	return 0;
+}
+
+static void xe_mmio_gem_release_dummy_page(struct drm_device *dev, void *res)
+{
+	__free_page((struct page *)res);
+}
+
+static vm_fault_t xe_mmio_gem_vm_fault_dummy_page(struct vm_area_struct *vma)
+{
+	struct drm_gem_object *base = vma->vm_private_data;
+	struct drm_device *dev = base->dev;
+	vm_fault_t ret = VM_FAULT_NOPAGE;
+	struct page *page;
+	unsigned long pfn;
+	unsigned long i;
+
+	page = alloc_page(GFP_KERNEL | __GFP_ZERO);
+	if (!page)
+		return VM_FAULT_OOM;
+
+	if (drmm_add_action_or_reset(dev, xe_mmio_gem_release_dummy_page, page))
+		return VM_FAULT_OOM;
+
+	pfn = page_to_pfn(page);
+
+	/* Map the entire VMA to the same dummy page */
+	for (i = 0; i < base->size; i += PAGE_SIZE) {
+		unsigned long addr = vma->vm_start + i;
+
+		ret = vmf_insert_pfn(vma, addr, pfn);
+		if (ret & VM_FAULT_ERROR)
+			break;
+	}
+
+	return ret;
+}
+
+static vm_fault_t xe_mmio_gem_vm_fault(struct vm_fault *vmf)
+{
+	struct vm_area_struct *vma = vmf->vma;
+	struct drm_gem_object *base = vma->vm_private_data;
+	struct xe_mmio_gem *obj = to_xe_mmio_gem(base);
+	struct drm_device *dev = base->dev;
+	vm_fault_t ret = VM_FAULT_NOPAGE;
+	unsigned long i;
+	int idx;
+
+	if (!drm_dev_enter(dev, &idx)) {
+		/*
+		 * Provide a dummy page to avoid SIGBUS for events such as hot-unplug.
+		 * This gives the userspace the option to recover instead of crashing.
+		 * It is assumed the userspace will receive the notification via some
+		 * other channel (e.g. drm uevent).
+		 */
+		return xe_mmio_gem_vm_fault_dummy_page(vma);
+	}
+
+	for (i = 0; i < base->size; i += PAGE_SIZE) {
+		unsigned long addr = vma->vm_start + i;
+		unsigned long phys_addr = obj->phys_addr + i;
+
+		ret = vmf_insert_pfn(vma, addr, PHYS_PFN(phys_addr));
+		if (ret & VM_FAULT_ERROR)
+			break;
+	}
+
+	drm_dev_exit(idx);
+	return ret;
+}
diff --git a/drivers/gpu/drm/xe/xe_mmio_gem.h b/drivers/gpu/drm/xe/xe_mmio_gem.h
new file mode 100644
index 000000000000..4b76d5586ebb
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_mmio_gem.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_MMIO_GEM_H_
+#define _XE_MMIO_GEM_H_
+
+#include <linux/types.h>
+
+struct drm_file;
+struct xe_device;
+struct xe_mmio_gem;
+
+struct xe_mmio_gem *xe_mmio_gem_create(struct xe_device *xe, struct drm_file *file,
+				       phys_addr_t phys_addr, size_t size);
+u64 xe_mmio_gem_mmap_offset(struct xe_mmio_gem *gem);
+void xe_mmio_gem_destroy(struct xe_mmio_gem *gem);
+
+#endif /* _XE_MMIO_GEM_H_ */

From 6c57634d734ac8b8f6fdadb58964ea2b010e5307 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Thu, 24 Jul 2025 15:59:54 +0530
Subject: [PATCH 0112/1455] drm/i915/vblank: Change log from err to debug

Let Potential update error just be a log instead of a big error
we already have Atomic Update error log which shouts out if
something really goes wrong.

--v2
-Fix typo in commit message [Mitul]

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Mitul Golani <mitulkumar.ajitkumar.golani@intel.com>
Link: https://lore.kernel.org/r/20250724102954.2573715-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_vblank.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c
index 70ba7aa26bf4..3e51deca0c21 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.c
+++ b/drivers/gpu/drm/i915/display/intel_vblank.c
@@ -724,9 +724,9 @@ int intel_vblank_evade(struct intel_vblank_evade_ctx *evade)
 			break;
 
 		if (!timeout) {
-			drm_err(display->drm,
-				"Potential atomic update failure on pipe %c\n",
-				pipe_name(crtc->pipe));
+			drm_dbg_kms(display->drm,
+				    "Potential atomic update failure on pipe %c\n",
+				    pipe_name(crtc->pipe));
 			break;
 		}
 

From 33f8f321e7aa7715ce19560801ee5223ba8b9a7d Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Mon, 23 Jun 2025 23:35:26 +0100
Subject: [PATCH 0113/1455] drm/vmwgfx: fix missing assignment to ts

The assignment to ts is missing on the call to ktime_to_timespec64.
Fix this by adding the missing assignment.

Fixes: db6a94b26354 ("drm/vmwgfx: Implement dma_fence_ops properly")
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Ian Forbes <ian.forbes@broadcom.com>
Signed-off-by: Zack Rusin <zack.rusin@broadcom.com>
Link: https://lore.kernel.org/r/20250623223526.281398-1-colin.i.king@gmail.com
---
 drivers/gpu/drm/vmwgfx/vmwgfx_fence.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
index c2294abbe753..00be92da5509 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c
@@ -538,7 +538,7 @@ static void vmw_event_fence_action_seq_passed(struct dma_fence *f,
 	if (likely(eaction->tv_sec != NULL)) {
 		struct timespec64 ts;
 
-		ktime_to_timespec64(f->timestamp);
+		ts = ktime_to_timespec64(f->timestamp);
 		/* monotonic time, so no y2038 overflow */
 		*eaction->tv_sec = ts.tv_sec;
 		*eaction->tv_usec = ts.tv_nsec / NSEC_PER_USEC;

From 3f92c7b329ef3685af8c0c19dba7566d7414fd77 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Mon, 7 Jul 2025 12:38:00 -0400
Subject: [PATCH 0114/1455] drm/amd/display: [FW Promotion] Release 0.1.18.0

Add new mode in struct ips_residency_mode

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index c587b3441e07..686de58a0412 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -882,7 +882,7 @@ enum dmub_shared_state_feature_id {
 /**
  * struct dmub_shared_state_ips_fw - Firmware signals for IPS.
  */
- union dmub_shared_state_ips_fw_signals {
+union dmub_shared_state_ips_fw_signals {
 	struct {
 		uint32_t ips1_commit : 1;  /**< 1 if in IPS1 or IPS0 RCG */
 		uint32_t ips2_commit : 1; /**< 1 if in IPS2 */
@@ -897,7 +897,7 @@ enum dmub_shared_state_feature_id {
 /**
  * struct dmub_shared_state_ips_signals - Firmware signals for IPS.
  */
- union dmub_shared_state_ips_driver_signals {
+union dmub_shared_state_ips_driver_signals {
 	struct {
 		uint32_t allow_pg : 1; /**< 1 if PG is allowed */
 		uint32_t allow_ips1 : 1; /**< 1 is IPS1 is allowed */
@@ -4099,7 +4099,6 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * Use for AUX-less ALPM LFPS wake operation
 	 */
 	struct dmub_alpm_auxless_data auxless_alpm_data;
-
 	/**
 	 * @pad: Align structure to 4 byte boundary.
 	 */
@@ -5914,6 +5913,7 @@ enum ips_residency_mode {
 	IPS_RESIDENCY__IPS2,
 	IPS_RESIDENCY__IPS1_RCG,
 	IPS_RESIDENCY__IPS1_ONO2_ON,
+	IPS_RESIDENCY__IPS1_Z8_RETENTION,
 };
 
 #define NUM_IPS_HISTOGRAM_BUCKETS 16

From eff8d30ac842fb118abc425c262424d75a658f95 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Wed, 9 Jul 2025 14:31:41 -0500
Subject: [PATCH 0115/1455] drm/amd/display: Promote DAL to 3.2.341

This version brings along following fixes:
- Fixes for DCN401, DCN32
- Initial support of SmartMux
- Improvements for Replay, IPS, and IPS2
- Refactor of DSC

Reviewed-by: Martin Leung <martin.leung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 59c07756130d..d0c78f94a7a3 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.340"
+#define DC_VER "3.2.341"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 19f76f2390be5abe8d5ed986780b73564ba2baca Mon Sep 17 00:00:00 2001
From: Ostrowski Rafal <rostrows@amd.com>
Date: Tue, 24 Jun 2025 14:13:53 +0200
Subject: [PATCH 0116/1455] drm/amd/display: Update tiled to tiled copy command

[Why & How]
Tiled command rect dimensions is 1 based, do rect_x/y - 1 internally

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Ostrowski Rafal <rostrows@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index f5ef1a07078e..714c468c010d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -2072,8 +2072,8 @@ bool dmub_lsdma_send_tiled_to_tiled_copy_command(
 	lsdma_data->u.tiled_copy_data.dst_swizzle_mode = params.swizzle_mode;
 	lsdma_data->u.tiled_copy_data.src_element_size = params.element_size;
 	lsdma_data->u.tiled_copy_data.dst_element_size = params.element_size;
-	lsdma_data->u.tiled_copy_data.rect_x           = params.rect_x;
-	lsdma_data->u.tiled_copy_data.rect_y           = params.rect_y;
+	lsdma_data->u.tiled_copy_data.rect_x           = params.rect_x - 1;
+	lsdma_data->u.tiled_copy_data.rect_y           = params.rect_y - 1;
 	lsdma_data->u.tiled_copy_data.dcc              = params.dcc;
 	lsdma_data->u.tiled_copy_data.tmz              = params.tmz;
 	lsdma_data->u.tiled_copy_data.read_compress    = params.read_compress;

From 1a6a3374ecb9899ccf0d209b5783a796bdba8cec Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Wed, 25 Jun 2025 10:11:22 -0400
Subject: [PATCH 0117/1455] drm/amd/display: fix condition for setting
 timing_adjust_pending

timing_adjust_pending is used to defer certain programming sequences
when OTG timing is about to be changed, like with VRR. Insufficient
checking for timing change in this case caused a regression which
reduces PSR Replay residency.

Reviewed-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Robin Chen <robin.chen@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 28aca7017f0f..60a78c53addd 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -446,7 +446,9 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc,
 	 * avoid conflicting with firmware updates.
 	 */
 	if (dc->ctx->dce_version > DCE_VERSION_MAX) {
-		if (dc->optimized_required || dc->wm_optimized_required) {
+		if ((dc->optimized_required || dc->wm_optimized_required) &&
+			(stream->adjust.v_total_max != adjust->v_total_max ||
+			stream->adjust.v_total_min != adjust->v_total_min)) {
 			stream->adjust.timing_adjust_pending = true;
 			return false;
 		}

From 92f68f6a1b297633159a3f3759e4dfc7e5b58abb Mon Sep 17 00:00:00 2001
From: Gabe Teeger <gabe.teeger@amd.com>
Date: Wed, 9 Jul 2025 14:12:22 -0400
Subject: [PATCH 0118/1455] drm/amd/display: Revert Add HPO encoder support to
 Replay

This reverts commits:
commit 1f26214d268b ("drm/amd/display: Add HPO encoder support to Replay")
commit 3bfce48b109f ("drm/amd/display: Add support for Panel Replay on DP1 eDP (panel_inst=1)")
due to visual confirm issue.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Gabe Teeger <gabe.teeger@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dce/dmub_replay.c  | 43 ++-----------------
 .../gpu/drm/amd/display/dc/dce/dmub_replay.h  |  2 +-
 .../link/protocols/link_edp_panel_control.c   |  2 +-
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 20 ---------
 4 files changed, 5 insertions(+), 62 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
index e7a318e26d38..fcd3d86ad517 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
@@ -4,7 +4,6 @@
 
 #include "dc.h"
 #include "dc_dmub_srv.h"
-#include "dc_dp_types.h"
 #include "dmub/dmub_srv.h"
 #include "core_types.h"
 #include "dmub_replay.h"
@@ -44,45 +43,21 @@ static void dmub_replay_get_state(struct dmub_replay *dmub, enum replay_state *s
 /*
  * Enable/Disable Replay.
  */
-static void dmub_replay_enable(struct dmub_replay *dmub, bool enable, bool wait, uint8_t panel_inst,
-			       struct dc_link *link)
+static void dmub_replay_enable(struct dmub_replay *dmub, bool enable, bool wait, uint8_t panel_inst)
 {
 	union dmub_rb_cmd cmd;
 	struct dc_context *dc = dmub->ctx;
 	uint32_t retry_count;
 	enum replay_state state = REPLAY_STATE_0;
-	struct pipe_ctx *pipe_ctx = NULL;
-	struct resource_context *res_ctx = &link->ctx->dc->current_state->res_ctx;
-	uint8_t i;
 
 	memset(&cmd, 0, sizeof(cmd));
 	cmd.replay_enable.header.type = DMUB_CMD__REPLAY;
 	cmd.replay_enable.data.panel_inst = panel_inst;
 
 	cmd.replay_enable.header.sub_type = DMUB_CMD__REPLAY_ENABLE;
-	if (enable) {
+	if (enable)
 		cmd.replay_enable.data.enable = REPLAY_ENABLE;
-		// hpo stream/link encoder assignments are not static, need to update everytime we try to enable replay
-		if (link->cur_link_settings.link_rate >= LINK_RATE_UHBR10) {
-			for (i = 0; i < MAX_PIPES; i++) {
-				if (res_ctx &&
-					res_ctx->pipe_ctx[i].stream &&
-					res_ctx->pipe_ctx[i].stream->link &&
-					res_ctx->pipe_ctx[i].stream->link == link &&
-					res_ctx->pipe_ctx[i].stream->link->connector_signal == SIGNAL_TYPE_EDP) {
-					pipe_ctx = &res_ctx->pipe_ctx[i];
-					//TODO: refactor for multi edp support
-					break;
-				}
-			}
-
-			if (!pipe_ctx)
-				return;
-
-			cmd.replay_enable.data.hpo_stream_enc_inst = pipe_ctx->stream_res.hpo_dp_stream_enc->inst;
-			cmd.replay_enable.data.hpo_link_enc_inst = pipe_ctx->link_res.hpo_dp_link_enc->inst;
-		}
-	} else
+	else
 		cmd.replay_enable.data.enable = REPLAY_DISABLE;
 
 	cmd.replay_enable.header.payload_bytes = sizeof(struct dmub_rb_cmd_replay_enable_data);
@@ -174,17 +149,6 @@ static bool dmub_replay_copy_settings(struct dmub_replay *dmub,
 	copy_settings_data->digbe_inst				= replay_context->digbe_inst;
 	copy_settings_data->digfe_inst				= replay_context->digfe_inst;
 
-	if (link->cur_link_settings.link_rate >= LINK_RATE_UHBR10) {
-		if (pipe_ctx->stream_res.hpo_dp_stream_enc)
-			copy_settings_data->hpo_stream_enc_inst = pipe_ctx->stream_res.hpo_dp_stream_enc->inst;
-		else
-			copy_settings_data->hpo_stream_enc_inst = 0;
-		if (pipe_ctx->link_res.hpo_dp_link_enc)
-			copy_settings_data->hpo_link_enc_inst = pipe_ctx->link_res.hpo_dp_link_enc->inst;
-		else
-			copy_settings_data->hpo_link_enc_inst = 0;
-	}
-
 	if (pipe_ctx->plane_res.dpp)
 		copy_settings_data->dpp_inst			= pipe_ctx->plane_res.dpp->inst;
 	else
@@ -247,7 +211,6 @@ static void dmub_replay_set_coasting_vtotal(struct dmub_replay *dmub,
 	pCmd->header.type = DMUB_CMD__REPLAY;
 	pCmd->header.sub_type = DMUB_CMD__REPLAY_SET_COASTING_VTOTAL;
 	pCmd->header.payload_bytes = sizeof(struct dmub_cmd_replay_set_coasting_vtotal_data);
-	pCmd->replay_set_coasting_vtotal_data.panel_inst = panel_inst;
 	pCmd->replay_set_coasting_vtotal_data.coasting_vtotal = (coasting_vtotal & 0xFFFF);
 	pCmd->replay_set_coasting_vtotal_data.coasting_vtotal_high = (coasting_vtotal & 0xFFFF0000) >> 16;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h
index ccbe385e132c..e6346c0ffc0e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.h
@@ -19,7 +19,7 @@ struct dmub_replay_funcs {
 	void (*replay_get_state)(struct dmub_replay *dmub, enum replay_state *state,
 		uint8_t panel_inst);
 	void (*replay_enable)(struct dmub_replay *dmub, bool enable, bool wait,
-		uint8_t panel_inst, struct dc_link *link);
+		uint8_t panel_inst);
 	bool (*replay_copy_settings)(struct dmub_replay *dmub, struct dc_link *link,
 		struct replay_context *replay_context, uint8_t panel_inst);
 	void (*replay_set_power_opt)(struct dmub_replay *dmub, unsigned int power_opt,
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index e7927b8f5ba3..98ec9b5a559c 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -944,7 +944,7 @@ bool edp_set_replay_allow_active(struct dc_link *link, const bool *allow_active,
 		// TODO: Handle mux change case if force_static is set
 		// If force_static is set, just change the replay_allow_active state directly
 		if (replay != NULL && link->replay_settings.replay_feature_enabled)
-			replay->funcs->replay_enable(replay, *allow_active, wait, panel_inst, link);
+			replay->funcs->replay_enable(replay, *allow_active, wait, panel_inst);
 		link->replay_settings.replay_allow_active = *allow_active;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 686de58a0412..1fb1472737ad 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -4047,14 +4047,6 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * DIG BE HW instance.
 	 */
 	uint8_t digbe_inst;
-	/**
-	 * @hpo_stream_enc_inst: HPO stream encoder instance
-	 */
-	uint8_t hpo_stream_enc_inst;
-	/**
-	 * @hpo_link_enc_inst: HPO link encoder instance
-	 */
-	uint8_t hpo_link_enc_inst;
 	/**
 	 * AUX HW instance.
 	 */
@@ -4158,18 +4150,6 @@ struct dmub_rb_cmd_replay_enable_data {
 	 * This does not support HDMI/DP2 for now.
 	 */
 	uint8_t phy_rate;
-	/**
-	 * @hpo_stream_enc_inst: HPO stream encoder instance
-	 */
-	uint8_t hpo_stream_enc_inst;
-	/**
-	 * @hpo_link_enc_inst: HPO link encoder instance
-	 */
-	uint8_t hpo_link_enc_inst;
-	/**
-	 * @pad: Align structure to 4 byte boundary.
-	 */
-	uint8_t pad[2];
 };
 
 /**

From ca74cc428f2b9d0170c56b473dbcfd7fa01daf2d Mon Sep 17 00:00:00 2001
From: Clay King <clayking@amd.com>
Date: Mon, 7 Jul 2025 13:21:30 -0400
Subject: [PATCH 0119/1455] drm/amd/display: ensure committing streams is
 seamless

[Why]
When transitioning between topologies such as multi-display to single
display ODM 2:1, pipes might not be freed before use.

[How]
In dc_commit_streams, commit an additional, minimal transition if
original transition is not seamless to ensure pipes are freed.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 60a78c53addd..f760e930ea1b 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -2399,6 +2399,18 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_commit_streams_params
 		goto fail;
 	}
 
+	/*
+	 * If not already seamless, make transition seamless by inserting intermediate minimal transition
+	 */
+	if (dc->hwss.is_pipe_topology_transition_seamless &&
+			!dc->hwss.is_pipe_topology_transition_seamless(dc, dc->current_state, context)) {
+		res = commit_minimal_transition_state(dc, context);
+		if (res != DC_OK) {
+			BREAK_TO_DEBUGGER();
+			goto fail;
+		}
+	}
+
 	res = dc_commit_state_no_check(dc, context);
 
 	for (i = 0; i < params->stream_count; i++) {

From 34c9cd82dc62641b9038815ebc1a397f24d5a6e2 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 25 Jun 2025 10:23:13 -0500
Subject: [PATCH 0120/1455] drm/amd/display: Drop unnecessary 'rc' variable in
 amdgpu_dm_backlight_get_level()

[Why]
amdgpu_dm_backlight_get_level() returns a boolean value but is assigned
to a variable named 'rc' which is generally used for return codes.

This can be confusing while looking at the code for other issues.

[How]
Drop the variable and just look directly at the return value of
amdgpu_dm_backlight_get_level() in the if statement.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 2a175fc0399c..bfe3ee4b07ff 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -4926,10 +4926,8 @@ static u32 amdgpu_dm_backlight_get_level(struct amdgpu_display_manager *dm,
 
 	if (caps.aux_support) {
 		u32 avg, peak;
-		bool rc;
 
-		rc = dc_link_get_backlight_level_nits(link, &avg, &peak);
-		if (!rc)
+		if (!dc_link_get_backlight_level_nits(link, &avg, &peak))
 			return dm->brightness[bl_idx];
 		return convert_brightness_to_user(&caps, avg);
 	}

From 9dd3146ff7c126f03b784116b5c3581977f3acd1 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Sun, 13 Jul 2025 11:28:43 -0400
Subject: [PATCH 0121/1455] drm/amd/display: [FW Promotion] Release 0.1.19.0

Update DMUB related command structure.

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 105 ++++++++++++++----
 1 file changed, 81 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 1fb1472737ad..a89bf08ffd37 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -1990,18 +1990,19 @@ struct dmub_cmd_lsdma_data {
 		struct lsdma_tiled_copy_data {
 			uint32_t src_addr_lo;
 			uint32_t src_addr_hi;
+
 			uint32_t dst_addr_lo;
 			uint32_t dst_addr_hi;
 
 			uint32_t src_x            : 16;
 			uint32_t src_y            : 16;
 
-			uint32_t src_width        : 16;
-			uint32_t src_height       : 16;
-
 			uint32_t dst_x            : 16;
 			uint32_t dst_y            : 16;
 
+			uint32_t src_width        : 16;
+			uint32_t src_height       : 16;
+
 			uint32_t dst_width        : 16;
 			uint32_t dst_height       : 16;
 
@@ -2034,41 +2035,58 @@ struct dmub_cmd_lsdma_data {
 			uint32_t padding          : 30;
 		} tiled_copy_data;
 		struct lsdma_linear_copy_data {
+			uint32_t src_lo;
+			uint32_t src_hi;
+
+			uint32_t dst_lo;
+			uint32_t dst_hi;
+
 			uint32_t count            : 30;
 			uint32_t cache_policy_dst : 2;
 
 			uint32_t tmz              : 1;
 			uint32_t cache_policy_src : 2;
 			uint32_t padding          : 29;
-
+		} linear_copy_data;
+		struct lsdma_linear_sub_window_copy_data {
 			uint32_t src_lo;
 			uint32_t src_hi;
+
 			uint32_t dst_lo;
 			uint32_t dst_hi;
-		} linear_copy_data;
+
+			uint32_t src_x        : 16;
+			uint32_t src_y        : 16;
+
+			uint32_t dst_x        : 16;
+			uint32_t dst_y        : 16;
+
+			uint32_t rect_x       : 16;
+			uint32_t rect_y       : 16;
+
+			uint32_t src_pitch    : 16;
+			uint32_t dst_pitch    : 16;
+
+			uint32_t src_slice_pitch;
+			uint32_t dst_slice_pitch;
+
+			uint32_t tmz              : 1;
+			uint32_t element_size     : 3;
+			uint32_t src_cache_policy : 3;
+			uint32_t dst_cache_policy : 3;
+			uint32_t reserved0        : 22;
+		} linear_sub_window_copy_data;
 		struct lsdma_reg_write_data {
 			uint32_t reg_addr;
 			uint32_t reg_data;
 		} reg_write_data;
 		struct lsdma_pio_copy_data {
-			union {
-				struct {
-					uint32_t byte_count      : 26;
-					uint32_t src_loc         : 1;
-					uint32_t dst_loc         : 1;
-					uint32_t src_addr_inc    : 1;
-					uint32_t dst_addr_inc    : 1;
-					uint32_t overlap_disable : 1;
-					uint32_t constant_fill   : 1;
-				} fields;
-				uint32_t raw;
-			} packet;
 			uint32_t src_lo;
 			uint32_t src_hi;
+
 			uint32_t dst_lo;
 			uint32_t dst_hi;
-		} pio_copy_data;
-		struct lsdma_pio_constfill_data {
+
 			union {
 				struct {
 					uint32_t byte_count      : 26;
@@ -2081,14 +2099,29 @@ struct dmub_cmd_lsdma_data {
 				} fields;
 				uint32_t raw;
 			} packet;
+		} pio_copy_data;
+		struct lsdma_pio_constfill_data {
 			uint32_t dst_lo;
 			uint32_t dst_hi;
+
+			union {
+				struct {
+					uint32_t byte_count      : 26;
+					uint32_t src_loc         : 1;
+					uint32_t dst_loc         : 1;
+					uint32_t src_addr_inc    : 1;
+					uint32_t dst_addr_inc    : 1;
+					uint32_t overlap_disable : 1;
+					uint32_t constant_fill   : 1;
+				} fields;
+				uint32_t raw;
+			} packet;
+
 			uint32_t data;
 		} pio_constfill_data;
 
 		uint32_t all[14];
 	} u;
-
 };
 
 struct dmub_rb_cmd_lsdma {
@@ -4047,6 +4080,14 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * DIG BE HW instance.
 	 */
 	uint8_t digbe_inst;
+	/**
+	 * @hpo_stream_enc_inst: HPO stream encoder instance
+	 */
+	uint8_t hpo_stream_enc_inst;
+	/**
+	 * @hpo_link_enc_inst: HPO link encoder instance
+	 */
+	uint8_t hpo_link_enc_inst;
 	/**
 	 * AUX HW instance.
 	 */
@@ -4150,6 +4191,18 @@ struct dmub_rb_cmd_replay_enable_data {
 	 * This does not support HDMI/DP2 for now.
 	 */
 	uint8_t phy_rate;
+	/**
+	 * @hpo_stream_enc_inst: HPO stream encoder instance
+	 */
+	uint8_t hpo_stream_enc_inst;
+	/**
+	 * @hpo_link_enc_inst: HPO link encoder instance
+	 */
+	uint8_t hpo_link_enc_inst;
+	/**
+	 * @pad: Align structure to 4 byte boundary.
+	 */
+	uint8_t pad[2];
 };
 
 /**
@@ -4663,22 +4716,26 @@ enum dmub_cmd_lsdma_type {
 	 * LSDMA copies data from source to destination linearly
 	 */
 	DMUB_CMD__LSDMA_LINEAR_COPY = 1,
+	/**
+	* LSDMA copies data from source to destination linearly in sub window
+	*/
+	DMUB_CMD__LSDMA_LINEAR_SUB_WINDOW_COPY = 2,
 	/**
 	 * Send the tiled-to-tiled copy command
 	 */
-	DMUB_CMD__LSDMA_TILED_TO_TILED_COPY = 2,
+	DMUB_CMD__LSDMA_TILED_TO_TILED_COPY = 3,
 	/**
 	 * Send the poll reg write command
 	 */
-	DMUB_CMD__LSDMA_POLL_REG_WRITE = 3,
+	DMUB_CMD__LSDMA_POLL_REG_WRITE = 4,
 	/**
 	 * Send the pio copy command
 	 */
-	DMUB_CMD__LSDMA_PIO_COPY = 4,
+	DMUB_CMD__LSDMA_PIO_COPY = 5,
 	/**
 	 * Send the pio constfill command
 	 */
-	DMUB_CMD__LSDMA_PIO_CONSTFILL = 5,
+	DMUB_CMD__LSDMA_PIO_CONSTFILL = 6,
 };
 
 struct abm_ace_curve {

From def57ea33fb79d72c9bd3a4216f2bb77e768d7ca Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Sun, 13 Jul 2025 12:42:07 -0500
Subject: [PATCH 0122/1455] drm/amd/display: Promote DAL to 3.2.342

This version brings along following fixes:
- Fix divide by zero when calculating min ODM factor
- Ensure committing streams is seamless when transitioning between
  topologies
- Fix condition for setting timing_adjust_pending
- Update DMUB tiled to tiled copy command

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index d0c78f94a7a3..779b39e33443 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.341"
+#define DC_VER "3.2.342"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 020ad3a4ed62e5e8d97db27008c1e470f91bfd8d Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Thu, 10 Jul 2025 13:51:57 +0800
Subject: [PATCH 0123/1455] drm/amdgpu: query the allocated vram address block
 info

The bad pages that need to be retired are not all allocated
in the same poison consumption process, so an interface is
needed to query the processes that allocate the bad pages.
By killing all the processes that allocate the bad pages,
the bad pages can be reserved immediately.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 38 ++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h | 17 +++++++++
 2 files changed, 55 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index abdc52b0895a..4bf3e99f47fe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -396,6 +396,35 @@ out:
 	return ret;
 }
 
+int amdgpu_vram_mgr_query_address_block_info(struct amdgpu_vram_mgr *mgr,
+			uint64_t address, struct amdgpu_vram_block_info *info)
+{
+	struct amdgpu_vram_mgr_resource *vres;
+	struct drm_buddy_block *block;
+	u64 start, size;
+	int ret = -ENOENT;
+
+	mutex_lock(&mgr->lock);
+	list_for_each_entry(vres, &mgr->allocated_vres_list, vres_node) {
+		list_for_each_entry(block, &vres->blocks, link) {
+			start = amdgpu_vram_mgr_block_start(block);
+			size = amdgpu_vram_mgr_block_size(block);
+			if ((start <= address) && (address < (start + size))) {
+				info->start = start;
+				info->size = size;
+				memcpy(&info->task, &vres->task, sizeof(vres->task));
+				ret = 0;
+				goto out;
+			}
+		}
+	}
+
+out:
+	mutex_unlock(&mgr->lock);
+
+	return ret;
+}
+
 static void amdgpu_dummy_vram_mgr_debug(struct ttm_resource_manager *man,
 				  struct drm_printer *printer)
 {
@@ -568,6 +597,10 @@ static int amdgpu_vram_mgr_new(struct ttm_resource_manager *man,
 			remaining_size -= size;
 	}
 
+	vres->task.pid = task_pid_nr(current);
+	get_task_comm(vres->task.comm, current);
+	list_add_tail(&vres->vres_node, &mgr->allocated_vres_list);
+
 	if (bo->flags & AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS && adjust_dcc_size) {
 		struct drm_buddy_block *dcc_block;
 		unsigned long dcc_start;
@@ -645,6 +678,10 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager *man,
 	uint64_t vis_usage = 0;
 
 	mutex_lock(&mgr->lock);
+
+	list_del(&vres->vres_node);
+	memset(&vres->task, 0, sizeof(vres->task));
+
 	list_for_each_entry(block, &vres->blocks, link)
 		vis_usage += amdgpu_vram_mgr_vis_size(adev, block);
 
@@ -917,6 +954,7 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev)
 	mutex_init(&mgr->lock);
 	INIT_LIST_HEAD(&mgr->reservations_pending);
 	INIT_LIST_HEAD(&mgr->reserved_pages);
+	INIT_LIST_HEAD(&mgr->allocated_vres_list);
 	mgr->default_page_size = PAGE_SIZE;
 
 	if (!adev->gmc.is_app_apu) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h
index 2c88d5fd87da..5f5fd9a911c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.h
@@ -35,12 +35,26 @@ struct amdgpu_vram_mgr {
 	struct list_head reserved_pages;
 	atomic64_t vis_usage;
 	u64 default_page_size;
+	struct list_head allocated_vres_list;
+};
+
+struct amdgpu_vres_task {
+	pid_t pid;
+	char  comm[TASK_COMM_LEN];
+};
+
+struct amdgpu_vram_block_info {
+	u64 start;
+	u64 size;
+	struct amdgpu_vres_task task;
 };
 
 struct amdgpu_vram_mgr_resource {
 	struct ttm_resource base;
 	struct list_head blocks;
 	unsigned long flags;
+	struct list_head vres_node;
+	struct amdgpu_vres_task task;
 };
 
 static inline u64 amdgpu_vram_mgr_block_start(struct drm_buddy_block *block)
@@ -72,4 +86,7 @@ static inline void amdgpu_vram_mgr_set_cleared(struct ttm_resource *res)
 	ares->flags |= DRM_BUDDY_CLEARED;
 }
 
+int amdgpu_vram_mgr_query_address_block_info(struct amdgpu_vram_mgr *mgr,
+		uint64_t address, struct amdgpu_vram_block_info *info);
+
 #endif

From a813437c33842c4e28a0656a9d8f20c3a8d35d6d Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Wed, 16 Jul 2025 11:16:20 +0800
Subject: [PATCH 0124/1455] drm/amdgpu: add command to check address validity

Add command to check address validity and remove
unused command codes.

v2:
 The command interface adds new parameters to support
 multiple check address strategies.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 63 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h |  3 ++
 2 files changed, 66 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index a0a14370745e..527ffad47563 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -128,6 +128,9 @@ const char *get_ras_block_str(struct ras_common_if *ras_block)
 
 #define MAX_FLUSH_RETIRE_DWORK_TIMES  100
 
+#define BYPASS_ALLOCATED_ADDRESS        0x0
+#define BYPASS_INITIALIZATION_ADDRESS   0x1
+
 enum amdgpu_ras_retire_page_reservation {
 	AMDGPU_RAS_RETIRE_PAGE_RESERVED,
 	AMDGPU_RAS_RETIRE_PAGE_PENDING,
@@ -207,6 +210,49 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre
 	return 0;
 }
 
+static int amdgpu_check_address_validity(struct amdgpu_device *adev,
+			uint64_t address, uint64_t flags)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct amdgpu_vram_block_info blk_info;
+	uint64_t page_pfns[32] = {0};
+	int i, ret, count;
+
+	if (amdgpu_ip_version(adev, UMC_HWIP, 0) < IP_VERSION(12, 0, 0))
+		return 0;
+
+	if ((address >= adev->gmc.mc_vram_size) ||
+	    (address >= RAS_UMC_INJECT_ADDR_LIMIT))
+		return -EFAULT;
+
+	count = amdgpu_umc_lookup_bad_pages_in_a_row(adev,
+				address, page_pfns, ARRAY_SIZE(page_pfns));
+	if (count <= 0)
+		return -EPERM;
+
+	for (i = 0; i < count; i++) {
+		memset(&blk_info, 0, sizeof(blk_info));
+		ret = amdgpu_vram_mgr_query_address_block_info(&adev->mman.vram_mgr,
+					page_pfns[i] << AMDGPU_GPU_PAGE_SHIFT, &blk_info);
+		if (!ret) {
+			/* The input address that needs to be checked is allocated by
+			 * current calling process, so it is necessary to exclude
+			 * the calling process.
+			 */
+			if ((flags == BYPASS_ALLOCATED_ADDRESS) &&
+			    ((blk_info.task.pid != task_pid_nr(current)) ||
+				strncmp(blk_info.task.comm, current->comm, TASK_COMM_LEN)))
+				return -EACCES;
+			else if ((flags == BYPASS_INITIALIZATION_ADDRESS) &&
+				(blk_info.task.pid == con->init_task_pid) &&
+				!strncmp(blk_info.task.comm, con->init_task_comm, TASK_COMM_LEN))
+				return -EACCES;
+		}
+	}
+
+	return 0;
+}
+
 static ssize_t amdgpu_ras_debugfs_read(struct file *f, char __user *buf,
 					size_t size, loff_t *pos)
 {
@@ -297,6 +343,8 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 		op = 2;
 	else if (strstr(str, "retire_page") != NULL)
 		op = 3;
+	else if (strstr(str, "check_address") != NULL)
+		op = 4;
 	else if (str[0] && str[1] && str[2] && str[3])
 		/* ascii string, but commands are not matched. */
 		return -EINVAL;
@@ -310,6 +358,15 @@ static int amdgpu_ras_debugfs_ctrl_parse_data(struct file *f,
 			data->op = op;
 			data->inject.address = address;
 
+			return 0;
+		} else if (op == 4) {
+			if (sscanf(str, "%*s 0x%llx 0x%llx", &address, &value) != 2 &&
+			    sscanf(str, "%*s %llu %llu", &address, &value) != 2)
+				return -EINVAL;
+
+			data->op = op;
+			data->inject.address = address;
+			data->inject.value = value;
 			return 0;
 		}
 
@@ -500,6 +557,9 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f,
 			return size;
 		else
 			return ret;
+	} else if (data.op == 4) {
+		ret = amdgpu_check_address_validity(adev, data.inject.address, data.inject.value);
+		return ret ? ret : size;
 	}
 
 	if (!amdgpu_ras_is_supported(adev, data.head.block))
@@ -4087,6 +4147,9 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 			goto release_con;
 	}
 
+	con->init_task_pid = task_pid_nr(current);
+	get_task_comm(con->init_task_comm, current);
+
 	dev_info(adev->dev, "RAS INFO: ras initialized successfully, "
 		 "hardware ability[%x] ras_mask[%x]\n",
 		 adev->ras_hw_enabled, adev->ras_enabled);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 927d6bff734a..7f10a7402160 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -570,6 +570,9 @@ struct amdgpu_ras {
 	struct ras_event_manager *event_mgr;
 
 	uint64_t reserved_pages_in_bytes;
+
+	pid_t init_task_pid;
+	char init_task_comm[TASK_COMM_LEN];
 };
 
 struct ras_fs_data {

From 2b17c240e8cd9ac61d3c82277fbed27edad7f002 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Wed, 2 Jul 2025 16:16:02 +0800
Subject: [PATCH 0125/1455] drm/amdgpu: add range check for RAS bad page
 address

Exclude invalid bad pages.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 60 ++++++++++++-------------
 1 file changed, 29 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 527ffad47563..f0ffa2f1d0e9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -139,9 +139,9 @@ enum amdgpu_ras_retire_page_reservation {
 
 atomic_t amdgpu_ras_in_intr = ATOMIC_INIT(0);
 
-static bool amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con,
+static int amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con,
 				uint64_t addr);
-static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
+static int amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
 				uint64_t addr);
 #ifdef CONFIG_X86_MCE_AMD
 static void amdgpu_register_bad_pages_mca_notifier(struct amdgpu_device *adev);
@@ -172,18 +172,16 @@ static int amdgpu_reserve_page_direct(struct amdgpu_device *adev, uint64_t addre
 	struct eeprom_table_record err_rec;
 	int ret;
 
-	if ((address >= adev->gmc.mc_vram_size) ||
-	    (address >= RAS_UMC_INJECT_ADDR_LIMIT)) {
+	ret = amdgpu_ras_check_bad_page(adev, address);
+	if (ret == -EINVAL) {
 		dev_warn(adev->dev,
-		         "RAS WARN: input address 0x%llx is invalid.\n",
-		         address);
+			"RAS WARN: input address 0x%llx is invalid.\n",
+			address);
 		return -EINVAL;
-	}
-
-	if (amdgpu_ras_check_bad_page(adev, address)) {
+	} else if (ret == 1) {
 		dev_warn(adev->dev,
-			 "RAS WARN: 0x%llx has already been marked as bad page!\n",
-			 address);
+			"RAS WARN: 0x%llx has already been marked as bad page!\n",
+			address);
 		return 0;
 	}
 
@@ -573,22 +571,16 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f,
 		ret = amdgpu_ras_feature_enable(adev, &data.head, 1);
 		break;
 	case 2:
-		if ((data.inject.address >= adev->gmc.mc_vram_size &&
-		    adev->gmc.mc_vram_size) ||
-		    (data.inject.address >= RAS_UMC_INJECT_ADDR_LIMIT)) {
-			dev_warn(adev->dev, "RAS WARN: input address "
-					"0x%llx is invalid.",
-					data.inject.address);
-			ret = -EINVAL;
-			break;
-		}
-
 		/* umc ce/ue error injection for a bad page is not allowed */
-		if ((data.head.block == AMDGPU_RAS_BLOCK__UMC) &&
-		    amdgpu_ras_check_bad_page(adev, data.inject.address)) {
-			dev_warn(adev->dev, "RAS WARN: inject: 0x%llx has "
-				 "already been marked as bad!\n",
-				 data.inject.address);
+		if (data.head.block == AMDGPU_RAS_BLOCK__UMC)
+			ret = amdgpu_ras_check_bad_page(adev, data.inject.address);
+		if (ret == -EINVAL) {
+			dev_warn(adev->dev, "RAS WARN: input address 0x%llx is invalid.",
+					data.inject.address);
+			break;
+		} else if (ret == 1) {
+			dev_warn(adev->dev, "RAS WARN: inject: 0x%llx has already been marked as bad!\n",
+					data.inject.address);
 			break;
 		}
 
@@ -3194,18 +3186,24 @@ out:
 	return ret;
 }
 
-static bool amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con,
+static int amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con,
 				uint64_t addr)
 {
 	struct ras_err_handler_data *data = con->eh_data;
+	struct amdgpu_device *adev = con->adev;
 	int i;
 
+	if ((addr >= adev->gmc.mc_vram_size &&
+	    adev->gmc.mc_vram_size) ||
+	    (addr >= RAS_UMC_INJECT_ADDR_LIMIT))
+		return -EINVAL;
+
 	addr >>= AMDGPU_GPU_PAGE_SHIFT;
 	for (i = 0; i < data->count; i++)
 		if (addr == data->bps[i].retired_page)
-			return true;
+			return 1;
 
-	return false;
+	return 0;
 }
 
 /*
@@ -3213,11 +3211,11 @@ static bool amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con,
  *
  * Note: this check is only for umc block
  */
-static bool amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
+static int amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
 				uint64_t addr)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
-	bool ret = false;
+	int ret = 0;
 
 	if (!con || !con->eh_data)
 		return ret;

From d45c5e6845a76169ef3d6076f0f04487e5776905 Mon Sep 17 00:00:00 2001
From: Tao Zhou <tao.zhou1@amd.com>
Date: Fri, 4 Jul 2025 17:12:24 +0800
Subject: [PATCH 0126/1455] drm/amdgpu: adjust the update of RAS bad page
 number

One eeprom record may not map to unit number of bad pages, the accurate
bad page number is gotten after bad page address check.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c       | 41 +++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h       |  3 ++
 .../gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c    |  6 +--
 3 files changed, 30 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index f0ffa2f1d0e9..d49671a2ecd5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2625,6 +2625,9 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device *adev,
 	}
 
 	for (; i < data->count; i++) {
+		if (!data->bps[i].ts)
+			continue;
+
 		(*bps)[i] = (struct ras_badpage){
 			.bp = data->bps[i].retired_page,
 			.size = AMDGPU_GPU_PAGE_SIZE,
@@ -2638,7 +2641,7 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device *adev,
 			(*bps)[i].flags = AMDGPU_RAS_RETIRE_PAGE_FAULT;
 	}
 
-	*count = data->count;
+	*count = con->bad_page_num;
 out:
 	mutex_unlock(&con->recovery_lock);
 	return ret;
@@ -2866,8 +2869,11 @@ static int __amdgpu_ras_restore_bad_pages(struct amdgpu_device *adev,
 
 	for (j = 0; j < count; j++) {
 		if (amdgpu_ras_check_bad_page_unlock(con,
-			bps[j].retired_page << AMDGPU_GPU_PAGE_SHIFT))
+			bps[j].retired_page << AMDGPU_GPU_PAGE_SHIFT)) {
+			data->count++;
+			data->space_left--;
 			continue;
+		}
 
 		if (!data->space_left &&
 		    amdgpu_ras_realloc_eh_data_space(adev, data, 256)) {
@@ -2880,6 +2886,7 @@ static int __amdgpu_ras_restore_bad_pages(struct amdgpu_device *adev,
 				sizeof(struct eeprom_table_record));
 		data->count++;
 		data->space_left--;
+		con->bad_page_num++;
 	}
 
 	return 0;
@@ -3026,7 +3033,7 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
 						ret = __amdgpu_ras_convert_rec_array_from_rom(adev,
 										&bps[i], &err_data, nps);
 						if (ret)
-							control->ras_num_bad_pages -= adev->umc.retire_unit;
+							con->bad_page_num -= adev->umc.retire_unit;
 						i += (adev->umc.retire_unit - 1);
 					} else {
 						break;
@@ -3040,8 +3047,10 @@ int amdgpu_ras_add_bad_pages(struct amdgpu_device *adev,
 			ret = __amdgpu_ras_convert_rec_from_rom(adev,
 				&bps[i], &err_data, nps);
 			if (ret)
-				control->ras_num_bad_pages -= adev->umc.retire_unit;
+				con->bad_page_num -= adev->umc.retire_unit;
 		}
+
+		con->eh_data->count_saved = con->eh_data->count;
 	} else {
 		ret = __amdgpu_ras_restore_bad_pages(adev, bps, pages);
 	}
@@ -3064,7 +3073,7 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev,
 	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
 	struct ras_err_handler_data *data;
 	struct amdgpu_ras_eeprom_control *control;
-	int save_count, unit_num, bad_page_num, i;
+	int save_count, unit_num, i;
 
 	if (!con || !con->eh_data) {
 		if (new_cnt)
@@ -3085,27 +3094,26 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev,
 	mutex_lock(&con->recovery_lock);
 	control = &con->eeprom_control;
 	data = con->eh_data;
-	bad_page_num = control->ras_num_bad_pages;
-	save_count = data->count - bad_page_num;
+	unit_num = data->count / adev->umc.retire_unit - control->ras_num_recs;
+	save_count = con->bad_page_num - control->ras_num_bad_pages;
 	mutex_unlock(&con->recovery_lock);
 
-	unit_num = save_count / adev->umc.retire_unit;
 	if (new_cnt)
 		*new_cnt = unit_num;
 
 	/* only new entries are saved */
-	if (save_count > 0) {
+	if (unit_num > 0) {
 		/*old asics only save pa to eeprom like before*/
 		if (IP_VERSION_MAJ(amdgpu_ip_version(adev, UMC_HWIP, 0)) < 12) {
 			if (amdgpu_ras_eeprom_append(control,
-					&data->bps[bad_page_num], save_count)) {
+					&data->bps[data->count_saved], unit_num)) {
 				dev_err(adev->dev, "Failed to save EEPROM table data!");
 				return -EIO;
 			}
 		} else {
 			for (i = 0; i < unit_num; i++) {
 				if (amdgpu_ras_eeprom_append(control,
-						&data->bps[bad_page_num +
+						&data->bps[data->count_saved +
 						i * adev->umc.retire_unit], 1)) {
 					dev_err(adev->dev, "Failed to save EEPROM table data!");
 					return -EIO;
@@ -3114,6 +3122,7 @@ int amdgpu_ras_save_bad_pages(struct amdgpu_device *adev,
 		}
 
 		dev_info(adev->dev, "Saved %d pages to EEPROM table.\n", save_count);
+		data->count_saved = data->count;
 	}
 
 	return 0;
@@ -3168,17 +3177,17 @@ static int amdgpu_ras_load_bad_pages(struct amdgpu_device *adev)
 			}
 		}
 
+		ret = amdgpu_ras_add_bad_pages(adev, bps, control->ras_num_recs, true);
+		if (ret)
+			goto out;
+
 		ret = amdgpu_ras_eeprom_check(control);
 		if (ret)
 			goto out;
 
 		/* HW not usable */
-		if (amdgpu_ras_is_rma(adev)) {
+		if (amdgpu_ras_is_rma(adev))
 			ret = -EHWPOISON;
-			goto out;
-		}
-
-		ret = amdgpu_ras_add_bad_pages(adev, bps, control->ras_num_recs, true);
 	}
 
 out:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 7f10a7402160..aa155cb5d58e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -573,6 +573,8 @@ struct amdgpu_ras {
 
 	pid_t init_task_pid;
 	char init_task_comm[TASK_COMM_LEN];
+
+	int bad_page_num;
 };
 
 struct ras_fs_data {
@@ -611,6 +613,7 @@ struct ras_err_handler_data {
 	struct eeprom_table_record *bps;
 	/* the count of entries */
 	int count;
+	int count_saved;
 	/* the space can place new entries */
 	int space_left;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index 9bda9ad13f88..c3c908cc0859 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -743,8 +743,7 @@ amdgpu_ras_eeprom_append_table(struct amdgpu_ras_eeprom_control *control,
 	else
 		control->ras_num_mca_recs += num;
 
-	control->ras_num_bad_pages = control->ras_num_pa_recs +
-				control->ras_num_mca_recs * adev->umc.retire_unit;
+	control->ras_num_bad_pages = con->bad_page_num;
 Out:
 	kfree(buf);
 	return res;
@@ -1457,8 +1456,7 @@ int amdgpu_ras_eeprom_check(struct amdgpu_ras_eeprom_control *control)
 	if (!__get_eeprom_i2c_addr(adev, control))
 		return -EINVAL;
 
-	control->ras_num_bad_pages = control->ras_num_pa_recs +
-			control->ras_num_mca_recs * adev->umc.retire_unit;
+	control->ras_num_bad_pages = ras->bad_page_num;
 
 	if (hdr->header == RAS_TABLE_HDR_VAL) {
 		dev_dbg(adev->dev,

From f3486918979030f8982e1af901561dbd6e2cd1bc Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Thu, 24 Jul 2025 15:34:29 +0800
Subject: [PATCH 0127/1455] drm/amdgpu: support ras critical address check

Support ras critical address check.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 89 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 14 ++++
 2 files changed, 103 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index d49671a2ecd5..b75a0dc22ace 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -143,6 +143,10 @@ static int amdgpu_ras_check_bad_page_unlock(struct amdgpu_ras *con,
 				uint64_t addr);
 static int amdgpu_ras_check_bad_page(struct amdgpu_device *adev,
 				uint64_t addr);
+
+static void amdgpu_ras_critical_region_init(struct amdgpu_device *adev);
+static void amdgpu_ras_critical_region_fini(struct amdgpu_device *adev);
+
 #ifdef CONFIG_X86_MCE_AMD
 static void amdgpu_register_bad_pages_mca_notifier(struct amdgpu_device *adev);
 struct mce_notifier_adev_list {
@@ -3728,6 +3732,8 @@ static int amdgpu_ras_recovery_fini(struct amdgpu_device *adev)
 	kfree(data);
 	mutex_unlock(&con->recovery_lock);
 
+	amdgpu_ras_critical_region_init(adev);
+
 	return 0;
 }
 /* recovery end */
@@ -4157,6 +4163,9 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 	con->init_task_pid = task_pid_nr(current);
 	get_task_comm(con->init_task_comm, current);
 
+	mutex_init(&con->critical_region_lock);
+	INIT_LIST_HEAD(&con->critical_region_head);
+
 	dev_info(adev->dev, "RAS INFO: ras initialized successfully, "
 		 "hardware ability[%x] ras_mask[%x]\n",
 		 adev->ras_hw_enabled, adev->ras_enabled);
@@ -4436,6 +4445,9 @@ int amdgpu_ras_fini(struct amdgpu_device *adev)
 	if (!adev->ras_enabled || !con)
 		return 0;
 
+	amdgpu_ras_critical_region_fini(adev);
+	mutex_destroy(&con->critical_region_lock);
+
 	list_for_each_entry_safe(ras_node, tmp, &adev->ras_list, node) {
 		if (ras_node->ras_obj) {
 			obj = ras_node->ras_obj;
@@ -5380,3 +5392,80 @@ bool amdgpu_ras_is_rma(struct amdgpu_device *adev)
 
 	return con->is_rma;
 }
+
+int amdgpu_ras_add_critical_region(struct amdgpu_device *adev,
+			struct amdgpu_bo *bo)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct amdgpu_vram_mgr_resource *vres;
+	struct ras_critical_region *region;
+	struct drm_buddy_block *block;
+	int ret = 0;
+
+	if (!bo || !bo->tbo.resource)
+		return -EINVAL;
+
+	vres = to_amdgpu_vram_mgr_resource(bo->tbo.resource);
+
+	mutex_lock(&con->critical_region_lock);
+
+	/* Check if the bo had been recorded */
+	list_for_each_entry(region, &con->critical_region_head, node)
+		if (region->bo == bo)
+			goto out;
+
+	/* Record new critical amdgpu bo */
+	list_for_each_entry(block, &vres->blocks, link) {
+		region = kzalloc(sizeof(*region), GFP_KERNEL);
+		if (!region) {
+			ret = -ENOMEM;
+			goto out;
+		}
+		region->bo = bo;
+		region->start = amdgpu_vram_mgr_block_start(block);
+		region->size = amdgpu_vram_mgr_block_size(block);
+		list_add_tail(&region->node, &con->critical_region_head);
+	}
+
+out:
+	mutex_unlock(&con->critical_region_lock);
+
+	return ret;
+}
+
+static void amdgpu_ras_critical_region_init(struct amdgpu_device *adev)
+{
+	amdgpu_ras_add_critical_region(adev, adev->mman.fw_reserved_memory);
+}
+
+static void amdgpu_ras_critical_region_fini(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_critical_region *region, *tmp;
+
+	mutex_lock(&con->critical_region_lock);
+	list_for_each_entry_safe(region, tmp, &con->critical_region_head, node) {
+		list_del(&region->node);
+		kfree(region);
+	}
+	mutex_unlock(&con->critical_region_lock);
+}
+
+bool amdgpu_ras_check_critical_address(struct amdgpu_device *adev, uint64_t addr)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_critical_region *region;
+	bool ret = false;
+
+	mutex_lock(&con->critical_region_lock);
+	list_for_each_entry(region, &con->critical_region_head, node) {
+		if ((region->start <= addr) &&
+		    (addr < (region->start + region->size))) {
+			ret = true;
+			break;
+		}
+	}
+	mutex_unlock(&con->critical_region_lock);
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index aa155cb5d58e..434e23c84962 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -496,6 +496,13 @@ struct ras_ecc_log_info {
 	uint64_t	prev_de_queried_count;
 };
 
+struct ras_critical_region {
+	struct list_head node;
+	struct amdgpu_bo *bo;
+	uint64_t start;
+	uint64_t size;
+};
+
 struct amdgpu_ras {
 	/* ras infrastructure */
 	/* for ras itself. */
@@ -575,6 +582,10 @@ struct amdgpu_ras {
 	char init_task_comm[TASK_COMM_LEN];
 
 	int bad_page_num;
+
+	struct list_head critical_region_head;
+	struct mutex critical_region_lock;
+
 };
 
 struct ras_fs_data {
@@ -979,6 +990,9 @@ int amdgpu_ras_mark_ras_event_caller(struct amdgpu_device *adev, enum ras_event_
 
 int amdgpu_ras_reserve_page(struct amdgpu_device *adev, uint64_t pfn);
 
+int amdgpu_ras_add_critical_region(struct amdgpu_device *adev, struct amdgpu_bo *bo);
+bool amdgpu_ras_check_critical_address(struct amdgpu_device *adev, uint64_t addr);
+
 int amdgpu_ras_put_poison_req(struct amdgpu_device *adev,
 		enum amdgpu_ras_block block, uint16_t pasid,
 		pasid_notify pasid_fn, void *data, uint32_t reset);

From 9d1ac25c7f830e0132aa816393b1e9f140e71148 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 8 Jul 2025 13:17:18 +0530
Subject: [PATCH 0128/1455] drm/amdgpu: Update supported modes for GC v9.5.0

For GC v9.5.0 SOCs, both CPX and QPX compute modes are also supported in
NPS2 mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
index 914cf4bfb033..811124ff88a8 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
@@ -227,6 +227,7 @@ static int __aqua_vanjaram_get_px_mode_info(struct amdgpu_xcp_mgr *xcp_mgr,
 					    uint16_t *nps_modes)
 {
 	struct amdgpu_device *adev = xcp_mgr->adev;
+	uint32_t gc_ver = amdgpu_ip_version(adev, GC_HWIP, 0);
 
 	if (!num_xcp || !nps_modes || !(xcp_mgr->supp_xcp_modes & BIT(px_mode)))
 		return -EINVAL;
@@ -250,12 +251,14 @@ static int __aqua_vanjaram_get_px_mode_info(struct amdgpu_xcp_mgr *xcp_mgr,
 		*num_xcp = 4;
 		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
 			     BIT(AMDGPU_NPS4_PARTITION_MODE);
+		if (gc_ver == IP_VERSION(9, 5, 0))
+			*nps_modes |= BIT(AMDGPU_NPS2_PARTITION_MODE);
 		break;
 	case AMDGPU_CPX_PARTITION_MODE:
 		*num_xcp = NUM_XCC(adev->gfx.xcc_mask);
 		*nps_modes = BIT(AMDGPU_NPS1_PARTITION_MODE) |
 			     BIT(AMDGPU_NPS4_PARTITION_MODE);
-		if (amdgpu_sriov_vf(adev))
+		if (gc_ver == IP_VERSION(9, 5, 0))
 			*nps_modes |= BIT(AMDGPU_NPS2_PARTITION_MODE);
 		break;
 	default:

From faab5ea0836733ef1c8e83cf6b05690a5c9066be Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Sun, 13 Jul 2025 01:28:02 +0530
Subject: [PATCH 0129/1455] drm/amdgpu: Check vcn sram load return value

Log an error when vcn sram load fails in indirect mode
and return the same error value.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c   | 10 ++++++++--
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c   | 10 ++++++++--
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c   | 10 ++++++++--
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c   | 10 ++++++++--
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 11 ++++++++---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 10 ++++++++--
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c |  9 +++++++--
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 11 ++++++++---
 8 files changed, 63 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index 68b4371df0f1..d1481e6d57ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -865,6 +865,7 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
 	uint32_t rb_bufsz, tmp;
+	int ret;
 
 	vcn_v2_0_enable_static_power_gating(vinst);
 
@@ -948,8 +949,13 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 		UVD, 0, mmUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, 0, 0);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, 0, 0);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	/* force RBC into idle state */
 	rb_bufsz = order_base_2(ring->ring_size);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index bc30a5326866..d7b2668ab0d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -1035,6 +1035,7 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t rb_bufsz, tmp;
+	int ret;
 
 	/* disable register anti-hang mechanism */
 	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), 1,
@@ -1125,8 +1126,13 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 		VCN, 0, mmUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_dec;
 	/* force RBC into idle state */
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 4b8f4407047f..a89662c97c9e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -1042,6 +1042,7 @@ static int vcn_v3_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t rb_bufsz, tmp;
+	int ret;
 
 	/* disable register anti-hang mechanism */
 	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, mmUVD_POWER_STATUS), 1,
@@ -1134,8 +1135,13 @@ static int vcn_v3_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 	WREG32_SOC15_DPG_MODE(inst_idx, SOC15_DPG_MODE_OFFSET(
 		VCN, inst_idx, mmUVD_VCPU_CNTL), tmp, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_dec;
 	/* force RBC into idle state */
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 1924e075b66f..e9f9cb169034 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1012,6 +1012,7 @@ static int vcn_v4_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
+	int ret;
 
 	/* disable register anti-hang mechanism */
 	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1,
@@ -1094,8 +1095,13 @@ static int vcn_v4_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 2a3663b551af..b904afc358ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -851,7 +851,7 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared =
 						adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
-	int vcn_inst;
+	int vcn_inst, ret;
 	uint32_t tmp;
 
 	vcn_inst = GET_INST(VCN, inst_idx);
@@ -944,8 +944,13 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 		VCN, 0, regUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index caf2d95a85d4..ed57e6431d4b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -926,6 +926,7 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 	volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
+	int ret;
 
 	/* disable register anti-hang mechanism */
 	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1,
@@ -1006,8 +1007,13 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 		VCN, inst_idx, regUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index 07a6e9582880..f8bb90fe764b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -713,6 +713,7 @@ static int vcn_v5_0_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 	volatile struct amdgpu_vcn5_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
+	int ret;
 
 	/* disable register anti-hang mechanism */
 	WREG32_P(SOC15_REG_OFFSET(VCN, inst_idx, regUVD_POWER_STATUS), 1,
@@ -766,8 +767,12 @@ static int vcn_v5_0_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 		VCN, inst_idx, regUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
+		dev_err(adev->dev, "%s: vcn sram load failed %d\n", __func__, ret);
+		if (ret)
+			return ret;
+	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index cdefd7fcb0da..d8bbb9376731 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -605,7 +605,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 		adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	struct dpg_pause_state state = {.fw_based = VCN_DPG_STATE__PAUSE};
-	int vcn_inst;
+	int vcn_inst, ret;
 	uint32_t tmp;
 
 	vcn_inst = GET_INST(VCN, inst_idx);
@@ -666,8 +666,13 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 		VCN, 0, regUVD_MASTINT_EN),
 		UVD_MASTINT_EN__VCPU_EN_MASK, 0, indirect);
 
-	if (indirect)
-		amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM);
+	if (indirect) {
+		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, AMDGPU_UCODE_ID_VCN0_RAM);
+		if (ret) {
+			dev_err(adev->dev, "vcn sram load failed %d\n", ret);
+			return ret;
+		}
+	}
 
 	/* resetting ring, fw should not check RB ring */
 	fw_shared->sq.queue_mode |= FW_QUEUE_RING_RESET;

From 3fc96f60b61cebf8649f557b5134795fbdb00e12 Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Tue, 22 Jul 2025 14:17:29 +0800
Subject: [PATCH 0130/1455] drm/amdgpu: add critical address check for bad page
 retirement

Add critical address check for bad page retirement.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index b75a0dc22ace..1e31ef5592f2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2637,6 +2637,11 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device *adev,
 			.size = AMDGPU_GPU_PAGE_SIZE,
 			.flags = AMDGPU_RAS_RETIRE_PAGE_RESERVED,
 		};
+
+		if (amdgpu_ras_check_critical_address(adev,
+			data->bps[i].retired_page << AMDGPU_GPU_PAGE_SHIFT))
+			continue;
+
 		status = amdgpu_vram_mgr_query_page_status(&adev->mman.vram_mgr,
 				data->bps[i].retired_page << AMDGPU_GPU_PAGE_SHIFT);
 		if (status == -EBUSY)
@@ -5356,6 +5361,9 @@ int amdgpu_ras_reserve_page(struct amdgpu_device *adev, uint64_t pfn)
 	uint64_t start = pfn << AMDGPU_GPU_PAGE_SHIFT;
 	int ret = 0;
 
+	if (amdgpu_ras_check_critical_address(adev, start))
+		return 0;
+
 	mutex_lock(&con->page_rsv_lock);
 	ret = amdgpu_vram_mgr_query_page_status(mgr, start);
 	if (ret == -ENOENT)

From 21c6764ed4bfaecad034bc4fd15dd64c5a436325 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 25 Jul 2025 10:21:10 +0530
Subject: [PATCH 0131/1455] drm/amdgpu: Update external revid for GC v9.5.0

Use different external revid for GC v9.5.0 SOCs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/soc15.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index c457be3a3c56..9e74c9822e62 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -1218,6 +1218,8 @@ static int soc15_common_early_init(struct amdgpu_ip_block *ip_block)
 			AMD_PG_SUPPORT_JPEG;
 		/*TODO: need a new external_rev_id for GC 9.4.4? */
 		adev->external_rev_id = adev->rev_id + 0x46;
+		if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 5, 0))
+			adev->external_rev_id = adev->rev_id + 0x50;
 		break;
 	default:
 		/* FIXME: not supported yet */

From 1cf1205ef2685cf43db3785706b017d1e54e0bec Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 24 Jul 2025 13:02:18 +0530
Subject: [PATCH 0132/1455] drm/amd/display: Reduce Stack Usage by moving
 'audio_output' into 'stream_res' v4

The function `dp_retrain_link_dp_test` currently allocates a large
audio_output array on the stack, causing the stack frame size to exceed
the compiler limit (1080 bytes > 1024 bytes).

This change prevents stack overflow issues:
amdgpu/../display/dc/link/accessories/link_dp_cts.c:65:13: warning: stack frame size (1080) exceeds limit (1024) in 'dp_retrain_link_dp_test' [-Wframe-larger-than]
static void dp_retrain_link_dp_test(struct dc_link *link,

v2: Move audio-related data like `audio_output` is kept "per pipe" to
    manage the audio for that specific display pipeline/display output path
    (stream). (Wenjing)

v3: Update in all the places where `build_audio_output` is currently
    called with a separate audio_output variable on the stack & wherever
    `audio_output` is passed to other functions
    `dce110_apply_single_controller_ctx_to_hw()` &
    `dce110_setup_audio_dto()` (like `az_configure`, `wall_dto_setup`)
    replace with usage of `pipe_ctx->stream_res.audio_output`
    to centralize audio data per pipe.

v4: Remove empty lines before `build_audio_output`. (Alex)

Fixes: 9c6669c2e21a ("drm/amd/display: Fix Link Override Sequencing When Switching Between DIO/HPO")
Cc: Wayne Lin <wayne.lin@amd.com>
Cc: George Shen <george.shen@amd.com>
Cc: Michael Strauss <michael.strauss@amd.com>
Cc: Alvin Lee <Alvin.Lee2@amd.com>
Cc: Ray Wu <ray.wu@amd.com>
Cc: Wenjing Liu <wenjing.liu@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Tom Chung <chiahsuan.chung@amd.com>
Cc: Roman Li <roman.li@amd.com>
Cc: Alex Hung <alex.hung@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/hwss/dce110/dce110_hwseq.c | 32 ++++++++-----------
 .../gpu/drm/amd/display/dc/inc/core_types.h   |  5 +--
 .../display/dc/link/accessories/link_dp_cts.c | 12 +++----
 .../dc/resource/dcn31/dcn31_resource.c        |  5 ++-
 .../dc/resource/dcn31/dcn31_resource.h        |  3 +-
 5 files changed, 26 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index 4ea13d0bf815..c69194e04ff9 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -1600,19 +1600,17 @@ enum dc_status dce110_apply_single_controller_ctx_to_hw(
 	}
 
 	if (pipe_ctx->stream_res.audio != NULL) {
-		struct audio_output audio_output = {0};
+		build_audio_output(context, pipe_ctx, &pipe_ctx->stream_res.audio_output);
 
-		build_audio_output(context, pipe_ctx, &audio_output);
-
-		link_hwss->setup_audio_output(pipe_ctx, &audio_output,
+		link_hwss->setup_audio_output(pipe_ctx, &pipe_ctx->stream_res.audio_output,
 				pipe_ctx->stream_res.audio->inst);
 
 		pipe_ctx->stream_res.audio->funcs->az_configure(
 				pipe_ctx->stream_res.audio,
 				pipe_ctx->stream->signal,
-				&audio_output.crtc_info,
+				&pipe_ctx->stream_res.audio_output.crtc_info,
 				&pipe_ctx->stream->audio_info,
-				&audio_output.dp_link_info);
+				&pipe_ctx->stream_res.audio_output.dp_link_info);
 
 		if (dc->config.disable_hbr_audio_dp2)
 			if (pipe_ctx->stream_res.audio->funcs->az_disable_hbr_audio &&
@@ -2386,9 +2384,7 @@ static void dce110_setup_audio_dto(
 		if (pipe_ctx->stream->signal != SIGNAL_TYPE_HDMI_TYPE_A)
 			continue;
 		if (pipe_ctx->stream_res.audio != NULL) {
-			struct audio_output audio_output;
-
-			build_audio_output(context, pipe_ctx, &audio_output);
+			build_audio_output(context, pipe_ctx, &pipe_ctx->stream_res.audio_output);
 
 			if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->set_audio_dtbclk_dto) {
 				struct dtbclk_dto_params dto_params = {0};
@@ -2399,14 +2395,14 @@ static void dce110_setup_audio_dto(
 				pipe_ctx->stream_res.audio->funcs->wall_dto_setup(
 						pipe_ctx->stream_res.audio,
 						pipe_ctx->stream->signal,
-						&audio_output.crtc_info,
-						&audio_output.pll_info);
+						&pipe_ctx->stream_res.audio_output.crtc_info,
+						&pipe_ctx->stream_res.audio_output.pll_info);
 			} else
 				pipe_ctx->stream_res.audio->funcs->wall_dto_setup(
 					pipe_ctx->stream_res.audio,
 					pipe_ctx->stream->signal,
-					&audio_output.crtc_info,
-					&audio_output.pll_info);
+					&pipe_ctx->stream_res.audio_output.crtc_info,
+					&pipe_ctx->stream_res.audio_output.pll_info);
 			break;
 		}
 	}
@@ -2426,15 +2422,15 @@ static void dce110_setup_audio_dto(
 				continue;
 
 			if (pipe_ctx->stream_res.audio != NULL) {
-				struct audio_output audio_output = {0};
-
-				build_audio_output(context, pipe_ctx, &audio_output);
+				build_audio_output(context,
+						   pipe_ctx,
+						   &pipe_ctx->stream_res.audio_output);
 
 				pipe_ctx->stream_res.audio->funcs->wall_dto_setup(
 					pipe_ctx->stream_res.audio,
 					pipe_ctx->stream->signal,
-					&audio_output.crtc_info,
-					&audio_output.pll_info);
+					&pipe_ctx->stream_res.audio_output.crtc_info,
+					&pipe_ctx->stream_res.audio_output.pll_info);
 				break;
 			}
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index f0d7185153b2..f896cce87b8d 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -228,8 +228,7 @@ struct resource_funcs {
 	enum dc_status (*update_dc_state_for_encoder_switch)(struct dc_link *link,
 		struct dc_link_settings *link_setting,
 		uint8_t pipe_count,
-		struct pipe_ctx *pipes,
-		struct audio_output *audio_output);
+		struct pipe_ctx *pipes);
 };
 
 struct audio_support{
@@ -361,6 +360,8 @@ struct stream_resource {
 	uint8_t gsl_group;
 
 	struct test_pattern_params test_pattern_params;
+
+	struct audio_output audio_output;
 };
 
 struct plane_resource {
diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
index 2956c2b3ad1a..b12d61701d4d 100644
--- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
+++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
@@ -75,7 +75,6 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 	bool is_hpo_acquired;
 	uint8_t count;
 	int i;
-	struct audio_output audio_output[MAX_PIPES];
 
 	needs_divider_update = (link->dc->link_srv->dp_get_encoding_format(link_setting) !=
 	link->dc->link_srv->dp_get_encoding_format((const struct dc_link_settings *) &link->cur_link_settings));
@@ -99,7 +98,7 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 	if (needs_divider_update && link->dc->res_pool->funcs->update_dc_state_for_encoder_switch) {
 		link->dc->res_pool->funcs->update_dc_state_for_encoder_switch(link,
 				link_setting, count,
-				*pipes, &audio_output[0]);
+				*pipes);
 		for (i = 0; i < count; i++) {
 			pipes[i]->clock_source->funcs->program_pix_clk(
 					pipes[i]->clock_source,
@@ -111,15 +110,16 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 				const struct link_hwss *link_hwss = get_link_hwss(
 					link, &pipes[i]->link_res);
 
-				link_hwss->setup_audio_output(pipes[i], &audio_output[i],
-						pipes[i]->stream_res.audio->inst);
+				link_hwss->setup_audio_output(pipes[i],
+							      &pipes[i]->stream_res.audio_output,
+							      pipes[i]->stream_res.audio->inst);
 
 				pipes[i]->stream_res.audio->funcs->az_configure(
 						pipes[i]->stream_res.audio,
 						pipes[i]->stream->signal,
-						&audio_output[i].crtc_info,
+						&pipes[i]->stream_res.audio_output.crtc_info,
 						&pipes[i]->stream->audio_info,
-						&audio_output[i].dp_link_info);
+						&pipes[i]->stream_res.audio_output.dp_link_info);
 
 				if (link->dc->config.disable_hbr_audio_dp2 &&
 						pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio &&
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
index 3ed7f50554e2..ca17e5d8fdc2 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
@@ -2239,8 +2239,7 @@ struct resource_pool *dcn31_create_resource_pool(
 enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link,
 	struct dc_link_settings *link_setting,
 	uint8_t pipe_count,
-	struct pipe_ctx *pipes,
-	struct audio_output *audio_output)
+	struct pipe_ctx *pipes)
 {
 	struct dc_state *state = link->dc->current_state;
 	int i;
@@ -2255,7 +2254,7 @@ enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link,
 
 		// Setup audio
 		if (pipes[i].stream_res.audio != NULL)
-			build_audio_output(state, &pipes[i], &audio_output[i]);
+			build_audio_output(state, &pipes[i], &pipes[i].stream_res.audio_output);
 	}
 #else
 	/* This DCN requires rate divider updates and audio reprogramming to allow DP1<-->DP2 link rate switching,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
index c32c85ef0ba4..7e8fde65528f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
@@ -69,8 +69,7 @@ unsigned int dcn31_get_det_buffer_size(
 enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link,
 	struct dc_link_settings *link_setting,
 	uint8_t pipe_count,
-	struct pipe_ctx *pipes,
-	struct audio_output *audio_output);
+	struct pipe_ctx *pipes);
 
 /*temp: B0 specific before switch to dcn313 headers*/
 #ifndef regPHYPLLF_PIXCLK_RESYNC_CNTL

From 04112dce53fc52438c8a1fb2b5ac779b6daeca5e Mon Sep 17 00:00:00 2001
From: Karthi Kandasamy <karthi.kandasamy@amd.com>
Date: Thu, 26 Jun 2025 15:10:01 +0200
Subject: [PATCH 0133/1455] drm/amd/display: Add DC EDID read policy struct

[Why & How]
Add a struct to allow DMs that utilize the EDID parser in DC to modify the
default settings.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Karthi Kandasamy <karthi.kandasamy@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_types.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 375ca2f13b7a..2a86058c3bfa 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -563,6 +563,12 @@ struct dc_info_packet_128 {
 	uint8_t sb[128];
 };
 
+struct dc_edid_read_policy {
+	uint32_t max_retry_count;
+	uint32_t delay_time_ms;
+	uint32_t ignore_checksum;
+};
+
 #define DC_PLANE_UPDATE_TIMES_MAX 10
 
 struct dc_plane_flip_time {

From d0e164f72e6a16e64f660023dc7ad25b31b8b08d Mon Sep 17 00:00:00 2001
From: Cruise Hung <Cruise.Hung@amd.com>
Date: Wed, 9 Jul 2025 19:04:31 +0800
Subject: [PATCH 0134/1455] drm/amd/display: Remove check DPIA HPD status for
 BW Allocation

[Why & How]
Link hpd_status is for embedded DPIA only.
Do not check hpd_status for BW allocation logic.

Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/link_validation.c |  6 +-
 .../dc/link/protocols/link_dp_dpia_bw.c       | 58 +++++++++----------
 2 files changed, 31 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.c b/drivers/gpu/drm/amd/display/dc/link/link_validation.c
index aecaf37eee35..acdc162de535 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_validation.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.c
@@ -408,8 +408,10 @@ enum dc_status link_validate_dp_tunnel_bandwidth(const struct dc *dc, const stru
 		link = stream->link;
 
 		if (!(link && (stream->signal == SIGNAL_TYPE_DISPLAY_PORT
-				|| stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)
-				&& link->hpd_status))
+				|| stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST)))
+			continue;
+
+		if ((link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) && (link->hpd_status == false))
 			continue;
 
 		dp_tunnel_settings = get_dp_tunnel_settings(new_ctx, stream);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
index 819bf2d8ba53..906d85ca8956 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
@@ -48,8 +48,7 @@
  */
 static bool link_dp_is_bw_alloc_available(struct dc_link *link)
 {
-	return (link && link->hpd_status
-		&& link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling
+	return (link && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling
 		&& link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dpia_bw_alloc
 		&& link->dpcd_caps.usb4_dp_tun_info.driver_bw_cap.bits.driver_bw_alloc_support);
 }
@@ -226,35 +225,35 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link)
 	bool ret = false;
 	uint8_t val;
 
-	if (link->hpd_status) {
-		val = DPTX_BW_ALLOC_MODE_ENABLE | DPTX_BW_ALLOC_UNMASK_IRQ;
+	val = DPTX_BW_ALLOC_MODE_ENABLE | DPTX_BW_ALLOC_UNMASK_IRQ;
 
-		if (core_link_write_dpcd(link, DPTX_BW_ALLOCATION_MODE_CONTROL, &val, sizeof(uint8_t)) == DC_OK) {
-			DC_LOG_DEBUG("%s:  link[%d] DPTX BW allocation mode enabled", __func__, link->link_index);
+	if (core_link_write_dpcd(link, DPTX_BW_ALLOCATION_MODE_CONTROL, &val, sizeof(uint8_t)) == DC_OK) {
+		DC_LOG_DEBUG("%s:  link[%d] DPTX BW allocation mode enabled", __func__, link->link_index);
 
-			retrieve_usb4_dp_bw_allocation_info(link);
+		retrieve_usb4_dp_bw_allocation_info(link);
 
-			if (link->dpia_bw_alloc_config.nrd_max_link_rate && link->dpia_bw_alloc_config.nrd_max_lane_count) {
-				link->reported_link_cap.link_rate = link->dpia_bw_alloc_config.nrd_max_link_rate;
-				link->reported_link_cap.lane_count = link->dpia_bw_alloc_config.nrd_max_lane_count;
-			}
+		if (
+				link->dpia_bw_alloc_config.nrd_max_link_rate
+				&& link->dpia_bw_alloc_config.nrd_max_lane_count) {
+			link->reported_link_cap.link_rate = link->dpia_bw_alloc_config.nrd_max_link_rate;
+			link->reported_link_cap.lane_count = link->dpia_bw_alloc_config.nrd_max_lane_count;
+		}
 
-			link->dpia_bw_alloc_config.bw_alloc_enabled = true;
-			ret = true;
+		link->dpia_bw_alloc_config.bw_alloc_enabled = true;
+		ret = true;
 
-			if (link->dc->debug.dpia_debug.bits.enable_usb4_bw_zero_alloc_patch) {
-				/*
-				 * During DP tunnel creation, the CM preallocates BW
-				 * and reduces the estimated BW of other DPIAs.
-				 * The CM releases the preallocation only when the allocation is complete.
-				 * Perform a zero allocation to make the CM release the preallocation
-				 * and correctly update the estimated BW for all DPIAs per host router.
-				 */
-				link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0);
-			}
-		} else
-			DC_LOG_DEBUG("%s:  link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index);
-	}
+		if (link->dc->debug.dpia_debug.bits.enable_usb4_bw_zero_alloc_patch) {
+			/*
+			 * During DP tunnel creation, the CM preallocates BW
+			 * and reduces the estimated BW of other DPIAs.
+			 * The CM releases the preallocation only when the allocation is complete.
+			 * Perform a zero allocation to make the CM release the preallocation
+			 * and correctly update the estimated BW for all DPIAs per host router.
+			 */
+			link_dp_dpia_allocate_usb4_bandwidth_for_stream(link, 0);
+		}
+	} else
+		DC_LOG_DEBUG("%s:  link[%d] failed to enable DPTX BW allocation mode", __func__, link->link_index);
 
 	return ret;
 }
@@ -297,15 +296,12 @@ void dpia_handle_usb4_bandwidth_allocation_for_link(struct dc_link *link, int pe
 {
 	if (link && link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.bits.dp_tunneling
 			&& link->dpia_bw_alloc_config.bw_alloc_enabled) {
-		//1. Hot Plug
-		if (link->hpd_status && peak_bw > 0) {
+		if (peak_bw > 0) {
 			// If DP over USB4 then we need to check BW allocation
 			link->dpia_bw_alloc_config.link_max_bw = peak_bw;
 
 			link_dpia_send_bw_alloc_request(link, peak_bw);
-		}
-		//2. Cold Unplug
-		else if (!link->hpd_status)
+		} else
 			dpia_bw_alloc_unplug(link);
 	}
 }

From 72b4aabdf1b0dd1096b55251e09ffe865e2e4877 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Fri, 11 Jul 2025 15:23:35 -0400
Subject: [PATCH 0135/1455] drm/amd/display: Add comma to last entry of enum
 for consistency

[Why&How]
Add comma to last entry of enum for consistency.

Reviewed-by: Jun Lei <jun.lei@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/inc/core_types.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index f896cce87b8d..4387de044469 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -434,7 +434,7 @@ enum p_state_switch_method {
 	P_STATE_V_ACTIVE,
 	P_STATE_SUB_VP,
 	P_STATE_DRR_SUB_VP,
-	P_STATE_V_BLANK_SUB_VP
+	P_STATE_V_BLANK_SUB_VP,
 };
 
 struct pipe_ctx {

From 2681bf4ae8d24df950138b8c9ea9c271cd62e414 Mon Sep 17 00:00:00 2001
From: Michael Strauss <michael.strauss@amd.com>
Date: Wed, 12 Feb 2025 14:08:08 -0500
Subject: [PATCH 0136/1455] drm/amd/display: Move setup_stream_attribute

[WHY]
If symclk RCO is enabled, stream encoder may not be receiving an ungated
clock by the time we attempt to set stream attributes when setting dpms
on. Since the clock is gated, register writes to the stream encoder fail.

[HOW]
Move set_stream_attribute call into enable_stream, just after the point
where symclk32_se is ungated.
Logically there is no need to set stream attributes as early as is
currently done in link_set_dpms_on, so this should have no impact beyond
the RCO fix.

Reviewed-by: Ovidiu (Ovi) Bunea <ovidiu.bunea@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c  | 1 +
 drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c    | 2 ++
 drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c  | 2 ++
 drivers/gpu/drm/amd/display/dc/link/link_dpms.c            | 3 ---
 .../drm/amd/display/dc/virtual/virtual_stream_encoder.c    | 7 +++++++
 5 files changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index c69194e04ff9..32fd6bdc18d7 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -671,6 +671,7 @@ void dce110_enable_stream(struct pipe_ctx *pipe_ctx)
 	uint32_t early_control = 0;
 	struct timing_generator *tg = pipe_ctx->stream_res.tg;
 
+	link_hwss->setup_stream_attribute(pipe_ctx);
 	link_hwss->setup_stream_encoder(pipe_ctx);
 
 	dc->hwss.update_info_frame(pipe_ctx);
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index 3207addbd4eb..7d24fa1517bf 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -3054,6 +3054,8 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx)
 						      link_enc->transmitter - TRANSMITTER_UNIPHY_A);
 	}
 
+	link_hwss->setup_stream_attribute(pipe_ctx);
+
 	if (dc->res_pool->dccg->funcs->set_pixel_rate_div)
 		dc->res_pool->dccg->funcs->set_pixel_rate_div(
 			dc->res_pool->dccg,
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index cc9f40d97af2..fb949aeb1244 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -965,6 +965,8 @@ void dcn401_enable_stream(struct pipe_ctx *pipe_ctx)
 		}
 	}
 
+	link_hwss->setup_stream_attribute(pipe_ctx);
+
 	if (dc->res_pool->dccg->funcs->set_pixel_rate_div) {
 		dc->res_pool->dccg->funcs->set_pixel_rate_div(
 			dc->res_pool->dccg,
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 8c8682f743d6..cb80b4599936 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -2458,7 +2458,6 @@ void link_set_dpms_on(
 	struct link_encoder *link_enc = pipe_ctx->link_res.dio_link_enc;
 	enum otg_out_mux_dest otg_out_dest = OUT_MUX_DIO;
 	struct vpg *vpg = pipe_ctx->stream_res.stream_enc->vpg;
-	const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res);
 	bool apply_edp_fast_boot_optimization =
 		pipe_ctx->stream->apply_edp_fast_boot_optimization;
 
@@ -2502,8 +2501,6 @@ void link_set_dpms_on(
 		pipe_ctx->stream_res.tg->funcs->set_out_mux(pipe_ctx->stream_res.tg, otg_out_dest);
 	}
 
-	link_hwss->setup_stream_attribute(pipe_ctx);
-
 	pipe_ctx->stream->apply_edp_fast_boot_optimization = false;
 
 	// Enable VPG before building infoframe
diff --git a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
index ad088d70e189..6ffc74fc9dcd 100644
--- a/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/virtual/virtual_stream_encoder.c
@@ -44,6 +44,11 @@ static void virtual_stream_encoder_dvi_set_stream_attribute(
 	struct dc_crtc_timing *crtc_timing,
 	bool is_dual_link) {}
 
+static void virtual_stream_encoder_lvds_set_stream_attribute(
+	struct stream_encoder *enc,
+	struct dc_crtc_timing *crtc_timing)
+{}
+
 static void virtual_stream_encoder_set_throttled_vcp_size(
 	struct stream_encoder *enc,
 	struct fixed31_32 avg_time_slots_per_mtp)
@@ -115,6 +120,8 @@ static const struct stream_encoder_funcs virtual_str_enc_funcs = {
 		virtual_stream_encoder_hdmi_set_stream_attribute,
 	.dvi_set_stream_attribute =
 		virtual_stream_encoder_dvi_set_stream_attribute,
+	.lvds_set_stream_attribute =
+		virtual_stream_encoder_lvds_set_stream_attribute,
 	.set_throttled_vcp_size =
 		virtual_stream_encoder_set_throttled_vcp_size,
 	.update_hdmi_info_packets =

From 82139760dcf7829ae2ca3f70442be9b53a0aff40 Mon Sep 17 00:00:00 2001
From: Jingwen Zhu <Jingwen.Zhu@amd.com>
Date: Mon, 14 Jul 2025 16:18:19 +0800
Subject: [PATCH 0137/1455] drm/amd/display: Add a config flag for
 limited_pll_vco

[Why/How]
Added a new config flag to pass to the DMUB during boot.
This workaround will solves black screen issue on reboot.

Reviewed-by: Hansen Dsouza <hansen.dsouza@amd.com>
Signed-off-by: Jingwen Zhu <Jingwen.Zhu@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h       | 1 +
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 3 ++-
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 0bafb6710761..87b761ac3135 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -316,6 +316,7 @@ struct dmub_srv_hw_params {
 	bool disable_sldo_opt;
 	bool enable_non_transparent_setconfig;
 	bool lower_hbr3_phy_ssc;
+	bool override_hbr3_pll_vco;
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index a89bf08ffd37..e2e5f71c03f2 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -843,7 +843,8 @@ union dmub_fw_boot_options {
 		uint32_t ips_sequential_ono: 1; /**< 1 to enable sequential ONO IPS sequence */
 		uint32_t disable_sldo_opt: 1; /**< 1 to disable SLDO optimizations */
 		uint32_t lower_hbr3_phy_ssc: 1; /**< 1 to lower hbr3 phy ssc to 0.125 percent */
-		uint32_t reserved : 6; /**< reserved */
+		uint32_t override_hbr3_pll_vco: 1; /**< 1 to override the hbr3 pll vco to 0 */
+		uint32_t reserved : 5; /**< reserved */
 	} bits; /**< boot bits */
 	uint32_t all; /**< 32-bit access to bits */
 };
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
index 3f38db752b84..4777c7203b2c 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
@@ -377,6 +377,7 @@ void dmub_dcn31_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmu
 	boot_options.bits.dpia_hpd_int_enable_supported = params->dpia_hpd_int_enable_supported;
 	boot_options.bits.power_optimization = params->power_optimization;
 	boot_options.bits.lower_hbr3_phy_ssc = params->lower_hbr3_phy_ssc;
+	boot_options.bits.override_hbr3_pll_vco = params->override_hbr3_pll_vco;
 
 	boot_options.bits.sel_mux_phy_c_d_phy_f_g = (dmub->asic == DMUB_ASIC_DCN31B) ? 1 : 0;
 

From b86cba11889ec4b784bfffc0a9424e1aa5eea56b Mon Sep 17 00:00:00 2001
From: Ovidiu Bunea <ovidiu.bunea@amd.com>
Date: Thu, 17 Jul 2025 10:41:41 -0400
Subject: [PATCH 0138/1455] drm/amd/display: Revert "Add a config flag for
 limited_pll_vco"

This reverts commit 82139760dcf7829ae2ca3f70442be9b53a0aff40.

[why & how]
DMUB header changes should be submitted to firmware branch first and
allowed to propagate to driver. Currently, this change breaks linux
builds so need to revert it until it's ready.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Ovidiu Bunea <ovidiu.bunea@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h       | 1 -
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 3 +--
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c | 1 -
 3 files changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 87b761ac3135..0bafb6710761 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -316,7 +316,6 @@ struct dmub_srv_hw_params {
 	bool disable_sldo_opt;
 	bool enable_non_transparent_setconfig;
 	bool lower_hbr3_phy_ssc;
-	bool override_hbr3_pll_vco;
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index e2e5f71c03f2..a89bf08ffd37 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -843,8 +843,7 @@ union dmub_fw_boot_options {
 		uint32_t ips_sequential_ono: 1; /**< 1 to enable sequential ONO IPS sequence */
 		uint32_t disable_sldo_opt: 1; /**< 1 to disable SLDO optimizations */
 		uint32_t lower_hbr3_phy_ssc: 1; /**< 1 to lower hbr3 phy ssc to 0.125 percent */
-		uint32_t override_hbr3_pll_vco: 1; /**< 1 to override the hbr3 pll vco to 0 */
-		uint32_t reserved : 5; /**< reserved */
+		uint32_t reserved : 6; /**< reserved */
 	} bits; /**< boot bits */
 	uint32_t all; /**< 32-bit access to bits */
 };
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
index 4777c7203b2c..3f38db752b84 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
@@ -377,7 +377,6 @@ void dmub_dcn31_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmu
 	boot_options.bits.dpia_hpd_int_enable_supported = params->dpia_hpd_int_enable_supported;
 	boot_options.bits.power_optimization = params->power_optimization;
 	boot_options.bits.lower_hbr3_phy_ssc = params->lower_hbr3_phy_ssc;
-	boot_options.bits.override_hbr3_pll_vco = params->override_hbr3_pll_vco;
 
 	boot_options.bits.sel_mux_phy_c_d_phy_f_g = (dmub->asic == DMUB_ASIC_DCN31B) ? 1 : 0;
 

From 2efe08a1a98bc797f12242beab9a3e3cc351d0b3 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Fri, 27 Jun 2025 16:58:26 -0500
Subject: [PATCH 0139/1455] drm/amd/display: Add missing SPDX license
 identifier

[Why]
All files should be properly classified.

[How]
Add missing SPDX-License-Identifier.

Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c            | 2 ++
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h            | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c      | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c        | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h        | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c    | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h    | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c       | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h       | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c    | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c        | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h        | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c  | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h  | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c     | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c        | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h        | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c     | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h     | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c   | 1 +
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h      | 1 +
 22 files changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index bfe3ee4b07ff..85f4a1bf5e6f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2015 Advanced Micro Devices, Inc.
  *
@@ -10395,6 +10396,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
 				if (amdgpu_dm_crc_window_is_activated(crtc)) {
 					uint8_t cnt;
+
 					spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 					for (cnt = 0; cnt < MAX_CRC_WINDOW_NUM; cnt++) {
 						if (acrtc->dm_irq_params.window_param[cnt].enable) {
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index b937da0a4e4a..67c3a7a967f2 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright (C) 2015-2020 Advanced Micro Devices, Inc. All rights reserved.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index ebabfe3a512f..ff97785b60a2 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2018 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
index 033bd817d871..e20aa7438066 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2015 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h
index 3da056c8d20b..95bdb8699d7f 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2019 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index b726bcd18e29..484691335fd6 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2018 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h
index 071200473c27..122cdc124b3b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2018 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
index c16962256514..4a973c85cb32 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2019 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h
index 69b445b011c8..4faa344f196e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2019 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
index 9e3e51a2dc49..fe100e4c9801 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2015 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
index b61e210f6246..a1c722112c22 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2015 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h
index ba17c23b2706..4f6b58f4f90d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2015 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
index 6c9de834455b..3c9995275cbd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq_params.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2020 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 7187d5aedf0a..137f18d41f1b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2012-15 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
index 600d6e221011..65f76a7d00db 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2012-15 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
index 848c5b4bb301..e5771f490f2e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2018 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
index ff7b867ae98b..de0a4968379b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2021 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h
index e2366321a3c1..4fb8626913cf 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2021 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c
index 41f07f13a7b5..82ea3fe5e764 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: MIT
 /*
  * Copyright 2023 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h
index 8126bdb1eb6b..73b6c67ae5e7 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.h
@@ -1,3 +1,4 @@
+/* SPDX-License-Identifier: MIT */
 /*
  * Copyright 2021 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
index 0005f5f8f34f..132de4071efd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
@@ -1,3 +1,4 @@
+//SPDX-License-Identifier: MIT
 /*
  * Copyright 2015 Advanced Micro Devices, Inc.
  *
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
index 95f890fda8aa..aa56fd6d56c3 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h
@@ -1,3 +1,4 @@
+//SPDX-License-Identifier: MIT
 /*
  * Copyright 2018 Advanced Micro Devices, Inc.
  *

From c491b1a2569e74ee57640494cc1ead2fa4072070 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Sun, 6 Jul 2025 09:55:58 -0500
Subject: [PATCH 0140/1455] drm/amd/display: Drop unused include

[Why]
Extra includes slow down compile time.

[How]
Drop unnecessary include.

Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
index de0a4968379b..fd491b7a3cd7 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c
@@ -27,7 +27,6 @@
 #include "amdgpu_dm_psr.h"
 #include "dc_dmub_srv.h"
 #include "dc.h"
-#include "dm_helpers.h"
 #include "amdgpu_dm.h"
 #include "modules/power/power_helpers.h"
 

From 29a5adc5e14c5f8682b0276ef44847f47f5fdf5e Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Fri, 27 Jun 2025 16:55:03 -0500
Subject: [PATCH 0141/1455] drm/amd/display: Remove unnecessary whitespace

[Why]
Whitespace before a new line is unnecessary.

[How]
Remove whitespace.

Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 85f4a1bf5e6f..2fde2065c5b6 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6389,13 +6389,15 @@ static void fill_stream_properties_from_drm_display_mode(
 							       (struct drm_connector *)connector,
 							       mode_in);
 		if (err < 0)
-			drm_warn_once(connector->dev, "Failed to setup avi infoframe on connector %s: %zd \n", connector->name, err);
+			drm_warn_once(connector->dev, "Failed to setup avi infoframe on connector %s: %zd\n",
+				      connector->name, err);
 		timing_out->vic = avi_frame.video_code;
 		err = drm_hdmi_vendor_infoframe_from_display_mode(&hv_frame,
 								  (struct drm_connector *)connector,
 								  mode_in);
 		if (err < 0)
-			drm_warn_once(connector->dev, "Failed to setup vendor infoframe on connector %s: %zd \n", connector->name, err);
+			drm_warn_once(connector->dev, "Failed to setup vendor infoframe on connector %s: %zd\n",
+				      connector->name, err);
 		timing_out->hdmi_vic = hv_frame.vic;
 	}
 

From 048bd938342835dd50363186cdb6f270a20f7aae Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Fri, 27 Jun 2025 16:53:39 -0500
Subject: [PATCH 0142/1455] drm/amd/display: Remove unnecessary includes

[Why]
Extra includes aren't needed for compilation and can slow down the
preprocessor.

[How]
Drop unneeded includes.

Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c    | 12 ------------
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c  |  1 -
 2 files changed, 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 2fde2065c5b6..c71167ffdb76 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -40,13 +40,11 @@
 #include "dc/dc_stat.h"
 #include "dc/dc_state.h"
 #include "amdgpu_dm_trace.h"
-#include "dpcd_defs.h"
 #include "link/protocols/link_dpcd.h"
 #include "link_service_types.h"
 #include "link/protocols/link_dp_capability.h"
 #include "link/protocols/link_ddc.h"
 
-#include "vid.h"
 #include "amdgpu.h"
 #include "amdgpu_display.h"
 #include "amdgpu_ucode.h"
@@ -57,7 +55,6 @@
 #include "amdgpu_dm_hdcp.h"
 #include <drm/display/drm_hdcp_helper.h>
 #include "amdgpu_dm_wb.h"
-#include "amdgpu_pm.h"
 #include "amdgpu_atombios.h"
 
 #include "amd_shared.h"
@@ -103,15 +100,6 @@
 
 #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h"
 
-#include "dcn/dcn_1_0_offset.h"
-#include "dcn/dcn_1_0_sh_mask.h"
-#include "soc15_hw_ip.h"
-#include "soc15_common.h"
-#include "vega10_ip_offset.h"
-
-#include "gc/gc_11_0_0_offset.h"
-#include "gc/gc_11_0_0_sh_mask.h"
-
 #include "modules/inc/mod_freesync.h"
 #include "modules/power/power_helpers.h"
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index ff97785b60a2..c7387af725d6 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -28,7 +28,6 @@
 #include "amdgpu_dm.h"
 #include "dc.h"
 #include "modules/color/color_gamma.h"
-#include "basics/conversion.h"
 
 /**
  * DOC: overview

From fa7cad4901931f0280a12f0bb375eda0294c6200 Mon Sep 17 00:00:00 2001
From: Duncan Ma <Duncan.Ma@amd.com>
Date: Wed, 4 Dec 2024 12:35:05 -0500
Subject: [PATCH 0143/1455] drm/amd/display: Add eDP AUXless ALPM

[Why & How]
Add AUX-less ALPM capability check and initialization

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Signed-off-by: Duncan Ma <Duncan.Ma@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/core/dc_link_exports.c |  7 ++++
 drivers/gpu/drm/amd/display/dc/dc.h           | 11 ++++++
 drivers/gpu/drm/amd/display/dc/dc_dp_types.h  | 16 +++++---
 drivers/gpu/drm/amd/display/dc/dc_types.h     |  8 ++++
 .../gpu/drm/amd/display/dc/dce/dmub_replay.c  | 13 +++++++
 drivers/gpu/drm/amd/display/dc/inc/link.h     |  5 ++-
 .../drm/amd/display/dc/link/link_factory.c    |  2 +
 .../dc/link/protocols/link_dp_capability.c    | 37 +++++++++++++++++++
 .../dc/link/protocols/link_dp_capability.h    |  6 +++
 .../link/protocols/link_edp_panel_control.c   |  8 +++-
 10 files changed, 105 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
index 130455f2802a..b7a5de4ecb61 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
@@ -520,3 +520,10 @@ enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, cons
 	return dc->link_srv->validate_dp_tunnel_bandwidth(dc, new_ctx);
 }
 
+void dc_link_get_alpm_support(struct dc_link *link,
+	bool *auxless_support,
+	bool *auxwake_support)
+{
+	link->dc->link_srv->edp_get_alpm_support(link, auxless_support, auxwake_support);
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 779b39e33443..fee54cc0f7d4 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1145,6 +1145,11 @@ struct dc_debug_options {
 	bool enable_hblank_borrow;
 	bool force_subvp_df_throttle;
 	uint32_t acpi_transition_bitmasks[MAX_PIPES];
+	unsigned int auxless_alpm_lfps_setup_ns;
+	unsigned int auxless_alpm_lfps_period_ns;
+	unsigned int auxless_alpm_lfps_silence_ns;
+	unsigned int auxless_alpm_lfps_t1t2_us;
+	short auxless_alpm_lfps_t1t2_offset_us;
 };
 
 
@@ -2447,6 +2452,12 @@ void dc_link_dp_dpia_handle_usb4_bandwidth_allocation_for_link(
  */
 enum dc_status dc_link_validate_dp_tunneling_bandwidth(const struct dc *dc, const struct dc_state *new_ctx);
 
+/*
+ * Get if ALPM is supported by the link
+ */
+void dc_link_get_alpm_support(struct dc_link *link, bool *auxless_support,
+	bool *auxwake_support);
+
 /* Sink Interfaces - A sink corresponds to a display output device */
 
 struct dc_container_id {
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index 5ce1be362534..3a3ec38cdf8b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -1021,7 +1021,8 @@ union dp_128b_132b_supported_lttpr_link_rates {
 union dp_alpm_lttpr_cap {
 	struct {
 		uint8_t AUX_LESS_ALPM_SUPPORTED	:1;
-		uint8_t RESERVED				:7;
+		uint8_t ASSR_SUPPORTED			:1;
+		uint8_t RESERVED			:6;
 	} bits;
 	uint8_t raw;
 };
@@ -1119,10 +1120,11 @@ union dp_128b_132b_training_aux_rd_interval {
 
 union edp_alpm_caps {
 	struct {
-		uint8_t AUX_WAKE_ALPM_CAP       :1;
-		uint8_t PM_STATE_2A_SUPPORT     :1;
-		uint8_t AUX_LESS_ALPM_CAP       :1;
-		uint8_t RESERVED                :5;
+		uint8_t AUX_WAKE_ALPM_CAP                               :1;
+		uint8_t PM_STATE_2A_SUPPORT                             :1;
+		uint8_t AUX_LESS_ALPM_CAP                               :1;
+		uint8_t AUX_LESS_ALPM_ML_PHY_SLEEP_STATUS_SUPPORTED     :1;
+		uint8_t RESERVED                                        :4;
 	} bits;
 	uint8_t raw;
 };
@@ -1347,7 +1349,9 @@ union dpcd_alpm_configuration {
 	struct {
 		unsigned char ENABLE                    : 1;
 		unsigned char IRQ_HPD_ENABLE            : 1;
-		unsigned char RESERVED                  : 6;
+		unsigned char ALPM_MODE_SEL             : 1;
+		unsigned char ACDS_PERIOD_DURATION      : 1;
+		unsigned char RESERVED                  : 4;
 	} bits;
 	unsigned char raw;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 2a86058c3bfa..2e2dea21b332 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -577,6 +577,12 @@ struct dc_plane_flip_time {
 	unsigned int prev_update_time_in_us;
 };
 
+enum dc_alpm_mode {
+	DC_ALPM_AUXWAKE = 0,
+	DC_ALPM_AUXLESS = 1,
+	DC_ALPM_UNSUPPORTED = 0xF,
+};
+
 enum dc_psr_state {
 	PSR_STATE0 = 0x0,
 	PSR_STATE1,
@@ -1143,6 +1149,8 @@ struct replay_config {
 	bool low_rr_supported;
 	/* Replay Video Conferencing Optimization Enabled */
 	bool replay_video_conferencing_optimization_enabled;
+	/* Replay alpm mode */
+	enum dc_alpm_mode alpm_mode;
 };
 
 /* Replay feature flags*/
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
index fcd3d86ad517..65b979617b0c 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
@@ -3,6 +3,7 @@
 // Copyright 2024 Advanced Micro Devices, Inc.
 
 #include "dc.h"
+#include "link.h"
 #include "dc_dmub_srv.h"
 #include "dmub/dmub_srv.h"
 #include "core_types.h"
@@ -189,6 +190,18 @@ static bool dmub_replay_copy_settings(struct dmub_replay *dmub,
 	else
 		copy_settings_data->flags.bitfields.force_wakeup_by_tps3 = 0;
 
+	copy_settings_data->flags.bitfields.alpm_mode = (enum dmub_alpm_mode)link->replay_settings.config.alpm_mode;
+	if (link->replay_settings.config.alpm_mode == DC_ALPM_AUXLESS) {
+		copy_settings_data->auxless_alpm_data.lfps_setup_ns = dc->dc->debug.auxless_alpm_lfps_setup_ns;
+		copy_settings_data->auxless_alpm_data.lfps_period_ns = dc->dc->debug.auxless_alpm_lfps_period_ns;
+		copy_settings_data->auxless_alpm_data.lfps_silence_ns = dc->dc->debug.auxless_alpm_lfps_silence_ns;
+		copy_settings_data->auxless_alpm_data.lfps_t1_t2_override_us =
+			dc->dc->debug.auxless_alpm_lfps_t1t2_us;
+		copy_settings_data->auxless_alpm_data.lfps_t1_t2_offset_us =
+			dc->dc->debug.auxless_alpm_lfps_t1t2_offset_us;
+		copy_settings_data->auxless_alpm_data.lttpr_count = link->dc->link_srv->dp_get_lttpr_count(link);
+	}
+
 	dc_wake_and_execute_dmub_cmd(dc, &cmd, DM_DMUB_WAIT_TYPE_WAIT);
 
 	return true;
diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link.h
index f2503402c10e..0cce49d95e26 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/link.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/link.h
@@ -218,7 +218,10 @@ struct link_service {
 	bool (*dp_overwrite_extended_receiver_cap)(struct dc_link *link);
 	enum lttpr_mode (*dp_decide_lttpr_mode)(struct dc_link *link,
 			struct dc_link_settings *link_setting);
-
+	uint8_t (*dp_get_lttpr_count)(struct dc_link *link);
+	void (*edp_get_alpm_support)(struct dc_link *link,
+		bool *auxless_support,
+		bool *auxwake_support);
 
 	/*************************** DP DPIA/PHY ******************************/
 	void (*dpia_handle_usb4_bandwidth_allocation_for_link)(
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.c b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
index de1143dbbd25..31a73867cd4c 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.c
@@ -165,6 +165,8 @@ static void construct_link_service_dp_capability(struct link_service *link_srv)
 	link_srv->dp_overwrite_extended_receiver_cap =
 			dp_overwrite_extended_receiver_cap;
 	link_srv->dp_decide_lttpr_mode = dp_decide_lttpr_mode;
+	link_srv->dp_get_lttpr_count = dp_get_lttpr_count;
+	link_srv->edp_get_alpm_support = edp_get_alpm_support;
 }
 
 /* link dp phy/dpia implements basic dp phy/dpia functionality such as
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index 651926e547b9..e0c4416993d9 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -2506,3 +2506,40 @@ bool dp_is_sink_present(struct dc_link *link)
 
 	return present;
 }
+
+uint8_t dp_get_lttpr_count(struct dc_link *link)
+{
+	if (dp_is_lttpr_present(link))
+		return dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
+
+	return 0;
+}
+
+void edp_get_alpm_support(struct dc_link *link,
+	bool *auxless_support,
+	bool *auxwake_support)
+{
+	bool lttpr_present = dp_is_lttpr_present(link);
+
+	if (auxless_support == NULL || auxwake_support == NULL)
+		return;
+
+	*auxless_support = false;
+	*auxwake_support = false;
+
+	if (!dc_is_embedded_signal(link->connector_signal))
+		return;
+
+	if (link->dpcd_caps.alpm_caps.bits.AUX_LESS_ALPM_CAP) {
+		if (lttpr_present) {
+			if (link->dpcd_caps.lttpr_caps.alpm.bits.AUX_LESS_ALPM_SUPPORTED)
+				*auxless_support = true;
+		} else
+			*auxless_support = true;
+	}
+
+	if (link->dpcd_caps.alpm_caps.bits.AUX_WAKE_ALPM_CAP) {
+		if (!lttpr_present)
+			*auxwake_support = true;
+	}
+}
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
index 940b147cc5d4..7170db5a1c13 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
@@ -108,4 +108,10 @@ uint32_t link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw);
 
 bool dp_overwrite_extended_receiver_cap(struct dc_link *link);
 
+uint8_t dp_get_lttpr_count(struct dc_link *link);
+
+void edp_get_alpm_support(struct dc_link *link,
+	bool *auxless_support,
+	bool *auxwake_support);
+
 #endif /* __DC_LINK_DP_CAPABILITY_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index 98ec9b5a559c..be714cbf6615 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -1042,7 +1042,13 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream
 			(uint8_t *)&(replay_config.raw), sizeof(uint8_t));
 
 		memset(&alpm_config, 0, sizeof(alpm_config));
-		alpm_config.bits.ENABLE = 1;
+		alpm_config.bits.ENABLE = link->replay_settings.config.alpm_mode != DC_ALPM_UNSUPPORTED ? 1 : 0;
+
+		if (link->replay_settings.config.alpm_mode == DC_ALPM_AUXLESS) {
+			alpm_config.bits.ALPM_MODE_SEL = 1;
+			alpm_config.bits.ACDS_PERIOD_DURATION = 1;
+		}
+
 		dm_helpers_dp_write_dpcd(
 			link->ctx,
 			link,

From 0352e5fab959208b9aba6c9c1020e813b8346472 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 16 Jul 2025 15:53:43 -0500
Subject: [PATCH 0144/1455] drm/amd/display: Rename dcn31 string shown to user

[Why]
DCN31 isn't a product, but DCN312 is. Matching against documentation users
might not understand the code.

[How]
Change DCN 3.1 string to be DCN 3.1.2.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_helper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index 7217de258851..51e41aed7316 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -732,7 +732,7 @@ char *dce_version_to_string(const int version)
 	case DCN_VERSION_3_03:
 		return "DCN 3.0.3";
 	case DCN_VERSION_3_1:
-		return "DCN 3.1";
+		return "DCN 3.1.2";
 	case DCN_VERSION_3_14:
 		return "DCN 3.1.4";
 	case DCN_VERSION_3_15:

From 9571dbaf98c8cb09ec1fa9721d94043b3ab7e740 Mon Sep 17 00:00:00 2001
From: "Chiang, Richard" <Richard.Chiang@amd.com>
Date: Tue, 15 Jul 2025 21:59:54 +0800
Subject: [PATCH 0145/1455] drm/amd/display: Remove update_planes_and_stream_v1
 sequence

[Why]/How]
Remove the update_planes_and_stream_v1 sequence to make the logic the same.

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Signed-off-by: Chiang, Richard <Richard.Chiang@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 129 +----------------------
 1 file changed, 2 insertions(+), 127 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index f760e930ea1b..cf3893a2f8ce 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -5102,129 +5102,6 @@ static bool fast_update_only(struct dc *dc,
 			&& !full_update_required(dc, srf_updates, surface_count, stream_update, stream);
 }
 
-static bool update_planes_and_stream_v1(struct dc *dc,
-		struct dc_surface_update *srf_updates, int surface_count,
-		struct dc_stream_state *stream,
-		struct dc_stream_update *stream_update,
-		struct dc_state *state)
-{
-	const struct dc_stream_status *stream_status;
-	enum surface_update_type update_type;
-	struct dc_state *context;
-	struct dc_context *dc_ctx = dc->ctx;
-	int i, j;
-	struct dc_fast_update fast_update[MAX_SURFACES] = {0};
-
-	dc_exit_ips_for_hw_access(dc);
-
-	populate_fast_updates(fast_update, srf_updates, surface_count, stream_update);
-	stream_status = dc_stream_get_status(stream);
-	context = dc->current_state;
-
-	update_type = dc_check_update_surfaces_for_stream(
-				dc, srf_updates, surface_count, stream_update, stream_status);
-	/* It is possible to receive a flip for one plane while there are multiple flip_immediate planes in the same stream.
-	 * E.g. Desktop and MPO plane are flip_immediate but only the MPO plane received a flip
-	 * Force the other flip_immediate planes to flip so GSL doesn't wait for a flip that won't come.
-	 */
-	force_immediate_gsl_plane_flip(dc, srf_updates, surface_count);
-
-	if (update_type >= UPDATE_TYPE_FULL) {
-
-		/* initialize scratch memory for building context */
-		context = dc_state_create_copy(state);
-		if (context == NULL) {
-			DC_ERROR("Failed to allocate new validate context!\n");
-			return false;
-		}
-
-		for (i = 0; i < dc->res_pool->pipe_count; i++) {
-			struct pipe_ctx *new_pipe = &context->res_ctx.pipe_ctx[i];
-			struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i];
-
-			if (new_pipe->plane_state && new_pipe->plane_state != old_pipe->plane_state)
-				new_pipe->plane_state->force_full_update = true;
-		}
-	} else if (update_type == UPDATE_TYPE_FAST) {
-		/*
-		 * Previous frame finished and HW is ready for optimization.
-		 */
-		dc_post_update_surfaces_to_stream(dc);
-	}
-
-	for (i = 0; i < surface_count; i++) {
-		struct dc_plane_state *surface = srf_updates[i].surface;
-
-		copy_surface_update_to_plane(surface, &srf_updates[i]);
-
-		if (update_type >= UPDATE_TYPE_MED) {
-			for (j = 0; j < dc->res_pool->pipe_count; j++) {
-				struct pipe_ctx *pipe_ctx =
-					&context->res_ctx.pipe_ctx[j];
-
-				if (pipe_ctx->plane_state != surface)
-					continue;
-
-				resource_build_scaling_params(pipe_ctx);
-			}
-		}
-	}
-
-	copy_stream_update_to_stream(dc, context, stream, stream_update);
-
-	if (update_type >= UPDATE_TYPE_FULL) {
-		if (dc->res_pool->funcs->validate_bandwidth(dc, context, DC_VALIDATE_MODE_AND_PROGRAMMING) != DC_OK) {
-			DC_ERROR("Mode validation failed for stream update!\n");
-			dc_state_release(context);
-			return false;
-		}
-	}
-
-	TRACE_DC_PIPE_STATE(pipe_ctx, i, MAX_PIPES);
-
-	if (fast_update_only(dc, fast_update, srf_updates, surface_count, stream_update, stream) &&
-			!dc->debug.enable_legacy_fast_update) {
-		commit_planes_for_stream_fast(dc,
-				srf_updates,
-				surface_count,
-				stream,
-				stream_update,
-				update_type,
-				context);
-	} else {
-		commit_planes_for_stream(
-				dc,
-				srf_updates,
-				surface_count,
-				stream,
-				stream_update,
-				update_type,
-				context);
-	}
-	/*update current_State*/
-	if (dc->current_state != context) {
-
-		struct dc_state *old = dc->current_state;
-
-		dc->current_state = context;
-		dc_state_release(old);
-
-		for (i = 0; i < dc->res_pool->pipe_count; i++) {
-			struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-
-			if (pipe_ctx->plane_state && pipe_ctx->stream == stream)
-				pipe_ctx->plane_state->force_full_update = false;
-		}
-	}
-
-	/* Legacy optimization path for DCE. */
-	if (update_type >= UPDATE_TYPE_FULL && dc_ctx->dce_version < DCE_VERSION_MAX) {
-		dc_post_update_surfaces_to_stream(dc);
-		TRACE_DCE_CLOCK_STATE(&context->bw_ctx.bw.dce);
-	}
-	return true;
-}
-
 static bool update_planes_and_stream_v2(struct dc *dc,
 		struct dc_surface_update *srf_updates, int surface_count,
 		struct dc_stream_state *stream,
@@ -5482,12 +5359,10 @@ void dc_commit_updates_for_stream(struct dc *dc,
 	if (dc->ctx->dce_version >= DCN_VERSION_4_01) {
 		ret = update_planes_and_stream_v3(dc, srf_updates, surface_count,
 				stream, stream_update);
-	} else if (dc->ctx->dce_version >= DCN_VERSION_3_2) {
+	} else {
 		ret = update_planes_and_stream_v2(dc, srf_updates, surface_count,
 				stream, stream_update);
-	} else
-		ret = update_planes_and_stream_v1(dc, srf_updates, surface_count, stream,
-				stream_update, state);
+	}
 
 	if (ret && dc->ctx->dce_version >= DCN_VERSION_3_2)
 		clear_update_flags(srf_updates, surface_count, stream);

From 7af124772fcf8422a55efa3a4d64484835db3beb Mon Sep 17 00:00:00 2001
From: Relja Vojvodic <rvojvodi@amd.com>
Date: Mon, 14 Jul 2025 11:56:50 -0400
Subject: [PATCH 0146/1455] drm/amd/display: Allow for sharing of some link and
 audio link functions

[Why&How]
Allow for sharing of some link and audio link functions by removing static
keyword from function definitions.
Expose those functions in the HWSEQ header.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Relja Vojvodic <rvojvodi@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c | 6 +++---
 drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h | 7 +++++++
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index 32fd6bdc18d7..153d68375fa3 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -1270,7 +1270,7 @@ void dce110_set_avmute(struct pipe_ctx *pipe_ctx, bool enable)
 		pipe_ctx->stream_res.stream_enc->funcs->set_avmute(pipe_ctx->stream_res.stream_enc, enable);
 }
 
-static enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id)
+enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id)
 {
 	switch (crtc_id) {
 	case CONTROLLER_ID_D0:
@@ -1290,7 +1290,7 @@ static enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id)
 	}
 }
 
-static void populate_audio_dp_link_info(
+void populate_audio_dp_link_info(
 	const struct pipe_ctx *pipe_ctx,
 	struct audio_dp_link_info *dp_link_info)
 {
@@ -2253,7 +2253,7 @@ static bool should_enable_fbc(struct dc *dc,
 /*
  *  Enable FBC
  */
-static void enable_fbc(
+void enable_fbc(
 		struct dc *dc,
 		struct dc_state *context)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h
index 7cd8c1576988..9c032e449481 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h
@@ -114,5 +114,12 @@ void build_audio_output(
 		struct dc_state *state,
 		const struct pipe_ctx *pipe_ctx,
 		struct audio_output *audio_output);
+enum audio_dto_source translate_to_dto_source(enum controller_id crtc_id);
+void populate_audio_dp_link_info(
+	const struct pipe_ctx *pipe_ctx,
+	struct audio_dp_link_info *dp_link_info);
+void enable_fbc(
+	struct dc *dc,
+	struct dc_state *context);
 #endif /* __DC_HWSS_DCE110_H__ */
 

From 9a885f1193efa4802504624b2df3aa5eaaa44a56 Mon Sep 17 00:00:00 2001
From: Cruise Hung <Cruise.Hung@amd.com>
Date: Tue, 15 Jul 2025 16:36:44 +0800
Subject: [PATCH 0147/1455] drm/amd/display: Add debug option to control BW
 Allocation mode

[Why & How]
Add debug option to control BW Allocation mode.

Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Reviewed-by: PeiChen (Pei-Chen) Huang <peichen.huang@amd.com>
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                          | 3 ++-
 .../gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c  | 5 +++++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index fee54cc0f7d4..57f4a9445b7e 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -839,7 +839,8 @@ union dpia_debug_options {
 		uint32_t enable_force_tbt3_work_around:1; /* bit 4 */
 		uint32_t disable_usb4_pm_support:1; /* bit 5 */
 		uint32_t enable_usb4_bw_zero_alloc_patch:1; /* bit 6 */
-		uint32_t reserved:25;
+		uint32_t enable_bw_allocation_mode:1; /* bit 7 */
+		uint32_t reserved:24;
 	} bits;
 	uint32_t raw;
 };
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
index 906d85ca8956..8a3c18ae97a7 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.c
@@ -225,6 +225,11 @@ bool link_dpia_enable_usb4_dp_bw_alloc_mode(struct dc_link *link)
 	bool ret = false;
 	uint8_t val;
 
+	if (link->dc->debug.dpia_debug.bits.enable_bw_allocation_mode == false) {
+		DC_LOG_DEBUG("%s:  link[%d] DPTX BW allocation mode disabled", __func__, link->link_index);
+		return false;
+	}
+
 	val = DPTX_BW_ALLOC_MODE_ENABLE | DPTX_BW_ALLOC_UNMASK_IRQ;
 
 	if (core_link_write_dpcd(link, DPTX_BW_ALLOCATION_MODE_CONTROL, &val, sizeof(uint8_t)) == DC_OK) {

From e3419e1e44b87d4176fb98679a77301b1ca40f63 Mon Sep 17 00:00:00 2001
From: Michael Strauss <michael.strauss@amd.com>
Date: Wed, 19 Mar 2025 18:04:01 -0400
Subject: [PATCH 0148/1455] drm/amd/display: Increase AUX Intra-Hop Done Max
 Wait Duration

[WHY]
In the worst case, AUX intra-hop done can take hundreds of milliseconds as
each retimer in a link might have to wait a full AUX_RD_INTERVAL to send
LT abort downstream.

[HOW]
Wait 300ms for each retimer in a link to allow time to propagate a LT abort
without infinitely waiting on intra-hop done.
For no-retimer case, keep the max duration at 10ms.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/protocols/link_dp_training.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
index 2dc1a660e504..134093ce5a8e 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
@@ -1018,7 +1018,12 @@ static enum link_training_result dpcd_exit_training_mode(struct dc_link *link, e
 {
 	enum dc_status status;
 	uint8_t sink_status = 0;
-	uint8_t i;
+	uint32_t i;
+	uint8_t lttpr_count = dp_parse_lttpr_repeater_count(link->dpcd_caps.lttpr_caps.phy_repeater_cnt);
+	uint32_t intra_hop_disable_time_ms = (lttpr_count > 0 ? lttpr_count * 300 : 10);
+
+	// Each hop could theoretically take over 256ms (max 128b/132b AUX RD INTERVAL)
+	// To be safe, allow 300ms per LTTPR and 10ms for no LTTPR case
 
 	/* clear training pattern set */
 	status = dpcd_set_training_pattern(link, DP_TRAINING_PATTERN_VIDEOIDLE);
@@ -1028,7 +1033,7 @@ static enum link_training_result dpcd_exit_training_mode(struct dc_link *link, e
 
 	if (encoding == DP_128b_132b_ENCODING) {
 		/* poll for intra-hop disable */
-		for (i = 0; i < 10; i++) {
+		for (i = 0; i < intra_hop_disable_time_ms; i++) {
 			if ((core_link_read_dpcd(link, DP_SINK_STATUS, &sink_status, 1) == DC_OK) &&
 					(sink_status & DP_INTRA_HOP_AUX_REPLY_INDICATION) == 0)
 				break;

From 640cd296f5877d20db9d4fbf6441eee114c1cb97 Mon Sep 17 00:00:00 2001
From: Ray Wu <ray.wu@amd.com>
Date: Wed, 11 Jun 2025 14:02:25 +0800
Subject: [PATCH 0149/1455] drm/amd/display: Add Replay residency in debugfs

[Why]
Users can access the replay residency to get PHY off percentage

[How]
Start capture residency:
sudo echo 1 /sys/kernel/debug/dri/0/eDP-1/replay_residency

Stop and Get replay residency:
sudo cat /sys/kernel/debug/dri/0/eDP-1/replay_residency

Reviewed-by: ChiaHsuan (Tom) Chung <chiahsuan.chung@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 34 ++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
index 484691335fd6..f263e1a4537e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c
@@ -3106,6 +3106,35 @@ static int replay_get_state(void *data, u64 *val)
 	return 0;
 }
 
+/*
+ *  Start / Stop capture Replay residency
+ */
+static int replay_set_residency(void *data, u64 val)
+{
+	struct amdgpu_dm_connector *connector = data;
+	struct dc_link *link = connector->dc_link;
+	bool is_start = (val != 0);
+	u32 residency = 0;
+
+	link->dc->link_srv->edp_replay_residency(link, &residency, is_start, PR_RESIDENCY_MODE_PHY);
+	return 0;
+}
+
+/*
+ *  Read Replay residency
+ */
+static int replay_get_residency(void *data, u64 *val)
+{
+	struct amdgpu_dm_connector *connector = data;
+	struct dc_link *link = connector->dc_link;
+	u32 residency = 0;
+
+	link->dc->link_srv->edp_replay_residency(link, &residency, false, PR_RESIDENCY_MODE_PHY);
+	*val = (u64)residency;
+
+	return 0;
+}
+
 /*
  *  Read PSR state
  */
@@ -3325,7 +3354,8 @@ DEFINE_DEBUGFS_ATTRIBUTE(dmcub_trace_event_state_fops, dmcub_trace_event_state_g
 			 dmcub_trace_event_state_set, "%llu\n");
 
 DEFINE_DEBUGFS_ATTRIBUTE(replay_state_fops, replay_get_state, NULL, "%llu\n");
-
+DEFINE_DEBUGFS_ATTRIBUTE(replay_residency_fops, replay_get_residency, replay_set_residency,
+			 "%llu\n");
 DEFINE_DEBUGFS_ATTRIBUTE(psr_fops, psr_get, NULL, "%llu\n");
 DEFINE_DEBUGFS_ATTRIBUTE(psr_residency_fops, psr_read_residency, NULL,
 			 "%llu\n");
@@ -3503,6 +3533,8 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector)
 		debugfs_create_file("replay_capability", 0444, dir, connector,
 					&replay_capability_fops);
 		debugfs_create_file("replay_state", 0444, dir, connector, &replay_state_fops);
+		debugfs_create_file_unsafe("replay_residency", 0444, dir,
+					   connector, &replay_residency_fops);
 		debugfs_create_file_unsafe("psr_capability", 0444, dir, connector, &psr_capability_fops);
 		debugfs_create_file_unsafe("psr_state", 0444, dir, connector, &psr_fops);
 		debugfs_create_file_unsafe("psr_residency", 0444, dir,

From 327aba7f558187e451636c77a1662a2858438dc9 Mon Sep 17 00:00:00 2001
From: Ovidiu Bunea <ovidiu.bunea@amd.com>
Date: Tue, 15 Jul 2025 17:26:39 -0400
Subject: [PATCH 0150/1455] drm/amd/display: Fix dmub_cmd header alignment

[why & how]
Header misalignment in struct dmub_cmd_replay_copy_settings_data and
struct dmub_alpm_auxless_data causes incorrect data read between driver
and dmub.
Fix the misalignment and ensure that everything is aligned to 4-byte
boundaries.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Ovidiu Bunea <ovidiu.bunea@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 20 +++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index a89bf08ffd37..dcae768c2cf4 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -4048,6 +4048,10 @@ struct dmub_alpm_auxless_data {
 	uint16_t lfps_t1_t2_override_us;
 	short lfps_t1_t2_offset_us;
 	uint8_t lttpr_count;
+	/*
+	 * Padding to align structure to 4 byte boundary.
+	 */
+	uint8_t pad[1];
 };
 
 /**
@@ -4080,14 +4084,6 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * DIG BE HW instance.
 	 */
 	uint8_t digbe_inst;
-	/**
-	 * @hpo_stream_enc_inst: HPO stream encoder instance
-	 */
-	uint8_t hpo_stream_enc_inst;
-	/**
-	 * @hpo_link_enc_inst: HPO link encoder instance
-	 */
-	uint8_t hpo_link_enc_inst;
 	/**
 	 * AUX HW instance.
 	 */
@@ -4132,6 +4128,14 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * Use for AUX-less ALPM LFPS wake operation
 	 */
 	struct dmub_alpm_auxless_data auxless_alpm_data;
+	/**
+	 * @hpo_stream_enc_inst: HPO stream encoder instance
+	 */
+	uint8_t hpo_stream_enc_inst;
+	/**
+	 * @hpo_link_enc_inst: HPO link encoder instance
+	 */
+	uint8_t hpo_link_enc_inst;
 	/**
 	 * @pad: Align structure to 4 byte boundary.
 	 */

From f5b69101f956f5b89605a13cb15f093a7906f2a1 Mon Sep 17 00:00:00 2001
From: Michael Strauss <michael.strauss@amd.com>
Date: Thu, 17 Jul 2025 16:18:58 -0400
Subject: [PATCH 0151/1455] drm/amd/display: Cache streams targeting link when
 performing LT automation

[WHY]
Last LT automation update can cause crash by referencing current_state and
calling into dc_update_planes_and_stream which may clobber current_state.

[HOW]
Cache relevant stream pointers and iterate through them instead of relying
on the current_state.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/link/accessories/link_dp_cts.c  | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
index b12d61701d4d..23f41c99fa38 100644
--- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
+++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
@@ -76,6 +76,9 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 	uint8_t count;
 	int i;
 
+	struct dc_stream_state *streams_on_link[MAX_PIPES];
+	int num_streams_on_link = 0;
+
 	needs_divider_update = (link->dc->link_srv->dp_get_encoding_format(link_setting) !=
 	link->dc->link_srv->dp_get_encoding_format((const struct dc_link_settings *) &link->cur_link_settings));
 
@@ -138,12 +141,19 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 		pipes[i]->stream_res.tg->funcs->enable_crtc(pipes[i]->stream_res.tg);
 
 	// Set DPMS on with stream update
-	for (i = 0; i < state->stream_count; i++)
-		if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link) {
-			stream_update.stream = state->streams[i];
+	// Cache all streams on current link since dc_update_planes_and_stream might kill current_state
+	for (i = 0; i < MAX_PIPES; i++) {
+		if (state->streams[i] && state->streams[i]->link && state->streams[i]->link == link)
+			streams_on_link[num_streams_on_link++] = state->streams[i];
+	}
+
+	for (i = 0; i < num_streams_on_link; i++) {
+		if (streams_on_link[i] && streams_on_link[i]->link && streams_on_link[i]->link == link) {
+			stream_update.stream = streams_on_link[i];
 			stream_update.dpms_off = &dpms_off;
-			dc_update_planes_and_stream(state->clk_mgr->ctx->dc, NULL, 0, state->streams[i], &stream_update);
+			dc_update_planes_and_stream(state->clk_mgr->ctx->dc, NULL, 0, streams_on_link[i], &stream_update);
 		}
+	}
 }
 
 static void dp_test_send_link_training(struct dc_link *link)

From c5fc24f126786dddcfdb87ac24afbecb1e4dd30d Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 18 Jul 2025 19:26:12 -0500
Subject: [PATCH 0152/1455] drm/amd/display: Promote DAL to 3.2.343

Summary:
* Fix caching streams for LT automation
* Fix DMUB command alignment
* Disabling DSC power gating on DCN314
* Add debugfs for Replay
* Add debug option for BW allocation mode
* Removal of unnecessary includes for faster compilation
* Refactor of code, including adding SPDX license to amdgpu_dm

Acked-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 57f4a9445b7e..5653c1673aec 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.342"
+#define DC_VER "3.2.343"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 216e4cff54c8719b2aed2f60833e1e84e041c574 Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Thu, 17 Jul 2025 11:39:58 -0400
Subject: [PATCH 0153/1455] drm/amdgpu: Add chain runlists support to GC9.4.2

Starting from MEC v97, GC 9.4.2 supports chain runlists of XNACK+/XNACK-
processes.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Philip Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c   |  3 +++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c | 12 ++++++++++++
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h |  1 +
 3 files changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 20b30f4b3c7d..6a9cf3587cc6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -2650,6 +2650,9 @@ static void gfx_v9_0_init_sq_config(struct amdgpu_device *adev)
 				!READ_ONCE(adev->barrier_has_auto_waitcnt));
 		WREG32_SOC15(GC, 0, mmSQ_CONFIG, tmp);
 		break;
+	case IP_VERSION(9, 4, 2):
+		gfx_v9_4_2_init_sq(adev);
+		break;
 	default:
 		break;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
index c48cd47b531f..8058ea91ecaf 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.c
@@ -748,6 +748,18 @@ void gfx_v9_4_2_init_golden_registers(struct amdgpu_device *adev,
 	}
 }
 
+void gfx_v9_4_2_init_sq(struct amdgpu_device *adev)
+{
+	uint32_t data;
+
+	if (adev->gfx.mec_fw_version >= 98) {
+		adev->gmc.xnack_flags |= AMDGPU_GMC_XNACK_FLAG_CHAIN;
+		data = RREG32_SOC15(GC, 0, regSQ_CONFIG1);
+		data = REG_SET_FIELD(data, SQ_CONFIG1, DISABLE_XNACK_CHECK_IN_RETRY_DISABLE, 1);
+		WREG32_SOC15(GC, 0, regSQ_CONFIG1, data);
+	}
+}
+
 void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev,
 				uint32_t first_vmid,
 				uint32_t last_vmid)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h
index 7584624b641c..a603724c1dfc 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2.h
@@ -28,6 +28,7 @@ void gfx_v9_4_2_debug_trap_config_init(struct amdgpu_device *adev,
 				uint32_t first_vmid, uint32_t last_vmid);
 void gfx_v9_4_2_init_golden_registers(struct amdgpu_device *adev,
 				      uint32_t die_id);
+void gfx_v9_4_2_init_sq(struct amdgpu_device *adev);
 void gfx_v9_4_2_set_power_brake_sequence(struct amdgpu_device *adev);
 int gfx_v9_4_2_do_edc_gpr_workarounds(struct amdgpu_device *adev);
 

From c2aa3089ad7e7fec3ec4a58d8d0904b5e9b392a1 Mon Sep 17 00:00:00 2001
From: Yann Dirson <ydirson@free.fr>
Date: Sun, 20 Jul 2025 16:13:16 +0200
Subject: [PATCH 0154/1455] Documentation/amdgpu: fix 'in the amdgfx'
 formulation

Clarify the mailing list.

Signed-off-by: Yann Dirson <ydirson@free.fr>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/display/dc-glossary.rst          | 2 +-
 Documentation/gpu/amdgpu/display/display-contributing.rst | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/gpu/amdgpu/display/dc-glossary.rst b/Documentation/gpu/amdgpu/display/dc-glossary.rst
index 7dc034e9e586..cbe737d1fcea 100644
--- a/Documentation/gpu/amdgpu/display/dc-glossary.rst
+++ b/Documentation/gpu/amdgpu/display/dc-glossary.rst
@@ -5,7 +5,7 @@ DC Glossary
 On this page, we try to keep track of acronyms related to the display
 component. If you do not find what you are looking for, look at the
 'Documentation/gpu/amdgpu/amdgpu-glossary.rst'; if you cannot find it anywhere,
-consider asking in the amdgfx and update this page.
+consider asking on the amd-gfx mailing list and update this page.
 
 .. glossary::
 
diff --git a/Documentation/gpu/amdgpu/display/display-contributing.rst b/Documentation/gpu/amdgpu/display/display-contributing.rst
index 36f3077eee00..2f741c52dce5 100644
--- a/Documentation/gpu/amdgpu/display/display-contributing.rst
+++ b/Documentation/gpu/amdgpu/display/display-contributing.rst
@@ -9,8 +9,8 @@ contribution to the display code, and for that, we say thank you :)
 
 This page summarizes some of the issues you can help with; keep in mind that
 this is a static page, and it is always a good idea to try to reach developers
-in the amdgfx or some of the maintainers. Finally, this page follows the DRM
-way of creating a TODO list; for more information, check
+on the amd-gfx mailing list or some of the maintainers. Finally, this page
+follows the DRM way of creating a TODO list; for more information, check
 'Documentation/gpu/todo.rst'.
 
 Gitlab issues

From 9f1f7cd467358e27205af0076e2f7b435a588893 Mon Sep 17 00:00:00 2001
From: Yann Dirson <ydirson@free.fr>
Date: Sun, 20 Jul 2025 16:13:17 +0200
Subject: [PATCH 0155/1455] drm/amdgpu: fix module parameter description

Fix dcdebugmask description.

Signed-off-by: Yann Dirson <ydirson@free.fr>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 395c6be901ce..efb54c23ba17 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -886,7 +886,7 @@ module_param_named(dcfeaturemask, amdgpu_dc_feature_mask, uint, 0444);
 
 /**
  * DOC: dcdebugmask (uint)
- * Override display features enabled. See enum DC_DEBUG_MASK in drivers/gpu/drm/amd/include/amd_shared.h.
+ * Display debug options. See enum DC_DEBUG_MASK in drivers/gpu/drm/amd/include/amd_shared.h.
  */
 MODULE_PARM_DESC(dcdebugmask, "all debug options disabled (default))");
 module_param_named(dcdebugmask, amdgpu_dc_debug_mask, uint, 0444);

From 45fbb51050e72723c2bdcedc1ce32305256c70ed Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Fri, 25 Jul 2025 19:43:37 -0700
Subject: [PATCH 0156/1455] drm/xe/guc: Add more GuC load error status codes

The GuC load process will abort if certain status codes (which are
indicative of a fatal error) are reported. Otherwise, it keeps waiting
until the 'success' code is returned. New error codes have been added
in recent GuC releases, so add support for aborting on those as well.

v2: Shuffle HWCONFIG_START to the front of the switch to keep the
ordering as per the enum define for clarity (review feedback by
Jonathan). Also add a description for the basic 'invalid init data'
code which was missing.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250726024337.4056272-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/abi/guc_errors_abi.h |  3 +++
 drivers/gpu/drm/xe/xe_guc.c             | 19 +++++++++++++++++--
 2 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/abi/guc_errors_abi.h b/drivers/gpu/drm/xe/abi/guc_errors_abi.h
index ecf748fd87df..ad76b4baf42e 100644
--- a/drivers/gpu/drm/xe/abi/guc_errors_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_errors_abi.h
@@ -63,6 +63,7 @@ enum xe_guc_load_status {
 	XE_GUC_LOAD_STATUS_HWCONFIG_START                   = 0x05,
 	XE_GUC_LOAD_STATUS_HWCONFIG_DONE                    = 0x06,
 	XE_GUC_LOAD_STATUS_HWCONFIG_ERROR                   = 0x07,
+	XE_GUC_LOAD_STATUS_BOOTROM_VERSION_MISMATCH         = 0x08,
 	XE_GUC_LOAD_STATUS_GDT_DONE                         = 0x10,
 	XE_GUC_LOAD_STATUS_IDT_DONE                         = 0x20,
 	XE_GUC_LOAD_STATUS_LAPIC_DONE                       = 0x30,
@@ -75,6 +76,8 @@ enum xe_guc_load_status {
 	XE_GUC_LOAD_STATUS_INVALID_INIT_DATA_RANGE_START,
 	XE_GUC_LOAD_STATUS_MPU_DATA_INVALID                 = 0x73,
 	XE_GUC_LOAD_STATUS_INIT_MMIO_SAVE_RESTORE_INVALID   = 0x74,
+	XE_GUC_LOAD_STATUS_KLV_WORKAROUND_INIT_ERROR        = 0x75,
+	XE_GUC_LOAD_STATUS_INVALID_FTR_FLAG                 = 0x76,
 	XE_GUC_LOAD_STATUS_INVALID_INIT_DATA_RANGE_END,
 
 	XE_GUC_LOAD_STATUS_READY                            = 0xF0,
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index b1d1d6da3758..9e34401e4489 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -992,11 +992,14 @@ static int guc_load_done(u32 status)
 	case XE_GUC_LOAD_STATUS_GUC_PREPROD_BUILD_MISMATCH:
 	case XE_GUC_LOAD_STATUS_ERROR_DEVID_INVALID_GUCTYPE:
 	case XE_GUC_LOAD_STATUS_HWCONFIG_ERROR:
+	case XE_GUC_LOAD_STATUS_BOOTROM_VERSION_MISMATCH:
 	case XE_GUC_LOAD_STATUS_DPC_ERROR:
 	case XE_GUC_LOAD_STATUS_EXCEPTION:
 	case XE_GUC_LOAD_STATUS_INIT_DATA_INVALID:
 	case XE_GUC_LOAD_STATUS_MPU_DATA_INVALID:
 	case XE_GUC_LOAD_STATUS_INIT_MMIO_SAVE_RESTORE_INVALID:
+	case XE_GUC_LOAD_STATUS_KLV_WORKAROUND_INIT_ERROR:
+	case XE_GUC_LOAD_STATUS_INVALID_FTR_FLAG:
 		return -1;
 	}
 
@@ -1136,17 +1139,29 @@ static void guc_wait_ucode(struct xe_guc *guc)
 		}
 
 		switch (ukernel) {
+		case XE_GUC_LOAD_STATUS_HWCONFIG_START:
+			xe_gt_err(gt, "still extracting hwconfig table.\n");
+			break;
+
 		case XE_GUC_LOAD_STATUS_EXCEPTION:
 			xe_gt_err(gt, "firmware exception. EIP: %#x\n",
 				  xe_mmio_read32(mmio, SOFT_SCRATCH(13)));
 			break;
 
+		case XE_GUC_LOAD_STATUS_INIT_DATA_INVALID:
+			xe_gt_err(gt, "illegal init/ADS data\n");
+			break;
+
 		case XE_GUC_LOAD_STATUS_INIT_MMIO_SAVE_RESTORE_INVALID:
 			xe_gt_err(gt, "illegal register in save/restore workaround list\n");
 			break;
 
-		case XE_GUC_LOAD_STATUS_HWCONFIG_START:
-			xe_gt_err(gt, "still extracting hwconfig table.\n");
+		case XE_GUC_LOAD_STATUS_KLV_WORKAROUND_INIT_ERROR:
+			xe_gt_err(gt, "illegal workaround KLV data\n");
+			break;
+
+		case XE_GUC_LOAD_STATUS_INVALID_FTR_FLAG:
+			xe_gt_err(gt, "illegal feature flag specified\n");
 			break;
 		}
 

From 4df0bd5eb497c59e14924452026d6d70505706b5 Mon Sep 17 00:00:00 2001
From: Priyanka Dandamudi <priyanka.dandamudi@intel.com>
Date: Mon, 28 Jul 2025 10:03:36 +0530
Subject: [PATCH 0157/1455] drm/xe/uapi: Add documentation for
 DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING

Add documentation for drm_xe_gem_create structure flag
DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING.

v2: Modified to be in a more generalised way.

Signed-off-by: Priyanka Dandamudi <priyanka.dandamudi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250728043336.3319521-1-priyanka.dandamudi@intel.com
Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
---
 include/uapi/drm/xe_drm.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index e2426413488f..c721e130c1d2 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -760,7 +760,11 @@ struct drm_xe_device_query {
  * gem creation
  *
  * The @flags can be:
- *  - %DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING
+ *  - %DRM_XE_GEM_CREATE_FLAG_DEFER_BACKING - Modify the GEM object
+ *    allocation strategy by deferring physical memory allocation
+ *    until the object is either bound to a virtual memory region via
+ *    VM_BIND or accessed by the CPU. As a result, no backing memory is
+ *    reserved at the time of GEM object creation.
  *  - %DRM_XE_GEM_CREATE_FLAG_SCANOUT
  *  - %DRM_XE_GEM_CREATE_FLAG_NEEDS_VISIBLE_VRAM - When using VRAM as a
  *    possible placement, ensure that the corresponding VRAM allocation

From b116bd3ad3a7b9ee3144ab70497b25b915b904c3 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 28 Jul 2025 13:21:13 +0300
Subject: [PATCH 0158/1455] drm/i915/display: remove superfluous
 <linux/types.h> includes

Commit f7a9dc796567 ("drm/i915/scaler: Use intel_display as argument to
skl_scaler_max_src_size") added superfluous includes. Remove them.

Cc: Suraj Kandpal <suraj.kandpal@intel.com>
Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250728102113.238730-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_pfit.h | 2 --
 drivers/gpu/drm/i915/display/skl_scaler.h | 2 --
 2 files changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_pfit.h b/drivers/gpu/drm/i915/display/intel_pfit.h
index 2f075748a34e..c1bb0d1f344e 100644
--- a/drivers/gpu/drm/i915/display/intel_pfit.h
+++ b/drivers/gpu/drm/i915/display/intel_pfit.h
@@ -6,8 +6,6 @@
 #ifndef __INTEL_PFIT_H__
 #define __INTEL_PFIT_H__
 
-#include <linux/types.h>
-
 enum drm_mode_status;
 struct drm_display_mode;
 struct drm_connector_state;
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h
index 692716dd7616..e3b35d2c13be 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.h
+++ b/drivers/gpu/drm/i915/display/skl_scaler.h
@@ -5,8 +5,6 @@
 #ifndef INTEL_SCALER_H
 #define INTEL_SCALER_H
 
-#include <linux/types.h>
-
 enum drm_mode_status;
 struct drm_display_mode;
 struct intel_atomic_state;

From d6a0311c37b0955d3b2e15137f40c65c9aeb8f20 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 25 Jul 2025 11:05:08 +0200
Subject: [PATCH 0159/1455] drm/xe/hw_engine_group: Don't use drm_warn to catch
 missed case

Since hwe->class is an enumeration we can rely on the compiler
to catch any unhandled engine class case at compile time thanks
to [-Werror=switch]. Any unexpected use of a special CLASS_MAX
enum case can be guarded by our xe_gt_assert() instead, which
will be compiled-out on the production builds.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Francois Dugast <francois.dugast@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Francois Dugast <francois.dugast@intel.com>
Link: https://lore.kernel.org/r/20250725090508.571-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_hw_engine_group.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_hw_engine_group.c b/drivers/gpu/drm/xe/xe_hw_engine_group.c
index c926f840c87b..58bee3ffe881 100644
--- a/drivers/gpu/drm/xe/xe_hw_engine_group.c
+++ b/drivers/gpu/drm/xe/xe_hw_engine_group.c
@@ -103,8 +103,8 @@ int xe_hw_engine_setup_groups(struct xe_gt *gt)
 			break;
 		case XE_ENGINE_CLASS_OTHER:
 			break;
-		default:
-			drm_warn(&xe->drm, "NOT POSSIBLE");
+		case XE_ENGINE_CLASS_MAX:
+			xe_gt_assert(gt, false);
 		}
 	}
 

From d94a2a00d2b8878678607c2969fee3b4e59126cb Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Mon, 21 Jul 2025 19:53:08 -0600
Subject: [PATCH 0160/1455] drm: Create mipi_dsi_dual* macros

Create mipi_dsi_dual, mipi_dsi_dual_dcs_write_seq_multi, and
mipi_dsi_dual_generic_write_seq_multi macros for panels which are driven
by two parallel serial interfaces. This allows for the reduction of code
duplication in drivers for these panels.

Remove mipi_dsi_dual_dcs_write_seq_multi definition from
panel-novatek-nt36523.c to avoid the duplicate definition. Make novatek
driver pass mipi_dsi_context struct as a pointer.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250722015313.561966-2-me@brighamcampbell.com
---
 drivers/gpu/drm/drm_mipi_dsi.c                |  48 ++
 drivers/gpu/drm/panel/panel-novatek-nt36523.c | 804 +++++++++---------
 include/drm/drm_mipi_dsi.h                    |  95 +++
 3 files changed, 541 insertions(+), 406 deletions(-)

diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index a00d76443128..5b387543de28 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -827,6 +827,30 @@ void mipi_dsi_generic_write_multi(struct mipi_dsi_multi_context *ctx,
 }
 EXPORT_SYMBOL(mipi_dsi_generic_write_multi);
 
+/**
+ * mipi_dsi_dual_generic_write_multi() - mipi_dsi_generic_write_multi() for
+ * two dsi channels, one after the other
+ * @ctx: Context for multiple DSI transactions
+ * @dsi1: First dsi channel to write buffer to
+ * @dsi2: Second dsi channel to write buffer to
+ * @payload: Buffer containing the payload
+ * @size: Size of payload buffer
+ *
+ * A wrapper around mipi_dsi_generic_write_multi() that allows the user to
+ * conveniently write to two dsi channels, one after the other.
+ */
+void mipi_dsi_dual_generic_write_multi(struct mipi_dsi_multi_context *ctx,
+				       struct mipi_dsi_device *dsi1,
+				       struct mipi_dsi_device *dsi2,
+				       const void *payload, size_t size)
+{
+	ctx->dsi = dsi1;
+	mipi_dsi_generic_write_multi(ctx, payload, size);
+	ctx->dsi = dsi2;
+	mipi_dsi_generic_write_multi(ctx, payload, size);
+}
+EXPORT_SYMBOL(mipi_dsi_dual_generic_write_multi);
+
 /**
  * mipi_dsi_generic_read() - receive data using a generic read packet
  * @dsi: DSI peripheral device
@@ -1006,6 +1030,30 @@ void mipi_dsi_dcs_write_buffer_multi(struct mipi_dsi_multi_context *ctx,
 }
 EXPORT_SYMBOL(mipi_dsi_dcs_write_buffer_multi);
 
+/**
+ * mipi_dsi_dual_dcs_write_buffer_multi - mipi_dsi_dcs_write_buffer_multi() for
+ * two dsi channels, one after the other
+ * @ctx: Context for multiple DSI transactions
+ * @dsi1: First dsi channel to write buffer to
+ * @dsi2: Second dsi channel to write buffer to
+ * @data: Buffer containing data to be transmitted
+ * @len: Size of transmission buffer
+ *
+ * A wrapper around mipi_dsi_dcs_write_buffer_multi() that allows the user to
+ * conveniently write to two dsi channels, one after the other.
+ */
+void mipi_dsi_dual_dcs_write_buffer_multi(struct mipi_dsi_multi_context *ctx,
+					  struct mipi_dsi_device *dsi1,
+					  struct mipi_dsi_device *dsi2,
+					  const void *data, size_t len)
+{
+	ctx->dsi = dsi1;
+	mipi_dsi_dcs_write_buffer_multi(ctx, data, len);
+	ctx->dsi = dsi2;
+	mipi_dsi_dcs_write_buffer_multi(ctx, data, len);
+}
+EXPORT_SYMBOL(mipi_dsi_dual_dcs_write_buffer_multi);
+
 /**
  * mipi_dsi_dcs_write() - send DCS write command
  * @dsi: DSI peripheral device
diff --git a/drivers/gpu/drm/panel/panel-novatek-nt36523.c b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
index 32cf64c7c18b..226d91daf8c7 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt36523.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt36523.c
@@ -23,14 +23,6 @@
 
 #define DSI_NUM_MIN 1
 
-#define mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, cmd, seq...)	\
-		do {								\
-			dsi_ctx.dsi = dsi0;					\
-			mipi_dsi_dcs_write_seq_multi(&dsi_ctx, cmd, seq);	\
-			dsi_ctx.dsi = dsi1;					\
-			mipi_dsi_dcs_write_seq_multi(&dsi_ctx, cmd, seq);	\
-		} while (0)
-
 struct panel_info {
 	struct drm_panel panel;
 	struct mipi_dsi_device *dsi[2];
@@ -71,217 +63,217 @@ static int elish_boe_init_sequence(struct panel_info *pinfo)
 	struct mipi_dsi_device *dsi1 = pinfo->dsi[1];
 	struct mipi_dsi_multi_context dsi_ctx = { .dsi = NULL };
 	/* No datasheet, so write magic init sequence directly */
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x05);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x18, 0x40);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x23);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0x84);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x05, 0x2d);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x06, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x07, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x08, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0x45);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x12, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x15, 0x83);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x16, 0x0c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29, 0x0a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x31, 0xfe);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x32, 0xfd);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x33, 0xfb);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x34, 0xf8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0xf5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x36, 0xf3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x37, 0xf2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x38, 0xf2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0xf2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0xef);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0xec);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3d, 0xe9);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3f, 0xe5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x40, 0xe5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x41, 0xe5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x13);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x45, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x46, 0xf4);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x47, 0xe7);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x48, 0xda);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x49, 0xcd);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4a, 0xc0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4b, 0xb3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4c, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4d, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4e, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x99);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x50, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x68);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x52, 0x66);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x66);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x54, 0x66);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0x0e);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x58, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x59, 0xfb);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5a, 0xf7);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5b, 0xf3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5c, 0xef);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5d, 0xe3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5e, 0xda);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5f, 0xd8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x60, 0xd8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x61, 0xd8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x62, 0xcb);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x63, 0xbf);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x64, 0xb3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x65, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x66, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x67, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x25, 0x47);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0x47);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0x47);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1a, 0xe0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0xe0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x84, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x85, 0x0c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x91, 0x1f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x92, 0x0f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x93, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x94, 0x18);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x95, 0x03);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x96, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb0, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x1f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x1b);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x24);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x28);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x27);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x31);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd1, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd2, 0x30);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xde, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdf, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x81);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x22);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9f, 0x50);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x6f, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x70, 0x11);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x73, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x74, 0x49);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x76, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x77, 0x49);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa0, 0x3f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa9, 0x50);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xaa, 0x28);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xab, 0x28);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xad, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x49);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xba, 0x49);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x49);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbe, 0x04);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbf, 0x49);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc0, 0x04);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc1, 0x59);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc2, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc5, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc6, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc7, 0x48);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xca, 0x43);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcb, 0x3c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xce, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcf, 0x43);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x3c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd3, 0x43);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x3c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd7, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdc, 0x43);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdd, 0x3c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xe1, 0x43);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xe2, 0x3c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xf2, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xf3, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xf4, 0x48);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x13, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x14, 0x23);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbc, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbd, 0x23);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x97, 0x3c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x98, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x99, 0x95);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9a, 0x03);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9b, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9c, 0x0b);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9d, 0x0a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9e, 0x90);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x22);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9f, 0x50);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x23);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa3, 0x50);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xe0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x14, 0x60);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x16, 0xc0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xd0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x02, 0xaf);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0xee);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x99);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1d, 0x09);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x0f, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x2c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x13);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb9, 0x05);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x18, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb9, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x00, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x01, 0x84);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x05, 0x2d);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x06, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x07, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x08, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x09, 0x45);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x11, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x12, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x15, 0x83);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x16, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x29, 0x0a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x30, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x31, 0xfe);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x32, 0xfd);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x33, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x34, 0xf8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x35, 0xf5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x36, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x37, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x38, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x39, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3a, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3b, 0xec);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3d, 0xe9);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3f, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x40, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x41, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2a, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x45, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x46, 0xf4);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x47, 0xe7);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x48, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x49, 0xcd);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4a, 0xc0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4b, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4c, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4d, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4e, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4f, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x50, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x51, 0x68);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x52, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x53, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x54, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2b, 0x0e);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x58, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x59, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5a, 0xf7);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5b, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5c, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5d, 0xe3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5e, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5f, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x60, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x61, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x62, 0xcb);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x63, 0xbf);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x64, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x65, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x66, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x67, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x25, 0x47);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x30, 0x47);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x39, 0x47);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x19, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1a, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1b, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1c, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2a, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2b, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x84, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x85, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x51, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x91, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x92, 0x0f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x93, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x94, 0x18);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x95, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x96, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb0, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x19, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1b, 0x1b);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x24);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb8, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x27);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd0, 0x31);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd1, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd2, 0x30);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd4, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xde, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xdf, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x00, 0x81);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x22);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9f, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x6f, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x70, 0x11);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x73, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x74, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x76, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x77, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xa0, 0x3f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xa9, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xaa, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xab, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xad, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb8, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb9, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xba, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbb, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbe, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbf, 0x49);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc0, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc1, 0x59);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc2, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc5, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc6, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc7, 0x48);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xca, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xcb, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xce, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xcf, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd0, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd3, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd4, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd7, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xdc, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xdd, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xe1, 0x43);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xe2, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xf2, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xf3, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xf4, 0x48);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x13, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x14, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbc, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbd, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x97, 0x3c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x98, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x99, 0x95);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9a, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9b, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9c, 0x0b);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9d, 0x0a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9e, 0x90);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x22);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9f, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xa3, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x14, 0x60);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x16, 0xc0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4f, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3a, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xd0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x02, 0xaf);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x09, 0xee);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1c, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1d, 0x09);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x51, 0x0f, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x53, 0x2c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x35, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbb, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x11);
 	mipi_dsi_msleep(&dsi_ctx, 70);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x29);
 
 	return dsi_ctx.accum_err;
 }
@@ -292,195 +284,195 @@ static int elish_csot_init_sequence(struct panel_info *pinfo)
 	struct mipi_dsi_device *dsi1 = pinfo->dsi[1];
 	struct mipi_dsi_multi_context dsi_ctx = { .dsi = NULL };
 	/* No datasheet, so write magic init sequence directly */
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x05);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x18, 0x40);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xd0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x02, 0xaf);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x30);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0xee);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x99);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1d, 0x09);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xe0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x58, 0x40);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x23);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0x84);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x05, 0x2d);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x06, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x07, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x08, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x09, 0x45);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x12, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x15, 0x83);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x16, 0x0c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29, 0x0a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x31, 0xfe);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x32, 0xfd);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x33, 0xfb);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x34, 0xf8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x35, 0xf5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x36, 0xf3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x37, 0xf2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x38, 0xf2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0xf2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3a, 0xef);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0xec);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3d, 0xe9);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3f, 0xe5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x40, 0xe5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x41, 0xe5);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x13);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x45, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x46, 0xf4);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x47, 0xe7);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x48, 0xda);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x49, 0xcd);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4a, 0xc0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4b, 0xb3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4c, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4d, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4e, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x4f, 0x99);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x50, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x68);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x52, 0x66);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x66);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x54, 0x66);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0x0e);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x58, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x59, 0xfb);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5a, 0xf7);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5b, 0xf3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5c, 0xef);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5d, 0xe3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5e, 0xda);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x5f, 0xd8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x60, 0xd8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x61, 0xd8);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x62, 0xcb);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x63, 0xbf);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x64, 0xb3);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x65, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x66, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x67, 0xb2);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x0f, 0xff);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x53, 0x2c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x55, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x13);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x25, 0x46);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x30, 0x46);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x39, 0x46);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1a, 0xe0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1c, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2a, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x2b, 0xe0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x84, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x85, 0x0c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x51, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x91, 0x1f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x92, 0x0f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x93, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x94, 0x18);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x95, 0x03);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x96, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb0, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x19, 0x1f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x1b, 0x1b);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x24);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x28);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x27);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x31);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd1, 0x20);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xde, 0x80);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xdf, 0x02);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x26);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x00, 0x81);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x22);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x6f, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x70, 0x11);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x73, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x74, 0x4d);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa0, 0x3f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xa9, 0x50);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xaa, 0x28);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xab, 0x28);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xad, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb8, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xb9, 0x4b);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xba, 0x96);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbb, 0x4b);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbe, 0x07);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbf, 0x4b);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc0, 0x07);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc1, 0x5c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc2, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc5, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc6, 0x3f);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xc7, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xca, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcb, 0x40);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xce, 0x00);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xcf, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd0, 0x40);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd3, 0x08);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xd4, 0x40);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x25);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbc, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xbd, 0x1c);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x9a, 0x03);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0xff, 0x10);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x11);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb9, 0x05);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x18, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb9, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xd0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x02, 0xaf);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x00, 0x30);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x09, 0xee);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1c, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1d, 0x09);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3a, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4f, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x58, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x35, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x23);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x00, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x01, 0x84);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x05, 0x2d);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x06, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x07, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x08, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x09, 0x45);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x11, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x12, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x15, 0x83);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x16, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x29, 0x0a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x30, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x31, 0xfe);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x32, 0xfd);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x33, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x34, 0xf8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x35, 0xf5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x36, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x37, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x38, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x39, 0xf2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3a, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3b, 0xec);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3d, 0xe9);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3f, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x40, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x41, 0xe5);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2a, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x45, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x46, 0xf4);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x47, 0xe7);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x48, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x49, 0xcd);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4a, 0xc0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4b, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4c, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4d, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4e, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x4f, 0x99);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x50, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x51, 0x68);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x52, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x53, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x54, 0x66);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2b, 0x0e);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x58, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x59, 0xfb);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5a, 0xf7);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5b, 0xf3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5c, 0xef);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5d, 0xe3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5e, 0xda);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x5f, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x60, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x61, 0xd8);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x62, 0xcb);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x63, 0xbf);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x64, 0xb3);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x65, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x66, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x67, 0xb2);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x51, 0x0f, 0xff);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x53, 0x2c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x55, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbb, 0x13);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x3b, 0x03, 0xac, 0x1a, 0x04, 0x04);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x25, 0x46);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x30, 0x46);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x39, 0x46);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x19, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1a, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1b, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1c, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2a, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x2b, 0xe0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0xf0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x84, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x85, 0x0c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x51, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x91, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x92, 0x0f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x93, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x94, 0x18);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x95, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x96, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb0, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x19, 0x1f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x1b, 0x1b);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x24);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb8, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x27);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd0, 0x31);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd1, 0x20);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd4, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xde, 0x80);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xdf, 0x02);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x26);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x00, 0x81);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x01, 0xb0);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x22);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x6f, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x70, 0x11);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x73, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x74, 0x4d);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xa0, 0x3f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xa9, 0x50);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xaa, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xab, 0x28);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xad, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb8, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xb9, 0x4b);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xba, 0x96);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbb, 0x4b);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbe, 0x07);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbf, 0x4b);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc0, 0x07);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc1, 0x5c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc2, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc5, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc6, 0x3f);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xc7, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xca, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xcb, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xce, 0x00);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xcf, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd0, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd3, 0x08);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xd4, 0x40);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x25);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbc, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xbd, 0x1c);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x2a);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xfb, 0x01);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x9a, 0x03);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0xff, 0x10);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x11);
 	mipi_dsi_msleep(&dsi_ctx, 70);
-	mipi_dsi_dual_dcs_write_seq_multi(dsi_ctx, dsi0, dsi1, 0x29);
+	mipi_dsi_dual_dcs_write_seq_multi(&dsi_ctx, dsi0, dsi1, 0x29);
 
 	return dsi_ctx.accum_err;
 }
diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h
index 369b0d8830c3..f1dc822f69d6 100644
--- a/include/drm/drm_mipi_dsi.h
+++ b/include/drm/drm_mipi_dsi.h
@@ -289,6 +289,10 @@ int mipi_dsi_generic_write_chatty(struct mipi_dsi_device *dsi,
 				  const void *payload, size_t size);
 void mipi_dsi_generic_write_multi(struct mipi_dsi_multi_context *ctx,
 				  const void *payload, size_t size);
+void mipi_dsi_dual_generic_write_multi(struct mipi_dsi_multi_context *ctx,
+				       struct mipi_dsi_device *dsi1,
+				       struct mipi_dsi_device *dsi2,
+				       const void *payload, size_t size);
 ssize_t mipi_dsi_generic_read(struct mipi_dsi_device *dsi, const void *params,
 			      size_t num_params, void *data, size_t size);
 u32 drm_mipi_dsi_get_input_bus_fmt(enum mipi_dsi_pixel_format dsi_format);
@@ -329,6 +333,10 @@ int mipi_dsi_dcs_write_buffer_chatty(struct mipi_dsi_device *dsi,
 				     const void *data, size_t len);
 void mipi_dsi_dcs_write_buffer_multi(struct mipi_dsi_multi_context *ctx,
 				     const void *data, size_t len);
+void mipi_dsi_dual_dcs_write_buffer_multi(struct mipi_dsi_multi_context *ctx,
+					  struct mipi_dsi_device *dsi1,
+					  struct mipi_dsi_device *dsi2,
+					  const void *data, size_t len);
 ssize_t mipi_dsi_dcs_write(struct mipi_dsi_device *dsi, u8 cmd,
 			   const void *data, size_t len);
 ssize_t mipi_dsi_dcs_read(struct mipi_dsi_device *dsi, u8 cmd, void *data,
@@ -431,6 +439,93 @@ void mipi_dsi_dcs_set_tear_off_multi(struct mipi_dsi_multi_context *ctx);
 		mipi_dsi_dcs_write_buffer_multi(ctx, d, ARRAY_SIZE(d)); \
 	} while (0)
 
+/**
+ * mipi_dsi_dual - send the same MIPI DSI command to two interfaces
+ *
+ * This macro will send the specified MIPI DSI command twice, once per each of
+ * the two interfaces supplied. This is useful for reducing duplication of code
+ * in panel drivers which use two parallel serial interfaces.
+ *
+ * Note that the _func parameter cannot accept a macro such as
+ * mipi_dsi_generic_write_multi() or mipi_dsi_dcs_write_buffer_multi(). See
+ * mipi_dsi_dual_generic_write_multi() and
+ * mipi_dsi_dual_dcs_write_buffer_multi() instead.
+ *
+ * WARNING: This macro reuses the _func argument and the optional trailing
+ * arguments twice each, which may cause unintended side effects. For example,
+ * adding the postfix increment ++ operator to one of the arguments to be
+ * passed to _func will cause the variable to be incremented twice instead of
+ * once and the variable will be its original value + 1 when sent to _dsi2.
+ *
+ * @_func: MIPI DSI function to pass context and arguments into
+ * @_ctx: Context for multiple DSI transactions
+ * @_dsi1: First DSI interface to act as recipient of the MIPI DSI command
+ * @_dsi2: Second DSI interface to act as recipient of the MIPI DSI command
+ * @...: Arguments to pass to MIPI DSI function or macro
+ */
+
+#define mipi_dsi_dual(_func, _ctx, _dsi1, _dsi2, ...)		 \
+	do {							 \
+		struct mipi_dsi_multi_context *_ctxcpy = (_ctx); \
+		_ctxcpy->dsi = (_dsi1);				 \
+		(_func)(_ctxcpy, ##__VA_ARGS__);		 \
+		_ctxcpy->dsi = (_dsi2);				 \
+		(_func)(_ctxcpy, ##__VA_ARGS__);		 \
+	} while (0)
+
+/**
+ * mipi_dsi_dual_generic_write_seq_multi - transmit data using a generic write
+ * packet to two dsi interfaces, one after the other
+ *
+ * This macro will send the specified generic packet twice, once per each of
+ * the two interfaces supplied. This is useful for reducing duplication of code
+ * in panel drivers which use two parallel serial interfaces.
+ *
+ * Note that if an error occurs while transmitting the packet to the first DSI
+ * interface, the packet will not be sent to the second DSI interface.
+ *
+ * This macro will print errors for you and error handling is optimized for
+ * callers that call this multiple times in a row.
+ *
+ * @_ctx: Context for multiple DSI transactions
+ * @_dsi1: First DSI interface to act as recipient of packet
+ * @_dsi2: Second DSI interface to act as recipient of packet
+ * @_seq: buffer containing the payload
+ */
+#define mipi_dsi_dual_generic_write_seq_multi(_ctx, _dsi1, _dsi2, _seq...) \
+	do {								   \
+		static const u8 d[] = { _seq };				   \
+		mipi_dsi_dual_generic_write_multi(_ctx, _dsi1, _dsi2, d,   \
+						  ARRAY_SIZE(d));	   \
+	} while (0)
+
+/**
+ * mipi_dsi_dual_dcs_write_seq_multi - transmit a DCS command with payload to
+ * two dsi interfaces, one after the other
+ *
+ * This macro will send the specified DCS command with payload twice, once per
+ * each of the two interfaces supplied. This is useful for reducing duplication
+ * of code in panel drivers which use two parallel serial interfaces.
+ *
+ * Note that if an error occurs while transmitting the payload to the first DSI
+ * interface, the payload will not be sent to the second DSI interface.
+ *
+ * This macro will print errors for you and error handling is optimized for
+ * callers that call this multiple times in a row.
+ *
+ * @_ctx: Context for multiple DSI transactions
+ * @_dsi1: First DSI interface to act as recipient of packet
+ * @_dsi2: Second DSI interface to act as recipient of packet
+ * @_cmd: Command
+ * @_seq: buffer containing the payload
+ */
+#define mipi_dsi_dual_dcs_write_seq_multi(_ctx, _dsi1, _dsi2, _cmd, _seq...) \
+	do {								     \
+		static const u8 d[] = { _cmd, _seq };			     \
+		mipi_dsi_dual_dcs_write_buffer_multi(_ctx, _dsi1, _dsi2, d,  \
+						     ARRAY_SIZE(d));	     \
+	} while (0)
+
 /**
  * struct mipi_dsi_driver - DSI driver
  * @driver: device driver model driver

From a6adf47d30cccaf3c1936ac9de94948c140e17dd Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Mon, 21 Jul 2025 19:53:09 -0600
Subject: [PATCH 0161/1455] drm/panel: jdi-lpm102a188a: Fix bug and clean up
 driver

Fix bug in unprepare() which causes the function's return value to be
that of the last mipi "enter sleep mode" command.

Update driver to use the "multi" variant of MIPI functions in order to
facilitate improved error handling and remove the panel's dependency on
deprecated MIPI functions.

Use the new mipi_dsi_dual macro to reduce code duplication.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Diogo Ivo <diogo.ivo@tecnico.ulisboa.pt>
Tested-by: Diogo Ivo <diogo.ivo@tecnico.ulisboa.pt>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250722015313.561966-3-me@brighamcampbell.com
---
 drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c | 196 ++++++------------
 1 file changed, 59 insertions(+), 137 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c b/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c
index 5f897e143758..83656bb4b0b2 100644
--- a/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c
+++ b/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c
@@ -81,25 +81,25 @@ static int jdi_panel_disable(struct drm_panel *panel)
 static int jdi_panel_unprepare(struct drm_panel *panel)
 {
 	struct jdi_panel *jdi = to_panel_jdi(panel);
-	int ret;
 
-	ret = mipi_dsi_dcs_set_display_off(jdi->link1);
-	if (ret < 0)
-		dev_err(panel->dev, "failed to set display off: %d\n", ret);
+	/*
+	 * One context per panel since we'll continue trying to shut down the
+	 * other panel even if one isn't responding.
+	 */
+	struct mipi_dsi_multi_context dsi_ctx1 = { .dsi = jdi->link1 };
+	struct mipi_dsi_multi_context dsi_ctx2 = { .dsi = jdi->link2 };
 
-	ret = mipi_dsi_dcs_set_display_off(jdi->link2);
-	if (ret < 0)
-		dev_err(panel->dev, "failed to set display off: %d\n", ret);
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx1);
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx2);
 
 	/* Specified by JDI @ 50ms, subject to change */
 	msleep(50);
 
-	ret = mipi_dsi_dcs_enter_sleep_mode(jdi->link1);
-	if (ret < 0)
-		dev_err(panel->dev, "failed to enter sleep mode: %d\n", ret);
-	ret = mipi_dsi_dcs_enter_sleep_mode(jdi->link2);
-	if (ret < 0)
-		dev_err(panel->dev, "failed to enter sleep mode: %d\n", ret);
+	/* Doesn't hurt to try sleep mode even if display off fails */
+	dsi_ctx1.accum_err = 0;
+	dsi_ctx2.accum_err = 0;
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx1);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx2);
 
 	/* Specified by JDI @ 150ms, subject to change */
 	msleep(150);
@@ -123,72 +123,46 @@ static int jdi_panel_unprepare(struct drm_panel *panel)
 	/* Specified by JDI @ 20ms, subject to change */
 	msleep(20);
 
-	return ret;
-}
-
-static int jdi_setup_symmetrical_split(struct mipi_dsi_device *left,
-				       struct mipi_dsi_device *right,
-				       const struct drm_display_mode *mode)
-{
-	int err;
-
-	err = mipi_dsi_dcs_set_column_address(left, 0, mode->hdisplay / 2 - 1);
-	if (err < 0) {
-		dev_err(&left->dev, "failed to set column address: %d\n", err);
-		return err;
-	}
-
-	err = mipi_dsi_dcs_set_column_address(right, 0, mode->hdisplay / 2 - 1);
-	if (err < 0) {
-		dev_err(&right->dev, "failed to set column address: %d\n", err);
-		return err;
-	}
-
-	err = mipi_dsi_dcs_set_page_address(left, 0, mode->vdisplay - 1);
-	if (err < 0) {
-		dev_err(&left->dev, "failed to set page address: %d\n", err);
-		return err;
-	}
-
-	err = mipi_dsi_dcs_set_page_address(right, 0, mode->vdisplay - 1);
-	if (err < 0) {
-		dev_err(&right->dev, "failed to set page address: %d\n", err);
-		return err;
-	}
-
 	return 0;
 }
 
-static int jdi_write_dcdc_registers(struct jdi_panel *jdi)
+static void jdi_setup_symmetrical_split(struct mipi_dsi_multi_context *dsi_ctx,
+					struct mipi_dsi_device *left,
+					struct mipi_dsi_device *right,
+					const struct drm_display_mode *mode)
+{
+	mipi_dsi_dual(mipi_dsi_dcs_set_column_address_multi,
+		      dsi_ctx, left, right,
+		      0, mode->hdisplay / 2 - 1);
+	mipi_dsi_dual(mipi_dsi_dcs_set_page_address_multi,
+		      dsi_ctx, left, right,
+		      0, mode->vdisplay - 1);
+}
+
+static void jdi_write_dcdc_registers(struct mipi_dsi_multi_context *dsi_ctx,
+				     struct jdi_panel *jdi)
 {
 	/* Clear the manufacturer command access protection */
-	mipi_dsi_generic_write_seq(jdi->link1, MCS_CMD_ACS_PROT,
-				   MCS_CMD_ACS_PROT_OFF);
-	mipi_dsi_generic_write_seq(jdi->link2, MCS_CMD_ACS_PROT,
-				   MCS_CMD_ACS_PROT_OFF);
+	mipi_dsi_dual_generic_write_seq_multi(dsi_ctx, jdi->link1, jdi->link2,
+					      MCS_CMD_ACS_PROT,
+					      MCS_CMD_ACS_PROT_OFF);
 	/*
-	 * Change the VGH/VGL divide rations to move the noise generated by the
+	 * Change the VGH/VGL divide ratios to move the noise generated by the
 	 * TCONN. This should hopefully avoid interaction with the backlight
 	 * controller.
 	 */
-	mipi_dsi_generic_write_seq(jdi->link1, MCS_PWR_CTRL_FUNC,
-				   MCS_PWR_CTRL_PARAM1_VGH_330_DIV |
-				   MCS_PWR_CTRL_PARAM1_DEFAULT,
-				   MCS_PWR_CTRL_PARAM2_VGL_410_DIV |
-				   MCS_PWR_CTRL_PARAM2_DEFAULT);
-
-	mipi_dsi_generic_write_seq(jdi->link2, MCS_PWR_CTRL_FUNC,
-				   MCS_PWR_CTRL_PARAM1_VGH_330_DIV |
-				   MCS_PWR_CTRL_PARAM1_DEFAULT,
-				   MCS_PWR_CTRL_PARAM2_VGL_410_DIV |
-				   MCS_PWR_CTRL_PARAM2_DEFAULT);
-
-	return 0;
+	mipi_dsi_dual_generic_write_seq_multi(dsi_ctx, jdi->link1, jdi->link2,
+					      MCS_PWR_CTRL_FUNC,
+					      MCS_PWR_CTRL_PARAM1_VGH_330_DIV |
+					      MCS_PWR_CTRL_PARAM1_DEFAULT,
+					      MCS_PWR_CTRL_PARAM2_VGL_410_DIV |
+					      MCS_PWR_CTRL_PARAM2_DEFAULT);
 }
 
 static int jdi_panel_prepare(struct drm_panel *panel)
 {
 	struct jdi_panel *jdi = to_panel_jdi(panel);
+	struct mipi_dsi_multi_context dsi_ctx = {};
 	int err;
 
 	/* Disable backlight to avoid showing random pixels
@@ -231,88 +205,36 @@ static int jdi_panel_prepare(struct drm_panel *panel)
 	 * put in place to communicate the configuration back to the DSI host
 	 * controller.
 	 */
-	err = jdi_setup_symmetrical_split(jdi->link1, jdi->link2,
-					  jdi->mode);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set up symmetrical split: %d\n",
-			err);
-		goto poweroff;
-	}
+	jdi_setup_symmetrical_split(&dsi_ctx, jdi->link1, jdi->link2,
+				    jdi->mode);
 
-	err = mipi_dsi_dcs_set_tear_scanline(jdi->link1,
-					     jdi->mode->vdisplay - 16);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set tear scanline: %d\n", err);
-		goto poweroff;
-	}
+	mipi_dsi_dual(mipi_dsi_dcs_set_tear_scanline_multi,
+		      &dsi_ctx, jdi->link1, jdi->link2,
+		      jdi->mode->vdisplay - 16);
 
-	err = mipi_dsi_dcs_set_tear_scanline(jdi->link2,
-					     jdi->mode->vdisplay - 16);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set tear scanline: %d\n", err);
-		goto poweroff;
-	}
+	mipi_dsi_dual(mipi_dsi_dcs_set_tear_on_multi,
+		      &dsi_ctx, jdi->link1, jdi->link2,
+		      MIPI_DSI_DCS_TEAR_MODE_VBLANK);
 
-	err = mipi_dsi_dcs_set_tear_on(jdi->link1,
-				       MIPI_DSI_DCS_TEAR_MODE_VBLANK);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set tear on: %d\n", err);
-		goto poweroff;
-	}
+	mipi_dsi_dual(mipi_dsi_dcs_set_pixel_format_multi,
+		      &dsi_ctx, jdi->link1, jdi->link2,
+		      MIPI_DCS_PIXEL_FMT_24BIT);
 
-	err = mipi_dsi_dcs_set_tear_on(jdi->link2,
-				       MIPI_DSI_DCS_TEAR_MODE_VBLANK);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set tear on: %d\n", err);
-		goto poweroff;
-	}
+	mipi_dsi_dual(mipi_dsi_dcs_exit_sleep_mode_multi,
+		      &dsi_ctx, jdi->link1, jdi->link2);
 
-	err = mipi_dsi_dcs_set_pixel_format(jdi->link1, MIPI_DCS_PIXEL_FMT_24BIT);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set pixel format: %d\n", err);
-		goto poweroff;
-	}
-
-	err = mipi_dsi_dcs_set_pixel_format(jdi->link2, MIPI_DCS_PIXEL_FMT_24BIT);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set pixel format: %d\n", err);
-		goto poweroff;
-	}
-
-	err = mipi_dsi_dcs_exit_sleep_mode(jdi->link1);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to exit sleep mode: %d\n", err);
-		goto poweroff;
-	}
-
-	err = mipi_dsi_dcs_exit_sleep_mode(jdi->link2);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to exit sleep mode: %d\n", err);
-		goto poweroff;
-	}
-
-	err = jdi_write_dcdc_registers(jdi);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to write dcdc registers: %d\n", err);
-		goto poweroff;
-	}
+	jdi_write_dcdc_registers(&dsi_ctx, jdi);
 	/*
-	 * We need to wait 150ms between mipi_dsi_dcs_exit_sleep_mode() and
-	 * mipi_dsi_dcs_set_display_on().
+	 * We need to wait 150ms between mipi_dsi_dcs_exit_sleep_mode_multi()
+	 * and mipi_dsi_dcs_set_display_on_multi().
 	 */
-	msleep(150);
+	mipi_dsi_msleep(&dsi_ctx, 150);
 
-	err = mipi_dsi_dcs_set_display_on(jdi->link1);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set display on: %d\n", err);
-		goto poweroff;
-	}
+	mipi_dsi_dual(mipi_dsi_dcs_set_display_on_multi,
+		      &dsi_ctx, jdi->link1, jdi->link2);
 
-	err = mipi_dsi_dcs_set_display_on(jdi->link2);
-	if (err < 0) {
-		dev_err(panel->dev, "failed to set display on: %d\n", err);
+	if (dsi_ctx.accum_err < 0)
 		goto poweroff;
-	}
 
 	jdi->link1->mode_flags &= ~MIPI_DSI_MODE_LPM;
 	jdi->link2->mode_flags &= ~MIPI_DSI_MODE_LPM;

From 79b6bb18f849818140dd351f6e76a097efe99e9f Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Mon, 21 Jul 2025 19:53:10 -0600
Subject: [PATCH 0162/1455] drm: Remove unused MIPI write seq and chatty
 functions

Remove the deprecated mipi_dsi_generic_write_seq() and
mipi_dsi_generic_write_chatty() functions now that they are no longer
used.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250722015313.561966-4-me@brighamcampbell.com
---
 drivers/gpu/drm/drm_mipi_dsi.c | 34 +++-------------------------------
 include/drm/drm_mipi_dsi.h     | 23 -----------------------
 2 files changed, 3 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index 5b387543de28..ca77f152404a 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -772,41 +772,13 @@ ssize_t mipi_dsi_generic_write(struct mipi_dsi_device *dsi, const void *payload,
 EXPORT_SYMBOL(mipi_dsi_generic_write);
 
 /**
- * mipi_dsi_generic_write_chatty() - mipi_dsi_generic_write() w/ an error log
- * @dsi: DSI peripheral device
- * @payload: buffer containing the payload
- * @size: size of payload buffer
- *
- * Like mipi_dsi_generic_write() but includes a dev_err()
- * call for you and returns 0 upon success, not the number of bytes sent.
- *
- * Return: 0 on success or a negative error code on failure.
- */
-int mipi_dsi_generic_write_chatty(struct mipi_dsi_device *dsi,
-				  const void *payload, size_t size)
-{
-	struct device *dev = &dsi->dev;
-	ssize_t ret;
-
-	ret = mipi_dsi_generic_write(dsi, payload, size);
-	if (ret < 0) {
-		dev_err(dev, "sending generic data %*ph failed: %zd\n",
-			(int)size, payload, ret);
-		return ret;
-	}
-
-	return 0;
-}
-EXPORT_SYMBOL(mipi_dsi_generic_write_chatty);
-
-/**
- * mipi_dsi_generic_write_multi() - mipi_dsi_generic_write_chatty() w/ accum_err
+ * mipi_dsi_generic_write_multi() - mipi_dsi_generic_write() w/ accum_err
  * @ctx: Context for multiple DSI transactions
  * @payload: buffer containing the payload
  * @size: size of payload buffer
  *
- * Like mipi_dsi_generic_write_chatty() but deals with errors in a way that
- * makes it convenient to make several calls in a row.
+ * A wrapper around mipi_dsi_generic_write() that deals with errors in a way
+ * that makes it convenient to make several calls in a row.
  */
 void mipi_dsi_generic_write_multi(struct mipi_dsi_multi_context *ctx,
 				  const void *payload, size_t size)
diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h
index f1dc822f69d6..ea523eb35b08 100644
--- a/include/drm/drm_mipi_dsi.h
+++ b/include/drm/drm_mipi_dsi.h
@@ -285,8 +285,6 @@ void mipi_dsi_picture_parameter_set_multi(struct mipi_dsi_multi_context *ctx,
 
 ssize_t mipi_dsi_generic_write(struct mipi_dsi_device *dsi, const void *payload,
 			       size_t size);
-int mipi_dsi_generic_write_chatty(struct mipi_dsi_device *dsi,
-				  const void *payload, size_t size);
 void mipi_dsi_generic_write_multi(struct mipi_dsi_multi_context *ctx,
 				  const void *payload, size_t size);
 void mipi_dsi_dual_generic_write_multi(struct mipi_dsi_multi_context *ctx,
@@ -387,27 +385,6 @@ void mipi_dsi_dcs_set_tear_scanline_multi(struct mipi_dsi_multi_context *ctx,
 					  u16 scanline);
 void mipi_dsi_dcs_set_tear_off_multi(struct mipi_dsi_multi_context *ctx);
 
-/**
- * mipi_dsi_generic_write_seq - transmit data using a generic write packet
- *
- * This macro will print errors for you and will RETURN FROM THE CALLING
- * FUNCTION (yes this is non-intuitive) upon error.
- *
- * Because of the non-intuitive return behavior, THIS MACRO IS DEPRECATED.
- * Please replace calls of it with mipi_dsi_generic_write_seq_multi().
- *
- * @dsi: DSI peripheral device
- * @seq: buffer containing the payload
- */
-#define mipi_dsi_generic_write_seq(dsi, seq...)                                \
-	do {                                                                   \
-		static const u8 d[] = { seq };                                 \
-		int ret;                                                       \
-		ret = mipi_dsi_generic_write_chatty(dsi, d, ARRAY_SIZE(d));    \
-		if (ret < 0)                                                   \
-			return ret;                                            \
-	} while (0)
-
 /**
  * mipi_dsi_generic_write_seq_multi - transmit data using a generic write packet
  *

From 85c23f28905cf20a86ceec3cfd7a0a5572c9eb13 Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Mon, 21 Jul 2025 19:53:11 -0600
Subject: [PATCH 0163/1455] drm: docs: Update task from drm TODO list

Update TODO item from drm documentation to contain more applicable
information regarding the removal of deprecated MIPI DSI functions and
no longer reference functions which have already been removed from the
kernel.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250722015313.561966-5-me@brighamcampbell.com
---
 Documentation/gpu/todo.rst | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst
index be8637da3fe9..92db80793bba 100644
--- a/Documentation/gpu/todo.rst
+++ b/Documentation/gpu/todo.rst
@@ -497,19 +497,19 @@ Contact: Douglas Anderson <dianders@chromium.org>
 
 Level: Intermediate
 
-Transition away from using mipi_dsi_*_write_seq()
--------------------------------------------------
+Transition away from using deprecated MIPI DSI functions
+--------------------------------------------------------
 
-The macros mipi_dsi_generic_write_seq() and mipi_dsi_dcs_write_seq() are
-non-intuitive because, if there are errors, they return out of the *caller's*
-function. We should move all callers to use mipi_dsi_generic_write_seq_multi()
-and mipi_dsi_dcs_write_seq_multi() macros instead.
+There are many functions defined in ``drm_mipi_dsi.c`` which have been
+deprecated. Each deprecated function was deprecated in favor of its `multi`
+variant (e.g. `mipi_dsi_generic_write()` and `mipi_dsi_generic_write_multi()`).
+The `multi` variant of a function includes improved error handling and logic
+which makes it more convenient to make several calls in a row, as most MIPI
+drivers do.
 
-Once all callers are transitioned, the macros and the functions that they call,
-mipi_dsi_generic_write_chatty() and mipi_dsi_dcs_write_buffer_chatty(), can
-probably be removed. Alternatively, if people feel like the _multi() variants
-are overkill for some use cases, we could keep the mipi_dsi_*_write_seq()
-variants but change them not to return out of the caller.
+Drivers should be updated to use undeprecated functions. Once all usages of the
+deprecated MIPI DSI functions have been removed, their definitions may be
+removed from ``drm_mipi_dsi.c``.
 
 Contact: Douglas Anderson <dianders@chromium.org>
 

From a843b9894705d5d171e97f167fc5290f6a3ecaee Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Tue, 29 Jul 2025 17:37:20 +0530
Subject: [PATCH 0164/1455] drm/xe/vf: Fix VM crash during VF driver release
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The VF CCS save/restore series (patchwork #149108) has a dependency
on the migration framework. A recent migration update in commit
d65ff1ec8535 ("drm/xe: Split xe_migrate allocation from initialization")
caused a VM crash during XE driver release for iGPU devices.

Oops: general protection fault, probably for non-canonical address
0x6b6b6b6b6b6b6b83: 0000 [#1] SMP NOPTI
RIP: 0010:xe_lrc_ring_head+0x12/0xb0 [xe]
Call Trace:
 xe_sriov_vf_ccs_fini+0x1e/0x40 [xe]
 devm_action_release+0x12/0x30
 release_nodes+0x3a/0x120
 devres_release_all+0x96/0xd0
 device_unbind_cleanup+0x12/0x80
 device_release_driver_internal+0x23a/0x280
 device_release_driver+0x12/0x20
 pci_stop_bus_device+0x69/0x90
 pci_stop_and_remove_bus_device+0x12/0x30
 pci_iov_remove_virtfn+0xbd/0x130
 sriov_disable+0x42/0x100
 pci_disable_sriov+0x34/0x50
 xe_pci_sriov_configure+0xf71/0x1020 [xe]

Update the VF CCS migration initialization sequence to align with the new
migration framework changes, resolving the release-time crash.

Fixes: f3009272ff2e ("drm/xe/vf: Create contexts for CCS read write")
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250729120720.13990-1-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/xe_gt.c           |  6 ++---
 drivers/gpu/drm/xe/xe_migrate.c      | 37 ++++++++++++++++------------
 drivers/gpu/drm/xe/xe_migrate.h      |  2 +-
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c |  7 +++++-
 4 files changed, 30 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index c8eda36546d3..5a79c6e3208b 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -564,11 +564,9 @@ static int gt_init_with_all_forcewake(struct xe_gt *gt)
 	if (xe_gt_is_main_type(gt)) {
 		struct xe_tile *tile = gt_to_tile(gt);
 
-		tile->migrate = xe_migrate_init(tile);
-		if (IS_ERR(tile->migrate)) {
-			err = PTR_ERR(tile->migrate);
+		err = xe_migrate_init(tile->migrate);
+		if (err)
 			goto err_force_wake;
-		}
 	}
 
 	err = xe_uc_load_hw(&gt->uc);
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 90065d7d29ff..3a276e2348a2 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -396,15 +396,15 @@ struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile)
 
 /**
  * xe_migrate_init() - Initialize a migrate context
- * @tile: Back-pointer to the tile we're initializing for.
+ * @m: The migration context
  *
- * Return: Pointer to a migrate context on success. Error pointer on error.
+ * Return: 0 if successful, negative error code on failure
  */
-struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
+int xe_migrate_init(struct xe_migrate *m)
 {
-	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_tile *tile = m->tile;
 	struct xe_gt *primary_gt = tile->primary_gt;
-	struct xe_migrate *m = tile->migrate;
+	struct xe_device *xe = tile_to_xe(tile);
 	struct xe_vm *vm;
 	int err;
 
@@ -412,15 +412,13 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 	vm = xe_vm_create(xe, XE_VM_FLAG_MIGRATION |
 			  XE_VM_FLAG_SET_TILE_ID(tile));
 	if (IS_ERR(vm))
-		return ERR_CAST(vm);
+		return PTR_ERR(vm);
 
 	xe_vm_lock(vm, false);
 	err = xe_migrate_prepare_vm(tile, m, vm);
 	xe_vm_unlock(vm);
-	if (err) {
-		xe_vm_close_and_put(vm);
-		return ERR_PTR(err);
-	}
+	if (err)
+		goto err_out;
 
 	if (xe->info.has_usm) {
 		struct xe_hw_engine *hwe = xe_gt_hw_engine(primary_gt,
@@ -429,8 +427,10 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 							   false);
 		u32 logical_mask = xe_migrate_usm_logical_mask(primary_gt);
 
-		if (!hwe || !logical_mask)
-			return ERR_PTR(-EINVAL);
+		if (!hwe || !logical_mask) {
+			err = -EINVAL;
+			goto err_out;
+		}
 
 		/*
 		 * XXX: Currently only reserving 1 (likely slow) BCS instance on
@@ -449,8 +449,8 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 						  EXEC_QUEUE_FLAG_MIGRATE, 0);
 	}
 	if (IS_ERR(m->q)) {
-		xe_vm_close_and_put(vm);
-		return ERR_CAST(m->q);
+		err = PTR_ERR(m->q);
+		goto err_out;
 	}
 
 	mutex_init(&m->job_mutex);
@@ -460,7 +460,7 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 
 	err = devm_add_action_or_reset(xe->drm.dev, xe_migrate_fini, m);
 	if (err)
-		return ERR_PTR(err);
+		return err;
 
 	if (IS_DGFX(xe)) {
 		if (xe_migrate_needs_ccs_emit(xe))
@@ -475,7 +475,12 @@ struct xe_migrate *xe_migrate_init(struct xe_tile *tile)
 			(unsigned long long)m->min_chunk_size);
 	}
 
-	return m;
+	return err;
+
+err_out:
+	xe_vm_close_and_put(vm);
+	return err;
+
 }
 
 static u64 max_mem_transfer_per_pass(struct xe_device *xe)
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 3758f9615484..e81ea6b27fb5 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -105,7 +105,7 @@ struct xe_migrate_pt_update {
 };
 
 struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile);
-struct xe_migrate *xe_migrate_init(struct xe_tile *tile);
+int xe_migrate_init(struct xe_migrate *m);
 
 struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
 				     unsigned long npages,
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index af43e04179aa..bf9fa1238462 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -270,11 +270,16 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 		ctx = &tile->sriov.vf.ccs[ctx_id];
 		ctx->ctx_id = ctx_id;
 
-		migrate = xe_migrate_init(tile);
+		migrate = xe_migrate_alloc(tile);
 		if (IS_ERR(migrate)) {
 			err = PTR_ERR(migrate);
 			goto err_ret;
 		}
+
+		err = xe_migrate_init(migrate);
+		if (err)
+			goto err_ret;
+
 		ctx->migrate = migrate;
 
 		err = alloc_bb_pool(tile, ctx);

From 82dde0407ab126f8413fd6c51429e5057ced5ba2 Mon Sep 17 00:00:00 2001
From: Vinod Govindapillai <vinod.govindapillai@intel.com>
Date: Tue, 29 Jul 2025 15:46:48 +0300
Subject: [PATCH 0165/1455] drm/i915/fbc: fix the implementation of
 wa_18038517565
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

As per the wa_18038517565, we need to disable FBC compressor
clock gating before enabling FBC and enable after disabling
FBC. Placing the enabling of clock gating in the fbc deactivate
function can make the above wa logic go wrong in case of
frontbuffer rendering FBC mechanism. FBC deactivate can get
called during fb invalidate and then the corresponding FBC
activate can get called without properly disabling the clock
gating and can result in compression stalled. So move the
enable clock gating at the end of one FBC session after FBC
is completely disabled for a pipe.

Bspec: 74212, 72197, 69741, 65555
Fixes: 010363c46189 ("drm/i915/display: implement wa_18038517565")
Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/20250729124648.288497-1-vinod.govindapillai@intel.com
---
 drivers/gpu/drm/i915/display/intel_fbc.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index e2e03af520b2..f82d392f1e22 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -552,10 +552,6 @@ static void ilk_fbc_deactivate(struct intel_fbc *fbc)
 	if (dpfc_ctl & DPFC_CTL_EN) {
 		dpfc_ctl &= ~DPFC_CTL_EN;
 		intel_de_write(display, ILK_DPFC_CONTROL(fbc->id), dpfc_ctl);
-
-		/* wa_18038517565 Enable DPFC clock gating after FBC disable */
-		if (display->platform.dg2 || DISPLAY_VER(display) >= 14)
-			fbc_compressor_clkgate_disable_wa(fbc, false);
 	}
 }
 
@@ -1710,6 +1706,10 @@ static void __intel_fbc_disable(struct intel_fbc *fbc)
 
 	__intel_fbc_cleanup_cfb(fbc);
 
+	/* wa_18038517565 Enable DPFC clock gating after FBC disable */
+	if (display->platform.dg2 || DISPLAY_VER(display) >= 14)
+		fbc_compressor_clkgate_disable_wa(fbc, false);
+
 	fbc->state.plane = NULL;
 	fbc->flip_pending = false;
 	fbc->busy_bits = 0;

From 552dbba1caaf0cb40ce961806d757615e26ec668 Mon Sep 17 00:00:00 2001
From: Lukasz Laguna <lukasz.laguna@intel.com>
Date: Tue, 29 Jul 2025 14:34:37 +0200
Subject: [PATCH 0166/1455] drm/xe/vf: Disable CSC support on VF

CSC is not accessible by VF drivers, so disable its support flag on VF
to prevent further initialization attempts.

Fixes: e02cea83d32d ("drm/xe/gsc: add Battlemage support")
Signed-off-by: Lukasz Laguna <lukasz.laguna@intel.com>
Cc: Alexander Usyskin <alexander.usyskin@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250729123437.5933-1-lukasz.laguna@intel.com
---
 drivers/gpu/drm/xe/xe_device.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index d04a0ae018e6..c754374f1a8d 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -682,6 +682,7 @@ static void sriov_update_device_info(struct xe_device *xe)
 	/* disable features that are not available/applicable to VFs */
 	if (IS_SRIOV_VF(xe)) {
 		xe->info.probe_display = 0;
+		xe->info.has_heci_cscfi = 0;
 		xe->info.has_heci_gscfi = 0;
 		xe->info.skip_guc_pc = 1;
 		xe->info.skip_pcode = 1;

From 784c99331c8d54a51d4f3e772c81f7fb82b7a1f8 Mon Sep 17 00:00:00 2001
From: Svyatoslav Ryhel <clamor95@gmail.com>
Date: Wed, 30 Jul 2025 08:54:23 +0300
Subject: [PATCH 0167/1455] dt-bindings: display: bridge: Document Solomon
 SSD2825

Add bindings for Solomon SSD2825 MIPI master bridge chip that connects an
application processor with traditional parallel LCD interface and an LCD
driver with MIPI slave interface. The SSD2825 supports both parallel RGB
interface and serial SPI interface.

Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20250730055424.6718-2-clamor95@gmail.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../display/bridge/solomon,ssd2825.yaml       | 141 ++++++++++++++++++
 1 file changed, 141 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/solomon,ssd2825.yaml

diff --git a/Documentation/devicetree/bindings/display/bridge/solomon,ssd2825.yaml b/Documentation/devicetree/bindings/display/bridge/solomon,ssd2825.yaml
new file mode 100644
index 000000000000..e2d293d623b8
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/bridge/solomon,ssd2825.yaml
@@ -0,0 +1,141 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/solomon,ssd2825.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Solomon SSD2825 RGB to MIPI-DSI bridge
+
+maintainers:
+  - Svyatoslav Ryhel <clamor95@gmail.com>
+
+allOf:
+  - $ref: /schemas/spi/spi-peripheral-props.yaml#
+
+properties:
+  compatible:
+    const: solomon,ssd2825
+
+  reg:
+    maxItems: 1
+
+  reset-gpios: true
+
+  dvdd-supply:
+    description: Regulator for 1.2V digital power supply.
+
+  avdd-supply:
+    description: Regulator for 1.2V analog power supply.
+
+  vddio-supply:
+    description: Regulator for 1.8V IO power supply.
+
+  spi-max-frequency:
+    maximum: 1000000
+
+  spi-cpha: true
+  spi-cpol: true
+
+  clocks:
+    maxItems: 1
+    description: Reference TX_CLK used before PLL is locked.
+
+  solomon,hs-zero-delay-ns:
+    description:
+      HS zero delay period
+    minimum: 0
+    maximum: 1700
+    default: 133
+
+  solomon,hs-prep-delay-ns:
+    description:
+      HS prep delay period
+    minimum: 0
+    maximum: 1728
+    default: 40
+
+  ports:
+    $ref: /schemas/graph.yaml#/properties/ports
+
+    properties:
+      port@0:
+        $ref: /schemas/graph.yaml#/$defs/port-base
+        unevaluatedProperties: false
+        description:
+          Video port for RGB input
+
+        properties:
+          endpoint:
+            $ref: /schemas/graph.yaml#/$defs/endpoint-base
+            unevaluatedProperties: false
+
+            properties:
+              bus-width:
+                enum: [ 16, 18, 24 ]
+
+      port@1:
+        $ref: /schemas/graph.yaml#/properties/port
+        description:
+          Video port for DSI output (panel or connector)
+
+    required:
+      - port@0
+      - port@1
+
+required:
+  - compatible
+  - ports
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    spi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        dsi@2 {
+            compatible = "solomon,ssd2825";
+            reg = <2>;
+
+            spi-max-frequency = <1000000>;
+
+            spi-cpha;
+            spi-cpol;
+
+            reset-gpios = <&gpio 114 GPIO_ACTIVE_LOW>;
+
+            dvdd-supply = <&vdd_1v2>;
+            avdd-supply = <&vdd_1v2>;
+            vddio-supply = <&vdd_1v8_io>;
+
+            solomon,hs-zero-delay-ns = <300>;
+            solomon,hs-prep-delay-ns = <65>;
+
+            clocks = <&ssd2825_tx_clk>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+
+                    bridge_input: endpoint {
+                        remote-endpoint = <&dpi_output>;
+                        bus-width = <24>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+
+                    bridge_output: endpoint {
+                        remote-endpoint = <&panel_input>;
+                    };
+                };
+            };
+        };
+    };

From 55023abe6a2921a8916b623c24208e1971b88729 Mon Sep 17 00:00:00 2001
From: Svyatoslav Ryhel <clamor95@gmail.com>
Date: Wed, 30 Jul 2025 08:54:24 +0300
Subject: [PATCH 0168/1455] drm: bridge: Add support for Solomon SSD2825
 RGB/DSI bridge

SSD2825 is a cost-effective MIPI Bridge Chip solution targeting mainly
smartphones. It can convert 24bit RGB interface into 4-lane MIPI-DSI
interface to drive display modules of up to 800 x 1366, while supporting
AMOLED, a-si LCD or LTPS panel technologies for smartphone applications.

Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250730055424.6718-3-clamor95@gmail.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/Kconfig   |  13 +
 drivers/gpu/drm/bridge/Makefile  |   1 +
 drivers/gpu/drm/bridge/ssd2825.c | 775 +++++++++++++++++++++++++++++++
 3 files changed, 789 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/ssd2825.c

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index b9e0ca85226a..c11444c80d88 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -316,6 +316,19 @@ config DRM_SIMPLE_BRIDGE
 	  Support for non-programmable DRM bridges, such as ADI ADV7123, TI
 	  THS8134 and THS8135 or passive resistor ladder DACs.
 
+config DRM_SOLOMON_SSD2825
+	tristate "SSD2825 RGB/DSI bridge"
+	depends on SPI_MASTER && OF
+	select DRM_MIPI_DSI
+	select DRM_KMS_HELPER
+	select DRM_PANEL
+	help
+	  Say Y here if you want support for the Solomon SSD2825 RGB/DSI
+	  SPI bridge driver.
+
+	  Say M here if you want to support this hardware as a module.
+	  The module will be named "ssd2825".
+
 config DRM_THINE_THC63LVD1024
 	tristate "Thine THC63LVD1024 LVDS decoder bridge"
 	depends on OF
diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
index 245e8a27e3fc..19f5a09d9291 100644
--- a/drivers/gpu/drm/bridge/Makefile
+++ b/drivers/gpu/drm/bridge/Makefile
@@ -27,6 +27,7 @@ obj-$(CONFIG_DRM_SIL_SII8620) += sil-sii8620.o
 obj-$(CONFIG_DRM_SII902X) += sii902x.o
 obj-$(CONFIG_DRM_SII9234) += sii9234.o
 obj-$(CONFIG_DRM_SIMPLE_BRIDGE) += simple-bridge.o
+obj-$(CONFIG_DRM_SOLOMON_SSD2825) += ssd2825.o
 obj-$(CONFIG_DRM_THINE_THC63LVD1024) += thc63lvd1024.o
 obj-$(CONFIG_DRM_TOSHIBA_TC358762) += tc358762.o
 obj-$(CONFIG_DRM_TOSHIBA_TC358764) += tc358764.o
diff --git a/drivers/gpu/drm/bridge/ssd2825.c b/drivers/gpu/drm/bridge/ssd2825.c
new file mode 100644
index 000000000000..f2fdbf7c117d
--- /dev/null
+++ b/drivers/gpu/drm/bridge/ssd2825.c
@@ -0,0 +1,775 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/clk.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/mutex.h>
+#include <linux/of.h>
+#include <linux/regulator/consumer.h>
+#include <linux/spi/spi.h>
+#include <linux/units.h>
+
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_of.h>
+#include <drm/drm_panel.h>
+#include <video/mipi_display.h>
+
+#define SSD2825_DEVICE_ID_REG			0xb0
+#define SSD2825_RGB_INTERFACE_CTRL_REG_1	0xb1
+#define SSD2825_RGB_INTERFACE_CTRL_REG_2	0xb2
+#define SSD2825_RGB_INTERFACE_CTRL_REG_3	0xb3
+#define SSD2825_RGB_INTERFACE_CTRL_REG_4	0xb4
+#define SSD2825_RGB_INTERFACE_CTRL_REG_5	0xb5
+#define SSD2825_RGB_INTERFACE_CTRL_REG_6	0xb6
+#define   SSD2825_NON_BURST_EV			BIT(2)
+#define   SSD2825_BURST				BIT(3)
+#define   SSD2825_PCKL_HIGH			BIT(13)
+#define   SSD2825_HSYNC_HIGH			BIT(14)
+#define   SSD2825_VSYNC_HIGH			BIT(15)
+#define SSD2825_CONFIGURATION_REG		0xb7
+#define   SSD2825_CONF_REG_HS			BIT(0)
+#define   SSD2825_CONF_REG_CKE			BIT(1)
+#define   SSD2825_CONF_REG_SLP			BIT(2)
+#define   SSD2825_CONF_REG_VEN			BIT(3)
+#define   SSD2825_CONF_REG_HCLK			BIT(4)
+#define   SSD2825_CONF_REG_CSS			BIT(5)
+#define   SSD2825_CONF_REG_DCS			BIT(6)
+#define   SSD2825_CONF_REG_REN			BIT(7)
+#define   SSD2825_CONF_REG_ECD			BIT(8)
+#define   SSD2825_CONF_REG_EOT			BIT(9)
+#define   SSD2825_CONF_REG_LPE			BIT(10)
+#define SSD2825_VC_CTRL_REG			0xb8
+#define SSD2825_PLL_CTRL_REG			0xb9
+#define SSD2825_PLL_CONFIGURATION_REG		0xba
+#define SSD2825_CLOCK_CTRL_REG			0xbb
+#define SSD2825_PACKET_SIZE_CTRL_REG_1		0xbc
+#define SSD2825_PACKET_SIZE_CTRL_REG_2		0xbd
+#define SSD2825_PACKET_SIZE_CTRL_REG_3		0xbe
+#define SSD2825_PACKET_DROP_REG			0xbf
+#define SSD2825_OPERATION_CTRL_REG		0xc0
+#define SSD2825_MAX_RETURN_SIZE_REG		0xc1
+#define SSD2825_RETURN_DATA_COUNT_REG		0xc2
+#define SSD2825_ACK_RESPONSE_REG		0xc3
+#define SSD2825_LINE_CTRL_REG			0xc4
+#define SSD2825_INTERRUPT_CTRL_REG		0xc5
+#define SSD2825_INTERRUPT_STATUS_REG		0xc6
+#define SSD2825_ERROR_STATUS_REG		0xc7
+#define SSD2825_DATA_FORMAT_REG			0xc8
+#define SSD2825_DELAY_ADJ_REG_1			0xc9
+#define SSD2825_DELAY_ADJ_REG_2			0xca
+#define SSD2825_DELAY_ADJ_REG_3			0xcb
+#define SSD2825_DELAY_ADJ_REG_4			0xcc
+#define SSD2825_DELAY_ADJ_REG_5			0xcd
+#define SSD2825_DELAY_ADJ_REG_6			0xce
+#define SSD2825_HS_TX_TIMER_REG_1		0xcf
+#define SSD2825_HS_TX_TIMER_REG_2		0xd0
+#define SSD2825_LP_RX_TIMER_REG_1		0xd1
+#define SSD2825_LP_RX_TIMER_REG_2		0xd2
+#define SSD2825_TE_STATUS_REG			0xd3
+#define SSD2825_SPI_READ_REG			0xd4
+#define   SSD2825_SPI_READ_REG_RESET		0xfa
+#define SSD2825_PLL_LOCK_REG			0xd5
+#define SSD2825_TEST_REG			0xd6
+#define SSD2825_TE_COUNT_REG			0xd7
+#define SSD2825_ANALOG_CTRL_REG_1		0xd8
+#define SSD2825_ANALOG_CTRL_REG_2		0xd9
+#define SSD2825_ANALOG_CTRL_REG_3		0xda
+#define SSD2825_ANALOG_CTRL_REG_4		0xdb
+#define SSD2825_INTERRUPT_OUT_CTRL_REG		0xdc
+#define SSD2825_RGB_INTERFACE_CTRL_REG_7	0xdd
+#define SSD2825_LANE_CONFIGURATION_REG		0xde
+#define SSD2825_DELAY_ADJ_REG_7			0xdf
+#define SSD2825_INPUT_PIN_CTRL_REG_1		0xe0
+#define SSD2825_INPUT_PIN_CTRL_REG_2		0xe1
+#define SSD2825_BIDIR_PIN_CTRL_REG_1		0xe2
+#define SSD2825_BIDIR_PIN_CTRL_REG_2		0xe3
+#define SSD2825_BIDIR_PIN_CTRL_REG_3		0xe4
+#define SSD2825_BIDIR_PIN_CTRL_REG_4		0xe5
+#define SSD2825_BIDIR_PIN_CTRL_REG_5		0xe6
+#define SSD2825_BIDIR_PIN_CTRL_REG_6		0xe7
+#define SSD2825_BIDIR_PIN_CTRL_REG_7		0xe8
+#define SSD2825_CABC_BRIGHTNESS_CTRL_REG_1	0xe9
+#define SSD2825_CABC_BRIGHTNESS_CTRL_REG_2	0xea
+#define SSD2825_CABC_BRIGHTNESS_STATUS_REG	0xeb
+#define SSD2825_READ_REG			0xff
+
+#define SSD2825_COM_BYTE			0x00
+#define SSD2825_DAT_BYTE			0x01
+
+#define SSD2828_LP_CLOCK_DIVIDER(n)		(((n) - 1) & 0x3f)
+#define SSD2825_LP_MIN_CLK			5000 /* KHz */
+#define SSD2825_REF_MIN_CLK			2000 /* KHz */
+
+static const struct regulator_bulk_data ssd2825_supplies[] = {
+	{ .supply = "dvdd" },
+	{ .supply = "avdd" },
+	{ .supply = "vddio" },
+};
+
+struct ssd2825_dsi_output {
+	struct mipi_dsi_device *dev;
+	struct drm_panel *panel;
+	struct drm_bridge *bridge;
+};
+
+struct ssd2825_priv {
+	struct spi_device *spi;
+	struct device *dev;
+
+	struct gpio_desc *reset_gpio;
+	struct regulator_bulk_data *supplies;
+
+	struct clk *tx_clk;
+
+	struct mipi_dsi_host dsi_host;
+	struct drm_bridge bridge;
+	struct ssd2825_dsi_output output;
+
+	struct mutex mlock;	/* for host transfer operations */
+
+	u32 pd_lines;		/* number of Parallel Port Input Data Lines */
+	u32 dsi_lanes;		/* number of DSI Lanes */
+
+	/* Parameters for PLL programming */
+	u32 pll_freq_kbps;	/* PLL in kbps */
+	u32 nibble_freq_khz;	/* PLL div by 4 */
+
+	u32 hzd;		/* HS Zero Delay in ns*/
+	u32 hpd;		/* HS Prepare Delay is ns */
+};
+
+static inline struct ssd2825_priv *dsi_host_to_ssd2825(struct mipi_dsi_host *host)
+{
+	return container_of(host, struct ssd2825_priv, dsi_host);
+}
+
+static inline struct ssd2825_priv *bridge_to_ssd2825(struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct ssd2825_priv, bridge);
+}
+
+static int ssd2825_write_raw(struct ssd2825_priv *priv, u8 high_byte, u8 low_byte)
+{
+	struct spi_device *spi = priv->spi;
+	u8 tx_buf[2];
+
+	/*
+	 * Low byte is the value, high byte defines type of
+	 * write cycle, 0 for command and 1 for data.
+	 */
+	tx_buf[0] = low_byte;
+	tx_buf[1] = high_byte;
+
+	return spi_write(spi, tx_buf, 2);
+}
+
+static int ssd2825_write_reg(struct ssd2825_priv *priv, u8 reg, u16 command)
+{
+	u8 datal = (command & 0x00FF);
+	u8 datah = (command & 0xFF00) >> 8;
+	int ret;
+
+	/* Command write cycle */
+	ret = ssd2825_write_raw(priv, SSD2825_COM_BYTE, reg);
+	if (ret)
+		return ret;
+
+	/* Data write cycle bits 7-0 */
+	ret = ssd2825_write_raw(priv, SSD2825_DAT_BYTE, datal);
+	if (ret)
+		return ret;
+
+	/* Data write cycle bits 15-8 */
+	ret = ssd2825_write_raw(priv, SSD2825_DAT_BYTE, datah);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int ssd2825_write_dsi(struct ssd2825_priv *priv, const u8 *command, int len)
+{
+	int ret, i;
+
+	ret = ssd2825_write_reg(priv, SSD2825_PACKET_SIZE_CTRL_REG_1, len);
+	if (ret)
+		return ret;
+
+	ret = ssd2825_write_raw(priv, SSD2825_COM_BYTE, SSD2825_PACKET_DROP_REG);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < len; i++) {
+		ret = ssd2825_write_raw(priv, SSD2825_DAT_BYTE, command[i]);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
+static int ssd2825_read_raw(struct ssd2825_priv *priv, u8 cmd, u16 *data)
+{
+	struct spi_device *spi = priv->spi;
+	struct spi_message msg;
+	struct spi_transfer xfer[2];
+	u8 tx_buf[2];
+	u8 rx_buf[2];
+	int ret;
+
+	memset(&xfer, 0, sizeof(xfer));
+
+	tx_buf[1] = (cmd & 0xFF00) >> 8;
+	tx_buf[0] = (cmd & 0x00FF);
+
+	xfer[0].tx_buf = tx_buf;
+	xfer[0].bits_per_word = 9;
+	xfer[0].len = 2;
+
+	xfer[1].rx_buf = rx_buf;
+	xfer[1].bits_per_word = 16;
+	xfer[1].len = 2;
+
+	spi_message_init(&msg);
+	spi_message_add_tail(&xfer[0], &msg);
+	spi_message_add_tail(&xfer[1], &msg);
+
+	ret = spi_sync(spi, &msg);
+	if (ret) {
+		dev_err(&spi->dev, "ssd2825 read raw failed %d\n", ret);
+		return ret;
+	}
+
+	*data = rx_buf[1] | (rx_buf[0] << 8);
+
+	return 0;
+}
+
+static int ssd2825_read_reg(struct ssd2825_priv *priv, u8 reg, u16 *data)
+{
+	int ret;
+
+	/* Reset the read register */
+	ret = ssd2825_write_reg(priv, SSD2825_SPI_READ_REG, SSD2825_SPI_READ_REG_RESET);
+	if (ret)
+		return ret;
+
+	/* Push the address to read */
+	ret = ssd2825_write_raw(priv, SSD2825_COM_BYTE, reg);
+	if (ret)
+		return ret;
+
+	/* Perform a reading cycle */
+	ret = ssd2825_read_raw(priv, SSD2825_SPI_READ_REG_RESET, data);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int ssd2825_dsi_host_attach(struct mipi_dsi_host *host, struct mipi_dsi_device *dev)
+{
+	struct ssd2825_priv *priv = dsi_host_to_ssd2825(host);
+	struct drm_bridge *bridge;
+	struct drm_panel *panel;
+	struct device_node *ep;
+	int ret;
+
+	if (dev->lanes > 4) {
+		dev_err(priv->dev, "unsupported number of data lanes(%u)\n", dev->lanes);
+		return -EINVAL;
+	}
+
+	/*
+	 * ssd2825 supports both Video and Pulse mode, but the driver only
+	 * implements Video (event) mode currently
+	 */
+	if (!(dev->mode_flags & MIPI_DSI_MODE_VIDEO)) {
+		dev_err(priv->dev, "Only MIPI_DSI_MODE_VIDEO is supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	ret = drm_of_find_panel_or_bridge(host->dev->of_node, 1, 0, &panel, &bridge);
+	if (ret)
+		return ret;
+
+	if (panel) {
+		bridge = drm_panel_bridge_add_typed(panel, DRM_MODE_CONNECTOR_DSI);
+		if (IS_ERR(bridge))
+			return PTR_ERR(bridge);
+	}
+
+	priv->output.dev = dev;
+	priv->output.bridge = bridge;
+	priv->output.panel = panel;
+
+	priv->dsi_lanes = dev->lanes;
+
+	/* get input ep (port0/endpoint0) */
+	ret = -EINVAL;
+	ep = of_graph_get_endpoint_by_regs(host->dev->of_node, 0, 0);
+	if (ep) {
+		ret = of_property_read_u32(ep, "bus-width", &priv->pd_lines);
+		of_node_put(ep);
+	}
+
+	if (ret)
+		priv->pd_lines = mipi_dsi_pixel_format_to_bpp(dev->format);
+
+	drm_bridge_add(&priv->bridge);
+
+	return 0;
+}
+
+static int ssd2825_dsi_host_detach(struct mipi_dsi_host *host, struct mipi_dsi_device *dev)
+{
+	struct ssd2825_priv *priv = dsi_host_to_ssd2825(host);
+
+	drm_bridge_remove(&priv->bridge);
+	if (priv->output.panel)
+		drm_panel_bridge_remove(priv->output.bridge);
+
+	return 0;
+}
+
+static ssize_t ssd2825_dsi_host_transfer(struct mipi_dsi_host *host,
+					 const struct mipi_dsi_msg *msg)
+{
+	struct ssd2825_priv *priv = dsi_host_to_ssd2825(host);
+	u16 config;
+	int ret;
+
+	if (msg->rx_len) {
+		dev_warn(priv->dev, "MIPI rx is not supported\n");
+		return -EOPNOTSUPP;
+	}
+
+	guard(mutex)(&priv->mlock);
+
+	ret = ssd2825_read_reg(priv, SSD2825_CONFIGURATION_REG, &config);
+	if (ret)
+		return ret;
+
+	switch (msg->type) {
+	case MIPI_DSI_DCS_SHORT_WRITE:
+	case MIPI_DSI_DCS_SHORT_WRITE_PARAM:
+	case MIPI_DSI_DCS_LONG_WRITE:
+		config |= SSD2825_CONF_REG_DCS;
+		break;
+	case MIPI_DSI_GENERIC_SHORT_WRITE_0_PARAM:
+	case MIPI_DSI_GENERIC_SHORT_WRITE_1_PARAM:
+	case MIPI_DSI_GENERIC_SHORT_WRITE_2_PARAM:
+	case MIPI_DSI_GENERIC_LONG_WRITE:
+		config &= ~SSD2825_CONF_REG_DCS;
+		break;
+	case MIPI_DSI_DCS_READ:
+	case MIPI_DSI_GENERIC_READ_REQUEST_0_PARAM:
+	case MIPI_DSI_GENERIC_READ_REQUEST_1_PARAM:
+	case MIPI_DSI_GENERIC_READ_REQUEST_2_PARAM:
+	default:
+		return 0;
+	}
+
+	ret = ssd2825_write_reg(priv, SSD2825_CONFIGURATION_REG, config);
+	if (ret)
+		return ret;
+
+	ret = ssd2825_write_reg(priv, SSD2825_VC_CTRL_REG, 0x0000);
+	if (ret)
+		return ret;
+
+	ret = ssd2825_write_dsi(priv, msg->tx_buf, msg->tx_len);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static const struct mipi_dsi_host_ops ssd2825_dsi_host_ops = {
+	.attach = ssd2825_dsi_host_attach,
+	.detach = ssd2825_dsi_host_detach,
+	.transfer = ssd2825_dsi_host_transfer,
+};
+
+static void ssd2825_hw_reset(struct ssd2825_priv *priv)
+{
+	gpiod_set_value_cansleep(priv->reset_gpio, 1);
+	usleep_range(5000, 6000);
+	gpiod_set_value_cansleep(priv->reset_gpio, 0);
+	usleep_range(5000, 6000);
+}
+
+/*
+ * PLL configuration register settings.
+ *
+ * See the "PLL Configuration Register Description" in the SSD2825 datasheet.
+ */
+static u16 construct_pll_config(struct ssd2825_priv *priv,
+				u32 desired_pll_freq_kbps, u32 reference_freq_khz)
+{
+	u32 div_factor = 1, mul_factor, fr = 0;
+
+	while (reference_freq_khz / (div_factor + 1) >= SSD2825_REF_MIN_CLK)
+		div_factor++;
+	if (div_factor > 31)
+		div_factor = 31;
+
+	mul_factor = DIV_ROUND_UP(desired_pll_freq_kbps * div_factor,
+				  reference_freq_khz);
+
+	priv->pll_freq_kbps = reference_freq_khz * mul_factor / div_factor;
+	priv->nibble_freq_khz = priv->pll_freq_kbps / 4;
+
+	if (priv->pll_freq_kbps >= 501000)
+		fr = 3;
+	else if (priv->pll_freq_kbps >= 251000)
+		fr = 2;
+	else if (priv->pll_freq_kbps >= 126000)
+		fr = 1;
+
+	return (fr << 14) | (div_factor << 8) | mul_factor;
+}
+
+static int ssd2825_setup_pll(struct ssd2825_priv *priv,
+			     const struct drm_display_mode *mode)
+{
+	u16 pll_config, lp_div;
+	u32 nibble_delay, pclk_mult, tx_freq_khz;
+	u8 hzd, hpd;
+
+	tx_freq_khz = clk_get_rate(priv->tx_clk) / KILO;
+	if (!tx_freq_khz)
+		tx_freq_khz = SSD2825_REF_MIN_CLK;
+
+	pclk_mult = priv->pd_lines / priv->dsi_lanes + 1;
+	pll_config = construct_pll_config(priv, pclk_mult * mode->clock,
+					  tx_freq_khz);
+
+	lp_div = priv->pll_freq_kbps / (SSD2825_LP_MIN_CLK * 8);
+
+	/* nibble_delay in nanoseconds */
+	nibble_delay = MICRO / priv->nibble_freq_khz;
+
+	hzd = priv->hzd / nibble_delay;
+	hpd = (priv->hpd - 4 * nibble_delay) / nibble_delay;
+
+	/* Disable PLL */
+	ssd2825_write_reg(priv, SSD2825_PLL_CTRL_REG, 0x0000);
+	ssd2825_write_reg(priv, SSD2825_LINE_CTRL_REG, 0x0001);
+
+	/* Set delays */
+	ssd2825_write_reg(priv, SSD2825_DELAY_ADJ_REG_1, (hzd << 8) | hpd);
+
+	/* Set PLL coefficients */
+	ssd2825_write_reg(priv, SSD2825_PLL_CONFIGURATION_REG, pll_config);
+
+	/* Clock Control Register */
+	ssd2825_write_reg(priv, SSD2825_CLOCK_CTRL_REG,
+			  SSD2828_LP_CLOCK_DIVIDER(lp_div));
+
+	/* Enable PLL */
+	ssd2825_write_reg(priv, SSD2825_PLL_CTRL_REG, 0x0001);
+	ssd2825_write_reg(priv, SSD2825_VC_CTRL_REG, 0);
+
+	return 0;
+}
+
+static void ssd2825_bridge_atomic_pre_enable(struct drm_bridge *bridge,
+					     struct drm_atomic_state *state)
+{
+	struct ssd2825_priv *priv = bridge_to_ssd2825(bridge);
+	struct mipi_dsi_device *dsi_dev = priv->output.dev;
+	const struct drm_crtc_state *crtc_state;
+	const struct drm_display_mode *mode;
+	struct drm_connector *connector;
+	struct drm_crtc *crtc;
+	u32 input_bus_flags = bridge->timings->input_bus_flags;
+	u16 flags = 0, config;
+	u8 pixel_format;
+	int ret;
+
+	/* Power Sequence */
+	ret = clk_prepare_enable(priv->tx_clk);
+	if (ret)
+		dev_err(priv->dev, "error enabling tx_clk (%d)\n", ret);
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(ssd2825_supplies), priv->supplies);
+	if (ret)
+		dev_err(priv->dev, "error enabling regulators (%d)\n", ret);
+
+	usleep_range(1000, 2000);
+
+	ssd2825_hw_reset(priv);
+
+	/* Perform SW reset */
+	ssd2825_write_reg(priv, SSD2825_OPERATION_CTRL_REG, 0x0100);
+
+	/* Set pixel format */
+	switch (dsi_dev->format) {
+	case MIPI_DSI_FMT_RGB565:
+		pixel_format = 0x00;
+		break;
+	case MIPI_DSI_FMT_RGB666_PACKED:
+		pixel_format = 0x01;
+		break;
+	case MIPI_DSI_FMT_RGB666:
+		pixel_format = 0x02;
+		break;
+	case MIPI_DSI_FMT_RGB888:
+	default:
+		pixel_format = 0x03;
+		break;
+	}
+
+	connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	crtc = drm_atomic_get_new_connector_state(state, connector)->crtc;
+	crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+	mode = &crtc_state->adjusted_mode;
+
+	/* Set panel timings */
+	ssd2825_write_reg(priv, SSD2825_RGB_INTERFACE_CTRL_REG_1,
+			  ((mode->vtotal - mode->vsync_end) << 8) |
+			  (mode->htotal - mode->hsync_end));
+	ssd2825_write_reg(priv, SSD2825_RGB_INTERFACE_CTRL_REG_2,
+			  ((mode->vtotal - mode->vsync_start) << 8) |
+			  (mode->htotal - mode->hsync_start));
+	ssd2825_write_reg(priv, SSD2825_RGB_INTERFACE_CTRL_REG_3,
+			  ((mode->vsync_start - mode->vdisplay) << 8) |
+			  (mode->hsync_start - mode->hdisplay));
+	ssd2825_write_reg(priv, SSD2825_RGB_INTERFACE_CTRL_REG_4, mode->hdisplay);
+	ssd2825_write_reg(priv, SSD2825_RGB_INTERFACE_CTRL_REG_5, mode->vdisplay);
+
+	if (mode->flags & DRM_MODE_FLAG_PHSYNC)
+		flags |= SSD2825_HSYNC_HIGH;
+
+	if (mode->flags & DRM_MODE_FLAG_PVSYNC)
+		flags |= SSD2825_VSYNC_HIGH;
+
+	if (dsi_dev->mode_flags & MIPI_DSI_MODE_VIDEO)
+		flags |= SSD2825_NON_BURST_EV;
+
+	if (input_bus_flags & DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE)
+		flags |= SSD2825_PCKL_HIGH;
+
+	ssd2825_write_reg(priv, SSD2825_RGB_INTERFACE_CTRL_REG_6, flags | pixel_format);
+	ssd2825_write_reg(priv, SSD2825_LANE_CONFIGURATION_REG, dsi_dev->lanes - 1);
+	ssd2825_write_reg(priv, SSD2825_TEST_REG, 0x0004);
+
+	/* Call PLL configuration */
+	ssd2825_setup_pll(priv, mode);
+
+	usleep_range(10000, 11000);
+
+	config = SSD2825_CONF_REG_HS | SSD2825_CONF_REG_CKE | SSD2825_CONF_REG_DCS |
+		 SSD2825_CONF_REG_ECD | SSD2825_CONF_REG_EOT;
+
+	if (dsi_dev->mode_flags & MIPI_DSI_MODE_LPM)
+		config &= ~SSD2825_CONF_REG_HS;
+
+	if (dsi_dev->mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET)
+		config &= ~SSD2825_CONF_REG_EOT;
+
+	/* Initial DSI configuration register set */
+	ssd2825_write_reg(priv, SSD2825_CONFIGURATION_REG, config);
+	ssd2825_write_reg(priv, SSD2825_VC_CTRL_REG, 0);
+
+	if (priv->output.panel)
+		drm_panel_enable(priv->output.panel);
+}
+
+static void ssd2825_bridge_atomic_enable(struct drm_bridge *bridge,
+					 struct drm_atomic_state *state)
+{
+	struct ssd2825_priv *priv = bridge_to_ssd2825(bridge);
+	struct mipi_dsi_device *dsi_dev = priv->output.dev;
+	u16 config;
+
+	config = SSD2825_CONF_REG_HS | SSD2825_CONF_REG_DCS |
+		 SSD2825_CONF_REG_ECD | SSD2825_CONF_REG_EOT;
+
+	if (dsi_dev->mode_flags & MIPI_DSI_MODE_VIDEO)
+		config |= SSD2825_CONF_REG_VEN;
+
+	if (dsi_dev->mode_flags & MIPI_DSI_MODE_NO_EOT_PACKET)
+		config &= ~SSD2825_CONF_REG_EOT;
+
+	/* Complete configuration after DSI commands were sent */
+	ssd2825_write_reg(priv, SSD2825_CONFIGURATION_REG, config);
+	ssd2825_write_reg(priv, SSD2825_PLL_CTRL_REG, 0x0001);
+	ssd2825_write_reg(priv, SSD2825_VC_CTRL_REG, 0x0000);
+}
+
+static void ssd2825_bridge_atomic_disable(struct drm_bridge *bridge,
+					  struct drm_atomic_state *state)
+{
+	struct ssd2825_priv *priv = bridge_to_ssd2825(bridge);
+	int ret;
+
+	msleep(100);
+
+	/* Exit DSI configuration register set */
+	ssd2825_write_reg(priv, SSD2825_CONFIGURATION_REG,
+			  SSD2825_CONF_REG_ECD | SSD2825_CONF_REG_EOT);
+	ssd2825_write_reg(priv, SSD2825_VC_CTRL_REG, 0);
+
+	/* HW disable */
+	gpiod_set_value_cansleep(priv->reset_gpio, 1);
+	usleep_range(5000, 6000);
+
+	ret = regulator_bulk_disable(ARRAY_SIZE(ssd2825_supplies),
+				     priv->supplies);
+	if (ret < 0)
+		dev_err(priv->dev, "error disabling regulators (%d)\n", ret);
+
+	clk_disable_unprepare(priv->tx_clk);
+}
+
+static int ssd2825_bridge_attach(struct drm_bridge *bridge, struct drm_encoder *encoder,
+				 enum drm_bridge_attach_flags flags)
+{
+	struct ssd2825_priv *priv = bridge_to_ssd2825(bridge);
+
+	return drm_bridge_attach(bridge->encoder, priv->output.bridge, bridge,
+				 flags);
+}
+
+static enum drm_mode_status
+ssd2825_bridge_mode_valid(struct drm_bridge *bridge,
+			  const struct drm_display_info *info,
+			  const struct drm_display_mode *mode)
+{
+	if (mode->hdisplay > 1366)
+		return MODE_H_ILLEGAL;
+
+	if (mode->vdisplay > 1366)
+		return MODE_V_ILLEGAL;
+
+	return MODE_OK;
+}
+
+static bool ssd2825_mode_fixup(struct drm_bridge *bridge,
+			       const struct drm_display_mode *mode,
+			       struct drm_display_mode *adjusted_mode)
+{
+	/* Default to positive sync */
+
+	if (!(adjusted_mode->flags &
+	      (DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NHSYNC)))
+		adjusted_mode->flags |= DRM_MODE_FLAG_PHSYNC;
+
+	if (!(adjusted_mode->flags &
+	      (DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_NVSYNC)))
+		adjusted_mode->flags |= DRM_MODE_FLAG_PVSYNC;
+
+	return true;
+}
+
+static const struct drm_bridge_funcs ssd2825_bridge_funcs = {
+	.attach = ssd2825_bridge_attach,
+	.mode_valid = ssd2825_bridge_mode_valid,
+	.mode_fixup = ssd2825_mode_fixup,
+
+	.atomic_pre_enable = ssd2825_bridge_atomic_pre_enable,
+	.atomic_enable = ssd2825_bridge_atomic_enable,
+	.atomic_disable = ssd2825_bridge_atomic_disable,
+
+	.atomic_reset = drm_atomic_helper_bridge_reset,
+	.atomic_duplicate_state = drm_atomic_helper_bridge_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
+};
+
+static const struct drm_bridge_timings default_ssd2825_timings = {
+	.input_bus_flags = DRM_BUS_FLAG_PIXDATA_SAMPLE_POSEDGE
+		 | DRM_BUS_FLAG_SYNC_SAMPLE_NEGEDGE
+		 | DRM_BUS_FLAG_DE_HIGH,
+};
+
+static int ssd2825_probe(struct spi_device *spi)
+{
+	struct ssd2825_priv *priv;
+	struct device *dev = &spi->dev;
+	struct device_node *np = dev->of_node;
+	int ret;
+
+	/* Driver supports only 8 bit 3 Wire mode */
+	spi->bits_per_word = 9;
+
+	ret = spi_setup(spi);
+	if (ret)
+		return ret;
+
+	priv = devm_drm_bridge_alloc(dev, struct ssd2825_priv, bridge, &ssd2825_bridge_funcs);
+	if (IS_ERR(priv))
+		return PTR_ERR(priv);
+
+	spi_set_drvdata(spi, priv);
+
+	priv->spi = spi;
+	priv->dev = dev;
+
+	mutex_init(&priv->mlock);
+
+	priv->tx_clk = devm_clk_get_optional(dev, NULL);
+	if (IS_ERR(priv->tx_clk))
+		return dev_err_probe(dev, PTR_ERR(priv->tx_clk),
+				     "can't retrieve bridge tx_clk\n");
+
+	priv->reset_gpio = devm_gpiod_get_optional(dev, "reset",
+						   GPIOD_OUT_HIGH);
+	if (IS_ERR(priv->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(priv->reset_gpio),
+				     "failed to get reset GPIO\n");
+
+	ret = devm_regulator_bulk_get_const(dev, ARRAY_SIZE(ssd2825_supplies),
+					    ssd2825_supplies, &priv->supplies);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to get regulators\n");
+
+	priv->hzd = 133; /* ns */
+	device_property_read_u32(dev, "solomon,hs-zero-delay-ns", &priv->hzd);
+
+	priv->hpd = 40; /* ns */
+	device_property_read_u32(dev, "solomon,hs-prep-delay-ns", &priv->hpd);
+
+	priv->dsi_host.dev = dev;
+	priv->dsi_host.ops = &ssd2825_dsi_host_ops;
+
+	priv->bridge.timings = &default_ssd2825_timings;
+	priv->bridge.of_node = np;
+
+	return mipi_dsi_host_register(&priv->dsi_host);
+}
+
+static void ssd2825_remove(struct spi_device *spi)
+{
+	struct ssd2825_priv *priv = spi_get_drvdata(spi);
+
+	mipi_dsi_host_unregister(&priv->dsi_host);
+}
+
+static const struct of_device_id ssd2825_of_match[] = {
+	{ .compatible = "solomon,ssd2825" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, ssd2825_of_match);
+
+static struct spi_driver ssd2825_driver = {
+	.driver = {
+		.name = "ssd2825",
+		.of_match_table = ssd2825_of_match,
+	},
+	.probe = ssd2825_probe,
+	.remove = ssd2825_remove,
+};
+module_spi_driver(ssd2825_driver);
+
+MODULE_AUTHOR("Svyatoslav Ryhel <clamor95@gmail.com>");
+MODULE_DESCRIPTION("Solomon SSD2825 RGB to MIPI-DSI bridge driver SPI");
+MODULE_LICENSE("GPL");

From 1cda3c755bb7770be07d75949bb0f45fb88651f6 Mon Sep 17 00:00:00 2001
From: Maarten Lankhorst <dev@lankhorst.se>
Date: Tue, 15 Jul 2025 17:20:58 +0200
Subject: [PATCH 0169/1455] drm/xe: Fix oops in xe_gem_fault when running
 core_hotunplug test.

I saw an oops in xe_gem_fault when running the xe-fast-feedback
testlist against the realtime kernel without debug options enabled.

The panic happens after core_hotunplug unbind-rebind finishes.
Presumably what happens is that a process mmaps, unlocks because
of the FAULT_FLAG_RETRY_NOWAIT logic, has no process memory left,
causing ttm_bo_vm_dummy_page() to return VM_FAULT_NOPAGE, since
there was nothing left to populate, and then oopses in
"mem_type_is_vram(tbo->resource->mem_type)" because tbo->resource
is NULL.

It's convoluted, but fits the data and explains the oops after
the test exits.

Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250715152057.23254-2-dev@lankhorst.se
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
---
 drivers/gpu/drm/xe/xe_bo.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index ffca1cea5585..1a72561f987c 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1732,22 +1732,26 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
 		ret = ttm_bo_vm_fault_reserved(vmf, vmf->vma->vm_page_prot,
 					       TTM_BO_VM_NUM_PREFAULT);
 		drm_dev_exit(idx);
+
+		if (ret == VM_FAULT_RETRY &&
+		    !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT))
+			goto out;
+
+		/*
+		 * ttm_bo_vm_reserve() already has dma_resv_lock.
+		 */
+		if (ret == VM_FAULT_NOPAGE &&
+		    mem_type_is_vram(tbo->resource->mem_type)) {
+			mutex_lock(&xe->mem_access.vram_userfault.lock);
+			if (list_empty(&bo->vram_userfault_link))
+				list_add(&bo->vram_userfault_link,
+					 &xe->mem_access.vram_userfault.list);
+			mutex_unlock(&xe->mem_access.vram_userfault.lock);
+		}
 	} else {
 		ret = ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot);
 	}
 
-	if (ret == VM_FAULT_RETRY && !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT))
-		goto out;
-	/*
-	 * ttm_bo_vm_reserve() already has dma_resv_lock.
-	 */
-	if (ret == VM_FAULT_NOPAGE && mem_type_is_vram(tbo->resource->mem_type)) {
-		mutex_lock(&xe->mem_access.vram_userfault.lock);
-		if (list_empty(&bo->vram_userfault_link))
-			list_add(&bo->vram_userfault_link, &xe->mem_access.vram_userfault.list);
-		mutex_unlock(&xe->mem_access.vram_userfault.lock);
-	}
-
 	dma_resv_unlock(tbo->base.resv);
 out:
 	if (needs_rpm)

From 934452cbb16e1e3609ba52acb48c503b9aaf3154 Mon Sep 17 00:00:00 2001
From: Robert Mader <robert.mader@collabora.com>
Date: Fri, 18 Jul 2025 14:14:42 +0200
Subject: [PATCH 0170/1455] drm/vkms: Add writeback encoders as possible clones
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since commit 41b4b11da0215 ("drm: Add valid clones check") setting
the `possible_clones` values is a hard requirement for cloning.
`vkms` supports cloning for writeback connectors in order to capture
CRTC content, however that broke with said commit.

Writeback connectors are created on a per-CRTC basis, thus mark
every non-writeback connector that is compatible with a given CRTC
as possible clone - and vice-versa.

Using a default configuration, the corresponding `drm_info` output
changes from:

├───Encoders
│   ├───Encoder 0
│   │   ├───Object ID: 40
│   │   ├───Type: virtual
│   │   ├───CRTCS: {0}
│   │   └───Clones: {0}
│   └───Encoder 1
│       ├───Object ID: 41
│       ├───Type: virtual
│       ├───CRTCS: {0}
│       └───Clones: {1}

into:

├───Encoders
│   ├───Encoder 0
│   │   ├───Object ID: 44
│   │   ├───Type: virtual
│   │   ├───CRTCS: {0}
│   │   └───Clones: {0, 1}
│   └───Encoder 1
│       ├───Object ID: 50
│       ├───Type: virtual
│       ├───CRTCS: {0}
│       └───Clones: {0, 1}

Fixes: dbd9d80c1b2e ("drm/vkms: Add support for writeback")
Signed-off-by: Robert Mader <robert.mader@collabora.com>
Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Link: https://lore.kernel.org/r/20250718121442.490634-1-robert.mader@collabora.com
[fixed Fixes line and changed to: to into: to avoid checkpatch warnings]
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_output.c    | 13 +++++++++++++
 drivers/gpu/drm/vkms/vkms_writeback.c |  2 ++
 2 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/vkms/vkms_output.c b/drivers/gpu/drm/vkms/vkms_output.c
index 8d7ca0cdd79f..2ee3749e2b28 100644
--- a/drivers/gpu/drm/vkms/vkms_output.c
+++ b/drivers/gpu/drm/vkms/vkms_output.c
@@ -77,9 +77,22 @@ int vkms_output_init(struct vkms_device *vkmsdev)
 			return ret;
 		}
 
+		encoder_cfg->encoder->possible_clones |=
+			drm_encoder_mask(encoder_cfg->encoder);
+
 		vkms_config_encoder_for_each_possible_crtc(encoder_cfg, idx, possible_crtc) {
 			encoder_cfg->encoder->possible_crtcs |=
 				drm_crtc_mask(&possible_crtc->crtc->crtc);
+
+			if (vkms_config_crtc_get_writeback(possible_crtc)) {
+				struct drm_encoder *wb_encoder =
+					&possible_crtc->crtc->wb_encoder;
+
+				encoder_cfg->encoder->possible_clones |=
+					drm_encoder_mask(wb_encoder);
+				wb_encoder->possible_clones |=
+					drm_encoder_mask(encoder_cfg->encoder);
+			}
 		}
 	}
 
diff --git a/drivers/gpu/drm/vkms/vkms_writeback.c b/drivers/gpu/drm/vkms/vkms_writeback.c
index fe163271d5b5..45d69a3b85f6 100644
--- a/drivers/gpu/drm/vkms/vkms_writeback.c
+++ b/drivers/gpu/drm/vkms/vkms_writeback.c
@@ -174,6 +174,8 @@ int vkms_enable_writeback_connector(struct vkms_device *vkmsdev,
 	if (ret)
 		return ret;
 	vkms_output->wb_encoder.possible_crtcs |= drm_crtc_mask(&vkms_output->crtc);
+	vkms_output->wb_encoder.possible_clones |=
+		drm_encoder_mask(&vkms_output->wb_encoder);
 
 	drm_connector_helper_add(&wb->base, &vkms_wb_conn_helper_funcs);
 

From 518867b09394217d13f6e05f704450bd9d2c8eeb Mon Sep 17 00:00:00 2001
From: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Date: Thu, 31 Jul 2025 18:59:39 +0800
Subject: [PATCH 0171/1455] drm/panel-edp: Add edp panels used by mt8189
 Chromebooks

Add a few generic edp panels used by mt8189 chromebooks, most of
them use the same general enable timing 50ms. For BOE-NV116WHM-N4B and
BOE-NV116WHM-T01 CMN-N140JCA-ELP the enable timing required 80ms. For
CMN-N116BCA-EAK, the enable timing required 200ms and disable timing
required 50ms. For CMN-N116BCL-EAK and CMN-N122JCA-ENK the enable timing
required 80ms and disable timing required 50ms. For TMA-TL140VDMS03-01,
the enable timing required 50ms and the disable timing required 100ms.

AUO B122UAN01.0:
edid-decode (hex):
00 ff ff ff ff ff ff 00 06 af a4 04 00 00 00 00
31 20 01 04 a5 1a 10 78 03 54 c5 9d 54 55 8f 25
22 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 fa 3c 80 b8 70 b0 24 40 10 10
3e 00 06 a4 10 00 00 18 00 00 00 fd 00 28 3c 4b
4b 10 01 0a 20 20 20 20 20 20 00 00 00 fe 00 41
55 4f 0a 20 20 20 20 20 20 20 20 20 00 00 00 fe
00 42 31 32 32 55 41 4e 30 31 2e 30 20 0a 01 7c

AUO B116XAK02.0:
edid-decode (hex):

00 ff ff ff ff ff ff 00 06 af b0 52 00 00 00 00
2e 21 01 04 95 1a 0e 78 03 5b 35 9f 59 55 8e 26
25 50 54 00 00 00 01 01 01 01
01 01 01 01 01 01
01 01 01 01 01 01 5d 1c 56 a0 50 00 19 30 30 20
46 00 00 90 10 00 00 18 00 00 00 fd 00 28 3c 30
30 07 01 0a 20 20 20 20 20 20 00 00 00 fe 00 41
55 4f 0a 20 20 20 20 20 20 20 20 20 00 00 00 fe
00 42 31 31 36 58 41 4b 30 32 2e 30 20 0a 00 bd

AUO B140UAN08.5:
edid-decode (hex):

00 ff ff ff ff ff ff 00 06 af ba 8b 00 00 00 00
10 23 01 04 a5 1e 13 78 03 7c f2 90 57 59 93 29
1d 51 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 26 3d 80 b8 70 b0 28 40 10 10
3e 00 2d bc 10 00 00 18 00 00 00 fd 00 28 3c 4b
4b 10 01 0a 20 20 20 20 20 20 00 00 00 fe 00 41
55 4f 0a 20 20 20 20 20 20 20 20 20 00 00 00 fc
00 42 31 34 30 55 41 4e 30 38 2e 35 20 0a 01 29

70 20 79 02 00 22 00 14 7b 63 02 85 7f 07 b7 00
0f 80 0f 00 af 04 27 00 02 00 0d 00 25 01 09 7b
63 02 7b 63 02 28 3c 80 2b 00 0c 27 00 28 3b 00
00 27 00 28 3b 00 00 81 00 15 74 1a 00 00 03 01
28 3c 00 00 60 51 60 51 3c 00 00 00 00 8d 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 31 90

AUO B140UAX01.2:
edid-decode (hex):

00 ff ff ff ff ff ff 00 06 af ba cd 00 00 00 00
32 23 01 04 a5 1e 13 78 02 ca 31 9b 5c 58 8d 26
21 4f 52 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 60 3f 80 a0 70 b0 64 40 30 20
96 00 2d bc 10 00 00 18 00 00 00 fd 00 28 3c 4e
4e 10 01 0a 20 20 20 20 20 20 00 00 00 fe 00 41
55 4f 0a 20 20 20 20 20 20 20 20 20 00 00 00 fc
00 42 31 34 30 55 41 58 30 31 2e 32 20 0a 00 46

BOE NV116WHM-N4B:
edid-decode (hex):
00 ff ff ff ff ff ff 00 09 e5 45 0d 00 00 00 00
1f 22 01 04 95 1a 0e 78 03 0b 55 9a 5f 58 95 28
1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 09 1e 56 dc 50 00 28 30 30 20
36 00 00 90 10 00 00 1a 00 00 00 fd 00 28 3c 30
30 08 01 0a 20 20 20 20 20 20 00 00 00 fe 00 42
4f 45 20 43 51 0a 20 20 20 20 20 20 00 00 00 fc
00 4e 56 31 31 36 57 48 4d 2d 4e 34 42 0a 01 c1

70 20 79 02 00 81 00 15 74 1a 00 00 03 01 28 3c
00 00 4b 51 4b 51 3c 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 65 90

BOE NV116WHM-T01:
edid-decode (hex):

00 ff ff ff ff ff ff 00 09 e5 df 0d 00 00 00 00
01 1c 01 04 95 1a 0e 78 0a 81 15 96 59 5a 9a 29
1f 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 6b 1b 56 64 50 00 1e 30 26 18
44 00 00 90 10 00 00 1a ef 15 56 64 50 00 1e 30
26 18 44 00 00 90 10 00 00 00 00 00 00 fe 00 42
4f 45 20 48 46 0a 20 20 20 20 20 20 00 00 00 fe
00 4e 56 31 31 36 57 48 32 2d 4d 30 30 0a 00 83

CMN N116BCL-EAK:
edid-decode (hex):

00 ff ff ff ff ff ff 00 0d ae 5f 11 00 00 00 00
08 22 01 04 95 1a 0e 78 03 46 a5 9c 5b 53 8b 24
1d 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 e6 1e 56 e2 50 00 3c 30 30 20
a6 00 00 90 10 00 00 1a 00 00 00 fd 00 28 3c 32
32 08 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43
4d 4e 0a 20 20 20 20 20 20 20 20 20 00 00 00 fe
00 4e 31 31 36 42 43 4c 2d 45 41 4b 0a 20 01 9b

70 20 79 02 00 25 01 09 fc 34 01 fc 34 01 28 3c
80 81 00 10 72 1a 00 00 03 01 28 3c 00 00 00 00
00 00 3c 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 2f 90

CMN N122JCA-ENK:
edid-decode (hex):

00 ff ff ff ff ff ff 00 0d ae 4c 12 00 00 00 00
11 20 01 04 a5 1a 10 78 03 0a f5 9e 5c 52 8b 24
1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 42 3c 80 a0 70 b0 24 40 30 20
a6 00 06 a4 10 00 00 18 00 00 00 fd 00 28 3c 4a
4a 10 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43
4d 4e 0a 20 20 20 20 20 20 20 20 20 00 00 00 fe
00 4e 31 32 32 4a 43 41 2d 45 4e 4b 0a 20 00 fd

CMN N140JCA-ELP:
edid-decode (hex):
00 ff ff ff ff ff ff 00 0d ae a8 14 00 00 00 00
1d 23 01 04 a5 1e 13 78 03 28 65 97 59 54 8e 27
1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 42 3c 80 a0 70 b0 24 40 30 20
a6 00 2d bc 10 00 00 18 00 00 00 fd 00 28 3c 4a
4a 10 01 0a 20 20 20 20 20 20 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fc
00 4e 31 34 30 4a 43 41 2d 45 4c 50 0a 20 01 c2

70 20 79 02 00 25 01 09 94 5a 02 94 5a 02 28 3c
80 81 00 15 74 1a 00 00 03 01 28 3c 00 00 00 00
00 00 3c 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 aa 90

CMN N116BCA-EAK:
edid-decode (hex):
00 ff ff ff ff ff ff 00 0d ae 02 74 00 00 00 00
2a 22 01 04 95 1a 0e 78 03 67 75 98 59 53 90 27
1c 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 da 1d 56 e2 50 00 20 30 30 20
a6 00 04 8c 10 00 00 1a 00 00 00 fd 00 28 3c 30
30 08 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43
4d 4e 0a 20 20 20 20 20 20 20 20 20 00 00 00 fe
00 4e 31 31 36 42 43 41 2d 45 41 4b 0a 20 00 ba

CSW MNE007QS5-2:
edid-decode (hex):
00 ff ff ff ff ff ff 00 0e 77 62 14 00 00 00 00
10 23 01 04 a5 1e 13 78 03 1c 2e 93 5f 58 95 28
1f 4f 58 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 ea 3d 80 c8 70 b0 2e 40 30 20
36 00 2e bc 10 00 00 1a 00 00 00 fd 00 28 3c 4b
4b 10 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43
53 4f 54 20 54 39 0a 20 20 20 20 20 00 00 00 fc
00 4d 4e 45 30 30 37 51 53 35 2d 32 0a 20 01 8e

70 20 79 02 00 81 00 15 74 1a 00 00 03 01 28 3c
00 00 00 00 00 00 3c 00 00 00 00 80 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 1d 90

CSW MNE007QB2-2:
edid-decode (hex):
00 ff ff ff ff ff ff 00 0e 77 68 14 00 00 00 00
00 23 01 04 a5 1e 13 78 03 90 e0 90 5e 59 86 25
14 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 35 3c 80 a0 70 b0 23 40 30 20
36 00 2d bc 10 00 00 18 00 00 00 fd 00 28 3c 4a
4a 10 01 0a 20 20 20 20 20 20 00 00 00 fc 00 4d
4e 46 33 30 37 51 42 32 2d 32 0a 20 00 00 00 fe
00 43 53 4f 54 20 54 33 0a 20 20 20 20 20 00 9a

TMA TM140VDXP01-04:
edid-decode (hex):

00 ff ff ff ff ff ff 00 51 a1 11 08 00 00 00 00
1a 22 01 04 a5 1e 13 78 03 83 3d 98 5b 57 8d 28
1f 4e 53 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 6d 3d 80 a0 70 b0 3c 40 30 20
36 00 2d bc 10 00 00 1a 00 00 00 10 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 fd 00 28
3c 4c 4c 10 01 0a 20 20 20 20 20 20 00 00 00 fc
00 54 4d 31 34 30 56 44 58 50 30 31 0a 20 01 f7

70 20 79 02 00 25 00 09 41 66 02 41 66 02 28 3c
80 81 00 14 73 1a 00 00 03 01 28 3c 00 00 00 00
00 00 3c 00 00 00 00 8d 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 ae 90

TMA TL140VDMS03-01:
edid-decode (hex):

00 ff ff ff ff ff ff 00 51 a1 94 20 00 00 00 00
0b 23 01 04 a5 1e 13 78 03 47 5a 9e 53 5e 8b 28
23 54 53 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 03 3e 80 a0 70 b0 48 40 30 20
66 0c 2e bd 10 00 00 1e 00 00 00 fd 00 28 3c 4d
4d 10 01 0a 20 20 20 20 20 20 00 00 00 10 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fc
00 54 4c 31 34 30 56 44 4d 53 30 33 0a 20 01 9c

70 20 79 02 00 20 00 0c 00 00 00 94 20 00 00 00
00 0b 19 00 21 00 1d c8 0b 5d 07 80 07 b0 04 00
48 c9 55 48 a5 90 7b 42 21 02 45 54 00 00 00 00
00 00 12 78 26 00 09 02 00 00 00 00 00 01 00 00
2b 00 06 04 00 28 3b 00 00 81 00 14 73 1a 00 00
03 01 28 3c 00 00 00 00 00 00 3c 00 00 00 00 8d
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 ae 90

Signed-off-by: Cong Yang <yangcong5@huaqin.corp-partner.google.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731105939.2692654-1-yangcong5@huaqin.corp-partner.google.com
---
 drivers/gpu/drm/panel/panel-edp.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index d0aa602ecc9d..91f530e91b40 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -1796,6 +1796,13 @@ static const struct panel_delay delay_200_500_e200_d10 = {
 	.disable = 10,
 };
 
+static const struct panel_delay delay_200_500_e200_d50 = {
+	.hpd_absent = 200,
+	.unprepare = 500,
+	.enable = 200,
+	.disable = 50,
+};
+
 static const struct panel_delay delay_200_150_e200 = {
 	.hpd_absent = 200,
 	.unprepare = 150,
@@ -1865,6 +1872,7 @@ static const struct panel_delay delay_200_500_e50_d100 = {
  * Sort first by vendor, then by product ID.
  */
 static const struct edp_panel_entry edp_panels[] = {
+	EDP_PANEL_ENTRY('A', 'U', 'O', 0x04a4, &delay_200_500_e50, "B122UAN01.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x105c, &delay_200_500_e50, "B116XTN01.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x1062, &delay_200_500_e50, "B120XAN01.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x125c, &delay_200_500_e50, "Unknown"),
@@ -1883,6 +1891,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY2('A', 'U', 'O', 0x405c, &auo_b116xak01.delay, "B116XAK01.0",
 			 &auo_b116xa3_mode),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x435c, &delay_200_500_e50, "Unknown"),
+	EDP_PANEL_ENTRY('A', 'U', 'O', 0x52b0, &delay_200_500_e50, "B116XAK02.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x582d, &delay_200_500_e50, "B133UAN01.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x615c, &delay_200_500_e50, "B116XAN06.1"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x635c, &delay_200_500_e50, "B116XAN06.3"),
@@ -1890,10 +1899,12 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x723c, &delay_200_500_e50, "B140XTN07.2"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x73aa, &delay_200_500_e50, "B116XTN02.3"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0x8594, &delay_200_500_e50, "B133UAN01.0"),
+	EDP_PANEL_ENTRY('A', 'U', 'O', 0x8bba, &delay_200_500_e50, "B140UAN08.5"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xa199, &delay_200_500_e50, "B116XAN06.1"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xa7b3, &delay_200_500_e50, "B140UAN04.4"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xc4b4, &delay_200_500_e50, "B116XAT04.1"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xc9a8, &delay_200_500_e50, "B140QAN08.H"),
+	EDP_PANEL_ENTRY('A', 'U', 'O', 0xcdba, &delay_200_500_e50, "B140UAX01.2"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xd497, &delay_200_500_e50, "B120XAN01.0"),
 	EDP_PANEL_ENTRY('A', 'U', 'O', 0xf390, &delay_200_500_e50, "B140XTN07.7"),
 
@@ -1956,7 +1967,9 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0c93, &delay_200_500_e200, "Unknown"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cb6, &delay_200_500_e200, "NT116WHM-N44"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cfa, &delay_200_500_e50, "NV116WHM-A4D"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0d45, &delay_200_500_e80, "NV116WHM-N4B"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0d73, &delay_200_500_e80, "NE140WUM-N6S"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ddf, &delay_200_500_e80, "NV116WHM-T01"),
 
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1130, &delay_200_500_e50, "N116BGE-EB2"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1132, &delay_200_500_e80_d50, "N116BGE-EA2"),
@@ -1974,18 +1987,22 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x115b, &delay_200_500_e80_d50, "N116BCN-EB1"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x115d, &delay_200_500_e80_d50, "N116BCA-EA2"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x115e, &delay_200_500_e80_d50, "N116BCA-EA1"),
+	EDP_PANEL_ENTRY('C', 'M', 'N', 0x115f, &delay_200_500_e80_d50, "N116BCL-EAK"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1160, &delay_200_500_e80_d50, "N116BCJ-EAK"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1161, &delay_200_500_e80, "N116BCP-EA2"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1163, &delay_200_500_e80_d50, "N116BCJ-EAK"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1247, &delay_200_500_e80_d50, "N120ACA-EA1"),
+	EDP_PANEL_ENTRY('C', 'M', 'N', 0x124c, &delay_200_500_e80_d50, "N122JCA-ENK"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x142b, &delay_200_500_e80_d50, "N140HCA-EAC"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x142e, &delay_200_500_e80_d50, "N140BGA-EA4"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x144f, &delay_200_500_e80_d50, "N140HGA-EA1"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1468, &delay_200_500_e80, "N140HGA-EA1"),
+	EDP_PANEL_ENTRY('C', 'M', 'N', 0x14a8, &delay_200_500_e80, "N140JCA-ELP"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x14d4, &delay_200_500_e80_d50, "N140HCA-EAC"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x14d6, &delay_200_500_e80_d50, "N140BGA-EA4"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x14e5, &delay_200_500_e80_d50, "N140HGA-EA1"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x162b, &delay_200_500_e80_d50, "N160JCE-ELL"),
+	EDP_PANEL_ENTRY('C', 'M', 'N', 0x7402, &delay_200_500_e200_d50, "N116BCA-EAK"),
 
 	EDP_PANEL_ENTRY('C', 'S', 'O', 0x1200, &delay_200_500_e50_d50_p2e200, "MNC207QS1-1"),
 	EDP_PANEL_ENTRY('C', 'S', 'O', 0x1413, &delay_200_500_e50_d50_p2e200, "MNE007JA1-2"),
@@ -1995,6 +2012,8 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50_d100, "MNB601LS1-4"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1457, &delay_80_500_e80_p2e200, "MNE007QS3-8"),
+	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1462, &delay_200_500_e50, "MNE007QS5-2"),
+	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1468, &delay_200_500_e50, "MNE007QB2-2"),
 
 	EDP_PANEL_ENTRY('E', 'T', 'C', 0x0000, &delay_50_500_e200_d200_po2e335, "LP079QX1-SP0V"),
 
@@ -2041,6 +2060,9 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('S', 'T', 'A', 0x0009, &delay_200_500_e250, "116QHD024002"),
 	EDP_PANEL_ENTRY('S', 'T', 'A', 0x0100, &delay_100_500_e200, "2081116HHD028001-51D"),
 
+	EDP_PANEL_ENTRY('T', 'M', 'A', 0x0811, &delay_200_500_e80_d50, "TM140VDXP01-04"),
+	EDP_PANEL_ENTRY('T', 'M', 'A', 0x2094, &delay_200_500_e50_d100, "TL140VDMS03-01"),
+
 	{ /* sentinal */ }
 };
 

From d680a7959b9d7cce9526b1b391cc7b13e313a096 Mon Sep 17 00:00:00 2001
From: Aleksandrs Vinarskis <alex.vinarskis@gmail.com>
Date: Thu, 31 Jul 2025 23:55:08 +0200
Subject: [PATCH 0172/1455] dt-bindings: display: panel: samsung,atna40cu11:
 document ATNA40CU11

The Samsung ATNA40CU11 panel is a 14" AMOLED eDP panel. It is
similar to the ATNA33XC20 except that it is larger and has a
different resolution. It is found in some arm64 laptops, eg.
Asus Zenbook A14 UX3407RA.

Raw panel edid:

00 ff ff ff ff ff ff 00 4c 83 9d 41 00 00 00 00
00 20 01 04 b5 1e 13 78 03 cf d1 ae 51 3e b6 23
0b 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 cb fe 40 64 b0 08 38 77 20 08
88 00 2e bd 10 00 00 1b 00 00 00 fd 00 30 78 da
da 42 01 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fe
00 41 54 4e 41 34 30 43 55 31 31 2d 30 20 01 2a

70 20 79 02 00 20 00 0c 4c 83 00 9d 41 00 00 00
00 00 20 00 21 00 1d b8 0b 6c 07 40 0b 08 07 00
ee ea 50 ec d3 b6 3d 42 0b 01 45 54 40 5e d0 60
18 10 23 78 26 00 09 07 06 03 00 00 00 50 00 00
22 00 14 e7 f3 09 85 3f 0b 63 00 1f 00 07 00 07
07 17 00 07 00 07 00 81 00 1f 73 1a 00 00 03 03
30 78 00 a0 74 02 60 02 78 00 00 00 00 8d e3 05
80 00 e6 06 05 01 74 60 02 00 00 00 00 00 91 90

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Aleksandrs Vinarskis <alex.vinarskis@gmail.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731215635.206702-2-alex.vinarskis@gmail.com
---
 .../devicetree/bindings/display/panel/samsung,atna33xc20.yaml   | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml b/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
index e36659340ef3..5e2ce200025f 100644
--- a/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
+++ b/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
@@ -21,6 +21,8 @@ properties:
           - enum:
               # Samsung 13" 3K (2880×1920 pixels) eDP AMOLED panel
               - samsung,atna30dw01
+              # Samsung 14" WQXGA+ (2880x1800 pixels) eDP AMOLED panel
+              - samsung,atna40cu11
               # Samsung 14" WQXGA+ (2880×1800 pixels) eDP AMOLED panel
               - samsung,atna40yk20
               # Samsung 14.5" WQXGA+ (2880x1800 pixels) eDP AMOLED panel

From d7c2aad1254a1b4e6d76947cbdd18813b02a0e24 Mon Sep 17 00:00:00 2001
From: Aleksandrs Vinarskis <alex.vinarskis@gmail.com>
Date: Thu, 31 Jul 2025 23:55:09 +0200
Subject: [PATCH 0173/1455] dt-bindings: display: panel: samsung,atna40ct06:
 document ATNA40CT06

The Samsung ATNA40CT06 panel is a 14" AMOLED eDP panel. It is
similar to the ATNA33XC20 except that it is larger and has a
different resolution. It is found in some arm64 laptops, eg.
Asus Zenbook A14 UX3407QA.

Raw panel edid:

00 ff ff ff ff ff ff 00 4c 83 0d 42 00 00 00 00
00 22 01 04 b5 1e 13 78 02 0c f1 ae 52 3c b9 23
0c 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 42 3c 80 a0 70 b0 24 40 30 20
88 00 2e bd 10 00 00 1b 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fc
00 41 54 4e 41 34 30 43 54 30 36 2d 30 20 01 7d

70 20 79 02 00 20 00 0c 4c 83 00 0d 42 00 00 00
00 00 22 00 21 00 1d ca 0b 5e 07 80 07 b0 04 00
e1 fa 51 cb 13 b9 3d d2 0c 01 45 54 40 5e d0 60
18 10 23 78 26 00 09 07 06 03 00 00 00 50 00 00
22 00 14 8d 5a 02 85 7f 07 9f 00 2f 00 1f 00 af
04 23 00 07 00 07 00 81 00 0b e3 05 80 00 e6 06
05 01 74 60 02 2e 00 06 00 45 40 5e d0 60 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 b0 90

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Aleksandrs Vinarskis <alex.vinarskis@gmail.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731215635.206702-3-alex.vinarskis@gmail.com
---
 .../devicetree/bindings/display/panel/samsung,atna33xc20.yaml   | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml b/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
index 5e2ce200025f..ccb574caed28 100644
--- a/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
+++ b/Documentation/devicetree/bindings/display/panel/samsung,atna33xc20.yaml
@@ -21,6 +21,8 @@ properties:
           - enum:
               # Samsung 13" 3K (2880×1920 pixels) eDP AMOLED panel
               - samsung,atna30dw01
+              # Samsung 14" FHD+ (1920x1200 pixels) eDP AMOLED panel
+              - samsung,atna40ct06
               # Samsung 14" WQXGA+ (2880x1800 pixels) eDP AMOLED panel
               - samsung,atna40cu11
               # Samsung 14" WQXGA+ (2880×1800 pixels) eDP AMOLED panel

From 82928cc1c2b2be16ea6ee9e23799ca182e1cd37c Mon Sep 17 00:00:00 2001
From: Aleksandrs Vinarskis <alex.vinarskis@gmail.com>
Date: Thu, 31 Jul 2025 23:55:10 +0200
Subject: [PATCH 0174/1455] drm/panel-edp: Add BOE NV140WUM-N64

Timings taken from NV140WUM-N41. It is found in some arm64 laptops,
eg. Asus Zenbook A14 UX3407QA.

The raw edid of the panel is:
00 ff ff ff ff ff ff 00 09 e5 f6 0c 00 00 00 00
10 22 01 04 a5 1e 13 78 07 8e 95 a6 52 4c 9d 26
0f 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 5d 30 80 a0 70 b0 28 40 30 20
36 00 2e bc 10 00 00 1a 00 00 00 fd 00 28 3c 4a
4a 0f 01 0a 20 20 20 20 20 20 00 00 00 fe 00 3d
4c 33 30 20 20 20 20 20 20 20 20 ff 00 00 00 fc
00 4e 56 31 34 30 57 55 4d 2d 4e 36 34 0a 01 f8

70 20 79 02 00 21 00 1d c8 0b 5d 07 80 07 b0 04
88 66 ea 51 cc 74 9d 66 52 0f 02 35 54 40 5e 40
5e 00 44 12 78 22 00 14 7f 5c 02 85 7f 07 9f 00
2f 00 1f 00 af 04 27 00 02 00 05 00 2b 00 0c 27
00 28 3b 00 00 27 00 28 2f 00 00 2e 00 06 00 44
40 5e 40 5e 81 00 1e 72 1a 00 00 03 71 28 3c 00
00 60 ff 60 ff 3c 00 00 00 00 e3 05 04 00 e6 06
01 01 60 60 ff 00 00 00 00 00 00 00 00 00 de 90

Signed-off-by: Aleksandrs Vinarskis <alex.vinarskis@gmail.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731215635.206702-4-alex.vinarskis@gmail.com
---
 drivers/gpu/drm/panel/panel-edp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index 91f530e91b40..3e4c3a439988 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -1966,6 +1966,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0c20, &delay_200_500_e80, "NT140FHM-N47"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0c93, &delay_200_500_e200, "Unknown"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cb6, &delay_200_500_e200, "NT116WHM-N44"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cf6, &delay_200_500_e200, "NV140WUM-N64"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0cfa, &delay_200_500_e50, "NV116WHM-A4D"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0d45, &delay_200_500_e80, "NV116WHM-N4B"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0d73, &delay_200_500_e80, "NE140WUM-N6S"),

From be001a68f075ec7b697e1c252b6659f3ab2a30de Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:08 +0300
Subject: [PATCH 0175/1455] drm/i915/hdmi: use intel_de_wait_for_set() instead
 of wait_for()

Prefer the register read specific wait function over i915 wait_for().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/c5d3044114b4464799a2fded18cda7946d95c4f6.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_hdmi.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 03045d188817..cbee628eb26b 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -1582,9 +1582,9 @@ bool intel_hdmi_hdcp_check_link_once(struct intel_digital_port *dig_port,
 	intel_de_write(display, HDCP_RPRIME(display, cpu_transcoder, port), ri.reg);
 
 	/* Wait for Ri prime match */
-	if (wait_for((intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) &
-		      (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC)) ==
-		     (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1)) {
+	ret = intel_de_wait_for_set(display, HDCP_STATUS(display, cpu_transcoder, port),
+				    HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC, 1);
+	if (ret) {
 		drm_dbg_kms(display->drm, "Ri' mismatch detected (%x)\n",
 			    intel_de_read(display, HDCP_STATUS(display, cpu_transcoder,
 							       port)));

From 7327b40c644487182969060894b2be952e2c4ab0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:09 +0300
Subject: [PATCH 0176/1455] drm/i915/ddi: use intel_de_wait_custom() instead of
 wait_for_us()

Prefer the register read specific wait function over i915 wait_for_us().

v2: Wait for bits to clear in mtl_ddi_disable_d2d()

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/232a554db6a327974c06f2491311b28f865467b9.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_ddi.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 0405396c7750..866ed3e46664 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2561,6 +2561,7 @@ mtl_ddi_enable_d2d(struct intel_encoder *encoder)
 	enum port port = encoder->port;
 	i915_reg_t reg;
 	u32 set_bits, wait_bits;
+	int ret;
 
 	if (DISPLAY_VER(display) < 14)
 		return;
@@ -2576,7 +2577,11 @@ mtl_ddi_enable_d2d(struct intel_encoder *encoder)
 	}
 
 	intel_de_rmw(display, reg, 0, set_bits);
-	if (wait_for_us(intel_de_read(display, reg) & wait_bits, 100)) {
+
+	ret = intel_de_wait_custom(display, reg,
+				   wait_bits, wait_bits,
+				   100, 0, NULL);
+	if (ret) {
 		drm_err(display->drm, "Timeout waiting for D2D Link enable for DDI/PORT_BUF_CTL %c\n",
 			port_name(port));
 	}
@@ -3058,6 +3063,7 @@ mtl_ddi_disable_d2d(struct intel_encoder *encoder)
 	enum port port = encoder->port;
 	i915_reg_t reg;
 	u32 clr_bits, wait_bits;
+	int ret;
 
 	if (DISPLAY_VER(display) < 14)
 		return;
@@ -3073,7 +3079,11 @@ mtl_ddi_disable_d2d(struct intel_encoder *encoder)
 	}
 
 	intel_de_rmw(display, reg, clr_bits, 0);
-	if (wait_for_us(!(intel_de_read(display, reg) & wait_bits), 100))
+
+	ret = intel_de_wait_custom(display, reg,
+				   wait_bits, 0,
+				   100, 0, NULL);
+	if (ret)
 		drm_err(display->drm, "Timeout waiting for D2D Link disable for DDI/PORT_BUF_CTL %c\n",
 			port_name(port));
 }

From 7c8d0cbe9808078063adf7112d2c7f9b836bc73d Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:10 +0300
Subject: [PATCH 0177/1455] drm/i915/dpll: use intel_de_wait_custom() instead
 of wait_for_us()

Prefer the register read specific wait function over i915 wait_for_us().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/d8c381524d721e01228b76b71080c6e4ccc528e9.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dpll_mgr.c | 20 +++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
index 33e0398120c8..8ea96cc524a1 100644
--- a/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
+++ b/drivers/gpu/drm/i915/display/intel_dpll_mgr.c
@@ -2046,6 +2046,7 @@ static void bxt_ddi_pll_enable(struct intel_display *display,
 	enum dpio_phy phy = DPIO_PHY0;
 	enum dpio_channel ch = DPIO_CH0;
 	u32 temp;
+	int ret;
 
 	bxt_port_to_phy_channel(display, port, &phy, &ch);
 
@@ -2056,8 +2057,10 @@ static void bxt_ddi_pll_enable(struct intel_display *display,
 		intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port),
 			     0, PORT_PLL_POWER_ENABLE);
 
-		if (wait_for_us((intel_de_read(display, BXT_PORT_PLL_ENABLE(port)) &
-				 PORT_PLL_POWER_STATE), 200))
+		ret = intel_de_wait_custom(display, BXT_PORT_PLL_ENABLE(port),
+					   PORT_PLL_POWER_STATE, PORT_PLL_POWER_STATE,
+					   200, 0, NULL);
+		if (ret)
 			drm_err(display->drm,
 				"Power state not set for PLL:%d\n", port);
 	}
@@ -2119,8 +2122,10 @@ static void bxt_ddi_pll_enable(struct intel_display *display,
 	intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port), 0, PORT_PLL_ENABLE);
 	intel_de_posting_read(display, BXT_PORT_PLL_ENABLE(port));
 
-	if (wait_for_us((intel_de_read(display, BXT_PORT_PLL_ENABLE(port)) & PORT_PLL_LOCK),
-			200))
+	ret = intel_de_wait_custom(display, BXT_PORT_PLL_ENABLE(port),
+				   PORT_PLL_LOCK, PORT_PLL_LOCK,
+				   200, 0, NULL);
+	if (ret)
 		drm_err(display->drm, "PLL %d not locked\n", port);
 
 	if (display->platform.geminilake) {
@@ -2144,6 +2149,7 @@ static void bxt_ddi_pll_disable(struct intel_display *display,
 				struct intel_dpll *pll)
 {
 	enum port port = (enum port)pll->info->id; /* 1:1 port->PLL mapping */
+	int ret;
 
 	intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port), PORT_PLL_ENABLE, 0);
 	intel_de_posting_read(display, BXT_PORT_PLL_ENABLE(port));
@@ -2152,8 +2158,10 @@ static void bxt_ddi_pll_disable(struct intel_display *display,
 		intel_de_rmw(display, BXT_PORT_PLL_ENABLE(port),
 			     PORT_PLL_POWER_ENABLE, 0);
 
-		if (wait_for_us(!(intel_de_read(display, BXT_PORT_PLL_ENABLE(port)) &
-				  PORT_PLL_POWER_STATE), 200))
+		ret = intel_de_wait_custom(display, BXT_PORT_PLL_ENABLE(port),
+					   PORT_PLL_POWER_STATE, 0,
+					   200, 0, NULL);
+		if (ret)
 			drm_err(display->drm,
 				"Power state not reset for PLL:%d\n", port);
 	}

From cc2cce4a8a1fbc5856139b27e1f7138220e395f0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:11 +0300
Subject: [PATCH 0178/1455] drm/i915/cdclk: use intel_de_wait_custom() instead
 of wait_for_us()

Prefer the register read specific wait function over i915 wait_for_us().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/fadd74e9450afff5e32bf921b192f19ea1629fff.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cdclk.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 228aa64c1349..2a2d6714e4f4 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -903,8 +903,10 @@ static void bdw_set_cdclk(struct intel_display *display,
 	 * According to the spec, it should be enough to poll for this 1 us.
 	 * However, extensive testing shows that this can take longer.
 	 */
-	if (wait_for_us(intel_de_read(display, LCPLL_CTL) &
-			LCPLL_CD_SOURCE_FCLK_DONE, 100))
+	ret = intel_de_wait_custom(display, LCPLL_CTL,
+				   LCPLL_CD_SOURCE_FCLK_DONE, LCPLL_CD_SOURCE_FCLK_DONE,
+				   100, 0, NULL);
+	if (ret)
 		drm_err(display->drm, "Switching to FCLK failed\n");
 
 	intel_de_rmw(display, LCPLL_CTL,
@@ -913,8 +915,10 @@ static void bdw_set_cdclk(struct intel_display *display,
 	intel_de_rmw(display, LCPLL_CTL,
 		     LCPLL_CD_SOURCE_FCLK, 0);
 
-	if (wait_for_us((intel_de_read(display, LCPLL_CTL) &
-			 LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1))
+	ret = intel_de_wait_custom(display, LCPLL_CTL,
+				   LCPLL_CD_SOURCE_FCLK_DONE, 0,
+				   1, 0, NULL);
+	if (ret)
 		drm_err(display->drm, "Switching back to LCPLL failed\n");
 
 	intel_pcode_write(display->drm, HSW_PCODE_DE_WRITE_FREQ_REQ,

From 6d13a797bb3b0101a17c276644128d57f24d86b1 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:12 +0300
Subject: [PATCH 0179/1455] drm/i915/power: use intel_de_wait_custom() instead
 of wait_for_us()

Prefer the register read specific wait function over i915 wait_for_us().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/154b681d9545b26453920b155656a65ce685da2a.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_power.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 273054c22325..2c2b23a138cc 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -1278,6 +1278,7 @@ static void hsw_disable_lcpll(struct intel_display *display,
 			      bool switch_to_fclk, bool allow_power_down)
 {
 	u32 val;
+	int ret;
 
 	assert_can_disable_lcpll(display);
 
@@ -1287,8 +1288,10 @@ static void hsw_disable_lcpll(struct intel_display *display,
 		val |= LCPLL_CD_SOURCE_FCLK;
 		intel_de_write(display, LCPLL_CTL, val);
 
-		if (wait_for_us(intel_de_read(display, LCPLL_CTL) &
-				LCPLL_CD_SOURCE_FCLK_DONE, 1))
+		ret = intel_de_wait_custom(display, LCPLL_CTL,
+					   LCPLL_CD_SOURCE_FCLK_DONE, LCPLL_CD_SOURCE_FCLK_DONE,
+					   1, 0, NULL);
+		if (ret)
 			drm_err(display->drm, "Switching to FCLK failed\n");
 
 		val = intel_de_read(display, LCPLL_CTL);
@@ -1324,6 +1327,7 @@ static void hsw_restore_lcpll(struct intel_display *display)
 {
 	struct drm_i915_private __maybe_unused *dev_priv = to_i915(display->drm);
 	u32 val;
+	int ret;
 
 	val = intel_de_read(display, LCPLL_CTL);
 
@@ -1358,8 +1362,10 @@ static void hsw_restore_lcpll(struct intel_display *display)
 	if (val & LCPLL_CD_SOURCE_FCLK) {
 		intel_de_rmw(display, LCPLL_CTL, LCPLL_CD_SOURCE_FCLK, 0);
 
-		if (wait_for_us((intel_de_read(display, LCPLL_CTL) &
-				 LCPLL_CD_SOURCE_FCLK_DONE) == 0, 1))
+		ret = intel_de_wait_custom(display, LCPLL_CTL,
+					   LCPLL_CD_SOURCE_FCLK_DONE, 0,
+					   1, 0, NULL);
+		if (ret)
 			drm_err(display->drm,
 				"Switching back to LCPLL failed\n");
 	}

From 5a37f08d49594b77e3344afaae30e4bca2c1abc6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:13 +0300
Subject: [PATCH 0180/1455] drm/i915/pch: use intel_de_wait_custom() instead of
 wait_for_us()

Prefer the register read specific wait function over i915 wait_for_us().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/67e0afa2c0c5ad39b9108af15d0496394e674518.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_pch_refclk.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_pch_refclk.c b/drivers/gpu/drm/i915/display/intel_pch_refclk.c
index d3c5255bf1a8..9ae53679a041 100644
--- a/drivers/gpu/drm/i915/display/intel_pch_refclk.c
+++ b/drivers/gpu/drm/i915/display/intel_pch_refclk.c
@@ -17,16 +17,22 @@
 
 static void lpt_fdi_reset_mphy(struct intel_display *display)
 {
+	int ret;
+
 	intel_de_rmw(display, SOUTH_CHICKEN2, 0, FDI_MPHY_IOSFSB_RESET_CTL);
 
-	if (wait_for_us(intel_de_read(display, SOUTH_CHICKEN2) &
-			FDI_MPHY_IOSFSB_RESET_STATUS, 100))
+	ret = intel_de_wait_custom(display, SOUTH_CHICKEN2,
+				   FDI_MPHY_IOSFSB_RESET_STATUS, FDI_MPHY_IOSFSB_RESET_STATUS,
+				   100, 0, NULL);
+	if (ret)
 		drm_err(display->drm, "FDI mPHY reset assert timeout\n");
 
 	intel_de_rmw(display, SOUTH_CHICKEN2, FDI_MPHY_IOSFSB_RESET_CTL, 0);
 
-	if (wait_for_us((intel_de_read(display, SOUTH_CHICKEN2) &
-			 FDI_MPHY_IOSFSB_RESET_STATUS) == 0, 100))
+	ret = intel_de_wait_custom(display, SOUTH_CHICKEN2,
+				   FDI_MPHY_IOSFSB_RESET_STATUS, 0,
+				   100, 0, NULL);
+	if (ret)
 		drm_err(display->drm, "FDI mPHY reset de-assert timeout\n");
 }
 

From 13a5a04ffd639602a4e705a3d6c38a5989989640 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 13:05:14 +0300
Subject: [PATCH 0181/1455] drm/i915/dsi: use intel_de_wait_custom() instead of
 wait_for_us()

Prefer the register read specific wait function over i915 wait_for_us().

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/1fe3d5ac314dd644573e9f59941e4c7f1d57b05d.1753956266.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/icl_dsi.c | 41 ++++++++++++++++++--------
 1 file changed, 28 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c
index 8d9cb73a93a7..44310984bb57 100644
--- a/drivers/gpu/drm/i915/display/icl_dsi.c
+++ b/drivers/gpu/drm/i915/display/icl_dsi.c
@@ -137,8 +137,11 @@ static void wait_for_cmds_dispatched_to_panel(struct intel_encoder *encoder)
 	/* wait for LP TX in progress bit to be cleared */
 	for_each_dsi_port(port, intel_dsi->ports) {
 		dsi_trans = dsi_port_to_transcoder(port);
-		if (wait_for_us(!(intel_de_read(display, DSI_LP_MSG(dsi_trans)) &
-				  LPTX_IN_PROGRESS), 20))
+
+		ret = intel_de_wait_custom(display, DSI_LP_MSG(dsi_trans),
+					   LPTX_IN_PROGRESS, 0,
+					   20, 0, NULL);
+		if (ret)
 			drm_err(display->drm, "LPTX bit not cleared\n");
 	}
 }
@@ -516,13 +519,15 @@ static void gen11_dsi_enable_ddi_buffer(struct intel_encoder *encoder)
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
+	int ret;
 
 	for_each_dsi_port(port, intel_dsi->ports) {
 		intel_de_rmw(display, DDI_BUF_CTL(port), 0, DDI_BUF_CTL_ENABLE);
 
-		if (wait_for_us(!(intel_de_read(display, DDI_BUF_CTL(port)) &
-				  DDI_BUF_IS_IDLE),
-				  500))
+		ret = intel_de_wait_custom(display, DDI_BUF_CTL(port),
+					   DDI_BUF_IS_IDLE, 0,
+					   500, 0, NULL);
+		if (ret)
 			drm_err(display->drm, "DDI port:%c buffer idle\n",
 				port_name(port));
 	}
@@ -838,9 +843,14 @@ gen11_dsi_configure_transcoder(struct intel_encoder *encoder,
 
 	/* wait for link ready */
 	for_each_dsi_port(port, intel_dsi->ports) {
+		int ret;
+
 		dsi_trans = dsi_port_to_transcoder(port);
-		if (wait_for_us((intel_de_read(display, DSI_TRANS_FUNC_CONF(dsi_trans)) &
-				 LINK_READY), 2500))
+
+		ret = intel_de_wait_custom(display, DSI_TRANS_FUNC_CONF(dsi_trans),
+					   LINK_READY, LINK_READY,
+					   2500, 0, NULL);
+		if (ret)
 			drm_err(display->drm, "DSI link not ready\n");
 	}
 }
@@ -1321,6 +1331,7 @@ static void gen11_dsi_deconfigure_trancoder(struct intel_encoder *encoder)
 	enum port port;
 	enum transcoder dsi_trans;
 	u32 tmp;
+	int ret;
 
 	/* disable periodic update mode */
 	if (is_cmd_mode(intel_dsi)) {
@@ -1337,9 +1348,10 @@ static void gen11_dsi_deconfigure_trancoder(struct intel_encoder *encoder)
 		tmp &= ~LINK_ULPS_TYPE_LP11;
 		intel_de_write(display, DSI_LP_MSG(dsi_trans), tmp);
 
-		if (wait_for_us((intel_de_read(display, DSI_LP_MSG(dsi_trans)) &
-				 LINK_IN_ULPS),
-				10))
+		ret = intel_de_wait_custom(display, DSI_LP_MSG(dsi_trans),
+					   LINK_IN_ULPS, LINK_IN_ULPS,
+					   10, 0, NULL);
+		if (ret)
 			drm_err(display->drm, "DSI link not in ULPS\n");
 	}
 
@@ -1367,14 +1379,17 @@ static void gen11_dsi_disable_port(struct intel_encoder *encoder)
 	struct intel_display *display = to_intel_display(encoder);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
+	int ret;
 
 	gen11_dsi_ungate_clocks(encoder);
 	for_each_dsi_port(port, intel_dsi->ports) {
 		intel_de_rmw(display, DDI_BUF_CTL(port), DDI_BUF_CTL_ENABLE, 0);
 
-		if (wait_for_us((intel_de_read(display, DDI_BUF_CTL(port)) &
-				 DDI_BUF_IS_IDLE),
-				 8))
+		ret = intel_de_wait_custom(display, DDI_BUF_CTL(port),
+					   DDI_BUF_IS_IDLE, DDI_BUF_IS_IDLE,
+					   8, 0, NULL);
+
+		if (ret)
 			drm_err(display->drm,
 				"DDI port:%c buffer not idle\n",
 				port_name(port));

From 7d3a5962d74ee7ce2d0051d4d1e106c06b432ce8 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Fri, 1 Aug 2025 16:32:35 +0300
Subject: [PATCH 0182/1455] drm/xe/vf: Fix IS_ERR() vs NULL check in
 xe_sriov_vf_ccs_init()

The xe_migrate_alloc() function returns NULL on error.  It doesn't return
error pointers.  Update the checking to match.

Fixes: a843b9894705 ("drm/xe/vf: Fix VM crash during VF driver release")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/aIzB8-Y6wtZvfNQT@stanley.mountain
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index bf9fa1238462..e363240a3455 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -271,8 +271,8 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 		ctx->ctx_id = ctx_id;
 
 		migrate = xe_migrate_alloc(tile);
-		if (IS_ERR(migrate)) {
-			err = PTR_ERR(migrate);
+		if (!migrate) {
+			err = -ENOMEM;
 			goto err_ret;
 		}
 

From 1fdc4c381ff765479d76ccf3134717c430c871b8 Mon Sep 17 00:00:00 2001
From: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Date: Fri, 1 Aug 2025 10:53:56 +0530
Subject: [PATCH 0183/1455] drm/xe/devcoredump: Defer devcoredump
 initialization during probe

Doing devcoredump initializing before GT though look harmless, it leads
to problem during driver unbind. Because of this order, GT/Engine
release functions will be called before xe devcoredump release function
(xe_driver_devcoredump_fini) leading to the following kernel crash[1]
because the devcoredump functions might still use GT/Engine
datastructures after those are freed.

The following crash is observed while running the IGT
xe_wedged@wedged-at-any-timeout. The test forces a wedged state by
submitting a workload which hangs. Then does a unbind/rebind of the
driver to recover from the wedged state.
The hanged workload leads to a devcoredump. The following crash is
noticed when the devcoredump capture races with the driver unbind.
During driver unbind, the release function hw_engine_fini() will be
called which assigns NULL to hwe->gt. But the same data structure is
accessed during the coredump capture in the function
xe_engine_snapshot_print by reading snapshot->hwe->gt.

With this patch, we make sure the devcoredump is stopped before
deinitializing the core driver functions.

[1]:
BUG: kernel NULL pointer dereference, address: 0000000000000000
Workqueue: events_unbound xe_devcoredump_deferred_snap_work [xe]
RIP: 0010:xe_engine_snapshot_print+0x47/0x420 [xe]
Call Trace:
 <TASK>
 ? drm_printf+0x64/0x90
 __xe_devcoredump_read+0x23f/0x2d0 [xe]
 ? __pfx___drm_printfn_coredump+0x10/0x10
 ? __pfx___drm_puts_coredump+0x10/0x10
 xe_devcoredump_deferred_snap_work+0x17a/0x190 [xe]
 process_one_work+0x22e/0x6f0
 worker_thread+0x1e8/0x3d0
 ? __pfx_worker_thread+0x10/0x10
 kthread+0x11f/0x250
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x47/0x70
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1a/0x30

v2: Detailed commit description (Rodrigo)
v3: FIXME added (Rodrigo, Stuart)

Fixes: 4209d635a823 ("drm/xe: Remove devcoredump during driver release")
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250731061300.14320-1-balasubramani.vivekanandan@intel.com
Signed-off-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://lore.kernel.org/r/20250801052356.21885-1-balasubramani.vivekanandan@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c      | 8 ++++----
 drivers/gpu/drm/xe/xe_guc_capture.c | 6 ++++++
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index c754374f1a8d..57edbc63da6f 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -822,10 +822,6 @@ int xe_device_probe(struct xe_device *xe)
 			return err;
 	}
 
-	err = xe_devcoredump_init(xe);
-	if (err)
-		return err;
-
 	/*
 	 * From here on, if a step fails, make sure a Driver-FLR is triggereed
 	 */
@@ -890,6 +886,10 @@ int xe_device_probe(struct xe_device *xe)
 	    XE_WA(xe->tiles->media_gt, 15015404425_disable))
 		XE_DEVICE_WA_DISABLE(xe, 15015404425);
 
+	err = xe_devcoredump_init(xe);
+	if (err)
+		return err;
+
 	xe_nvm_init(xe);
 
 	err = xe_heci_gsc_init(xe);
diff --git a/drivers/gpu/drm/xe/xe_guc_capture.c b/drivers/gpu/drm/xe/xe_guc_capture.c
index 859a3ba91be5..243dad3e2418 100644
--- a/drivers/gpu/drm/xe/xe_guc_capture.c
+++ b/drivers/gpu/drm/xe/xe_guc_capture.c
@@ -1817,6 +1817,12 @@ void xe_engine_snapshot_print(struct xe_hw_engine_snapshot *snapshot, struct drm
 		   str_yes_no(snapshot->kernel_reserved));
 
 	for (type = GUC_STATE_CAPTURE_TYPE_GLOBAL; type < GUC_STATE_CAPTURE_TYPE_MAX; type++) {
+		/*
+		 * FIXME: During devcoredump print we should avoid accessing the
+		 * driver pointers for gt or engine. Printing should be done only
+		 * using the snapshot captured. Here we are accessing the gt
+		 * pointer. It should be fixed.
+		 */
 		list = xe_guc_capture_get_reg_desc_list(gt, GUC_CAPTURE_LIST_INDEX_PF, type,
 							capture_class, false);
 		snapshot_print_by_list_order(snapshot, p, type, list);

From b52f8d7a8f38629e3a9d1d372f2fe43e3a27d6b2 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 1 Aug 2025 16:44:18 +0200
Subject: [PATCH 0184/1455] drm/xe/pf: Skip LMTT update if no LMEM was
 provisioned
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

During VF unprovisioning, if VF was not provisioned with LMEM,
there is no need to trigger LMTT update, as VF LMTT was never
set. This will spare us sending full TLB invalidation requests.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250801144418.180584-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index f00cce81f1ff..c8f0320d032f 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -1434,7 +1434,8 @@ fail:
 	return err;
 }
 
-static void pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_config *config)
+/* Return: %true if there was an LMEM provisioned, %false otherwise */
+static bool pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_config *config)
 {
 	xe_gt_assert(gt, IS_DGFX(gt_to_xe(gt)));
 	xe_gt_assert(gt, xe_gt_is_main_type(gt));
@@ -1443,7 +1444,9 @@ static void pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_confi
 	if (config->lmem_obj) {
 		xe_bo_unpin_map_no_vm(config->lmem_obj);
 		config->lmem_obj = NULL;
+		return true;
 	}
+	return false;
 }
 
 static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size)
@@ -2021,12 +2024,13 @@ static void pf_release_vf_config(struct xe_gt *gt, unsigned int vfid)
 {
 	struct xe_gt_sriov_config *config = pf_pick_vf_config(gt, vfid);
 	struct xe_device *xe = gt_to_xe(gt);
+	bool released;
 
 	if (xe_gt_is_main_type(gt)) {
 		pf_release_vf_config_ggtt(gt, config);
 		if (IS_DGFX(xe)) {
-			pf_release_vf_config_lmem(gt, config);
-			if (xe_device_has_lmtt(xe))
+			released = pf_release_vf_config_lmem(gt, config);
+			if (released && xe_device_has_lmtt(xe))
 				pf_update_vf_lmtt(xe, vfid);
 		}
 	}

From a424353937c24554bb242a6582ed8f018b4a411c Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 1 Aug 2025 16:28:20 +0200
Subject: [PATCH 0185/1455] drm/xe/pf: Disable PF restart worker on device
 removal
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We can't let restart worker run once device is removed, since other
data that it might want to access could be already released.
Explicitly disable worker as part of device cleanup action.

Fixes: a4d1c5d0b99b ("drm/xe/pf: Move VFs reprovisioning to worker")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250801142822.180530-2-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_sriov_pf.c | 32 ++++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
index 35489fa81825..2ea81d81c0ae 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
@@ -47,9 +47,16 @@ static int pf_alloc_metadata(struct xe_gt *gt)
 
 static void pf_init_workers(struct xe_gt *gt)
 {
+	xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt)));
 	INIT_WORK(&gt->sriov.pf.workers.restart, pf_worker_restart_func);
 }
 
+static void pf_fini_workers(struct xe_gt *gt)
+{
+	xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt)));
+	disable_work_sync(&gt->sriov.pf.workers.restart);
+}
+
 /**
  * xe_gt_sriov_pf_init_early - Prepare SR-IOV PF data structures on PF.
  * @gt: the &xe_gt to initialize
@@ -79,6 +86,21 @@ int xe_gt_sriov_pf_init_early(struct xe_gt *gt)
 	return 0;
 }
 
+static void pf_fini_action(void *arg)
+{
+	struct xe_gt *gt = arg;
+
+	pf_fini_workers(gt);
+}
+
+static int pf_init_late(struct xe_gt *gt)
+{
+	struct xe_device *xe = gt_to_xe(gt);
+
+	xe_gt_assert(gt, IS_SRIOV_PF(xe));
+	return devm_add_action_or_reset(xe->drm.dev, pf_fini_action, gt);
+}
+
 /**
  * xe_gt_sriov_pf_init - Prepare SR-IOV PF data structures on PF.
  * @gt: the &xe_gt to initialize
@@ -95,7 +117,15 @@ int xe_gt_sriov_pf_init(struct xe_gt *gt)
 	if (err)
 		return err;
 
-	return xe_gt_sriov_pf_migration_init(gt);
+	err = xe_gt_sriov_pf_migration_init(gt);
+	if (err)
+		return err;
+
+	err = pf_init_late(gt);
+	if (err)
+		return err;
+
+	return 0;
 }
 
 static bool pf_needs_enable_ggtt_guest_update(struct xe_device *xe)

From c6c86441c465ea440dfb5039f1c26e629a6fd64c Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 1 Aug 2025 16:28:21 +0200
Subject: [PATCH 0186/1455] drm/xe/pf: Make sure PF is ready to configure VFs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The PF driver might be resumed just to configure VFs, but since
it is doing some asynchronous GuC reconfigurations after fresh
reset, we should wait until all pending works are completed.

This is especially important in case of LMEM provisioning, since
we also need to update the LMTT and send invalidation requests
to all GuCs, which are expected to be already in the VGT mode.

Fixes: 68ae022278a1 ("drm/xe/pf: Force GuC virtualization mode")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250801142822.180530-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_sriov_pf.c         | 25 +++++++++++++++++++
 drivers/gpu/drm/xe/xe_gt_sriov_pf.h         |  1 +
 drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c |  4 ++-
 drivers/gpu/drm/xe/xe_pci_sriov.c           |  7 +++++-
 drivers/gpu/drm/xe/xe_sriov_pf.c            | 27 +++++++++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_pf.h            |  1 +
 6 files changed, 63 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
index 2ea81d81c0ae..bdbd15f3afe3 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
@@ -16,6 +16,7 @@
 #include "xe_gt_sriov_pf_migration.h"
 #include "xe_gt_sriov_pf_service.h"
 #include "xe_gt_sriov_printk.h"
+#include "xe_guc_submit.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
 
@@ -260,3 +261,27 @@ void xe_gt_sriov_pf_restart(struct xe_gt *gt)
 {
 	pf_queue_restart(gt);
 }
+
+static void pf_flush_restart(struct xe_gt *gt)
+{
+	xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt)));
+	flush_work(&gt->sriov.pf.workers.restart);
+}
+
+/**
+ * xe_gt_sriov_pf_wait_ready() - Wait until per-GT PF SR-IOV support is ready.
+ * @gt: the &xe_gt
+ *
+ * This function can only be called on PF.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int xe_gt_sriov_pf_wait_ready(struct xe_gt *gt)
+{
+	/* don't wait if there is another ongoing reset */
+	if (xe_guc_read_stopped(&gt->uc.guc))
+		return -EBUSY;
+
+	pf_flush_restart(gt);
+	return 0;
+}
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.h b/drivers/gpu/drm/xe/xe_gt_sriov_pf.h
index e2b2ff8132dc..e7fde3f9937a 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.h
@@ -11,6 +11,7 @@ struct xe_gt;
 #ifdef CONFIG_PCI_IOV
 int xe_gt_sriov_pf_init_early(struct xe_gt *gt);
 int xe_gt_sriov_pf_init(struct xe_gt *gt);
+int xe_gt_sriov_pf_wait_ready(struct xe_gt *gt);
 void xe_gt_sriov_pf_init_hw(struct xe_gt *gt);
 void xe_gt_sriov_pf_sanitize_hw(struct xe_gt *gt, unsigned int vfid);
 void xe_gt_sriov_pf_stop_prepare(struct xe_gt *gt);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c
index bf679b21f485..3ed245e04d0c 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_debugfs.c
@@ -22,6 +22,7 @@
 #include "xe_gt_sriov_pf_policy.h"
 #include "xe_gt_sriov_pf_service.h"
 #include "xe_pm.h"
+#include "xe_sriov_pf.h"
 
 /*
  *      /sys/kernel/debug/dri/0/
@@ -205,7 +206,8 @@ static int CONFIG##_set(void *data, u64 val)					\
 		return -EOVERFLOW;						\
 										\
 	xe_pm_runtime_get(xe);							\
-	err = xe_gt_sriov_pf_config_set_##CONFIG(gt, vfid, val);		\
+	err = xe_sriov_pf_wait_ready(xe) ?:					\
+	      xe_gt_sriov_pf_config_set_##CONFIG(gt, vfid, val);		\
 	xe_pm_runtime_put(xe);							\
 										\
 	return err;								\
diff --git a/drivers/gpu/drm/xe/xe_pci_sriov.c b/drivers/gpu/drm/xe/xe_pci_sriov.c
index 8813efdcafbb..447a7867eecb 100644
--- a/drivers/gpu/drm/xe/xe_pci_sriov.c
+++ b/drivers/gpu/drm/xe/xe_pci_sriov.c
@@ -12,6 +12,7 @@
 #include "xe_pci_sriov.h"
 #include "xe_pm.h"
 #include "xe_sriov.h"
+#include "xe_sriov_pf.h"
 #include "xe_sriov_pf_helpers.h"
 #include "xe_sriov_printk.h"
 
@@ -138,6 +139,10 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
 	xe_assert(xe, num_vfs <= total_vfs);
 	xe_sriov_dbg(xe, "enabling %u VF%s\n", num_vfs, str_plural(num_vfs));
 
+	err = xe_sriov_pf_wait_ready(xe);
+	if (err)
+		goto out;
+
 	/*
 	 * We must hold additional reference to the runtime PM to keep PF in D0
 	 * during VFs lifetime, as our VFs do not implement the PM capability.
@@ -169,7 +174,7 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
 failed:
 	pf_unprovision_vfs(xe, num_vfs);
 	xe_pm_runtime_put(xe);
-
+out:
 	xe_sriov_notice(xe, "Failed to enable %u VF%s (%pe)\n",
 			num_vfs, str_plural(num_vfs), ERR_PTR(err));
 	return err;
diff --git a/drivers/gpu/drm/xe/xe_sriov_pf.c b/drivers/gpu/drm/xe/xe_sriov_pf.c
index afbdd894bd6e..27ddf3cc80e9 100644
--- a/drivers/gpu/drm/xe/xe_sriov_pf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_pf.c
@@ -9,6 +9,7 @@
 
 #include "xe_assert.h"
 #include "xe_device.h"
+#include "xe_gt_sriov_pf.h"
 #include "xe_module.h"
 #include "xe_sriov.h"
 #include "xe_sriov_pf.h"
@@ -102,6 +103,32 @@ int xe_sriov_pf_init_early(struct xe_device *xe)
 	return 0;
 }
 
+/**
+ * xe_sriov_pf_wait_ready() - Wait until PF is ready to operate.
+ * @xe: the &xe_device to test
+ *
+ * This function can only be called on PF.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int xe_sriov_pf_wait_ready(struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	unsigned int id;
+	int err;
+
+	if (xe_device_wedged(xe))
+		return -ECANCELED;
+
+	for_each_gt(gt, xe, id) {
+		err = xe_gt_sriov_pf_wait_ready(gt);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 /**
  * xe_sriov_pf_print_vfs_summary - Print SR-IOV PF information.
  * @xe: the &xe_device to print info from
diff --git a/drivers/gpu/drm/xe/xe_sriov_pf.h b/drivers/gpu/drm/xe/xe_sriov_pf.h
index c392c3fcf085..e3b34f8f5e04 100644
--- a/drivers/gpu/drm/xe/xe_sriov_pf.h
+++ b/drivers/gpu/drm/xe/xe_sriov_pf.h
@@ -15,6 +15,7 @@ struct xe_device;
 #ifdef CONFIG_PCI_IOV
 bool xe_sriov_pf_readiness(struct xe_device *xe);
 int xe_sriov_pf_init_early(struct xe_device *xe);
+int xe_sriov_pf_wait_ready(struct xe_device *xe);
 void xe_sriov_pf_debugfs_register(struct xe_device *xe, struct dentry *root);
 void xe_sriov_pf_print_vfs_summary(struct xe_device *xe, struct drm_printer *p);
 #else

From 9fd9f221440024b7451678898facfb34af054310 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 1 Aug 2025 16:28:22 +0200
Subject: [PATCH 0187/1455] drm/xe/pf: Don't resume device from restart worker
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The PF's restart worker shouldn't attempt to resume the device on
its own, since its goal is to finish PF and VFs reprovisioning on
the recently reset GuC. Take extra RPM reference while scheduling
a work and release it from the worker or when we cancel a work.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250801142822.180530-4-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_sriov_pf.c | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
index bdbd15f3afe3..c4dda87b47cc 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf.c
@@ -55,7 +55,12 @@ static void pf_init_workers(struct xe_gt *gt)
 static void pf_fini_workers(struct xe_gt *gt)
 {
 	xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt)));
-	disable_work_sync(&gt->sriov.pf.workers.restart);
+
+	if (disable_work_sync(&gt->sriov.pf.workers.restart)) {
+		xe_gt_sriov_dbg_verbose(gt, "pending restart disabled!\n");
+		/* release an rpm reference taken on the worker's behalf */
+		xe_pm_runtime_put(gt_to_xe(gt));
+	}
 }
 
 /**
@@ -207,8 +212,11 @@ static void pf_cancel_restart(struct xe_gt *gt)
 {
 	xe_gt_assert(gt, IS_SRIOV_PF(gt_to_xe(gt)));
 
-	if (cancel_work_sync(&gt->sriov.pf.workers.restart))
+	if (cancel_work_sync(&gt->sriov.pf.workers.restart)) {
 		xe_gt_sriov_dbg_verbose(gt, "pending restart canceled!\n");
+		/* release an rpm reference taken on the worker's behalf */
+		xe_pm_runtime_put(gt_to_xe(gt));
+	}
 }
 
 /**
@@ -226,9 +234,12 @@ static void pf_restart(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 
-	xe_pm_runtime_get(xe);
+	xe_gt_assert(gt, !xe_pm_runtime_suspended(xe));
+
 	xe_gt_sriov_pf_config_restart(gt);
 	xe_gt_sriov_pf_control_restart(gt);
+
+	/* release an rpm reference taken on our behalf */
 	xe_pm_runtime_put(xe);
 
 	xe_gt_sriov_dbg(gt, "restart completed\n");
@@ -247,8 +258,13 @@ static void pf_queue_restart(struct xe_gt *gt)
 
 	xe_gt_assert(gt, IS_SRIOV_PF(xe));
 
-	if (!queue_work(xe->sriov.wq, &gt->sriov.pf.workers.restart))
+	/* take an rpm reference on behalf of the worker */
+	xe_pm_runtime_get_noresume(xe);
+
+	if (!queue_work(xe->sriov.wq, &gt->sriov.pf.workers.restart)) {
 		xe_gt_sriov_dbg(gt, "restart already in queue!\n");
+		xe_pm_runtime_put(xe);
+	}
 }
 
 /**

From f8b99c61a8ac1dc94520fea7d5660ae09a047c9f Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Fri, 1 Aug 2025 17:46:58 +0100
Subject: [PATCH 0188/1455] drm/i915/bw: Remove space before newline

There is an extraneous space before a newline in a drm_dbg_kms message.
Remove the space.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250801164658.2438212-1-colin.i.king@gmail.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bw.c b/drivers/gpu/drm/i915/display/intel_bw.c
index d29a755612de..ac6da20d9529 100644
--- a/drivers/gpu/drm/i915/display/intel_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_bw.c
@@ -359,7 +359,7 @@ static int icl_get_qgv_points(struct intel_display *display,
 
 		for (i = 0; i < qi->num_psf_points; i++)
 			drm_dbg_kms(display->drm,
-				    "PSF GV %d: CLK=%d \n",
+				    "PSF GV %d: CLK=%d\n",
 				    i, qi->psf_points[i].clk);
 	}
 

From 218b15a3e9755c9cae79941b4bec22a4e8ad8758 Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Sat, 2 Aug 2025 13:25:46 -0600
Subject: [PATCH 0189/1455] accel/rocket: Fix Rockchip NPU compilation

Replace DRM_GPU_SCHED_STAT_NOMINAL with GPU_DRM_SCHED_STAT_RESET, in
accordance with commit 0a5dc1b67ef5 ("drm/sched: Rename
DRM_GPU_SCHED_STAT_NOMINAL to DRM_GPU_SCHED_STAT_RESET")

Pass extra parameter to drm_sched_job_init, as required by commit
2956554823ce ("drm/sched: Store the drm client_id in drm_sched_fence")

Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250802-fix-rockchip-npu-build-v1-1-fb0f0dacb3fe@brighamcampbell.com
---
 drivers/accel/rocket/rocket_job.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/rocket/rocket_job.c b/drivers/accel/rocket/rocket_job.c
index e731da15ebff..5d4afd692306 100644
--- a/drivers/accel/rocket/rocket_job.c
+++ b/drivers/accel/rocket/rocket_job.c
@@ -390,7 +390,7 @@ static enum drm_gpu_sched_stat rocket_job_timedout(struct drm_sched_job *sched_j
 	atomic_set(&core->reset.pending, 1);
 	rocket_reset(core, sched_job);
 
-	return DRM_GPU_SCHED_STAT_NOMINAL;
+	return DRM_GPU_SCHED_STAT_RESET;
 }
 
 static void rocket_reset_work(struct work_struct *work)
@@ -552,7 +552,7 @@ static int rocket_ioctl_submit_job(struct drm_device *dev, struct drm_file *file
 
 	ret = drm_sched_job_init(&rjob->base,
 				 &file_priv->sched_entity,
-				 1, NULL);
+				 1, NULL, file->client_id);
 	if (ret)
 		goto out_put_job;
 

From d9c1b06f3553977c50017e7ab40f8cecf80a30c9 Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Sat, 2 Aug 2025 13:25:47 -0600
Subject: [PATCH 0190/1455] accel/rocket: Fix undeclared const rocket_pm_ops

Fix sparse warning regarding an undeclared const rocket_pm_ops, which is
used in rocket_drv.c.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202508030021.uwdr4P08-lkp@intel.com/
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250802-fix-rockchip-npu-build-v1-2-fb0f0dacb3fe@brighamcampbell.com
---
 drivers/accel/rocket/rocket_drv.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/accel/rocket/rocket_drv.h b/drivers/accel/rocket/rocket_drv.h
index f50634935b60..2c673bb99ccc 100644
--- a/drivers/accel/rocket/rocket_drv.h
+++ b/drivers/accel/rocket/rocket_drv.h
@@ -9,6 +9,8 @@
 
 #include "rocket_device.h"
 
+extern const struct dev_pm_ops rocket_pm_ops;
+
 struct rocket_iommu_domain {
 	struct iommu_domain *domain;
 	struct kref kref;

From 904eef7dd0aecd71f7c3b264439275cb9cdab6af Mon Sep 17 00:00:00 2001
From: Arun R Murthy <arun.r.murthy@intel.com>
Date: Wed, 30 Jul 2025 16:03:20 +0530
Subject: [PATCH 0191/1455] drm/i915/display: Use the recomended min_hblank
 values

Use recommended values as per wa_14021694213 to compare with the
calculated value and choose minimum of them.

v2: corrected checkpatch warning and retain the restriction for
min_hblank (Jani)
v3: use calculated value to compare with recomended value and choose
minimum of them (Imre)
v4: As driver supported min bpc is 8, omit the condition check for
bpc6 with ycbcr420. Added a note for the same (Imre)
v5: Add a warn for the unexpected case of 6bpc + uhbr + ycbcr420
v6: Reworded the comments and check fo anything < compressed bpp 8(Imre)
v7: Fix checkpatch warning. (Ankit)

Bspec: 74379
Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250730-min_hblank-v7-1-179360220ced@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 54d88f24b689..625036c47bdf 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -3182,7 +3182,26 @@ int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state,
 	 */
 	min_hblank = min_hblank - 2;
 
-	min_hblank = min(10, min_hblank);
+	/*
+	 * min_hblank formula is undergoing a change, to avoid underrun use the
+	 * recomended value in spec to compare with the calculated one and use the
+	 * minimum value
+	 */
+	if (intel_dp_is_uhbr(crtc_state)) {
+		/*
+		 * Note: Bspec requires a min_hblank of 2 for YCBCR420
+		 * with compressed bpp 6, but the minimum compressed bpp
+		 * supported by the driver is 8.
+		 */
+		drm_WARN_ON(display->drm,
+			    (crtc_state->dsc.compression_enable &&
+			     crtc_state->output_format == INTEL_OUTPUT_FORMAT_YCBCR420 &&
+			     crtc_state->dsc.compressed_bpp_x16 < fxp_q4_from_int(8)));
+		min_hblank = min(3, min_hblank);
+	} else {
+		min_hblank = min(10, min_hblank);
+	}
+
 	crtc_state->min_hblank = min_hblank;
 
 	return 0;

From 0cc207494fdce38977c3fc82b6e826ca4d6d6c9f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 15:36:15 +0300
Subject: [PATCH 0192/1455] drm/xe/compat: remove unused platform macros

After refactors, a lot of platform macros have become unused. Remove
them before new users have a chance to pop up.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/4507b49ead12c997de4615fa6ec277e666e5226a.1753965351.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../gpu/drm/xe/compat-i915-headers/i915_drv.h | 21 -------------------
 1 file changed, 21 deletions(-)

diff --git a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
index 9b7572e06f34..d8cd66a96621 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
@@ -26,34 +26,13 @@ static inline struct drm_i915_private *to_i915(const struct drm_device *dev)
 #define IS_I915G(dev_priv)	(dev_priv && 0)
 #define IS_I915GM(dev_priv)	(dev_priv && 0)
 #define IS_PINEVIEW(dev_priv)	(dev_priv && 0)
-#define IS_IVYBRIDGE(dev_priv)	(dev_priv && 0)
 #define IS_VALLEYVIEW(dev_priv)	(dev_priv && 0)
 #define IS_CHERRYVIEW(dev_priv)	(dev_priv && 0)
 #define IS_HASWELL(dev_priv)	(dev_priv && 0)
 #define IS_BROADWELL(dev_priv)	(dev_priv && 0)
-#define IS_SKYLAKE(dev_priv)	(dev_priv && 0)
 #define IS_BROXTON(dev_priv)	(dev_priv && 0)
-#define IS_KABYLAKE(dev_priv)	(dev_priv && 0)
 #define IS_GEMINILAKE(dev_priv)	(dev_priv && 0)
-#define IS_COFFEELAKE(dev_priv)	(dev_priv && 0)
-#define IS_COMETLAKE(dev_priv)	(dev_priv && 0)
-#define IS_ICELAKE(dev_priv)	(dev_priv && 0)
-#define IS_JASPERLAKE(dev_priv)	(dev_priv && 0)
-#define IS_ELKHARTLAKE(dev_priv)	(dev_priv && 0)
-#define IS_TIGERLAKE(dev_priv)	IS_PLATFORM(dev_priv, XE_TIGERLAKE)
-#define IS_ROCKETLAKE(dev_priv)	IS_PLATFORM(dev_priv, XE_ROCKETLAKE)
-#define IS_DG1(dev_priv)        IS_PLATFORM(dev_priv, XE_DG1)
-#define IS_ALDERLAKE_S(dev_priv) IS_PLATFORM(dev_priv, XE_ALDERLAKE_S)
-#define IS_ALDERLAKE_P(dev_priv) (IS_PLATFORM(dev_priv, XE_ALDERLAKE_P) || \
-				  IS_PLATFORM(dev_priv, XE_ALDERLAKE_N))
 #define IS_DG2(dev_priv)	IS_PLATFORM(dev_priv, XE_DG2)
-#define IS_METEORLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_METEORLAKE)
-#define IS_LUNARLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_LUNARLAKE)
-#define IS_BATTLEMAGE(dev_priv)  IS_PLATFORM(dev_priv, XE_BATTLEMAGE)
-#define IS_PANTHERLAKE(dev_priv) IS_PLATFORM(dev_priv, XE_PANTHERLAKE)
-
-#define IS_HASWELL_ULT(dev_priv) (dev_priv && 0)
-#define IS_BROADWELL_ULT(dev_priv) (dev_priv && 0)
 
 #define IS_MOBILE(xe) (xe && 0)
 

From 4274a2b75696115f5a5d026a93a3a56f0f99c928 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 15:36:16 +0300
Subject: [PATCH 0193/1455] drm/xe/compat: stop including i915_utils.h from
 compat i915_drv.h

Expose the places that need i915_utils.h, and include it where needed.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/6338c8524e600e048b56c5484624cfb51ed49d1d.1753965351.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.c               | 1 +
 drivers/gpu/drm/i915/display/intel_cdclk.c              | 1 +
 drivers/gpu/drm/i915/display/intel_display_driver.c     | 1 +
 drivers/gpu/drm/i915/display/intel_display_power.c      | 1 +
 drivers/gpu/drm/i915/display/intel_display_power_well.c | 1 +
 drivers/gpu/drm/i915/display/intel_fb.c                 | 1 +
 drivers/gpu/drm/i915/display/intel_gmbus.c              | 1 +
 drivers/gpu/drm/i915/display/intel_hotplug.c            | 1 +
 drivers/gpu/drm/i915/display/intel_vblank.c             | 1 +
 drivers/gpu/drm/i915/display/skl_universal_plane.c      | 1 +
 drivers/gpu/drm/i915/soc/intel_dram.c                   | 1 +
 drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h       | 1 -
 drivers/gpu/drm/xe/display/ext/i915_utils.c             | 1 +
 13 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 9c268bed091d..276ac29e4daf 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -36,6 +36,7 @@
 #include "soc/intel_rom.h"
 
 #include "i915_drv.h"
+#include "i915_utils.h"
 #include "intel_display.h"
 #include "intel_display_core.h"
 #include "intel_display_rpm.h"
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 2a2d6714e4f4..59997dc95984 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -31,6 +31,7 @@
 #include "hsw_ips.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_atomic.h"
 #include "intel_audio.h"
 #include "intel_bw.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_driver.c b/drivers/gpu/drm/i915/display/intel_display_driver.c
index 8586ba102605..cf1c14412abe 100644
--- a/drivers/gpu/drm/i915/display/intel_display_driver.c
+++ b/drivers/gpu/drm/i915/display/intel_display_driver.c
@@ -18,6 +18,7 @@
 #include <drm/drm_vblank.h>
 
 #include "i915_drv.h"
+#include "i915_utils.h"
 #include "i9xx_wm.h"
 #include "intel_acpi.h"
 #include "intel_atomic.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 2c2b23a138cc..d1d3b6f89e2a 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -10,6 +10,7 @@
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_backlight_regs.h"
 #include "intel_cdclk.h"
 #include "intel_clock_gating.h"
diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 48cac225a809..6efe5524cbbf 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -6,6 +6,7 @@
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_backlight_regs.h"
 #include "intel_combo_phy.h"
 #include "intel_combo_phy_regs.h"
diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 79811f998e38..71b4fa19f1e5 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -11,6 +11,7 @@
 #include <drm/drm_modeset_helper.h>
 
 #include "i915_drv.h"
+#include "i915_utils.h"
 #include "intel_bo.h"
 #include "intel_display.h"
 #include "intel_display_core.h"
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index 637f0f23f163..6a74805570e1 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -36,6 +36,7 @@
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_regs.h"
 #include "intel_display_types.h"
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index 265aa97fcc75..740e5c930521 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -28,6 +28,7 @@
 
 #include "i915_drv.h"
 #include "i915_irq.h"
+#include "i915_utils.h"
 #include "intel_connector.h"
 #include "intel_display_power.h"
 #include "intel_display_core.h"
diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c
index 3e51deca0c21..46d6db5fed11 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.c
+++ b/drivers/gpu/drm/i915/display/intel_vblank.c
@@ -6,6 +6,7 @@
 #include <drm/drm_vblank.h>
 
 #include "i915_drv.h"
+#include "i915_utils.h"
 #include "intel_color.h"
 #include "intel_crtc.h"
 #include "intel_de.h"
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index ec3fe75f2c6c..950dc79dbdd4 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -10,6 +10,7 @@
 
 #include "pxp/intel_pxp.h"
 #include "i915_drv.h"
+#include "i915_utils.h"
 #include "intel_bo.h"
 #include "intel_de.h"
 #include "intel_display_irq.h"
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index deb159548a09..3eeaabdf59e8 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -11,6 +11,7 @@
 
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_utils.h"
 #include "intel_dram.h"
 #include "intel_mchbar_regs.h"
 #include "intel_pcode.h"
diff --git a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
index d8cd66a96621..b8269391bc69 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/i915_drv.h
@@ -12,7 +12,6 @@
 
 #include <drm/drm_drv.h>
 
-#include "i915_utils.h"
 #include "xe_device.h" /* for xe_device_has_flat_ccs() */
 #include "xe_device_types.h"
 
diff --git a/drivers/gpu/drm/xe/display/ext/i915_utils.c b/drivers/gpu/drm/xe/display/ext/i915_utils.c
index 43b10a2cc508..1421c2a7b64d 100644
--- a/drivers/gpu/drm/xe/display/ext/i915_utils.c
+++ b/drivers/gpu/drm/xe/display/ext/i915_utils.c
@@ -4,6 +4,7 @@
  */
 
 #include "i915_drv.h"
+#include "i915_utils.h"
 
 bool i915_vtd_active(struct drm_i915_private *i915)
 {

From 7b4106517fe651fdb4dd3a0c341a68e9c63d85dd Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 14:12:14 +0300
Subject: [PATCH 0194/1455] drm/xe: fix stale comment about unordered_wq usage

Display has switched to its own workqueue, no longer using
xe->unordered_wq.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250731111214.1130130-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/xe/xe_device_types.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 003afb279a5e..4f795157f711 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -453,7 +453,7 @@ struct xe_device {
 	/** @ordered_wq: used to serialize compute mode resume */
 	struct workqueue_struct *ordered_wq;
 
-	/** @unordered_wq: used to serialize unordered work, mostly display */
+	/** @unordered_wq: used to serialize unordered work */
 	struct workqueue_struct *unordered_wq;
 
 	/** @destroy_wq: used to serialize user destroy work, like queue */

From 0521a868222ffe636bf202b6e9d29292c1e19c62 Mon Sep 17 00:00:00 2001
From: Simon Richter <Simon.Richter@hogyros.de>
Date: Sat, 2 Aug 2025 11:40:36 +0900
Subject: [PATCH 0195/1455] Mark xe driver as BROKEN if kernel page size is not
 4kB
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This driver, for the time being, assumes that the kernel page size is 4kB,
so it fails on loong64 and aarch64 with 16kB pages, and ppc64el with 64kB
pages.

Signed-off-by: Simon Richter <Simon.Richter@hogyros.de>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Cc: stable@vger.kernel.org # v6.8+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250802024152.3021-1-Simon.Richter@hogyros.de
---
 drivers/gpu/drm/xe/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/Kconfig b/drivers/gpu/drm/xe/Kconfig
index 2bb2bc052120..714d5702dfd7 100644
--- a/drivers/gpu/drm/xe/Kconfig
+++ b/drivers/gpu/drm/xe/Kconfig
@@ -5,6 +5,7 @@ config DRM_XE
 	depends on KUNIT || !KUNIT
 	depends on INTEL_VSEC || !INTEL_VSEC
 	depends on X86_PLATFORM_DEVICES || !(X86 && ACPI)
+	depends on PAGE_SIZE_4KB || COMPILE_TEST || BROKEN
 	select INTERVAL_TREE
 	# we need shmfs for the swappable backing store, and in particular
 	# the shmem_readpage() which depends upon tmpfs

From 6e2c8fbc66a002c90bf9aa217ec2ffb52b027154 Mon Sep 17 00:00:00 2001
From: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Date: Thu, 24 Jul 2025 14:09:27 +0530
Subject: [PATCH 0196/1455] drm/{i915,xe}/display: Block hpd during suspend

It has been observed that during `xe_display_pm_suspend()` execution,
an HPD interrupt can still be triggered, resulting in `dig_port_work`
being scheduled. The issue arises when this work executes after
`xe_display_pm_suspend_late()`, by which time the display is fully
suspended.

This can lead to errors such as "DC state mismatch", as the dig_port
work accesses display resources that are no longer available or
powered.

To address this, introduce  'intel_encoder_block_all_hpds' and
'intel_encoder_unblock_all_hpds' functions, which iterate over all
encoders and block/unblock HPD respectively.

These are used to:
- Block HPD IRQs before calling 'intel_hpd_cancel_work' in suspend
  and shutdown
- Unblock HPD IRQs after 'intel_hpd_init' in resume

This will prevent 'dig_port_work' being scheduled during display
suspend.

Continuation of previous patch discussion:
https://patchwork.freedesktop.org/patch/663964/

Changes in v2:
 - Add 'intel_encoder_block_all_hpds' to 'xe_display_pm_shutdown'.(Imre
   Deak)
 - Add 'intel_hpd_cancel_work' to 'xe_display_fini_early' to cancel
   any HPD pending work at late driver removal. (Imre Deak)

Changes in v3:
 - Move 'intel_encoder_block_all_hpds' after intel_dp_mst_suspend
   in 'xe_display_pm_shutdown'.(Imre Deak)

Signed-off-by: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Acked-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250724083928.2298199-1-dibin.moolakadan.subrahmanian@intel.com
---
 drivers/gpu/drm/i915/display/intel_encoder.c | 23 ++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_encoder.h |  3 +++
 drivers/gpu/drm/i915/display/intel_hotplug.c |  2 --
 drivers/gpu/drm/xe/display/xe_display.c      |  6 +++++
 4 files changed, 32 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_encoder.c b/drivers/gpu/drm/i915/display/intel_encoder.c
index 0b7bd26f4339..4e2b77b87678 100644
--- a/drivers/gpu/drm/i915/display/intel_encoder.c
+++ b/drivers/gpu/drm/i915/display/intel_encoder.c
@@ -8,6 +8,7 @@
 #include "intel_display_core.h"
 #include "intel_display_types.h"
 #include "intel_encoder.h"
+#include "intel_hotplug.h"
 
 static void intel_encoder_link_check_work_fn(struct work_struct *work)
 {
@@ -37,6 +38,28 @@ void intel_encoder_link_check_queue_work(struct intel_encoder *encoder, int dela
 			 &encoder->link_check_work, msecs_to_jiffies(delay_ms));
 }
 
+void intel_encoder_unblock_all_hpds(struct intel_display *display)
+{
+	struct intel_encoder *encoder;
+
+	if (!HAS_DISPLAY(display))
+		return;
+
+	for_each_intel_encoder(display->drm, encoder)
+		intel_hpd_unblock(encoder);
+}
+
+void intel_encoder_block_all_hpds(struct intel_display *display)
+{
+	struct intel_encoder *encoder;
+
+	if (!HAS_DISPLAY(display))
+		return;
+
+	for_each_intel_encoder(display->drm, encoder)
+		intel_hpd_block(encoder);
+}
+
 void intel_encoder_suspend_all(struct intel_display *display)
 {
 	struct intel_encoder *encoder;
diff --git a/drivers/gpu/drm/i915/display/intel_encoder.h b/drivers/gpu/drm/i915/display/intel_encoder.h
index 3fa5589f0b1c..e1d3aeab7c00 100644
--- a/drivers/gpu/drm/i915/display/intel_encoder.h
+++ b/drivers/gpu/drm/i915/display/intel_encoder.h
@@ -17,4 +17,7 @@ void intel_encoder_link_check_flush_work(struct intel_encoder *encoder);
 void intel_encoder_suspend_all(struct intel_display *display);
 void intel_encoder_shutdown_all(struct intel_display *display);
 
+void intel_encoder_block_all_hpds(struct intel_display *display);
+void intel_encoder_unblock_all_hpds(struct intel_display *display);
+
 #endif /* __INTEL_ENCODER_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index 740e5c930521..bd8851fd2ee9 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -972,8 +972,6 @@ void intel_hpd_cancel_work(struct intel_display *display)
 
 	spin_lock_irq(&display->irq.lock);
 
-	drm_WARN_ON(display->drm, get_blocked_hpd_pin_mask(display));
-
 	display->hotplug.long_hpd_pin_mask = 0;
 	display->hotplug.short_hpd_pin_mask = 0;
 	display->hotplug.event_bits = 0;
diff --git a/drivers/gpu/drm/xe/display/xe_display.c b/drivers/gpu/drm/xe/display/xe_display.c
index e2e0771cf274..8b68d70db6c8 100644
--- a/drivers/gpu/drm/xe/display/xe_display.c
+++ b/drivers/gpu/drm/xe/display/xe_display.c
@@ -96,6 +96,7 @@ static void xe_display_fini_early(void *arg)
 	if (!xe->info.probe_display)
 		return;
 
+	intel_hpd_cancel_work(display);
 	intel_display_driver_remove_nogem(display);
 	intel_display_driver_remove_noirq(display);
 	intel_opregion_cleanup(display);
@@ -340,6 +341,8 @@ void xe_display_pm_suspend(struct xe_device *xe)
 
 	xe_display_flush_cleanup_work(xe);
 
+	intel_encoder_block_all_hpds(display);
+
 	intel_hpd_cancel_work(display);
 
 	if (has_display(xe)) {
@@ -370,6 +373,7 @@ void xe_display_pm_shutdown(struct xe_device *xe)
 
 	xe_display_flush_cleanup_work(xe);
 	intel_dp_mst_suspend(display);
+	intel_encoder_block_all_hpds(display);
 	intel_hpd_cancel_work(display);
 
 	if (has_display(xe))
@@ -471,6 +475,8 @@ void xe_display_pm_resume(struct xe_device *xe)
 
 	intel_hpd_init(display);
 
+	intel_encoder_unblock_all_hpds(display);
+
 	if (has_display(xe)) {
 		intel_display_driver_resume(display);
 		drm_kms_helper_poll_enable(&xe->drm);

From d47cc89d810d7ec19fbb34331551a3439da99c01 Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:38 +0200
Subject: [PATCH 0197/1455] drm/xe/sa: Avoid caching GGTT address within the
 manager
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Non-virtualized resources require fixups after SRIOV VF migration.
Caching GGTT references rather than re-computing them from the
underlying Buffer Object is something we want to avoid, as such
code would require additional fixup step and additional locking
around all the places where the address is accessed.

This change removes the cached GPU address from the Sub-Allocation
Manager, and introduces a function which recomputes and returns
the address instead.

v2: renamed xe_sa_manager_gpu_addr(), added kerneldoc

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-2-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_debugfs.c   |  5 +++--
 drivers/gpu/drm/xe/xe_guc_buf.c      |  2 +-
 drivers/gpu/drm/xe/xe_sa.c           |  1 -
 drivers/gpu/drm/xe/xe_sa.h           | 15 ++++++++++++++-
 drivers/gpu/drm/xe/xe_sa_types.h     |  1 -
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c |  2 +-
 6 files changed, 19 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index 8d4b66ca38b7..bf3a67b5951c 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -29,6 +29,7 @@
 #include "xe_pm.h"
 #include "xe_reg_sr.h"
 #include "xe_reg_whitelist.h"
+#include "xe_sa.h"
 #include "xe_sriov.h"
 #include "xe_tuning.h"
 #include "xe_uc_debugfs.h"
@@ -128,7 +129,7 @@ static int sa_info(struct xe_gt *gt, struct drm_printer *p)
 
 	xe_pm_runtime_get(gt_to_xe(gt));
 	drm_suballoc_dump_debug_info(&tile->mem.kernel_bb_pool->base, p,
-				     tile->mem.kernel_bb_pool->gpu_addr);
+				     xe_sa_manager_gpu_addr(tile->mem.kernel_bb_pool));
 	xe_pm_runtime_put(gt_to_xe(gt));
 
 	return 0;
@@ -152,7 +153,7 @@ static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p)
 
 		drm_printf(p, "ccs %s bb suballoc info\n", ctx_id ? "write" : "read");
 		drm_printf(p, "-------------------------\n");
-		drm_suballoc_dump_debug_info(&bb_pool->base, p, bb_pool->gpu_addr);
+		drm_suballoc_dump_debug_info(&bb_pool->base, p, xe_sa_manager_gpu_addr(bb_pool));
 		drm_puts(p, "\n");
 	}
 
diff --git a/drivers/gpu/drm/xe/xe_guc_buf.c b/drivers/gpu/drm/xe/xe_guc_buf.c
index 14a07dca48e7..502ca3a4ee60 100644
--- a/drivers/gpu/drm/xe/xe_guc_buf.c
+++ b/drivers/gpu/drm/xe/xe_guc_buf.c
@@ -164,7 +164,7 @@ u64 xe_guc_cache_gpu_addr_from_ptr(struct xe_guc_buf_cache *cache, const void *p
 	if (offset < 0 || offset + size > cache->sam->base.size)
 		return 0;
 
-	return cache->sam->gpu_addr + offset;
+	return xe_sa_manager_gpu_addr(cache->sam) + offset;
 }
 
 #if IS_BUILTIN(CONFIG_DRM_XE_KUNIT_TEST)
diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c
index 1d43e183ca21..fedd017d6dd3 100644
--- a/drivers/gpu/drm/xe/xe_sa.c
+++ b/drivers/gpu/drm/xe/xe_sa.c
@@ -69,7 +69,6 @@ struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u3
 	}
 	sa_manager->bo = bo;
 	sa_manager->is_iomem = bo->vmap.is_iomem;
-	sa_manager->gpu_addr = xe_bo_ggtt_addr(bo);
 
 	if (bo->vmap.is_iomem) {
 		sa_manager->cpu_ptr = kvzalloc(managed_size, GFP_KERNEL);
diff --git a/drivers/gpu/drm/xe/xe_sa.h b/drivers/gpu/drm/xe/xe_sa.h
index 1170ee5a81a8..99dbf0eea540 100644
--- a/drivers/gpu/drm/xe/xe_sa.h
+++ b/drivers/gpu/drm/xe/xe_sa.h
@@ -7,6 +7,8 @@
 
 #include <linux/sizes.h>
 #include <linux/types.h>
+
+#include "xe_bo.h"
 #include "xe_sa_types.h"
 
 struct dma_fence;
@@ -43,9 +45,20 @@ to_xe_sa_manager(struct drm_suballoc_manager *mng)
 	return container_of(mng, struct xe_sa_manager, base);
 }
 
+/**
+ * xe_sa_manager_gpu_addr - Retrieve GPU address of a back storage BO
+ * within suballocator.
+ * @sa_manager: the &xe_sa_manager struct instance
+ * Return: GGTT address of the back storage BO.
+ */
+static inline u64 xe_sa_manager_gpu_addr(struct xe_sa_manager *sa_manager)
+{
+	return xe_bo_ggtt_addr(sa_manager->bo);
+}
+
 static inline u64 xe_sa_bo_gpu_addr(struct drm_suballoc *sa)
 {
-	return to_xe_sa_manager(sa->manager)->gpu_addr +
+	return xe_sa_manager_gpu_addr(to_xe_sa_manager(sa->manager)) +
 		drm_suballoc_soffset(sa);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_sa_types.h b/drivers/gpu/drm/xe/xe_sa_types.h
index 2b070ff1292e..cb7238799dcb 100644
--- a/drivers/gpu/drm/xe/xe_sa_types.h
+++ b/drivers/gpu/drm/xe/xe_sa_types.h
@@ -12,7 +12,6 @@ struct xe_bo;
 struct xe_sa_manager {
 	struct drm_suballoc_manager base;
 	struct xe_bo *bo;
-	u64 gpu_addr;
 	void *cpu_ptr;
 	bool is_iomem;
 };
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index e363240a3455..f0ca2a9b2bb7 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -169,7 +169,7 @@ static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx)
 static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
 {
 	struct xe_lrc *lrc = xe_migrate_lrc(ctx->migrate);
-	u64 addr = ctx->mem.ccs_bb_pool->gpu_addr;
+	u64 addr = xe_sa_manager_gpu_addr(ctx->mem.ccs_bb_pool);
 	u32 dw[10], i = 0;
 
 	dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;

From f1193b864c08213d90e50c20ec3aaf8a413607b6 Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:39 +0200
Subject: [PATCH 0198/1455] drm/xe/vf: Pause submissions during RESFIX fixups
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

While applying post-migration fixups to VF, GuC will not respond
to any commands. This means submissions have no way of finishing.

To avoid acquiring additional resources and then stalling
on hardware access, pause the submission work. This will
decrease the chance of depleting resources, and speed up
the recovery.

v2: Commented xe_irq_resume() call
v3: Typo fix

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-3-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_gpu_scheduler.c | 13 ++++++++
 drivers/gpu/drm/xe/xe_gpu_scheduler.h |  1 +
 drivers/gpu/drm/xe/xe_guc_submit.c    | 35 ++++++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_submit.h    |  2 ++
 drivers/gpu/drm/xe/xe_sriov_vf.c      | 46 +++++++++++++++++++++++++++
 5 files changed, 97 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.c b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
index 869b43a4151d..455ccaf17314 100644
--- a/drivers/gpu/drm/xe/xe_gpu_scheduler.c
+++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.c
@@ -101,6 +101,19 @@ void xe_sched_submission_stop(struct xe_gpu_scheduler *sched)
 	cancel_work_sync(&sched->work_process_msg);
 }
 
+/**
+ * xe_sched_submission_stop_async - Stop further runs of submission tasks on a scheduler.
+ * @sched: the &xe_gpu_scheduler struct instance
+ *
+ * This call disables further runs of scheduling work queue. It does not wait
+ * for any in-progress runs to finish, only makes sure no further runs happen
+ * afterwards.
+ */
+void xe_sched_submission_stop_async(struct xe_gpu_scheduler *sched)
+{
+	drm_sched_wqueue_stop(&sched->base);
+}
+
 void xe_sched_submission_resume_tdr(struct xe_gpu_scheduler *sched)
 {
 	drm_sched_resume_timeout(&sched->base, sched->base.timeout);
diff --git a/drivers/gpu/drm/xe/xe_gpu_scheduler.h b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
index 308061f0cf37..e548b2aed95a 100644
--- a/drivers/gpu/drm/xe/xe_gpu_scheduler.h
+++ b/drivers/gpu/drm/xe/xe_gpu_scheduler.h
@@ -21,6 +21,7 @@ void xe_sched_fini(struct xe_gpu_scheduler *sched);
 
 void xe_sched_submission_start(struct xe_gpu_scheduler *sched);
 void xe_sched_submission_stop(struct xe_gpu_scheduler *sched);
+void xe_sched_submission_stop_async(struct xe_gpu_scheduler *sched);
 
 void xe_sched_submission_resume_tdr(struct xe_gpu_scheduler *sched);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 679ae229cc82..ebc137be4de2 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1836,6 +1836,19 @@ void xe_guc_submit_stop(struct xe_guc *guc)
 
 }
 
+/**
+ * xe_guc_submit_pause - Stop further runs of submission tasks on given GuC.
+ * @guc: the &xe_guc struct instance whose scheduler is to be disabled
+ */
+void xe_guc_submit_pause(struct xe_guc *guc)
+{
+	struct xe_exec_queue *q;
+	unsigned long index;
+
+	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
+		xe_sched_submission_stop_async(&q->guc->sched);
+}
+
 static void guc_exec_queue_start(struct xe_exec_queue *q)
 {
 	struct xe_gpu_scheduler *sched = &q->guc->sched;
@@ -1876,6 +1889,28 @@ int xe_guc_submit_start(struct xe_guc *guc)
 	return 0;
 }
 
+static void guc_exec_queue_unpause(struct xe_exec_queue *q)
+{
+	struct xe_gpu_scheduler *sched = &q->guc->sched;
+
+	xe_sched_submission_start(sched);
+}
+
+/**
+ * xe_guc_submit_unpause - Allow further runs of submission tasks on given GuC.
+ * @guc: the &xe_guc struct instance whose scheduler is to be enabled
+ */
+void xe_guc_submit_unpause(struct xe_guc *guc)
+{
+	struct xe_exec_queue *q;
+	unsigned long index;
+
+	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
+		guc_exec_queue_unpause(q);
+
+	wake_up_all(&guc->ct.wq);
+}
+
 static struct xe_exec_queue *
 g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id)
 {
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 8f64e799283b..ff44500f3da2 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -18,6 +18,8 @@ int xe_guc_submit_reset_prepare(struct xe_guc *guc);
 void xe_guc_submit_reset_wait(struct xe_guc *guc);
 void xe_guc_submit_stop(struct xe_guc *guc);
 int xe_guc_submit_start(struct xe_guc *guc);
+void xe_guc_submit_pause(struct xe_guc *guc);
+void xe_guc_submit_unpause(struct xe_guc *guc);
 void xe_guc_submit_wedge(struct xe_guc *guc);
 
 int xe_guc_read_stopped(struct xe_guc *guc);
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 26e243c28994..c66b17da1ce7 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -11,6 +11,8 @@
 #include "xe_gt_sriov_printk.h"
 #include "xe_gt_sriov_vf.h"
 #include "xe_guc_ct.h"
+#include "xe_guc_submit.h"
+#include "xe_irq.h"
 #include "xe_pm.h"
 #include "xe_sriov.h"
 #include "xe_sriov_printk.h"
@@ -147,6 +149,48 @@ void xe_sriov_vf_init_early(struct xe_device *xe)
 		xe_sriov_info(xe, "migration not supported by this module version\n");
 }
 
+/**
+ * vf_post_migration_shutdown - Stop the driver activities after VF migration.
+ * @xe: the &xe_device struct instance
+ *
+ * After this VM is migrated and assigned to a new VF, it is running on a new
+ * hardware, and therefore many hardware-dependent states and related structures
+ * require fixups. Without fixups, the hardware cannot do any work, and therefore
+ * all GPU pipelines are stalled.
+ * Stop some of kernel activities to make the fixup process faster.
+ */
+static void vf_post_migration_shutdown(struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	unsigned int id;
+
+	for_each_gt(gt, xe, id)
+		xe_guc_submit_pause(&gt->uc.guc);
+}
+
+/**
+ * vf_post_migration_kickstart - Re-start the driver activities under new hardware.
+ * @xe: the &xe_device struct instance
+ *
+ * After we have finished with all post-migration fixups, restart the driver
+ * activities to continue feeding the GPU with workloads.
+ */
+static void vf_post_migration_kickstart(struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	unsigned int id;
+
+	/*
+	 * Make sure interrupts on the new HW are properly set. The GuC IRQ
+	 * must be working at this point, since the recovery did started,
+	 * but the rest was not enabled using the procedure from spec.
+	 */
+	xe_irq_resume(xe);
+
+	for_each_gt(gt, xe, id)
+		xe_guc_submit_unpause(&gt->uc.guc);
+}
+
 static bool gt_vf_post_migration_needed(struct xe_gt *gt)
 {
 	return test_bit(gt->info.id, &gt_to_xe(gt)->sriov.vf.migration.gt_flags);
@@ -230,6 +274,7 @@ static void vf_post_migration_recovery(struct xe_device *xe)
 
 	drm_dbg(&xe->drm, "migration recovery in progress\n");
 	xe_pm_runtime_get(xe);
+	vf_post_migration_shutdown(xe);
 
 	if (!vf_migration_supported(xe)) {
 		xe_sriov_err(xe, "migration not supported by this module version\n");
@@ -247,6 +292,7 @@ static void vf_post_migration_recovery(struct xe_device *xe)
 		set_bit(id, &fixed_gts);
 	}
 
+	vf_post_migration_kickstart(xe);
 	err = vf_post_migration_notify_resfix_done(xe, fixed_gts);
 	if (err)
 		goto fail;

From a0840b1ce95e37a1a870a3e794ed25c69080cecf Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:40 +0200
Subject: [PATCH 0199/1455] drm/xe: Block reset while recovering from VF
 migration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Resetting GuC during recovery could interfere with the recovery
process. Such reset might be also triggered without justification,
due to migration taking time, rather than due to the workload not
progressing.

Doing GuC reset during the recovery would cause exit of RESFIX state,
and therefore continuation of GuC work while fixups are still being
applied. To avoid that, reset needs to be blocked during the recovery.

This patch blocks the reset during recovery. Reset request in that
time range will be stalled, and unblocked only after GuC goes out
of RESFIX state.

In case a reset procedure already started while the recovery is
triggered, there isn't much we can do - we cannot wait for it to
finish as it involves waiting for hardware, and we can't be sure
at which exact point of the reset procedure the GPU got switched.
Therefore, the rare cases where migration happens while reset is
in progress, are still dangerous. Resets are not a part of the
standard flow, and cause unfinished workloads - that will happen
during the reset interrupted by migration as well, so it doesn't
diverge that much from what normally happens during such resets.

v2: Introduce a new atomic for reset blocking, as we cannot reuse
  `stopped` atomic (that could lead to losing a workload).
v3: Switched atomic functs to ones which include proper barriers

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michal Winiarski <michal.winiarski@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-4-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_gt.c         | 10 ++++++++
 drivers/gpu/drm/xe/xe_guc_submit.c | 37 ++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_submit.h |  3 +++
 drivers/gpu/drm/xe/xe_guc_types.h  |  6 +++++
 drivers/gpu/drm/xe/xe_sriov_vf.c   | 12 ++++++++--
 5 files changed, 66 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 5a79c6e3208b..390394bbaadc 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -41,6 +41,7 @@
 #include "xe_gt_topology.h"
 #include "xe_guc_exec_queue_types.h"
 #include "xe_guc_pc.h"
+#include "xe_guc_submit.h"
 #include "xe_hw_fence.h"
 #include "xe_hw_engine_class_sysfs.h"
 #include "xe_irq.h"
@@ -802,6 +803,11 @@ static int do_gt_restart(struct xe_gt *gt)
 	return 0;
 }
 
+static int gt_wait_reset_unblock(struct xe_gt *gt)
+{
+	return xe_guc_wait_reset_unblock(&gt->uc.guc);
+}
+
 static int gt_reset(struct xe_gt *gt)
 {
 	unsigned int fw_ref;
@@ -816,6 +822,10 @@ static int gt_reset(struct xe_gt *gt)
 
 	xe_gt_info(gt, "reset started\n");
 
+	err = gt_wait_reset_unblock(gt);
+	if (!err)
+		xe_gt_warn(gt, "reset block failed to get lifted");
+
 	xe_pm_runtime_get(gt_to_xe(gt));
 
 	if (xe_fault_inject_gt_reset()) {
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index ebc137be4de2..3274048e6a38 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1783,6 +1783,43 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q)
 	}
 }
 
+/**
+ * xe_guc_submit_reset_block - Disallow reset calls on given GuC.
+ * @guc: the &xe_guc struct instance
+ */
+int xe_guc_submit_reset_block(struct xe_guc *guc)
+{
+	return atomic_fetch_or(1, &guc->submission_state.reset_blocked);
+}
+
+/**
+ * xe_guc_submit_reset_unblock - Allow back reset calls on given GuC.
+ * @guc: the &xe_guc struct instance
+ */
+void xe_guc_submit_reset_unblock(struct xe_guc *guc)
+{
+	atomic_set_release(&guc->submission_state.reset_blocked, 0);
+	wake_up_all(&guc->ct.wq);
+}
+
+static int guc_submit_reset_is_blocked(struct xe_guc *guc)
+{
+	return atomic_read_acquire(&guc->submission_state.reset_blocked);
+}
+
+/* Maximum time of blocking reset */
+#define RESET_BLOCK_PERIOD_MAX (HZ * 5)
+
+/**
+ * xe_guc_wait_reset_unblock - Wait until reset blocking flag is lifted, or timeout.
+ * @guc: the &xe_guc struct instance
+ */
+int xe_guc_wait_reset_unblock(struct xe_guc *guc)
+{
+	return wait_event_timeout(guc->ct.wq,
+				  !guc_submit_reset_is_blocked(guc), RESET_BLOCK_PERIOD_MAX);
+}
+
 int xe_guc_submit_reset_prepare(struct xe_guc *guc)
 {
 	int ret;
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index ff44500f3da2..4dd278f5f61b 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -20,6 +20,9 @@ void xe_guc_submit_stop(struct xe_guc *guc);
 int xe_guc_submit_start(struct xe_guc *guc);
 void xe_guc_submit_pause(struct xe_guc *guc);
 void xe_guc_submit_unpause(struct xe_guc *guc);
+int xe_guc_submit_reset_block(struct xe_guc *guc);
+void xe_guc_submit_reset_unblock(struct xe_guc *guc);
+int xe_guc_wait_reset_unblock(struct xe_guc *guc);
 void xe_guc_submit_wedge(struct xe_guc *guc);
 
 int xe_guc_read_stopped(struct xe_guc *guc);
diff --git a/drivers/gpu/drm/xe/xe_guc_types.h b/drivers/gpu/drm/xe/xe_guc_types.h
index 1fde7614fcc5..c7b9642b41ba 100644
--- a/drivers/gpu/drm/xe/xe_guc_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_types.h
@@ -85,6 +85,12 @@ struct xe_guc {
 		struct xarray exec_queue_lookup;
 		/** @submission_state.stopped: submissions are stopped */
 		atomic_t stopped;
+		/**
+		 * @submission_state.reset_blocked: reset attempts are blocked;
+		 * blocking reset in order to delay it may be required if running
+		 * an operation which is sensitive to resets.
+		 */
+		atomic_t reset_blocked;
 		/** @submission_state.lock: protects submission state */
 		struct mutex lock;
 		/** @submission_state.enabled: submission is enabled */
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index c66b17da1ce7..71d28b30de43 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -163,9 +163,15 @@ static void vf_post_migration_shutdown(struct xe_device *xe)
 {
 	struct xe_gt *gt;
 	unsigned int id;
+	int ret = 0;
 
-	for_each_gt(gt, xe, id)
+	for_each_gt(gt, xe, id) {
 		xe_guc_submit_pause(&gt->uc.guc);
+		ret |= xe_guc_submit_reset_block(&gt->uc.guc);
+	}
+
+	if (ret)
+		drm_info(&xe->drm, "migration recovery encountered ongoing reset\n");
 }
 
 /**
@@ -187,8 +193,10 @@ static void vf_post_migration_kickstart(struct xe_device *xe)
 	 */
 	xe_irq_resume(xe);
 
-	for_each_gt(gt, xe, id)
+	for_each_gt(gt, xe, id) {
+		xe_guc_submit_reset_unblock(&gt->uc.guc);
 		xe_guc_submit_unpause(&gt->uc.guc);
+	}
 }
 
 static bool gt_vf_post_migration_needed(struct xe_gt *gt)

From b46ef766737c4ecc11aad9742fd7f70f8f922246 Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:41 +0200
Subject: [PATCH 0200/1455] drm/xe/vf: Rebase HWSP of all contexts after
 migration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

All contexts require an update due to GGTT range shift, as that
affects their HWSP.

The HW status page of a context contains GGTT references, which
need to be shifted to a new range (or re-computed using the
previously updated vma nodes). The references include ring start
address and indirect state address.

v2: move some functions to better matched files
v3: Add missing kerneldocs
v4: Style fix

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Acked-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Michal Winiarski <michal.winiarski@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-5-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_exec_queue.c | 13 +++++++++++++
 drivers/gpu/drm/xe/xe_exec_queue.h |  2 ++
 drivers/gpu/drm/xe/xe_guc_submit.c | 16 ++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_submit.h |  2 ++
 drivers/gpu/drm/xe/xe_lrc.c        | 17 +++++++++++++++++
 drivers/gpu/drm/xe/xe_lrc.h        |  1 +
 drivers/gpu/drm/xe/xe_sriov_vf.c   |  1 +
 7 files changed, 52 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index d5958518ef00..ae36a35ec80b 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -1076,3 +1076,16 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q, struct xe_vm *vm)
 
 	return err;
 }
+
+/**
+ * xe_exec_queue_contexts_hwsp_rebase - Re-compute GGTT references
+ * within all LRCs of a queue.
+ * @q: the &xe_exec_queue struct instance containing target LRCs
+ */
+void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q)
+{
+	int i;
+
+	for (i = 0; i < q->width; ++i)
+		xe_lrc_update_hwctx_regs_with_address(q->lrc[i]);
+}
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index 17bc50a7f05a..1d399a33c5c0 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -90,4 +90,6 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q,
 				      struct xe_vm *vm);
 void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q);
 
+void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q);
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 3274048e6a38..f756b3855873 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2505,3 +2505,19 @@ void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p)
 		guc_exec_queue_print(q, p);
 	mutex_unlock(&guc->submission_state.lock);
 }
+
+/**
+ * xe_guc_contexts_hwsp_rebase - Re-compute GGTT references within all
+ * exec queues registered to given GuC.
+ * @guc: the &xe_guc struct instance
+ */
+void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc)
+{
+	struct xe_exec_queue *q;
+	unsigned long index;
+
+	mutex_lock(&guc->submission_state.lock);
+	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
+		xe_exec_queue_contexts_hwsp_rebase(q);
+	mutex_unlock(&guc->submission_state.lock);
+}
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 4dd278f5f61b..d22f2d36dca1 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -46,4 +46,6 @@ xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapsh
 void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p);
 void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type);
 
+void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc);
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 540f044bf425..baf926ad3a66 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -1439,6 +1439,23 @@ void xe_lrc_destroy(struct kref *ref)
 	kfree(lrc);
 }
 
+/**
+ * xe_lrc_update_hwctx_regs_with_address - Re-compute GGTT references within given LRC.
+ * @lrc: the &xe_lrc struct instance
+ */
+void xe_lrc_update_hwctx_regs_with_address(struct xe_lrc *lrc)
+{
+	if (xe_lrc_has_indirect_ring_state(lrc)) {
+		xe_lrc_write_ctx_reg(lrc, CTX_INDIRECT_RING_STATE,
+				     __xe_lrc_indirect_ring_ggtt_addr(lrc));
+
+		xe_lrc_write_indirect_ctx_reg(lrc, INDIRECT_CTX_RING_START,
+					      __xe_lrc_ring_ggtt_addr(lrc));
+	} else {
+		xe_lrc_write_ctx_reg(lrc, CTX_RING_START, __xe_lrc_ring_ggtt_addr(lrc));
+	}
+}
+
 void xe_lrc_set_ring_tail(struct xe_lrc *lrc, u32 tail)
 {
 	if (xe_lrc_has_indirect_ring_state(lrc))
diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h
index b6c8053c581b..cc231bec709c 100644
--- a/drivers/gpu/drm/xe/xe_lrc.h
+++ b/drivers/gpu/drm/xe/xe_lrc.h
@@ -88,6 +88,7 @@ bool xe_lrc_ring_is_idle(struct xe_lrc *lrc);
 u32 xe_lrc_indirect_ring_ggtt_addr(struct xe_lrc *lrc);
 u32 xe_lrc_ggtt_addr(struct xe_lrc *lrc);
 u32 *xe_lrc_regs(struct xe_lrc *lrc);
+void xe_lrc_update_hwctx_regs_with_address(struct xe_lrc *lrc);
 
 u32 xe_lrc_read_ctx_reg(struct xe_lrc *lrc, int reg_nr);
 void xe_lrc_write_ctx_reg(struct xe_lrc *lrc, int reg_nr, u32 val);
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 71d28b30de43..54a6218e48b0 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -269,6 +269,7 @@ static int gt_vf_post_migration_fixups(struct xe_gt *gt)
 	shift = xe_gt_sriov_vf_ggtt_shift(gt);
 	if (shift) {
 		xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
+		xe_guc_contexts_hwsp_rebase(&gt->uc.guc);
 		/* FIXME: add the recovery steps */
 		xe_guc_ct_fixup_messages_with_ggtt(&gt->uc.guc.ct, shift);
 	}

From 30d137ddcee514d3ee5a498713628bd50d41e87b Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:42 +0200
Subject: [PATCH 0201/1455] drm/xe/vf: Rebase MEMIRQ structures for all
 contexts after migration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

All contexts require an update of state data, as the data includes
GGTT references to memirq-related buffers.

Default contexts need these references updated as well, because they
are not refreshed when a new context is created from them.

The way we write to vram requires scratch buffer to be used
before the whole block is memcopied. Since using kalloc() within
specific recovery functions would lead to unintended relations
between locks, we are allocating the buffer earlier, before
any locks are taken. The same buffer will be used for other steps
of the recovery.

v2: Update addresses by xe_lrc_write_ctx_reg() rather than
  set_memory_based_intr()
v3: Renamed parameter, reordered parameters in some functs
v4: Check if have MEMIRQ, move `xe_gt*` funct to proper file
v5: Revert back to requiring scratch buffer, but allocate it
  earlier this time

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Acked-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Michal Winiarski <michal.winiarski@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-6-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_exec_queue.c  |  7 ++--
 drivers/gpu/drm/xe/xe_exec_queue.h  |  2 +-
 drivers/gpu/drm/xe/xe_gt_sriov_vf.c | 14 ++++++++
 drivers/gpu/drm/xe/xe_gt_sriov_vf.h |  1 +
 drivers/gpu/drm/xe/xe_guc_submit.c  |  5 +--
 drivers/gpu/drm/xe/xe_guc_submit.h  |  2 +-
 drivers/gpu/drm/xe/xe_lrc.c         | 51 +++++++++++++++++++++++++++--
 drivers/gpu/drm/xe/xe_lrc.h         |  4 +++
 drivers/gpu/drm/xe/xe_sriov_vf.c    | 20 +++++++++--
 9 files changed, 96 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index ae36a35ec80b..0beb6388acb0 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -1081,11 +1081,14 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q, struct xe_vm *vm)
  * xe_exec_queue_contexts_hwsp_rebase - Re-compute GGTT references
  * within all LRCs of a queue.
  * @q: the &xe_exec_queue struct instance containing target LRCs
+ * @scratch: scratch buffer to be used as temporary storage
  */
-void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q)
+void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch)
 {
 	int i;
 
-	for (i = 0; i < q->width; ++i)
+	for (i = 0; i < q->width; ++i) {
+		xe_lrc_update_memirq_regs_with_address(q->lrc[i], q->hwe, scratch);
 		xe_lrc_update_hwctx_regs_with_address(q->lrc[i]);
+	}
 }
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index 1d399a33c5c0..da720197929b 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -90,6 +90,6 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q,
 				      struct xe_vm *vm);
 void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q);
 
-void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q);
+void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
index b282838d59e6..0461d5513487 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.c
@@ -25,6 +25,7 @@
 #include "xe_guc.h"
 #include "xe_guc_hxg_helpers.h"
 #include "xe_guc_relay.h"
+#include "xe_lrc.h"
 #include "xe_mmio.h"
 #include "xe_sriov.h"
 #include "xe_sriov_vf.h"
@@ -750,6 +751,19 @@ failed:
 	return err;
 }
 
+/**
+ * xe_gt_sriov_vf_default_lrcs_hwsp_rebase - Update GGTT references in HWSP of default LRCs.
+ * @gt: the &xe_gt struct instance
+ */
+void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt)
+{
+	struct xe_hw_engine *hwe;
+	enum xe_hw_engine_id id;
+
+	for_each_hw_engine(hwe, gt, id)
+		xe_default_lrc_update_memirq_regs_with_address(hwe);
+}
+
 /**
  * xe_gt_sriov_vf_migrated_event_handler - Start a VF migration recovery,
  *   or just mark that a GuC is ready for it.
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
index e0357f341a2d..0af1dc769fe0 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_vf.h
@@ -21,6 +21,7 @@ void xe_gt_sriov_vf_guc_versions(struct xe_gt *gt,
 int xe_gt_sriov_vf_query_config(struct xe_gt *gt);
 int xe_gt_sriov_vf_connect(struct xe_gt *gt);
 int xe_gt_sriov_vf_query_runtime(struct xe_gt *gt);
+void xe_gt_sriov_vf_default_lrcs_hwsp_rebase(struct xe_gt *gt);
 int xe_gt_sriov_vf_notify_resfix_done(struct xe_gt *gt);
 void xe_gt_sriov_vf_migrated_event_handler(struct xe_gt *gt);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index f756b3855873..300dac6dab31 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2510,14 +2510,15 @@ void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p)
  * xe_guc_contexts_hwsp_rebase - Re-compute GGTT references within all
  * exec queues registered to given GuC.
  * @guc: the &xe_guc struct instance
+ * @scratch: scratch buffer to be used as temporary storage
  */
-void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc)
+void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch)
 {
 	struct xe_exec_queue *q;
 	unsigned long index;
 
 	mutex_lock(&guc->submission_state.lock);
 	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
-		xe_exec_queue_contexts_hwsp_rebase(q);
+		xe_exec_queue_contexts_hwsp_rebase(q, scratch);
 	mutex_unlock(&guc->submission_state.lock);
 }
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index d22f2d36dca1..9a2718c81d43 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -46,6 +46,6 @@ xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapsh
 void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p);
 void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type);
 
-void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc);
+void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index baf926ad3a66..9a67f4ae235a 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -697,7 +697,13 @@ u32 xe_lrc_regs_offset(struct xe_lrc *lrc)
 	return xe_lrc_pphwsp_offset(lrc) + LRC_PPHWSP_SIZE;
 }
 
-static size_t lrc_reg_size(struct xe_device *xe)
+/**
+ * xe_lrc_reg_size() - Get size of the LRC registers area within queues
+ * @xe: the &xe_device struct instance
+ *
+ * Returns: Size of the LRC registers area for current platform
+ */
+size_t xe_lrc_reg_size(struct xe_device *xe)
 {
 	if (GRAPHICS_VERx100(xe) >= 1250)
 		return 96 * sizeof(u32);
@@ -707,7 +713,7 @@ static size_t lrc_reg_size(struct xe_device *xe)
 
 size_t xe_lrc_skip_size(struct xe_device *xe)
 {
-	return LRC_PPHWSP_SIZE + lrc_reg_size(xe);
+	return LRC_PPHWSP_SIZE + xe_lrc_reg_size(xe);
 }
 
 static inline u32 __xe_lrc_seqno_offset(struct xe_lrc *lrc)
@@ -948,6 +954,47 @@ static void *empty_lrc_data(struct xe_hw_engine *hwe)
 	return data;
 }
 
+/**
+ * xe_default_lrc_update_memirq_regs_with_address - Re-compute GGTT references in default LRC
+ * of given engine.
+ * @hwe: the &xe_hw_engine struct instance
+ */
+void xe_default_lrc_update_memirq_regs_with_address(struct xe_hw_engine *hwe)
+{
+	struct xe_gt *gt = hwe->gt;
+	u32 *regs;
+
+	if (!gt->default_lrc[hwe->class])
+		return;
+
+	regs = gt->default_lrc[hwe->class] + LRC_PPHWSP_SIZE;
+	set_memory_based_intr(regs, hwe);
+}
+
+/**
+ * xe_lrc_update_memirq_regs_with_address - Re-compute GGTT references in mem interrupt data
+ * for given LRC.
+ * @lrc: the &xe_lrc struct instance
+ * @hwe: the &xe_hw_engine struct instance
+ * @regs: scratch buffer to be used as temporary storage
+ */
+void xe_lrc_update_memirq_regs_with_address(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
+					    u32 *regs)
+{
+	struct xe_gt *gt = hwe->gt;
+	struct iosys_map map;
+	size_t regs_len;
+
+	if (!xe_device_uses_memirq(gt_to_xe(gt)))
+		return;
+
+	map = __xe_lrc_regs_map(lrc);
+	regs_len = xe_lrc_reg_size(gt_to_xe(gt));
+	xe_map_memcpy_from(gt_to_xe(gt), regs, &map, 0, regs_len);
+	set_memory_based_intr(regs, hwe);
+	xe_map_memcpy_to(gt_to_xe(gt), &map, 0, regs, regs_len);
+}
+
 static void xe_lrc_set_ppgtt(struct xe_lrc *lrc, struct xe_vm *vm)
 {
 	u64 desc = xe_vm_pdp4_descriptor(vm, gt_to_tile(lrc->gt));
diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h
index cc231bec709c..eceeeee6c021 100644
--- a/drivers/gpu/drm/xe/xe_lrc.h
+++ b/drivers/gpu/drm/xe/xe_lrc.h
@@ -89,6 +89,9 @@ u32 xe_lrc_indirect_ring_ggtt_addr(struct xe_lrc *lrc);
 u32 xe_lrc_ggtt_addr(struct xe_lrc *lrc);
 u32 *xe_lrc_regs(struct xe_lrc *lrc);
 void xe_lrc_update_hwctx_regs_with_address(struct xe_lrc *lrc);
+void xe_default_lrc_update_memirq_regs_with_address(struct xe_hw_engine *hwe);
+void xe_lrc_update_memirq_regs_with_address(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
+					    u32 *regs);
 
 u32 xe_lrc_read_ctx_reg(struct xe_lrc *lrc, int reg_nr);
 void xe_lrc_write_ctx_reg(struct xe_lrc *lrc, int reg_nr, u32 val);
@@ -107,6 +110,7 @@ s32 xe_lrc_start_seqno(struct xe_lrc *lrc);
 u32 xe_lrc_parallel_ggtt_addr(struct xe_lrc *lrc);
 struct iosys_map xe_lrc_parallel_map(struct xe_lrc *lrc);
 
+size_t xe_lrc_reg_size(struct xe_device *xe);
 size_t xe_lrc_skip_size(struct xe_device *xe);
 
 void xe_lrc_dump_default(struct drm_printer *p,
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 54a6218e48b0..43ac73e432d4 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -13,6 +13,7 @@
 #include "xe_guc_ct.h"
 #include "xe_guc_submit.h"
 #include "xe_irq.h"
+#include "xe_lrc.h"
 #include "xe_pm.h"
 #include "xe_sriov.h"
 #include "xe_sriov_printk.h"
@@ -244,6 +245,11 @@ static int vf_get_next_migrated_gt_id(struct xe_device *xe)
 	return -1;
 }
 
+static size_t post_migration_scratch_size(struct xe_device *xe)
+{
+	return xe_lrc_reg_size(xe);
+}
+
 /**
  * Perform post-migration fixups on a single GT.
  *
@@ -260,19 +266,29 @@ static int vf_get_next_migrated_gt_id(struct xe_device *xe)
 static int gt_vf_post_migration_fixups(struct xe_gt *gt)
 {
 	s64 shift;
+	void *buf;
 	int err;
 
+	buf = kmalloc(post_migration_scratch_size(gt_to_xe(gt)), GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
 	err = xe_gt_sriov_vf_query_config(gt);
-	if (err)
+	if (err) {
+		kfree(buf);
 		return err;
+	}
 
 	shift = xe_gt_sriov_vf_ggtt_shift(gt);
 	if (shift) {
 		xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
-		xe_guc_contexts_hwsp_rebase(&gt->uc.guc);
+		xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
+		xe_guc_contexts_hwsp_rebase(&gt->uc.guc, buf);
 		/* FIXME: add the recovery steps */
 		xe_guc_ct_fixup_messages_with_ggtt(&gt->uc.guc.ct, shift);
 	}
+
+	kfree(buf);
 	return 0;
 }
 

From a0dda25d24e636df5c30a9370464b7cebc709faf Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:43 +0200
Subject: [PATCH 0202/1455] drm/xe/vf: Post migration, repopulate ring area for
 pending request
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The commands within ring area allocated for a request may contain
references to GGTT. These references require update after VF
migration, in order to continue any preempted LRCs, or jobs which
were emitted to the ring but not sent to GuC yet.

This change calls the emit function again for all such jobs,
as part of post-migration recovery.

v2: Moved few functions to better files
v3: Take job_list_lock
v4: Rephrased comments

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michal Winiarski <michal.winiarski@intel.com>
Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-7-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_exec_queue.c | 24 ++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_exec_queue.h |  2 ++
 drivers/gpu/drm/xe/xe_guc_submit.c | 24 ++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_submit.h |  2 ++
 drivers/gpu/drm/xe/xe_sriov_vf.c   |  2 +-
 5 files changed, 53 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 0beb6388acb0..05a1bd6b55bf 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -1092,3 +1092,27 @@ void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch)
 		xe_lrc_update_hwctx_regs_with_address(q->lrc[i]);
 	}
 }
+
+/**
+ * xe_exec_queue_jobs_ring_restore - Re-emit ring commands of requests pending on given queue.
+ * @q: the &xe_exec_queue struct instance
+ */
+void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q)
+{
+	struct xe_gpu_scheduler *sched = &q->guc->sched;
+	struct xe_sched_job *job;
+
+	/*
+	 * This routine is used within VF migration recovery. This means
+	 * using the lock here introduces a restriction: we cannot wait
+	 * for any GFX HW response while the lock is taken.
+	 */
+	spin_lock(&sched->base.job_list_lock);
+	list_for_each_entry(job, &sched->base.pending_list, drm.list) {
+		if (xe_sched_job_is_error(job))
+			continue;
+
+		q->ring_ops->emit_job(job);
+	}
+	spin_unlock(&sched->base.job_list_lock);
+}
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index da720197929b..0ffc0cb03aa6 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -92,4 +92,6 @@ void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q);
 
 void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch);
 
+void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q);
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 300dac6dab31..7a54890ce9d0 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -781,6 +781,30 @@ guc_exec_queue_run_job(struct drm_sched_job *drm_job)
 	return fence;
 }
 
+/**
+ * xe_guc_jobs_ring_rebase - Re-emit ring commands of requests pending
+ * on all queues under a guc.
+ * @guc: the &xe_guc struct instance
+ */
+void xe_guc_jobs_ring_rebase(struct xe_guc *guc)
+{
+	struct xe_exec_queue *q;
+	unsigned long index;
+
+	/*
+	 * This routine is used within VF migration recovery. This means
+	 * using the lock here introduces a restriction: we cannot wait
+	 * for any GFX HW response while the lock is taken.
+	 */
+	mutex_lock(&guc->submission_state.lock);
+	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) {
+		if (exec_queue_killed_or_banned_or_wedged(q))
+			continue;
+		xe_exec_queue_jobs_ring_restore(q);
+	}
+	mutex_unlock(&guc->submission_state.lock);
+}
+
 static void guc_exec_queue_free_job(struct drm_sched_job *drm_job)
 {
 	struct xe_sched_job *job = to_xe_sched_job(drm_job);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 9a2718c81d43..92a6f0ade615 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -34,6 +34,8 @@ int xe_guc_exec_queue_memory_cat_error_handler(struct xe_guc *guc, u32 *msg,
 int xe_guc_exec_queue_reset_failure_handler(struct xe_guc *guc, u32 *msg, u32 len);
 int xe_guc_error_capture_handler(struct xe_guc *guc, u32 *msg, u32 len);
 
+void xe_guc_jobs_ring_rebase(struct xe_guc *guc);
+
 struct xe_guc_submit_exec_queue_snapshot *
 xe_guc_exec_queue_snapshot_capture(struct xe_exec_queue *q);
 void
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 43ac73e432d4..a219395c15de 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -284,7 +284,7 @@ static int gt_vf_post_migration_fixups(struct xe_gt *gt)
 		xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
 		xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
 		xe_guc_contexts_hwsp_rebase(&gt->uc.guc, buf);
-		/* FIXME: add the recovery steps */
+		xe_guc_jobs_ring_rebase(&gt->uc.guc);
 		xe_guc_ct_fixup_messages_with_ggtt(&gt->uc.guc.ct, shift);
 	}
 

From 168b5867318bff761c9095d12115c302c483d66a Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:44 +0200
Subject: [PATCH 0203/1455] drm/xe/vf: Refresh utilization buffer during
 migration recovery
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The WA buffer we use to capture context utilization contains GGTT
references. This means its instructions have to be either fixed or
re-emitted during VF post-migration recovery.

This patch adds re-emitting content of the utilization WA BB during
the recovery.

The way we write to vram requires scratch buffer to be used before
the whole block is memcopied. We are re-using a scratch buffer
introduced in earlier part of the recovery. This is not a performance
optimization, but a necessity to avoid creating dependencies between
locks.

v2: Notable rebase after "Prepare WA BB setup for more users" patch
v3: Added error propagation

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michal Winiarski <michal.winiarski@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-8-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_exec_queue.c | 10 +++++++-
 drivers/gpu/drm/xe/xe_exec_queue.h |  2 +-
 drivers/gpu/drm/xe/xe_guc_submit.c | 14 ++++++++---
 drivers/gpu/drm/xe/xe_guc_submit.h |  2 +-
 drivers/gpu/drm/xe/xe_lrc.c        | 39 ++++++++++++++++++++++++------
 drivers/gpu/drm/xe/xe_lrc.h        |  4 +++
 drivers/gpu/drm/xe/xe_sriov_vf.c   | 15 ++++++------
 7 files changed, 66 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 05a1bd6b55bf..6c176183ed58 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -1082,15 +1082,23 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q, struct xe_vm *vm)
  * within all LRCs of a queue.
  * @q: the &xe_exec_queue struct instance containing target LRCs
  * @scratch: scratch buffer to be used as temporary storage
+ *
+ * Returns: zero on success, negative error code on failure
  */
-void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch)
+int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch)
 {
 	int i;
+	int err = 0;
 
 	for (i = 0; i < q->width; ++i) {
 		xe_lrc_update_memirq_regs_with_address(q->lrc[i], q->hwe, scratch);
 		xe_lrc_update_hwctx_regs_with_address(q->lrc[i]);
+		err = xe_lrc_setup_wa_bb_with_scratch(q->lrc[i], q->hwe, scratch);
+		if (err)
+			break;
 	}
+
+	return err;
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index 0ffc0cb03aa6..4d416f23001c 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -90,7 +90,7 @@ int xe_exec_queue_last_fence_test_dep(struct xe_exec_queue *q,
 				      struct xe_vm *vm);
 void xe_exec_queue_update_run_ticks(struct xe_exec_queue *q);
 
-void xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch);
+int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch);
 
 void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 7a54890ce9d0..20f5b9820264 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2535,14 +2535,22 @@ void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p)
  * exec queues registered to given GuC.
  * @guc: the &xe_guc struct instance
  * @scratch: scratch buffer to be used as temporary storage
+ *
+ * Returns: zero on success, negative error code on failure.
  */
-void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch)
+int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch)
 {
 	struct xe_exec_queue *q;
 	unsigned long index;
+	int err = 0;
 
 	mutex_lock(&guc->submission_state.lock);
-	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
-		xe_exec_queue_contexts_hwsp_rebase(q, scratch);
+	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) {
+		err = xe_exec_queue_contexts_hwsp_rebase(q, scratch);
+		if (err)
+			break;
+	}
 	mutex_unlock(&guc->submission_state.lock);
+
+	return err;
 }
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 92a6f0ade615..6b5df5d0956b 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -48,6 +48,6 @@ xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapsh
 void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p);
 void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type);
 
-void xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch);
+int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 9a67f4ae235a..8760c4c2ca38 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -41,7 +41,6 @@
 #define LRC_PPHWSP_SIZE				SZ_4K
 #define LRC_INDIRECT_CTX_BO_SIZE		SZ_4K
 #define LRC_INDIRECT_RING_STATE_SIZE		SZ_4K
-#define LRC_WA_BB_SIZE				SZ_4K
 
 /*
  * Layout of the LRC and associated data allocated as
@@ -1149,13 +1148,11 @@ static int setup_bo(struct bo_setup_state *state)
 	ssize_t remain;
 
 	if (state->lrc->bo->vmap.is_iomem) {
-		state->buffer = kmalloc(state->max_size, GFP_KERNEL);
 		if (!state->buffer)
 			return -ENOMEM;
 		state->ptr = state->buffer;
 	} else {
 		state->ptr = state->lrc->bo->vmap.vaddr + state->offset;
-		state->buffer = NULL;
 	}
 
 	remain = state->max_size / sizeof(u32);
@@ -1180,7 +1177,6 @@ static int setup_bo(struct bo_setup_state *state)
 	return 0;
 
 fail:
-	kfree(state->buffer);
 	return -ENOSPC;
 }
 
@@ -1192,10 +1188,16 @@ static void finish_bo(struct bo_setup_state *state)
 	xe_map_memcpy_to(gt_to_xe(state->lrc->gt), &state->lrc->bo->vmap,
 			 state->offset, state->buffer,
 			 state->written * sizeof(u32));
-	kfree(state->buffer);
 }
 
-static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
+/**
+ * xe_lrc_setup_wa_bb_with_scratch - Execute all wa bb setup callbacks.
+ * @lrc: the &xe_lrc struct instance
+ * @hwe: the &xe_hw_engine struct instance
+ * @scratch: preallocated scratch buffer for temporary storage
+ * Return: 0 on success, negative error code on failure
+ */
+int xe_lrc_setup_wa_bb_with_scratch(struct xe_lrc *lrc, struct xe_hw_engine *hwe, u32 *scratch)
 {
 	static const struct bo_setup funcs[] = {
 		{ .setup = setup_timestamp_wa },
@@ -1206,6 +1208,7 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 		.lrc = lrc,
 		.hwe = hwe,
 		.max_size = LRC_WA_BB_SIZE,
+		.buffer = scratch,
 		.reserve_dw = 1,
 		.offset = __xe_lrc_wa_bb_offset(lrc),
 		.funcs = funcs,
@@ -1228,6 +1231,21 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 	return 0;
 }
 
+static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
+{
+	u32 *buf = NULL;
+	int ret;
+
+	if (lrc->bo->vmap.is_iomem)
+		buf = kmalloc(LRC_WA_BB_SIZE, GFP_KERNEL);
+
+	ret = xe_lrc_setup_wa_bb_with_scratch(lrc, hwe, buf);
+
+	kfree(buf);
+
+	return ret;
+}
+
 static int
 setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 {
@@ -1238,6 +1256,7 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 		.lrc = lrc,
 		.hwe = hwe,
 		.max_size = (63 * 64) /* max 63 cachelines */,
+		.buffer = NULL,
 		.offset = __xe_lrc_indirect_ctx_offset(lrc),
 	};
 	int ret;
@@ -1254,9 +1273,14 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 	if (xe_gt_WARN_ON(lrc->gt, !state.funcs))
 		return 0;
 
+	if (lrc->bo->vmap.is_iomem)
+		state.buffer = kmalloc(state.max_size, GFP_KERNEL);
+
 	ret = setup_bo(&state);
-	if (ret)
+	if (ret) {
+		kfree(state.buffer);
 		return ret;
+	}
 
 	/*
 	 * Align to 64B cacheline so there's no garbage at the end for CS to
@@ -1268,6 +1292,7 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 	}
 
 	finish_bo(&state);
+	kfree(state.buffer);
 
 	xe_lrc_write_ctx_reg(lrc,
 			     CTX_CS_INDIRECT_CTX,
diff --git a/drivers/gpu/drm/xe/xe_lrc.h b/drivers/gpu/drm/xe/xe_lrc.h
index eceeeee6c021..188565465779 100644
--- a/drivers/gpu/drm/xe/xe_lrc.h
+++ b/drivers/gpu/drm/xe/xe_lrc.h
@@ -42,6 +42,8 @@ struct xe_lrc_snapshot {
 #define LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR (0x34 * 4)
 #define LRC_PPHWSP_PXP_INVAL_SCRATCH_ADDR (0x40 * 4)
 
+#define LRC_WA_BB_SIZE SZ_4K
+
 #define XE_LRC_CREATE_RUNALONE 0x1
 #define XE_LRC_CREATE_PXP 0x2
 struct xe_lrc *xe_lrc_create(struct xe_hw_engine *hwe, struct xe_vm *vm,
@@ -129,6 +131,8 @@ u32 xe_lrc_ctx_timestamp_udw_ggtt_addr(struct xe_lrc *lrc);
 u64 xe_lrc_ctx_timestamp(struct xe_lrc *lrc);
 u32 xe_lrc_ctx_job_timestamp_ggtt_addr(struct xe_lrc *lrc);
 u32 xe_lrc_ctx_job_timestamp(struct xe_lrc *lrc);
+int xe_lrc_setup_wa_bb_with_scratch(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
+				    u32 *scratch);
 
 /**
  * xe_lrc_update_timestamp - readout LRC timestamp and update cached value
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index a219395c15de..5de81f213d83 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -247,7 +247,7 @@ static int vf_get_next_migrated_gt_id(struct xe_device *xe)
 
 static size_t post_migration_scratch_size(struct xe_device *xe)
 {
-	return xe_lrc_reg_size(xe);
+	return max(xe_lrc_reg_size(xe), LRC_WA_BB_SIZE);
 }
 
 /**
@@ -274,22 +274,23 @@ static int gt_vf_post_migration_fixups(struct xe_gt *gt)
 		return -ENOMEM;
 
 	err = xe_gt_sriov_vf_query_config(gt);
-	if (err) {
-		kfree(buf);
-		return err;
-	}
+	if (err)
+		goto out;
 
 	shift = xe_gt_sriov_vf_ggtt_shift(gt);
 	if (shift) {
 		xe_tile_sriov_vf_fixup_ggtt_nodes(gt_to_tile(gt), shift);
 		xe_gt_sriov_vf_default_lrcs_hwsp_rebase(gt);
-		xe_guc_contexts_hwsp_rebase(&gt->uc.guc, buf);
+		err = xe_guc_contexts_hwsp_rebase(&gt->uc.guc, buf);
+		if (err)
+			goto out;
 		xe_guc_jobs_ring_rebase(&gt->uc.guc);
 		xe_guc_ct_fixup_messages_with_ggtt(&gt->uc.guc.ct, shift);
 	}
 
+out:
 	kfree(buf);
-	return 0;
+	return err;
 }
 
 static void vf_post_migration_recovery(struct xe_device *xe)

From ba180a362128cb71d16c3f0ce6645448011d2607 Mon Sep 17 00:00:00 2001
From: Tomasz Lis <tomasz.lis@intel.com>
Date: Sat, 2 Aug 2025 05:10:45 +0200
Subject: [PATCH 0204/1455] drm/xe/vf: Rebase exec queue parallel commands
 during migration recovery
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Parallel exec queues have an additional command streamer buffer which holds
a GGTT reference to data within context status. The GGTT references have to
be fixed after VF migration.

v2: Properly handle nop entry, verify if parsing goes ok
v3: Improve error/warn logging, add propagation of errors,
 give names to magic offsets

Signed-off-by: Tomasz Lis <tomasz.lis@intel.com>
Cc: Michal Winiarski <michal.winiarski@intel.com>
Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>
Link: https://lore.kernel.org/r/20250802031045.1127138-9-tomasz.lis@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/abi/guc_actions_abi.h |  8 ++++
 drivers/gpu/drm/xe/xe_guc_submit.c       | 54 ++++++++++++++++++++++++
 2 files changed, 62 insertions(+)

diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
index 81eb046aeebf..d8cf68a0516d 100644
--- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
@@ -193,6 +193,14 @@ enum xe_guc_register_context_multi_lrc_param_offsets {
 	XE_GUC_REGISTER_CONTEXT_MULTI_LRC_MSG_MIN_LEN = 11,
 };
 
+enum xe_guc_context_wq_item_offsets {
+	XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN = 0,
+	XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW,
+	XE_GUC_CONTEXT_WQ_EL_INFO_DATA_2_GUCCTX_RINGTAIL_FREEZEPOCS,
+	XE_GUC_CONTEXT_WQ_EL_INFO_DATA_3_WI_FENCE_ID,
+	XE_GUC_CONTEXT_WQ_EL_CHILD_LIST_DATA_4_RINGTAIL,
+};
+
 enum xe_guc_report_status {
 	XE_GUC_REPORT_STATUS_UNKNOWN = 0x0,
 	XE_GUC_REPORT_STATUS_ACKED = 0x1,
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 20f5b9820264..554a25ca77d3 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -671,12 +671,18 @@ static void wq_item_append(struct xe_exec_queue *q)
 	if (wq_wait_for_space(q, wqi_size))
 		return;
 
+	xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN);
 	wqi[i++] = FIELD_PREP(WQ_TYPE_MASK, WQ_TYPE_MULTI_LRC) |
 		FIELD_PREP(WQ_LEN_MASK, len_dw);
+	xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW);
 	wqi[i++] = xe_lrc_descriptor(q->lrc[0]);
+	xe_gt_assert(guc_to_gt(guc), i ==
+		     XE_GUC_CONTEXT_WQ_EL_INFO_DATA_2_GUCCTX_RINGTAIL_FREEZEPOCS);
 	wqi[i++] = FIELD_PREP(WQ_GUC_ID_MASK, q->guc->id) |
 		FIELD_PREP(WQ_RING_TAIL_MASK, q->lrc[0]->ring.tail / sizeof(u64));
+	xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_INFO_DATA_3_WI_FENCE_ID);
 	wqi[i++] = 0;
+	xe_gt_assert(guc_to_gt(guc), i == XE_GUC_CONTEXT_WQ_EL_CHILD_LIST_DATA_4_RINGTAIL);
 	for (j = 1; j < q->width; ++j) {
 		struct xe_lrc *lrc = q->lrc[j];
 
@@ -697,6 +703,50 @@ static void wq_item_append(struct xe_exec_queue *q)
 	parallel_write(xe, map, wq_desc.tail, q->guc->wqi_tail);
 }
 
+static int wq_items_rebase(struct xe_exec_queue *q)
+{
+	struct xe_guc *guc = exec_queue_to_guc(q);
+	struct xe_device *xe = guc_to_xe(guc);
+	struct iosys_map map = xe_lrc_parallel_map(q->lrc[0]);
+	int i = q->guc->wqi_head;
+
+	/* the ring starts after a header struct */
+	iosys_map_incr(&map, offsetof(struct guc_submit_parallel_scratch, wq[0]));
+
+	while ((i % WQ_SIZE) != (q->guc->wqi_tail % WQ_SIZE)) {
+		u32 len_dw, type, val;
+
+		if (drm_WARN_ON_ONCE(&xe->drm, i < 0 || i > 2 * WQ_SIZE))
+			break;
+
+		val = xe_map_rd_ring_u32(xe, &map, i / sizeof(u32) +
+					 XE_GUC_CONTEXT_WQ_HEADER_DATA_0_TYPE_LEN,
+					 WQ_SIZE / sizeof(u32));
+		len_dw = FIELD_GET(WQ_LEN_MASK, val);
+		type = FIELD_GET(WQ_TYPE_MASK, val);
+
+		if (drm_WARN_ON_ONCE(&xe->drm, len_dw >= WQ_SIZE / sizeof(u32)))
+			break;
+
+		if (type == WQ_TYPE_MULTI_LRC) {
+			val = xe_lrc_descriptor(q->lrc[0]);
+			xe_map_wr_ring_u32(xe, &map, i / sizeof(u32) +
+					   XE_GUC_CONTEXT_WQ_EL_INFO_DATA_1_CTX_DESC_LOW,
+					   WQ_SIZE / sizeof(u32), val);
+		} else if (drm_WARN_ON_ONCE(&xe->drm, type != WQ_TYPE_NOOP)) {
+			break;
+		}
+
+		i += (len_dw + 1) * sizeof(u32);
+	}
+
+	if ((i % WQ_SIZE) != (q->guc->wqi_tail % WQ_SIZE)) {
+		xe_gt_err(q->gt, "Exec queue fixups incomplete - wqi parse failed\n");
+		return -EBADMSG;
+	}
+	return 0;
+}
+
 #define RESUME_PENDING	~0x0ull
 static void submit_exec_queue(struct xe_exec_queue *q)
 {
@@ -2549,6 +2599,10 @@ int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch)
 		err = xe_exec_queue_contexts_hwsp_rebase(q, scratch);
 		if (err)
 			break;
+		if (xe_exec_queue_is_parallel(q))
+			err = wq_items_rebase(q);
+		if (err)
+			break;
 	}
 	mutex_unlock(&guc->submission_state.lock);
 

From 1da71a08080bb5cf5601f237e94c3dd5cede446b Mon Sep 17 00:00:00 2001
From: Paul Kocialkowski <paulk@sys-base.io>
Date: Wed, 2 Jul 2025 10:22:29 +0200
Subject: [PATCH 0205/1455] dt-bindings: display: simple: Add Olimex
 LCD-OLinuXino-5CTS

Add the Olimex LCD-OLinuXino-5CTS, a 5-inch TFT LCD panel.

Signed-off-by: Paul Kocialkowski <paulk@sys-base.io>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250702082230.1291953-1-paulk@sys-base.io
---
 .../devicetree/bindings/display/panel/panel-simple.yaml         | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index 1ac1f0219079..a03a9f1dd9f6 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -236,6 +236,8 @@ properties:
       - okaya,rs800480t-7x0gp
         # Olimex 4.3" TFT LCD panel
       - olimex,lcd-olinuxino-43-ts
+        # Olimex 5.0" TFT LCD panel
+      - olimex,lcd-olinuxino-5-cts
         # On Tat Industrial Company 5" DPI TFT panel.
       - ontat,kd50g21-40nt-a1
         # On Tat Industrial Company 7" DPI TFT panel.

From 26232780077b3e6147fabb6111afd0bd9a60f973 Mon Sep 17 00:00:00 2001
From: Paul Kocialkowski <paulk@sys-base.io>
Date: Wed, 2 Jul 2025 10:22:30 +0200
Subject: [PATCH 0206/1455] drm/panel: simple: Add Olimex LCD-OLinuXino-5CTS
 support

Add support for the Olimex LCD-OLinuXino-5CTS, a 5-inch TFT LCD panel
with a mode operating at 33.3 MHz.

Signed-off-by: Paul Kocialkowski <paulk@sys-base.io>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250702082230.1291953-2-paulk@sys-base.io
---
 drivers/gpu/drm/panel/panel-simple.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 47222d2d8129..fbd10396a374 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -3679,6 +3679,29 @@ static const struct panel_desc olimex_lcd_olinuxino_43ts = {
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
 };
 
+static const struct drm_display_mode olimex_lcd_olinuxino_5cts_mode = {
+	.clock = 33300,
+	.hdisplay = 800,
+	.hsync_start = 800 + 210,
+	.hsync_end = 800 + 210 + 20,
+	.htotal = 800 + 210 + 20 + 26,
+	.vdisplay = 480,
+	.vsync_start = 480 + 22,
+	.vsync_end = 480 + 22 + 10,
+	.vtotal = 480 + 22 + 10 + 13,
+};
+
+static const struct panel_desc olimex_lcd_olinuxino_5cts = {
+	.modes = &olimex_lcd_olinuxino_5cts_mode,
+	.num_modes = 1,
+	.size = {
+		.width = 154,
+		.height = 86,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB888_1X24,
+};
+
+
 static const struct display_timing ontat_kd50g21_40nt_a1_timing = {
 	.pixelclock = { 30000000, 30000000, 50000000 },
 	.hactive = { 800, 800, 800 },
@@ -5241,6 +5264,9 @@ static const struct of_device_id platform_of_match[] = {
 	}, {
 		.compatible = "olimex,lcd-olinuxino-43-ts",
 		.data = &olimex_lcd_olinuxino_43ts,
+	}, {
+		.compatible = "olimex,lcd-olinuxino-5-cts",
+		.data = &olimex_lcd_olinuxino_5cts,
 	}, {
 		.compatible = "ontat,kd50g21-40nt-a1",
 		.data = &ontat_kd50g21_40nt_a1,

From f3e1caef597e7f29dd2326dae7fa5964ae46c91d Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Mon, 21 Jul 2025 20:08:59 +0530
Subject: [PATCH 0207/1455] dt-bindings: display: panel: document Samsung
 AMS561RA01 panel with S6E8AA5X01 controller

Samsung AMS561RA01 is an AMOLED panel, using the Samsung S6E8AA5X01 MIPI
DSI panel controller. Document the compatible and devicetree properties
of this hardware. It has a reset GPIO and two voltage regulators.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250721-panel-samsung-s6e8aa5x01-v5-1-1a315aba530b@disroot.org
---
 .../panel/samsung,s6e8aa5x01-ams561ra01.yaml  | 55 +++++++++++++++++++
 1 file changed, 55 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/panel/samsung,s6e8aa5x01-ams561ra01.yaml

diff --git a/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa5x01-ams561ra01.yaml b/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa5x01-ams561ra01.yaml
new file mode 100644
index 000000000000..eccfc66d7fe2
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa5x01-ams561ra01.yaml
@@ -0,0 +1,55 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/samsung,s6e8aa5x01-ams561ra01.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Samsung AMS561RA01 panel with S6E8AA5X01 controller
+
+maintainers:
+  - Kaustabh Chakraborty <kauschluss@disroot.org>
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    const: samsung,s6e8aa5x01-ams561ra01
+
+  reg:
+    maxItems: 1
+
+  vdd-supply:
+    description: core voltage supply
+
+  vci-supply:
+    description: voltage supply for analog circuits
+
+  reset-gpios: true
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "samsung,s6e8aa5x01-ams561ra01";
+            reg = <0>;
+
+            vdd-supply = <&panel_vdd_reg>;
+            vci-supply = <&panel_vci_reg>;
+
+            reset-gpios = <&gpd3 4 GPIO_ACTIVE_HIGH>;
+        };
+    };
+
+...

From 47d7953d8e1e774265ec2c79e37871af97abadb2 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Mon, 21 Jul 2025 20:09:00 +0530
Subject: [PATCH 0208/1455] drm: panel: add support for Samsung AMS561RA01
 panel with S6E8AA5X01 controller

Samsung AMS561RA01 is an AMOLED panel, using the Samsung S6E8AA5X01 MIPI
DSI controller. Implement a basic panel driver for such panels.

The driver also initializes a backlight device, which works by changing
the panel's gamma values and aid brightness levels appropriately, with
the help of look-up tables acquired from downstream kernel sources.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250721-panel-samsung-s6e8aa5x01-v5-2-1a315aba530b@disroot.org
---
 drivers/gpu/drm/panel/Kconfig                 |  11 +
 drivers/gpu/drm/panel/Makefile                |   1 +
 .../panel-samsung-s6e8aa5x01-ams561ra01.c     | 981 ++++++++++++++++++
 3 files changed, 993 insertions(+)
 create mode 100644 drivers/gpu/drm/panel/panel-samsung-s6e8aa5x01-ams561ra01.c

diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index 09b9f7ff9340..a5d89b13de06 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -843,6 +843,17 @@ config DRM_PANEL_SAMSUNG_S6E8AA0
 	select DRM_MIPI_DSI
 	select VIDEOMODE_HELPERS
 
+config DRM_PANEL_SAMSUNG_S6E8AA5X01_AMS561RA01
+	tristate "Samsung AMS561RA01 panel with S6E8AA5X01 controller"
+	depends on GPIOLIB && OF && REGULATOR
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for Samsung AMS561RA01
+	  panel, which uses Samsung's S6E8AA5X01 controller. The panel has a
+	  ~5.6 inch AMOLED display, and the controller is driven by the MIPI
+	  DSI protocol with 4 lanes.
+
 config DRM_PANEL_SAMSUNG_SOFEF00
 	tristate "Samsung sofef00/s6e3fc2x01 OnePlus 6/6T DSI cmd mode panels"
 	depends on OF
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index 957555b49996..c425eb2940b0 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -87,6 +87,7 @@ obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E63M0_DSI) += panel-samsung-s6e63m0-dsi.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E88A0_AMS427AP24) += panel-samsung-s6e88a0-ams427ap24.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E88A0_AMS452EF01) += panel-samsung-s6e88a0-ams452ef01.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E8AA0) += panel-samsung-s6e8aa0.o
+obj-$(CONFIG_DRM_PANEL_SAMSUNG_S6E8AA5X01_AMS561RA01) += panel-samsung-s6e8aa5x01-ams561ra01.o
 obj-$(CONFIG_DRM_PANEL_SAMSUNG_SOFEF00) += panel-samsung-sofef00.o
 obj-$(CONFIG_DRM_PANEL_SEIKO_43WVF1G) += panel-seiko-43wvf1g.o
 obj-$(CONFIG_DRM_PANEL_SHARP_LQ101R1SX01) += panel-sharp-lq101r1sx01.o
diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e8aa5x01-ams561ra01.c b/drivers/gpu/drm/panel/panel-samsung-s6e8aa5x01-ams561ra01.c
new file mode 100644
index 000000000000..56e10c7c3a76
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e8aa5x01-ams561ra01.c
@@ -0,0 +1,981 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Samsung AMS561RA01 panel with S6E8AA5X01 controller.
+ *
+ * Copyright (C) 2025 Kaustabh Chakraborty <kauschluss@disroot.org>
+ */
+
+#include <linux/backlight.h>
+#include <linux/gpio/consumer.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/regulator/consumer.h>
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+#include <drm/drm_probe_helper.h>
+
+/* Manufacturer Command Set */
+#define MCS_AIDCTL		0xb2
+#define MCS_ADAPTIVECTL		0xb5
+#define MCS_ELVSS		0xb6
+#define MCS_TEMPERCTL		0xb8
+#define MCS_PENTILE		0xc0
+#define MCS_GAMMACTL		0xca
+#define MCS_LTPSCTL		0xcb
+#define MCS_PCD			0xcc
+#define MCS_ERRFLAG		0xe7
+#define MCS_ACCESSPROT		0xf0
+#define MCS_DISPCTL		0xf2
+#define MCS_GAMMAUPD		0xf7
+
+#define GAMMA_CMD_LEN	34
+#define AID_CMD_LEN	3
+
+static const struct {
+	u8 gamma[GAMMA_CMD_LEN];
+	u8 aid[AID_CMD_LEN];
+} s6e8aa5x01_ams561ra01_cmds[] = {
+	{
+		/* 5 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x94,
+		  0x88, 0x89, 0x8a, 0x87, 0x87, 0x89,
+		  0x8d, 0x8c, 0x8d, 0x89, 0x8c, 0x8e,
+		  0x8e, 0x8f, 0x90, 0xa3, 0xa2, 0x9a,
+		  0xcf, 0xca, 0x9f, 0xe6, 0xff, 0xb4,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0xa5 },
+	}, {
+		/* 6 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x95,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x8c, 0x8a, 0x8c, 0x85, 0x88, 0x8c,
+		  0x8b, 0x8c, 0x8e, 0xa2, 0xa2, 0x9a,
+		  0xd0, 0xcc, 0xa2, 0xed, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x95 },
+	}, {
+		/* 7 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x95,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x8c, 0x8a, 0x8c, 0x85, 0x88, 0x8c,
+		  0x8b, 0x8c, 0x8e, 0xa2, 0xa2, 0x99,
+		  0xc8, 0xc4, 0x9d, 0xed, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x89 },
+	}, {
+		/* 8 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8a, 0x87, 0x87, 0x89,
+		  0x8a, 0x88, 0x8b, 0x83, 0x86, 0x8b,
+		  0x8c, 0x8b, 0x8d, 0x9d, 0x9f, 0x97,
+		  0xc7, 0xc3, 0x9c, 0xf5, 0xff, 0xbb,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x7e },
+	}, {
+		/* 9 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8a, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x8a, 0x82, 0x84, 0x88,
+		  0x90, 0x8f, 0x91, 0x95, 0x97, 0x94,
+		  0xc6, 0xc2, 0x9d, 0xf5, 0xff, 0xbb,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x73 },
+	}, {
+		/* 10 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8a, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x8a, 0x82, 0x84, 0x88,
+		  0x90, 0x8f, 0x91, 0x94, 0x97, 0x93,
+		  0xc6, 0xc2, 0x9e, 0xec, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x67 },
+	}, {
+		/* 11 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8a, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x8a, 0x82, 0x84, 0x88,
+		  0x8b, 0x8b, 0x8d, 0x90, 0x93, 0x92,
+		  0xc5, 0xc1, 0x9c, 0xf5, 0xff, 0xbb,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x56 },
+	}, {
+		/* 12 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x82, 0x84, 0x88,
+		  0x87, 0x86, 0x8a, 0x8c, 0x90, 0x8f,
+		  0xcd, 0xc9, 0xa1, 0xec, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x4a },
+	}, {
+		/* 13 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x82, 0x84, 0x88,
+		  0x87, 0x86, 0x8a, 0x8c, 0x90, 0x8e,
+		  0xc4, 0xbf, 0x9c, 0xf5, 0xff, 0xbb,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x3b },
+	}, {
+		/* 14 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x82, 0x84, 0x88,
+		  0x87, 0x86, 0x89, 0x8c, 0x90, 0x8f,
+		  0xc2, 0xbf, 0x9c, 0xec, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x35 },
+	}, {
+		/* 15 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x82, 0x84, 0x88,
+		  0x87, 0x86, 0x89, 0x8c, 0x90, 0x8f,
+		  0xb7, 0xb6, 0x96, 0xec, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x25 },
+	}, {
+		/* 16 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x82, 0x84, 0x88,
+		  0x88, 0x86, 0x89, 0x8c, 0x90, 0x8f,
+		  0xb7, 0xb6, 0x96, 0xec, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x20 },
+	}, {
+		/* 17 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x7f, 0x80, 0x86,
+		  0x86, 0x85, 0x89, 0x88, 0x8c, 0x8e,
+		  0xbf, 0xbe, 0x9c, 0xec, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x05, 0x11 },
+	}, {
+		/* 19 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x7f, 0x80, 0x86,
+		  0x87, 0x85, 0x89, 0x88, 0x8c, 0x8e,
+		  0xb3, 0xb4, 0x97, 0xeb, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0xf2 },
+	}, {
+		/* 20 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x95,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x89, 0x86, 0x89, 0x7f, 0x80, 0x86,
+		  0x87, 0x85, 0x89, 0x89, 0x8c, 0x8e,
+		  0xb3, 0xb4, 0x97, 0xeb, 0xff, 0xb7,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0xe4 },
+	}, {
+		/* 21 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x96,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x8a, 0x88, 0x8b, 0x7d, 0x7e, 0x84,
+		  0x8c, 0x8a, 0x8c, 0x8e, 0x90, 0x8f,
+		  0xb6, 0xb6, 0x97, 0xe3, 0xff, 0xb3,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0xd5 },
+	}, {
+		/* 22 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x97,
+		  0x88, 0x89, 0x8b, 0x87, 0x87, 0x89,
+		  0x8a, 0x88, 0x8b, 0x81, 0x82, 0x86,
+		  0x87, 0x86, 0x88, 0x8e, 0x90, 0x8f,
+		  0xb6, 0xb6, 0x95, 0xe3, 0xff, 0xb3,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0xc5 },
+	}, {
+		/* 24 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x97,
+		  0x88, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8a, 0x81, 0x82, 0x86,
+		  0x87, 0x86, 0x88, 0x8e, 0x90, 0x8f,
+		  0xb6, 0xb6, 0x94, 0xe3, 0xff, 0xb3,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0xa7 },
+	}, {
+		/* 25 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x98,
+		  0x88, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8a, 0x81, 0x82, 0x86,
+		  0x87, 0x86, 0x87, 0x8e, 0x90, 0x8f,
+		  0xbf, 0xbf, 0x9a, 0xda, 0xfa, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0x95 },
+	}, {
+		/* 27 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x99,
+		  0x88, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8a, 0x83, 0x86, 0x8a,
+		  0x88, 0x87, 0x87, 0x88, 0x8b, 0x8c,
+		  0xbf, 0xbf, 0x9a, 0xda, 0xfa, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0x76 },
+	}, {
+		/* 29 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x99,
+		  0x88, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x83, 0x86, 0x89,
+		  0x88, 0x87, 0x88, 0x88, 0x8b, 0x8b,
+		  0xbf, 0xbf, 0x9a, 0xda, 0xfa, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0x54 },
+	}, {
+		/* 30 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9a,
+		  0x88, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8a, 0x84, 0x86, 0x8a,
+		  0x87, 0x87, 0x87, 0x88, 0x8b, 0x8b,
+		  0xbf, 0xbf, 0x99, 0xda, 0xfa, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0x44 },
+	}, {
+		/* 32 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9a,
+		  0x89, 0x89, 0x8c, 0x88, 0x88, 0x8a,
+		  0x89, 0x87, 0x8a, 0x84, 0x86, 0x8a,
+		  0x87, 0x87, 0x87, 0x89, 0x8b, 0x8b,
+		  0xbf, 0xbf, 0x98, 0xd2, 0xf2, 0xac,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x04, 0x1f },
+	}, {
+		/* 34 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9b,
+		  0x88, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8b, 0x87, 0x8b, 0x83, 0x86, 0x89,
+		  0x87, 0x87, 0x88, 0x88, 0x8b, 0x8a,
+		  0xbf, 0xbf, 0x98, 0xd2, 0xf2, 0xac,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0xff },
+	}, {
+		/* 37 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9b,
+		  0x89, 0x89, 0x8c, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8a, 0x81, 0x82, 0x86,
+		  0x86, 0x86, 0x86, 0x8d, 0x90, 0x8d,
+		  0xc0, 0xbf, 0x9a, 0xd2, 0xf2, 0xac,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0xd3 },
+	}, {
+		/* 39 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9b,
+		  0x89, 0x89, 0x8c, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8a, 0x81, 0x82, 0x86,
+		  0x87, 0x86, 0x87, 0x8d, 0x90, 0x8d,
+		  0xb6, 0xb6, 0x93, 0xda, 0xf9, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0xb3 },
+	}, {
+		/* 41 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9b,
+		  0x89, 0x89, 0x8c, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x81, 0x82, 0x85,
+		  0x87, 0x86, 0x87, 0x8d, 0x90, 0x8d,
+		  0xb6, 0xb6, 0x94, 0xda, 0xf9, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0x93 },
+	}, {
+		/* 44 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9b,
+		  0x89, 0x89, 0x8c, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x81, 0x82, 0x86,
+		  0x87, 0x86, 0x86, 0x85, 0x87, 0x8a,
+		  0xbe, 0xbe, 0x99, 0xda, 0xf9, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0x66 },
+	}, {
+		/* 47 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9b,
+		  0x89, 0x89, 0x8c, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x81, 0x82, 0x86,
+		  0x88, 0x86, 0x87, 0x84, 0x87, 0x89,
+		  0xb4, 0xb4, 0x94, 0xe2, 0xff, 0xb3,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0x40 },
+	}, {
+		/* 50 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9c,
+		  0x89, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x81, 0x82, 0x86,
+		  0x88, 0x86, 0x87, 0x84, 0x87, 0x89,
+		  0xb4, 0xb4, 0x95, 0xe2, 0xff, 0xb3,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x03, 0x0e },
+	}, {
+		/* 53 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9c,
+		  0x89, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x81, 0x82, 0x86,
+		  0x88, 0x86, 0x87, 0x85, 0x87, 0x8a,
+		  0xb4, 0xb4, 0x96, 0xe2, 0xff, 0xb3,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0xe2 },
+	}, {
+		/* 56 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9c,
+		  0x89, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x81, 0x82, 0x86,
+		  0x88, 0x86, 0x87, 0x85, 0x87, 0x8a,
+		  0xab, 0xab, 0x90, 0xdd, 0xf7, 0xaf,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0xb5 },
+	}, {
+		/* 60 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9c,
+		  0x89, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x82, 0x82, 0x87,
+		  0x83, 0x81, 0x84, 0x81, 0x84, 0x88,
+		  0xb3, 0xb3, 0x96, 0xcf, 0xe5, 0xa8,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x77 },
+	}, {
+		/* 64 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x98, 0x00, 0xa4, 0x00, 0x9c,
+		  0x89, 0x89, 0x8b, 0x88, 0x88, 0x8a,
+		  0x8a, 0x87, 0x8b, 0x82, 0x82, 0x87,
+		  0x83, 0x81, 0x84, 0x82, 0x84, 0x88,
+		  0xb2, 0xb3, 0x97, 0xcf, 0xe5, 0xa8,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x36 },
+	}, {
+		/* 68 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x9b, 0x00, 0xa6, 0x00, 0x9d,
+		  0x88, 0x88, 0x89, 0x89, 0x89, 0x8b,
+		  0x8a, 0x88, 0x8b, 0x7f, 0x80, 0x86,
+		  0x88, 0x86, 0x87, 0x7d, 0x7f, 0x85,
+		  0xb2, 0xb3, 0x97, 0xcf, 0xe5, 0xa8,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 72 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0x9c, 0x00, 0xa9, 0x00, 0xa0,
+		  0x88, 0x88, 0x89, 0x88, 0x88, 0x8a,
+		  0x8c, 0x8a, 0x8d, 0x7f, 0x81, 0x85,
+		  0x84, 0x82, 0x84, 0x85, 0x87, 0x8a,
+		  0xaa, 0xab, 0x93, 0xcf, 0xe5, 0xa8,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 77 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xa1, 0x00, 0xad, 0x00, 0xa5,
+		  0x89, 0x89, 0x8a, 0x88, 0x87, 0x89,
+		  0x8c, 0x89, 0x8d, 0x7f, 0x81, 0x85,
+		  0x84, 0x83, 0x84, 0x81, 0x83, 0x86,
+		  0xaa, 0xab, 0x93, 0xc0, 0xd3, 0xa1,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 82 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xa5, 0x00, 0xb0, 0x00, 0xa9,
+		  0x88, 0x89, 0x89, 0x85, 0x86, 0x89,
+		  0x8a, 0x88, 0x8b, 0x82, 0x82, 0x87,
+		  0x81, 0x80, 0x82, 0x89, 0x8b, 0x8b,
+		  0xa2, 0xa3, 0x8e, 0xc0, 0xd3, 0xa1,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 87 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xab, 0x00, 0xb4, 0x00, 0xad,
+		  0x88, 0x89, 0x8a, 0x84, 0x86, 0x88,
+		  0x8a, 0x88, 0x8b, 0x7f, 0x7f, 0x84,
+		  0x86, 0x84, 0x85, 0x85, 0x86, 0x88,
+		  0xa2, 0xa3, 0x8f, 0xc0, 0xd3, 0xa1,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 93 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xaf, 0x00, 0xb9, 0x00, 0xb1,
+		  0x88, 0x89, 0x8a, 0x84, 0x85, 0x87,
+		  0x8a, 0x89, 0x8b, 0x7e, 0x7e, 0x83,
+		  0x87, 0x86, 0x86, 0x88, 0x8a, 0x89,
+		  0x9c, 0x9c, 0x8b, 0xc0, 0xd3, 0xa1,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 98 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xb3, 0x00, 0xbc, 0x00, 0xb5,
+		  0x88, 0x88, 0x88, 0x84, 0x84, 0x86,
+		  0x8a, 0x88, 0x8a, 0x7f, 0x7f, 0x84,
+		  0x84, 0x83, 0x84, 0x88, 0x8a, 0x89,
+		  0x9c, 0x9c, 0x8b, 0xc0, 0xd3, 0xa1,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 105 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xb7, 0x00, 0xc0, 0x00, 0xba,
+		  0x87, 0x87, 0x88, 0x85, 0x85, 0x87,
+		  0x89, 0x88, 0x89, 0x7f, 0x7f, 0x83,
+		  0x81, 0x80, 0x82, 0x88, 0x8a, 0x89,
+		  0x9c, 0x9c, 0x8c, 0xb2, 0xc2, 0x9a,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 111 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xbb, 0x00, 0xc3, 0x00, 0xbe,
+		  0x87, 0x87, 0x88, 0x85, 0x85, 0x88,
+		  0x88, 0x87, 0x89, 0x80, 0x80, 0x84,
+		  0x81, 0x81, 0x82, 0x85, 0x86, 0x87,
+		  0x9c, 0x9c, 0x8b, 0xb2, 0xc2, 0x9a,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x15 },
+	}, {
+		/* 119 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc4,
+		  0x87, 0x87, 0x88, 0x82, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x82, 0x81, 0x84,
+		  0x83, 0x82, 0x83, 0x80, 0x81, 0x84,
+		  0x9c, 0x9c, 0x8c, 0xb2, 0xc2, 0x9a,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x02, 0x14 },
+	}, {
+		/* 126 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc4,
+		  0x87, 0x87, 0x88, 0x82, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x82, 0x81, 0x84,
+		  0x83, 0x82, 0x83, 0x80, 0x81, 0x84,
+		  0x9c, 0x9c, 0x8d, 0xb2, 0xc2, 0x9a,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x01, 0xde },
+	}, {
+		/* 134 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc4,
+		  0x87, 0x87, 0x88, 0x82, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x82, 0x81, 0x84,
+		  0x83, 0x82, 0x83, 0x80, 0x81, 0x84,
+		  0x9c, 0x9c, 0x8d, 0xa4, 0xb0, 0x92,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x01, 0x94 },
+	}, {
+		/* 143 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc3,
+		  0x87, 0x87, 0x88, 0x82, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x82, 0x81, 0x85,
+		  0x83, 0x82, 0x83, 0x80, 0x81, 0x84,
+		  0x92, 0x92, 0x89, 0xab, 0xb6, 0x96,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x01, 0x46 },
+	}, {
+		/* 152 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc3,
+		  0x87, 0x87, 0x88, 0x83, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x81, 0x81, 0x85,
+		  0x84, 0x82, 0x83, 0x80, 0x81, 0x83,
+		  0x92, 0x92, 0x8b, 0xab, 0xb6, 0x96,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0xfa },
+	}, {
+		/* 162 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc3,
+		  0x87, 0x87, 0x88, 0x83, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x81, 0x81, 0x84,
+		  0x84, 0x82, 0x84, 0x80, 0x81, 0x83,
+		  0x92, 0x92, 0x8b, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0xac },
+	}, {
+		/* 172 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc0, 0x00, 0xc8, 0x00, 0xc3,
+		  0x87, 0x87, 0x88, 0x83, 0x84, 0x86,
+		  0x87, 0x85, 0x87, 0x81, 0x81, 0x84,
+		  0x84, 0x82, 0x83, 0x80, 0x81, 0x84,
+		  0x93, 0x92, 0x8c, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x57 },
+	}, {
+		/* 183 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc2, 0x00, 0xca, 0x00, 0xc5,
+		  0x86, 0x86, 0x87, 0x85, 0x84, 0x87,
+		  0x87, 0x86, 0x88, 0x7e, 0x80, 0x83,
+		  0x84, 0x82, 0x83, 0x80, 0x81, 0x83,
+		  0x93, 0x92, 0x8c, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 195 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xc7, 0x00, 0xce, 0x00, 0xc9,
+		  0x86, 0x87, 0x86, 0x83, 0x83, 0x85,
+		  0x85, 0x84, 0x86, 0x82, 0x82, 0x85,
+		  0x80, 0x80, 0x81, 0x81, 0x81, 0x84,
+		  0x93, 0x92, 0x8c, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 207 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xcc, 0x00, 0xd2, 0x00, 0xce,
+		  0x86, 0x86, 0x87, 0x81, 0x83, 0x84,
+		  0x84, 0x82, 0x84, 0x83, 0x83, 0x85,
+		  0x81, 0x81, 0x82, 0x7c, 0x7d, 0x81,
+		  0x93, 0x92, 0x8c, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 220 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xd1, 0x00, 0xd6, 0x00, 0xd3,
+		  0x86, 0x86, 0x86, 0x81, 0x83, 0x84,
+		  0x84, 0x82, 0x84, 0x80, 0x80, 0x83,
+		  0x81, 0x81, 0x82, 0x7c, 0x7d, 0x81,
+		  0x93, 0x92, 0x8c, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 234 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xd6, 0x00, 0xdb, 0x00, 0xd8,
+		  0x85, 0x85, 0x85, 0x81, 0x83, 0x84,
+		  0x83, 0x82, 0x83, 0x80, 0x80, 0x82,
+		  0x84, 0x82, 0x83, 0x79, 0x79, 0x7e,
+		  0x93, 0x92, 0x8d, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 249 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xdc, 0x00, 0xe0, 0x00, 0xdd,
+		  0x84, 0x84, 0x84, 0x81, 0x82, 0x83,
+		  0x84, 0x82, 0x84, 0x7f, 0x7f, 0x82,
+		  0x81, 0x80, 0x81, 0x80, 0x81, 0x82,
+		  0x8c, 0x8c, 0x86, 0x9d, 0xa4, 0x8e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 265 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xe2, 0x00, 0xe5, 0x00, 0xe3,
+		  0x83, 0x83, 0x83, 0x81, 0x82, 0x83,
+		  0x82, 0x82, 0x83, 0x82, 0x81, 0x83,
+		  0x7f, 0x7e, 0x80, 0x7c, 0x7d, 0x80,
+		  0x8c, 0x8c, 0x86, 0x8e, 0x92, 0x87,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 282 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xe8, 0x00, 0xea, 0x00, 0xe9,
+		  0x83, 0x83, 0x83, 0x80, 0x82, 0x82,
+		  0x81, 0x82, 0x82, 0x82, 0x81, 0x82,
+		  0x81, 0x80, 0x81, 0x80, 0x80, 0x81,
+		  0x85, 0x85, 0x83, 0x8e, 0x92, 0x87,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 300 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xed, 0x00, 0xef, 0x00, 0xed,
+		  0x81, 0x82, 0x81, 0x81, 0x81, 0x82,
+		  0x82, 0x82, 0x83, 0x80, 0x80, 0x81,
+		  0x81, 0x81, 0x82, 0x83, 0x83, 0x83,
+		  0x80, 0x80, 0x7f, 0x8e, 0x92, 0x87,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 316 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xf3, 0x00, 0xf4, 0x00, 0xf3,
+		  0x80, 0x81, 0x80, 0x81, 0x81, 0x81,
+		  0x82, 0x82, 0x82, 0x81, 0x80, 0x81,
+		  0x82, 0x82, 0x83, 0x80, 0x80, 0x80,
+		  0x80, 0x80, 0x7f, 0x80, 0x80, 0x80,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 333 nits */
+		{ MCS_GAMMACTL,
+		  0x00, 0xf8, 0x00, 0xf8, 0x00, 0xf8,
+		  0x80, 0x81, 0x80, 0x81, 0x80, 0x81,
+		  0x81, 0x82, 0x82, 0x81, 0x80, 0x81,
+		  0x83, 0x83, 0x83, 0x7e, 0x7d, 0x7e,
+		  0x80, 0x80, 0x7f, 0x80, 0x80, 0x80,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 360 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x00, 0x01, 0x00, 0x01, 0x00,
+		  0x80, 0x80, 0x80, 0x80, 0x80, 0x80,
+		  0x80, 0x80, 0x80, 0x80, 0x80, 0x80,
+		  0x80, 0x80, 0x80, 0x80, 0x80, 0x80,
+		  0x80, 0x80, 0x80, 0x80, 0x80, 0x80,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 378 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x04, 0x01, 0x03, 0x01, 0x04,
+		  0x7f, 0x7f, 0x80, 0x7f, 0x7f, 0x7f,
+		  0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f,
+		  0x80, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f,
+		  0x80, 0x80, 0x7f, 0x7f, 0x7f, 0x7f,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 395 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x09, 0x01, 0x07, 0x01, 0x08,
+		  0x7e, 0x7f, 0x80, 0x7f, 0x7f, 0x7f,
+		  0x7f, 0x7f, 0x7f, 0x7f, 0x7f, 0x7f,
+		  0x80, 0x7f, 0x7f, 0x7e, 0x7e, 0x7e,
+		  0x80, 0x80, 0x7f, 0x7e, 0x7e, 0x7f,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 413 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x0e, 0x01, 0x0b, 0x01, 0x0c,
+		  0x7e, 0x7f, 0x80, 0x7e, 0x7e, 0x7e,
+		  0x7e, 0x7e, 0x7e, 0x7f, 0x7f, 0x7f,
+		  0x80, 0x7f, 0x7f, 0x7d, 0x7d, 0x7d,
+		  0x80, 0x80, 0x7f, 0x7d, 0x7e, 0x7e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 430 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x13, 0x01, 0x0f, 0x01, 0x10,
+		  0x7d, 0x7f, 0x80, 0x7e, 0x7e, 0x7e,
+		  0x7e, 0x7e, 0x7e, 0x7f, 0x7f, 0x7f,
+		  0x80, 0x7f, 0x7f, 0x7d, 0x7d, 0x7d,
+		  0x80, 0x80, 0x7f, 0x7c, 0x7d, 0x7e,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 448 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x18, 0x01, 0x13, 0x01, 0x14,
+		  0x7c, 0x7e, 0x80, 0x7e, 0x7e, 0x7e,
+		  0x7e, 0x7e, 0x7d, 0x7e, 0x7f, 0x7e,
+		  0x80, 0x7f, 0x7f, 0x7c, 0x7c, 0x7c,
+		  0x80, 0x80, 0x7e, 0x7b, 0x7c, 0x7d,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 465 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x1d, 0x01, 0x17, 0x01, 0x18,
+		  0x7c, 0x7e, 0x80, 0x7d, 0x7d, 0x7d,
+		  0x7d, 0x7d, 0x7d, 0x7e, 0x7f, 0x7e,
+		  0x80, 0x7f, 0x7f, 0x7b, 0x7b, 0x7b,
+		  0x80, 0x80, 0x7e, 0x7a, 0x7c, 0x7d,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 483 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x22, 0x01, 0x1b, 0x01, 0x1c,
+		  0x7b, 0x7e, 0x80, 0x7d, 0x7d, 0x7d,
+		  0x7d, 0x7d, 0x7c, 0x7e, 0x7f, 0x7e,
+		  0x80, 0x7f, 0x7f, 0x7a, 0x7a, 0x7a,
+		  0x80, 0x80, 0x7e, 0x79, 0x7b, 0x7c,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	}, {
+		/* 500 nits */
+		{ MCS_GAMMACTL,
+		  0x01, 0x27, 0x01, 0x1f, 0x01, 0x20,
+		  0x7b, 0x7e, 0x80, 0x7d, 0x7d, 0x7d,
+		  0x7d, 0x7d, 0x7c, 0x7e, 0x7f, 0x7e,
+		  0x80, 0x7f, 0x7f, 0x7a, 0x7a, 0x7a,
+		  0x81, 0x80, 0x7e, 0x79, 0x7b, 0x7c,
+		  0x00, 0x00, 0x00, },
+		{ MCS_AIDCTL, 0x00, 0x10 },
+	},
+};
+
+struct s6e8aa5x01_ams561ra01_ctx {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct backlight_device *bl;
+	struct gpio_desc *reset_gpio;
+	struct regulator_bulk_data *supplies;
+	u32 nr_supplies;
+};
+
+static const struct regulator_bulk_data s6e8aa5x01_ams561ra01_supplies[] = {
+	{ .supply = "vdd" },
+	{ .supply = "vci" },
+};
+
+static inline struct s6e8aa5x01_ams561ra01_ctx *to_ctx(struct drm_panel *panel)
+{
+	return container_of(panel, struct s6e8aa5x01_ams561ra01_ctx, panel);
+}
+
+static int s6e8aa5x01_ams561ra01_update_status(struct backlight_device *bl)
+{
+	struct s6e8aa5x01_ams561ra01_ctx *ctx = bl_get_data(bl);
+	struct mipi_dsi_multi_context dsi = { .dsi = ctx->dsi };
+	u16 lvl = backlight_get_brightness(bl);
+
+	if (!ctx->panel.enabled)
+		return 0;
+
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_ACCESSPROT, 0x5a, 0x5a);
+
+	mipi_dsi_dcs_write_buffer_multi(&dsi,
+					s6e8aa5x01_ams561ra01_cmds[lvl].gamma,
+					GAMMA_CMD_LEN);
+	mipi_dsi_dcs_write_buffer_multi(&dsi,
+					s6e8aa5x01_ams561ra01_cmds[lvl].aid,
+					AID_CMD_LEN);
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_GAMMAUPD, 0x03);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_ACCESSPROT, 0xa5, 0xa5);
+
+	return dsi.accum_err;
+}
+
+static int s6e8aa5x01_ams561ra01_prepare(struct drm_panel *panel)
+{
+	struct s6e8aa5x01_ams561ra01_ctx *ctx = to_ctx(panel);
+	struct device *dev = &ctx->dsi->dev;
+	int ret;
+
+	ret = regulator_bulk_enable(ctx->nr_supplies, ctx->supplies);
+	if (ret < 0) {
+		dev_err(dev, "failed to enable regulators: %d\n", ret);
+		return ret;
+	}
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	usleep_range(5000, 6000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(5000, 6000);
+	gpiod_set_value_cansleep(ctx->reset_gpio, 0);
+	usleep_range(10000, 11000);
+
+	return 0;
+}
+
+static int s6e8aa5x01_ams561ra01_unprepare(struct drm_panel *panel)
+{
+	struct s6e8aa5x01_ams561ra01_ctx *ctx = to_ctx(panel);
+
+	gpiod_set_value_cansleep(ctx->reset_gpio, 1);
+	usleep_range(5000, 6000);
+
+	regulator_bulk_disable(ctx->nr_supplies, ctx->supplies);
+
+	return 0;
+}
+
+static int s6e8aa5x01_ams561ra01_enable(struct drm_panel *panel)
+{
+	struct s6e8aa5x01_ams561ra01_ctx *ctx = to_ctx(panel);
+	struct mipi_dsi_multi_context dsi = { .dsi = ctx->dsi };
+
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi);
+	mipi_dsi_msleep(&dsi, 100);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_ACCESSPROT, 0x5a, 0x5a);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_PENTILE, 0xd8, 0xd8, 0x00);
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_PCD, 0x5c);
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_ERRFLAG, 0xed, 0xc7, 0x23, 0x67);
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_DISPCTL, 0x0c, 0x0c, 0xb9, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_LTPSCTL,
+				0x00, 0x45, 0x10, 0x10, 0x08, 0x32, 0x54, 0x00,
+				0x00, 0x00, 0x00, 0x07, 0x06, 0x00, 0x00, 0x00,
+				0x00, 0x00, 0x48, 0x5e, 0x00, 0x00, 0x00, 0x00,
+				0x00, 0x03, 0x00, 0x00, 0x00, 0xad, 0x00, 0x00,
+				0x08, 0x05, 0x2a, 0x54, 0x03, 0xcc, 0x00, 0xff,
+				0xfb, 0x03, 0x0d, 0x00, 0x11, 0x0f, 0x02, 0x03,
+				0x0b, 0x0c, 0x13, 0x13, 0x13, 0x13, 0x13, 0x13,
+				0x13, 0x13, 0x13, 0x13, 0x00, 0x02, 0x03, 0x0b,
+				0x0c, 0x13, 0x13, 0x13, 0x13, 0x13, 0x13, 0x13,
+				0x13, 0x13);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi, MCS_ACCESSPROT, 0xa5, 0xa5);
+
+	mipi_dsi_dcs_set_display_on_multi(&dsi);
+
+	return dsi.accum_err;
+}
+
+static int s6e8aa5x01_ams561ra01_disable(struct drm_panel *panel)
+{
+	struct s6e8aa5x01_ams561ra01_ctx *ctx = to_ctx(panel);
+	struct mipi_dsi_multi_context dsi = { .dsi = ctx->dsi };
+
+	mipi_dsi_dcs_set_display_off_multi(&dsi);
+	mipi_dsi_msleep(&dsi, 100);
+
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi);
+	mipi_dsi_msleep(&dsi, 150);
+
+	return dsi.accum_err;
+}
+
+static const struct drm_display_mode s6e8aa5x01_ams561ra01_mode = {
+	.clock = (720 + 62 + 2 + 26) * (1480 + 12 + 2 + 10) * 60 / 1000,
+	.hdisplay = 720,
+	.hsync_start = 720 + 62,
+	.hsync_end = 720 + 62 + 2,
+	.htotal = 720 + 62 + 2 + 26,
+	.vdisplay = 1480,
+	.vsync_start = 1480 + 12,
+	.vsync_end = 1480 + 12 + 2,
+	.vtotal = 1480 + 12 + 2 + 10,
+	.width_mm = 62,
+	.height_mm = 128,
+	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
+};
+
+static int s6e8aa5x01_ams561ra01_get_modes(struct drm_panel *panel,
+					   struct drm_connector *connector)
+{
+	return drm_connector_helper_get_modes_fixed(connector,
+						    &s6e8aa5x01_ams561ra01_mode);
+}
+
+static const struct backlight_ops s6e8aa5x01_ams561ra01_bl_ops = {
+	.update_status = s6e8aa5x01_ams561ra01_update_status,
+};
+
+static const struct drm_panel_funcs s6e8aa5x01_ams561ra01_panel_funcs = {
+	.prepare = s6e8aa5x01_ams561ra01_prepare,
+	.unprepare = s6e8aa5x01_ams561ra01_unprepare,
+	.enable = s6e8aa5x01_ams561ra01_enable,
+	.disable = s6e8aa5x01_ams561ra01_disable,
+	.get_modes = s6e8aa5x01_ams561ra01_get_modes,
+};
+
+static int s6e8aa5x01_ams561ra01_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct s6e8aa5x01_ams561ra01_ctx *ctx;
+	int ret;
+
+	ctx = devm_drm_panel_alloc(dev, struct s6e8aa5x01_ams561ra01_ctx, panel,
+				   &s6e8aa5x01_ams561ra01_panel_funcs,
+				   DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
+
+	ctx->dsi = dsi;
+	mipi_dsi_set_drvdata(dsi, ctx);
+
+	ctx->nr_supplies = ARRAY_SIZE(s6e8aa5x01_ams561ra01_supplies);
+	ret = devm_regulator_bulk_get_const(dev, ctx->nr_supplies,
+					    s6e8aa5x01_ams561ra01_supplies,
+					    &ctx->supplies);
+	if (ret < 0)
+		return dev_err_probe(dev, ret, "failed to get regulators\n");
+
+	ctx->reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_ASIS);
+	if (IS_ERR(ctx->reset_gpio))
+		return dev_err_probe(dev, PTR_ERR(ctx->reset_gpio),
+				     "failed to get reset-gpios\n");
+
+	ctx->bl = devm_backlight_device_register(dev, dev_name(dev), dev, ctx,
+						 &s6e8aa5x01_ams561ra01_bl_ops,
+						 NULL);
+	if (IS_ERR(ctx->bl))
+		return dev_err_probe(dev, PTR_ERR(ctx->bl),
+				     "failed to register backlight device\n");
+
+	ctx->bl->props.type = BACKLIGHT_PLATFORM;
+	ctx->bl->props.brightness = ARRAY_SIZE(s6e8aa5x01_ams561ra01_cmds) - 1;
+	ctx->bl->props.max_brightness = ctx->bl->props.brightness;
+
+	dsi->lanes = 4;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
+			  MIPI_DSI_MODE_VIDEO_NO_HFP;
+
+	ctx->panel.prepare_prev_first = true;
+	drm_panel_add(&ctx->panel);
+
+	ret = devm_mipi_dsi_attach(dev, dsi);
+	if (ret < 0) {
+		drm_panel_remove(&ctx->panel);
+		return dev_err_probe(dev, ret, "failed to attach to DSI host\n");
+	}
+
+	return 0;
+}
+
+static void s6e8aa5x01_ams561ra01_remove(struct mipi_dsi_device *dsi)
+{
+	struct s6e8aa5x01_ams561ra01_ctx *ctx = mipi_dsi_get_drvdata(dsi);
+
+	drm_panel_remove(&ctx->panel);
+}
+
+static const struct of_device_id s6e8aa5x01_ams561ra01_of_device_id[] = {
+	{ .compatible = "samsung,s6e8aa5x01-ams561ra01" },
+	{ }
+};
+MODULE_DEVICE_TABLE(of, s6e8aa5x01_ams561ra01_of_device_id);
+
+static struct mipi_dsi_driver s6e8aa5x01_ams561ra01_dsi_driver = {
+	.probe = s6e8aa5x01_ams561ra01_probe,
+	.remove = s6e8aa5x01_ams561ra01_remove,
+	.driver = {
+		.name = "panel-samsung-s6e8aa5x01-ams561ra01",
+		.of_match_table = s6e8aa5x01_ams561ra01_of_device_id,
+	},
+};
+module_mipi_dsi_driver(s6e8aa5x01_ams561ra01_dsi_driver);
+
+MODULE_AUTHOR("Kaustabh Chakraborty <kauschluss@disroot.org>");
+MODULE_DESCRIPTION("Samsung AMS561RA01 Panel with S6E8AA5X01 Controller");
+MODULE_LICENSE("GPL");

From 368ea3f33f01b618bd645aa61b673f92e9aae767 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Thu, 24 Jul 2025 12:23:58 +0100
Subject: [PATCH 0209/1455] drm/panel: Kconfig: Fix spelling mistake "pannel"
 -> "panel"

There is a spelling mistake in the LEDS_BD2606MVV config. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250724112358.142351-1-colin.i.king@gmail.com
---
 drivers/gpu/drm/panel/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index a5d89b13de06..e2a43c33ade1 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -982,7 +982,7 @@ config DRM_PANEL_STARTEK_KD070FHFID015
 	depends on BACKLIGHT_CLASS_DEVICE
 	help
 	  Say Y here if you want to enable support for STARTEK KD070FHFID015 DSI panel
-	  based on RENESAS-R69429 controller. The pannel is a 7-inch TFT LCD display
+	  based on RENESAS-R69429 controller. The panel is a 7-inch TFT LCD display
 	  with a resolution of 1024 x 600 pixels. It provides a MIPI DSI interface to
 	  the host, a built-in LED backlight and touch controller.
 

From 640d512caa64b569e58a08b540d9c400e1aa8f94 Mon Sep 17 00:00:00 2001
From: Akhilesh Patil <akhilesh@ee.iitb.ac.in>
Date: Thu, 24 Jul 2025 21:38:32 +0530
Subject: [PATCH 0210/1455] drm: panel: orisetech: improve error handling
 during probe

Use dev_err_probe() helper as directed by core driver model to handle
driver probe error. Use standard helper defined at drivers/base/core.c
to maintain code consistency.

Inspired by,
commit a787e5400a1ce ("driver core: add device probe log helper")

Signed-off-by: Akhilesh Patil <akhilesh@ee.iitb.ac.in>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/aIJagJ/RnhSCtb2t@bhairav-test.ee.iitb.ac.in
---
 drivers/gpu/drm/panel/panel-orisetech-ota5601a.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-orisetech-ota5601a.c b/drivers/gpu/drm/panel/panel-orisetech-ota5601a.c
index 3231e84dc66c..8a608972fc41 100644
--- a/drivers/gpu/drm/panel/panel-orisetech-ota5601a.c
+++ b/drivers/gpu/drm/panel/panel-orisetech-ota5601a.c
@@ -276,11 +276,8 @@ static int ota5601a_probe(struct spi_device *spi)
 	}
 
 	err = drm_panel_of_backlight(&panel->drm_panel);
-	if (err) {
-		if (err != -EPROBE_DEFER)
-			dev_err(dev, "Failed to get backlight handle\n");
-		return err;
-	}
+	if (err)
+		return dev_err_probe(dev, err, "Failed to get backlight handle\n");
 
 	drm_panel_add(&panel->drm_panel);
 

From bd068333ef090d63a7cf952d70040e768b0aa4cf Mon Sep 17 00:00:00 2001
From: Svyatoslav Ryhel <clamor95@gmail.com>
Date: Thu, 17 Jul 2025 16:57:51 +0300
Subject: [PATCH 0211/1455] dt-bindings: display: panel: Document Hydis
 HV101HD1 DSI panel

Hydis HV101HD1 is a 2-lane 1366x768 MIPI DSI panel found in ASUS VivoTab RT
TF600T tablet.

Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250717135752.55958-2-clamor95@gmail.com
---
 .../display/panel/hydis,hv101hd1.yaml         | 60 +++++++++++++++++++
 1 file changed, 60 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/panel/hydis,hv101hd1.yaml

diff --git a/Documentation/devicetree/bindings/display/panel/hydis,hv101hd1.yaml b/Documentation/devicetree/bindings/display/panel/hydis,hv101hd1.yaml
new file mode 100644
index 000000000000..f429e84ee65d
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/hydis,hv101hd1.yaml
@@ -0,0 +1,60 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/panel/hydis,hv101hd1.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Hydis HV101HD1 DSI Display Panel
+
+maintainers:
+  - Svyatoslav Ryhel <clamor95@gmail.com>
+
+allOf:
+  - $ref: panel-common.yaml#
+
+properties:
+  compatible:
+    const: hydis,hv101hd1
+
+  reg:
+    maxItems: 1
+
+  vdd-supply: true
+  vio-supply: true
+
+  backlight: true
+  port: true
+
+required:
+  - compatible
+  - vdd-supply
+  - vio-supply
+  - backlight
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+
+    dsi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+
+        panel@0 {
+            compatible = "hydis,hv101hd1";
+            reg = <0>;
+
+            vdd-supply = <&vdd_lcd>;
+            vio-supply = <&vddio_lcd>;
+
+            backlight = <&backlight>;
+
+            port {
+                panel_in: endpoint {
+                    remote-endpoint = <&dsi_out>;
+                };
+            };
+        };
+    };
+...

From fdb4e289d2d6fd172591d802db71cef3134c1cf8 Mon Sep 17 00:00:00 2001
From: Svyatoslav Ryhel <clamor95@gmail.com>
Date: Thu, 17 Jul 2025 16:57:52 +0300
Subject: [PATCH 0212/1455] drm: panel: Add support for Hydis HV101HD1 MIPI DSI
 panel

HV101HD1-1E1 is a color active matrix TFT LCD module using amorphous
silicon TFT's (Thin Film Transistors) as an active switching devices. This
module has a 10.1 inch diagonally measured active area with HD resolutions
(1366 horizontal by 768 vertical pixel array).

Signed-off-by: Svyatoslav Ryhel <clamor95@gmail.com>
Reviewed-by: David Heidelberg <david@ixit.cz>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250717135752.55958-3-clamor95@gmail.com
---
 drivers/gpu/drm/panel/Kconfig                |  13 ++
 drivers/gpu/drm/panel/Makefile               |   1 +
 drivers/gpu/drm/panel/panel-hydis-hv101hd1.c | 188 +++++++++++++++++++
 3 files changed, 202 insertions(+)
 create mode 100644 drivers/gpu/drm/panel/panel-hydis-hv101hd1.c

diff --git a/drivers/gpu/drm/panel/Kconfig b/drivers/gpu/drm/panel/Kconfig
index e2a43c33ade1..407c5f6a268b 100644
--- a/drivers/gpu/drm/panel/Kconfig
+++ b/drivers/gpu/drm/panel/Kconfig
@@ -215,6 +215,19 @@ config DRM_PANEL_HIMAX_HX8394
 
 	  If M is selected the module will be called panel-himax-hx8394.
 
+config DRM_PANEL_HYDIS_HV101HD1
+	tristate "Hydis HV101HD1 panel"
+	depends on OF
+	depends on DRM_MIPI_DSI
+	depends on BACKLIGHT_CLASS_DEVICE
+	help
+	  Say Y here if you want to enable support for the Hydis HV101HD1
+	  2-lane 1366x768 MIPI DSI panel found in ASUS VivoTab RT TF600T.
+	  HV101HD1 is a color active matrix TFT LCD module using amorphous
+	  silicon TFT's (Thin Film Transistors) as an active switching devices.
+
+	  If M is selected the module will be called panel-hydis-hv101hd1
+
 config DRM_PANEL_ILITEK_IL9322
 	tristate "Ilitek ILI9322 320x240 QVGA panels"
 	depends on OF && SPI
diff --git a/drivers/gpu/drm/panel/Makefile b/drivers/gpu/drm/panel/Makefile
index c425eb2940b0..3615a761b44f 100644
--- a/drivers/gpu/drm/panel/Makefile
+++ b/drivers/gpu/drm/panel/Makefile
@@ -22,6 +22,7 @@ obj-$(CONFIG_DRM_PANEL_HIMAX_HX83102) += panel-himax-hx83102.o
 obj-$(CONFIG_DRM_PANEL_HIMAX_HX83112A) += panel-himax-hx83112a.o
 obj-$(CONFIG_DRM_PANEL_HIMAX_HX83112B) += panel-himax-hx83112b.o
 obj-$(CONFIG_DRM_PANEL_HIMAX_HX8394) += panel-himax-hx8394.o
+obj-$(CONFIG_DRM_PANEL_HYDIS_HV101HD1) += panel-hydis-hv101hd1.o
 obj-$(CONFIG_DRM_PANEL_ILITEK_IL9322) += panel-ilitek-ili9322.o
 obj-$(CONFIG_DRM_PANEL_ILITEK_ILI9341) += panel-ilitek-ili9341.o
 obj-$(CONFIG_DRM_PANEL_ILITEK_ILI9805) += panel-ilitek-ili9805.o
diff --git a/drivers/gpu/drm/panel/panel-hydis-hv101hd1.c b/drivers/gpu/drm/panel/panel-hydis-hv101hd1.c
new file mode 100644
index 000000000000..46426c388932
--- /dev/null
+++ b/drivers/gpu/drm/panel/panel-hydis-hv101hd1.c
@@ -0,0 +1,188 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/array_size.h>
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/gpio/consumer.h>
+#include <linux/mod_devicetable.h>
+#include <linux/module.h>
+#include <linux/property.h>
+#include <linux/regulator/consumer.h>
+
+#include <video/mipi_display.h>
+
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_modes.h>
+#include <drm/drm_panel.h>
+
+struct hv101hd1 {
+	struct drm_panel panel;
+	struct mipi_dsi_device *dsi;
+	struct regulator_bulk_data *supplies;
+};
+
+static const struct regulator_bulk_data hv101hd1_supplies[] = {
+	{ .supply = "vdd" },
+	{ .supply = "vio" },
+};
+
+static inline struct hv101hd1 *to_hv101hd1(struct drm_panel *panel)
+{
+	return container_of(panel, struct hv101hd1, panel);
+}
+
+static int hv101hd1_prepare(struct drm_panel *panel)
+{
+	struct hv101hd1 *hv = to_hv101hd1(panel);
+	struct mipi_dsi_multi_context ctx = { .dsi = hv->dsi };
+	struct device *dev = &hv->dsi->dev;
+	int ret;
+
+	ret = regulator_bulk_enable(ARRAY_SIZE(hv101hd1_supplies), hv->supplies);
+	if (ret) {
+		dev_err(dev, "error enabling regulators (%d)\n", ret);
+		return ret;
+	}
+
+	mipi_dsi_dcs_exit_sleep_mode_multi(&ctx);
+	mipi_dsi_msleep(&ctx, 20);
+
+	mipi_dsi_dcs_set_display_on_multi(&ctx);
+	mipi_dsi_msleep(&ctx, 20);
+
+	return 0;
+}
+
+static int hv101hd1_disable(struct drm_panel *panel)
+{
+	struct hv101hd1 *hv = to_hv101hd1(panel);
+	struct mipi_dsi_multi_context ctx = { .dsi = hv->dsi };
+
+	mipi_dsi_dcs_set_display_off_multi(&ctx);
+	mipi_dsi_msleep(&ctx, 120);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&ctx);
+	mipi_dsi_msleep(&ctx, 20);
+
+	return 0;
+}
+
+static int hv101hd1_unprepare(struct drm_panel *panel)
+{
+	struct hv101hd1 *hv = to_hv101hd1(panel);
+
+	return regulator_bulk_disable(ARRAY_SIZE(hv101hd1_supplies),
+				      hv->supplies);
+}
+
+static const struct drm_display_mode hv101hd1_mode = {
+	.clock = (1366 + 74 + 36 + 24) * (768 + 21 + 7 + 4) * 60 / 1000,
+	.hdisplay = 1366,
+	.hsync_start = 1366 + 74,
+	.hsync_end = 1366 + 74 + 36,
+	.htotal = 1366 + 74 + 36 + 24,
+	.vdisplay = 768,
+	.vsync_start = 768 + 21,
+	.vsync_end = 768 + 21 + 7,
+	.vtotal = 768 + 21 + 7 + 4,
+	.width_mm = 140,
+	.height_mm = 220,
+};
+
+static int hv101hd1_get_modes(struct drm_panel *panel, struct drm_connector *connector)
+{
+	struct drm_display_mode *mode;
+
+	mode = drm_mode_duplicate(connector->dev, &hv101hd1_mode);
+	if (!mode)
+		return -ENOMEM;
+
+	drm_mode_set_name(mode);
+
+	mode->type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED;
+
+	connector->display_info.width_mm = mode->width_mm;
+	connector->display_info.height_mm = mode->height_mm;
+
+	drm_mode_probed_add(connector, mode);
+
+	return 1;
+}
+
+static const struct drm_panel_funcs hv101hd1_panel_funcs = {
+	.prepare = hv101hd1_prepare,
+	.disable = hv101hd1_disable,
+	.unprepare = hv101hd1_unprepare,
+	.get_modes = hv101hd1_get_modes,
+};
+
+static int hv101hd1_probe(struct mipi_dsi_device *dsi)
+{
+	struct device *dev = &dsi->dev;
+	struct hv101hd1 *hv;
+	int ret;
+
+	hv = devm_drm_panel_alloc(dev, struct hv101hd1, panel,
+				  &hv101hd1_panel_funcs,
+				  DRM_MODE_CONNECTOR_DSI);
+	if (IS_ERR(hv))
+		return PTR_ERR(hv);
+
+	ret = devm_regulator_bulk_get_const(dev, ARRAY_SIZE(hv101hd1_supplies),
+					    hv101hd1_supplies, &hv->supplies);
+	if (ret)
+		return dev_err_probe(dev, ret, "failed to get regulators\n");
+
+	hv->dsi = dsi;
+	mipi_dsi_set_drvdata(dsi, hv);
+
+	dsi->lanes = 2;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_LPM;
+
+	ret = drm_panel_of_backlight(&hv->panel);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to get backlight\n");
+
+	drm_panel_add(&hv->panel);
+
+	ret = mipi_dsi_attach(dsi);
+	if (ret) {
+		drm_panel_remove(&hv->panel);
+		return dev_err_probe(dev, ret, "Failed to attach to DSI host\n");
+	}
+
+	return 0;
+}
+
+static void hv101hd1_remove(struct mipi_dsi_device *dsi)
+{
+	struct hv101hd1 *hv = mipi_dsi_get_drvdata(dsi);
+	int ret;
+
+	ret = mipi_dsi_detach(dsi);
+	if (ret < 0)
+		dev_err(&dsi->dev,
+			"Failed to detach from DSI host: %d\n", ret);
+
+	drm_panel_remove(&hv->panel);
+}
+
+static const struct of_device_id hv101hd1_of_match[] = {
+	{ .compatible = "hydis,hv101hd1" },
+	{ /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, hv101hd1_of_match);
+
+static struct mipi_dsi_driver hv101hd1_driver = {
+	.driver = {
+		.name = "panel-hv101hd1",
+		.of_match_table = hv101hd1_of_match,
+	},
+	.probe = hv101hd1_probe,
+	.remove = hv101hd1_remove,
+};
+module_mipi_dsi_driver(hv101hd1_driver);
+
+MODULE_AUTHOR("Svyatoslav Ryhel <clamor95@gmail.com>");
+MODULE_DESCRIPTION("DRM driver for Hydis HV101HD1 panel");
+MODULE_LICENSE("GPL");

From 125459e19ec654924e472f3ff5aeea40358dbebf Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Wed, 30 Jul 2025 21:23:41 -0600
Subject: [PATCH 0213/1455] drm/panel: novatek-nt35560: Fix invalid return
 value

Fix bug in nt35560_set_brightness() which causes the function to
erroneously report an error. mipi_dsi_dcs_write() returns either a
negative value when an error occurred or a positive number of bytes
written when no error occurred. The buggy code reports an error under
either condition.

Fixes: 8152c2bfd780 ("drm/panel: Add driver for Sony ACX424AKP panel")
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250731032343.1258366-2-me@brighamcampbell.com
---
 drivers/gpu/drm/panel/panel-novatek-nt35560.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35560.c b/drivers/gpu/drm/panel/panel-novatek-nt35560.c
index 98f0782c8411..17898a29efe8 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt35560.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt35560.c
@@ -161,7 +161,7 @@ static int nt35560_set_brightness(struct backlight_device *bl)
 		par = 0x00;
 		ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY,
 					 &par, 1);
-		if (ret) {
+		if (ret < 0) {
 			dev_err(nt->dev, "failed to disable display backlight (%d)\n", ret);
 			return ret;
 		}

From 05efa7f1d14febf0cd230c4950abfa6add1344a2 Mon Sep 17 00:00:00 2001
From: Chen Ni <nichen@iscas.ac.cn>
Date: Tue, 29 Jul 2025 13:42:14 +0800
Subject: [PATCH 0214/1455] drm/panel: himax-hx8279: Remove unneeded semicolon

Remove unnecessary semicolons reported by Coccinelle/coccicheck and the
semantic patch at scripts/coccinelle/misc/semicolon.cocci.

Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250729054214.2264377-1-nichen@iscas.ac.cn
---
 drivers/gpu/drm/panel/panel-himax-hx8279.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/panel-himax-hx8279.c b/drivers/gpu/drm/panel/panel-himax-hx8279.c
index fb302d1f91b9..9e443c719843 100644
--- a/drivers/gpu/drm/panel/panel-himax-hx8279.c
+++ b/drivers/gpu/drm/panel/panel-himax-hx8279.c
@@ -935,7 +935,7 @@ static int hx8279_check_dig_gamma(struct hx8279 *hx, struct device *dev, const u
 			j++;
 			x++;
 		} while (x < 4);
-	};
+	}
 
 	return 0;
 }

From cf9710a0a2ed490bd3c6933ecfc22556894a1df4 Mon Sep 17 00:00:00 2001
From: Hugo Villeneuve <hvilleneuve@dimonoff.com>
Date: Mon, 21 Jul 2025 11:28:15 -0400
Subject: [PATCH 0215/1455] drm/panel: sitronix-st7703: fix typo in comments

Fix typo in comments:
    souch -> such.

Signed-off-by: Hugo Villeneuve <hvilleneuve@dimonoff.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250721152818.1891212-1-hugo@hugovil.com
---
 drivers/gpu/drm/panel/panel-sitronix-st7703.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/panel-sitronix-st7703.c b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
index 1a007a244d84..6c348fe28955 100644
--- a/drivers/gpu/drm/panel/panel-sitronix-st7703.c
+++ b/drivers/gpu/drm/panel/panel-sitronix-st7703.c
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 /*
- * Driver for panels based on Sitronix ST7703 controller, souch as:
+ * Driver for panels based on Sitronix ST7703 controller, such as:
  *
  * - Rocktech jh057n00900 5.5" MIPI-DSI panel
  *

From bcddb12c027434fdf0491c1a05a3fe4fd2263d71 Mon Sep 17 00:00:00 2001
From: Tangudu Tilak Tirumalesh <tilak.tirumalesh.tangudu@intel.com>
Date: Thu, 31 Jul 2025 22:01:44 +0000
Subject: [PATCH 0216/1455] drm/xe: Extend wa_13012615864 to additional Xe2 and
 Xe3 platforms

Extend WA 13012615864 to Graphics Versions 20.01,20.02,20.04
and 30.03.

Signed-off-by: Tangudu Tilak Tirumalesh <tilak.tirumalesh.tangudu@intel.com>
Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250731220143.72942-2-jonathan.cavitt@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_wa.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index 22a98600fd8f..535067e7fb0c 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -538,6 +538,11 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	  XE_RTP_RULES(GRAPHICS_VERSION(2004), ENGINE_CLASS(RENDER)),
 	  XE_RTP_ACTIONS(SET(HALF_SLICE_CHICKEN7, CLEAR_OPTIMIZATION_DISABLE))
 	},
+	{ XE_RTP_NAME("13012615864"),
+	  XE_RTP_RULES(GRAPHICS_VERSION(2004),
+		       FUNC(xe_rtp_match_first_render_or_compute)),
+	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS))
+	},
 
 	/* Xe2_HPG */
 
@@ -602,6 +607,11 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 		       FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, STK_ID_RESTRICT))
 	},
+	{ XE_RTP_NAME("13012615864"),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(2001, 2002),
+		       FUNC(xe_rtp_match_first_render_or_compute)),
+	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS))
+	},
 
 	/* Xe2_LPM */
 
@@ -647,7 +657,8 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	  XE_RTP_ACTIONS(SET(TDL_CHICKEN, QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE))
 	},
 	{ XE_RTP_NAME("13012615864"),
-	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001), OR,
+		       GRAPHICS_VERSION(3003),
 		       FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS))
 	},

From 6716a823d18d422c356c2cd0c087f46d84c9e713 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Fri, 6 Jun 2025 14:13:37 +0200
Subject: [PATCH 0217/1455] drm/amdgpu: rework how PTE flags are generated v3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Previously we tried to keep the HW specific PTE flags in each mapping,
but for CRIU that isn't sufficient any more since the original value is
needed for the checkpoint procedure.

So rework the whole handling, nuke the early mapping function, keep the
UAPI flags in each mapping instead of the HW flags and translate them to
the HW flags while filling in the PTEs.

Only tested on Navi 23 for now, so probably needs quite a bit of more
work.

v2: fix KFD and SVN handling
v3: one more SVN fix pointed out by Felix
v4: squash in gfx12 fix from David

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  |  7 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c       | 37 +----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h       |  1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h       | 15 ++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h    |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c     |  8 +--
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c        | 17 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h        |  4 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c        | 57 +++++++++--------
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c        | 57 +++++++++--------
 drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c        | 47 +++++++-------
 drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c         |  4 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c         |  4 +-
 drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c         | 10 ++-
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c         | 61 +++++++++----------
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c          | 13 ++--
 16 files changed, 158 insertions(+), 186 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 260165bbe373..37d8a7034a7e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -494,7 +494,8 @@ static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync)
 	return amdgpu_sync_fence(sync, vm->last_update, GFP_KERNEL);
 }
 
-static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
+static uint64_t get_pte_flags(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			      struct kgd_mem *mem)
 {
 	uint32_t mapping_flags = AMDGPU_VM_PAGE_READABLE |
 				 AMDGPU_VM_MTYPE_DEFAULT;
@@ -504,7 +505,7 @@ static uint64_t get_pte_flags(struct amdgpu_device *adev, struct kgd_mem *mem)
 	if (mem->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_EXECUTABLE)
 		mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE;
 
-	return amdgpu_gem_va_map_flags(adev, mapping_flags);
+	return mapping_flags;
 }
 
 /**
@@ -961,7 +962,7 @@ static int kfd_mem_attach(struct amdgpu_device *adev, struct kgd_mem *mem,
 			goto unwind;
 		}
 		attachment[i]->va = va;
-		attachment[i]->pte_flags = get_pte_flags(adev, mem);
+		attachment[i]->pte_flags = get_pte_flags(adev, vm, mem);
 		attachment[i]->adev = adev;
 		list_add(&attachment[i]->list, &mem->attachments);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 6626a6e64ff5..d5e685c5e28b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -790,36 +790,6 @@ error:
 	return fence;
 }
 
-/**
- * amdgpu_gem_va_map_flags - map GEM UAPI flags into hardware flags
- *
- * @adev: amdgpu_device pointer
- * @flags: GEM UAPI flags
- *
- * Returns the GEM UAPI flags mapped into hardware for the ASIC.
- */
-uint64_t amdgpu_gem_va_map_flags(struct amdgpu_device *adev, uint32_t flags)
-{
-	uint64_t pte_flag = 0;
-
-	if (flags & AMDGPU_VM_PAGE_EXECUTABLE)
-		pte_flag |= AMDGPU_PTE_EXECUTABLE;
-	if (flags & AMDGPU_VM_PAGE_READABLE)
-		pte_flag |= AMDGPU_PTE_READABLE;
-	if (flags & AMDGPU_VM_PAGE_WRITEABLE)
-		pte_flag |= AMDGPU_PTE_WRITEABLE;
-	if (flags & AMDGPU_VM_PAGE_PRT)
-		pte_flag |= AMDGPU_PTE_PRT_FLAG(adev);
-	if (flags & AMDGPU_VM_PAGE_NOALLOC)
-		pte_flag |= AMDGPU_PTE_NOALLOC;
-
-	if (adev->gmc.gmc_funcs->map_mtype)
-		pte_flag |= amdgpu_gmc_map_mtype(adev,
-						 flags & AMDGPU_VM_MTYPE_MASK);
-
-	return pte_flag;
-}
-
 int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 			  struct drm_file *filp)
 {
@@ -840,7 +810,6 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	struct dma_fence_chain *timeline_chain = NULL;
 	struct dma_fence *fence;
 	struct drm_exec exec;
-	uint64_t va_flags;
 	uint64_t vm_size;
 	int r = 0;
 
@@ -944,10 +913,9 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 
 	switch (args->operation) {
 	case AMDGPU_VA_OP_MAP:
-		va_flags = amdgpu_gem_va_map_flags(adev, args->flags);
 		r = amdgpu_vm_bo_map(adev, bo_va, args->va_address,
 				     args->offset_in_bo, args->map_size,
-				     va_flags);
+				     args->flags);
 		break;
 	case AMDGPU_VA_OP_UNMAP:
 		r = amdgpu_vm_bo_unmap(adev, bo_va, args->va_address);
@@ -959,10 +927,9 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 						args->map_size);
 		break;
 	case AMDGPU_VA_OP_REPLACE:
-		va_flags = amdgpu_gem_va_map_flags(adev, args->flags);
 		r = amdgpu_vm_bo_replace_map(adev, bo_va, args->va_address,
 					     args->offset_in_bo, args->map_size,
-					     va_flags);
+					     args->flags);
 		break;
 	default:
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
index 3a8f57900a3a..b51e8f95ee86 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
@@ -63,7 +63,6 @@ int amdgpu_gem_mmap_ioctl(struct drm_device *dev, void *data,
 			  struct drm_file *filp);
 int amdgpu_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *filp);
-uint64_t amdgpu_gem_va_map_flags(struct amdgpu_device *adev, uint32_t flags);
 int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 			  struct drm_file *filp);
 int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 397c6ccdb903..55097ca10738 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -154,15 +154,15 @@ struct amdgpu_gmc_funcs {
 				   unsigned pasid);
 	/* enable/disable PRT support */
 	void (*set_prt)(struct amdgpu_device *adev, bool enable);
-	/* map mtype to hardware flags */
-	uint64_t (*map_mtype)(struct amdgpu_device *adev, uint32_t flags);
 	/* get the pde for a given mc addr */
 	void (*get_vm_pde)(struct amdgpu_device *adev, int level,
 			   u64 *dst, u64 *flags);
-	/* get the pte flags to use for a BO VA mapping */
+	/* get the pte flags to use for PTEs */
 	void (*get_vm_pte)(struct amdgpu_device *adev,
-			   struct amdgpu_bo_va_mapping *mapping,
-			   uint64_t *flags);
+			   struct amdgpu_vm *vm,
+			   struct amdgpu_bo *bo,
+			   uint32_t vm_flags,
+			   uint64_t *pte_flags);
 	/* override per-page pte flags */
 	void (*override_vm_pte_flags)(struct amdgpu_device *dev,
 				      struct amdgpu_vm *vm,
@@ -356,9 +356,10 @@ struct amdgpu_gmc {
 
 #define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr))
 #define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid))
-#define amdgpu_gmc_map_mtype(adev, flags) (adev)->gmc.gmc_funcs->map_mtype((adev),(flags))
 #define amdgpu_gmc_get_vm_pde(adev, level, dst, flags) (adev)->gmc.gmc_funcs->get_vm_pde((adev), (level), (dst), (flags))
-#define amdgpu_gmc_get_vm_pte(adev, mapping, flags) (adev)->gmc.gmc_funcs->get_vm_pte((adev), (mapping), (flags))
+#define amdgpu_gmc_get_vm_pte(adev, vm, bo, vm_flags, pte_flags) \
+	((adev)->gmc.gmc_funcs->get_vm_pte((adev), (vm), (bo), (vm_flags), \
+					   (pte_flags)))
 #define amdgpu_gmc_override_vm_pte_flags(adev, vm, addr, pte_flags)	\
 	(adev)->gmc.gmc_funcs->override_vm_pte_flags			\
 		((adev), (vm), (addr), (pte_flags))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index c316920f3450..87523fcd4386 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -69,7 +69,7 @@ struct amdgpu_bo_va_mapping {
 	uint64_t			last;
 	uint64_t			__subtree_last;
 	uint64_t			offset;
-	uint64_t			flags;
+	uint32_t			flags;
 };
 
 /* User space allocated BO in a VM */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c
index d45ebfb642ca..a0b479d5fff1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c
@@ -67,9 +67,9 @@ static inline u64 amdgpu_seq64_get_va_base(struct amdgpu_device *adev)
 int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		     struct amdgpu_bo_va **bo_va)
 {
-	u64 seq64_addr, va_flags;
 	struct amdgpu_bo *bo;
 	struct drm_exec exec;
+	u64 seq64_addr;
 	int r;
 
 	bo = adev->seq64.sbo;
@@ -94,9 +94,9 @@ int amdgpu_seq64_map(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 
 	seq64_addr = amdgpu_seq64_get_va_base(adev) & AMDGPU_GMC_HOLE_MASK;
 
-	va_flags = amdgpu_gem_va_map_flags(adev, AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_MTYPE_UC);
-	r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0, AMDGPU_VA_RESERVED_SEQ64_SIZE,
-			     va_flags);
+	r = amdgpu_vm_bo_map(adev, *bo_va, seq64_addr, 0,
+			     AMDGPU_VA_RESERVED_SEQ64_SIZE,
+			     AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_MTYPE_UC);
 	if (r) {
 		DRM_ERROR("failed to do bo_map on userq sem, err=%d\n", r);
 		amdgpu_vm_bo_del(adev, *bo_va);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 5cacf5717016..39b4250ede0f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1332,13 +1332,14 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
 		/* normally,bo_va->flags only contians READABLE and WIRTEABLE bit go here
 		 * but in case of something, we filter the flags in first place
 		 */
-		if (!(mapping->flags & AMDGPU_PTE_READABLE))
+		if (!(mapping->flags & AMDGPU_VM_PAGE_READABLE))
 			update_flags &= ~AMDGPU_PTE_READABLE;
-		if (!(mapping->flags & AMDGPU_PTE_WRITEABLE))
+		if (!(mapping->flags & AMDGPU_VM_PAGE_WRITEABLE))
 			update_flags &= ~AMDGPU_PTE_WRITEABLE;
 
 		/* Apply ASIC specific mapping flags */
-		amdgpu_gmc_get_vm_pte(adev, mapping, &update_flags);
+		amdgpu_gmc_get_vm_pte(adev, vm, bo, mapping->flags,
+				      &update_flags);
 
 		trace_amdgpu_vm_bo_update(mapping);
 
@@ -1479,7 +1480,7 @@ static void amdgpu_vm_free_mapping(struct amdgpu_device *adev,
 				   struct amdgpu_bo_va_mapping *mapping,
 				   struct dma_fence *fence)
 {
-	if (mapping->flags & AMDGPU_PTE_PRT_FLAG(adev))
+	if (mapping->flags & AMDGPU_VM_PAGE_PRT)
 		amdgpu_vm_add_prt_cb(adev, fence);
 	kfree(mapping);
 }
@@ -1758,7 +1759,7 @@ static void amdgpu_vm_bo_insert_map(struct amdgpu_device *adev,
 	list_add(&mapping->list, &bo_va->invalids);
 	amdgpu_vm_it_insert(mapping, &vm->va);
 
-	if (mapping->flags & AMDGPU_PTE_PRT_FLAG(adev))
+	if (mapping->flags & AMDGPU_VM_PAGE_PRT)
 		amdgpu_vm_prt_get(adev);
 
 	if (amdgpu_vm_is_bo_always_valid(vm, bo) && !bo_va->base.moved)
@@ -1818,7 +1819,7 @@ static int amdgpu_vm_verify_parameters(struct amdgpu_device *adev,
 int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		     struct amdgpu_bo_va *bo_va,
 		     uint64_t saddr, uint64_t offset,
-		     uint64_t size, uint64_t flags)
+		     uint64_t size, uint32_t flags)
 {
 	struct amdgpu_bo_va_mapping *mapping, *tmp;
 	struct amdgpu_bo *bo = bo_va->base.bo;
@@ -1877,7 +1878,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 			     struct amdgpu_bo_va *bo_va,
 			     uint64_t saddr, uint64_t offset,
-			     uint64_t size, uint64_t flags)
+			     uint64_t size, uint32_t flags)
 {
 	struct amdgpu_bo_va_mapping *mapping;
 	struct amdgpu_bo *bo = bo_va->base.bo;
@@ -2734,7 +2735,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	dma_fence_put(vm->last_tlb_flush);
 
 	list_for_each_entry_safe(mapping, tmp, &vm->freed, list) {
-		if (mapping->flags & AMDGPU_PTE_PRT_FLAG(adev) && prt_fini_needed) {
+		if (mapping->flags & AMDGPU_VM_PAGE_PRT && prt_fini_needed) {
 			amdgpu_vm_prt_fini(adev, vm);
 			prt_fini_needed = false;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index fd086efd8457..3b4fa3246675 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -538,11 +538,11 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		     struct amdgpu_bo_va *bo_va,
 		     uint64_t addr, uint64_t offset,
-		     uint64_t size, uint64_t flags);
+		     uint64_t size, uint32_t flags);
 int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev,
 			     struct amdgpu_bo_va *bo_va,
 			     uint64_t addr, uint64_t offset,
-			     uint64_t size, uint64_t flags);
+			     uint64_t size, uint32_t flags);
 int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 		       struct amdgpu_bo_va *bo_va,
 		       uint64_t addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 7923f491cf73..7031dd8c3c5e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -466,24 +466,6 @@ static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int
  * 0 valid
  */
 
-static uint64_t gmc_v10_0_map_mtype(struct amdgpu_device *adev, uint32_t flags)
-{
-	switch (flags) {
-	case AMDGPU_VM_MTYPE_DEFAULT:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_NC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_WC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_WC);
-	case AMDGPU_VM_MTYPE_CC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_CC);
-	case AMDGPU_VM_MTYPE_UC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_UC);
-	default:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC);
-	}
-}
-
 static void gmc_v10_0_get_vm_pde(struct amdgpu_device *adev, int level,
 				 uint64_t *addr, uint64_t *flags)
 {
@@ -508,21 +490,39 @@ static void gmc_v10_0_get_vm_pde(struct amdgpu_device *adev, int level,
 }
 
 static void gmc_v10_0_get_vm_pte(struct amdgpu_device *adev,
-				 struct amdgpu_bo_va_mapping *mapping,
+				 struct amdgpu_vm *vm,
+				 struct amdgpu_bo *bo,
+				 uint32_t vm_flags,
 				 uint64_t *flags)
 {
-	struct amdgpu_bo *bo = mapping->bo_va->base.bo;
+	if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE)
+		*flags |= AMDGPU_PTE_EXECUTABLE;
+	else
+		*flags &= ~AMDGPU_PTE_EXECUTABLE;
 
-	*flags &= ~AMDGPU_PTE_EXECUTABLE;
-	*flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;
+	switch (vm_flags & AMDGPU_VM_MTYPE_MASK) {
+	case AMDGPU_VM_MTYPE_DEFAULT:
+	case AMDGPU_VM_MTYPE_NC:
+	default:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_NC);
+		break;
+	case AMDGPU_VM_MTYPE_WC:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_WC);
+		break;
+	case AMDGPU_VM_MTYPE_CC:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_CC);
+		break;
+	case AMDGPU_VM_MTYPE_UC:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_UC);
+		break;
+	}
 
-	*flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK;
-	*flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK);
+	if (vm_flags & AMDGPU_VM_PAGE_NOALLOC)
+		*flags |= AMDGPU_PTE_NOALLOC;
+	else
+		*flags &= ~AMDGPU_PTE_NOALLOC;
 
-	*flags &= ~AMDGPU_PTE_NOALLOC;
-	*flags |= (mapping->flags & AMDGPU_PTE_NOALLOC);
-
-	if (mapping->flags & AMDGPU_PTE_PRT) {
+	if (vm_flags & AMDGPU_VM_PAGE_PRT) {
 		*flags |= AMDGPU_PTE_PRT;
 		*flags |= AMDGPU_PTE_SNOOPED;
 		*flags |= AMDGPU_PTE_LOG;
@@ -563,7 +563,6 @@ static const struct amdgpu_gmc_funcs gmc_v10_0_gmc_funcs = {
 	.flush_gpu_tlb_pasid = gmc_v10_0_flush_gpu_tlb_pasid,
 	.emit_flush_gpu_tlb = gmc_v10_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v10_0_emit_pasid_mapping,
-	.map_mtype = gmc_v10_0_map_mtype,
 	.get_vm_pde = gmc_v10_0_get_vm_pde,
 	.get_vm_pte = gmc_v10_0_get_vm_pte,
 	.get_vbios_fb_size = gmc_v10_0_get_vbios_fb_size,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index f15d691e9a20..93d2b0bbe641 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -430,24 +430,6 @@ static void gmc_v11_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int
  * 0 valid
  */
 
-static uint64_t gmc_v11_0_map_mtype(struct amdgpu_device *adev, uint32_t flags)
-{
-	switch (flags) {
-	case AMDGPU_VM_MTYPE_DEFAULT:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_NC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_WC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_WC);
-	case AMDGPU_VM_MTYPE_CC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_CC);
-	case AMDGPU_VM_MTYPE_UC:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_UC);
-	default:
-		return AMDGPU_PTE_MTYPE_NV10(0ULL, MTYPE_NC);
-	}
-}
-
 static void gmc_v11_0_get_vm_pde(struct amdgpu_device *adev, int level,
 				 uint64_t *addr, uint64_t *flags)
 {
@@ -472,21 +454,39 @@ static void gmc_v11_0_get_vm_pde(struct amdgpu_device *adev, int level,
 }
 
 static void gmc_v11_0_get_vm_pte(struct amdgpu_device *adev,
-				 struct amdgpu_bo_va_mapping *mapping,
+				 struct amdgpu_vm *vm,
+				 struct amdgpu_bo *bo,
+				 uint32_t vm_flags,
 				 uint64_t *flags)
 {
-	struct amdgpu_bo *bo = mapping->bo_va->base.bo;
+	if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE)
+		*flags |= AMDGPU_PTE_EXECUTABLE;
+	else
+		*flags &= ~AMDGPU_PTE_EXECUTABLE;
 
-	*flags &= ~AMDGPU_PTE_EXECUTABLE;
-	*flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;
+	switch (vm_flags & AMDGPU_VM_MTYPE_MASK) {
+	case AMDGPU_VM_MTYPE_DEFAULT:
+	case AMDGPU_VM_MTYPE_NC:
+	default:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_NC);
+		break;
+	case AMDGPU_VM_MTYPE_WC:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_WC);
+		break;
+	case AMDGPU_VM_MTYPE_CC:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_CC);
+		break;
+	case AMDGPU_VM_MTYPE_UC:
+		*flags = AMDGPU_PTE_MTYPE_NV10(*flags, MTYPE_UC);
+		break;
+	}
 
-	*flags &= ~AMDGPU_PTE_MTYPE_NV10_MASK;
-	*flags |= (mapping->flags & AMDGPU_PTE_MTYPE_NV10_MASK);
+	if (vm_flags & AMDGPU_VM_PAGE_NOALLOC)
+		*flags |= AMDGPU_PTE_NOALLOC;
+	else
+		*flags &= ~AMDGPU_PTE_NOALLOC;
 
-	*flags &= ~AMDGPU_PTE_NOALLOC;
-	*flags |= (mapping->flags & AMDGPU_PTE_NOALLOC);
-
-	if (mapping->flags & AMDGPU_PTE_PRT) {
+	if (vm_flags & AMDGPU_VM_PAGE_PRT) {
 		*flags |= AMDGPU_PTE_PRT;
 		*flags |= AMDGPU_PTE_SNOOPED;
 		*flags |= AMDGPU_PTE_LOG;
@@ -527,7 +527,6 @@ static const struct amdgpu_gmc_funcs gmc_v11_0_gmc_funcs = {
 	.flush_gpu_tlb_pasid = gmc_v11_0_flush_gpu_tlb_pasid,
 	.emit_flush_gpu_tlb = gmc_v11_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v11_0_emit_pasid_mapping,
-	.map_mtype = gmc_v11_0_map_mtype,
 	.get_vm_pde = gmc_v11_0_get_vm_pde,
 	.get_vm_pte = gmc_v11_0_get_vm_pte,
 	.get_vbios_fb_size = gmc_v11_0_get_vbios_fb_size,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
index de763105fdfd..a0a5367f9dc4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
@@ -453,20 +453,6 @@ static void gmc_v12_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid
  * 0 valid
  */
 
-static uint64_t gmc_v12_0_map_mtype(struct amdgpu_device *adev, uint32_t flags)
-{
-	switch (flags) {
-	case AMDGPU_VM_MTYPE_DEFAULT:
-		return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_NC:
-		return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_UC:
-		return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_UC);
-	default:
-		return AMDGPU_PTE_MTYPE_GFX12(0ULL, MTYPE_NC);
-	}
-}
-
 static void gmc_v12_0_get_vm_pde(struct amdgpu_device *adev, int level,
 				 uint64_t *addr, uint64_t *flags)
 {
@@ -490,19 +476,35 @@ static void gmc_v12_0_get_vm_pde(struct amdgpu_device *adev, int level,
 }
 
 static void gmc_v12_0_get_vm_pte(struct amdgpu_device *adev,
-				 struct amdgpu_bo_va_mapping *mapping,
+				 struct amdgpu_vm *vm,
+				 struct amdgpu_bo *bo,
+				 uint32_t vm_flags,
 				 uint64_t *flags)
 {
-	struct amdgpu_bo *bo = mapping->bo_va->base.bo;
+	if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE)
+		*flags |= AMDGPU_PTE_EXECUTABLE;
+	else
+		*flags &= ~AMDGPU_PTE_EXECUTABLE;
 
-	*flags &= ~AMDGPU_PTE_EXECUTABLE;
-	*flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;
+	switch (vm_flags & AMDGPU_VM_MTYPE_MASK) {
+	case AMDGPU_VM_MTYPE_DEFAULT:
+		*flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_NC);
+		break;
+	case AMDGPU_VM_MTYPE_NC:
+	default:
+		*flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_NC);
+		break;
+	case AMDGPU_VM_MTYPE_UC:
+		*flags = AMDGPU_PTE_MTYPE_GFX12(*flags, MTYPE_UC);
+		break;
+	}
 
-	*flags &= ~AMDGPU_PTE_MTYPE_GFX12_MASK;
-	*flags |= (mapping->flags & AMDGPU_PTE_MTYPE_GFX12_MASK);
+	if (vm_flags & AMDGPU_VM_PAGE_NOALLOC)
+		*flags |= AMDGPU_PTE_NOALLOC;
+	else
+		*flags &= ~AMDGPU_PTE_NOALLOC;
 
-	if (mapping->flags & AMDGPU_PTE_PRT_GFX12) {
-		*flags |= AMDGPU_PTE_PRT_GFX12;
+	if (vm_flags & AMDGPU_VM_PAGE_PRT) {
 		*flags |= AMDGPU_PTE_SNOOPED;
 		*flags |= AMDGPU_PTE_SYSTEM;
 		*flags |= AMDGPU_PTE_IS_PTE;
@@ -543,7 +545,6 @@ static const struct amdgpu_gmc_funcs gmc_v12_0_gmc_funcs = {
 	.flush_gpu_tlb_pasid = gmc_v12_0_flush_gpu_tlb_pasid,
 	.emit_flush_gpu_tlb = gmc_v12_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v12_0_emit_pasid_mapping,
-	.map_mtype = gmc_v12_0_map_mtype,
 	.get_vm_pde = gmc_v12_0_get_vm_pde,
 	.get_vm_pte = gmc_v12_0_get_vm_pte,
 	.get_vbios_fb_size = gmc_v12_0_get_vbios_fb_size,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index 8030fcd64210..f6ad7911f1e6 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -382,7 +382,9 @@ static void gmc_v6_0_get_vm_pde(struct amdgpu_device *adev, int level,
 }
 
 static void gmc_v6_0_get_vm_pte(struct amdgpu_device *adev,
-				struct amdgpu_bo_va_mapping *mapping,
+				struct amdgpu_vm *vm,
+				struct amdgpu_bo *bo,
+				uint32_t vm_flags,
 				uint64_t *flags)
 {
 	*flags &= ~AMDGPU_PTE_EXECUTABLE;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index a8d5795084fc..93d7ccb7d013 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -504,7 +504,9 @@ static void gmc_v7_0_get_vm_pde(struct amdgpu_device *adev, int level,
 }
 
 static void gmc_v7_0_get_vm_pte(struct amdgpu_device *adev,
-				struct amdgpu_bo_va_mapping *mapping,
+				struct amdgpu_vm *vm,
+				struct amdgpu_bo *bo,
+				uint32_t vm_flags,
 				uint64_t *flags)
 {
 	*flags &= ~AMDGPU_PTE_EXECUTABLE;
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index b45fa0cea9d2..c5e2a2c41e06 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -716,11 +716,15 @@ static void gmc_v8_0_get_vm_pde(struct amdgpu_device *adev, int level,
 }
 
 static void gmc_v8_0_get_vm_pte(struct amdgpu_device *adev,
-				struct amdgpu_bo_va_mapping *mapping,
+				struct amdgpu_vm *vm,
+				struct amdgpu_bo *bo,
+				uint32_t vm_flags,
 				uint64_t *flags)
 {
-	*flags &= ~AMDGPU_PTE_EXECUTABLE;
-	*flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;
+	if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE)
+		*flags |= AMDGPU_PTE_EXECUTABLE;
+	else
+		*flags &= ~AMDGPU_PTE_EXECUTABLE;
 	*flags &= ~AMDGPU_PTE_PRT;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index c4d69cf4e06c..8404695eb13f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1073,27 +1073,6 @@ static void gmc_v9_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned int v
  * 0 valid
  */
 
-static uint64_t gmc_v9_0_map_mtype(struct amdgpu_device *adev, uint32_t flags)
-
-{
-	switch (flags) {
-	case AMDGPU_VM_MTYPE_DEFAULT:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_NC:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_NC);
-	case AMDGPU_VM_MTYPE_WC:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_WC);
-	case AMDGPU_VM_MTYPE_RW:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_RW);
-	case AMDGPU_VM_MTYPE_CC:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_CC);
-	case AMDGPU_VM_MTYPE_UC:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_UC);
-	default:
-		return AMDGPU_PTE_MTYPE_VG10(0ULL, MTYPE_NC);
-	}
-}
-
 static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level,
 				uint64_t *addr, uint64_t *flags)
 {
@@ -1123,6 +1102,7 @@ static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level,
 static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 					 struct amdgpu_vm *vm,
 					 struct amdgpu_bo *bo,
+					 uint32_t vm_flags,
 					 uint64_t *flags)
 {
 	struct amdgpu_device *bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
@@ -1236,25 +1216,43 @@ static void gmc_v9_0_get_coherence_flags(struct amdgpu_device *adev,
 }
 
 static void gmc_v9_0_get_vm_pte(struct amdgpu_device *adev,
-				struct amdgpu_bo_va_mapping *mapping,
+				struct amdgpu_vm *vm,
+				struct amdgpu_bo *bo,
+				uint32_t vm_flags,
 				uint64_t *flags)
 {
-	struct amdgpu_bo *bo = mapping->bo_va->base.bo;
+	if (vm_flags & AMDGPU_VM_PAGE_EXECUTABLE)
+		*flags |= AMDGPU_PTE_EXECUTABLE;
+	else
+		*flags &= ~AMDGPU_PTE_EXECUTABLE;
 
-	*flags &= ~AMDGPU_PTE_EXECUTABLE;
-	*flags |= mapping->flags & AMDGPU_PTE_EXECUTABLE;
+	switch (vm_flags & AMDGPU_VM_MTYPE_MASK) {
+	case AMDGPU_VM_MTYPE_DEFAULT:
+	case AMDGPU_VM_MTYPE_NC:
+	default:
+		*flags = AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_NC);
+		break;
+	case AMDGPU_VM_MTYPE_WC:
+		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_WC);
+		break;
+	case AMDGPU_VM_MTYPE_RW:
+		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_RW);
+		break;
+	case AMDGPU_VM_MTYPE_CC:
+		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_CC);
+		break;
+	case AMDGPU_VM_MTYPE_UC:
+		*flags |= AMDGPU_PTE_MTYPE_VG10(*flags, MTYPE_UC);
+		break;
+	}
 
-	*flags &= ~AMDGPU_PTE_MTYPE_VG10_MASK;
-	*flags |= mapping->flags & AMDGPU_PTE_MTYPE_VG10_MASK;
-
-	if (mapping->flags & AMDGPU_PTE_PRT) {
+	if (vm_flags & AMDGPU_VM_PAGE_PRT) {
 		*flags |= AMDGPU_PTE_PRT;
 		*flags &= ~AMDGPU_PTE_VALID;
 	}
 
 	if ((*flags & AMDGPU_PTE_VALID) && bo)
-		gmc_v9_0_get_coherence_flags(adev, mapping->bo_va->base.vm, bo,
-					     flags);
+		gmc_v9_0_get_coherence_flags(adev, vm, bo, vm_flags, flags);
 }
 
 static void gmc_v9_0_override_vm_pte_flags(struct amdgpu_device *adev,
@@ -1391,7 +1389,6 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
 	.flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid,
 	.emit_flush_gpu_tlb = gmc_v9_0_emit_flush_gpu_tlb,
 	.emit_pasid_mapping = gmc_v9_0_emit_pasid_mapping,
-	.map_mtype = gmc_v9_0_map_mtype,
 	.get_vm_pde = gmc_v9_0_get_vm_pde,
 	.get_vm_pte = gmc_v9_0_get_vm_pte,
 	.override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index a0f22ea6d15a..e23b5a0f31f2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1189,7 +1189,7 @@ svm_nodes_in_same_hive(struct kfd_node *node_a, struct kfd_node *node_b)
 }
 
 static uint64_t
-svm_range_get_pte_flags(struct kfd_node *node,
+svm_range_get_pte_flags(struct kfd_node *node, struct amdgpu_vm *vm,
 			struct svm_range *prange, int domain)
 {
 	struct kfd_node *bo_node;
@@ -1292,10 +1292,6 @@ svm_range_get_pte_flags(struct kfd_node *node,
 			AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC;
 	}
 
-	mapping_flags |= AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE;
-
-	if (flags & KFD_IOCTL_SVM_FLAG_GPU_RO)
-		mapping_flags &= ~AMDGPU_VM_PAGE_WRITEABLE;
 	if (flags & KFD_IOCTL_SVM_FLAG_GPU_EXEC)
 		mapping_flags |= AMDGPU_VM_PAGE_EXECUTABLE;
 
@@ -1305,7 +1301,10 @@ svm_range_get_pte_flags(struct kfd_node *node,
 	if (gc_ip_version >= IP_VERSION(12, 0, 0))
 		pte_flags |= AMDGPU_PTE_IS_PTE;
 
-	pte_flags |= amdgpu_gem_va_map_flags(node->adev, mapping_flags);
+	amdgpu_gmc_get_vm_pte(node->adev, vm, NULL, mapping_flags, &pte_flags);
+	pte_flags |= AMDGPU_PTE_READABLE;
+	if (!(flags & KFD_IOCTL_SVM_FLAG_GPU_RO))
+		pte_flags |= AMDGPU_PTE_WRITEABLE;
 	return pte_flags;
 }
 
@@ -1412,7 +1411,7 @@ svm_range_map_to_gpu(struct kfd_process_device *pdd, struct svm_range *prange,
 		pr_debug("Mapping range [0x%lx 0x%llx] on domain: %s\n",
 			 last_start, prange->start + i, last_domain ? "GPU" : "CPU");
 
-		pte_flags = svm_range_get_pte_flags(pdd->dev, prange, last_domain);
+		pte_flags = svm_range_get_pte_flags(pdd->dev, vm, prange, last_domain);
 		if (readonly)
 			pte_flags &= ~AMDGPU_PTE_WRITEABLE;
 

From 8e3967a71e6fca9c871f98b9289b59c82b88b729 Mon Sep 17 00:00:00 2001
From: Perry Yuan <perry.yuan@amd.com>
Date: Mon, 7 Jul 2025 10:45:28 +0800
Subject: [PATCH 0218/1455] drm/amdgpu: Fix build error when CONFIG_SUSPEND is
 disabled

The variable `pm_suspend_target_state` is conditionally defined only when
`CONFIG_SUSPEND` is enabled (see `include/linux/suspend.h`). Directly
referencing it without guarding by `#ifdef CONFIG_SUSPEND` causes build
failures when suspend functionality is disabled (e.g., `CONFIG_SUSPEND=n`).

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Perry Yuan <perry.yuan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index efb54c23ba17..8e8b7d2c3f42 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2597,6 +2597,7 @@ static int amdgpu_pmops_suspend(struct device *dev)
 	else if (amdgpu_acpi_is_s3_active(adev))
 		adev->in_s3 = true;
 	if (!adev->in_s0ix && !adev->in_s3) {
+#if IS_ENABLED(CONFIG_SUSPEND)
 		/* don't allow going deep first time followed by s2idle the next time */
 		if (adev->last_suspend_state != PM_SUSPEND_ON &&
 		    adev->last_suspend_state != pm_suspend_target_state) {
@@ -2604,11 +2605,14 @@ static int amdgpu_pmops_suspend(struct device *dev)
 				     pm_suspend_target_state);
 			return -EINVAL;
 		}
+#endif
 		return 0;
 	}
 
+#if IS_ENABLED(CONFIG_SUSPEND)
 	/* cache the state last used for suspend */
 	adev->last_suspend_state = pm_suspend_target_state;
+#endif
 
 	return amdgpu_device_suspend(drm_dev, true);
 }

From 26a609e053a6fc494403e95403bc6a2470383bec Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 24 Jul 2025 22:12:21 -0500
Subject: [PATCH 0219/1455] drm/amd: Restore cached power limit during resume

The power limit will be cached in smu->current_power_limit but
if the ASIC goes into S3 this value won't be restored.

Restore the value during SMU resume.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250725031222.3015095-2-superm1@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 756afe78a6e5..310f51ff05b9 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -2226,6 +2226,12 @@ static int smu_resume(struct amdgpu_ip_block *ip_block)
 
 	adev->pm.dpm_enabled = true;
 
+	if (smu->current_power_limit) {
+		ret = smu_set_power_limit(smu, smu->current_power_limit);
+		if (ret && ret != -EOPNOTSUPP)
+			return ret;
+	}
+
 	dev_info(adev->dev, "SMU is resumed successfully!\n");
 
 	return 0;

From 4e9526924d09057a9ba854305e17eded900ced82 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 24 Jul 2025 22:12:22 -0500
Subject: [PATCH 0220/1455] drm/amd: Restore cached manual clock settings
 during resume

If the SCLK limits have been set before S3 they will not
be restored. The limits are however cached in the driver and so
they can be restored by running a commit sequence during resume.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250725031222.3015095-3-superm1@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 310f51ff05b9..b47cb4a5f488 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -77,6 +77,9 @@ static void smu_power_profile_mode_get(struct smu_context *smu,
 static void smu_power_profile_mode_put(struct smu_context *smu,
 				       enum PP_SMC_POWER_PROFILE profile_mode);
 static enum smu_clk_type smu_convert_to_smuclk(enum pp_clock_type type);
+static int smu_od_edit_dpm_table(void *handle,
+				 enum PP_OD_DPM_TABLE_COMMAND type,
+				 long *input, uint32_t size);
 
 static int smu_sys_get_pp_feature_mask(void *handle,
 				       char *buf)
@@ -2195,6 +2198,7 @@ static int smu_resume(struct amdgpu_ip_block *ip_block)
 	int ret;
 	struct amdgpu_device *adev = ip_block->adev;
 	struct smu_context *smu = adev->powerplay.pp_handle;
+	struct smu_dpm_context *smu_dpm_ctx = &(smu->smu_dpm);
 
 	if (amdgpu_sriov_multi_vf_mode(adev))
 		return 0;
@@ -2232,6 +2236,12 @@ static int smu_resume(struct amdgpu_ip_block *ip_block)
 			return ret;
 	}
 
+	if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) {
+		ret = smu_od_edit_dpm_table(smu, PP_OD_COMMIT_DPM_TABLE, NULL, 0);
+		if (ret)
+			return ret;
+	}
+
 	dev_info(adev->dev, "SMU is resumed successfully!\n");
 
 	return 0;

From 810a8809ccc69a67af74f3bd63f4d99da08049e7 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 28 Jul 2025 11:15:40 -0400
Subject: [PATCH 0221/1455] Documentation: update APU and dGPU tables with
 MP0/1 info

Add MP1 for APUs and MP0 and MP1 details for dGPUs.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3905
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/amdgpu/amd-hardware-list-info.rst     |  4 +-
 .../gpu/amdgpu/apu-asic-info-table.csv        | 34 +++++------
 .../gpu/amdgpu/dgpu-asic-info-table.csv       | 56 +++++++++----------
 3 files changed, 47 insertions(+), 47 deletions(-)

diff --git a/Documentation/gpu/amdgpu/amd-hardware-list-info.rst b/Documentation/gpu/amdgpu/amd-hardware-list-info.rst
index 1786544fe7c1..e72f4ff770c4 100644
--- a/Documentation/gpu/amdgpu/amd-hardware-list-info.rst
+++ b/Documentation/gpu/amdgpu/amd-hardware-list-info.rst
@@ -10,7 +10,7 @@ Accelerated Processing Units (APU) Info
 
 .. csv-table::
    :header-rows: 1
-   :widths: 3, 2, 2, 1, 1, 1, 1
+   :widths: 3, 2, 2, 1, 1, 1, 1, 1
    :file: ./apu-asic-info-table.csv
 
 Discrete GPU Info
@@ -18,6 +18,6 @@ Discrete GPU Info
 
 .. csv-table::
    :header-rows: 1
-   :widths: 3, 2, 2, 1, 1, 1
+   :widths: 3, 2, 2, 1, 1, 1, 1, 1
    :file: ./dgpu-asic-info-table.csv
 
diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
index 1d50b539677f..b479c5629146 100644
--- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
@@ -1,17 +1,17 @@
-Product Name, Code Reference, DCN/DCE version, GC version, VCE/UVD/VCN version, SDMA version, MP0 version
-Radeon R* Graphics, CARRIZO/STONEY, DCE 11, 8, VCE 3 / UVD 6, 3, n/a
-Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0, 10.0.0
-Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2, 11.0.3
-Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1, 10.0.1
-SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0
-Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1
-Ryzen 6000 series / Ryzen 7x35 series / Ryzen 7x36 series, YELLOW CARP / Rembrandt / Rembrandt-R, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3
-Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
-Ryzen 9000 series (AM5), Granite Ridge, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
-Ryzen 7x45 series (FL1), Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5
-Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8
-Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11
-Ryzen 8x40 series, Hawk Point, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11
-Ryzen AI 300 series, Strix Point, 3.5.0, 11.5.0, 4.0.5, 6.1.0, 14.0.0
-Ryzen AI 350 series, Krackan Point, 3.5.0, 11.5.2, 4.0.5, 6.1.2, 14.0.4
-Ryzen AI Max 300 series, Strix Halo, 3.5.1, 11.5.1, 4.0.6, 6.1.1, 14.0.1
+Product Name, Code Reference, DCN/DCE version, GC version, VCE/UVD/VCN version, SDMA version, MP0 version, MP1 version
+Radeon R* Graphics, CARRIZO/STONEY, DCE 11, 8, VCE 3 / UVD 6, 3, n/a, 8
+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN/PICASSO, DCN 1.0, 9.1.0, VCN 1.0, 4.1.0, 10.0.0, 10.0.0
+Ryzen 4000 series, RENOIR, DCN 2.1, 9.3, VCN 2.2, 4.1.2, 11.0.3, 12.0.1
+Ryzen 3000 series / AMD Ryzen Embedded V1*/R1* with Radeon Vega Gfx, RAVEN2, DCN 1.0, 9.2.2, VCN 1.0.1, 4.1.1, 10.0.1, 10.0.1
+SteamDeck, VANGOGH, DCN 3.0.1, 10.3.1, VCN 3.1.0, 5.2.1, 11.5.0, 11.5.0
+Ryzen 5000 series / Ryzen 7x30 series, GREEN SARDINE / Cezanne / Barcelo / Barcelo-R, DCN 2.1, 9.3, VCN 2.2, 4.1.1, 12.0.1, 12.0.1
+Ryzen 6000 series / Ryzen 7x35 series / Ryzen 7x36 series, YELLOW CARP / Rembrandt / Rembrandt-R, 3.1.2, 10.3.3, VCN 3.1.1, 5.2.3, 13.0.3, 13.0.3
+Ryzen 7000 series (AM5), Raphael, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5, 13.0.5
+Ryzen 9000 series (AM5), Granite Ridge, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5, 13.0.5
+Ryzen 7x45 series (FL1), Dragon Range, 3.1.5, 10.3.6, 3.1.2, 5.2.6, 13.0.5, 13.0.5
+Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8, 13.0.8
+Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11, 13.0.4 / 13.0.11
+Ryzen 8x40 series, Hawk Point, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11, 13.0.4 / 13.0.11
+Ryzen AI 300 series, Strix Point, 3.5.0, 11.5.0, 4.0.5, 6.1.0, 14.0.0, 14.0.0
+Ryzen AI 350 series, Krackan Point, 3.5.0, 11.5.2, 4.0.5, 6.1.2, 14.0.4, 14.0.4
+Ryzen AI Max 300 series, Strix Halo, 3.5.1, 11.5.1, 4.0.6, 6.1.1, 14.0.1, 14.0.1
diff --git a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
index d2f10ee69dfc..0f822a252fcb 100644
--- a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
@@ -1,28 +1,28 @@
-Product Name, Code Reference, DCN/DCE version, GC version, VCN version, SDMA version
-AMD Radeon (TM) HD 8500M/ 8600M /M200 /M320 /M330 /M335 Series, HAINAN, --,  6, --, --
-AMD Radeon HD 7800 /7900 /FireGL Series, TAHITI, DCE 6, 6, VCE 1 / UVD 3, --
-AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, VCE 1 / UVD 3, --
-AMD Radeon (TM) (HD|R7) 7800 /7970 /8800 /8970 /370/ Series, PITCAIRN, DCE 6, 6, VCE 1 / UVD 3, --
-AMD Radeon (TM|R7|R9|HD) E8860 /M360 /7700 /7800 /8800 /9000(M) /W4100 Series, VERDE, DCE 6, 6, VCE 1 / UVD 3, --
-AMD Radeon HD M280X /M380 /7700 /8950 /W5100, BONAIRE, DCE 8, 7, VCE 2 / UVD 4.2, 1
-AMD Radeon (R9|TM) 200 /390 /W8100 /W9100 Series, HAWAII, DCE 8, 7, VCE 2 / UVD 4.2, 1
-AMD Radeon (TM) R(5|7) M315 /M340 /M360, TOPAZ, *, 8, --, 2
-AMD Radeon (TM) R9 200 /380 /W7100 /S7150 /M390 /M395 Series, TONGA, DCE 10, 8, VCE 3 / UVD 5, 3
-AMD Radeon (FirePro) (TM) R9 Fury Series, FIJI, DCE 10, 8, VCE 3 / UVD 6, 3
-Radeon RX 470 /480 /570 /580 /590 Series - AMD Radeon (TM) (Pro WX) 5100 /E9390 /E9560 /E9565 /V7350 /7100 /P30PH, POLARIS10, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3
-Radeon (TM) (RX|Pro WX) E9260 /460 /V5300X /550 /560(X) Series, POLARIS11, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3
-Radeon (RX/Pro) 500 /540(X) /550 /640 /WX2100 /WX3100 /WX200 Series, POLARIS12, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3
-Radeon (RX|TM) (PRO|WX) Vega /MI25 /V320 /V340L /8200 /9100 /SSG MxGPU, VEGA10, DCE 12, 9.0.1, VCE 4.0.0 / UVD 7.0.0, 4.0.0
-AMD Radeon (Pro) VII /MI50 /MI60, VEGA20, DCE 12, 9.4.0, VCE 4.1.0 / UVD 7.2.0, 4.2.0
-MI100, ARCTURUS, *, 9.4.1, VCN 2.5.0, 4.2.2
-MI200 Series, ALDEBARAN, *, 9.4.2, VCN 2.6.0, 4.4.0
-MI300 Series, AQUA_VANJARAM, *, 9.4.3, VCN 4.0.3, 4.4.2
-AMD Radeon (RX|Pro) 5600(M|XT) /5700 (M|XT|XTB) /W5700, NAVI10, DCN 2.0.0, 10.1.10, VCN 2.0.0, 5.0.0
-AMD Radeon (Pro) 5300 /5500XTB/5500(XT|M) /W5500M /W5500, NAVI14, DCN 2.0.0, 10.1.1, VCN 2.0.2, 5.0.2
-AMD Radeon RX 6800(XT) /6900(XT) /W6800, SIENNA_CICHLID, DCN 3.0.0, 10.3.0, VCN 3.0.0, 5.2.0
-AMD Radeon RX 6700 XT / 6800M / 6700M, NAVY_FLOUNDER, DCN 3.0.0, 10.3.2, VCN 3.0.0, 5.2.2
-AMD Radeon RX 6600(XT) /6600M /W6600 /W6600M, DIMGREY_CAVEFISH, DCN 3.0.2, 10.3.4, VCN 3.0.16, 5.2.4
-AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3.0.33, 5.2.5
-AMD Radeon RX 7900 XT /XTX, , DCN 3.2.0, 11.0.0, VCN 4.0.0, 6.0.0
-AMD Radeon RX 7800 XT, , DCN 3.2.0, 11.0.3, VCN 4.0.0, 6.0.3
-AMD Radeon RX 7600M (XT) /7700S /7600S, , DCN 3.2.1, 11.0.2, VCN 4.0.4, 6.0.2
+Product Name, Code Reference, DCN/DCE version, GC version, VCN version, SDMA version, MP0 version, MP1 version
+AMD Radeon (TM) HD 8500M/ 8600M /M200 /M320 /M330 /M335 Series, HAINAN, --,  6, --, --, --, 6
+AMD Radeon HD 7800 /7900 /FireGL Series, TAHITI, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
+AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
+AMD Radeon (TM) (HD|R7) 7800 /7970 /8800 /8970 /370/ Series, PITCAIRN, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
+AMD Radeon (TM|R7|R9|HD) E8860 /M360 /7700 /7800 /8800 /9000(M) /W4100 Series, VERDE, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
+AMD Radeon HD M280X /M380 /7700 /8950 /W5100, BONAIRE, DCE 8, 7, VCE 2 / UVD 4.2, 1, --, 7
+AMD Radeon (R9|TM) 200 /390 /W8100 /W9100 Series, HAWAII, DCE 8, 7, VCE 2 / UVD 4.2, 1, --, 7
+AMD Radeon (TM) R(5|7) M315 /M340 /M360, TOPAZ, *, 8, --, 2, n/a, 7
+AMD Radeon (TM) R9 200 /380 /W7100 /S7150 /M390 /M395 Series, TONGA, DCE 10, 8, VCE 3 / UVD 5, 3, n/a, 7
+AMD Radeon (FirePro) (TM) R9 Fury Series, FIJI, DCE 10, 8, VCE 3 / UVD 6, 3, n/a, 7
+Radeon RX 470 /480 /570 /580 /590 Series - AMD Radeon (TM) (Pro WX) 5100 /E9390 /E9560 /E9565 /V7350 /7100 /P30PH, POLARIS10, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3, n/a, 7
+Radeon (TM) (RX|Pro WX) E9260 /460 /V5300X /550 /560(X) Series, POLARIS11, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3, n/a, 7
+Radeon (RX/Pro) 500 /540(X) /550 /640 /WX2100 /WX3100 /WX200 Series, POLARIS12, DCE 11.2, 8, VCE 3.4 / UVD 6.3, 3, n/a, 7
+Radeon (RX|TM) (PRO|WX) Vega /MI25 /V320 /V340L /8200 /9100 /SSG MxGPU, VEGA10, DCE 12, 9.0.1, VCE 4.0.0 / UVD 7.0.0, 4.0.0, 9.0.0, 9.0.0
+AMD Radeon (Pro) VII /MI50 /MI60, VEGA20, DCE 12, 9.4.0, VCE 4.1.0 / UVD 7.2.0, 4.2.0, 11.0.2, 11.0.2
+MI100, ARCTURUS, *, 9.4.1, VCN 2.5.0, 4.2.2, 11.0.4, 11.0.2
+MI200 Series, ALDEBARAN, *, 9.4.2, VCN 2.6.0, 4.4.0, 13.0.2, 13.0.2
+MI300 Series, AQUA_VANJARAM, *, 9.4.3, VCN 4.0.3, 4.4.2, 13.0.6, 13.0.6
+AMD Radeon (RX|Pro) 5600(M|XT) /5700 (M|XT|XTB) /W5700, NAVI10, DCN 2.0.0, 10.1.10, VCN 2.0.0, 5.0.0, 11.0.0, 11.0.0
+AMD Radeon (Pro) 5300 /5500XTB/5500(XT|M) /W5500M /W5500, NAVI14, DCN 2.0.0, 10.1.1, VCN 2.0.2, 5.0.2, 11.0.5, 11.0.5
+AMD Radeon RX 6800(XT) /6900(XT) /W6800, SIENNA_CICHLID, DCN 3.0.0, 10.3.0, VCN 3.0.0, 5.2.0, 11.0.7, 11.0.7
+AMD Radeon RX 6700 XT / 6800M / 6700M, NAVY_FLOUNDER, DCN 3.0.0, 10.3.2, VCN 3.0.0, 5.2.2, 11.0.11, 11.0.11
+AMD Radeon RX 6600(XT) /6600M /W6600 /W6600M, DIMGREY_CAVEFISH, DCN 3.0.2, 10.3.4, VCN 3.0.16, 5.2.4, 11.0.12, 11.0.12
+AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3.0.33, 5.2.5, 11.0.13, 11.0.13
+AMD Radeon RX 7900 XT /XTX, , DCN 3.2.0, 11.0.0, VCN 4.0.0, 6.0.0, 13.0.0, 13.0.0
+AMD Radeon RX 7800 XT, , DCN 3.2.0, 11.0.3, VCN 4.0.0, 6.0.3, 13.0.10, 13.0.10
+AMD Radeon RX 7600M (XT) /7700S /7600S, , DCN 3.2.1, 11.0.2, VCN 4.0.4, 6.0.2, 13.0.7, 13.0.7

From 8f249ba6ec607313c2dd5d6c8f99ad647c99af02 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 28 Jul 2025 11:21:19 -0400
Subject: [PATCH 0222/1455] Documentation: add RDNA4 dGPUs

Add RDNA4 dGPUs to the dGPU table.

Link: https://www.amd.com/en/products/graphics/desktops/radeon/9000-series/amd-radeon-rx-9070xt.html
Link: https://www.amd.com/en/products/graphics/desktops/radeon/9000-series/amd-radeon-rx-9070.html
Link: https://www.amd.com/en/products/graphics/desktops/radeon/9000-series/amd-radeon-rx-9060xt.html
Link: https://www.amd.com/en/products/graphics/desktops/radeon/9000-series/amd-radeon-rx-9060xt-8gb.html
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/dgpu-asic-info-table.csv | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
index 0f822a252fcb..a3796daec9ec 100644
--- a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
@@ -26,3 +26,5 @@ AMD Radeon RX 6500M /6300M /W6500M /W6300M, BEIGE_GOBY, DCN 3.0.3, 10.3.5, VCN 3
 AMD Radeon RX 7900 XT /XTX, , DCN 3.2.0, 11.0.0, VCN 4.0.0, 6.0.0, 13.0.0, 13.0.0
 AMD Radeon RX 7800 XT, , DCN 3.2.0, 11.0.3, VCN 4.0.0, 6.0.3, 13.0.10, 13.0.10
 AMD Radeon RX 7600M (XT) /7700S /7600S, , DCN 3.2.1, 11.0.2, VCN 4.0.4, 6.0.2, 13.0.7, 13.0.7
+AMD Radeon RX 9070 (XT), , DCN 4.0.1, 12.0.1, VCN 5.0.0, 7.0.1, 14.0.3, 14.0.3
+AMD Radeon RX 9060 XT, , DCN 4.0.1, 12.0.0, VCN 5.0.0, 7.0.0, 14.0.2, 14.0.2

From a578f2a58c3ab38f0643b1b6e7534af860233cb1 Mon Sep 17 00:00:00 2001
From: David Yat Sin <David.YatSin@amd.com>
Date: Wed, 16 Jul 2025 22:04:28 +0000
Subject: [PATCH 0223/1455] drm/amdkfd: Fix checkpoint-restore on multi-xcc

GPUs with multi-xcc have multiple MQDs per queue. This patch saves and
restores all the MQDs within the partition.

Signed-off-by: David Yat Sin <David.YatSin@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/amdkfd/kfd_device_queue_manager.c |  2 +-
 .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c   | 61 ++++++++++++++++---
 .../amd/amdkfd/kfd_process_queue_manager.c    | 20 ++++--
 3 files changed, 67 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index 2d91027e2a74..6c5c7c1bf5ed 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -2725,7 +2725,7 @@ static void get_queue_checkpoint_info(struct device_queue_manager *dqm,
 
 	dqm_lock(dqm);
 	mqd_mgr = dqm->mqd_mgrs[mqd_type];
-	*mqd_size = mqd_mgr->mqd_size;
+	*mqd_size = mqd_mgr->mqd_size * NUM_XCC(mqd_mgr->dev->xcc_mask);
 	*ctl_stack_size = 0;
 
 	if (q->properties.type == KFD_QUEUE_TYPE_COMPUTE && mqd_mgr->get_checkpoint_info)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index 97933d2a3803..f2dee320fada 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -373,7 +373,7 @@ static void get_checkpoint_info(struct mqd_manager *mm, void *mqd, u32 *ctl_stac
 {
 	struct v9_mqd *m = get_mqd(mqd);
 
-	*ctl_stack_size = m->cp_hqd_cntl_stack_size;
+	*ctl_stack_size = m->cp_hqd_cntl_stack_size * NUM_XCC(mm->dev->xcc_mask);
 }
 
 static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, void *ctl_stack_dst)
@@ -388,6 +388,24 @@ static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, voi
 	memcpy(ctl_stack_dst, ctl_stack, m->cp_hqd_cntl_stack_size);
 }
 
+static void checkpoint_mqd_v9_4_3(struct mqd_manager *mm,
+								  void *mqd,
+								  void *mqd_dst,
+								  void *ctl_stack_dst)
+{
+	struct v9_mqd *m;
+	int xcc;
+	uint64_t size = get_mqd(mqd)->cp_mqd_stride_size;
+
+	for (xcc = 0; xcc < NUM_XCC(mm->dev->xcc_mask); xcc++) {
+		m = get_mqd(mqd + size * xcc);
+
+		checkpoint_mqd(mm, m,
+				(uint8_t *)mqd_dst + sizeof(*m) * xcc,
+				(uint8_t *)ctl_stack_dst + m->cp_hqd_cntl_stack_size * xcc);
+	}
+}
+
 static void restore_mqd(struct mqd_manager *mm, void **mqd,
 			struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr,
 			struct queue_properties *qp,
@@ -764,13 +782,35 @@ static void restore_mqd_v9_4_3(struct mqd_manager *mm, void **mqd,
 			const void *mqd_src,
 			const void *ctl_stack_src, u32 ctl_stack_size)
 {
-	restore_mqd(mm, mqd, mqd_mem_obj, gart_addr, qp, mqd_src, ctl_stack_src, ctl_stack_size);
-	if (amdgpu_sriov_multi_vf_mode(mm->dev->adev)) {
-		struct v9_mqd *m;
+	struct kfd_mem_obj xcc_mqd_mem_obj;
+	u32 mqd_ctl_stack_size;
+	struct v9_mqd *m;
+	u32 num_xcc;
+	int xcc;
 
-		m = (struct v9_mqd *) mqd_mem_obj->cpu_ptr;
-		m->cp_hqd_pq_doorbell_control |= 1 <<
-				CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_MODE__SHIFT;
+	uint64_t offset = mm->mqd_stride(mm, qp);
+
+	mm->dev->dqm->current_logical_xcc_start++;
+
+	num_xcc = NUM_XCC(mm->dev->xcc_mask);
+	mqd_ctl_stack_size = ctl_stack_size / num_xcc;
+
+	memset(&xcc_mqd_mem_obj, 0x0, sizeof(struct kfd_mem_obj));
+
+	/* Set the MQD pointer and gart address to XCC0 MQD */
+	*mqd = mqd_mem_obj->cpu_ptr;
+	if (gart_addr)
+		*gart_addr = mqd_mem_obj->gpu_addr;
+
+	for (xcc = 0; xcc < num_xcc; xcc++) {
+		get_xcc_mqd(mqd_mem_obj, &xcc_mqd_mem_obj, offset * xcc);
+		restore_mqd(mm, (void **)&m,
+					&xcc_mqd_mem_obj,
+					NULL,
+					qp,
+					(uint8_t *)mqd_src + xcc * sizeof(*m),
+					(uint8_t *)ctl_stack_src + xcc *  mqd_ctl_stack_size,
+					mqd_ctl_stack_size);
 	}
 }
 static int destroy_mqd_v9_4_3(struct mqd_manager *mm, void *mqd,
@@ -906,7 +946,6 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
 		mqd->free_mqd = kfd_free_mqd_cp;
 		mqd->is_occupied = kfd_is_occupied_cp;
 		mqd->get_checkpoint_info = get_checkpoint_info;
-		mqd->checkpoint_mqd = checkpoint_mqd;
 		mqd->mqd_size = sizeof(struct v9_mqd);
 		mqd->mqd_stride = mqd_stride_v9;
 #if defined(CONFIG_DEBUG_FS)
@@ -918,16 +957,18 @@ struct mqd_manager *mqd_manager_init_v9(enum KFD_MQD_TYPE type,
 			mqd->init_mqd = init_mqd_v9_4_3;
 			mqd->load_mqd = load_mqd_v9_4_3;
 			mqd->update_mqd = update_mqd_v9_4_3;
-			mqd->restore_mqd = restore_mqd_v9_4_3;
 			mqd->destroy_mqd = destroy_mqd_v9_4_3;
 			mqd->get_wave_state = get_wave_state_v9_4_3;
+			mqd->checkpoint_mqd = checkpoint_mqd_v9_4_3;
+			mqd->restore_mqd = restore_mqd_v9_4_3;
 		} else {
 			mqd->init_mqd = init_mqd;
 			mqd->load_mqd = load_mqd;
 			mqd->update_mqd = update_mqd;
-			mqd->restore_mqd = restore_mqd;
 			mqd->destroy_mqd = kfd_destroy_mqd_cp;
 			mqd->get_wave_state = get_wave_state;
+			mqd->checkpoint_mqd = checkpoint_mqd;
+			mqd->restore_mqd = restore_mqd;
 		}
 		break;
 	case KFD_MQD_TYPE_HIQ:
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
index c643e0ccec52..7fbb5c274ccc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c
@@ -914,7 +914,10 @@ static int criu_checkpoint_queues_device(struct kfd_process_device *pdd,
 
 		q_data = (struct kfd_criu_queue_priv_data *)q_private_data;
 
-		/* data stored in this order: priv_data, mqd, ctl_stack */
+		/*
+		 * data stored in this order:
+		 * priv_data, mqd[xcc0], mqd[xcc1],..., ctl_stack[xcc0], ctl_stack[xcc1]...
+		 */
 		q_data->mqd_size = mqd_size;
 		q_data->ctl_stack_size = ctl_stack_size;
 
@@ -963,7 +966,7 @@ int kfd_criu_checkpoint_queues(struct kfd_process *p,
 }
 
 static void set_queue_properties_from_criu(struct queue_properties *qp,
-					  struct kfd_criu_queue_priv_data *q_data)
+					  struct kfd_criu_queue_priv_data *q_data, uint32_t num_xcc)
 {
 	qp->is_interop = false;
 	qp->queue_percent = q_data->q_percent;
@@ -976,7 +979,11 @@ static void set_queue_properties_from_criu(struct queue_properties *qp,
 	qp->eop_ring_buffer_size = q_data->eop_ring_buffer_size;
 	qp->ctx_save_restore_area_address = q_data->ctx_save_restore_area_address;
 	qp->ctx_save_restore_area_size = q_data->ctx_save_restore_area_size;
-	qp->ctl_stack_size = q_data->ctl_stack_size;
+	if (q_data->type == KFD_QUEUE_TYPE_COMPUTE)
+		qp->ctl_stack_size = q_data->ctl_stack_size / num_xcc;
+	else
+		qp->ctl_stack_size = q_data->ctl_stack_size;
+
 	qp->type = q_data->type;
 	qp->format = q_data->format;
 }
@@ -1036,12 +1043,15 @@ int kfd_criu_restore_queue(struct kfd_process *p,
 		goto exit;
 	}
 
-	/* data stored in this order: mqd, ctl_stack */
+	/*
+	 * data stored in this order:
+	 * mqd[xcc0], mqd[xcc1],..., ctl_stack[xcc0], ctl_stack[xcc1]...
+	 */
 	mqd = q_extra_data;
 	ctl_stack = mqd + q_data->mqd_size;
 
 	memset(&qp, 0, sizeof(qp));
-	set_queue_properties_from_criu(&qp, q_data);
+	set_queue_properties_from_criu(&qp, q_data, NUM_XCC(pdd->dev->adev->gfx.xcc_mask));
 
 	print_queue_properties(&qp);
 

From 70e33073d941b4fc1e773769bd7d72220c14d41d Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Mon, 21 Jul 2025 19:04:34 +0530
Subject: [PATCH 0224/1455] drm/amdgpu: Fix kdoc style in amdgpu_fence.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The initial comment block before
amdgpu_fence_driver_guilty_force_completion() incorrectly used '/**' but
is not a kernel-doc comment, causing build warnings.

Fixes the below with gcc W=1:
drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c:742: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Kernel queue reset handling

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 9e7506965cab..bcb74286a78a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -738,7 +738,7 @@ void amdgpu_fence_driver_force_completion(struct amdgpu_ring *ring)
 }
 
 
-/**
+/*
  * Kernel queue reset handling
  *
  * The driver can reset individual queues for most engines, but those queues

From 8f0245ee95c5ba65a2fe03f60386868353c6a3a0 Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Wed, 23 Jul 2025 14:28:35 +0800
Subject: [PATCH 0225/1455] drm/amdgpu: Update IPID value for bad page
 threshold CPER

Update the IPID register value for bad page threshold CPER according to
the latest definition.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
index 25252231a68a..6c266f18c598 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
@@ -206,6 +206,7 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev
 {
 	struct cper_sec_desc *section_desc;
 	struct cper_sec_nonstd_err *section;
+	uint32_t socket_id;
 
 	section_desc = (struct cper_sec_desc *)((uint8_t *)hdr + SEC_DESC_OFFSET(idx));
 	section = (struct cper_sec_nonstd_err *)((uint8_t *)hdr +
@@ -224,6 +225,9 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev
 	section->ctx.reg_arr_size = sizeof(section->ctx.reg_dump);
 
 	/* Hardcoded Reg dump for bad page threshold CPER */
+	socket_id = (adev->smuio.funcs && adev->smuio.funcs->get_socket_id) ?
+				adev->smuio.funcs->get_socket_id(adev) :
+				0;
 	section->ctx.reg_dump[CPER_ACA_REG_CTL_LO]    = 0x1;
 	section->ctx.reg_dump[CPER_ACA_REG_CTL_HI]    = 0x0;
 	section->ctx.reg_dump[CPER_ACA_REG_STATUS_LO] = 0x137;
@@ -234,8 +238,8 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev
 	section->ctx.reg_dump[CPER_ACA_REG_MISC0_HI]  = 0x0;
 	section->ctx.reg_dump[CPER_ACA_REG_CONFIG_LO] = 0x2;
 	section->ctx.reg_dump[CPER_ACA_REG_CONFIG_HI] = 0x1ff;
-	section->ctx.reg_dump[CPER_ACA_REG_IPID_LO]   = 0x0;
-	section->ctx.reg_dump[CPER_ACA_REG_IPID_HI]   = 0x96;
+	section->ctx.reg_dump[CPER_ACA_REG_IPID_LO]   = (socket_id / 4) & 0x01;
+	section->ctx.reg_dump[CPER_ACA_REG_IPID_HI]   = 0x096 | (((socket_id % 4) & 0x3) << 12);
 	section->ctx.reg_dump[CPER_ACA_REG_SYND_LO]   = 0x0;
 	section->ctx.reg_dump[CPER_ACA_REG_SYND_HI]   = 0x0;
 

From 21c0ffa612c98bcc6dab5bd9d977a18d565ee28e Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Sun, 27 Jul 2025 12:06:55 +0800
Subject: [PATCH 0226/1455] drm/amdgpu: Avoid rma causes GPU duplicate reset

Try to ensure poison creation handle is completed in time
to set device rma value.

Signed-off-by: Ce Sun <cesun102@amd.com>
Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 17 ++++++++++-------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h |  1 +
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 1e31ef5592f2..a2af7c835962 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -3363,7 +3363,6 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work)
 					      page_retirement_dwork.work);
 	struct amdgpu_device *adev = con->adev;
 	struct ras_err_data err_data;
-	unsigned long err_cnt;
 
 	/* If gpu reset is ongoing, delay retiring the bad pages */
 	if (amdgpu_in_reset(adev) || amdgpu_ras_in_recovery(adev)) {
@@ -3375,13 +3374,9 @@ static void amdgpu_ras_do_page_retirement(struct work_struct *work)
 	amdgpu_ras_error_data_init(&err_data);
 
 	amdgpu_umc_handle_bad_pages(adev, &err_data);
-	err_cnt = err_data.err_addr_cnt;
 
 	amdgpu_ras_error_data_fini(&err_data);
 
-	if (err_cnt && amdgpu_ras_is_rma(adev))
-		amdgpu_ras_reset_gpu(adev);
-
 	amdgpu_ras_schedule_retirement_dwork(con,
 			AMDGPU_RAS_RETIRE_PAGE_INTERVAL);
 }
@@ -3435,6 +3430,9 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev,
 	if (total_detect_count)
 		schedule_delayed_work(&ras->page_retirement_dwork, 0);
 
+	if (amdgpu_ras_is_rma(adev) && atomic_cmpxchg(&ras->rma_in_recovery, 0, 1) == 0)
+		amdgpu_ras_reset_gpu(adev);
+
 	return 0;
 }
 
@@ -3470,6 +3468,12 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev,
 		reset_flags |= msg.reset;
 	}
 
+	/*
+	 * Try to ensure poison creation handler is completed first
+	 * to set rma if bad page exceed threshold.
+	 */
+	flush_delayed_work(&con->page_retirement_dwork);
+
 	/* for RMA, amdgpu_ras_poison_creation_handler will trigger gpu reset */
 	if (reset_flags && !amdgpu_ras_is_rma(adev)) {
 		if (reset_flags & AMDGPU_RAS_GPU_RESET_MODE1_RESET)
@@ -3479,8 +3483,6 @@ static int amdgpu_ras_poison_consumption_handler(struct amdgpu_device *adev,
 		else
 			reset = reset_flags;
 
-		flush_delayed_work(&con->page_retirement_dwork);
-
 		con->gpu_reset_flags |= reset;
 		amdgpu_ras_reset_gpu(adev);
 
@@ -3648,6 +3650,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info)
 	mutex_init(&con->recovery_lock);
 	INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery);
 	atomic_set(&con->in_recovery, 0);
+	atomic_set(&con->rma_in_recovery, 0);
 	con->eeprom_control.bad_channel_bitmap = 0;
 
 	max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(&con->eeprom_control);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 434e23c84962..ff63020f9c6c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -522,6 +522,7 @@ struct amdgpu_ras {
 	/* gpu recovery */
 	struct work_struct recovery_work;
 	atomic_t in_recovery;
+	atomic_t rma_in_recovery;
 	struct amdgpu_device *adev;
 	/* error handler data */
 	struct ras_err_handler_data *eh_data;

From da467352296f8e50c7ab7057ead44a1df1c81496 Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Sat, 26 Jul 2025 20:16:24 +0800
Subject: [PATCH 0227/1455] drm/amdgpu: Effective health check before reset

Move amdgpu_device_health_check into amdgpu_device_gpu_recover to
ensure that if the device is present can be checked before reset

The reason is:
1.During the dpc event, the device where the dpc event occurs is not
present on the bus
2.When both dpc event and ATHUB event occur simultaneously,the dpc thread
holds the reset domain lock when detecting error,and the gpu recover thread
acquires the hive lock.The device is simultaneously in the states of
amdgpu_ras_in_recovery and occurs_dpc,so gpu recover thread will not go to
amdgpu_device_health_check.It waits for the reset domain lock held by the
dpc thread, but dpc thread has not released the reset domain lock.In the dpc
callback slot_reset,to obtain the hive lock, the hive lock is held by the
gpu recover thread at this time.So a deadlock occurred

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 26 +++++++---------------
 1 file changed, 8 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index a43ce3404849..8a9488a98646 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -6134,12 +6134,11 @@ static int amdgpu_device_health_check(struct list_head *device_list_handle)
 	return ret;
 }
 
-static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev,
+static void amdgpu_device_recovery_prepare(struct amdgpu_device *adev,
 					  struct list_head *device_list,
 					  struct amdgpu_hive_info *hive)
 {
 	struct amdgpu_device *tmp_adev = NULL;
-	int r;
 
 	/*
 	 * Build list of devices to reset.
@@ -6159,14 +6158,6 @@ static int amdgpu_device_recovery_prepare(struct amdgpu_device *adev,
 	} else {
 		list_add_tail(&adev->reset_list, device_list);
 	}
-
-	if (!amdgpu_sriov_vf(adev) && (!adev->pcie_reset_ctx.occurs_dpc)) {
-		r = amdgpu_device_health_check(device_list);
-		if (r)
-			return r;
-	}
-
-	return 0;
 }
 
 static void amdgpu_device_recovery_get_reset_lock(struct amdgpu_device *adev,
@@ -6459,8 +6450,13 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 	reset_context->hive = hive;
 	INIT_LIST_HEAD(&device_list);
 
-	if (amdgpu_device_recovery_prepare(adev, &device_list, hive))
-		goto end_reset;
+	amdgpu_device_recovery_prepare(adev, &device_list, hive);
+
+	if (!amdgpu_sriov_vf(adev)) {
+		r = amdgpu_device_health_check(&device_list);
+		if (r)
+			goto end_reset;
+	}
 
 	/* We need to lock reset domain only once both for XGMI and single device */
 	amdgpu_device_recovery_get_reset_lock(adev, &device_list);
@@ -6967,12 +6963,6 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
 	int r = 0, i;
 	u32 memsize;
 
-	/* PCI error slot reset should be skipped During RAS recovery */
-	if ((amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) ||
-	    amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) &&
-	    amdgpu_ras_in_recovery(adev))
-		return PCI_ERS_RESULT_RECOVERED;
-
 	dev_info(adev->dev, "PCI error: slot reset callback!!\n");
 
 	memset(&reset_context, 0, sizeof(reset_context));

From bbddcbe36a686af03e91341b9bbfcca94bd45fb6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Tue, 22 Jul 2025 17:58:29 +0200
Subject: [PATCH 0228/1455] drm/amd/display: Don't overwrite dce60_clk_mgr
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

dc_clk_mgr_create accidentally overwrites the dce60_clk_mgr
with the dce_clk_mgr, causing incorrect behaviour on DCE6.
Fix it by removing the extra dce_clk_mgr_construct.

Fixes: 62eab49faae7 ("drm/amd/display: hide VGH asic specific structs")
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
index 33b9d36619ff..4071851f9e86 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
@@ -158,7 +158,6 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p
 			return NULL;
 		}
 		dce60_clk_mgr_construct(ctx, clk_mgr);
-		dce_clk_mgr_construct(ctx, clk_mgr);
 		return &clk_mgr->base;
 	}
 #endif

From 35222b5934ec8d762473592ece98659baf6bc48e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Tue, 22 Jul 2025 17:58:30 +0200
Subject: [PATCH 0229/1455] drm/amd/display: Fix DCE 6.0 and 6.4 PLL
 programming.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Apparently, both DCE 6.0 and 6.4 have 3 PLLs, but PLL0 can only
be used for DP. Make sure to initialize the correct amount of PLLs
in DC for these DCE versions and use PLL0 only for DP.

Also, on DCE 6.0 and 6.4, the PLL0 needs to be powered on at
initialization as opposed to DCE 6.1 and 7.x which use a different
clock source for DFS.

The following functions were used as reference from the	old
radeon driver implementation of	DCE 6.x:
- radeon_atom_pick_pll
- atombios_crtc_set_disp_eng_pll

Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dce100/dce_clk_mgr.c   |  5 +++
 .../dc/resource/dce60/dce60_resource.c        | 34 +++++++++++--------
 2 files changed, 25 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
index 26feefbb8990..f5ad0a177038 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
@@ -245,6 +245,11 @@ int dce_set_clock(
 	pxl_clk_params.target_pixel_clock_100hz = requested_clk_khz * 10;
 	pxl_clk_params.pll_id = CLOCK_SOURCE_ID_DFS;
 
+	/* DCE 6.0, DCE 6.4: engine clock is the same as PLL0 */
+	if (clk_mgr_base->ctx->dce_version == DCE_VERSION_6_0 ||
+	    clk_mgr_base->ctx->dce_version == DCE_VERSION_6_4)
+		pxl_clk_params.pll_id = CLOCK_SOURCE_ID_PLL0;
+
 	if (clk_mgr_dce->dfs_bypass_active)
 		pxl_clk_params.flags.SET_DISPCLK_DFS_BYPASS = true;
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
index 58b59d52dc9d..53b60044653f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
@@ -373,7 +373,7 @@ static const struct resource_caps res_cap = {
 		.num_timing_generator = 6,
 		.num_audio = 6,
 		.num_stream_encoder = 6,
-		.num_pll = 2,
+		.num_pll = 3,
 		.num_ddc = 6,
 };
 
@@ -389,7 +389,7 @@ static const struct resource_caps res_cap_64 = {
 		.num_timing_generator = 2,
 		.num_audio = 2,
 		.num_stream_encoder = 2,
-		.num_pll = 2,
+		.num_pll = 3,
 		.num_ddc = 2,
 };
 
@@ -973,21 +973,24 @@ static bool dce60_construct(
 
 	if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) {
 		pool->base.dp_clock_source =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
 
+		/* DCE 6.0 and 6.4: PLL0 can only be used with DP. Don't initialize it here. */
 		pool->base.clock_sources[0] =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], false);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
 		pool->base.clock_sources[1] =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[2], false);
 		pool->base.clk_src_count = 2;
 
 	} else {
 		pool->base.dp_clock_source =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], true);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], true);
 
 		pool->base.clock_sources[0] =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
-		pool->base.clk_src_count = 1;
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+		pool->base.clock_sources[1] =
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[2], false);
+		pool->base.clk_src_count = 2;
 	}
 
 	if (pool->base.dp_clock_source == NULL) {
@@ -1365,21 +1368,24 @@ static bool dce64_construct(
 
 	if (bp->fw_info_valid && bp->fw_info.external_clock_source_frequency_for_dp != 0) {
 		pool->base.dp_clock_source =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_EXTERNAL, NULL, true);
 
+		/* DCE 6.0 and 6.4: PLL0 can only be used with DP. Don't initialize it here. */
 		pool->base.clock_sources[0] =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[0], false);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
 		pool->base.clock_sources[1] =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[1], false);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[2], false);
 		pool->base.clk_src_count = 2;
 
 	} else {
 		pool->base.dp_clock_source =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[0], true);
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL0, &clk_src_regs[0], true);
 
 		pool->base.clock_sources[0] =
-				dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[1], false);
-		pool->base.clk_src_count = 1;
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL1, &clk_src_regs[1], false);
+		pool->base.clock_sources[1] =
+			dce60_clock_source_create(ctx, bp, CLOCK_SOURCE_ID_PLL2, &clk_src_regs[2], false);
+		pool->base.clk_src_count = 2;
 	}
 
 	if (pool->base.dp_clock_source == NULL) {

From 46b0e6b9d749cfa891e6969d6565be1131c53aa2 Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Mon, 28 Jul 2025 18:27:06 +0530
Subject: [PATCH 0230/1455] drm/amdgpu: Fix unintended error log in VCN5_0_0

The error log is supposed to be gaurded under if failure condition.

Fixes: faab5ea08367 ("drm/amdgpu: Check vcn sram load return value")
Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index f8bb90fe764b..f9dbd2757d83 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -769,9 +769,10 @@ static int vcn_v5_0_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 
 	if (indirect) {
 		ret = amdgpu_vcn_psp_update_sram(adev, inst_idx, 0);
-		dev_err(adev->dev, "%s: vcn sram load failed %d\n", __func__, ret);
-		if (ret)
+		if (ret) {
+			dev_err(adev->dev, "%s: vcn sram load failed %d\n", __func__, ret);
 			return ret;
+		}
 	}
 
 	ring = &adev->vcn.inst[inst_idx].ring_enc[0];

From 130c7ed88f62f72bd99db69e634a5f0ac10a2362 Mon Sep 17 00:00:00 2001
From: Yunshui Jiang <jiangyunshui@kylinos.cn>
Date: Thu, 24 Jul 2025 09:37:53 +0800
Subject: [PATCH 0231/1455] drm/amdgpu: use kmalloc_array() instead of
 kmalloc()

Use kmalloc_array() instead of kmalloc() with multiplication.
kmalloc_array() is a safer way because of its multiply overflow check.

Signed-off-by: Yunshui Jiang <jiangyunshui@kylinos.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index a2af7c835962..e6b8c2cf80ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2622,7 +2622,7 @@ static int amdgpu_ras_badpages_read(struct amdgpu_device *adev,
 		goto out;
 	}
 
-	*bps = kmalloc(sizeof(struct ras_badpage) * data->count, GFP_KERNEL);
+	*bps = kmalloc_array(data->count, sizeof(struct ras_badpage), GFP_KERNEL);
 	if (!*bps) {
 		ret = -ENOMEM;
 		goto out;
@@ -2786,7 +2786,7 @@ static int amdgpu_ras_realloc_eh_data_space(struct amdgpu_device *adev,
 	unsigned int old_space = data->count + data->space_left;
 	unsigned int new_space = old_space + pages;
 	unsigned int align_space = ALIGN(new_space, 512);
-	void *bps = kmalloc(align_space * sizeof(*data->bps), GFP_KERNEL);
+	void *bps = kmalloc_array(align_space, sizeof(*data->bps), GFP_KERNEL);
 
 	if (!bps) {
 		return -ENOMEM;

From cc51bbc7d7a7ff63de3e095e3681e78fe63ddda6 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Tue, 15 Jul 2025 16:24:20 -0500
Subject: [PATCH 0232/1455] drm/amd: Use drm_*() macros instead of DRM_*() for
 amdgpu_cs

Some of the IOCTL messages can be called for different GPUs and it might
not be obvious which one called them from a problem.  Using the drm_*()
macros the correct device will be shown in the messages.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250715212420.2254925-1-superm1@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index a2adaacf6adb..27f8f316f6c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -396,7 +396,7 @@ static int amdgpu_cs_p2_ib(struct amdgpu_cs_parser *p,
 			   chunk_ib->ib_bytes : 0,
 			   AMDGPU_IB_POOL_DELAYED, ib);
 	if (r) {
-		DRM_ERROR("Failed to get ib !\n");
+		drm_err(adev_to_drm(p->adev), "Failed to get ib !\n");
 		return r;
 	}
 
@@ -468,7 +468,7 @@ static int amdgpu_syncobj_lookup_and_add(struct amdgpu_cs_parser *p,
 
 	r = drm_syncobj_find_fence(p->filp, handle, point, flags, &fence);
 	if (r) {
-		DRM_ERROR("syncobj %u failed to find fence @ %llu (%d)!\n",
+		drm_err(adev_to_drm(p->adev), "syncobj %u failed to find fence @ %llu (%d)!\n",
 			  handle, point, r);
 		return r;
 	}
@@ -902,7 +902,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 					 sizeof(struct page *),
 					 GFP_KERNEL);
 		if (!e->user_pages) {
-			DRM_ERROR("kvmalloc_array failure\n");
+			drm_err(adev_to_drm(p->adev), "kvmalloc_array failure\n");
 			r = -ENOMEM;
 			goto out_free_user_pages;
 		}
@@ -983,7 +983,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 	r = amdgpu_vm_validate(p->adev, &fpriv->vm, NULL,
 			       amdgpu_cs_bo_validate, p);
 	if (r) {
-		DRM_ERROR("amdgpu_vm_validate() failed.\n");
+		drm_err(adev_to_drm(p->adev), "amdgpu_vm_validate() failed.\n");
 		goto out_free_user_pages;
 	}
 
@@ -1061,13 +1061,13 @@ static int amdgpu_cs_patch_ibs(struct amdgpu_cs_parser *p,
 		va_start = ib->gpu_addr & AMDGPU_GMC_HOLE_MASK;
 		r = amdgpu_cs_find_mapping(p, va_start, &aobj, &m);
 		if (r) {
-			DRM_ERROR("IB va_start is invalid\n");
+			drm_err(adev_to_drm(p->adev), "IB va_start is invalid\n");
 			return r;
 		}
 
 		if ((va_start + ib->length_dw * 4) >
 		    (m->last + 1) * AMDGPU_GPU_PAGE_SIZE) {
-			DRM_ERROR("IB va_start+ib_bytes is invalid\n");
+			drm_err(adev_to_drm(p->adev), "IB va_start+ib_bytes is invalid\n");
 			return -EINVAL;
 		}
 
@@ -1235,7 +1235,7 @@ static int amdgpu_cs_sync_rings(struct amdgpu_cs_parser *p)
 	r = amdgpu_ctx_wait_prev_fence(p->ctx, p->entities[p->gang_leader_idx]);
 	if (r) {
 		if (r != -ERESTARTSYS)
-			DRM_ERROR("amdgpu_ctx_wait_prev_fence failed.\n");
+			drm_err(adev_to_drm(p->adev), "amdgpu_ctx_wait_prev_fence failed.\n");
 		return r;
 	}
 
@@ -1448,7 +1448,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 
 	r = amdgpu_cs_parser_init(&parser, adev, filp, data);
 	if (r) {
-		DRM_ERROR_RATELIMITED("Failed to initialize parser %d!\n", r);
+		drm_err_ratelimited(dev, "Failed to initialize parser %d!\n", r);
 		return r;
 	}
 
@@ -1463,9 +1463,9 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	r = amdgpu_cs_parser_bos(&parser, data);
 	if (r) {
 		if (r == -ENOMEM)
-			DRM_ERROR("Not enough memory for command submission!\n");
+			drm_err(dev, "Not enough memory for command submission!\n");
 		else if (r != -ERESTARTSYS && r != -EAGAIN)
-			DRM_DEBUG("Failed to process the buffer list %d!\n", r);
+			drm_dbg(dev, "Failed to process the buffer list %d!\n", r);
 		goto error_fini;
 	}
 

From ed76936c6b10b547c6df4ca75412331e9ef6d339 Mon Sep 17 00:00:00 2001
From: YuanShang <YuanShang.Mao@amd.com>
Date: Wed, 23 Jul 2025 16:44:49 +0800
Subject: [PATCH 0233/1455] drm/amdgpu: Retain job->vm in
 amdgpu_job_prepare_job

The field job->vm is used in function amdgpu_job_run to get the page
table re-generation counter and decide whether the job should be skipped.

Specifically, function amdgpu_vm_generation checks if the VM is valid for this job to use.
For instance, if a gfx job depends on a cancelled sdma job from entity vm->delayed,
then the gfx job should be skipped.

Fixes: 26c95e838e63 ("drm/amdgpu: set the VM pointer to NULL in amdgpu_job_prepare")
Signed-off-by: YuanShang <YuanShang.Mao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index e6061d45f142..9b1c55115921 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -365,13 +365,6 @@ amdgpu_job_prepare_job(struct drm_sched_job *sched_job,
 			dev_err(ring->adev->dev, "Error getting VM ID (%d)\n", r);
 			goto error;
 		}
-		/*
-		 * The VM structure might be released after the VMID is
-		 * assigned, we had multiple problems with people trying to use
-		 * the VM pointer so better set it to NULL.
-		 */
-		if (!fence)
-			job->vm = NULL;
 		return fence;
 	}
 

From e6c2b0f23221ed43c4cc6f636e9ab7862954d562 Mon Sep 17 00:00:00 2001
From: Meng Li <li.meng@amd.com>
Date: Fri, 9 May 2025 13:44:24 +0800
Subject: [PATCH 0234/1455] drm/amd/amdgpu: Release xcp drm memory after unplug

Add a new API amdgpu_xcp_drm_dev_free().
After unplug xcp device, need to release xcp drm memory etc.

Co-developed-by: Jiang Liu <gerry@linux.alibaba.com>
Signed-off-by: Jiang Liu <gerry@linux.alibaba.com>
Signed-off-by: Meng Li <li.meng@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c     |  1 +
 drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c | 56 +++++++++++++++++----
 drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h |  1 +
 3 files changed, 49 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index c417f8689220..699acc1b46b5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -406,6 +406,7 @@ void amdgpu_xcp_dev_unplug(struct amdgpu_device *adev)
 		p_ddev->primary->dev = adev->xcp_mgr->xcp[i].pdev;
 		p_ddev->driver =  adev->xcp_mgr->xcp[i].driver;
 		p_ddev->vma_offset_manager = adev->xcp_mgr->xcp[i].vma_offset_manager;
+		amdgpu_xcp_drm_dev_free(p_ddev);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c
index 8bc36f04b1b7..44009aa8216e 100644
--- a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c
+++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.c
@@ -46,18 +46,29 @@ static const struct drm_driver amdgpu_xcp_driver = {
 
 static int8_t pdev_num;
 static struct xcp_device *xcp_dev[MAX_XCP_PLATFORM_DEVICE];
+static DEFINE_MUTEX(xcp_mutex);
 
 int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev)
 {
 	struct platform_device *pdev;
 	struct xcp_device *pxcp_dev;
 	char dev_name[20];
-	int ret;
+	int ret, i;
+
+	guard(mutex)(&xcp_mutex);
 
 	if (pdev_num >= MAX_XCP_PLATFORM_DEVICE)
 		return -ENODEV;
 
-	snprintf(dev_name, sizeof(dev_name), "amdgpu_xcp_%d", pdev_num);
+	for (i = 0; i < MAX_XCP_PLATFORM_DEVICE; i++) {
+		if (!xcp_dev[i])
+			break;
+	}
+
+	if (i >= MAX_XCP_PLATFORM_DEVICE)
+		return -ENODEV;
+
+	snprintf(dev_name, sizeof(dev_name), "amdgpu_xcp_%d", i);
 	pdev = platform_device_register_simple(dev_name, -1, NULL, 0);
 	if (IS_ERR(pdev))
 		return PTR_ERR(pdev);
@@ -73,8 +84,8 @@ int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev)
 		goto out_devres;
 	}
 
-	xcp_dev[pdev_num] = pxcp_dev;
-	xcp_dev[pdev_num]->pdev = pdev;
+	xcp_dev[i] = pxcp_dev;
+	xcp_dev[i]->pdev = pdev;
 	*ddev = &pxcp_dev->drm;
 	pdev_num++;
 
@@ -89,16 +100,43 @@ out_unregister:
 }
 EXPORT_SYMBOL(amdgpu_xcp_drm_dev_alloc);
 
-void amdgpu_xcp_drv_release(void)
+static void free_xcp_dev(int8_t index)
 {
-	for (--pdev_num; pdev_num >= 0; --pdev_num) {
-		struct platform_device *pdev = xcp_dev[pdev_num]->pdev;
+	if ((index < MAX_XCP_PLATFORM_DEVICE) && (xcp_dev[index])) {
+		struct platform_device *pdev = xcp_dev[index]->pdev;
 
 		devres_release_group(&pdev->dev, NULL);
 		platform_device_unregister(pdev);
-		xcp_dev[pdev_num] = NULL;
+
+		xcp_dev[index] = NULL;
+		pdev_num--;
+	}
+}
+
+void amdgpu_xcp_drm_dev_free(struct drm_device *ddev)
+{
+	int8_t i;
+
+	guard(mutex)(&xcp_mutex);
+
+	for (i = 0; i < MAX_XCP_PLATFORM_DEVICE; i++) {
+		if ((xcp_dev[i]) && (&xcp_dev[i]->drm == ddev)) {
+			free_xcp_dev(i);
+			break;
+		}
+	}
+}
+EXPORT_SYMBOL(amdgpu_xcp_drm_dev_free);
+
+void amdgpu_xcp_drv_release(void)
+{
+	int8_t i;
+
+	guard(mutex)(&xcp_mutex);
+
+	for (i = 0; pdev_num && i < MAX_XCP_PLATFORM_DEVICE; i++) {
+		free_xcp_dev(i);
 	}
-	pdev_num = 0;
 }
 EXPORT_SYMBOL(amdgpu_xcp_drv_release);
 
diff --git a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h
index c1c4b679bf95..580a1602c8e3 100644
--- a/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h
+++ b/drivers/gpu/drm/amd/amdxcp/amdgpu_xcp_drv.h
@@ -25,5 +25,6 @@
 #define _AMDGPU_XCP_DRV_H_
 
 int amdgpu_xcp_drm_dev_alloc(struct drm_device **ddev);
+void amdgpu_xcp_drm_dev_free(struct drm_device *ddev);
 void amdgpu_xcp_drv_release(void);
 #endif /* _AMDGPU_XCP_DRV_H_ */

From de55cbff5ce93c316b0113535752e43079761f2c Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Thu, 17 Jul 2025 11:30:52 +0530
Subject: [PATCH 0235/1455] drm/amdgpu/vcn: Add regdump helper functions

Add generic helper functions for vcn devcoredump support
which can be re-used for all vcn versions.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 87 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h |  8 +++
 2 files changed, 95 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index f1f67521c29c..b497a6714138 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -92,6 +92,7 @@ MODULE_FIRMWARE(FIRMWARE_VCN5_0_0);
 MODULE_FIRMWARE(FIRMWARE_VCN5_0_1);
 
 static void amdgpu_vcn_idle_work_handler(struct work_struct *work);
+static void amdgpu_vcn_reg_dump_fini(struct amdgpu_device *adev);
 
 int amdgpu_vcn_early_init(struct amdgpu_device *adev, int i)
 {
@@ -285,6 +286,10 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i)
 		amdgpu_ucode_release(&adev->vcn.inst[0].fw);
 		adev->vcn.inst[i].fw = NULL;
 	}
+
+	if (adev->vcn.reg_list)
+		amdgpu_vcn_reg_dump_fini(adev);
+
 	mutex_destroy(&adev->vcn.inst[i].vcn_pg_lock);
 	mutex_destroy(&adev->vcn.inst[i].vcn1_jpeg1_workaround);
 
@@ -1527,3 +1532,85 @@ int amdgpu_vcn_ring_reset(struct amdgpu_ring *ring,
 
 	return amdgpu_vcn_reset_engine(adev, ring->me);
 }
+
+int amdgpu_vcn_reg_dump_init(struct amdgpu_device *adev,
+			     const struct amdgpu_hwip_reg_entry *reg, u32 count)
+{
+	adev->vcn.ip_dump = kcalloc(adev->vcn.num_vcn_inst * count,
+				     sizeof(uint32_t), GFP_KERNEL);
+	if (!adev->vcn.ip_dump)
+		return -ENOMEM;
+	adev->vcn.reg_list = reg;
+	adev->vcn.reg_count = count;
+
+	return 0;
+}
+
+static void amdgpu_vcn_reg_dump_fini(struct amdgpu_device *adev)
+{
+	kfree(adev->vcn.ip_dump);
+	adev->vcn.reg_list = NULL;
+	adev->vcn.reg_count = 0;
+}
+
+void amdgpu_vcn_dump_ip_state(struct amdgpu_ip_block *ip_block)
+{
+	struct amdgpu_device *adev = ip_block->adev;
+	int i, j;
+	bool is_powered;
+	u32 inst_off;
+
+	if (!adev->vcn.ip_dump)
+		return;
+
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+		if (adev->vcn.harvest_config & (1 << i))
+			continue;
+
+		inst_off = i * adev->vcn.reg_count;
+		/* mmUVD_POWER_STATUS is always readable and is the first in reg_list */
+		adev->vcn.ip_dump[inst_off] =
+			RREG32(SOC15_REG_ENTRY_OFFSET_INST(adev->vcn.reg_list[0], i));
+		is_powered = (adev->vcn.ip_dump[inst_off] &
+			      UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF) !=
+			      UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF;
+
+		if (is_powered)
+			for (j = 1; j < adev->vcn.reg_count; j++)
+				adev->vcn.ip_dump[inst_off + j] =
+				RREG32(SOC15_REG_ENTRY_OFFSET_INST(adev->vcn.reg_list[j], i));
+	}
+}
+
+void amdgpu_vcn_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
+{
+	struct amdgpu_device *adev = ip_block->adev;
+	int i, j;
+	bool is_powered;
+	u32 inst_off;
+
+	if (!adev->vcn.ip_dump)
+		return;
+
+	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+		if (adev->vcn.harvest_config & (1 << i)) {
+			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
+			continue;
+		}
+
+		inst_off = i * adev->vcn.reg_count;
+		is_powered = (adev->vcn.ip_dump[inst_off] &
+			      UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF) !=
+			      UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF;
+
+		if (is_powered) {
+			drm_printf(p, "\nActive Instance:VCN%d\n", i);
+			for (j = 0; j < adev->vcn.reg_count; j++)
+				drm_printf(p, "%-50s \t 0x%08x\n", adev->vcn.reg_list[j].reg_name,
+					   adev->vcn.ip_dump[inst_off + j]);
+		} else {
+			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
+		}
+	}
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 0bc0a94d7cf0..b3fb1d0e43fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -237,6 +237,8 @@
 
 #define AMDGPU_DRM_KEY_INJECT_WORKAROUND_VCNFW_ASD_HANDSHAKING 2
 
+struct amdgpu_hwip_reg_entry;
+
 enum amdgpu_vcn_caps {
 	AMDGPU_VCN_RRMT_ENABLED,
 };
@@ -362,6 +364,8 @@ struct amdgpu_vcn {
 
 	bool			workload_profile_active;
 	struct mutex            workload_profile_mutex;
+	u32 reg_count;
+	const struct amdgpu_hwip_reg_entry *reg_list;
 };
 
 struct amdgpu_fw_shared_rb_ptrs_struct {
@@ -557,4 +561,8 @@ int vcn_set_powergating_state(struct amdgpu_ip_block *ip_block,
 int amdgpu_vcn_ring_reset(struct amdgpu_ring *ring,
 			  unsigned int vmid,
 			  struct amdgpu_fence *guilty_fence);
+int amdgpu_vcn_reg_dump_init(struct amdgpu_device *adev,
+			     const struct amdgpu_hwip_reg_entry *reg, u32 count);
+void amdgpu_vcn_dump_ip_state(struct amdgpu_ip_block *ip_block);
+void amdgpu_vcn_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p);
 #endif

From b1b29aa88f5367d0367c8eeef643635bc6009a9a Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Mon, 28 Jul 2025 19:33:50 +0800
Subject: [PATCH 0236/1455] drm/amdgpu: Fix vcn v5.0.1 poison irq call trace

Why:
    [13014.890792] Call Trace:
    [13014.890793]  <TASK>
    [13014.890795]  ? show_trace_log_lvl+0x1d6/0x2ea
    [13014.890799]  ? show_trace_log_lvl+0x1d6/0x2ea
    [13014.890800]  ? vcn_v5_0_1_hw_fini+0xe9/0x110 [amdgpu]
    [13014.890872]  ? show_regs.part.0+0x23/0x29
    [13014.890873]  ? show_regs.cold+0x8/0xd
    [13014.890874]  ? amdgpu_irq_put+0xc6/0xe0 [amdgpu]
    [13014.890934]  ? __warn+0x8c/0x100
    [13014.890936]  ? amdgpu_irq_put+0xc6/0xe0 [amdgpu]
    [13014.890995]  ? report_bug+0xa4/0xd0
    [13014.890999]  ? handle_bug+0x39/0x90
    [13014.891001]  ? exc_invalid_op+0x19/0x70
    [13014.891003]  ? asm_exc_invalid_op+0x1b/0x20
    [13014.891005]  ? amdgpu_irq_put+0xc6/0xe0 [amdgpu]
    [13014.891065]  ? amdgpu_irq_put+0x63/0xe0 [amdgpu]
    [13014.891124]  vcn_v5_0_1_hw_fini+0xe9/0x110 [amdgpu]
    [13014.891189]  amdgpu_ip_block_hw_fini+0x3b/0x78 [amdgpu]
    [13014.891309]  amdgpu_device_fini_hw+0x3c1/0x479 [amdgpu]
How:
    Add omitted vcn poison irq get call.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c  |  7 +++++++
 2 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
index 54523dc1f702..03ec4b741d19 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -1058,6 +1058,11 @@ static int jpeg_v5_0_1_ras_late_init(struct amdgpu_device *adev, struct ras_comm
 	if (r)
 		return r;
 
+	r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__JPEG,
+				&jpeg_v5_0_1_aca_info, NULL);
+	if (r)
+		goto late_fini;
+
 	if (amdgpu_ras_is_supported(adev, ras_block->block) &&
 		adev->jpeg.inst->ras_poison_irq.funcs) {
 		r = amdgpu_irq_get(adev, &adev->jpeg.inst->ras_poison_irq, 0);
@@ -1065,11 +1070,6 @@ static int jpeg_v5_0_1_ras_late_init(struct amdgpu_device *adev, struct ras_comm
 			goto late_fini;
 	}
 
-	r = amdgpu_ras_bind_aca(adev, AMDGPU_RAS_BLOCK__JPEG,
-				&jpeg_v5_0_1_aca_info, NULL);
-	if (r)
-		goto late_fini;
-
 	return 0;
 
 late_fini:
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index d8bbb9376731..cb560d64da08 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -1608,6 +1608,13 @@ static int vcn_v5_0_1_ras_late_init(struct amdgpu_device *adev, struct ras_commo
 	if (r)
 		goto late_fini;
 
+	if (amdgpu_ras_is_supported(adev, ras_block->block) &&
+		adev->vcn.inst->ras_poison_irq.funcs) {
+		r = amdgpu_irq_get(adev, &adev->vcn.inst->ras_poison_irq, 0);
+		if (r)
+			goto late_fini;
+	}
+
 	return 0;
 
 late_fini:

From 08e27c9d9209ec4daaba3aa9b7c69e9c75be3eee Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Mon, 28 Jul 2025 19:49:24 +0800
Subject: [PATCH 0237/1455] drm/amdgpu: Add new error code for VCN/JPEG new
 chain

Add VIDS and JPEG8/9 S|D chain error code for VCN/JPEG v5.0.1.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c | 5 +++--
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c  | 2 +-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
index 03ec4b741d19..8f389cb2f6c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -1016,8 +1016,9 @@ static int jpeg_v5_0_1_aca_bank_parser(struct aca_handle *handle, struct aca_ban
 
 /* reference to smu driver if header file */
 static int jpeg_v5_0_1_err_codes[] = {
-	16, 17, 18, 19, 20, 21, 22, 23, /* JPEG[0-7][S|D] */
-	24, 25, 26, 27, 28, 29, 30, 31
+	16, 17, 18, 19, 20, 21, 22, 23, /* JPEG[0-9][S|D] */
+	24, 25, 26, 27, 28, 29, 30, 31,
+	48, 49, 50, 51,
 };
 
 static bool jpeg_v5_0_1_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index cb560d64da08..74f811fed79a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -1562,7 +1562,7 @@ static int vcn_v5_0_1_aca_bank_parser(struct aca_handle *handle, struct aca_bank
 
 /* reference to smu driver if header file */
 static int vcn_v5_0_1_err_codes[] = {
-	14, 15, /* VCN */
+	14, 15, 47, /* VCN [D|V|S] */
 };
 
 static bool vcn_v5_0_1_aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank,

From f4c3be28d55cb544133f0d8bff2564041dba719e Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 00:27:50 +0530
Subject: [PATCH 0238/1455] drm/amdgpu/vcn: Register dump cleanup in VCN5

Use generic vcn devcoredump helper functions for VCN5

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 86 ++-----------------------
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h |  5 --
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 44 +++++++++++--
 3 files changed, 44 insertions(+), 91 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index f9dbd2757d83..455f829b8bb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -115,21 +115,6 @@ static int vcn_v5_0_0_early_init(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
-void vcn_v5_0_0_alloc_ip_dump(struct amdgpu_device *adev)
-{
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0);
-	uint32_t *ptr;
-
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (!ptr) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
-}
-
 /**
  * vcn_v5_0_0_sw_init - sw init for VCN block
  *
@@ -201,7 +186,9 @@ static int vcn_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block)
 	if (!amdgpu_sriov_vf(adev))
 		adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
 
-	vcn_v5_0_0_alloc_ip_dump(adev);
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_5_0, ARRAY_SIZE(vcn_reg_list_5_0));
+	if (r)
+		return r;
 
 	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
 	if (r)
@@ -251,8 +238,6 @@ static int vcn_v5_0_0_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	kfree(adev->vcn.ip_dump);
-
 	return 0;
 }
 
@@ -1434,67 +1419,6 @@ static void vcn_v5_0_0_set_irq_funcs(struct amdgpu_device *adev)
 	}
 }
 
-void vcn_v5_0_0_print_ip_state(struct amdgpu_ip_block *ip_block,
-			       struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0);
-	uint32_t inst_off, is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_5_0[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-void vcn_v5_0_0_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, regUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_5_0[j], i));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v5_0_0_ip_funcs = {
 	.name = "vcn_v5_0_0",
 	.early_init = vcn_v5_0_0_early_init,
@@ -1508,8 +1432,8 @@ static const struct amd_ip_funcs vcn_v5_0_0_ip_funcs = {
 	.wait_for_idle = vcn_v5_0_0_wait_for_idle,
 	.set_clockgating_state = vcn_v5_0_0_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v5_0_0_dump_ip_state,
-	.print_ip_state = vcn_v5_0_0_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v5_0_0_ip_block = {
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h
index b8927652bc50..51bbccd4360f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.h
@@ -32,11 +32,6 @@
 #define VCN_VID_IP_ADDRESS                         0x0
 #define VCN_AON_IP_ADDRESS                         0x30000
 
-void vcn_v5_0_0_alloc_ip_dump(struct amdgpu_device *adev);
-void vcn_v5_0_0_print_ip_state(struct amdgpu_ip_block *ip_block,
-			       struct drm_printer *p);
-void vcn_v5_0_0_dump_ip_state(struct amdgpu_ip_block *ip_block);
-
 extern const struct amdgpu_ip_block_version vcn_v5_0_0_ip_block;
 
 #endif /* __VCN_V5_0_0_H__ */
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 74f811fed79a..02a0516edabe 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -40,6 +40,40 @@
 
 #include <drm/drm_drv.h>
 
+static const struct amdgpu_hwip_reg_entry vcn_reg_list_5_0_1[] = {
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_POWER_STATUS),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_STATUS),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_CONTEXT_ID),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_CONTEXT_ID2),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_GPCOM_VCPU_DATA0),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_GPCOM_VCPU_DATA1),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_GPCOM_VCPU_CMD),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI2),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO2),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI3),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO3),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_HI4),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_BASE_LO4),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR2),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR2),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR3),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR3),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_RPTR4),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_WPTR4),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE2),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE3),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_RB_SIZE4),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_LMA_CTL),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_LMA_DATA),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_LMA_MASK),
+	SOC15_REG_ENTRY_STR(VCN, 0, regUVD_DPG_PAUSE)
+};
+
 static int vcn_v5_0_1_start_sriov(struct amdgpu_device *adev);
 static void vcn_v5_0_1_set_unified_ring_funcs(struct amdgpu_device *adev);
 static void vcn_v5_0_1_set_irq_funcs(struct amdgpu_device *adev);
@@ -163,7 +197,9 @@ static int vcn_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	vcn_v5_0_0_alloc_ip_dump(adev);
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_5_0_1, ARRAY_SIZE(vcn_reg_list_5_0_1));
+	if (r)
+		return r;
 
 	return amdgpu_vcn_sysfs_reset_mask_init(adev);
 }
@@ -209,8 +245,6 @@ static int vcn_v5_0_1_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	amdgpu_vcn_sysfs_reset_mask_fini(adev);
 
-	kfree(adev->vcn.ip_dump);
-
 	return 0;
 }
 
@@ -1480,8 +1514,8 @@ static const struct amd_ip_funcs vcn_v5_0_1_ip_funcs = {
 	.post_soft_reset = NULL,
 	.set_clockgating_state = vcn_v5_0_1_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v5_0_0_dump_ip_state,
-	.print_ip_state = vcn_v5_0_0_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v5_0_1_ip_block = {

From 4e011af912fb51e1b0d12f22e6d6b883b6a700e7 Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 12:18:30 +0530
Subject: [PATCH 0239/1455] drm/amdgpu/vcn: Register dump cleanup in VCN4_0_0

Use generic vcn devcoredump helper functions for VCN4_0_0

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 80 ++-------------------------
 1 file changed, 5 insertions(+), 75 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index e9f9cb169034..1785786a72f8 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -183,8 +183,6 @@ static int vcn_v4_0_sw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_ring *ring;
 	struct amdgpu_device *adev = ip_block->adev;
 	int i, r;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0);
-	uint32_t *ptr;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 		if (adev->vcn.harvest_config & (1 << i))
@@ -255,14 +253,9 @@ static int vcn_v4_0_sw_init(struct amdgpu_ip_block *ip_block)
 	if (r)
 		return r;
 
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (!ptr) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_4_0, ARRAY_SIZE(vcn_reg_list_4_0));
+	if (r)
+		return r;
 
 	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
 	if (r)
@@ -315,8 +308,6 @@ static int vcn_v4_0_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	kfree(adev->vcn.ip_dump);
-
 	return 0;
 }
 
@@ -2252,67 +2243,6 @@ static void vcn_v4_0_set_irq_funcs(struct amdgpu_device *adev)
 	}
 }
 
-static void vcn_v4_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0);
-	uint32_t inst_off, is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_4_0[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-static void vcn_v4_0_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, regUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_4_0[j],
-									   i));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v4_0_ip_funcs = {
 	.name = "vcn_v4_0",
 	.early_init = vcn_v4_0_early_init,
@@ -2326,8 +2256,8 @@ static const struct amd_ip_funcs vcn_v4_0_ip_funcs = {
 	.wait_for_idle = vcn_v4_0_wait_for_idle,
 	.set_clockgating_state = vcn_v4_0_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v4_0_dump_ip_state,
-	.print_ip_state = vcn_v4_0_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v4_0_ip_block = {

From 793b97c4ad33a3c35739c8ac59ecfdae22a4cb18 Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 12:38:49 +0530
Subject: [PATCH 0240/1455] drm/amdgpu/vcn: Register dump cleanup in VCN4_0_5

Use generic vcn devcoredump helper functions for VCN4_0_5

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 80 ++-----------------------
 1 file changed, 5 insertions(+), 75 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index ed57e6431d4b..f785467370d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -147,9 +147,6 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_ring *ring;
 	struct amdgpu_device *adev = ip_block->adev;
 	int i, r;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5);
-	uint32_t *ptr;
-
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
 		volatile struct amdgpu_vcn4_fw_shared *fw_shared;
@@ -233,15 +230,9 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (!ptr) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
-	return 0;
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_4_0_5, ARRAY_SIZE(vcn_reg_list_4_0_5));
+
+	return r;
 }
 
 /**
@@ -1710,67 +1701,6 @@ static void vcn_v4_0_5_set_irq_funcs(struct amdgpu_device *adev)
 	}
 }
 
-static void vcn_v4_0_5_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5);
-	uint32_t inst_off, is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_4_0_5[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-static void vcn_v4_0_5_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, regUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_4_0_5[j],
-									   i));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v4_0_5_ip_funcs = {
 	.name = "vcn_v4_0_5",
 	.early_init = vcn_v4_0_5_early_init,
@@ -1784,8 +1714,8 @@ static const struct amd_ip_funcs vcn_v4_0_5_ip_funcs = {
 	.wait_for_idle = vcn_v4_0_5_wait_for_idle,
 	.set_clockgating_state = vcn_v4_0_5_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v4_0_5_dump_ip_state,
-	.print_ip_state = vcn_v4_0_5_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v4_0_5_ip_block = {

From 69cc37647b3ba8aa8b4fdbde34598ce56281bf0d Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 12:57:04 +0530
Subject: [PATCH 0241/1455] drm/amdgpu/vcn: Register dump cleanup in VCN4_0_3

Use generic vcn devcoredump helper functions for VCN4_0_3

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 79 ++-----------------------
 1 file changed, 5 insertions(+), 74 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index b904afc358ae..c51dc401226a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -160,8 +160,6 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	struct amdgpu_ring *ring;
 	int i, r, vcn_inst;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3);
-	uint32_t *ptr;
 
 	/* VCN DEC TRAP */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
@@ -231,14 +229,9 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block)
 		}
 	}
 
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (!ptr) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_4_0_3, ARRAY_SIZE(vcn_reg_list_4_0_3));
+	if (r)
+		return r;
 
 	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
 	if (r)
@@ -1877,68 +1870,6 @@ static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 	adev->vcn.inst->ras_poison_irq.funcs = &vcn_v4_0_3_ras_irq_funcs;
 }
 
-static void vcn_v4_0_3_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3);
-	uint32_t inst_off, is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_4_0_3[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-static void vcn_v4_0_3_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off, inst_id;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_id = GET_INST(VCN, i);
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, inst_id, regUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_4_0_3[j],
-									   inst_id));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = {
 	.name = "vcn_v4_0_3",
 	.early_init = vcn_v4_0_3_early_init,
@@ -1952,8 +1883,8 @@ static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = {
 	.wait_for_idle = vcn_v4_0_3_wait_for_idle,
 	.set_clockgating_state = vcn_v4_0_3_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v4_0_3_dump_ip_state,
-	.print_ip_state = vcn_v4_0_3_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v4_0_3_ip_block = {

From b2d532b588b313e882ee9b9169c9f31f743747dc Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 13:08:00 +0530
Subject: [PATCH 0242/1455] drm/amdgpu/vcn: Register dump cleanup in VCN3_0

Use generic vcn devcoredump helper functions for VCN3_0

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 78 ++-------------------------
 1 file changed, 5 insertions(+), 73 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index a89662c97c9e..ff2a85619f23 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -175,8 +175,6 @@ static int vcn_v3_0_sw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_ring *ring;
 	int i, j, r;
 	int vcn_doorbell_index = 0;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0);
-	uint32_t *ptr;
 	struct amdgpu_device *adev = ip_block->adev;
 
 	/*
@@ -304,14 +302,9 @@ static int vcn_v3_0_sw_init(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (ptr == NULL) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_3_0, ARRAY_SIZE(vcn_reg_list_3_0));
+	if (r)
+		return r;
 
 	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
 	if (r)
@@ -2348,67 +2341,6 @@ static void vcn_v3_0_set_irq_funcs(struct amdgpu_device *adev)
 	}
 }
 
-static void vcn_v3_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0);
-	uint32_t inst_off;
-	bool is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-			      UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_3_0[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-static void vcn_v3_0_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, mmUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-			      UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_3_0[j], i));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v3_0_ip_funcs = {
 	.name = "vcn_v3_0",
 	.early_init = vcn_v3_0_early_init,
@@ -2422,8 +2354,8 @@ static const struct amd_ip_funcs vcn_v3_0_ip_funcs = {
 	.wait_for_idle = vcn_v3_0_wait_for_idle,
 	.set_clockgating_state = vcn_v3_0_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v3_0_dump_ip_state,
-	.print_ip_state = vcn_v3_0_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v3_0_ip_block = {

From 53c4be7a59dfbc5f6581727e536b62a322a21a36 Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 13:15:00 +0530
Subject: [PATCH 0243/1455] drm/amdgpu/vcn: Register dump cleanup in VCN2_0_0

Use generic vcn devcoredump helper functions for VCN2_0_0

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c | 79 ++-------------------------
 1 file changed, 5 insertions(+), 74 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index d1481e6d57ec..b115137ab2d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -136,8 +136,6 @@ static int vcn_v2_0_sw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_ring *ring;
 	int i, r;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0);
-	uint32_t *ptr;
 	struct amdgpu_device *adev = ip_block->adev;
 	volatile struct amdgpu_fw_shared *fw_shared;
 
@@ -232,14 +230,9 @@ static int vcn_v2_0_sw_init(struct amdgpu_ip_block *ip_block)
 	if (amdgpu_vcnfw_log)
 		amdgpu_vcn_fwlog_init(adev->vcn.inst);
 
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (!ptr) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_2_0, ARRAY_SIZE(vcn_reg_list_2_0));
+	if (r)
+		return r;
 
 	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
 	if (r)
@@ -276,8 +269,6 @@ static int vcn_v2_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	r = amdgpu_vcn_sw_fini(adev, 0);
 
-	kfree(adev->vcn.ip_dump);
-
 	return r;
 }
 
@@ -2101,66 +2092,6 @@ static int vcn_v2_0_start_sriov(struct amdgpu_device *adev)
 	return vcn_v2_0_start_mmsch(adev, &adev->virt.mm_table);
 }
 
-static void vcn_v2_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0);
-	uint32_t inst_off, is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_2_0[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-static void vcn_v2_0_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, mmUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_2_0[j], i));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v2_0_ip_funcs = {
 	.name = "vcn_v2_0",
 	.early_init = vcn_v2_0_early_init,
@@ -2174,8 +2105,8 @@ static const struct amd_ip_funcs vcn_v2_0_ip_funcs = {
 	.wait_for_idle = vcn_v2_0_wait_for_idle,
 	.set_clockgating_state = vcn_v2_0_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v2_0_dump_ip_state,
-	.print_ip_state = vcn_v2_0_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 static const struct amdgpu_ring_funcs vcn_v2_0_dec_ring_vm_funcs = {

From 26a63590fea63e0b462f3db6860fd454b088897b Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Fri, 18 Jul 2025 13:23:53 +0530
Subject: [PATCH 0244/1455] drm/amdgpu/vcn: Register dump cleanup in VCN2_5

Use generic vcn devcoredump helper functions for VCN2_5 and VCN2_6

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Acked-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 83 +++------------------------
 1 file changed, 7 insertions(+), 76 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index d7b2668ab0d9..3a7c137a83ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -297,8 +297,6 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_ring *ring;
 	int i, j, r;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5);
-	uint32_t *ptr;
 	struct amdgpu_device *adev = ip_block->adev;
 
 	for (j = 0; j < adev->vcn.num_vcn_inst; j++) {
@@ -423,14 +421,9 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block)
 	if (r)
 		return r;
 
-	/* Allocate memory for VCN IP Dump buffer */
-	ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL);
-	if (!ptr) {
-		DRM_ERROR("Failed to allocate memory for VCN IP Dump\n");
-		adev->vcn.ip_dump = NULL;
-	} else {
-		adev->vcn.ip_dump = ptr;
-	}
+	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_2_5, ARRAY_SIZE(vcn_reg_list_2_5));
+	if (r)
+		return r;
 
 	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
 	if (r)
@@ -477,8 +470,6 @@ static int vcn_v2_5_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	kfree(adev->vcn.ip_dump);
-
 	return 0;
 }
 
@@ -2133,66 +2124,6 @@ static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev)
 	}
 }
 
-static void vcn_v2_5_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5);
-	uint32_t inst_off, is_powered;
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	drm_printf(p, "num_instances:%d\n", adev->vcn.num_vcn_inst);
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i)) {
-			drm_printf(p, "\nHarvested Instance:VCN%d Skipping dump\n", i);
-			continue;
-		}
-
-		inst_off = i * reg_count;
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered) {
-			drm_printf(p, "\nActive Instance:VCN%d\n", i);
-			for (j = 0; j < reg_count; j++)
-				drm_printf(p, "%-50s \t 0x%08x\n", vcn_reg_list_2_5[j].reg_name,
-					   adev->vcn.ip_dump[inst_off + j]);
-		} else {
-			drm_printf(p, "\nInactive Instance:VCN%d\n", i);
-		}
-	}
-}
-
-static void vcn_v2_5_dump_ip_state(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int i, j;
-	bool is_powered;
-	uint32_t inst_off;
-	uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5);
-
-	if (!adev->vcn.ip_dump)
-		return;
-
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		if (adev->vcn.harvest_config & (1 << i))
-			continue;
-
-		inst_off = i * reg_count;
-		/* mmUVD_POWER_STATUS is always readable and is first element of the array */
-		adev->vcn.ip_dump[inst_off] = RREG32_SOC15(VCN, i, mmUVD_POWER_STATUS);
-		is_powered = (adev->vcn.ip_dump[inst_off] &
-				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK) != 1;
-
-		if (is_powered)
-			for (j = 1; j < reg_count; j++)
-				adev->vcn.ip_dump[inst_off + j] =
-					RREG32(SOC15_REG_ENTRY_OFFSET_INST(vcn_reg_list_2_5[j], i));
-	}
-}
-
 static const struct amd_ip_funcs vcn_v2_5_ip_funcs = {
 	.name = "vcn_v2_5",
 	.early_init = vcn_v2_5_early_init,
@@ -2206,8 +2137,8 @@ static const struct amd_ip_funcs vcn_v2_5_ip_funcs = {
 	.wait_for_idle = vcn_v2_5_wait_for_idle,
 	.set_clockgating_state = vcn_v2_5_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v2_5_dump_ip_state,
-	.print_ip_state = vcn_v2_5_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 static const struct amd_ip_funcs vcn_v2_6_ip_funcs = {
@@ -2223,8 +2154,8 @@ static const struct amd_ip_funcs vcn_v2_6_ip_funcs = {
         .wait_for_idle = vcn_v2_5_wait_for_idle,
         .set_clockgating_state = vcn_v2_5_set_clockgating_state,
 	.set_powergating_state = vcn_set_powergating_state,
-	.dump_ip_state = vcn_v2_5_dump_ip_state,
-	.print_ip_state = vcn_v2_5_print_ip_state,
+	.dump_ip_state = amdgpu_vcn_dump_ip_state,
+	.print_ip_state = amdgpu_vcn_print_ip_state,
 };
 
 const struct amdgpu_ip_block_version vcn_v2_5_ip_block =

From 2a2681eda73b99a2c1ee8cdb006099ea5d0c2505 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 18 Jul 2025 15:52:04 -0400
Subject: [PATCH 0245/1455] drm/amdgpu: update mmhub 3.0.1 client id mappings

Update the client id mapping so the correct clients
get printed when there is a mmhub page fault.

Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c | 57 +++++++++++++----------
 1 file changed, 32 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
index 134c4ec10887..910337dc28d1 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c
@@ -36,40 +36,47 @@
 
 static const char *mmhub_client_ids_v3_0_1[][2] = {
 	[0][0] = "VMC",
+	[1][0] = "ISPXT",
+	[2][0] = "ISPIXT",
 	[4][0] = "DCEDMC",
 	[5][0] = "DCEVGA",
 	[6][0] = "MP0",
 	[7][0] = "MP1",
-	[8][0] = "MPIO",
-	[16][0] = "HDP",
-	[17][0] = "LSDMA",
-	[18][0] = "JPEG",
-	[19][0] = "VCNU0",
-	[21][0] = "VSCH",
-	[22][0] = "VCNU1",
-	[23][0] = "VCN1",
-	[32+20][0] = "VCN0",
-	[2][1] = "DBGUNBIO",
+	[8][0] = "MPM",
+	[12][0] = "ISPTNR",
+	[14][0] = "ISPCRD0",
+	[15][0] = "ISPCRD1",
+	[16][0] = "ISPCRD2",
+	[22][0] = "HDP",
+	[23][0] = "LSDMA",
+	[24][0] = "JPEG",
+	[27][0] = "VSCH",
+	[28][0] = "VCNU",
+	[29][0] = "VCN",
+	[1][1] = "ISPXT",
+	[2][1] = "ISPIXT",
 	[3][1] = "DCEDWB",
 	[4][1] = "DCEDMC",
 	[5][1] = "DCEVGA",
 	[6][1] = "MP0",
 	[7][1] = "MP1",
-	[8][1] = "MPIO",
-	[10][1] = "DBGU0",
-	[11][1] = "DBGU1",
-	[12][1] = "DBGU2",
-	[13][1] = "DBGU3",
-	[14][1] = "XDP",
-	[15][1] = "OSSSYS",
-	[16][1] = "HDP",
-	[17][1] = "LSDMA",
-	[18][1] = "JPEG",
-	[19][1] = "VCNU0",
-	[20][1] = "VCN0",
-	[21][1] = "VSCH",
-	[22][1] = "VCNU1",
-	[23][1] = "VCN1",
+	[8][1] = "MPM",
+	[10][1] = "ISPMWR0",
+	[11][1] = "ISPMWR1",
+	[12][1] = "ISPTNR",
+	[13][1] = "ISPSWR",
+	[14][1] = "ISPCWR0",
+	[15][1] = "ISPCWR1",
+	[16][1] = "ISPCWR2",
+	[17][1] = "ISPCWR3",
+	[18][1] = "XDP",
+	[21][1] = "OSSSYS",
+	[22][1] = "HDP",
+	[23][1] = "LSDMA",
+	[24][1] = "JPEG",
+	[27][1] = "VSCH",
+	[28][1] = "VCNU",
+	[29][1] = "VCN",
 };
 
 static uint32_t mmhub_v3_0_1_get_invalidate_req(unsigned int vmid,

From e932f4779a2d329841bb9ca70bb80a4bb2d707b6 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 18 Jul 2025 15:53:21 -0400
Subject: [PATCH 0246/1455] drm/amdgpu: update mmhub 3.3 client id mappings

Update the client id mapping so the correct clients
get printed when there is a mmhub page fault.

v2: fix typos spotted by David Wu.
v3: fix additional typo spotted by David.

Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c | 121 ++++++++++++++++++++++--
 1 file changed, 112 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c
index bc3d6c2fc87a..f6fc9778bc30 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_3.c
@@ -39,6 +39,64 @@
 #define regDAGB1_L1TLB_REG_RW_3_3_BASE_IDX          1
 
 static const char *mmhub_client_ids_v3_3[][2] = {
+	[0][0] = "VMC",
+	[1][0] = "ISPXT",
+	[2][0] = "ISPIXT",
+	[4][0] = "DCEDMC",
+	[6][0] = "MP0",
+	[7][0] = "MP1",
+	[8][0] = "MPM",
+	[9][0] = "ISPPDPRD",
+	[10][0] = "ISPCSTATRD",
+	[11][0] = "ISPBYRPRD",
+	[12][0] = "ISPRGBPRD",
+	[13][0] = "ISPMCFPRD",
+	[14][0] = "ISPMCFPRD1",
+	[15][0] = "ISPYUVPRD",
+	[16][0] = "ISPMCSCRD",
+	[17][0] = "ISPGDCRD",
+	[18][0] = "ISPLMERD",
+	[22][0] = "ISPXT1",
+	[23][0] = "ISPIXT1",
+	[24][0] = "HDP",
+	[25][0] = "LSDMA",
+	[26][0] = "JPEG",
+	[27][0] = "VPE",
+	[28][0] = "VSCH",
+	[29][0] = "VCNU",
+	[30][0] = "VCN",
+	[1][1] = "ISPXT",
+	[2][1] = "ISPIXT",
+	[3][1] = "DCEDWB",
+	[4][1] = "DCEDMC",
+	[5][1] = "ISPCSISWR",
+	[6][1] = "MP0",
+	[7][1] = "MP1",
+	[8][1] = "MPM",
+	[9][1] = "ISPPDPWR",
+	[10][1] = "ISPCSTATWR",
+	[11][1] = "ISPBYRPWR",
+	[12][1] = "ISPRGBPWR",
+	[13][1] = "ISPMCFPWR",
+	[14][1] = "ISPMWR0",
+	[15][1] = "ISPYUVPWR",
+	[16][1] = "ISPMCSCWR",
+	[17][1] = "ISPGDCWR",
+	[18][1] = "ISPLMEWR",
+	[20][1] = "ISPMWR2",
+	[21][1] = "OSSSYS",
+	[22][1] = "ISPXT1",
+	[23][1] = "ISPIXT1",
+	[24][1] = "HDP",
+	[25][1] = "LSDMA",
+	[26][1] = "JPEG",
+	[27][1] = "VPE",
+	[28][1] = "VSCH",
+	[29][1] = "VCNU",
+	[30][1] = "VCN",
+};
+
+static const char *mmhub_client_ids_v3_3_1[][2] = {
 	[0][0] = "VMC",
 	[4][0] = "DCEDMC",
 	[6][0] = "MP0",
@@ -46,10 +104,29 @@ static const char *mmhub_client_ids_v3_3[][2] = {
 	[8][0] = "MPM",
 	[24][0] = "HDP",
 	[25][0] = "LSDMA",
-	[26][0] = "JPEG",
-	[27][0] = "VPE",
-	[29][0] = "VCNU",
-	[30][0] = "VCN",
+	[26][0] = "JPEG0",
+	[27][0] = "VPE0",
+	[28][0] = "VSCH",
+	[29][0] = "VCNU0",
+	[30][0] = "VCN0",
+	[32+1][0] = "ISPXT",
+	[32+2][0] = "ISPIXT",
+	[32+9][0] = "ISPPDPRD",
+	[32+10][0] = "ISPCSTATRD",
+	[32+11][0] = "ISPBYRPRD",
+	[32+12][0] = "ISPRGBPRD",
+	[32+13][0] = "ISPMCFPRD",
+	[32+14][0] = "ISPMCFPRD1",
+	[32+15][0] = "ISPYUVPRD",
+	[32+16][0] = "ISPMCSCRD",
+	[32+17][0] = "ISPGDCRD",
+	[32+18][0] = "ISPLMERD",
+	[32+22][0] = "ISPXT1",
+	[32+23][0] = "ISPIXT1",
+	[32+26][0] = "JPEG1",
+	[32+27][0] = "VPE1",
+	[32+29][0] = "VCNU1",
+	[32+30][0] = "VCN1",
 	[3][1] = "DCEDWB",
 	[4][1] = "DCEDMC",
 	[6][1] = "MP0",
@@ -58,10 +135,32 @@ static const char *mmhub_client_ids_v3_3[][2] = {
 	[21][1] = "OSSSYS",
 	[24][1] = "HDP",
 	[25][1] = "LSDMA",
-	[26][1] = "JPEG",
-	[27][1] = "VPE",
-	[29][1] = "VCNU",
-	[30][1] = "VCN",
+	[26][1] = "JPEG0",
+	[27][1] = "VPE0",
+	[28][1] = "VSCH",
+	[29][1] = "VCNU0",
+	[30][1] = "VCN0",
+	[32+1][1] = "ISPXT",
+	[32+2][1] = "ISPIXT",
+	[32+5][1] = "ISPCSISWR",
+	[32+9][1] = "ISPPDPWR",
+	[32+10][1] = "ISPCSTATWR",
+	[32+11][1] = "ISPBYRPWR",
+	[32+12][1] = "ISPRGBPWR",
+	[32+13][1] = "ISPMCFPWR",
+	[32+14][1] = "ISPMWR0",
+	[32+15][1] = "ISPYUVPWR",
+	[32+16][1] = "ISPMCSCWR",
+	[32+17][1] = "ISPGDCWR",
+	[32+18][1] = "ISPLMEWR",
+	[32+19][1] = "ISPMWR1",
+	[32+20][1] = "ISPMWR2",
+	[32+22][1] = "ISPXT1",
+	[32+23][1] = "ISPIXT1",
+	[32+26][1] = "JPEG1",
+	[32+27][1] = "VPE1",
+	[32+29][1] = "VCNU1",
+	[32+30][1] = "VCN1",
 };
 
 static uint32_t mmhub_v3_3_get_invalidate_req(unsigned int vmid,
@@ -102,12 +201,16 @@ mmhub_v3_3_print_l2_protection_fault_status(struct amdgpu_device *adev,
 
 	switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) {
 	case IP_VERSION(3, 3, 0):
-	case IP_VERSION(3, 3, 1):
 	case IP_VERSION(3, 3, 2):
 		mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v3_3) ?
 			    mmhub_client_ids_v3_3[cid][rw] :
 			    cid == 0x140 ? "UMSCH" : NULL;
 		break;
+	case IP_VERSION(3, 3, 1):
+		mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v3_3_1) ?
+			    mmhub_client_ids_v3_3_1[cid][rw] :
+			    cid == 0x140 ? "UMSCH" : NULL;
+		break;
 	default:
 		mmhub_cid = NULL;
 		break;

From 3cf06bd4cf2512d564fdb451b07de0cebe7b138d Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 27 Jun 2025 10:10:31 -0400
Subject: [PATCH 0247/1455] drm/amd/display: add more cyan skillfish devices

Add PCI IDs to support display probe for cyan skillfish
family of SOCs.

Acked-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 8 +++++++-
 drivers/gpu/drm/amd/display/include/dal_asic_id.h | 5 +++++
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 4d6181e7c612..d712548b1927 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -165,7 +165,13 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id)
 
 	case FAMILY_NV:
 		dc_version = DCN_VERSION_2_0;
-		if (asic_id.chip_id == DEVICE_ID_NV_13FE || asic_id.chip_id == DEVICE_ID_NV_143F) {
+		if (asic_id.chip_id == DEVICE_ID_NV_13FE ||
+		    asic_id.chip_id == DEVICE_ID_NV_143F ||
+		    asic_id.chip_id == DEVICE_ID_NV_13F9 ||
+		    asic_id.chip_id == DEVICE_ID_NV_13FA ||
+		    asic_id.chip_id == DEVICE_ID_NV_13FB ||
+		    asic_id.chip_id == DEVICE_ID_NV_13FC ||
+		    asic_id.chip_id == DEVICE_ID_NV_13DB) {
 			dc_version = DCN_VERSION_2_01;
 			break;
 		}
diff --git a/drivers/gpu/drm/amd/display/include/dal_asic_id.h b/drivers/gpu/drm/amd/display/include/dal_asic_id.h
index 5fc29164e4b4..8aea50aa9533 100644
--- a/drivers/gpu/drm/amd/display/include/dal_asic_id.h
+++ b/drivers/gpu/drm/amd/display/include/dal_asic_id.h
@@ -213,6 +213,11 @@ enum {
 #endif
 #define DEVICE_ID_NV_13FE 0x13FE  // CYAN_SKILLFISH
 #define DEVICE_ID_NV_143F 0x143F
+#define DEVICE_ID_NV_13F9 0x13F9
+#define DEVICE_ID_NV_13FA 0x13FA
+#define DEVICE_ID_NV_13FB 0x13FB
+#define DEVICE_ID_NV_13FC 0x13FC
+#define DEVICE_ID_NV_13DB 0x13DB
 #define FAMILY_VGH 144
 #define DEVICE_ID_VGH_163F 0x163F
 #define DEVICE_ID_VGH_1435 0x1435

From 4d22db6d070ed3934f02ed15391283f6feb258ad Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 14 Jul 2025 10:16:25 +0200
Subject: [PATCH 0248/1455] drm/amdgpu: fix link error for !PM_SLEEP

When power management is not enabled in the kernel build, the newly
added hibernation changes cause a link failure:

arm-linux-gnueabi-ld: drivers/gpu/drm/amd/amdgpu/amdgpu_drv.o: in function `amdgpu_pmops_thaw':
amdgpu_drv.c:(.text+0x1514): undefined reference to `pm_hibernate_is_recovering'

Make the power management code in this driver conditional on
CONFIG_PM and CONFIG_PM_SLEEP

Fixes: 530694f54dd5 ("drm/amdgpu: do not resume device in thaw for normal hibernation")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20250714081635.4071570-1-arnd@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 8e8b7d2c3f42..141ba1f100d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2968,15 +2968,15 @@ out:
 }
 
 static const struct dev_pm_ops amdgpu_pm_ops = {
-	.prepare = amdgpu_pmops_prepare,
-	.complete = amdgpu_pmops_complete,
-	.suspend = amdgpu_pmops_suspend,
-	.suspend_noirq = amdgpu_pmops_suspend_noirq,
-	.resume = amdgpu_pmops_resume,
-	.freeze = amdgpu_pmops_freeze,
-	.thaw = amdgpu_pmops_thaw,
-	.poweroff = amdgpu_pmops_poweroff,
-	.restore = amdgpu_pmops_restore,
+	.prepare = pm_sleep_ptr(amdgpu_pmops_prepare),
+	.complete = pm_sleep_ptr(amdgpu_pmops_complete),
+	.suspend = pm_sleep_ptr(amdgpu_pmops_suspend),
+	.suspend_noirq = pm_sleep_ptr(amdgpu_pmops_suspend_noirq),
+	.resume = pm_sleep_ptr(amdgpu_pmops_resume),
+	.freeze = pm_sleep_ptr(amdgpu_pmops_freeze),
+	.thaw = pm_sleep_ptr(amdgpu_pmops_thaw),
+	.poweroff = pm_sleep_ptr(amdgpu_pmops_poweroff),
+	.restore = pm_sleep_ptr(amdgpu_pmops_restore),
 	.runtime_suspend = amdgpu_pmops_runtime_suspend,
 	.runtime_resume = amdgpu_pmops_runtime_resume,
 	.runtime_idle = amdgpu_pmops_runtime_idle,
@@ -3121,7 +3121,7 @@ static struct pci_driver amdgpu_kms_pci_driver = {
 	.probe = amdgpu_pci_probe,
 	.remove = amdgpu_pci_remove,
 	.shutdown = amdgpu_pci_shutdown,
-	.driver.pm = &amdgpu_pm_ops,
+	.driver.pm = pm_ptr(&amdgpu_pm_ops),
 	.err_handler = &amdgpu_pci_err_handler,
 	.dev_groups = amdgpu_sysfs_groups,
 };

From 8e8e08c831f088ed581444c58a635c49ea1222ab Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Wed, 30 Jul 2025 11:07:43 +0800
Subject: [PATCH 0249/1455] drm/amdgpu: Skip poison aca bank from UE channel

Avoid GFX poison consumption errors logged when fatal error occurs.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c | 51 +++++++++++++++----------
 1 file changed, 30 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
index cbc40cad581b..d1e431818212 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
@@ -130,6 +130,27 @@ static void aca_smu_bank_dump(struct amdgpu_device *adev, int idx, int total, st
 		RAS_EVENT_LOG(adev, event_id, HW_ERR "hardware error logged by the scrubber\n");
 }
 
+static bool aca_bank_hwip_is_matched(struct aca_bank *bank, enum aca_hwip_type type)
+{
+
+	struct aca_hwip *hwip;
+	int hwid, mcatype;
+	u64 ipid;
+
+	if (!bank || type == ACA_HWIP_TYPE_UNKNOW)
+		return false;
+
+	hwip = &aca_hwid_mcatypes[type];
+	if (!hwip->hwid)
+		return false;
+
+	ipid = bank->regs[ACA_REG_IDX_IPID];
+	hwid = ACA_REG__IPID__HARDWAREID(ipid);
+	mcatype = ACA_REG__IPID__MCATYPE(ipid);
+
+	return hwip->hwid == hwid && hwip->mcatype == mcatype;
+}
+
 static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_type type,
 				       int start, int count,
 				       struct aca_banks *banks, struct ras_query_context *qctx)
@@ -168,6 +189,15 @@ static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_
 
 		bank.smu_err_type = type;
 
+		/*
+		 * Poison being consumed when injecting a UE while running background workloads,
+		 * which are unexpected.
+		 */
+		if (type == ACA_SMU_TYPE_UE &&
+		    ACA_REG__STATUS__POISON(bank.regs[ACA_REG_IDX_STATUS]) &&
+		    !aca_bank_hwip_is_matched(&bank, ACA_HWIP_TYPE_UMC))
+			continue;
+
 		aca_smu_bank_dump(adev, i, count, &bank, qctx);
 
 		ret = aca_banks_add_bank(banks, &bank);
@@ -178,27 +208,6 @@ static int aca_smu_get_valid_aca_banks(struct amdgpu_device *adev, enum aca_smu_
 	return 0;
 }
 
-static bool aca_bank_hwip_is_matched(struct aca_bank *bank, enum aca_hwip_type type)
-{
-
-	struct aca_hwip *hwip;
-	int hwid, mcatype;
-	u64 ipid;
-
-	if (!bank || type == ACA_HWIP_TYPE_UNKNOW)
-		return false;
-
-	hwip = &aca_hwid_mcatypes[type];
-	if (!hwip->hwid)
-		return false;
-
-	ipid = bank->regs[ACA_REG_IDX_IPID];
-	hwid = ACA_REG__IPID__HARDWAREID(ipid);
-	mcatype = ACA_REG__IPID__MCATYPE(ipid);
-
-	return hwip->hwid == hwid && hwip->mcatype == mcatype;
-}
-
 static bool aca_bank_is_valid(struct aca_handle *handle, struct aca_bank *bank, enum aca_smu_type type)
 {
 	const struct aca_bank_ops *bank_ops = handle->bank_ops;

From 01fa9758c8498d8930df56eca36c88ba3e9493d4 Mon Sep 17 00:00:00 2001
From: Mangesh Gadre <Mangesh.Gadre@amd.com>
Date: Tue, 22 Jul 2025 00:27:52 +0800
Subject: [PATCH 0250/1455] drm/amdgpu: Initialize jpeg v5_0_1 ras function

Initialize jpeg v5_0_1 ras function

Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
index 8f389cb2f6c7..a2d781898767 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -196,6 +196,14 @@ static int jpeg_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 		}
 	}
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG)) {
+		r = amdgpu_jpeg_ras_sw_init(adev);
+		if (r) {
+			dev_err(adev->dev, "Failed to initialize jpeg ras block!\n");
+			return r;
+		}
+	}
+
 	r = amdgpu_jpeg_reg_dump_init(adev, jpeg_reg_list_5_0_1, ARRAY_SIZE(jpeg_reg_list_5_0_1));
 	if (r)
 		return r;

From ba5e322b2617157edb757055252a33587b6729e0 Mon Sep 17 00:00:00 2001
From: Yunxiang Li <Yunxiang.Li@amd.com>
Date: Fri, 25 Jul 2025 12:56:35 -0400
Subject: [PATCH 0251/1455] drm/amdgpu: skip mgpu fan boost for multi-vf

On multi-vf setup if the VM have two vf assigned, perhaps from two
different gpus, mgpu fan boost will fail.

Signed-off-by: Yunxiang Li <Yunxiang.Li@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 8a9488a98646..5949cc747c9f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3389,7 +3389,7 @@ static int amdgpu_device_enable_mgpu_fan_boost(void)
 	for (i = 0; i < mgpu_info.num_dgpu; i++) {
 		gpu_ins = &(mgpu_info.gpu_ins[i]);
 		adev = gpu_ins->adev;
-		if (!(adev->flags & AMD_IS_APU) &&
+		if (!(adev->flags & AMD_IS_APU || amdgpu_sriov_multi_vf_mode(adev)) &&
 		    !gpu_ins->mgpu_fan_enabled) {
 			ret = amdgpu_dpm_enable_mgpu_fan_boost(adev);
 			if (ret)

From cc7bfba95966251b254cb970c21627124da3b7f4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michel=20D=C3=A4nzer?= <mdaenzer@redhat.com>
Date: Wed, 30 Jul 2025 10:09:02 +0200
Subject: [PATCH 0252/1455] drm/amd/display: Add primary plane to commits for
 correct VRR handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

amdgpu_dm_commit_planes calls update_freesync_state_on_stream only for
the primary plane. If a commit affects a CRTC but not its primary plane,
it would previously not trigger a refresh cycle or affect LFC, violating
current UAPI semantics.

Fixes e.g. atomic commits affecting only the cursor plane being limited
to the minimum refresh rate.

Don't do this for the legacy cursor ioctls though, it would break the
UAPI semantics for those.

Suggested-by: Xaver Hugl <xaver.hugl@kde.org>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3034
Signed-off-by: Michel Dänzer <mdaenzer@redhat.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index 2551823382f8..010172f930ae 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -661,6 +661,15 @@ static int amdgpu_dm_crtc_helper_atomic_check(struct drm_crtc *crtc,
 		return -EINVAL;
 	}
 
+	if (!state->legacy_cursor_update && amdgpu_dm_crtc_vrr_active(dm_crtc_state)) {
+		struct drm_plane_state *primary_state;
+
+		/* Pull in primary plane for correct VRR handling */
+		primary_state = drm_atomic_get_plane_state(state, crtc->primary);
+		if (IS_ERR(primary_state))
+			return PTR_ERR(primary_state);
+	}
+
 	/* In some use cases, like reset, no stream is attached */
 	if (!dm_crtc_state->stream)
 		return 0;

From 82594ac858e863b5203df951b6ed198e6fb9628c Mon Sep 17 00:00:00 2001
From: Mangesh Gadre <Mangesh.Gadre@amd.com>
Date: Mon, 21 Jul 2025 23:32:34 +0800
Subject: [PATCH 0253/1455] drm/amdgpu: Initialize vcn v5_0_1 ras function

Initialize vcn v5_0_1 ras function

Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 02a0516edabe..7cb21e2b4eb0 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -197,6 +197,14 @@ static int vcn_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN)) {
+		r = amdgpu_vcn_ras_sw_init(adev);
+		if (r) {
+			dev_err(adev->dev, "Failed to initialize vcn ras block!\n");
+			return r;
+		}
+	}
+
 	r = amdgpu_vcn_reg_dump_init(adev, vcn_reg_list_5_0_1, ARRAY_SIZE(vcn_reg_list_5_0_1));
 	if (r)
 		return r;

From 9dd8e2ba268c636c240a918e0a31e6feaee19404 Mon Sep 17 00:00:00 2001
From: Siyang Liu <Security@tencent.com>
Date: Fri, 4 Jul 2025 11:16:22 +0800
Subject: [PATCH 0254/1455] drm/amd/display: fix a Null pointer dereference
 vulnerability

[Why]
A null pointer dereference vulnerability exists in the AMD display driver's
(DC module) cleanup function dc_destruct().
When display control context (dc->ctx) construction fails
(due to memory allocation failure), this pointer remains NULL.
During subsequent error handling when dc_destruct() is called,
there's no NULL check before dereferencing the perf_trace member
(dc->ctx->perf_trace), causing a kernel null pointer dereference crash.

[How]
Check if dc->ctx is non-NULL before dereferencing.

Link: https://lore.kernel.org/r/tencent_54FF4252EDFB6533090A491A25EEF3EDBF06@qq.com
Co-developed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
(Updated commit text and removed unnecessary error message)
Signed-off-by: Siyang Liu <Security@tencent.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index cf3893a2f8ce..e42cc37eae11 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -940,17 +940,18 @@ static void dc_destruct(struct dc *dc)
 	if (dc->link_srv)
 		link_destroy_link_service(&dc->link_srv);
 
-	if (dc->ctx->gpio_service)
-		dal_gpio_service_destroy(&dc->ctx->gpio_service);
+	if (dc->ctx) {
+		if (dc->ctx->gpio_service)
+			dal_gpio_service_destroy(&dc->ctx->gpio_service);
 
-	if (dc->ctx->created_bios)
-		dal_bios_parser_destroy(&dc->ctx->dc_bios);
+		if (dc->ctx->created_bios)
+			dal_bios_parser_destroy(&dc->ctx->dc_bios);
+		kfree(dc->ctx->logger);
+		dc_perf_trace_destroy(&dc->ctx->perf_trace);
 
-	kfree(dc->ctx->logger);
-	dc_perf_trace_destroy(&dc->ctx->perf_trace);
-
-	kfree(dc->ctx);
-	dc->ctx = NULL;
+		kfree(dc->ctx);
+		dc->ctx = NULL;
+	}
 
 	kfree(dc->bw_vbios);
 	dc->bw_vbios = NULL;

From fd20627c74993cf4d81e3848f2833c7198c884d7 Mon Sep 17 00:00:00 2001
From: Duncan Ma <Duncan.Ma@amd.com>
Date: Tue, 22 Jul 2025 12:22:15 -0400
Subject: [PATCH 0255/1455] drm/amd/display: Adjust AUX-less ALPM setting

[Why & How]
Change ACDS period to support LTTPR.

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Signed-off-by: Duncan Ma <Duncan.Ma@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/protocols/link_edp_panel_control.c  | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index be714cbf6615..2c3e2945124a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -1046,7 +1046,7 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream
 
 		if (link->replay_settings.config.alpm_mode == DC_ALPM_AUXLESS) {
 			alpm_config.bits.ALPM_MODE_SEL = 1;
-			alpm_config.bits.ACDS_PERIOD_DURATION = 1;
+			alpm_config.bits.ACDS_PERIOD_DURATION = 0;
 		}
 
 		dm_helpers_dp_write_dpcd(

From c210b757b400959577a5a17b783b5959b82baed8 Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Mon, 21 Jul 2025 11:03:39 -0400
Subject: [PATCH 0256/1455] drm/amd/display: fix dmub access race condition

Accessing DC from amdgpu_dm is usually preceded by acquisition of
dc_lock mutex. Most of the DC API that DM calls are under a DC lock.
However, there are a few that are not. Some DC API called from interrupt
context end up sending DMUB commands via a DC API, while other threads were
using DMUB. This was apparent from a race between calls for setting idle
optimization enable/disable and the DC API to set vmin/vmax.

Offload the call to dc_stream_adjust_vmin_vmax() to a thread instead
of directly calling them from the interrupt handler such that it waits
for dc_lock.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 55 +++++++++++++++++--
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 14 +++++
 2 files changed, 63 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c71167ffdb76..6762fc7de769 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -530,6 +530,50 @@ static void dm_pflip_high_irq(void *interrupt_params)
 		      amdgpu_crtc->crtc_id, amdgpu_crtc, vrr_active, (int)!e);
 }
 
+static void dm_handle_vmin_vmax_update(struct work_struct *offload_work)
+{
+	struct vupdate_offload_work *work = container_of(offload_work, struct vupdate_offload_work, work);
+	struct amdgpu_device *adev = work->adev;
+	struct dc_stream_state *stream = work->stream;
+	struct dc_crtc_timing_adjust *adjust = work->adjust;
+
+	mutex_lock(&adev->dm.dc_lock);
+	dc_stream_adjust_vmin_vmax(adev->dm.dc, stream, adjust);
+	mutex_unlock(&adev->dm.dc_lock);
+
+	dc_stream_release(stream);
+	kfree(work->adjust);
+	kfree(work);
+}
+
+static void schedule_dc_vmin_vmax(struct amdgpu_device *adev,
+	struct dc_stream_state *stream,
+	struct dc_crtc_timing_adjust *adjust)
+{
+	struct vupdate_offload_work *offload_work = kzalloc(sizeof(*offload_work), GFP_KERNEL);
+	if (!offload_work) {
+		drm_dbg_driver(adev_to_drm(adev), "Failed to allocate vupdate_offload_work\n");
+		return;
+	}
+
+	struct dc_crtc_timing_adjust *adjust_copy = kzalloc(sizeof(*adjust_copy), GFP_KERNEL);
+	if (!adjust_copy) {
+		drm_dbg_driver(adev_to_drm(adev), "Failed to allocate adjust_copy\n");
+		kfree(offload_work);
+		return;
+	}
+
+	dc_stream_retain(stream);
+	memcpy(adjust_copy, adjust, sizeof(*adjust_copy));
+
+	INIT_WORK(&offload_work->work, dm_handle_vmin_vmax_update);
+	offload_work->adev = adev;
+	offload_work->stream = stream;
+	offload_work->adjust = adjust_copy;
+
+	queue_work(system_wq, &offload_work->work);
+}
+
 static void dm_vupdate_high_irq(void *interrupt_params)
 {
 	struct common_irq_params *irq_params = interrupt_params;
@@ -579,10 +623,9 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 				    acrtc->dm_irq_params.stream,
 				    &acrtc->dm_irq_params.vrr_params);
 
-				dc_stream_adjust_vmin_vmax(
-				    adev->dm.dc,
-				    acrtc->dm_irq_params.stream,
-				    &acrtc->dm_irq_params.vrr_params.adjust);
+				schedule_dc_vmin_vmax(adev,
+					acrtc->dm_irq_params.stream,
+					&acrtc->dm_irq_params.vrr_params.adjust);
 				spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 			}
 		}
@@ -672,8 +715,8 @@ static void dm_crtc_high_irq(void *interrupt_params)
 					     acrtc->dm_irq_params.stream,
 					     &acrtc->dm_irq_params.vrr_params);
 
-		dc_stream_adjust_vmin_vmax(adev->dm.dc, acrtc->dm_irq_params.stream,
-					   &acrtc->dm_irq_params.vrr_params.adjust);
+		schedule_dc_vmin_vmax(adev, acrtc->dm_irq_params.stream,
+				&acrtc->dm_irq_params.vrr_params.adjust);
 	}
 
 	/*
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 67c3a7a967f2..94f312bae9ac 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -153,6 +153,20 @@ struct idle_workqueue {
 	bool running;
 };
 
+/**
+ * struct dm_vupdate_work - Work data for periodic action in idle
+ * @work: Kernel work data for the work event
+ * @adev: amdgpu_device back pointer
+ * @stream: DC stream associated with the crtc
+ * @adjust: DC CRTC timing adjust to be applied to the crtc
+ */
+struct vupdate_offload_work {
+	struct work_struct work;
+	struct amdgpu_device *adev;
+	struct dc_stream_state *stream;
+	struct dc_crtc_timing_adjust *adjust;
+};
+
 #define MAX_LUMINANCE_DATA_POINTS 99
 
 /**

From 6d31602a9f57a7bb3c6c8dbde1d00af67e250a3f Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Wed, 16 Apr 2025 11:26:54 -0400
Subject: [PATCH 0257/1455] drm/amd/display: more liberal vmin/vmax update for
 freesync

[Why]
FAMS2 expects vmin/vmax to be updated in the case when freesync is
off, but supported. But we only update it when freesync is enabled.

[How]
Change the vsync handler such that dc_stream_adjust_vmin_vmax() its called
irrespective of whether freesync is enabled. If freesync is supported,
then there is no harm in updating vmin/vmax registers.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3546
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Reviewed-by: ChiaHsuan Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 28 +++++++++++++------
 1 file changed, 20 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 6762fc7de769..16347ca2396a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -612,6 +612,11 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 		 * if a pageflip happened inside front-porch.
 		 */
 		if (vrr_active) {
+			bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled;
+			bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled;
+			bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state
+				== VRR_STATE_ACTIVE_VARIABLE;
+
 			amdgpu_dm_crtc_handle_vblank(acrtc);
 
 			/* BTR processing for pre-DCE12 ASICs */
@@ -623,9 +628,11 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 				    acrtc->dm_irq_params.stream,
 				    &acrtc->dm_irq_params.vrr_params);
 
-				schedule_dc_vmin_vmax(adev,
-					acrtc->dm_irq_params.stream,
-					&acrtc->dm_irq_params.vrr_params.adjust);
+				if (fs_active_var_en || (!fs_active_var_en && !replay_en && !psr_en)) {
+					schedule_dc_vmin_vmax(adev,
+						acrtc->dm_irq_params.stream,
+						&acrtc->dm_irq_params.vrr_params.adjust);
+				}
 				spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
 			}
 		}
@@ -708,15 +715,20 @@ static void dm_crtc_high_irq(void *interrupt_params)
 	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 
 	if (acrtc->dm_irq_params.stream &&
-	    acrtc->dm_irq_params.vrr_params.supported &&
-	    acrtc->dm_irq_params.freesync_config.state ==
-		    VRR_STATE_ACTIVE_VARIABLE) {
+		acrtc->dm_irq_params.vrr_params.supported) {
+		bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled;
+		bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled;
+		bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state == VRR_STATE_ACTIVE_VARIABLE;
+
 		mod_freesync_handle_v_update(adev->dm.freesync_module,
 					     acrtc->dm_irq_params.stream,
 					     &acrtc->dm_irq_params.vrr_params);
 
-		schedule_dc_vmin_vmax(adev, acrtc->dm_irq_params.stream,
-				&acrtc->dm_irq_params.vrr_params.adjust);
+		/* update vmin_vmax only if freesync is enabled, or only if PSR and REPLAY are disabled */
+		if (fs_active_var_en || (!fs_active_var_en && !replay_en && !psr_en)) {
+			schedule_dc_vmin_vmax(adev, acrtc->dm_irq_params.stream,
+					&acrtc->dm_irq_params.vrr_params.adjust);
+		}
 	}
 
 	/*

From 2e72fdba8a32ce062a86571edff4592710c26215 Mon Sep 17 00:00:00 2001
From: Paul Hsieh <Paul.Hsieh@amd.com>
Date: Wed, 23 Jul 2025 11:51:42 +0800
Subject: [PATCH 0258/1455] drm/amd/display: update dpp/disp clock from smu
 clock table

[Why]
The reason some high-resolution monitors fail to display properly
is that this platform does not support sufficiently high DPP and
DISP clock frequencies

[How]
Update DISP and DPP clocks from the smu clock table then DML can
filter these mode if not support.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Paul Hsieh <Paul.Hsieh@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn301/vg_clk_mgr.c    | 16 +++++++++++++++
 .../amd/display/dc/dml/dcn301/dcn301_fpu.c    | 20 ++++++++++++++++---
 2 files changed, 33 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c
index 9e2ef0e724fc..7aee02d56292 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c
@@ -563,6 +563,7 @@ static void vg_clk_mgr_helper_populate_bw_params(
 {
 	int i, j;
 	struct clk_bw_params *bw_params = clk_mgr->base.bw_params;
+	uint32_t max_dispclk = 0, max_dppclk = 0;
 
 	j = -1;
 
@@ -584,6 +585,15 @@ static void vg_clk_mgr_helper_populate_bw_params(
 		return;
 	}
 
+	/* dispclk and dppclk can be max at any voltage, same number of levels for both */
+	if (clock_table->NumDispClkLevelsEnabled <= VG_NUM_DISPCLK_DPM_LEVELS &&
+	    clock_table->NumDispClkLevelsEnabled <= VG_NUM_DPPCLK_DPM_LEVELS) {
+		max_dispclk = find_max_clk_value(clock_table->DispClocks, clock_table->NumDispClkLevelsEnabled);
+		max_dppclk = find_max_clk_value(clock_table->DppClocks, clock_table->NumDispClkLevelsEnabled);
+	} else {
+		ASSERT(0);
+	}
+
 	bw_params->clk_table.num_entries = j + 1;
 
 	for (i = 0; i < bw_params->clk_table.num_entries - 1; i++, j--) {
@@ -591,11 +601,17 @@ static void vg_clk_mgr_helper_populate_bw_params(
 		bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[j].memclk;
 		bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[j].voltage;
 		bw_params->clk_table.entries[i].dcfclk_mhz = find_dcfclk_for_voltage(clock_table, clock_table->DfPstateTable[j].voltage);
+
+		/* Now update clocks we do read */
+		bw_params->clk_table.entries[i].dispclk_mhz = max_dispclk;
+		bw_params->clk_table.entries[i].dppclk_mhz = max_dppclk;
 	}
 	bw_params->clk_table.entries[i].fclk_mhz = clock_table->DfPstateTable[j].fclk;
 	bw_params->clk_table.entries[i].memclk_mhz = clock_table->DfPstateTable[j].memclk;
 	bw_params->clk_table.entries[i].voltage = clock_table->DfPstateTable[j].voltage;
 	bw_params->clk_table.entries[i].dcfclk_mhz = find_max_clk_value(clock_table->DcfClocks, VG_NUM_DCFCLK_DPM_LEVELS);
+	bw_params->clk_table.entries[i].dispclk_mhz = find_max_clk_value(clock_table->DispClocks, VG_NUM_DISPCLK_DPM_LEVELS);
+	bw_params->clk_table.entries[i].dppclk_mhz = find_max_clk_value(clock_table->DppClocks, VG_NUM_DPPCLK_DPM_LEVELS);
 
 	bw_params->vram_type = bios_info->memory_type;
 	bw_params->num_channels = bios_info->ma_channel_number;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
index 0c0b2d67c9cd..2066a65c69bb 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
@@ -326,7 +326,7 @@ void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p
 	struct dcn301_resource_pool *pool = TO_DCN301_RES_POOL(dc->res_pool);
 	struct clk_limit_table *clk_table = &bw_params->clk_table;
 	unsigned int i, closest_clk_lvl;
-	int j;
+	int j = 0, max_dispclk_mhz = 0, max_dppclk_mhz = 0;
 
 	dc_assert_fp_enabled();
 
@@ -338,6 +338,15 @@ void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p
 	dcn3_01_soc.num_chans = bw_params->num_channels;
 
 	ASSERT(clk_table->num_entries);
+
+	/* Prepass to find max clocks independent of voltage level. */
+	for (i = 0; i < clk_table->num_entries; ++i) {
+		if (clk_table->entries[i].dispclk_mhz > max_dispclk_mhz)
+			max_dispclk_mhz = clk_table->entries[i].dispclk_mhz;
+		if (clk_table->entries[i].dppclk_mhz > max_dppclk_mhz)
+			max_dppclk_mhz = clk_table->entries[i].dppclk_mhz;
+	}
+
 	for (i = 0; i < clk_table->num_entries; i++) {
 		/* loop backwards*/
 		for (closest_clk_lvl = 0, j = dcn3_01_soc.num_states - 1; j >= 0; j--) {
@@ -353,8 +362,13 @@ void dcn301_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p
 		s[i].socclk_mhz = clk_table->entries[i].socclk_mhz;
 		s[i].dram_speed_mts = clk_table->entries[i].memclk_mhz * 2;
 
-		s[i].dispclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
-		s[i].dppclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+		/* Clocks independent of voltage level. */
+		s[i].dispclk_mhz = max_dispclk_mhz ? max_dispclk_mhz :
+			dcn3_01_soc.clock_limits[closest_clk_lvl].dispclk_mhz;
+
+		s[i].dppclk_mhz = max_dppclk_mhz ? max_dppclk_mhz :
+			dcn3_01_soc.clock_limits[closest_clk_lvl].dppclk_mhz;
+
 		s[i].dram_bw_per_chan_gbps =
 			dcn3_01_soc.clock_limits[closest_clk_lvl].dram_bw_per_chan_gbps;
 		s[i].dscclk_mhz = dcn3_01_soc.clock_limits[closest_clk_lvl].dscclk_mhz;

From 6ec8a5cbec751625133461600d0d4950ffd3a214 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Sun, 20 Jul 2025 23:39:41 -0500
Subject: [PATCH 0259/1455] drm/amd/display: Revert "drm/amd/display: Fix
 AMDGPU_MAX_BL_LEVEL value"

This reverts commit 66abb996999de0d440a02583a6e70c2c24deab45.
This broke custom brightness curves but it wasn't obvious because
of other related changes. Custom brightness curves are always
from a 0-255 input signal. The correct fix was to fix the default
value which was done by [1].

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4412
Link: https://lore.kernel.org/amd-gfx/0f094c4b-d2a3-42cd-824c-dc2858a5618d@kernel.org/T/#m69f875a7e69aa22df3370b3e3a9e69f4a61fdaf2
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 16347ca2396a..31ea57edeb45 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -4800,16 +4800,16 @@ static int get_brightness_range(const struct amdgpu_dm_backlight_caps *caps,
 	return 1;
 }
 
-/* Rescale from [min..max] to [0..MAX_BACKLIGHT_LEVEL] */
+/* Rescale from [min..max] to [0..AMDGPU_MAX_BL_LEVEL] */
 static inline u32 scale_input_to_fw(int min, int max, u64 input)
 {
-	return DIV_ROUND_CLOSEST_ULL(input * MAX_BACKLIGHT_LEVEL, max - min);
+	return DIV_ROUND_CLOSEST_ULL(input * AMDGPU_MAX_BL_LEVEL, max - min);
 }
 
-/* Rescale from [0..MAX_BACKLIGHT_LEVEL] to [min..max] */
+/* Rescale from [0..AMDGPU_MAX_BL_LEVEL] to [min..max] */
 static inline u32 scale_fw_to_input(int min, int max, u64 input)
 {
-	return min + DIV_ROUND_CLOSEST_ULL(input * (max - min), MAX_BACKLIGHT_LEVEL);
+	return min + DIV_ROUND_CLOSEST_ULL(input * (max - min), AMDGPU_MAX_BL_LEVEL);
 }
 
 static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *caps,

From fa5f99ee7260e067e4e3c1350243471862fad11c Mon Sep 17 00:00:00 2001
From: Fangzhi Zuo <Jerry.Zuo@amd.com>
Date: Wed, 9 Jul 2025 21:42:54 -0400
Subject: [PATCH 0260/1455] drm/amd/display: Avoid Read Remote DPCD Many Times

Reading remote dpcd is time consuming. Instead of reading each byte
one by one, read 16 bytes together.

Reviewed-by: ChiaHsuan (Tom) Chung <chiahsuan.chung@amd.com>
Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c  | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 137f18d41f1b..32281bf62f70 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -1763,14 +1763,17 @@ static bool dp_get_link_current_set_bw(struct drm_dp_aux *aux, uint32_t *cur_lin
 	union lane_count_set lane_count;
 	u8 dp_link_encoding;
 	u8 link_bw_set = 0;
+	u8 data[16] = {0};
 
 	*cur_link_bw = 0;
 
-	if (drm_dp_dpcd_read(aux, DP_MAIN_LINK_CHANNEL_CODING_SET, &dp_link_encoding, 1) != 1 ||
-		drm_dp_dpcd_read(aux, DP_LANE_COUNT_SET, &lane_count.raw, 1) != 1 ||
-		drm_dp_dpcd_read(aux, DP_LINK_BW_SET, &link_bw_set, 1) != 1)
+	if (drm_dp_dpcd_read(aux, DP_LINK_BW_SET, data, 16) != 16)
 		return false;
 
+	dp_link_encoding = data[DP_MAIN_LINK_CHANNEL_CODING_SET - DP_LINK_BW_SET];
+	link_bw_set = data[DP_LINK_BW_SET - DP_LINK_BW_SET];
+	lane_count.raw = data[DP_LANE_COUNT_SET - DP_LINK_BW_SET];
+
 	switch (dp_link_encoding) {
 	case DP_8b_10b_ENCODING:
 		link_rate = link_bw_set;

From 3df957517f8c17cd6c9be9c5a12979caf653a763 Mon Sep 17 00:00:00 2001
From: Jingwen Zhu <Jingwen.Zhu@amd.com>
Date: Mon, 14 Jul 2025 16:18:19 +0800
Subject: [PATCH 0261/1455] drm/amd/display: limited pll vco w/a v2

[Why/How]
The w/a will cause reboot black screen issue.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Jingwen Zhu <Jingwen.Zhu@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h       | 1 +
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 3 ++-
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 0bafb6710761..87b761ac3135 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -316,6 +316,7 @@ struct dmub_srv_hw_params {
 	bool disable_sldo_opt;
 	bool enable_non_transparent_setconfig;
 	bool lower_hbr3_phy_ssc;
+	bool override_hbr3_pll_vco;
 };
 
 /**
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index dcae768c2cf4..79b5b1bb9b93 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -843,7 +843,8 @@ union dmub_fw_boot_options {
 		uint32_t ips_sequential_ono: 1; /**< 1 to enable sequential ONO IPS sequence */
 		uint32_t disable_sldo_opt: 1; /**< 1 to disable SLDO optimizations */
 		uint32_t lower_hbr3_phy_ssc: 1; /**< 1 to lower hbr3 phy ssc to 0.125 percent */
-		uint32_t reserved : 6; /**< reserved */
+		uint32_t override_hbr3_pll_vco: 1; /**< 1 to override the hbr3 pll vco to 0 */
+		uint32_t reserved : 5; /**< reserved */
 	} bits; /**< boot bits */
 	uint32_t all; /**< 32-bit access to bits */
 };
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
index 3f38db752b84..4777c7203b2c 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn31.c
@@ -377,6 +377,7 @@ void dmub_dcn31_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmu
 	boot_options.bits.dpia_hpd_int_enable_supported = params->dpia_hpd_int_enable_supported;
 	boot_options.bits.power_optimization = params->power_optimization;
 	boot_options.bits.lower_hbr3_phy_ssc = params->lower_hbr3_phy_ssc;
+	boot_options.bits.override_hbr3_pll_vco = params->override_hbr3_pll_vco;
 
 	boot_options.bits.sel_mux_phy_c_d_phy_f_g = (dmub->asic == DMUB_ASIC_DCN31B) ? 1 : 0;
 

From e63e9f8b3d1c5d17fa4db9241905a8baf1e4bda8 Mon Sep 17 00:00:00 2001
From: Reza Amini <reza.amini@amd.com>
Date: Mon, 14 Jul 2025 16:22:38 -0400
Subject: [PATCH 0262/1455] drm/amd/display: Fixing hubp programming of 3dlut
 fast load

[why]
HUBP needs to know the size of the lut's destination in MPC.
This is currently defaulted to 17, and needs to be set for specific
lut size.

[how]
Define and apply the missing hubp field. Taking this opportunity
to consolidate the programming of 3dlut into a hubp and mpc function.

Reviewed-by: Krunoslav Kovac <krunoslav.kovac@amd.com>
Signed-off-by: Reza Amini <reza.amini@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h           | 26 ++++++++
 .../amd/display/dc/hubp/dcn10/dcn10_hubp.h    |  1 +
 .../amd/display/dc/hubp/dcn20/dcn20_hubp.h    |  1 +
 .../amd/display/dc/hubp/dcn401/dcn401_hubp.c  | 38 +++++++++++
 .../amd/display/dc/hubp/dcn401/dcn401_hubp.h  |  4 ++
 .../amd/display/dc/hwss/dcn10/dcn10_hwseq.c   | 63 ++++++++++++++-----
 drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h  | 19 +++++-
 drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h   | 11 ++++
 8 files changed, 148 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 5653c1673aec..c013b28207a1 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1311,6 +1311,32 @@ union dc_3dlut_state {
 };
 
 
+#define MATRIX_9C__DIM_128_ALIGNED_LEN   16 // 9+8 :  9 * 8 +  7 * 8 = 72  + 56  = 128 % 128 = 0
+#define MATRIX_17C__DIM_128_ALIGNED_LEN  32 //17+15:  17 * 8 + 15 * 8 = 136 + 120 = 256 % 128 = 0
+#define MATRIX_33C__DIM_128_ALIGNED_LEN  64 //17+47:  17 * 8 + 47 * 8 = 136 + 376 = 512 % 128 = 0
+
+struct lut_rgb {
+	uint16_t b;
+	uint16_t g;
+	uint16_t r;
+	uint16_t padding;
+};
+
+//this structure maps directly to how the lut will read it from memory
+struct lut_mem_mapping {
+	union {
+		//NATIVE MODE 1, 2
+		//RGB layout          [b][g][r]      //red  is 128 byte aligned
+		//BGR layout          [r][g][b]      //blue is 128 byte aligned
+		struct lut_rgb rgb_17c[17][17][MATRIX_17C__DIM_128_ALIGNED_LEN];
+		struct lut_rgb rgb_33c[33][33][MATRIX_33C__DIM_128_ALIGNED_LEN];
+
+		//TRANSFORMED
+		uint16_t linear_rgb[(33*33*33*4/128+1)*128];
+	};
+	uint16_t size;
+};
+
 struct dc_rmcm_3dlut {
 	bool isInUse;
 	const struct dc_stream_state *stream;
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h
index f8f991785d4f..0b7547d5b488 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h
@@ -671,6 +671,7 @@ struct dcn_fl_regs_st {
 	uint32_t lut_done;
 	uint32_t lut_addr_mode;
 	uint32_t lut_width;
+	uint32_t lut_mpc_width;
 	uint32_t lut_tmz;
 	uint32_t lut_crossbar_sel_r;
 	uint32_t lut_crossbar_sel_g;
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h
index 62369be070ea..f325db555102 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn20/dcn20_hubp.h
@@ -264,6 +264,7 @@
 	type HUBP_3DLUT_DONE;\
 	type HUBP_3DLUT_ADDRESSING_MODE;\
 	type HUBP_3DLUT_WIDTH;\
+	type HUBP_3DLUT_MPC_WIDTH;\
 	type HUBP_3DLUT_TMZ;\
 	type HUBP_3DLUT_CROSSBAR_SELECT_Y_G;\
 	type HUBP_3DLUT_CROSSBAR_SELECT_CB_B;\
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
index 705b98b1b6cc..5028180ad80a 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
@@ -127,6 +127,43 @@ void hubp401_program_3dlut_fl_format(struct hubp *hubp, enum hubp_3dlut_fl_forma
 	REG_UPDATE(_3DLUT_FL_CONFIG, HUBP0_3DLUT_FL_FORMAT, format);
 }
 
+void hubp401_program_3dlut_fl_config(
+	struct hubp *hubp,
+	struct hubp_fl_3dlut_config *cfg)
+{
+	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
+
+	uint32_t mpc_width = {(cfg->width == 17) ? 0 : 1};
+	uint32_t width = {cfg->width};
+
+	if (cfg->layout == DC_CM2_GPU_MEM_LAYOUT_1D_PACKED_LINEAR)
+		width = (cfg->width == 17) ? 4916 : 35940;
+
+	REG_UPDATE_2(_3DLUT_FL_CONFIG,
+		HUBP0_3DLUT_FL_MODE, cfg->mode,
+		HUBP0_3DLUT_FL_FORMAT, cfg->format);
+
+	REG_UPDATE_2(_3DLUT_FL_BIAS_SCALE,
+		HUBP0_3DLUT_FL_BIAS, cfg->bias,
+		HUBP0_3DLUT_FL_SCALE, cfg->scale);
+
+	REG_UPDATE(HUBP_3DLUT_ADDRESS_HIGH,
+		HUBP_3DLUT_ADDRESS_HIGH, cfg->address.lut3d.addr.high_part);
+	REG_UPDATE(HUBP_3DLUT_ADDRESS_LOW,
+		HUBP_3DLUT_ADDRESS_LOW, cfg->address.lut3d.addr.low_part);
+
+	//cross bar
+	REG_UPDATE_8(HUBP_3DLUT_CONTROL,
+		HUBP_3DLUT_MPC_WIDTH, mpc_width,
+		HUBP_3DLUT_WIDTH, width,
+		HUBP_3DLUT_CROSSBAR_SELECT_CR_R, cfg->crossbar_bit_slice_cr_r,
+		HUBP_3DLUT_CROSSBAR_SELECT_Y_G, cfg->crossbar_bit_slice_y_g,
+		HUBP_3DLUT_CROSSBAR_SELECT_CB_B, cfg->crossbar_bit_slice_cb_b,
+		HUBP_3DLUT_ADDRESSING_MODE, cfg->addr_mode,
+		HUBP_3DLUT_TMZ, cfg->protection_bits,
+		HUBP_3DLUT_ENABLE, cfg->enabled ? 1 : 0);
+}
+
 void hubp401_update_mall_sel(struct hubp *hubp, uint32_t mall_sel, bool c_cursor)
 {
 	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
@@ -1033,6 +1070,7 @@ static struct hubp_funcs dcn401_hubp_funcs = {
 	.hubp_program_3dlut_fl_crossbar = hubp401_program_3dlut_fl_crossbar,
 	.hubp_get_3dlut_fl_done = hubp401_get_3dlut_fl_done,
 	.hubp_clear_tiling = hubp401_clear_tiling,
+	.hubp_program_3dlut_fl_config = hubp401_program_3dlut_fl_config,
 };
 
 bool hubp401_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h
index 608e6153fa68..887b479ed1d7 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h
@@ -349,6 +349,10 @@ void hubp401_program_3dlut_fl_format(struct hubp *hubp, enum hubp_3dlut_fl_forma
 
 void hubp401_program_3dlut_fl_mode(struct hubp *hubp, enum hubp_3dlut_fl_mode mode);
 
+void hubp401_program_3dlut_fl_config(
+	struct hubp *hubp,
+	struct hubp_fl_3dlut_config *cfg);
+
 void hubp401_clear_tiling(struct hubp *hubp);
 
 void hubp401_vready_at_or_After_vsync(struct hubp *hubp,
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index 39910f73ecd0..79c9bea78c47 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -328,19 +328,25 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx)
 	}
 
 	DTN_INFO("\n=======HUBP FL======\n");
-	DTN_INFO(
-		"HUBP FL:  Enabled  Done  adr_mode  width  tmz  xbar_sel_R  xbar_sel_G  xbar_sel_B  adr_hi  adr_low  REFCYC  Bias   Scale       Mode      Format\n");
+	char pLabels[18][50] = {
+		"inst", "Enabled ", "Done ", "adr_mode ", "width ", "mpc_width ",
+		"tmz", "xbar_sel_R", "xbar_sel_G", "xbar_sel_B", "adr_hi ",
+		"adr_low", "REFCYC", "Bias", "Scale", "Mode",
+		"Format", "prefetch"};
+
 	for (i = 0; i < pool->pipe_count; i++) {
 		struct dcn_hubp_state *s = &(TO_DCN10_HUBP(pool->hubps[i])->state);
 		struct dcn_fl_regs_st *fl_regs = &s->fl_regs;
+		struct _vcs_dpi_display_dlg_regs_st *dlg_regs = &s->dlg_attr;
 
 		if (!s->blank_en) {
-			DTN_INFO("[%2d]:  %5xh  %6xh  %5d  %6d  %8xh  %2xh  %6xh  %6d  %8d  %8d  %7d  %8xh %5x %5x %5x",
+			uint32_t values[] = {
 					pool->hubps[i]->inst,
 					fl_regs->lut_enable,
 					fl_regs->lut_done,
 					fl_regs->lut_addr_mode,
 					fl_regs->lut_width,
+					fl_regs->lut_mpc_width,
 					fl_regs->lut_tmz,
 					fl_regs->lut_crossbar_sel_r,
 					fl_regs->lut_crossbar_sel_g,
@@ -351,8 +357,13 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx)
 					fl_regs->lut_fl_bias,
 					fl_regs->lut_fl_scale,
 					fl_regs->lut_fl_mode,
-					fl_regs->lut_fl_format);
-			DTN_INFO("\n");
+					fl_regs->lut_fl_format,
+					dlg_regs->dst_y_prefetch};
+
+			int num_elements = 18;
+
+			for (int j = 0; j < num_elements; j++)
+				DTN_INFO("%s \t %8xh\n", pLabels[j], values[j]);
 		}
 	}
 
@@ -541,19 +552,43 @@ static void dcn10_log_color_state(struct dc *dc,
 		 dc->caps.color.mpc.ogam_ram,
 		 dc->caps.color.mpc.ocsc);
 	DTN_INFO("===== MPC RMCM 3DLUT =====\n");
-	DTN_INFO("MPCC:  SIZE  MODE  MODE_CUR  RD_SEL  30BIT_EN  WR_EN_MASK  RAM_SEL  OUT_NORM_FACTOR	FL_SEL	OUT_OFFSET	OUT_SCALE	FL_DONE	SOFT_UNDERFLOW	HARD_UNDERFLOW MEM_PWR_ST	FORCE	DIS	MODE\n");
+	char pLabels[19][50] = {
+		"MPCC", "SIZE", "MODE", "MODE_CUR", "RD_SEL",
+		"30BIT_EN", "WR_EN_MASK", "RAM_SEL", "OUT_NORM_FACTOR", "FL_SEL",
+		"OUT_OFFSET", "OUT_SCALE", "FL_DONE", "SOFT_UNDERFLOW", "HARD_UNDERFLOW",
+		"MEM_PWR_ST", "FORCE", "DIS", "MODE"};
+
 	for (i = 0; i < pool->mpcc_count; i++) {
 		struct mpcc_state s = {0};
 
 		pool->mpc->funcs->read_mpcc_state(pool->mpc, i, &s);
-		if (s.opp_id != 0xf)
-			DTN_INFO("[%2d]:  %4xh  %4xh  %6xh  %4x  %4x  %4x  %4x  %4x %4xh  %4xh  %6xh  %4x  %4x  %4x  %4x  %4x  %4x  %4x\n",
-				i, s.rmcm_regs.rmcm_3dlut_size, s.rmcm_regs.rmcm_3dlut_mode, s.rmcm_regs.rmcm_3dlut_mode_cur,
-				s.rmcm_regs.rmcm_3dlut_read_sel, s.rmcm_regs.rmcm_3dlut_30bit_en, s.rmcm_regs.rmcm_3dlut_wr_en_mask,
-				s.rmcm_regs.rmcm_3dlut_ram_sel, s.rmcm_regs.rmcm_3dlut_out_norm_factor, s.rmcm_regs.rmcm_3dlut_fl_sel,
-				s.rmcm_regs.rmcm_3dlut_out_offset_r, s.rmcm_regs.rmcm_3dlut_out_scale_r, s.rmcm_regs.rmcm_3dlut_fl_done,
-				s.rmcm_regs.rmcm_3dlut_fl_soft_underflow, s.rmcm_regs.rmcm_3dlut_fl_hard_underflow, s.rmcm_regs.rmcm_3dlut_mem_pwr_state,
-				s.rmcm_regs.rmcm_3dlut_mem_pwr_force, s.rmcm_regs.rmcm_3dlut_mem_pwr_dis, s.rmcm_regs.rmcm_3dlut_mem_pwr_mode);
+		if (s.opp_id != 0xf) {
+			uint32_t values[] = {
+				i,
+				s.rmcm_regs.rmcm_3dlut_size,
+				s.rmcm_regs.rmcm_3dlut_mode,
+				s.rmcm_regs.rmcm_3dlut_mode_cur,
+				s.rmcm_regs.rmcm_3dlut_read_sel,
+				s.rmcm_regs.rmcm_3dlut_30bit_en,
+				s.rmcm_regs.rmcm_3dlut_wr_en_mask,
+				s.rmcm_regs.rmcm_3dlut_ram_sel,
+				s.rmcm_regs.rmcm_3dlut_out_norm_factor,
+				s.rmcm_regs.rmcm_3dlut_fl_sel,
+				s.rmcm_regs.rmcm_3dlut_out_offset_r,
+				s.rmcm_regs.rmcm_3dlut_out_scale_r,
+				s.rmcm_regs.rmcm_3dlut_fl_done,
+				s.rmcm_regs.rmcm_3dlut_fl_soft_underflow,
+				s.rmcm_regs.rmcm_3dlut_fl_hard_underflow,
+				s.rmcm_regs.rmcm_3dlut_mem_pwr_state,
+				s.rmcm_regs.rmcm_3dlut_mem_pwr_force,
+				s.rmcm_regs.rmcm_3dlut_mem_pwr_dis,
+				s.rmcm_regs.rmcm_3dlut_mem_pwr_mode};
+
+			int num_elements = 19;
+
+			for (int j = 0; j < num_elements; j++)
+				DTN_INFO("%s \t %8xh\n", pLabels[j], values[j]);
+		}
 	}
 	DTN_INFO("\n");
 	DTN_INFO("===== MPC RMCM Shaper =====\n");
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
index cee29e89ec5c..198a28bd8e28 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
@@ -89,7 +89,7 @@ enum hubp_3dlut_fl_addressing_mode {
 enum hubp_3dlut_fl_width {
 	hubp_3dlut_fl_width_17 = 17,
 	hubp_3dlut_fl_width_33 = 33,
-	hubp_3dlut_fl_width_transformed = 4916
+	hubp_3dlut_fl_width_transformed    = 4916, //mpc default
 };
 
 enum hubp_3dlut_fl_crossbar_bit_slice {
@@ -99,6 +99,22 @@ enum hubp_3dlut_fl_crossbar_bit_slice {
 	hubp_3dlut_fl_crossbar_bit_slice_48_63 = 3
 };
 
+struct hubp_fl_3dlut_config {
+	bool enabled;
+	enum hubp_3dlut_fl_width width;
+	enum hubp_3dlut_fl_mode mode;
+	enum hubp_3dlut_fl_format format;
+	uint16_t bias;
+	uint16_t scale;
+	struct dc_plane_address address;
+	enum hubp_3dlut_fl_addressing_mode addr_mode;
+	enum dc_cm2_gpu_mem_layout layout;
+	uint8_t protection_bits;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_y_g;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cb_b;
+	enum hubp_3dlut_fl_crossbar_bit_slice crossbar_bit_slice_cr_r;
+};
+
 struct hubp {
 	const struct hubp_funcs *funcs;
 	struct dc_context *ctx;
@@ -288,6 +304,7 @@ struct hubp_funcs {
 			enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_cb_b,
 			enum hubp_3dlut_fl_crossbar_bit_slice bit_slice_cr_r);
 	int (*hubp_get_3dlut_fl_done)(struct hubp *hubp);
+	void (*hubp_program_3dlut_fl_config)(struct hubp *hubp, struct hubp_fl_3dlut_config *cfg);
 	void (*hubp_clear_tiling)(struct hubp *hubp);
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
index 7641439f6ca0..14f0304e3eb9 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
@@ -115,6 +115,16 @@ enum MCM_LUT_ID {
 	MCM_LUT_SHAPER
 };
 
+struct mpc_fl_3dlut_config {
+	bool enabled;
+	uint16_t width;
+	bool select_lut_bank_a;
+	uint16_t bit_depth;
+	int hubp_index;
+	uint16_t bias;
+	uint16_t scale;
+};
+
 union mcm_lut_params {
 	const struct pwl_params *pwl;
 	const struct tetrahedral_params *lut3d;
@@ -1098,6 +1108,7 @@ struct mpc_funcs {
 	 * MPC RMCM new HW sequential programming functions
 	 */
 	struct {
+		void (*fl_3dlut_configure)(struct mpc *mpc, struct mpc_fl_3dlut_config *cfg, int mpcc_id);
 		void (*enable_3dlut_fl)(struct mpc *mpc, bool enable, int mpcc_id);
 		void (*update_3dlut_fast_load_select)(struct mpc *mpc, int mpcc_id, int hubp_idx);
 		void (*program_lut_read_write_control)(struct mpc *mpc, const enum MCM_LUT_ID id,

From 20ea30a793f29fd9026406dc4de3c8d17da120c2 Mon Sep 17 00:00:00 2001
From: Ryan Seto <ryanseto@amd.com>
Date: Thu, 24 Jul 2025 14:57:52 -0400
Subject: [PATCH 0263/1455] drm/amd/display: Toggle for Disable Force Pstate
 Allow on Disable

[Why & How]
In theory, driver should be able to support disabling force pstate allow
after hardware release however this behavior is not tested yet.
Introducing a new toggle to disable the force on the fly.

Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Ryan Seto <ryanseto@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h           |  1 +
 .../amd/display/dc/hwss/dcn401/dcn401_hwseq.c | 32 ++++++++++++-------
 .../dc/resource/dcn401/dcn401_resource.c      |  1 +
 3 files changed, 22 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index c013b28207a1..a83227f6bb14 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1073,6 +1073,7 @@ struct dc_debug_options {
 	unsigned int force_mall_ss_num_ways;
 	bool alloc_extra_way_for_cursor;
 	uint32_t subvp_extra_lines;
+	bool disable_force_pstate_allow_on_hw_release;
 	bool force_usr_allow;
 	/* uses value at boot and disables switch */
 	bool disable_dtb_ref_clk_switch;
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index fb949aeb1244..d5b5e2ce6ff6 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -1621,20 +1621,28 @@ void dcn401_unblank_stream(struct pipe_ctx *pipe_ctx,
 
 void dcn401_hardware_release(struct dc *dc)
 {
-	dc_dmub_srv_fams2_update_config(dc, dc->current_state, false);
+	if (!dc->debug.disable_force_pstate_allow_on_hw_release) {
+		dc_dmub_srv_fams2_update_config(dc, dc->current_state, false);
 
-	/* If pstate unsupported, or still supported
-	 * by firmware, force it supported by dcn
-	 */
-	if (dc->current_state) {
-		if ((!dc->clk_mgr->clks.p_state_change_support ||
-				dc->current_state->bw_ctx.bw.dcn.fams2_global_config.features.bits.enable) &&
-				dc->res_pool->hubbub->funcs->force_pstate_change_control)
-			dc->res_pool->hubbub->funcs->force_pstate_change_control(
-					dc->res_pool->hubbub, true, true);
+		/* If pstate unsupported, or still supported
+		* by firmware, force it supported by dcn
+		*/
+		if (dc->current_state) {
+			if ((!dc->clk_mgr->clks.p_state_change_support ||
+					dc->current_state->bw_ctx.bw.dcn.fams2_global_config.features.bits.enable) &&
+					dc->res_pool->hubbub->funcs->force_pstate_change_control)
+				dc->res_pool->hubbub->funcs->force_pstate_change_control(
+						dc->res_pool->hubbub, true, true);
 
-		dc->current_state->bw_ctx.bw.dcn.clk.p_state_change_support = true;
-		dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, dc->current_state, true);
+			dc->current_state->bw_ctx.bw.dcn.clk.p_state_change_support = true;
+			dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, dc->current_state, true);
+		}
+	} else {
+		if (dc->current_state) {
+			dc->clk_mgr->clks.p_state_change_support = false;
+			dc->clk_mgr->funcs->update_clocks(dc->clk_mgr, dc->current_state, true);
+		}
+		dc_dmub_srv_fams2_update_config(dc, dc->current_state, false);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
index b3988e38d0a6..068c123ea8a8 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
@@ -708,6 +708,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	},
 	.use_max_lb = true,
 	.force_disable_subvp = false,
+	.disable_force_pstate_allow_on_hw_release = false,
 	.exit_idle_opt_for_cursor_updates = true,
 	.using_dml2 = true,
 	.using_dml21 = true,

From 5dc0ec782ef17ed10b87fa0962388040a86602c8 Mon Sep 17 00:00:00 2001
From: Muhammad Ahmed <Muhammad.Ahmed@amd.com>
Date: Thu, 24 Jul 2025 21:50:25 -0400
Subject: [PATCH 0264/1455] drm/amd/display: Adding interface to log hw state
 when underflow happens

[why]
Will help us better debug underflow issues.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Muhammad Ahmed <Muhammad.Ahmed@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c      | 18 +++++++
 drivers/gpu/drm/amd/display/dc/dc.h           | 30 ++++++++++++
 .../display/dc/hubbub/dcn30/dcn30_hubbub.c    | 31 ++++++++++++
 .../display/dc/hubbub/dcn30/dcn30_hubbub.h    |  6 +++
 .../display/dc/hubbub/dcn31/dcn31_hubbub.c    |  2 +
 .../display/dc/hubbub/dcn32/dcn32_hubbub.c    |  2 +
 .../display/dc/hubbub/dcn35/dcn35_hubbub.c    |  2 +
 .../display/dc/hubbub/dcn401/dcn401_hubbub.c  |  2 +
 .../amd/display/dc/hubp/dcn10/dcn10_hubp.h    |  8 +++-
 .../amd/display/dc/hubp/dcn30/dcn30_hubp.c    | 26 ++++++++++
 .../amd/display/dc/hubp/dcn30/dcn30_hubp.h    |  8 +++-
 .../amd/display/dc/hubp/dcn31/dcn31_hubp.c    | 15 ++++++
 .../amd/display/dc/hubp/dcn31/dcn31_hubp.h    |  6 ++-
 .../amd/display/dc/hubp/dcn32/dcn32_hubp.c    |  3 ++
 .../amd/display/dc/hubp/dcn35/dcn35_hubp.c    |  3 ++
 .../amd/display/dc/hubp/dcn401/dcn401_hubp.c  |  3 ++
 .../amd/display/dc/hubp/dcn401/dcn401_hubp.h  |  4 +-
 .../amd/display/dc/hwss/dcn30/dcn30_hwseq.c   | 48 +++++++++++++++++++
 .../amd/display/dc/hwss/dcn30/dcn30_hwseq.h   |  5 ++
 .../amd/display/dc/hwss/dcn30/dcn30_init.c    |  1 +
 .../amd/display/dc/hwss/dcn31/dcn31_init.c    |  1 +
 .../amd/display/dc/hwss/dcn314/dcn314_init.c  |  1 +
 .../amd/display/dc/hwss/dcn32/dcn32_init.c    |  1 +
 .../amd/display/dc/hwss/dcn35/dcn35_init.c    |  1 +
 .../amd/display/dc/hwss/dcn351/dcn351_init.c  |  1 +
 .../amd/display/dc/hwss/dcn401/dcn401_init.c  |  1 +
 .../drm/amd/display/dc/hwss/hw_sequencer.h    |  4 ++
 .../gpu/drm/amd/display/dc/inc/hw/dchubbub.h  |  2 +
 drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h  |  2 +
 .../dc/resource/dcn32/dcn32_resource.h        |  3 +-
 .../dc/resource/dcn401/dcn401_resource.h      |  3 +-
 31 files changed, 236 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index e42cc37eae11..242bcb30dd34 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -6339,3 +6339,21 @@ bool dc_can_clear_cursor_limit(struct dc *dc)
 
 	return false;
 }
+
+void dc_get_underflow_debug_data_for_otg(struct dc *dc, int primary_otg_inst,
+				struct dc_underflow_debug_data *out_data)
+{
+	struct timing_generator *tg = NULL;
+
+	for (int i = 0; i < MAX_PIPES; i++) {
+		if (dc->res_pool->timing_generators[i] &&
+			dc->res_pool->timing_generators[i]->inst == primary_otg_inst) {
+				tg = dc->res_pool->timing_generators[i];
+				break;
+		}
+	}
+
+	dc_exit_ips_for_hw_access(dc);
+	if (dc->hwss.get_underflow_debug_data)
+		dc->hwss.get_underflow_debug_data(dc, tg, out_data);
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index a83227f6bb14..b68fbdd73dca 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1819,6 +1819,23 @@ struct dc_surface_update {
 	struct dc_bias_and_scale bias_and_scale;
 };
 
+struct dc_underflow_debug_data {
+	uint32_t otg_inst;
+	uint32_t otg_underflow;
+	uint32_t h_position;
+	uint32_t v_position;
+	uint32_t otg_frame_count;
+	struct dc_underflow_per_hubp_debug_data {
+		uint32_t hubp_underflow;
+		uint32_t hubp_in_blank;
+		uint32_t hubp_readline;
+		uint32_t det_config_error;
+	} hubps[MAX_PIPES];
+	uint32_t curr_det_sizes[MAX_PIPES];
+	uint32_t target_det_sizes[MAX_PIPES];
+	uint32_t compbuf_config_error;
+};
+
 /*
  * Create a new surface with default parameters;
  */
@@ -2713,4 +2730,17 @@ bool dc_is_timing_changed(struct dc_stream_state *cur_stream,
 bool dc_is_cursor_limit_pending(struct dc *dc);
 bool dc_can_clear_cursor_limit(struct dc *dc);
 
+/**
+ * dc_get_underflow_debug_data_for_otg() - Retrieve underflow debug data.
+ *
+ * @dc: Pointer to the display core context.
+ * @primary_otg_inst: Instance index of the primary OTG that underflowed.
+ * @out_data: Pointer to a dc_underflow_debug_data struct to be filled with debug information.
+ *
+ * This function collects and logs underflow-related HW states when underflow happens,
+ * including OTG underflow status, current read positions, frame count, and per-HUBP debug data.
+ * The results are stored in the provided out_data structure for further analysis or logging.
+ */
+void dc_get_underflow_debug_data_for_otg(struct dc *dc, int primary_otg_inst, struct dc_underflow_debug_data *out_data);
+
 #endif /* DC_INTERFACE_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c
index d347bb06577a..e7e5f6d4778e 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.c
@@ -440,6 +440,35 @@ void hubbub3_init_watermarks(struct hubbub *hubbub)
 	REG_WRITE(DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, reg);
 }
 
+void hubbub3_get_det_sizes(struct hubbub *hubbub, uint32_t *curr_det_sizes, uint32_t *target_det_sizes)
+{
+	struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub);
+
+	REG_GET_2(DCHUBBUB_DET0_CTRL, DET0_SIZE_CURRENT, &curr_det_sizes[0],
+		DET0_SIZE, &target_det_sizes[0]);
+
+	REG_GET_2(DCHUBBUB_DET1_CTRL, DET1_SIZE_CURRENT, &curr_det_sizes[1],
+		DET1_SIZE, &target_det_sizes[1]);
+
+	REG_GET_2(DCHUBBUB_DET2_CTRL, DET2_SIZE_CURRENT, &curr_det_sizes[2],
+		DET2_SIZE, &target_det_sizes[2]);
+
+	REG_GET_2(DCHUBBUB_DET3_CTRL, DET3_SIZE_CURRENT, &curr_det_sizes[3],
+		DET3_SIZE, &target_det_sizes[3]);
+
+}
+
+uint32_t hubbub3_compbuf_config_error(struct hubbub *hubbub)
+{
+	struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub);
+	uint32_t compbuf_config_error = 0;
+
+	REG_GET(DCHUBBUB_COMPBUF_CTRL, CONFIG_ERROR,
+		&compbuf_config_error);
+
+	return compbuf_config_error;
+}
+
 static const struct hubbub_funcs hubbub30_funcs = {
 	.update_dchub = hubbub2_update_dchub,
 	.init_dchub_sys_ctx = hubbub3_init_dchub_sys_ctx,
@@ -457,6 +486,8 @@ static const struct hubbub_funcs hubbub30_funcs = {
 	.force_pstate_change_control = hubbub3_force_pstate_change_control,
 	.init_watermarks = hubbub3_init_watermarks,
 	.hubbub_read_state = hubbub2_read_state,
+	.get_det_sizes = hubbub3_get_det_sizes,
+	.compbuf_config_error = hubbub3_compbuf_config_error,
 };
 
 void hubbub3_construct(struct dcn20_hubbub *hubbub3,
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h
index ca6233e8f1f4..49a469969d36 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn30/dcn30_hubbub.h
@@ -133,4 +133,10 @@ void hubbub3_force_pstate_change_control(struct hubbub *hubbub,
 
 void hubbub3_init_watermarks(struct hubbub *hubbub);
 
+void hubbub3_get_det_sizes(struct hubbub *hubbub,
+	uint32_t *curr_det_sizes,
+	uint32_t *target_det_sizes);
+
+uint32_t hubbub3_compbuf_config_error(struct hubbub *hubbub);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c
index b98505b240a7..cdb20251a154 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn31/dcn31_hubbub.c
@@ -1071,6 +1071,8 @@ static const struct hubbub_funcs hubbub31_funcs = {
 	.program_compbuf_size = dcn31_program_compbuf_size,
 	.init_crb = dcn31_init_crb,
 	.hubbub_read_state = hubbub2_read_state,
+	.get_det_sizes = hubbub3_get_det_sizes,
+	.compbuf_config_error = hubbub3_compbuf_config_error,
 };
 
 void hubbub31_construct(struct dcn20_hubbub *hubbub31,
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
index 32a6be543105..92957398ac0a 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
@@ -1009,6 +1009,8 @@ static const struct hubbub_funcs hubbub32_funcs = {
 	.force_usr_retraining_allow = hubbub32_force_usr_retraining_allow,
 	.set_request_limit = hubbub32_set_request_limit,
 	.get_mall_en = hubbub32_get_mall_en,
+	.get_det_sizes = hubbub3_get_det_sizes,
+	.compbuf_config_error = hubbub3_compbuf_config_error,
 };
 
 void hubbub32_construct(struct dcn20_hubbub *hubbub2,
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c
index 6d41953011f5..a443722a8632 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn35/dcn35_hubbub.c
@@ -589,6 +589,8 @@ static const struct hubbub_funcs hubbub35_funcs = {
 	.hubbub_read_state = hubbub2_read_state,
 	.force_usr_retraining_allow = hubbub32_force_usr_retraining_allow,
 	.dchubbub_init = hubbub35_init,
+	.get_det_sizes = hubbub3_get_det_sizes,
+	.compbuf_config_error = hubbub3_compbuf_config_error,
 };
 
 void hubbub35_construct(struct dcn20_hubbub *hubbub2,
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c
index 92fab471b183..a36273a52880 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn401/dcn401_hubbub.c
@@ -1247,6 +1247,8 @@ static const struct hubbub_funcs hubbub4_01_funcs = {
 	.program_compbuf_segments = dcn401_program_compbuf_segments,
 	.wait_for_det_update = dcn401_wait_for_det_update,
 	.program_arbiter = dcn401_program_arbiter,
+	.get_det_sizes = hubbub3_get_det_sizes,
+	.compbuf_config_error = hubbub3_compbuf_config_error,
 };
 
 void hubbub401_construct(struct dcn20_hubbub *hubbub2,
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h
index 0b7547d5b488..cf2eb9793008 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn10/dcn10_hubp.h
@@ -104,7 +104,8 @@
 	SRI(DCN_SURF1_TTU_CNTL1, HUBPREQ, id),\
 	SRI(DCN_CUR0_TTU_CNTL0, HUBPREQ, id),\
 	SRI(DCN_CUR0_TTU_CNTL1, HUBPREQ, id),\
-	SRI(HUBP_CLK_CNTL, HUBP, id)
+	SRI(HUBP_CLK_CNTL, HUBP, id),\
+	SRI(HUBPRET_READ_LINE_VALUE, HUBPRET, id)
 
 /* Register address initialization macro for ASICs with VM */
 #define HUBP_REG_LIST_DCN_VM(id)\
@@ -249,7 +250,8 @@
 	uint32_t CURSOR_POSITION; \
 	uint32_t CURSOR_HOT_SPOT; \
 	uint32_t CURSOR_DST_OFFSET; \
-	uint32_t HUBP_CLK_CNTL
+	uint32_t HUBP_CLK_CNTL; \
+	uint32_t HUBPRET_READ_LINE_VALUE
 
 #define HUBP_SF(reg_name, field_name, post_fix)\
 	.field_name = reg_name ## __ ## field_name ## post_fix
@@ -622,6 +624,8 @@
 	type DCN_VM_SYSTEM_APERTURE_DEFAULT_SYSTEM;\
 	type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_MSB;\
 	type DCN_VM_SYSTEM_APERTURE_DEFAULT_ADDR_LSB;\
+	type PIPE_READ_LINE;\
+	type HUBP_SEG_ALLOC_ERR_STATUS;\
 	/* todo:  get these from GVM instead of reading registers ourselves */\
 	type PAGE_DIRECTORY_ENTRY_HI32;\
 	type PAGE_DIRECTORY_ENTRY_LO32;\
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c
index 0da70b50e86d..556214b2227d 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.c
@@ -505,6 +505,30 @@ void hubp3_init(struct hubp *hubp)
 	hubp_reset(hubp);
 }
 
+uint32_t hubp3_get_current_read_line(struct hubp *hubp)
+{
+	uint32_t read_line = 0;
+	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
+
+	REG_GET(HUBPRET_READ_LINE_VALUE,
+		PIPE_READ_LINE,
+		&read_line);
+
+	return read_line;
+}
+
+unsigned int hubp3_get_underflow_status(struct hubp *hubp)
+{
+	uint32_t hubp_underflow = 0;
+	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
+
+	REG_GET(DCHUBP_CNTL,
+		HUBP_UNDERFLOW_STATUS,
+		&hubp_underflow);
+
+	return hubp_underflow;
+}
+
 static struct hubp_funcs dcn30_hubp_funcs = {
 	.hubp_enable_tripleBuffer = hubp2_enable_triplebuffer,
 	.hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled,
@@ -534,6 +558,8 @@ static struct hubp_funcs dcn30_hubp_funcs = {
 	.hubp_soft_reset = hubp1_soft_reset,
 	.hubp_set_flip_int = hubp1_set_flip_int,
 	.hubp_clear_tiling = hubp3_clear_tiling,
+	.hubp_get_underflow_status = hubp3_get_underflow_status,
+	.hubp_get_current_read_line = hubp3_get_current_read_line,
 };
 
 bool hubp3_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h
index b7d7adf0b58c..842f4eb72cc8 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn30/dcn30_hubp.h
@@ -243,7 +243,8 @@
 	HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_6, REFCYC_PER_META_CHUNK_FLIP_C, mask_sh),\
 	HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, mask_sh),\
 	HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_6, REFCYC_PER_VM_REQ_VBLANK, mask_sh),\
-	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh)
+	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh),\
+	HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_VALUE, PIPE_READ_LINE, mask_sh)
 
 bool hubp3_construct(
 		struct dcn20_hubp *hubp2,
@@ -299,6 +300,11 @@ void hubp3_init(struct hubp *hubp);
 
 void hubp3_clear_tiling(struct hubp *hubp);
 
+uint32_t hubp3_get_current_read_line(struct hubp *hubp);
+
+uint32_t hubp3_get_underflow_status(struct hubp *hubp);
+
+
 #endif /* __DC_HUBP_DCN30_H__ */
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c
index 7fd582a8a4ba..47101847c2b7 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.c
@@ -68,6 +68,18 @@ void hubp31_program_extended_blank_value(
 	hubp31_program_extended_blank(hubp, min_dst_y_next_start_optimized);
 }
 
+uint32_t hubp31_get_det_config_error(struct hubp *hubp)
+{
+	uint32_t config_error = 0;
+	struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp);
+
+	REG_GET(DCHUBP_CNTL,
+		HUBP_SEG_ALLOC_ERR_STATUS,
+		&config_error);
+
+	return config_error;
+}
+
 static struct hubp_funcs dcn31_hubp_funcs = {
 	.hubp_enable_tripleBuffer = hubp2_enable_triplebuffer,
 	.hubp_is_triplebuffer_enabled = hubp2_is_triplebuffer_enabled,
@@ -98,6 +110,9 @@ static struct hubp_funcs dcn31_hubp_funcs = {
 	.hubp_in_blank = hubp1_in_blank,
 	.program_extended_blank = hubp31_program_extended_blank,
 	.hubp_clear_tiling = hubp3_clear_tiling,
+	.hubp_get_underflow_status = hubp3_get_underflow_status,
+	.hubp_get_current_read_line = hubp3_get_current_read_line,
+	.hubp_get_det_config_error = hubp31_get_det_config_error,
 };
 
 bool hubp31_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h
index d688db79b750..5952c4671507 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn31/dcn31_hubp.h
@@ -228,7 +228,9 @@
 	HUBP_SF(HUBPREQ0_FLIP_PARAMETERS_6, REFCYC_PER_META_CHUNK_FLIP_C, mask_sh),\
 	HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_5, REFCYC_PER_VM_GROUP_VBLANK, mask_sh),\
 	HUBP_SF(HUBPREQ0_VBLANK_PARAMETERS_6, REFCYC_PER_VM_REQ_VBLANK, mask_sh),\
-	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh)
+	HUBP_SF(HUBP0_DCHUBP_REQ_SIZE_CONFIG, VM_GROUP_SIZE, mask_sh),\
+	HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_VALUE, PIPE_READ_LINE, mask_sh),\
+	HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_SEG_ALLOC_ERR_STATUS, mask_sh)
 
 
 bool hubp31_construct(
@@ -246,4 +248,6 @@ void hubp31_set_unbounded_requesting(struct hubp *hubp, bool enable);
 void hubp31_program_extended_blank_value(
 	struct hubp *hubp, unsigned int min_dst_y_next_start_optimized);
 
+uint32_t hubp31_get_det_config_error(struct hubp *hubp);
+
 #endif /* __DC_HUBP_DCN31_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c
index f3a21c623f44..a5f23bb2a76a 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn32/dcn32_hubp.c
@@ -206,6 +206,9 @@ static struct hubp_funcs dcn32_hubp_funcs = {
 	.hubp_update_mall_sel = hubp32_update_mall_sel,
 	.hubp_prepare_subvp_buffering = hubp32_prepare_subvp_buffering,
 	.hubp_clear_tiling = hubp3_clear_tiling,
+	.hubp_get_underflow_status = hubp3_get_underflow_status,
+	.hubp_get_current_read_line = hubp3_get_current_read_line,
+	.hubp_get_det_config_error = hubp31_get_det_config_error,
 };
 
 bool hubp32_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c
index 6d060ba12da8..b140808f21af 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn35/dcn35_hubp.c
@@ -218,6 +218,9 @@ static struct hubp_funcs dcn35_hubp_funcs = {
 	.hubp_in_blank = hubp1_in_blank,
 	.program_extended_blank = hubp31_program_extended_blank_value,
 	.hubp_clear_tiling = hubp3_clear_tiling,
+	.hubp_get_underflow_status = hubp3_get_underflow_status,
+	.hubp_get_current_read_line = hubp3_get_current_read_line,
+	.hubp_get_det_config_error = hubp31_get_det_config_error,
 };
 
 bool hubp35_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
index 5028180ad80a..0fcbc6a35be6 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.c
@@ -1071,6 +1071,9 @@ static struct hubp_funcs dcn401_hubp_funcs = {
 	.hubp_get_3dlut_fl_done = hubp401_get_3dlut_fl_done,
 	.hubp_clear_tiling = hubp401_clear_tiling,
 	.hubp_program_3dlut_fl_config = hubp401_program_3dlut_fl_config,
+	.hubp_get_underflow_status = hubp3_get_underflow_status,
+	.hubp_get_current_read_line = hubp3_get_current_read_line,
+	.hubp_get_det_config_error = hubp31_get_det_config_error,
 };
 
 bool hubp401_construct(
diff --git a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h
index 887b479ed1d7..fdabbeec8ffa 100644
--- a/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/hubp/dcn401/dcn401_hubp.h
@@ -252,7 +252,9 @@
 	HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_1H_P0, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_2H_P0, mask_sh),\
 	HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_1H_P1, mask_sh),\
-	HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_2H_P1, mask_sh)
+	HUBP_SF(HUBP0_DCHUBP_MCACHEID_CONFIG, MCACHEID_MALL_PREF_2H_P1, mask_sh),\
+	HUBP_SF(HUBPRET0_HUBPRET_READ_LINE_VALUE, PIPE_READ_LINE, mask_sh),\
+	HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_SEG_ALLOC_ERR_STATUS, mask_sh)
 
 void hubp401_update_mall_sel(struct hubp *hubp, uint32_t mall_sel, bool c_cursor);
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c
index 37a239219dfe..139a63101488 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c
@@ -1228,3 +1228,51 @@ void dcn30_wait_for_all_pending_updates(const struct pipe_ctx *pipe_ctx)
 		}
 	}
 }
+
+void dcn30_get_underflow_debug_data(const struct dc *dc,
+	struct timing_generator *tg,
+	struct dc_underflow_debug_data *out_data)
+{
+	struct hubbub *hubbub = dc->res_pool->hubbub;
+
+	if (tg) {
+		uint32_t v_blank_start = 0, v_blank_end = 0;
+
+		out_data->otg_inst = tg->inst;
+
+		tg->funcs->get_scanoutpos(tg,
+					  &v_blank_start,
+					  &v_blank_end,
+					  &out_data->h_position,
+					  &out_data->v_position);
+
+		out_data->otg_frame_count = tg->funcs->get_frame_count(tg);
+
+		out_data->otg_underflow = tg->funcs->is_optc_underflow_occurred(tg);
+	}
+
+	for (int i = 0; i < MAX_PIPES; i++) {
+		struct hubp *hubp = dc->res_pool->hubps[i];
+
+		if (hubp) {
+			if (hubp->funcs->hubp_get_underflow_status)
+				out_data->hubps[i].hubp_underflow = hubp->funcs->hubp_get_underflow_status(hubp);
+
+			if (hubp->funcs->hubp_in_blank)
+				out_data->hubps[i].hubp_in_blank = hubp->funcs->hubp_in_blank(hubp);
+
+			if (hubp->funcs->hubp_get_current_read_line)
+				out_data->hubps[i].hubp_readline = hubp->funcs->hubp_get_current_read_line(hubp);
+
+			if (hubp->funcs->hubp_get_det_config_error)
+				out_data->hubps[i].det_config_error = hubp->funcs->hubp_get_det_config_error(hubp);
+		}
+	}
+
+	if (hubbub->funcs->get_det_sizes)
+		hubbub->funcs->get_det_sizes(hubbub, out_data->curr_det_sizes, out_data->target_det_sizes);
+
+	if (hubbub->funcs->compbuf_config_error)
+		out_data->compbuf_config_error = hubbub->funcs->compbuf_config_error(hubbub);
+
+}
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h
index 4b90b781c4f2..40afbbfb5b9c 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h
@@ -29,6 +29,7 @@
 #include "hw_sequencer_private.h"
 
 struct dc;
+struct dc_underflow_debug_data;
 
 void dcn30_init_hw(struct dc *dc);
 void dcn30_program_all_writeback_pipes_in_tree(
@@ -98,4 +99,8 @@ void dcn30_prepare_bandwidth(struct dc *dc,
 
 void dcn30_wait_for_all_pending_updates(const struct pipe_ctx *pipe_ctx);
 
+void dcn30_get_underflow_debug_data(const struct dc *dc,
+	struct timing_generator *tg,
+	struct dc_underflow_debug_data *out_data);
+
 #endif /* __DC_HWSS_DCN30_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c
index 2ac5d54d1626..d7ff55669bac 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c
@@ -110,6 +110,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = {
 	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.is_abm_supported = dcn21_is_abm_supported,
 	.wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn30_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c
index 556f4fe57eda..5a6a459da224 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_init.c
@@ -112,6 +112,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = {
 	.exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state,
 	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.setup_hpo_hw_control = dcn31_setup_hpo_hw_control,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn31_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c
index f5112742edf9..a99145a30230 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c
@@ -115,6 +115,7 @@ static const struct hw_sequencer_funcs dcn314_funcs = {
 	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.calculate_pix_rate_divider = dcn314_calculate_pix_rate_divider,
 	.setup_hpo_hw_control = dcn31_setup_hpo_hw_control,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn314_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c
index b971356d30b1..c19ef075c882 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c
@@ -121,6 +121,7 @@ static const struct hw_sequencer_funcs dcn32_funcs = {
 	.calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider,
 	.program_outstanding_updates = dcn32_program_outstanding_updates,
 	.wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn32_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
index a3ccf805bd16..52cc488416ac 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
@@ -128,6 +128,7 @@ static const struct hw_sequencer_funcs dcn35_funcs = {
 	.enable_plane = dcn20_enable_plane,
 	.update_dchubp_dpp = dcn20_update_dchubp_dpp,
 	.post_unlock_reset_opp = dcn20_post_unlock_reset_opp,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn35_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
index 58f2be2a326b..e34efcb7bde5 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
@@ -123,6 +123,7 @@ static const struct hw_sequencer_funcs dcn351_funcs = {
 	.set_long_vtotal = dcn35_set_long_vblank,
 	.calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider,
 	.setup_hpo_hw_control = dcn35_setup_hpo_hw_control,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn351_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c
index fe7aceb2f510..d6e11b7e4fce 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c
@@ -104,6 +104,7 @@ static const struct hw_sequencer_funcs dcn401_funcs = {
 	.enable_plane = dcn20_enable_plane,
 	.update_dchubp_dpp = dcn20_update_dchubp_dpp,
 	.post_unlock_reset_opp = dcn20_post_unlock_reset_opp,
+	.get_underflow_debug_data = dcn30_get_underflow_debug_data,
 };
 
 static const struct hwseq_private_funcs dcn401_private_funcs = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h
index 9df8030e37f7..1723bbcf2c46 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h
@@ -47,6 +47,7 @@ struct link_resource;
 struct dc_dmub_cmd;
 struct pg_block_update;
 struct drr_params;
+struct dc_underflow_debug_data;
 
 struct subvp_pipe_control_lock_fast_params {
 	struct dc *dc;
@@ -475,6 +476,9 @@ struct hw_sequencer_funcs {
 			struct dc_state *context);
 	void (*post_unlock_reset_opp)(struct dc *dc,
 			struct pipe_ctx *opp_head);
+	void (*get_underflow_debug_data)(const struct dc *dc,
+			struct timing_generator *tg,
+			struct dc_underflow_debug_data *out_data);
 };
 
 void color_space_to_black_color(
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
index 52b745667ef7..9bee45b36629 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
@@ -229,6 +229,8 @@ struct hubbub_funcs {
 	void (*program_compbuf_segments)(struct hubbub *hubbub, unsigned compbuf_size_seg, bool safe_to_increase);
 	void (*wait_for_det_update)(struct hubbub *hubbub, int hubp_inst);
 	bool (*program_arbiter)(struct hubbub *hubbub, struct dml2_display_arb_regs *arb_regs, bool safe_to_lower);
+	void (*get_det_sizes)(struct hubbub *hubbub, uint32_t *curr_det_sizes, uint32_t *target_det_sizes);
+	uint32_t (*compbuf_config_error)(struct hubbub *hubbub);
 };
 
 struct hubbub {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
index 198a28bd8e28..2b874d2cc61c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h
@@ -306,6 +306,8 @@ struct hubp_funcs {
 	int (*hubp_get_3dlut_fl_done)(struct hubp *hubp);
 	void (*hubp_program_3dlut_fl_config)(struct hubp *hubp, struct hubp_fl_3dlut_config *cfg);
 	void (*hubp_clear_tiling)(struct hubp *hubp);
+	uint32_t (*hubp_get_current_read_line)(struct hubp *hubp);
+	uint32_t (*hubp_get_det_config_error)(struct hubp *hubp);
 };
 
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
index 82f966cf4ed2..20d714596021 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
@@ -1141,7 +1141,8 @@ unsigned int dcn32_get_max_hw_cursor_size(const struct dc *dc,
       SRI_ARR(DCN_SURF1_TTU_CNTL1, HUBPREQ, id),                               \
       SRI_ARR(DCN_CUR0_TTU_CNTL0, HUBPREQ, id),                                \
       SRI_ARR(DCN_CUR0_TTU_CNTL1, HUBPREQ, id),                                \
-      SRI_ARR(HUBP_CLK_CNTL, HUBP, id)
+      SRI_ARR(HUBP_CLK_CNTL, HUBP, id),                                        \
+      SRI_ARR(HUBPRET_READ_LINE_VALUE, HUBPRET, id)
 #define HUBP_REG_LIST_DCN2_COMMON_RI(id)                                       \
   HUBP_REG_LIST_DCN_RI(id), HUBP_REG_LIST_DCN_VM_RI(id),                       \
       SRI_ARR(PREFETCH_SETTINGS, HUBPREQ, id),                                 \
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h
index 2ae6831c31ef..0fc66487d800 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h
@@ -140,7 +140,8 @@ void dcn401_prepare_mcache_programming(struct dc *dc, struct dc_state *context);
 	SRI_ARR(UCLK_PSTATE_FORCE, HUBPREQ, id),                                 \
 	HUBP_3DLUT_FL_REG_LIST_DCN401(id),                                       \
 	SRI_ARR(DCSURF_VIEWPORT_MCACHE_SPLIT_COORDINATE, HUBP, id),              \
-	SRI_ARR(DCHUBP_MCACHEID_CONFIG, HUBP, id)
+	SRI_ARR(DCHUBP_MCACHEID_CONFIG, HUBP, id),								 \
+	SRI_ARR(HUBPRET_READ_LINE_VALUE, HUBPRET, id)
 
 /* ABM */
 #define ABM_DCN401_REG_LIST_RI(id)                                            \

From 712d98c7da689447596e388b4fb05ec4fa74d70c Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 25 Jul 2025 17:14:58 -0500
Subject: [PATCH 0265/1455] drm/amd/display: Promote DC to 3.2.344

Summary:
* Add interface to log hw state when underflow happens
* Fix hubp programming of 3dlut fast load
* Avoid Read Remote DPCD Many Times
* More liberal vmin/vmax update for freesync
* Fix dmub access race condition

Acked-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Roman Li <roman.li@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index b68fbdd73dca..18c68e531f71 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.343"
+#define DC_VER "3.2.344"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 582bf7c5158dce16f7dc5b8345b7876bd8031224 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 18 Jul 2025 09:25:21 +0530
Subject: [PATCH 0266/1455] drm/amdgpu: Add NULL check for asic_funcs

If driver load fails too early, asic_funcs pointer remains unassigned.
Add NULL check to sanitize unwind path.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c
index e56ba93a8df6..a974265837f0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.c
@@ -55,7 +55,8 @@ u64 amdgpu_nbio_get_pcie_replay_count(struct amdgpu_device *adev)
 
 bool amdgpu_nbio_is_replay_cnt_supported(struct amdgpu_device *adev)
 {
-	if (amdgpu_sriov_vf(adev) || !adev->asic_funcs->get_pcie_replay_count ||
+	if (amdgpu_sriov_vf(adev) || !adev->asic_funcs ||
+	    !adev->asic_funcs->get_pcie_replay_count ||
 	    (!adev->nbio.funcs || !adev->nbio.funcs->get_pcie_replay_count))
 		return false;
 

From e87577ef6daa0cfb10ca139c720f0c57bd894174 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 11 Jul 2025 12:15:45 +0530
Subject: [PATCH 0267/1455] drm/amd/pm: Use cached metrics data on aldebaran

Cached metrics data validity is 1ms on aldebaran. It's not reasonable
for any client to query gpu_metrics at a faster rate and constantly
interrupt PMFW.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
index c63d2e28954d..b067147b7c41 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
@@ -1781,7 +1781,7 @@ static ssize_t aldebaran_get_gpu_metrics(struct smu_context *smu,
 
 	ret = smu_cmn_get_metrics_table(smu,
 					&metrics,
-					true);
+					false);
 	if (ret)
 		return ret;
 

From 2f3b1ccf83be83a3330e38194ddfd1a91fec69be Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 11 Jul 2025 12:18:04 +0530
Subject: [PATCH 0268/1455] drm/amd/pm: Use cached metrics data on arcturus

Cached metrics data validity is 1ms on arcturus. It's not reasonable for
any client to query gpu_metrics at a faster rate and constantly
interrupt PMFW.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
index 9ad46f545d15..599eddb5a67d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
@@ -1897,7 +1897,7 @@ static ssize_t arcturus_get_gpu_metrics(struct smu_context *smu,
 
 	ret = smu_cmn_get_metrics_table(smu,
 					&metrics,
-					true);
+					false);
 	if (ret)
 		return ret;
 

From 92e2449241516c95aab95eea91faecd0fa2b7ed5 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Mon, 4 Aug 2025 08:43:15 +0800
Subject: [PATCH 0269/1455] drm/amdgpu: Update SDMA firmware version check for
 user queue support

This commit fixes a firmware version check for enabling user queue
support in SDMA v7.0. The previous version check (7836028) was
incorrect and could lead to issues with PROTECTED_FENCE_SIGNAL
commands causing register conflicts between MCU_DBG0 and MCU_DBG1.

Fixes: 8c011408ed84 ("drm/amdgpu/sdma7: add ucode version checks for userq support")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
index b8b06d4c5882..326ecc8d37d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c
@@ -1353,7 +1353,7 @@ static int sdma_v7_0_sw_init(struct amdgpu_ip_block *ip_block)
 	switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) {
 	case IP_VERSION(7, 0, 0):
 	case IP_VERSION(7, 0, 1):
-		if ((adev->sdma.instance[0].fw_version >= 7836028) && !adev->sdma.disable_uq)
+		if ((adev->sdma.instance[0].fw_version >= 7966358) && !adev->sdma.disable_uq)
 			adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
 		break;
 	default:

From 9b331f0f60832c01939b4c46f66d24eca3917c3d Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 1 Aug 2025 23:42:01 +0530
Subject: [PATCH 0270/1455] drm/amd/pm: Allow static metrics table query in VF

Allow statics metrics table to be queried on SMUv13.0.6 SOCs in VF mode.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 9cc294f4708b..148941d7ba51 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -177,7 +177,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU
 	MSG_MAP(SetThrottlingPolicy,                 PPSMC_MSG_SetThrottlingPolicy,             0),
 	MSG_MAP(ResetSDMA,                           PPSMC_MSG_ResetSDMA,                       0),
 	MSG_MAP(ResetVCN,                            PPSMC_MSG_ResetVCN,                       0),
-	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           0),
+	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           1),
 };
 
 // clang-format on

From 5c2b3226d00b5da63080ad0387481f66a024c635 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 24 Jul 2025 12:52:56 +0530
Subject: [PATCH 0271/1455] drm/amdgpu: Add wrapper function for dpc state

Use wrapper functions to set/indicate dpc status.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Ce Sun <cesun102@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 17 ++++++++---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h  | 11 +++++++++++
 2 files changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 5949cc747c9f..7c7eecdd68d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5699,7 +5699,7 @@ int amdgpu_device_link_reset(struct amdgpu_device *adev)
 
 	dev_info(adev->dev, "GPU link reset\n");
 
-	if (!adev->pcie_reset_ctx.occurs_dpc)
+	if (!amdgpu_reset_in_dpc(adev))
 		ret = amdgpu_dpm_link_reset(adev);
 
 	if (ret)
@@ -6150,7 +6150,7 @@ static void amdgpu_device_recovery_prepare(struct amdgpu_device *adev,
 			list_add_tail(&tmp_adev->reset_list, device_list);
 			if (adev->shutdown)
 				tmp_adev->shutdown = true;
-			if (adev->pcie_reset_ctx.occurs_dpc)
+			if (amdgpu_reset_in_dpc(adev))
 				tmp_adev->pcie_reset_ctx.in_link_reset = true;
 		}
 		if (!list_is_first(&adev->reset_list, device_list))
@@ -6226,9 +6226,8 @@ static void amdgpu_device_halt_activities(struct amdgpu_device *adev,
 		drm_client_dev_suspend(adev_to_drm(tmp_adev), false);
 
 		/* disable ras on ALL IPs */
-		if (!need_emergency_restart &&
-		      (!adev->pcie_reset_ctx.occurs_dpc) &&
-		      amdgpu_device_ip_need_full_reset(tmp_adev))
+		if (!need_emergency_restart && !amdgpu_reset_in_dpc(adev) &&
+		    amdgpu_device_ip_need_full_reset(tmp_adev))
 			amdgpu_ras_suspend(tmp_adev);
 
 		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
@@ -6256,10 +6255,10 @@ static int amdgpu_device_asic_reset(struct amdgpu_device *adev,
 
 retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 	list_for_each_entry(tmp_adev, device_list, reset_list) {
-		if (adev->pcie_reset_ctx.occurs_dpc)
+		if (amdgpu_reset_in_dpc(adev))
 			tmp_adev->no_hw_access = true;
 		r = amdgpu_device_pre_asic_reset(tmp_adev, reset_context);
-		if (adev->pcie_reset_ctx.occurs_dpc)
+		if (amdgpu_reset_in_dpc(adev))
 			tmp_adev->no_hw_access = false;
 		/*TODO Should we stop ?*/
 		if (r) {
@@ -6901,7 +6900,7 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 
 		if (hive)
 			mutex_lock(&hive->hive_lock);
-		adev->pcie_reset_ctx.occurs_dpc = true;
+		amdgpu_reset_set_dpc_status(adev, true);
 		memset(&reset_context, 0, sizeof(reset_context));
 		INIT_LIST_HEAD(&device_list);
 
@@ -7064,7 +7063,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 	amdgpu_device_sched_resume(&device_list, NULL, NULL);
 	amdgpu_device_gpu_resume(adev, &device_list, false);
 	amdgpu_device_recovery_put_reset_lock(adev, &device_list);
-	adev->pcie_reset_ctx.occurs_dpc = false;
+	amdgpu_reset_set_dpc_status(adev, false);
 
 	if (hive) {
 		mutex_unlock(&hive->hive_lock);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
index 4d9b9701139b..3a806953338f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
@@ -160,4 +160,15 @@ int amdgpu_reset_do_xgmi_reset_on_init(
 
 bool amdgpu_reset_in_recovery(struct amdgpu_device *adev);
 
+static inline void amdgpu_reset_set_dpc_status(struct amdgpu_device *adev,
+					       bool status)
+{
+	adev->pcie_reset_ctx.occurs_dpc = status;
+}
+
+static inline bool amdgpu_reset_in_dpc(struct amdgpu_device *adev)
+{
+	return adev->pcie_reset_ctx.occurs_dpc;
+}
+
 #endif

From d3d73bdb02e8cc4a1b2b721a42908504cd18ebf9 Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Thu, 31 Jul 2025 14:28:26 +0800
Subject: [PATCH 0272/1455] drm/amdgpu: Fix jpeg v4.0.3 poison irq call trace
 on sriov guest

Sriov guest side doesn't init ras feature hence the poison irq shouldn't
be put during hw fini.

[25209.467154] Call Trace:
[25209.467156]  <TASK>
[25209.467158]  ? srso_alias_return_thunk+0x5/0x7f
[25209.467162]  ? show_trace_log_lvl+0x28e/0x2ea
[25209.467166]  ? show_trace_log_lvl+0x28e/0x2ea
[25209.467171]  ? jpeg_v4_0_3_hw_fini+0x6f/0x90 [amdgpu]
[25209.467300]  ? show_regs.part.0+0x23/0x29
[25209.467303]  ? show_regs.cold+0x8/0xd
[25209.467304]  ? amdgpu_irq_put+0x9e/0xc0 [amdgpu]
[25209.467403]  ? __warn+0x8c/0x100
[25209.467407]  ? amdgpu_irq_put+0x9e/0xc0 [amdgpu]
[25209.467503]  ? report_bug+0xa4/0xd0
[25209.467508]  ? handle_bug+0x39/0x90
[25209.467511]  ? exc_invalid_op+0x19/0x70
[25209.467513]  ? asm_exc_invalid_op+0x1b/0x20
[25209.467518]  ? amdgpu_irq_put+0x9e/0xc0 [amdgpu]
[25209.467613]  ? amdgpu_irq_put+0x5f/0xc0 [amdgpu]
[25209.467709]  jpeg_v4_0_3_hw_fini+0x6f/0x90 [amdgpu]
[25209.467805]  amdgpu_ip_block_hw_fini+0x34/0x61 [amdgpu]
[25209.467971]  amdgpu_device_fini_hw+0x3b3/0x467 [amdgpu]

Fixes: 1b2231de4163 ("drm/amdgpu: Register aqua vanjaram jpeg poison irq")
Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index b86288a69e7b..a78144773fab 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -444,7 +444,7 @@ static int jpeg_v4_0_3_hw_fini(struct amdgpu_ip_block *ip_block)
 			ret = jpeg_v4_0_3_set_powergating_state(ip_block, AMD_PG_STATE_GATE);
 	}
 
-	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG))
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG) && !amdgpu_sriov_vf(adev))
 		amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0);
 
 	return ret;

From 58364f01db4a155356f92cce1474761d7a0eda3d Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Thu, 31 Jul 2025 14:54:50 +0800
Subject: [PATCH 0273/1455] drm/amdgpu: Fix vcn v4.0.3 poison irq call trace on
 sriov guest

Sriov guest side doesn't init ras feature hence the poison irq shouldn't
be put during hw fini.

[25209.468816] Call Trace:
[25209.468817]  <TASK>
[25209.468818]  ? srso_alias_return_thunk+0x5/0x7f
[25209.468820]  ? show_trace_log_lvl+0x28e/0x2ea
[25209.468822]  ? show_trace_log_lvl+0x28e/0x2ea
[25209.468825]  ? vcn_v4_0_3_hw_fini+0xaf/0xe0 [amdgpu]
[25209.468936]  ? show_regs.part.0+0x23/0x29
[25209.468939]  ? show_regs.cold+0x8/0xd
[25209.468940]  ? amdgpu_irq_put+0x9e/0xc0 [amdgpu]
[25209.469038]  ? __warn+0x8c/0x100
[25209.469040]  ? amdgpu_irq_put+0x9e/0xc0 [amdgpu]
[25209.469135]  ? report_bug+0xa4/0xd0
[25209.469138]  ? handle_bug+0x39/0x90
[25209.469140]  ? exc_invalid_op+0x19/0x70
[25209.469142]  ? asm_exc_invalid_op+0x1b/0x20
[25209.469146]  ? amdgpu_irq_put+0x9e/0xc0 [amdgpu]
[25209.469241]  vcn_v4_0_3_hw_fini+0xaf/0xe0 [amdgpu]
[25209.469343]  amdgpu_ip_block_hw_fini+0x34/0x61 [amdgpu]
[25209.469511]  amdgpu_device_fini_hw+0x3b3/0x467 [amdgpu]

Fixes: 4c4a89149608 ("drm/amdgpu: Register aqua vanjaram vcn poison irq")
Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Stanley.Yang <Stanley.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index c51dc401226a..c6450ed65c12 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -384,7 +384,7 @@ static int vcn_v4_0_3_hw_fini(struct amdgpu_ip_block *ip_block)
 			vinst->set_pg_state(vinst, AMD_PG_STATE_GATE);
 	}
 
-	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN))
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN) && !amdgpu_sriov_vf(adev))
 		amdgpu_irq_put(adev, &adev->vcn.inst->ras_poison_irq, 0);
 
 	return 0;

From 1c2efae2f855e348d772c071c8c30217d8a0dbae Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 4 Aug 2025 10:34:06 +0530
Subject: [PATCH 0274/1455] drm/amd/pm: Make static table support conditional

Add PMFW version check for static table support on SMU v13.0.6 VFs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 28 ++++++++++++++-----
 1 file changed, 21 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 148941d7ba51..a0ce108aab44 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -402,14 +402,28 @@ static void smu_v13_0_6_init_caps(struct smu_context *smu)
 		if ((pgm == 7 && fw_ver >= 0x7550E00) ||
 		    (pgm == 0 && fw_ver >= 0x00557E00))
 			smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS));
-		if ((pgm == 0 && fw_ver >= 0x00557F01) ||
-		    (pgm == 7 && fw_ver >= 0x7551000)) {
-			smu_v13_0_6_cap_set(smu, SMU_CAP(STATIC_METRICS));
-			smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE));
+
+		if (amdgpu_sriov_vf(adev)) {
+			if ((pgm == 0 && fw_ver >= 0x00558000) ||
+			    (pgm == 7 && fw_ver >= 0x7551000)) {
+				smu_v13_0_6_cap_set(smu,
+						    SMU_CAP(STATIC_METRICS));
+				smu_v13_0_6_cap_set(smu,
+						    SMU_CAP(BOARD_VOLTAGE));
+				smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION));
+			}
+		} else {
+			if ((pgm == 0 && fw_ver >= 0x00557F01) ||
+			    (pgm == 7 && fw_ver >= 0x7551000)) {
+				smu_v13_0_6_cap_set(smu,
+						    SMU_CAP(STATIC_METRICS));
+				smu_v13_0_6_cap_set(smu,
+						    SMU_CAP(BOARD_VOLTAGE));
+			}
+			if ((pgm == 0 && fw_ver >= 0x00558000) ||
+			    (pgm == 7 && fw_ver >= 0x7551000))
+				smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION));
 		}
-		if ((pgm == 0 && fw_ver >= 0x00558000) ||
-		    (pgm == 7 && fw_ver >= 0x7551000))
-			smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION));
 	}
 	if (((pgm == 7) && (fw_ver >= 0x7550700)) ||
 	    ((pgm == 0) && (fw_ver >= 0x00557900)) ||

From 73309ed9d598d35576431c4691813e080980df68 Mon Sep 17 00:00:00 2001
From: Nemesa Garg <nemesa.garg@intel.com>
Date: Fri, 1 Aug 2025 18:28:35 +0530
Subject: [PATCH 0275/1455] drm/i915/display: WA_14011503117

Mask the ERR_FATAL_MASK before scaler initialization.
After enabling the scaler and waiting for one frame,
unmask the previously masked bits, PS_ECC and
ERR_FATAL_MASK
Unmasking of ERR_FATAL_MASK bit is use for
validation purpose. There is no functional
impact.

v2: Remove intel_display_need_wa[Jani]
    Optimize the ecc_unmask call[Animesh]
v3: Add intel_display_wa[Jani]

Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
Reviewed-by: Animesh Manna <animesh.manna@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250801125835.2337614-1-nemesa.garg@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c  |  6 ++++
 .../gpu/drm/i915/display/intel_display_wa.c   |  8 +++++
 .../gpu/drm/i915/display/intel_display_wa.h   |  1 +
 drivers/gpu/drm/i915/display/skl_scaler.c     | 29 +++++++++++++++++++
 drivers/gpu/drm/i915/display/skl_scaler.h     |  3 ++
 5 files changed, 47 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index 7035c1fc9033..c1a3a95c65f0 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -76,6 +76,7 @@
 #include "intel_display_regs.h"
 #include "intel_display_rpm.h"
 #include "intel_display_types.h"
+#include "intel_display_wa.h"
 #include "intel_dmc.h"
 #include "intel_dp.h"
 #include "intel_dp_link_training.h"
@@ -1081,6 +1082,11 @@ static void intel_post_plane_update(struct intel_atomic_state *state,
 	if (audio_enabling(old_crtc_state, new_crtc_state))
 		intel_encoders_audio_enable(state, crtc);
 
+	if (intel_display_wa(display, 14011503117)) {
+		if (old_crtc_state->pch_pfit.enabled != new_crtc_state->pch_pfit.enabled)
+			adl_scaler_ecc_unmask(new_crtc_state);
+	}
+
 	intel_alpm_post_plane_update(state, crtc);
 
 	intel_psr_post_plane_update(state, crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.c b/drivers/gpu/drm/i915/display/intel_display_wa.c
index 399c08902413..31cd2c9cd488 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.c
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.c
@@ -52,6 +52,12 @@ static bool intel_display_needs_wa_16025573575(struct intel_display *display)
 	return DISPLAY_VERx100(display) == 3000 || DISPLAY_VERx100(display) == 3002;
 }
 
+/*
+ * Wa_14011503117:
+ * Fixes: Before enabling the scaler DE fatal error is masked
+ * Workaround: Unmask the DE fatal error register after enabling the scaler
+ * and after waiting of at least 1 frame.
+ */
 bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name)
 {
 	switch (wa) {
@@ -59,6 +65,8 @@ bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa,
 		return intel_display_needs_wa_16023588340(display);
 	case INTEL_DISPLAY_WA_16025573575:
 		return intel_display_needs_wa_16025573575(display);
+	case INTEL_DISPLAY_WA_14011503117:
+		return DISPLAY_VER(display) == 13;
 	default:
 		drm_WARN(display->drm, 1, "Missing Wa number: %s\n", name);
 		break;
diff --git a/drivers/gpu/drm/i915/display/intel_display_wa.h b/drivers/gpu/drm/i915/display/intel_display_wa.h
index aedea4cfa3ce..abc1df83f066 100644
--- a/drivers/gpu/drm/i915/display/intel_display_wa.h
+++ b/drivers/gpu/drm/i915/display/intel_display_wa.h
@@ -24,6 +24,7 @@ bool intel_display_needs_wa_16023588340(struct intel_display *display);
 enum intel_display_wa {
 	INTEL_DISPLAY_WA_16023588340,
 	INTEL_DISPLAY_WA_16025573575,
+	INTEL_DISPLAY_WA_14011503117,
 };
 
 bool __intel_display_wa(struct intel_display *display, enum intel_display_wa wa, const char *name);
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index 4cc55f4e1f9f..1374fa94ff5c 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -10,6 +10,7 @@
 #include "intel_display_regs.h"
 #include "intel_display_trace.h"
 #include "intel_display_types.h"
+#include "intel_display_wa.h"
 #include "intel_fb.h"
 #include "skl_scaler.h"
 #include "skl_universal_plane.h"
@@ -762,6 +763,9 @@ void skl_pfit_enable(const struct intel_crtc_state *crtc_state)
 			crtc_state->scaler_state.scaler_id < 0))
 		return;
 
+	if (intel_display_wa(display, 14011503117))
+		adl_scaler_ecc_mask(crtc_state);
+
 	drm_rect_init(&src, 0, 0,
 		      drm_rect_width(&crtc_state->pipe_src) << 16,
 		      drm_rect_height(&crtc_state->pipe_src) << 16);
@@ -938,3 +942,28 @@ void skl_scaler_get_config(struct intel_crtc_state *crtc_state)
 	else
 		scaler_state->scaler_users &= ~(1 << SKL_CRTC_INDEX);
 }
+
+void adl_scaler_ecc_mask(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+
+	if (!crtc_state->pch_pfit.enabled)
+		return;
+
+	intel_de_write(display, XELPD_DISPLAY_ERR_FATAL_MASK, ~0);
+}
+
+void adl_scaler_ecc_unmask(const struct intel_crtc_state *crtc_state)
+{
+	struct intel_display *display = to_intel_display(crtc_state);
+	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
+	const struct intel_crtc_scaler_state *scaler_state =
+		&crtc_state->scaler_state;
+	int id;
+
+	if (!scaler_state && scaler_state->scaler_id == -1)
+		return;
+
+	intel_de_write_fw(display, SKL_PS_ECC_STAT(crtc->pipe, id), 1);
+	intel_de_write(display, XELPD_DISPLAY_ERR_FATAL_MASK, 0);
+}
diff --git a/drivers/gpu/drm/i915/display/skl_scaler.h b/drivers/gpu/drm/i915/display/skl_scaler.h
index e3b35d2c13be..12a19016c5f6 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.h
+++ b/drivers/gpu/drm/i915/display/skl_scaler.h
@@ -42,4 +42,7 @@ skl_scaler_mode_valid(struct intel_display *display,
 		      enum intel_output_format output_format,
 		      int num_joined_pipes);
 
+void adl_scaler_ecc_mask(const struct intel_crtc_state *crtc_state);
+
+void adl_scaler_ecc_unmask(const struct intel_crtc_state *crtc_state);
 #endif

From 7c9de25efa7677acd4feae905165571be05b6f29 Mon Sep 17 00:00:00 2001
From: Jonathan Cavitt <jonathan.cavitt@intel.com>
Date: Mon, 28 Jul 2025 19:48:06 +0000
Subject: [PATCH 0276/1455] drm/xe/xe_guc_ads: Consolidate guc_waklv_enable
 functions

Presently, multiple versions of the guc_waklv_enable_.* function exist,
all with different numbers of dwords added to the klv_entry array.  This
is not extensible, and more duplicates of the function will need to be
created if it ever becomes necessary to support 3 or more dwords per wa
in the future.

Consolidate the disparate guc_waklv_enable functions into a single
guc_waklv_enable function that can take an arbitrary number of dword
values.

v2:
- Update length value properly (Shuicheng)

v3: (Harrison)
- Use data as a term instead of dwords or arr
- Reformat warning message to use hex values
- Eliminate need for kzalloc and klv_entry array
- Reorder function parameters to fix line wrapping

v4:
- Miscellaneous formatting fixes (Cavitt)

v5: (Harrison)
- s/data_range/data_len_dw
- Use data_len_dw to calculate size for xe_map_memcpy_to

Suggested-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Cc: Lucas De Marchi <lucas.demarch@intel.com>
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Cc: John Harrison <john.c.harrison@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250728194806.68176-2-jonathan.cavitt@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_guc_ads.c | 138 +++++++++-----------------------
 1 file changed, 40 insertions(+), 98 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
index 8ff8626227ae..2ceaa197cb2f 100644
--- a/drivers/gpu/drm/xe/xe_guc_ads.c
+++ b/drivers/gpu/drm/xe/xe_guc_ads.c
@@ -284,81 +284,26 @@ static size_t calculate_golden_lrc_size(struct xe_guc_ads *ads)
 	return total_size;
 }
 
-static void guc_waklv_enable_two_word(struct xe_guc_ads *ads,
-				      enum xe_guc_klv_ids klv_id,
-				      u32 value1,
-				      u32 value2,
-				      u32 *offset, u32 *remain)
+static void guc_waklv_enable(struct xe_guc_ads *ads,
+			     u32 data[], u32 data_len_dw,
+			     u32 *offset, u32 *remain,
+			     enum xe_guc_klv_ids klv_id)
 {
-	u32 size;
-	u32 klv_entry[] = {
-			/* 16:16 key/length */
-			FIELD_PREP(GUC_KLV_0_KEY, klv_id) |
-			FIELD_PREP(GUC_KLV_0_LEN, 2),
-			value1,
-			value2,
-			/* 2 dword data */
-	};
-
-	size = sizeof(klv_entry);
+	size_t size = sizeof(u32) * (1 + data_len_dw);
 
 	if (*remain < size) {
 		drm_warn(&ads_to_xe(ads)->drm,
-			 "w/a klv buffer too small to add klv id %d\n", klv_id);
-	} else {
-		xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), *offset,
-				 klv_entry, size);
-		*offset += size;
-		*remain -= size;
-	}
-}
-
-static void guc_waklv_enable_one_word(struct xe_guc_ads *ads,
-				      enum xe_guc_klv_ids klv_id,
-				      u32 value,
-				      u32 *offset, u32 *remain)
-{
-	u32 size;
-	u32 klv_entry[] = {
-		/* 16:16 key/length */
-		FIELD_PREP(GUC_KLV_0_KEY, klv_id) |
-		FIELD_PREP(GUC_KLV_0_LEN, 1),
-		value,
-		/* 1 dword data */
-	};
-
-	size = sizeof(klv_entry);
-
-	if (*remain < size) {
-		drm_warn(&ads_to_xe(ads)->drm,
-			 "w/a klv buffer too small to add klv id %d\n", klv_id);
-	} else {
-		xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), *offset,
-				 klv_entry, size);
-		*offset += size;
-		*remain -= size;
-	}
-}
-
-static void guc_waklv_enable_simple(struct xe_guc_ads *ads,
-				    enum xe_guc_klv_ids klv_id, u32 *offset, u32 *remain)
-{
-	u32 klv_entry[] = {
-		/* 16:16 key/length */
-		FIELD_PREP(GUC_KLV_0_KEY, klv_id) |
-		FIELD_PREP(GUC_KLV_0_LEN, 0),
-		/* 0 dwords data */
-	};
-	u32 size;
-
-	size = sizeof(klv_entry);
-
-	if (xe_gt_WARN(ads_to_gt(ads), *remain < size,
-		       "w/a klv buffer too small to add klv id %d\n", klv_id))
+			 "w/a klv buffer too small to add klv id 0x%04X\n", klv_id);
 		return;
+	}
+
+	/* 16:16 key/length */
+	xe_map_wr(ads_to_xe(ads), ads_to_map(ads), *offset, u32,
+		  FIELD_PREP(GUC_KLV_0_KEY, klv_id) | FIELD_PREP(GUC_KLV_0_LEN, data_len_dw));
+	/* data_len_dw dwords of data */
+	xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads),
+			 *offset + sizeof(u32), data, data_len_dw * sizeof(u32));
 
-	xe_map_memcpy_to(ads_to_xe(ads), ads_to_map(ads), *offset,
-			 klv_entry, size);
 	*offset += size;
 	*remain -= size;
 }
@@ -373,49 +318,46 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 	remain = guc_ads_waklv_size(ads);
 
 	if (XE_WA(gt, 14019882105) || XE_WA(gt, 16021333562))
-		guc_waklv_enable_simple(ads,
-					GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED,
-					&offset, &remain);
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED);
 	if (XE_WA(gt, 18024947630))
-		guc_waklv_enable_simple(ads,
-					GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING,
-					&offset, &remain);
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING);
 	if (XE_WA(gt, 16022287689))
-		guc_waklv_enable_simple(ads,
-					GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE,
-					&offset, &remain);
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE);
 
 	if (XE_WA(gt, 14022866841))
-		guc_waklv_enable_simple(ads,
-					GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO,
-					&offset, &remain);
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO);
 
 	/*
 	 * On RC6 exit, GuC will write register 0xB04 with the default value provided. As of now,
 	 * the default value for this register is determined to be 0xC40. This could change in the
 	 * future, so GuC depends on KMD to send it the correct value.
 	 */
-	if (XE_WA(gt, 13011645652))
-		guc_waklv_enable_one_word(ads,
-					  GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE,
-					  0xC40,
-					  &offset, &remain);
+	if (XE_WA(gt, 13011645652)) {
+		u32 data = 0xC40;
+
+		guc_waklv_enable(ads, &data, sizeof(data) / sizeof(u32), &offset, &remain,
+				 GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE);
+	}
 
 	if (XE_WA(gt, 14022293748) || XE_WA(gt, 22019794406))
-		guc_waklv_enable_simple(ads,
-					GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET,
-					&offset, &remain);
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET);
 
 	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_WA(gt, 16026508708))
-		guc_waklv_enable_simple(ads,
-					GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH,
-					&offset, &remain);
-	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 47, 0) && XE_WA(gt, 16026007364))
-		guc_waklv_enable_two_word(ads,
-					  GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG,
-					  0x0,
-					  0xF,
-					  &offset, &remain);
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH);
+	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 47, 0) && XE_WA(gt, 16026007364)) {
+		u32 data[] = {
+			0x0,
+			0xF,
+		};
+		guc_waklv_enable(ads, data, sizeof(data) / sizeof(u32), &offset, &remain,
+				 GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG);
+	}
 
 	size = guc_ads_waklv_size(ads) - remain;
 	if (!size)

From 90759cddaceab1a7cfd0128d9421abf2d9288d09 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:29 +0200
Subject: [PATCH 0277/1455] drm/xe: Simplify module initialization code

There is no need to have extra checks and WARN() in the helpers
as instead of an index of the entry with function pointers, we
can pass pointer to the entry which we prepare directly in the
main loop, that is guaranteed to be valid.

  add/remove: 0/0 grow/shrink: 0/4 up/down: 0/-180 (-180)
  Function                                     old     new   delta
  xe_exit                                      109      79     -30
  cleanup_module                               109      79     -30
  xe_init                                      248     188     -60
  init_module                                  248     188     -60
  Total: Before=2774145, After=2773965, chg -0.01%

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-2-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_module.c | 27 ++++++++++-----------------
 1 file changed, 10 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c
index d9391bd08194..cfed36361613 100644
--- a/drivers/gpu/drm/xe/xe_module.c
+++ b/drivers/gpu/drm/xe/xe_module.c
@@ -135,24 +135,17 @@ static const struct init_funcs init_funcs[] = {
 	},
 };
 
-static int __init xe_call_init_func(unsigned int i)
+static int __init xe_call_init_func(const struct init_funcs *func)
 {
-	if (WARN_ON(i >= ARRAY_SIZE(init_funcs)))
-		return 0;
-	if (!init_funcs[i].init)
-		return 0;
-
-	return init_funcs[i].init();
+	if (func->init)
+		return func->init();
+	return 0;
 }
 
-static void xe_call_exit_func(unsigned int i)
+static void xe_call_exit_func(const struct init_funcs *func)
 {
-	if (WARN_ON(i >= ARRAY_SIZE(init_funcs)))
-		return;
-	if (!init_funcs[i].exit)
-		return;
-
-	init_funcs[i].exit();
+	if (func->exit)
+		func->exit();
 }
 
 static int __init xe_init(void)
@@ -160,10 +153,10 @@ static int __init xe_init(void)
 	int err, i;
 
 	for (i = 0; i < ARRAY_SIZE(init_funcs); i++) {
-		err = xe_call_init_func(i);
+		err = xe_call_init_func(init_funcs + i);
 		if (err) {
 			while (i--)
-				xe_call_exit_func(i);
+				xe_call_exit_func(init_funcs + i);
 			return err;
 		}
 	}
@@ -176,7 +169,7 @@ static void __exit xe_exit(void)
 	int i;
 
 	for (i = ARRAY_SIZE(init_funcs) - 1; i >= 0; i--)
-		xe_call_exit_func(i);
+		xe_call_exit_func(init_funcs + i);
 }
 
 module_init(xe_init);

From 823301c847bd9c57dd8f7c684046778281744c77 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:30 +0200
Subject: [PATCH 0278/1455] drm/xe: Print module init abort code

We should provide a hint to the user why the module refused to
load. This will also allow us to drop individual error messages
from init steps.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-3-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_module.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_module.c b/drivers/gpu/drm/xe/xe_module.c
index cfed36361613..d08338fc3bc1 100644
--- a/drivers/gpu/drm/xe/xe_module.c
+++ b/drivers/gpu/drm/xe/xe_module.c
@@ -155,6 +155,8 @@ static int __init xe_init(void)
 	for (i = 0; i < ARRAY_SIZE(init_funcs); i++) {
 		err = xe_call_init_func(init_funcs + i);
 		if (err) {
+			pr_info("%s: module_init aborted at %ps %pe\n",
+				DRIVER_NAME, init_funcs[i].init, ERR_PTR(err));
 			while (i--)
 				xe_call_exit_func(init_funcs + i);
 			return err;

From b90613fb02179a01c6042d8a952e2c68e81d3cf7 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:31 +0200
Subject: [PATCH 0279/1455] drm/xe/configfs: Destroy xe_configfs.su_mutex on
 exit/error

While mutex_destroy() is NOP when CONFIG_DEBUG_MUTEXES is not
enabled, we should still call it.

While around, drop a trailing line.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-4-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 36e2b45b305f..56fbf4c1c37f 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -401,6 +401,7 @@ int __init xe_configfs_init(void)
 	if (ret) {
 		pr_err("Error %d while registering %s subsystem\n",
 		       ret, root->cg_item.ci_namebuf);
+		mutex_destroy(&xe_configfs.su_mutex);
 		return ret;
 	}
 
@@ -410,5 +411,5 @@ int __init xe_configfs_init(void)
 void __exit xe_configfs_exit(void)
 {
 	configfs_unregister_subsystem(&xe_configfs);
+	mutex_destroy(&xe_configfs.su_mutex);
 }
-

From c4b1dde0630e40fa773b1613495d4df4955f44b4 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:32 +0200
Subject: [PATCH 0280/1455] drm/xe/configfs: Drop redundant init() error
 message

There is no need to print separate error message since we will
also print one in xe_init(). Also drop temporary variable, which
was likely just taken from the example code.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-5-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 56fbf4c1c37f..6aa0531bcf76 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -392,15 +392,12 @@ u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 
 int __init xe_configfs_init(void)
 {
-	struct config_group *root = &xe_configfs.su_group;
 	int ret;
 
-	config_group_init(root);
+	config_group_init(&xe_configfs.su_group);
 	mutex_init(&xe_configfs.su_mutex);
 	ret = configfs_register_subsystem(&xe_configfs);
 	if (ret) {
-		pr_err("Error %d while registering %s subsystem\n",
-		       ret, root->cg_item.ci_namebuf);
 		mutex_destroy(&xe_configfs.su_mutex);
 		return ret;
 	}

From 88df7939d728a90ddb722c8113a4529948b3d2ef Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:33 +0200
Subject: [PATCH 0281/1455] drm/xe/configfs: Rename struct xe_config_device

Rename it to struct xe_config_group_device to better match its
purpose. It will also help us to reintroduce in the upcoming patch
the same struct name but this time to hold only configuration data.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-6-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 6aa0531bcf76..d69b5f248e59 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -85,7 +85,7 @@
  *	rmdir /sys/kernel/config/xe/0000:03:00.0/
  */
 
-struct xe_config_device {
+struct xe_config_group_device {
 	struct config_group group;
 
 	bool survivability_mode;
@@ -113,21 +113,21 @@ static const struct engine_info engine_info[] = {
 	{ .cls = "gsccs", .mask = XE_HW_ENGINE_GSCCS_MASK },
 };
 
-static struct xe_config_device *to_xe_config_device(struct config_item *item)
+static struct xe_config_group_device *to_xe_config_group_device(struct config_item *item)
 {
-	return container_of(to_config_group(item), struct xe_config_device, group);
+	return container_of(to_config_group(item), struct xe_config_group_device, group);
 }
 
 static ssize_t survivability_mode_show(struct config_item *item, char *page)
 {
-	struct xe_config_device *dev = to_xe_config_device(item);
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 
 	return sprintf(page, "%d\n", dev->survivability_mode);
 }
 
 static ssize_t survivability_mode_store(struct config_item *item, const char *page, size_t len)
 {
-	struct xe_config_device *dev = to_xe_config_device(item);
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 	bool survivability_mode;
 	int ret;
 
@@ -144,7 +144,7 @@ static ssize_t survivability_mode_store(struct config_item *item, const char *pa
 
 static ssize_t engines_allowed_show(struct config_item *item, char *page)
 {
-	struct xe_config_device *dev = to_xe_config_device(item);
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 	char *p = page;
 
 	for (size_t i = 0; i < ARRAY_SIZE(engine_info); i++) {
@@ -199,7 +199,7 @@ static bool lookup_engine_mask(const char *pattern, u64 *mask)
 static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 				     size_t len)
 {
-	struct xe_config_device *dev = to_xe_config_device(item);
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 	size_t patternlen, p;
 	u64 mask, val = 0;
 
@@ -237,7 +237,7 @@ static struct configfs_attribute *xe_config_device_attrs[] = {
 
 static void xe_config_device_release(struct config_item *item)
 {
-	struct xe_config_device *dev = to_xe_config_device(item);
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 
 	mutex_destroy(&dev->lock);
 	kfree(dev);
@@ -257,7 +257,7 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 							const char *name)
 {
 	unsigned int domain, bus, slot, function;
-	struct xe_config_device *dev;
+	struct xe_config_group_device *dev;
 	struct pci_dev *pdev;
 	char canonical[16];
 	int ret;
@@ -309,7 +309,7 @@ static struct configfs_subsystem xe_configfs = {
 	},
 };
 
-static struct xe_config_device *configfs_find_group(struct pci_dev *pdev)
+static struct xe_config_group_device *configfs_find_group(struct pci_dev *pdev)
 {
 	struct config_item *item;
 
@@ -320,7 +320,7 @@ static struct xe_config_device *configfs_find_group(struct pci_dev *pdev)
 	if (!item)
 		return NULL;
 
-	return to_xe_config_device(item);
+	return to_xe_config_group_device(item);
 }
 
 /**
@@ -334,7 +334,7 @@ static struct xe_config_device *configfs_find_group(struct pci_dev *pdev)
  */
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 {
-	struct xe_config_device *dev = configfs_find_group(pdev);
+	struct xe_config_group_device *dev = configfs_find_group(pdev);
 	bool mode;
 
 	if (!dev)
@@ -355,7 +355,7 @@ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
  */
 void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
 {
-	struct xe_config_device *dev = configfs_find_group(pdev);
+	struct xe_config_group_device *dev = configfs_find_group(pdev);
 
 	if (!dev)
 		return;
@@ -378,7 +378,7 @@ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
  */
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 {
-	struct xe_config_device *dev = configfs_find_group(pdev);
+	struct xe_config_group_device *dev = configfs_find_group(pdev);
 	u64 engines_allowed;
 
 	if (!dev)

From ae3184d5f980d92377545ba58030e47ce48dd09f Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:34 +0200
Subject: [PATCH 0282/1455] drm/xe/configfs: Rename configfs_find_group()
 helper

This helper name shouldn't suggest that it iss a part of the core
configfs API family. While around switch to use different helper
to release a reference.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-7-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index d69b5f248e59..acdee616d118 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -309,7 +309,7 @@ static struct configfs_subsystem xe_configfs = {
 	},
 };
 
-static struct xe_config_group_device *configfs_find_group(struct pci_dev *pdev)
+static struct xe_config_group_device *find_xe_config_group_device(struct pci_dev *pdev)
 {
 	struct config_item *item;
 
@@ -334,14 +334,14 @@ static struct xe_config_group_device *configfs_find_group(struct pci_dev *pdev)
  */
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 {
-	struct xe_config_group_device *dev = configfs_find_group(pdev);
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
 	bool mode;
 
 	if (!dev)
 		return false;
 
 	mode = dev->survivability_mode;
-	config_item_put(&dev->group.cg_item);
+	config_group_put(&dev->group);
 
 	return mode;
 }
@@ -355,7 +355,7 @@ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
  */
 void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
 {
-	struct xe_config_group_device *dev = configfs_find_group(pdev);
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
 
 	if (!dev)
 		return;
@@ -364,7 +364,7 @@ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
 	dev->survivability_mode = 0;
 	mutex_unlock(&dev->lock);
 
-	config_item_put(&dev->group.cg_item);
+	config_group_put(&dev->group);
 }
 
 /**
@@ -378,14 +378,14 @@ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
  */
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 {
-	struct xe_config_group_device *dev = configfs_find_group(pdev);
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
 	u64 engines_allowed;
 
 	if (!dev)
 		return U64_MAX;
 
 	engines_allowed = dev->engines_allowed;
-	config_item_put(&dev->group.cg_item);
+	config_group_put(&dev->group);
 
 	return engines_allowed;
 }

From 3c643f6216215c7e4e30f7a647024a20847b27af Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:35 +0200
Subject: [PATCH 0283/1455] drm/xe/configfs: Reintroduce struct
 xe_config_device

This time it will hold just pure configuration parameters, without
any configfs related stuff. This will help us define defaults data
without wasting space for unneeded data.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-8-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index acdee616d118..7ad9fc65e21d 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -88,8 +88,10 @@
 struct xe_config_group_device {
 	struct config_group group;
 
-	bool survivability_mode;
-	u64 engines_allowed;
+	struct xe_config_device {
+		bool survivability_mode;
+		u64 engines_allowed;
+	} config;
 
 	/* protects attributes */
 	struct mutex lock;
@@ -118,9 +120,14 @@ static struct xe_config_group_device *to_xe_config_group_device(struct config_it
 	return container_of(to_config_group(item), struct xe_config_group_device, group);
 }
 
+static struct xe_config_device *to_xe_config_device(struct config_item *item)
+{
+	return &to_xe_config_group_device(item)->config;
+}
+
 static ssize_t survivability_mode_show(struct config_item *item, char *page)
 {
-	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+	struct xe_config_device *dev = to_xe_config_device(item);
 
 	return sprintf(page, "%d\n", dev->survivability_mode);
 }
@@ -136,7 +143,7 @@ static ssize_t survivability_mode_store(struct config_item *item, const char *pa
 		return ret;
 
 	mutex_lock(&dev->lock);
-	dev->survivability_mode = survivability_mode;
+	dev->config.survivability_mode = survivability_mode;
 	mutex_unlock(&dev->lock);
 
 	return len;
@@ -144,7 +151,7 @@ static ssize_t survivability_mode_store(struct config_item *item, const char *pa
 
 static ssize_t engines_allowed_show(struct config_item *item, char *page)
 {
-	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+	struct xe_config_device *dev = to_xe_config_device(item);
 	char *p = page;
 
 	for (size_t i = 0; i < ARRAY_SIZE(engine_info); i++) {
@@ -220,7 +227,7 @@ static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 	}
 
 	mutex_lock(&dev->lock);
-	dev->engines_allowed = val;
+	dev->config.engines_allowed = val;
 	mutex_unlock(&dev->lock);
 
 	return len;
@@ -282,7 +289,7 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 		return ERR_PTR(-ENOMEM);
 
 	/* Default values */
-	dev->engines_allowed = U64_MAX;
+	dev->config.engines_allowed = U64_MAX;
 
 	config_group_init_type_name(&dev->group, name, &xe_config_device_type);
 
@@ -340,7 +347,7 @@ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 	if (!dev)
 		return false;
 
-	mode = dev->survivability_mode;
+	mode = dev->config.survivability_mode;
 	config_group_put(&dev->group);
 
 	return mode;
@@ -361,7 +368,7 @@ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
 		return;
 
 	mutex_lock(&dev->lock);
-	dev->survivability_mode = 0;
+	dev->config.survivability_mode = 0;
 	mutex_unlock(&dev->lock);
 
 	config_group_put(&dev->group);
@@ -384,7 +391,7 @@ u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 	if (!dev)
 		return U64_MAX;
 
-	engines_allowed = dev->engines_allowed;
+	engines_allowed = dev->config.engines_allowed;
 	config_group_put(&dev->group);
 
 	return engines_allowed;

From 737a72d7e0231123ae3f1578845d9a23b6d02a60 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:36 +0200
Subject: [PATCH 0284/1455] drm/xe/configfs: Keep default device config
 settings together

For easier maintenance add a placeholder where we can keep all
default device configuration settings in one place.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-9-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 7ad9fc65e21d..3fc01bef0114 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -97,6 +97,16 @@ struct xe_config_group_device {
 	struct mutex lock;
 };
 
+static const struct xe_config_device device_defaults = {
+	.survivability_mode = false,
+	.engines_allowed = U64_MAX,
+};
+
+static void set_device_defaults(struct xe_config_device *config)
+{
+	*config = device_defaults;
+}
+
 struct engine_info {
 	const char *cls;
 	u64 mask;
@@ -288,8 +298,7 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
-	/* Default values */
-	dev->config.engines_allowed = U64_MAX;
+	set_device_defaults(&dev->config);
 
 	config_group_init_type_name(&dev->group, name, &xe_config_device_type);
 
@@ -345,7 +354,7 @@ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 	bool mode;
 
 	if (!dev)
-		return false;
+		return device_defaults.survivability_mode;
 
 	mode = dev->config.survivability_mode;
 	config_group_put(&dev->group);
@@ -389,7 +398,7 @@ u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 	u64 engines_allowed;
 
 	if (!dev)
-		return U64_MAX;
+		return device_defaults.engines_allowed;
 
 	engines_allowed = dev->config.engines_allowed;
 	config_group_put(&dev->group);

From b4687422c322e5afc0f5d9425b29080c06dc17b9 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 21:33:38 +0200
Subject: [PATCH 0285/1455] drm/xe/configfs: Only allow configurations for
 supported devices

Since we already lookup for the real PCI device before we allow
to create its directory config, we might also check if the found
device matches our driver PCI ID list. This will prevent creation
of the directory configs for the unsupported devices.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731193339.179829-11-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 3fc01bef0114..5c13749733af 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -270,11 +270,22 @@ static const struct config_item_type xe_config_device_type = {
 	.ct_owner	= THIS_MODULE,
 };
 
+static const struct xe_device_desc *xe_match_desc(struct pci_dev *pdev)
+{
+	struct device_driver *driver = driver_find("xe", &pci_bus_type);
+	struct pci_driver *drv = to_pci_driver(driver);
+	const struct pci_device_id *ids = drv ? drv->id_table : NULL;
+	const struct pci_device_id *found = pci_match_id(ids, pdev);
+
+	return found ? (const void *)found->driver_data : NULL;
+}
+
 static struct config_group *xe_config_make_device_group(struct config_group *group,
 							const char *name)
 {
 	unsigned int domain, bus, slot, function;
 	struct xe_config_group_device *dev;
+	const struct xe_device_desc *match;
 	struct pci_dev *pdev;
 	char canonical[16];
 	int ret;
@@ -292,8 +303,16 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 	pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function));
 	if (!pdev)
 		return ERR_PTR(-ENODEV);
+
+	match = xe_match_desc(pdev);
+	if (!match)
+		pci_info(pdev, "xe driver does not support configuration of this device\n");
+
 	pci_dev_put(pdev);
 
+	if (!match)
+		return ERR_PTR(-ENOENT);
+
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
 	if (!dev)
 		return ERR_PTR(-ENOMEM);

From ca0ed3b10f5dba790808107896c568cc037a74f4 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 31 Jul 2025 23:21:45 +0200
Subject: [PATCH 0286/1455] drm/xe/configfs: Allow adding configurations for
 future VFs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since we are expecting that all configuration directory names
will match some of the existing devices, we can't provide any
configuration for the VFs until they are actually enabled.

But we can relax that restriction by just checking if there
is a PF device that could create given VF. This is easy since
all our PF devices are always present at function 0 and we can
query PF device for number of VFs it could support.

Then for some system with PF device at 0000:00:02.0 we can add
configs for all VFs:

  /sys/kernel/config/xe/
  ├── 0000:00:02.0
  │   └── ...
  ├── 0000:00:02.1
  │   └── ...
  ├── 0000:00:02.2
  │   └── ...
  :
  └── 0000:00:02.7
      └── ...

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250731212145.179898-1-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 33 +++++++++++++++++++++++++++++---
 1 file changed, 30 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 5c13749733af..853da2ee837a 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -12,9 +12,9 @@
 #include <linux/string.h>
 
 #include "xe_configfs.h"
-#include "xe_module.h"
-
 #include "xe_hw_engine_types.h"
+#include "xe_module.h"
+#include "xe_pci_types.h"
 
 /**
  * DOC: Xe Configfs
@@ -280,6 +280,15 @@ static const struct xe_device_desc *xe_match_desc(struct pci_dev *pdev)
 	return found ? (const void *)found->driver_data : NULL;
 }
 
+static struct pci_dev *get_physfn_instead(struct pci_dev *virtfn)
+{
+	struct pci_dev *physfn = pci_physfn(virtfn);
+
+	pci_dev_get(physfn);
+	pci_dev_put(virtfn);
+	return physfn;
+}
+
 static struct config_group *xe_config_make_device_group(struct config_group *group,
 							const char *name)
 {
@@ -288,6 +297,7 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 	const struct xe_device_desc *match;
 	struct pci_dev *pdev;
 	char canonical[16];
+	int vfnumber = 0;
 	int ret;
 
 	ret = sscanf(name, "%x:%x:%x.%x", &domain, &bus, &slot, &function);
@@ -301,12 +311,29 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 		return ERR_PTR(-EINVAL);
 
 	pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function));
+	if (!pdev && function)
+		pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, 0));
+	if (!pdev && slot)
+		pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(0, 0));
 	if (!pdev)
 		return ERR_PTR(-ENODEV);
 
+	if (PCI_DEVFN(slot, function) != pdev->devfn) {
+		pdev = get_physfn_instead(pdev);
+		vfnumber = PCI_DEVFN(slot, function) - pdev->devfn;
+		if (!dev_is_pf(&pdev->dev) || vfnumber > pci_sriov_get_totalvfs(pdev)) {
+			pci_dev_put(pdev);
+			return ERR_PTR(-ENODEV);
+		}
+	}
+
 	match = xe_match_desc(pdev);
-	if (!match)
+	if (match && vfnumber && !match->has_sriov) {
+		pci_info(pdev, "xe driver does not support VFs on this device\n");
+		match = NULL;
+	} else if (!match) {
 		pci_info(pdev, "xe driver does not support configuration of this device\n");
+	}
 
 	pci_dev_put(pdev);
 

From cdaecac4fbcb2680e0c4e89f772f20b5bf132ea6 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Wed, 6 Aug 2025 08:38:56 +0530
Subject: [PATCH 0287/1455] drm/i915/scaler: Fix WA_14011503117

This introduces and uses a variable id which is just uninitialized.
What really needs to be used is the scaler_id.

Fixes: 73309ed9d598 ("drm/i915/display: WA_14011503117")
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Nemesa Garg <nemesa.garg@intel.com>
Link: https://lore.kernel.org/r/20250806030856.3514127-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index 1374fa94ff5c..cd7ebbeb9508 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -959,11 +959,12 @@ void adl_scaler_ecc_unmask(const struct intel_crtc_state *crtc_state)
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 	const struct intel_crtc_scaler_state *scaler_state =
 		&crtc_state->scaler_state;
-	int id;
 
 	if (!scaler_state && scaler_state->scaler_id == -1)
 		return;
 
-	intel_de_write_fw(display, SKL_PS_ECC_STAT(crtc->pipe, id), 1);
+	intel_de_write_fw(display,
+			  SKL_PS_ECC_STAT(crtc->pipe, scaler_state->scaler_id),
+			  1);
 	intel_de_write(display, XELPD_DISPLAY_ERR_FATAL_MASK, 0);
 }

From 01048738d6b618fc1dba564b38b3df06b0937eb0 Mon Sep 17 00:00:00 2001
From: Joseph Guo <qijian.guo@nxp.com>
Date: Wed, 6 Aug 2025 10:33:32 +0900
Subject: [PATCH 0288/1455] dt-bindings: display: bridge: Add waveshare DSI2DPI
 unit support

Add dt-binding documentation for waveshare DSI2DPI unit

Signed-off-by: Joseph Guo <qijian.guo@nxp.com>
Suggested-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250806-waveshare-v3-1-fd28e01f064f@nxp.com
---
 .../display/bridge/waveshare,dsi2dpi.yaml     | 103 ++++++++++++++++++
 1 file changed, 103 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/bridge/waveshare,dsi2dpi.yaml

diff --git a/Documentation/devicetree/bindings/display/bridge/waveshare,dsi2dpi.yaml b/Documentation/devicetree/bindings/display/bridge/waveshare,dsi2dpi.yaml
new file mode 100644
index 000000000000..5e8498c8303d
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/bridge/waveshare,dsi2dpi.yaml
@@ -0,0 +1,103 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/bridge/waveshare,dsi2dpi.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Waveshare MIPI-DSI to DPI Converter bridge
+
+maintainers:
+  - Joseph Guo <qijian.guo@nxp.com>
+
+description:
+  Waveshare bridge board is part of Waveshare panel which converts DSI to DPI.
+
+properties:
+  compatible:
+    const: waveshare,dsi2dpi
+
+  reg:
+    maxItems: 1
+    description: base I2C address of the device
+
+  power-supply: true
+
+  ports:
+    $ref: /schemas/graph.yaml#/properties/ports
+
+    properties:
+      port@0:
+        $ref: /schemas/graph.yaml#/$defs/port-base
+        unevaluatedProperties: false
+        description:
+          Video port for MIPI DSI input
+
+        properties:
+          endpoint:
+            $ref: /schemas/media/video-interfaces.yaml#
+            unevaluatedProperties: false
+
+            properties:
+              data-lanes:
+                description: array of physical DSI data lane indexes.
+                items:
+                  - const: 1
+                  - const: 2
+
+            required:
+              - data-lanes
+
+      port@1:
+        $ref: /schemas/graph.yaml#/properties/port
+        description:
+          Video port for MIPI DPI output panel.
+
+    required:
+      - port@0
+      - port@1
+
+required:
+  - compatible
+  - reg
+  - ports
+  - power-supply
+
+additionalProperties: false
+
+examples:
+  - |
+
+    i2c {
+      #address-cells = <1>;
+      #size-cells = <0>;
+
+      bridge@45 {
+        compatible = "waveshare,dsi2dpi";
+        reg = <0x45>;
+        power-supply = <&reg_3p3v>;
+
+        ports {
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          port@0 {
+            reg = <0>;
+
+            waveshare_from_dsim: endpoint {
+              data-lanes = <1 2>;
+              remote-endpoint = <&dsim_to_waveshare>;
+            };
+          };
+
+          port@1 {
+            reg = <1>;
+
+            waveshare_to_panel: endpoint {
+              remote-endpoint = <&panel_to_waveshare>;
+            };
+          };
+        };
+      };
+    };
+
+...

From 80b0eb11f8e0a504078c3b405b54cddaf535ff97 Mon Sep 17 00:00:00 2001
From: Joseph Guo <qijian.guo@nxp.com>
Date: Wed, 6 Aug 2025 10:33:33 +0900
Subject: [PATCH 0289/1455] dt-bindings: display: panel: Add waveshare DPI
 panel support

Add dt-binding documentation for waveshare DPI panel

Signed-off-by: Joseph Guo <qijian.guo@nxp.com>
Suggested-by: Liu Ying <victor.liu@nxp.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Liu Ying <victor.liu@nxp.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250806-waveshare-v3-2-fd28e01f064f@nxp.com
---
 .../devicetree/bindings/display/panel/panel-simple.yaml       | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
index a03a9f1dd9f6..5e8dc9afa1fd 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml
@@ -323,6 +323,10 @@ properties:
       - vivax,tpc9150-panel
         # VXT 800x480 color TFT LCD panel
       - vxt,vl050-8048nt-c01
+        # Waveshare 13.3" FHD (1920x1080) LCD panel
+      - waveshare,13.3inch-panel
+        # Waveshare 7.0" WSVGA (1024x600) LCD panel
+      - waveshare,7.0inch-c-panel
         # Winstar Display Corporation 3.5" QVGA (320x240) TFT LCD panel
       - winstar,wf35ltiacd
         # Yes Optoelectronics YTC700TLAG-05-201C 7" TFT LCD panel

From dbdea37add132a0f5d96006ea01fa73280984b88 Mon Sep 17 00:00:00 2001
From: Joseph Guo <qijian.guo@nxp.com>
Date: Wed, 6 Aug 2025 10:33:34 +0900
Subject: [PATCH 0290/1455] drm: bridge: Add waveshare DSI2DPI unit driver

Waveshare touchscreen consists of a DPI panel and a driver board.
The waveshare driver board consists of ICN6211 and a MCU to
convert DSI to DPI and control the backlight.
This driver treats the MCU and ICN6211 board as a whole unit.
It can support all resolution waveshare DSI2DPI based panel,
the timing table should come from 'panel-dpi' panel in the device tree.

Signed-off-by: Joseph Guo <qijian.guo@nxp.com>
Suggested-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250806-waveshare-v3-3-fd28e01f064f@nxp.com
---
 drivers/gpu/drm/bridge/Kconfig         |  12 ++
 drivers/gpu/drm/bridge/Makefile        |   1 +
 drivers/gpu/drm/bridge/waveshare-dsi.c | 203 +++++++++++++++++++++++++
 3 files changed, 216 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/waveshare-dsi.c

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index c11444c80d88..9cefbd15177a 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -451,6 +451,18 @@ config DRM_TI_TPD12S015
 	  Texas Instruments TPD12S015 HDMI level shifter and ESD protection
 	  driver.
 
+config DRM_WAVESHARE_BRIDGE
+	tristate "Waveshare DSI bridge"
+	depends on OF
+	depends on BACKLIGHT_CLASS_DEVICE
+	select DRM_PANEL_BRIDGE
+	select DRM_KMS_HELPER
+	select DRM_MIPI_DSI
+	select REGMAP_I2C
+	help
+	  Driver for waveshare DSI to DPI bridge board.
+	  Please say Y if you have such hardware
+
 source "drivers/gpu/drm/bridge/analogix/Kconfig"
 
 source "drivers/gpu/drm/bridge/adv7511/Kconfig"
diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile
index 19f5a09d9291..c7dc03182e59 100644
--- a/drivers/gpu/drm/bridge/Makefile
+++ b/drivers/gpu/drm/bridge/Makefile
@@ -41,6 +41,7 @@ obj-$(CONFIG_DRM_TI_SN65DSI86) += ti-sn65dsi86.o
 obj-$(CONFIG_DRM_TI_TDP158) += ti-tdp158.o
 obj-$(CONFIG_DRM_TI_TFP410) += ti-tfp410.o
 obj-$(CONFIG_DRM_TI_TPD12S015) += ti-tpd12s015.o
+obj-$(CONFIG_DRM_WAVESHARE_BRIDGE) += waveshare-dsi.o
 obj-$(CONFIG_DRM_NWL_MIPI_DSI) += nwl-dsi.o
 obj-$(CONFIG_DRM_ITE_IT66121) += ite-it66121.o
 
diff --git a/drivers/gpu/drm/bridge/waveshare-dsi.c b/drivers/gpu/drm/bridge/waveshare-dsi.c
new file mode 100644
index 000000000000..01c70e7d3d3b
--- /dev/null
+++ b/drivers/gpu/drm/bridge/waveshare-dsi.c
@@ -0,0 +1,203 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2025 NXP
+ * Based on panel-raspberrypi-touchscreen by Broadcom
+ */
+
+#include <linux/backlight.h>
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/of_graph.h>
+#include <linux/regmap.h>
+
+#include <drm/drm_bridge.h>
+#include <drm/drm_mipi_dsi.h>
+#include <drm/drm_of.h>
+#include <drm/drm_panel.h>
+
+struct ws_bridge {
+	struct drm_bridge bridge;
+	struct drm_bridge *next_bridge;
+	struct backlight_device *backlight;
+	struct device *dev;
+	struct regmap *reg_map;
+};
+
+static const struct regmap_config ws_regmap_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+	.max_register = 0xff,
+};
+
+static struct ws_bridge *bridge_to_ws_bridge(struct drm_bridge *bridge)
+{
+	return container_of(bridge, struct ws_bridge, bridge);
+}
+
+static int ws_bridge_attach_dsi(struct ws_bridge *ws)
+{
+	const struct mipi_dsi_device_info info = {
+		.type = "ws-bridge",
+		.channel = 0,
+		.node = NULL,
+	};
+	struct device_node *dsi_host_node;
+	struct device *dev = ws->dev;
+	struct mipi_dsi_device *dsi;
+	struct mipi_dsi_host *host;
+	int ret;
+
+	dsi_host_node = of_graph_get_remote_node(dev->of_node, 0, 0);
+	if (!dsi_host_node) {
+		dev_err(dev, "Failed to get remote port\n");
+		return -ENODEV;
+	}
+	host = of_find_mipi_dsi_host_by_node(dsi_host_node);
+	of_node_put(dsi_host_node);
+	if (!host)
+		return dev_err_probe(dev, -EPROBE_DEFER, "Failed to find dsi_host\n");
+
+	dsi = devm_mipi_dsi_device_register_full(dev, host, &info);
+	if (IS_ERR(dsi))
+		return dev_err_probe(dev, PTR_ERR(dsi), "Failed to create dsi device\n");
+
+	dsi->mode_flags = MIPI_DSI_MODE_VIDEO_HSE | MIPI_DSI_MODE_VIDEO |
+			  MIPI_DSI_CLOCK_NON_CONTINUOUS;
+	dsi->format = MIPI_DSI_FMT_RGB888;
+	dsi->lanes = 2;
+
+	ret = devm_mipi_dsi_attach(dev, dsi);
+	if (ret < 0)
+		return dev_err_probe(dev, ret, "Failed to attach dsi to host\n");
+
+	return 0;
+}
+
+static int ws_bridge_bridge_attach(struct drm_bridge *bridge,
+				   struct drm_encoder *encoder,
+				   enum drm_bridge_attach_flags flags)
+{
+	struct ws_bridge *ws = bridge_to_ws_bridge(bridge);
+	int ret;
+
+	ret = ws_bridge_attach_dsi(ws);
+	if (ret)
+		return ret;
+
+	return drm_bridge_attach(encoder, ws->next_bridge,
+				 &ws->bridge, flags);
+}
+
+static void ws_bridge_bridge_enable(struct drm_bridge *bridge)
+{
+	struct ws_bridge *ws = bridge_to_ws_bridge(bridge);
+
+	regmap_write(ws->reg_map, 0xad, 0x01);
+	backlight_enable(ws->backlight);
+}
+
+static void ws_bridge_bridge_disable(struct drm_bridge *bridge)
+{
+	struct ws_bridge *ws = bridge_to_ws_bridge(bridge);
+
+	backlight_disable(ws->backlight);
+	regmap_write(ws->reg_map, 0xad, 0x00);
+}
+
+static const struct drm_bridge_funcs ws_bridge_bridge_funcs = {
+	.enable = ws_bridge_bridge_enable,
+	.disable = ws_bridge_bridge_disable,
+	.attach = ws_bridge_bridge_attach,
+};
+
+static int ws_bridge_bl_update_status(struct backlight_device *bl)
+{
+	struct ws_bridge *ws = bl_get_data(bl);
+
+	regmap_write(ws->reg_map, 0xab, 0xff - backlight_get_brightness(bl));
+	regmap_write(ws->reg_map, 0xaa, 0x01);
+
+	return 0;
+}
+
+static const struct backlight_ops ws_bridge_bl_ops = {
+	.update_status = ws_bridge_bl_update_status,
+};
+
+static struct backlight_device *ws_bridge_create_backlight(struct ws_bridge *ws)
+{
+	const struct backlight_properties props = {
+		.type = BACKLIGHT_RAW,
+		.brightness = 255,
+		.max_brightness = 255,
+	};
+	struct device *dev = ws->dev;
+
+	return devm_backlight_device_register(dev, dev_name(dev), dev, ws,
+					      &ws_bridge_bl_ops, &props);
+}
+
+static int ws_bridge_probe(struct i2c_client *i2c)
+{
+	struct device *dev = &i2c->dev;
+	struct drm_panel *panel;
+	struct ws_bridge *ws;
+	int ret;
+
+	ws = devm_drm_bridge_alloc(dev, struct ws_bridge, bridge, &ws_bridge_bridge_funcs);
+	if (!ws)
+		return -ENOMEM;
+
+	ws->dev = dev;
+
+	ws->reg_map = devm_regmap_init_i2c(i2c, &ws_regmap_config);
+	if (IS_ERR(ws->reg_map))
+		return dev_err_probe(dev, PTR_ERR(ws->reg_map), "Failed to allocate regmap\n");
+
+	ret = drm_of_find_panel_or_bridge(dev->of_node, 1, -1, &panel, NULL);
+	if (ret)
+		return dev_err_probe(dev, ret, "Failed to find remote panel\n");
+
+	ws->next_bridge = devm_drm_panel_bridge_add(dev, panel);
+	if (IS_ERR(ws->next_bridge))
+		return PTR_ERR(ws->next_bridge);
+
+	ws->backlight = ws_bridge_create_backlight(ws);
+	if (IS_ERR(ws->backlight)) {
+		ret = PTR_ERR(ws->backlight);
+		dev_err(dev, "Failed to create backlight: %d\n", ret);
+		return ret;
+	}
+
+	regmap_write(ws->reg_map, 0xc0, 0x01);
+	regmap_write(ws->reg_map, 0xc2, 0x01);
+	regmap_write(ws->reg_map, 0xac, 0x01);
+
+	ws->bridge.type = DRM_MODE_CONNECTOR_DPI;
+	ws->bridge.of_node = dev->of_node;
+	devm_drm_bridge_add(dev, &ws->bridge);
+
+	return 0;
+}
+
+static const struct of_device_id ws_bridge_of_ids[] = {
+	{.compatible = "waveshare,dsi2dpi",},
+	{ }
+};
+
+MODULE_DEVICE_TABLE(of, ws_bridge_of_ids);
+
+static struct i2c_driver ws_bridge_driver = {
+	.driver = {
+		.name = "ws_dsi2dpi",
+		.of_match_table = ws_bridge_of_ids,
+	},
+	.probe = ws_bridge_probe,
+};
+module_i2c_driver(ws_bridge_driver);
+
+MODULE_AUTHOR("Joseph Guo <qijian.guo@nxp.com>");
+MODULE_DESCRIPTION("Waveshare DSI2DPI bridge driver");
+MODULE_LICENSE("GPL");

From 6098dcfa5d1d997083978ed53c31116220bc14bf Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 12:19:20 +0300
Subject: [PATCH 0291/1455] drm/i915/display: hide global state iterators,
 remove unused

for_each_{new,old,oldnew}_global_obj_in_state() are only used within
intel_global_state.c, hide them there. intel_for_each_global_obj() is
unused, remove it.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/a23212d9298423d8971d6ad62f961386f7f927cc.1753953530.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../gpu/drm/i915/display/intel_global_state.c | 25 +++++++++++++++++
 .../gpu/drm/i915/display/intel_global_state.h | 28 -------------------
 2 files changed, 25 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c
index 000a898c9480..3df55ab353f2 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.c
+++ b/drivers/gpu/drm/i915/display/intel_global_state.c
@@ -13,6 +13,31 @@
 #include "intel_display_types.h"
 #include "intel_global_state.h"
 
+#define for_each_new_global_obj_in_state(__state, obj, new_obj_state, __i) \
+	for ((__i) = 0; \
+	     (__i) < (__state)->num_global_objs && \
+		     ((obj) = (__state)->global_objs[__i].ptr, \
+		      (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \
+	     (__i)++) \
+		for_each_if(obj)
+
+#define for_each_old_global_obj_in_state(__state, obj, old_obj_state, __i) \
+	for ((__i) = 0; \
+	     (__i) < (__state)->num_global_objs && \
+		     ((obj) = (__state)->global_objs[__i].ptr, \
+		      (old_obj_state) = (__state)->global_objs[__i].old_state, 1); \
+	     (__i)++) \
+		for_each_if(obj)
+
+#define for_each_oldnew_global_obj_in_state(__state, obj, old_obj_state, new_obj_state, __i) \
+	for ((__i) = 0; \
+	     (__i) < (__state)->num_global_objs && \
+		     ((obj) = (__state)->global_objs[__i].ptr, \
+		      (old_obj_state) = (__state)->global_objs[__i].old_state, \
+		      (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \
+	     (__i)++) \
+		for_each_if(obj)
+
 struct intel_global_commit {
 	struct kref ref;
 	struct completion done;
diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h
index d42fb2547ee9..e24fea1c9dbc 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.h
+++ b/drivers/gpu/drm/i915/display/intel_global_state.h
@@ -26,34 +26,6 @@ struct intel_global_obj {
 	const struct intel_global_state_funcs *funcs;
 };
 
-#define intel_for_each_global_obj(obj, dev_priv) \
-	list_for_each_entry(obj, &(dev_priv)->display.global.obj_list, head)
-
-#define for_each_new_global_obj_in_state(__state, obj, new_obj_state, __i) \
-	for ((__i) = 0; \
-	     (__i) < (__state)->num_global_objs && \
-		     ((obj) = (__state)->global_objs[__i].ptr, \
-		      (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \
-	     (__i)++) \
-		for_each_if(obj)
-
-#define for_each_old_global_obj_in_state(__state, obj, old_obj_state, __i) \
-	for ((__i) = 0; \
-	     (__i) < (__state)->num_global_objs && \
-		     ((obj) = (__state)->global_objs[__i].ptr, \
-		      (old_obj_state) = (__state)->global_objs[__i].old_state, 1); \
-	     (__i)++) \
-		for_each_if(obj)
-
-#define for_each_oldnew_global_obj_in_state(__state, obj, old_obj_state, new_obj_state, __i) \
-	for ((__i) = 0; \
-	     (__i) < (__state)->num_global_objs && \
-		     ((obj) = (__state)->global_objs[__i].ptr, \
-		      (old_obj_state) = (__state)->global_objs[__i].old_state, \
-		      (new_obj_state) = (__state)->global_objs[__i].new_state, 1); \
-	     (__i)++) \
-		for_each_if(obj)
-
 struct intel_global_commit;
 
 struct intel_global_state {

From 1a449b5882e93beca3ae580d20cb52b9e7c8a765 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 12:19:21 +0300
Subject: [PATCH 0292/1455] drm/i915/display: make struct
 __intel_global_objs_state opaque

With struct __intel_global_objs_state only being accessed in
intel_global_state.c, we can make it opaque. The double underscore to
indicate internal becomes redundant, drop it.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/06cc4d1c506e3a5b1cc50e01c4bd1135bbf0f7bd.1753953530.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_types.h | 4 ++--
 drivers/gpu/drm/i915/display/intel_global_state.c  | 7 ++++++-
 drivers/gpu/drm/i915/display/intel_global_state.h  | 5 -----
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 4d9df803ad47..35596f3921e8 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -54,10 +54,10 @@
 
 struct cec_notifier;
 struct drm_printer;
-struct __intel_global_objs_state;
 struct intel_connector;
 struct intel_ddi_buf_trans;
 struct intel_fbc;
+struct intel_global_objs_state;
 struct intel_hdcp_shim;
 struct intel_tc_port;
 
@@ -593,7 +593,7 @@ struct intel_atomic_state {
 
 	struct ref_tracker *wakeref;
 
-	struct __intel_global_objs_state *global_objs;
+	struct intel_global_objs_state *global_objs;
 	int num_global_objs;
 
 	/* Internal commit, as opposed to userspace/client initiated one */
diff --git a/drivers/gpu/drm/i915/display/intel_global_state.c b/drivers/gpu/drm/i915/display/intel_global_state.c
index 3df55ab353f2..30eff6009e87 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.c
+++ b/drivers/gpu/drm/i915/display/intel_global_state.c
@@ -38,6 +38,11 @@
 	     (__i)++) \
 		for_each_if(obj)
 
+struct intel_global_objs_state {
+	struct intel_global_obj *ptr;
+	struct intel_global_state *state, *old_state, *new_state;
+};
+
 struct intel_global_commit {
 	struct kref ref;
 	struct completion done;
@@ -173,7 +178,7 @@ intel_atomic_get_global_obj_state(struct intel_atomic_state *state,
 	struct intel_display *display = to_intel_display(state);
 	int index, num_objs, i;
 	size_t size;
-	struct __intel_global_objs_state *arr;
+	struct intel_global_objs_state *arr;
 	struct intel_global_state *obj_state;
 
 	for (i = 0; i < state->num_global_objs; i++)
diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h
index e24fea1c9dbc..4f304e2c47e0 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.h
+++ b/drivers/gpu/drm/i915/display/intel_global_state.h
@@ -36,11 +36,6 @@ struct intel_global_state {
 	bool changed, serialized;
 };
 
-struct __intel_global_objs_state {
-	struct intel_global_obj *ptr;
-	struct intel_global_state *state, *old_state, *new_state;
-};
-
 void intel_atomic_global_obj_init(struct intel_display *display,
 				  struct intel_global_obj *obj,
 				  struct intel_global_state *state,

From b5de222cb0394e21ced9e7e7cc48415636ae782f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 31 Jul 2025 12:19:22 +0300
Subject: [PATCH 0293/1455] drm/i915/display: keep forward declarations
 together

Adhere to prevalent style.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/2c74fa7f2b7d5ecf8247aa5bff05d104ad60cf9e.1753953530.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_global_state.h | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_global_state.h b/drivers/gpu/drm/i915/display/intel_global_state.h
index 4f304e2c47e0..e1efa530cc86 100644
--- a/drivers/gpu/drm/i915/display/intel_global_state.h
+++ b/drivers/gpu/drm/i915/display/intel_global_state.h
@@ -11,6 +11,7 @@
 
 struct intel_atomic_state;
 struct intel_display;
+struct intel_global_commit;
 struct intel_global_obj;
 struct intel_global_state;
 
@@ -26,8 +27,6 @@ struct intel_global_obj {
 	const struct intel_global_state_funcs *funcs;
 };
 
-struct intel_global_commit;
-
 struct intel_global_state {
 	struct intel_global_obj *obj;
 	struct intel_atomic_state *state;

From 81aa3c7c62049b42959bd4054c3a3ed34b2d5bf4 Mon Sep 17 00:00:00 2001
From: Francois Dugast <francois.dugast@intel.com>
Date: Tue, 5 Aug 2025 15:59:02 +0200
Subject: [PATCH 0294/1455] drm/pagemap: Rename drm_pagemap_device_addr to
 drm_pagemap_addr

Rename this struct to the more generic name drm_pagemap_addr so it can be
used in a broader context, such as DMA mappings of CPU memory.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/20250805140028.599361-2-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 drivers/gpu/drm/drm_gpusvm.c       |  4 ++--
 drivers/gpu/drm/xe/xe_res_cursor.h | 10 +++++-----
 drivers/gpu/drm/xe/xe_svm.c        |  4 ++--
 drivers/gpu/drm/xe/xe_svm.h        |  4 ++--
 include/drm/drm_gpusvm.h           |  4 ++--
 include/drm/drm_pagemap.h          | 32 +++++++++++++++---------------
 6 files changed, 29 insertions(+), 29 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index 5bb4c77db2c3..661306da6b2d 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -1048,7 +1048,7 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 		};
 
 		for (i = 0, j = 0; i < npages; j++) {
-			struct drm_pagemap_device_addr *addr = &range->dma_addr[j];
+			struct drm_pagemap_addr *addr = &range->dma_addr[j];
 
 			if (addr->proto == DRM_INTERCONNECT_SYSTEM)
 				dma_unmap_page(dev,
@@ -1400,7 +1400,7 @@ map_pages:
 				goto err_unmap;
 			}
 
-			range->dma_addr[j] = drm_pagemap_device_addr_encode
+			range->dma_addr[j] = drm_pagemap_addr_encode
 				(addr, DRM_INTERCONNECT_SYSTEM, order,
 				 DMA_BIDIRECTIONAL);
 		}
diff --git a/drivers/gpu/drm/xe/xe_res_cursor.h b/drivers/gpu/drm/xe/xe_res_cursor.h
index d1a403cfb628..4e00008b7081 100644
--- a/drivers/gpu/drm/xe/xe_res_cursor.h
+++ b/drivers/gpu/drm/xe/xe_res_cursor.h
@@ -55,8 +55,8 @@ struct xe_res_cursor {
 	u32 mem_type;
 	/** @sgl: Scatterlist for cursor */
 	struct scatterlist *sgl;
-	/** @dma_addr: Current element in a struct drm_pagemap_device_addr array */
-	const struct drm_pagemap_device_addr *dma_addr;
+	/** @dma_addr: Current element in a struct drm_pagemap_addr array */
+	const struct drm_pagemap_addr *dma_addr;
 	/** @mm: Buddy allocator for VRAM cursor */
 	struct drm_buddy *mm;
 	/**
@@ -170,7 +170,7 @@ static inline void __xe_res_sg_next(struct xe_res_cursor *cur)
  */
 static inline void __xe_res_dma_next(struct xe_res_cursor *cur)
 {
-	const struct drm_pagemap_device_addr *addr = cur->dma_addr;
+	const struct drm_pagemap_addr *addr = cur->dma_addr;
 	u64 start = cur->start;
 
 	while (start >= cur->dma_seg_size) {
@@ -222,14 +222,14 @@ static inline void xe_res_first_sg(const struct sg_table *sg,
 /**
  * xe_res_first_dma - initialize a xe_res_cursor with dma_addr array
  *
- * @dma_addr: struct drm_pagemap_device_addr array to walk
+ * @dma_addr: struct drm_pagemap_addr array to walk
  * @start: Start of the range
  * @size: Size of the range
  * @cur: cursor object to initialize
  *
  * Start walking over the range of allocations between @start and @size.
  */
-static inline void xe_res_first_dma(const struct drm_pagemap_device_addr *dma_addr,
+static inline void xe_res_first_dma(const struct drm_pagemap_addr *dma_addr,
 				    u64 start, u64 size,
 				    struct xe_res_cursor *cur)
 {
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 10c8a1bcb86e..3e4f4a8760ee 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -1014,7 +1014,7 @@ int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range,
 				       ctx->timeslice_ms);
 }
 
-static struct drm_pagemap_device_addr
+static struct drm_pagemap_addr
 xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,
 			  struct device *dev,
 			  struct page *page,
@@ -1033,7 +1033,7 @@ xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,
 		prot = 0;
 	}
 
-	return drm_pagemap_device_addr_encode(addr, prot, order, dir);
+	return drm_pagemap_addr_encode(addr, prot, order, dir);
 }
 
 static const struct drm_pagemap_ops xe_drm_pagemap_ops = {
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index da9a69ea0bb1..4bdccb56d25f 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -163,7 +163,7 @@ void xe_svm_flush(struct xe_vm *vm);
 #else
 #include <linux/interval_tree.h>
 
-struct drm_pagemap_device_addr;
+struct drm_pagemap_addr;
 struct drm_gpusvm_ctx;
 struct drm_gpusvm_range;
 struct xe_bo;
@@ -178,7 +178,7 @@ struct xe_vram_region;
 struct xe_svm_range {
 	struct {
 		struct interval_tree_node itree;
-		const struct drm_pagemap_device_addr *dma_addr;
+		const struct drm_pagemap_addr *dma_addr;
 	} base;
 	u32 tile_present;
 	u32 tile_invalidated;
diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h
index 4aedc5423aff..8d613e9b2690 100644
--- a/include/drm/drm_gpusvm.h
+++ b/include/drm/drm_gpusvm.h
@@ -17,7 +17,7 @@ struct drm_gpusvm_notifier;
 struct drm_gpusvm_ops;
 struct drm_gpusvm_range;
 struct drm_pagemap;
-struct drm_pagemap_device_addr;
+struct drm_pagemap_addr;
 
 /**
  * struct drm_gpusvm_ops - Operations structure for GPU SVM
@@ -154,7 +154,7 @@ struct drm_gpusvm_range {
 	struct interval_tree_node itree;
 	struct list_head entry;
 	unsigned long notifier_seq;
-	struct drm_pagemap_device_addr *dma_addr;
+	struct drm_pagemap_addr *dma_addr;
 	struct drm_pagemap *dpagemap;
 	struct drm_gpusvm_range_flags flags;
 };
diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h
index e5f20a1235be..69d6ee49a3de 100644
--- a/include/drm/drm_pagemap.h
+++ b/include/drm/drm_pagemap.h
@@ -23,7 +23,7 @@ enum drm_interconnect_protocol {
 };
 
 /**
- * struct drm_pagemap_device_addr - Device address representation.
+ * struct drm_pagemap_addr - Address representation.
  * @addr: The dma address or driver-defined address for driver private interconnects.
  * @proto: The interconnect protocol.
  * @order: The page order of the device mapping. (Size is PAGE_SIZE << order).
@@ -32,7 +32,7 @@ enum drm_interconnect_protocol {
  * Note: There is room for improvement here. We should be able to pack into
  * 64 bits.
  */
-struct drm_pagemap_device_addr {
+struct drm_pagemap_addr {
 	dma_addr_t addr;
 	u64 proto : 54;
 	u64 order : 8;
@@ -40,21 +40,21 @@ struct drm_pagemap_device_addr {
 };
 
 /**
- * drm_pagemap_device_addr_encode() - Encode a dma address with metadata
+ * drm_pagemap_addr_encode() - Encode a dma address with metadata
  * @addr: The dma address or driver-defined address for driver private interconnects.
  * @proto: The interconnect protocol.
  * @order: The page order of the dma mapping. (Size is PAGE_SIZE << order).
  * @dir: The DMA direction.
  *
- * Return: A struct drm_pagemap_device_addr encoding the above information.
+ * Return: A struct drm_pagemap_addr encoding the above information.
  */
-static inline struct drm_pagemap_device_addr
-drm_pagemap_device_addr_encode(dma_addr_t addr,
-			       enum drm_interconnect_protocol proto,
-			       unsigned int order,
-			       enum dma_data_direction dir)
+static inline struct drm_pagemap_addr
+drm_pagemap_addr_encode(dma_addr_t addr,
+			enum drm_interconnect_protocol proto,
+			unsigned int order,
+			enum dma_data_direction dir)
 {
-	return (struct drm_pagemap_device_addr) {
+	return (struct drm_pagemap_addr) {
 		.addr = addr,
 		.proto = proto,
 		.order = order,
@@ -75,11 +75,11 @@ struct drm_pagemap_ops {
 	 * @order: The page order of the device mapping. (Size is PAGE_SIZE << order).
 	 * @dir: The transfer direction.
 	 */
-	struct drm_pagemap_device_addr (*device_map)(struct drm_pagemap *dpagemap,
-						     struct device *dev,
-						     struct page *page,
-						     unsigned int order,
-						     enum dma_data_direction dir);
+	struct drm_pagemap_addr (*device_map)(struct drm_pagemap *dpagemap,
+					      struct device *dev,
+					      struct page *page,
+					      unsigned int order,
+					      enum dma_data_direction dir);
 
 	/**
 	 * @device_unmap: Unmap a device address previously obtained using @device_map.
@@ -90,7 +90,7 @@ struct drm_pagemap_ops {
 	 */
 	void (*device_unmap)(struct drm_pagemap *dpagemap,
 			     struct device *dev,
-			     struct drm_pagemap_device_addr addr);
+			     struct drm_pagemap_addr addr);
 
 	/**
 	 * @populate_mm: Populate part of the mm with @dpagemap memory,

From f35a6cdf8a6d69f2fb35ece202a09f13fe7c87b2 Mon Sep 17 00:00:00 2001
From: Francois Dugast <francois.dugast@intel.com>
Date: Tue, 5 Aug 2025 15:59:03 +0200
Subject: [PATCH 0295/1455] drm/pagemap: Use struct drm_pagemap_addr in mapping
 and copy functions

This struct embeds more information than just the DMA address. This will
help later to support folio orders greater than zero. At this point, there
is no functional change as the only struct member used is addr.

In Xe, adapt to the new drm_gpusvm_devmem_ops type signatures using struct
drm_pagemap_addr, as well as the internal xe SVM functions implementing
those operations. The use of this struct is propagated to xe_migrate as it
makes indexed accesses to the next DMA address but they are no longer
contiguous.

v2:
- Rename drm_pagemap_device_addr to drm_pagemap_addr (Matthew Brost)
- Squash with patch for Xe (Matthew Brost)
- Set proto and dir for completeness (Matthew Brost)
- Assess DMA map protocol (Matthew Brost)

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/20250805140028.599361-3-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 drivers/gpu/drm/drm_pagemap.c   | 65 +++++++++++++++++--------------
 drivers/gpu/drm/xe/xe_migrate.c | 69 +++++++++++++++++++--------------
 drivers/gpu/drm/xe/xe_migrate.h |  5 ++-
 drivers/gpu/drm/xe/xe_svm.c     | 37 ++++++++++--------
 include/drm/drm_pagemap.h       |  8 ++--
 5 files changed, 104 insertions(+), 80 deletions(-)

diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
index 1da55322af12..e5c135d793ed 100644
--- a/drivers/gpu/drm/drm_pagemap.c
+++ b/drivers/gpu/drm/drm_pagemap.c
@@ -202,7 +202,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
 /**
  * drm_pagemap_migrate_map_pages() - Map migration pages for GPU SVM migration
  * @dev: The device for which the pages are being mapped
- * @dma_addr: Array to store DMA addresses corresponding to mapped pages
+ * @pagemap_addr: Array to store DMA information corresponding to mapped pages
  * @migrate_pfn: Array of migrate page frame numbers to map
  * @npages: Number of pages to map
  * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL)
@@ -215,7 +215,7 @@ static void drm_pagemap_get_devmem_page(struct page *page,
  * Returns: 0 on success, -EFAULT if an error occurs during mapping.
  */
 static int drm_pagemap_migrate_map_pages(struct device *dev,
-					 dma_addr_t *dma_addr,
+					 struct drm_pagemap_addr *pagemap_addr,
 					 unsigned long *migrate_pfn,
 					 unsigned long npages,
 					 enum dma_data_direction dir)
@@ -224,6 +224,7 @@ static int drm_pagemap_migrate_map_pages(struct device *dev,
 
 	for (i = 0; i < npages; ++i) {
 		struct page *page = migrate_pfn_to_page(migrate_pfn[i]);
+		dma_addr_t dma_addr;
 
 		if (!page)
 			continue;
@@ -231,9 +232,14 @@ static int drm_pagemap_migrate_map_pages(struct device *dev,
 		if (WARN_ON_ONCE(is_zone_device_page(page)))
 			return -EFAULT;
 
-		dma_addr[i] = dma_map_page(dev, page, 0, PAGE_SIZE, dir);
-		if (dma_mapping_error(dev, dma_addr[i]))
+		dma_addr = dma_map_page(dev, page, 0, PAGE_SIZE, dir);
+		if (dma_mapping_error(dev, dma_addr))
 			return -EFAULT;
+
+		pagemap_addr[i] =
+			drm_pagemap_addr_encode(dma_addr,
+						DRM_INTERCONNECT_SYSTEM,
+						0, dir);
 	}
 
 	return 0;
@@ -242,7 +248,7 @@ static int drm_pagemap_migrate_map_pages(struct device *dev,
 /**
  * drm_pagemap_migrate_unmap_pages() - Unmap pages previously mapped for GPU SVM migration
  * @dev: The device for which the pages were mapped
- * @dma_addr: Array of DMA addresses corresponding to mapped pages
+ * @pagemap_addr: Array of DMA information corresponding to mapped pages
  * @npages: Number of pages to unmap
  * @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL)
  *
@@ -251,17 +257,17 @@ static int drm_pagemap_migrate_map_pages(struct device *dev,
  * if it's valid and not already unmapped, and unmaps the corresponding page.
  */
 static void drm_pagemap_migrate_unmap_pages(struct device *dev,
-					    dma_addr_t *dma_addr,
+					    struct drm_pagemap_addr *pagemap_addr,
 					    unsigned long npages,
 					    enum dma_data_direction dir)
 {
 	unsigned long i;
 
 	for (i = 0; i < npages; ++i) {
-		if (!dma_addr[i] || dma_mapping_error(dev, dma_addr[i]))
+		if (!pagemap_addr[i].addr || dma_mapping_error(dev, pagemap_addr[i].addr))
 			continue;
 
-		dma_unmap_page(dev, dma_addr[i], PAGE_SIZE, dir);
+		dma_unmap_page(dev, pagemap_addr[i].addr, PAGE_SIZE, dir);
 	}
 }
 
@@ -314,7 +320,7 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation,
 	struct vm_area_struct *vas;
 	struct drm_pagemap_zdd *zdd = NULL;
 	struct page **pages;
-	dma_addr_t *dma_addr;
+	struct drm_pagemap_addr *pagemap_addr;
 	void *buf;
 	int err;
 
@@ -340,14 +346,14 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation,
 		goto err_out;
 	}
 
-	buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) +
+	buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*pagemap_addr) +
 		       sizeof(*pages), GFP_KERNEL);
 	if (!buf) {
 		err = -ENOMEM;
 		goto err_out;
 	}
-	dma_addr = buf + (2 * sizeof(*migrate.src) * npages);
-	pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages;
+	pagemap_addr = buf + (2 * sizeof(*migrate.src) * npages);
+	pages = buf + (2 * sizeof(*migrate.src) + sizeof(*pagemap_addr)) * npages;
 
 	zdd = drm_pagemap_zdd_alloc(pgmap_owner);
 	if (!zdd) {
@@ -377,8 +383,9 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation,
 	if (err)
 		goto err_finalize;
 
-	err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, dma_addr,
+	err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, pagemap_addr,
 					    migrate.src, npages, DMA_TO_DEVICE);
+
 	if (err)
 		goto err_finalize;
 
@@ -390,7 +397,7 @@ int drm_pagemap_migrate_to_devmem(struct drm_pagemap_devmem *devmem_allocation,
 		drm_pagemap_get_devmem_page(page, zdd);
 	}
 
-	err = ops->copy_to_devmem(pages, dma_addr, npages);
+	err = ops->copy_to_devmem(pages, pagemap_addr, npages);
 	if (err)
 		goto err_finalize;
 
@@ -404,7 +411,7 @@ err_finalize:
 		drm_pagemap_migration_unlock_put_pages(npages, migrate.dst);
 	migrate_vma_pages(&migrate);
 	migrate_vma_finalize(&migrate);
-	drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages,
+	drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, pagemap_addr, npages,
 					DMA_TO_DEVICE);
 err_free:
 	if (zdd)
@@ -509,7 +516,7 @@ int drm_pagemap_evict_to_ram(struct drm_pagemap_devmem *devmem_allocation)
 	unsigned long npages, mpages = 0;
 	struct page **pages;
 	unsigned long *src, *dst;
-	dma_addr_t *dma_addr;
+	struct drm_pagemap_addr *pagemap_addr;
 	void *buf;
 	int i, err = 0;
 	unsigned int retry_count = 2;
@@ -520,7 +527,7 @@ retry:
 	if (!mmget_not_zero(devmem_allocation->mm))
 		return -EFAULT;
 
-	buf = kvcalloc(npages, 2 * sizeof(*src) + sizeof(*dma_addr) +
+	buf = kvcalloc(npages, 2 * sizeof(*src) + sizeof(*pagemap_addr) +
 		       sizeof(*pages), GFP_KERNEL);
 	if (!buf) {
 		err = -ENOMEM;
@@ -528,8 +535,8 @@ retry:
 	}
 	src = buf;
 	dst = buf + (sizeof(*src) * npages);
-	dma_addr = buf + (2 * sizeof(*src) * npages);
-	pages = buf + (2 * sizeof(*src) + sizeof(*dma_addr)) * npages;
+	pagemap_addr = buf + (2 * sizeof(*src) * npages);
+	pages = buf + (2 * sizeof(*src) + sizeof(*pagemap_addr)) * npages;
 
 	err = ops->populate_devmem_pfn(devmem_allocation, npages, src);
 	if (err)
@@ -544,7 +551,7 @@ retry:
 	if (err || !mpages)
 		goto err_finalize;
 
-	err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, dma_addr,
+	err = drm_pagemap_migrate_map_pages(devmem_allocation->dev, pagemap_addr,
 					    dst, npages, DMA_FROM_DEVICE);
 	if (err)
 		goto err_finalize;
@@ -552,7 +559,7 @@ retry:
 	for (i = 0; i < npages; ++i)
 		pages[i] = migrate_pfn_to_page(src[i]);
 
-	err = ops->copy_to_ram(pages, dma_addr, npages);
+	err = ops->copy_to_ram(pages, pagemap_addr, npages);
 	if (err)
 		goto err_finalize;
 
@@ -561,7 +568,7 @@ err_finalize:
 		drm_pagemap_migration_unlock_put_pages(npages, dst);
 	migrate_device_pages(src, dst, npages);
 	migrate_device_finalize(src, dst, npages);
-	drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, dma_addr, npages,
+	drm_pagemap_migrate_unmap_pages(devmem_allocation->dev, pagemap_addr, npages,
 					DMA_FROM_DEVICE);
 err_free:
 	kvfree(buf);
@@ -612,7 +619,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas,
 	struct device *dev = NULL;
 	unsigned long npages, mpages = 0;
 	struct page **pages;
-	dma_addr_t *dma_addr;
+	struct drm_pagemap_addr *pagemap_addr;
 	unsigned long start, end;
 	void *buf;
 	int i, err = 0;
@@ -637,14 +644,14 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas,
 	migrate.end = end;
 	npages = npages_in_range(start, end);
 
-	buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*dma_addr) +
+	buf = kvcalloc(npages, 2 * sizeof(*migrate.src) + sizeof(*pagemap_addr) +
 		       sizeof(*pages), GFP_KERNEL);
 	if (!buf) {
 		err = -ENOMEM;
 		goto err_out;
 	}
-	dma_addr = buf + (2 * sizeof(*migrate.src) * npages);
-	pages = buf + (2 * sizeof(*migrate.src) + sizeof(*dma_addr)) * npages;
+	pagemap_addr = buf + (2 * sizeof(*migrate.src) * npages);
+	pages = buf + (2 * sizeof(*migrate.src) + sizeof(*pagemap_addr)) * npages;
 
 	migrate.vma = vas;
 	migrate.src = buf;
@@ -680,7 +687,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas,
 	if (err)
 		goto err_finalize;
 
-	err = drm_pagemap_migrate_map_pages(dev, dma_addr, migrate.dst, npages,
+	err = drm_pagemap_migrate_map_pages(dev, pagemap_addr, migrate.dst, npages,
 					    DMA_FROM_DEVICE);
 	if (err)
 		goto err_finalize;
@@ -688,7 +695,7 @@ static int __drm_pagemap_migrate_to_ram(struct vm_area_struct *vas,
 	for (i = 0; i < npages; ++i)
 		pages[i] = migrate_pfn_to_page(migrate.src[i]);
 
-	err = ops->copy_to_ram(pages, dma_addr, npages);
+	err = ops->copy_to_ram(pages, pagemap_addr, npages);
 	if (err)
 		goto err_finalize;
 
@@ -698,7 +705,7 @@ err_finalize:
 	migrate_vma_pages(&migrate);
 	migrate_vma_finalize(&migrate);
 	if (dev)
-		drm_pagemap_migrate_unmap_pages(dev, dma_addr, npages,
+		drm_pagemap_migrate_unmap_pages(dev, pagemap_addr, npages,
 						DMA_FROM_DEVICE);
 err_free:
 	kvfree(buf);
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 3a276e2348a2..6a0f04411ae0 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -9,6 +9,7 @@
 #include <linux/sizes.h>
 
 #include <drm/drm_managed.h>
+#include <drm/drm_pagemap.h>
 #include <drm/ttm/ttm_tt.h>
 #include <uapi/drm/xe_drm.h>
 
@@ -1738,7 +1739,8 @@ static u32 pte_update_cmd_size(u64 size)
 
 static void build_pt_update_batch_sram(struct xe_migrate *m,
 				       struct xe_bb *bb, u32 pt_offset,
-				       dma_addr_t *sram_addr, u32 size)
+				       struct drm_pagemap_addr *sram_addr,
+				       u32 size)
 {
 	u16 pat_index = tile_to_xe(m->tile)->pat.idx[XE_CACHE_WB];
 	u32 ptes;
@@ -1756,14 +1758,18 @@ static void build_pt_update_batch_sram(struct xe_migrate *m,
 		ptes -= chunk;
 
 		while (chunk--) {
-			u64 addr = sram_addr[i++] & PAGE_MASK;
+			u64 addr = sram_addr[i].addr & PAGE_MASK;
 
+			xe_tile_assert(m->tile, sram_addr[i].proto ==
+				       DRM_INTERCONNECT_SYSTEM);
 			xe_tile_assert(m->tile, addr);
 			addr = m->q->vm->pt_ops->pte_encode_addr(m->tile->xe,
 								 addr, pat_index,
 								 0, false, 0);
 			bb->cs[bb->len++] = lower_32_bits(addr);
 			bb->cs[bb->len++] = upper_32_bits(addr);
+
+			i++;
 		}
 	}
 }
@@ -1779,7 +1785,8 @@ enum xe_migrate_copy_dir {
 static struct dma_fence *xe_migrate_vram(struct xe_migrate *m,
 					 unsigned long len,
 					 unsigned long sram_offset,
-					 dma_addr_t *sram_addr, u64 vram_addr,
+					 struct drm_pagemap_addr *sram_addr,
+					 u64 vram_addr,
 					 const enum xe_migrate_copy_dir dir)
 {
 	struct xe_gt *gt = m->tile->primary_gt;
@@ -1861,7 +1868,7 @@ err:
  * xe_migrate_to_vram() - Migrate to VRAM
  * @m: The migration context.
  * @npages: Number of pages to migrate.
- * @src_addr: Array of dma addresses (source of migrate)
+ * @src_addr: Array of DMA information (source of migrate)
  * @dst_addr: Device physical address of VRAM (destination of migrate)
  *
  * Copy from an array dma addresses to a VRAM device physical address
@@ -1871,7 +1878,7 @@ err:
  */
 struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
 				     unsigned long npages,
-				     dma_addr_t *src_addr,
+				     struct drm_pagemap_addr *src_addr,
 				     u64 dst_addr)
 {
 	return xe_migrate_vram(m, npages * PAGE_SIZE, 0, src_addr, dst_addr,
@@ -1883,7 +1890,7 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
  * @m: The migration context.
  * @npages: Number of pages to migrate.
  * @src_addr: Device physical address of VRAM (source of migrate)
- * @dst_addr: Array of dma addresses (destination of migrate)
+ * @dst_addr: Array of DMA information (destination of migrate)
  *
  * Copy from a VRAM device physical address to an array dma addresses
  *
@@ -1893,61 +1900,65 @@ struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
 struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m,
 				       unsigned long npages,
 				       u64 src_addr,
-				       dma_addr_t *dst_addr)
+				       struct drm_pagemap_addr *dst_addr)
 {
 	return xe_migrate_vram(m, npages * PAGE_SIZE, 0, dst_addr, src_addr,
 			       XE_MIGRATE_COPY_TO_SRAM);
 }
 
-static void xe_migrate_dma_unmap(struct xe_device *xe, dma_addr_t *dma_addr,
+static void xe_migrate_dma_unmap(struct xe_device *xe,
+				 struct drm_pagemap_addr *pagemap_addr,
 				 int len, int write)
 {
 	unsigned long i, npages = DIV_ROUND_UP(len, PAGE_SIZE);
 
 	for (i = 0; i < npages; ++i) {
-		if (!dma_addr[i])
+		if (!pagemap_addr[i].addr)
 			break;
 
-		dma_unmap_page(xe->drm.dev, dma_addr[i], PAGE_SIZE,
+		dma_unmap_page(xe->drm.dev, pagemap_addr[i].addr, PAGE_SIZE,
 			       write ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
 	}
-	kfree(dma_addr);
+	kfree(pagemap_addr);
 }
 
-static dma_addr_t *xe_migrate_dma_map(struct xe_device *xe,
-				      void *buf, int len, int write)
+static struct drm_pagemap_addr *xe_migrate_dma_map(struct xe_device *xe,
+						   void *buf, int len,
+						   int write)
 {
-	dma_addr_t *dma_addr;
+	struct drm_pagemap_addr *pagemap_addr;
 	unsigned long i, npages = DIV_ROUND_UP(len, PAGE_SIZE);
 
-	dma_addr = kcalloc(npages, sizeof(*dma_addr), GFP_KERNEL);
-	if (!dma_addr)
+	pagemap_addr = kcalloc(npages, sizeof(*pagemap_addr), GFP_KERNEL);
+	if (!pagemap_addr)
 		return ERR_PTR(-ENOMEM);
 
 	for (i = 0; i < npages; ++i) {
 		dma_addr_t addr;
 		struct page *page;
+		enum dma_data_direction dir = write ? DMA_TO_DEVICE :
+						      DMA_FROM_DEVICE;
 
 		if (is_vmalloc_addr(buf))
 			page = vmalloc_to_page(buf);
 		else
 			page = virt_to_page(buf);
 
-		addr = dma_map_page(xe->drm.dev,
-				    page, 0, PAGE_SIZE,
-				    write ? DMA_TO_DEVICE :
-				    DMA_FROM_DEVICE);
+		addr = dma_map_page(xe->drm.dev, page, 0, PAGE_SIZE, dir);
 		if (dma_mapping_error(xe->drm.dev, addr))
 			goto err_fault;
 
-		dma_addr[i] = addr;
+		pagemap_addr[i] =
+			drm_pagemap_addr_encode(addr,
+						DRM_INTERCONNECT_SYSTEM,
+						0, dir);
 		buf += PAGE_SIZE;
 	}
 
-	return dma_addr;
+	return pagemap_addr;
 
 err_fault:
-	xe_migrate_dma_unmap(xe, dma_addr, len, write);
+	xe_migrate_dma_unmap(xe, pagemap_addr, len, write);
 	return ERR_PTR(-EFAULT);
 }
 
@@ -1976,7 +1987,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 	struct xe_device *xe = tile_to_xe(tile);
 	struct xe_res_cursor cursor;
 	struct dma_fence *fence = NULL;
-	dma_addr_t *dma_addr;
+	struct drm_pagemap_addr *pagemap_addr;
 	unsigned long page_offset = (unsigned long)buf & ~PAGE_MASK;
 	int bytes_left = len, current_page = 0;
 	void *orig_buf = buf;
@@ -2031,9 +2042,9 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 		return 0;
 	}
 
-	dma_addr = xe_migrate_dma_map(xe, buf, len + page_offset, write);
-	if (IS_ERR(dma_addr))
-		return PTR_ERR(dma_addr);
+	pagemap_addr = xe_migrate_dma_map(xe, buf, len + page_offset, write);
+	if (IS_ERR(pagemap_addr))
+		return PTR_ERR(pagemap_addr);
 
 	xe_res_first(bo->ttm.resource, offset, xe_bo_size(bo) - offset, &cursor);
 
@@ -2054,7 +2065,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 
 		__fence = xe_migrate_vram(m, current_bytes,
 					  (unsigned long)buf & ~PAGE_MASK,
-					  dma_addr + current_page,
+					  &pagemap_addr[current_page],
 					  vram_addr, write ?
 					  XE_MIGRATE_COPY_TO_VRAM :
 					  XE_MIGRATE_COPY_TO_SRAM);
@@ -2078,7 +2089,7 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 	dma_fence_put(fence);
 
 out_err:
-	xe_migrate_dma_unmap(xe, dma_addr, len + page_offset, write);
+	xe_migrate_dma_unmap(xe, pagemap_addr, len + page_offset, write);
 	return IS_ERR(fence) ? PTR_ERR(fence) : 0;
 }
 
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index e81ea6b27fb5..a3f3fa281e04 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -9,6 +9,7 @@
 #include <linux/types.h>
 
 struct dma_fence;
+struct drm_pagemap_addr;
 struct iosys_map;
 struct ttm_resource;
 
@@ -109,13 +110,13 @@ int xe_migrate_init(struct xe_migrate *m);
 
 struct dma_fence *xe_migrate_to_vram(struct xe_migrate *m,
 				     unsigned long npages,
-				     dma_addr_t *src_addr,
+				     struct drm_pagemap_addr *src_addr,
 				     u64 dst_addr);
 
 struct dma_fence *xe_migrate_from_vram(struct xe_migrate *m,
 				       unsigned long npages,
 				       u64 src_addr,
-				       dma_addr_t *dst_addr);
+				       struct drm_pagemap_addr *dst_addr);
 
 struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 				  struct xe_bo *src_bo,
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 3e4f4a8760ee..1d097e76aabc 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -328,7 +328,8 @@ enum xe_svm_copy_dir {
 	XE_SVM_COPY_TO_SRAM,
 };
 
-static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
+static int xe_svm_copy(struct page **pages,
+		       struct drm_pagemap_addr *pagemap_addr,
 		       unsigned long npages, const enum xe_svm_copy_dir dir)
 {
 	struct xe_vram_region *vr = NULL;
@@ -360,7 +361,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 		last = (i + 1) == npages;
 
 		/* No CPU page and no device pages queue'd to copy */
-		if (!dma_addr[i] && vram_addr == XE_VRAM_ADDR_INVALID)
+		if (!pagemap_addr[i].addr && vram_addr == XE_VRAM_ADDR_INVALID)
 			continue;
 
 		if (!vr && spage) {
@@ -374,7 +375,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 		 * first device page, check if physical contiguous on subsequent
 		 * device pages.
 		 */
-		if (dma_addr[i] && spage) {
+		if (pagemap_addr[i].addr && spage) {
 			__vram_addr = xe_vram_region_page_to_dpa(vr, spage);
 			if (vram_addr == XE_VRAM_ADDR_INVALID) {
 				vram_addr = __vram_addr;
@@ -399,18 +400,20 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 				if (sram) {
 					vm_dbg(&xe->drm,
 					       "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld",
-					       vram_addr, (u64)dma_addr[pos], i - pos + incr);
+					       vram_addr,
+					       (u64)pagemap_addr[pos].addr, i - pos + incr);
 					__fence = xe_migrate_from_vram(vr->migrate,
 								       i - pos + incr,
 								       vram_addr,
-								       dma_addr + pos);
+								       &pagemap_addr[pos]);
 				} else {
 					vm_dbg(&xe->drm,
 					       "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld",
-					       (u64)dma_addr[pos], vram_addr, i - pos + incr);
+					       (u64)pagemap_addr[pos].addr, vram_addr,
+					       i - pos + incr);
 					__fence = xe_migrate_to_vram(vr->migrate,
 								     i - pos + incr,
-								     dma_addr + pos,
+								     &pagemap_addr[pos],
 								     vram_addr);
 				}
 				if (IS_ERR(__fence)) {
@@ -423,7 +426,7 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 			}
 
 			/* Setup physical address of next device page */
-			if (dma_addr[i] && spage) {
+			if (pagemap_addr[i].addr && spage) {
 				vram_addr = __vram_addr;
 				pos = i;
 			} else {
@@ -435,16 +438,16 @@ static int xe_svm_copy(struct page **pages, dma_addr_t *dma_addr,
 				if (sram) {
 					vm_dbg(&xe->drm,
 					       "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%d",
-					       vram_addr, (u64)dma_addr[pos], 1);
+					       vram_addr, (u64)pagemap_addr[pos].addr, 1);
 					__fence = xe_migrate_from_vram(vr->migrate, 1,
 								       vram_addr,
-								       dma_addr + pos);
+								       &pagemap_addr[pos]);
 				} else {
 					vm_dbg(&xe->drm,
 					       "COPY TO VRAM - 0x%016llx -> 0x%016llx, NPAGES=%d",
-					       (u64)dma_addr[pos], vram_addr, 1);
+					       (u64)pagemap_addr[pos].addr, vram_addr, 1);
 					__fence = xe_migrate_to_vram(vr->migrate, 1,
-								     dma_addr + pos,
+								     &pagemap_addr[pos],
 								     vram_addr);
 				}
 				if (IS_ERR(__fence)) {
@@ -470,16 +473,18 @@ err_out:
 #undef XE_VRAM_ADDR_INVALID
 }
 
-static int xe_svm_copy_to_devmem(struct page **pages, dma_addr_t *dma_addr,
+static int xe_svm_copy_to_devmem(struct page **pages,
+				 struct drm_pagemap_addr *pagemap_addr,
 				 unsigned long npages)
 {
-	return xe_svm_copy(pages, dma_addr, npages, XE_SVM_COPY_TO_VRAM);
+	return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_VRAM);
 }
 
-static int xe_svm_copy_to_ram(struct page **pages, dma_addr_t *dma_addr,
+static int xe_svm_copy_to_ram(struct page **pages,
+			      struct drm_pagemap_addr *pagemap_addr,
 			      unsigned long npages)
 {
-	return xe_svm_copy(pages, dma_addr, npages, XE_SVM_COPY_TO_SRAM);
+	return xe_svm_copy(pages, pagemap_addr, npages, XE_SVM_COPY_TO_SRAM);
 }
 
 static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation)
diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h
index 69d6ee49a3de..1d5919a99139 100644
--- a/include/drm/drm_pagemap.h
+++ b/include/drm/drm_pagemap.h
@@ -170,7 +170,7 @@ struct drm_pagemap_devmem_ops {
 	/**
 	 * @copy_to_devmem: Copy to device memory (required for migration)
 	 * @pages: Pointer to array of device memory pages (destination)
-	 * @dma_addr: Pointer to array of DMA addresses (source)
+	 * @pagemap_addr: Pointer to array of DMA information (source)
 	 * @npages: Number of pages to copy
 	 *
 	 * Copy pages to device memory.
@@ -178,13 +178,13 @@ struct drm_pagemap_devmem_ops {
 	 * Return: 0 on success, a negative error code on failure.
 	 */
 	int (*copy_to_devmem)(struct page **pages,
-			      dma_addr_t *dma_addr,
+			      struct drm_pagemap_addr *pagemap_addr,
 			      unsigned long npages);
 
 	/**
 	 * @copy_to_ram: Copy to system RAM (required for migration)
 	 * @pages: Pointer to array of device memory pages (source)
-	 * @dma_addr: Pointer to array of DMA addresses (destination)
+	 * @pagemap_addr: Pointer to array of DMA information (destination)
 	 * @npages: Number of pages to copy
 	 *
 	 * Copy pages to system RAM.
@@ -192,7 +192,7 @@ struct drm_pagemap_devmem_ops {
 	 * Return: 0 on success, a negative error code on failure.
 	 */
 	int (*copy_to_ram)(struct page **pages,
-			   dma_addr_t *dma_addr,
+			   struct drm_pagemap_addr *pagemap_addr,
 			   unsigned long npages);
 };
 

From d755ff6063852cbd43c666726b69333d33d0d379 Mon Sep 17 00:00:00 2001
From: Francois Dugast <francois.dugast@intel.com>
Date: Tue, 5 Aug 2025 15:59:04 +0200
Subject: [PATCH 0296/1455] drm/pagemap: DMA map folios when possible

If the page is part of a folio, DMA map the whole folio at once instead of
mapping individual pages one after the other. For example if 2MB folios
are used instead of 4KB pages, this reduces the number of DMA mappings by
512.

The folio order (and consequently, the size) is persisted in the struct
drm_pagemap_device_addr to be available at the time of unmapping.

v2:
- Initialize order variable (Matthew Brost)
- Set proto and dir for completeness (Matthew Brost)
- Do not populate drm_pagemap_addr, document it (Matthew Brost)
- Add and use macro NR_PAGES(order) (Matthew Brost)

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/20250805140028.599361-4-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 drivers/gpu/drm/drm_pagemap.c | 25 ++++++++++++++++++-------
 include/drm/drm_pagemap.h     | 10 ++++++++--
 2 files changed, 26 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
index e5c135d793ed..0312edb8d4a8 100644
--- a/drivers/gpu/drm/drm_pagemap.c
+++ b/drivers/gpu/drm/drm_pagemap.c
@@ -222,24 +222,32 @@ static int drm_pagemap_migrate_map_pages(struct device *dev,
 {
 	unsigned long i;
 
-	for (i = 0; i < npages; ++i) {
+	for (i = 0; i < npages;) {
 		struct page *page = migrate_pfn_to_page(migrate_pfn[i]);
 		dma_addr_t dma_addr;
+		struct folio *folio;
+		unsigned int order = 0;
 
 		if (!page)
-			continue;
+			goto next;
 
 		if (WARN_ON_ONCE(is_zone_device_page(page)))
 			return -EFAULT;
 
-		dma_addr = dma_map_page(dev, page, 0, PAGE_SIZE, dir);
+		folio = page_folio(page);
+		order = folio_order(folio);
+
+		dma_addr = dma_map_page(dev, page, 0, page_size(page), dir);
 		if (dma_mapping_error(dev, dma_addr))
 			return -EFAULT;
 
 		pagemap_addr[i] =
 			drm_pagemap_addr_encode(dma_addr,
 						DRM_INTERCONNECT_SYSTEM,
-						0, dir);
+						order, dir);
+
+next:
+		i += NR_PAGES(order);
 	}
 
 	return 0;
@@ -263,11 +271,14 @@ static void drm_pagemap_migrate_unmap_pages(struct device *dev,
 {
 	unsigned long i;
 
-	for (i = 0; i < npages; ++i) {
+	for (i = 0; i < npages;) {
 		if (!pagemap_addr[i].addr || dma_mapping_error(dev, pagemap_addr[i].addr))
-			continue;
+			goto next;
 
-		dma_unmap_page(dev, pagemap_addr[i].addr, PAGE_SIZE, dir);
+		dma_unmap_page(dev, pagemap_addr[i].addr, PAGE_SIZE << pagemap_addr[i].order, dir);
+
+next:
+		i += NR_PAGES(pagemap_addr[i].order);
 	}
 }
 
diff --git a/include/drm/drm_pagemap.h b/include/drm/drm_pagemap.h
index 1d5919a99139..f6e7e234c089 100644
--- a/include/drm/drm_pagemap.h
+++ b/include/drm/drm_pagemap.h
@@ -6,6 +6,8 @@
 #include <linux/hmm.h>
 #include <linux/types.h>
 
+#define NR_PAGES(order) (1U << (order))
+
 struct drm_pagemap;
 struct drm_pagemap_zdd;
 struct device;
@@ -173,7 +175,9 @@ struct drm_pagemap_devmem_ops {
 	 * @pagemap_addr: Pointer to array of DMA information (source)
 	 * @npages: Number of pages to copy
 	 *
-	 * Copy pages to device memory.
+	 * Copy pages to device memory. If the order of a @pagemap_addr entry
+	 * is greater than 0, the entry is populated but subsequent entries
+	 * within the range of that order are not populated.
 	 *
 	 * Return: 0 on success, a negative error code on failure.
 	 */
@@ -187,7 +191,9 @@ struct drm_pagemap_devmem_ops {
 	 * @pagemap_addr: Pointer to array of DMA information (destination)
 	 * @npages: Number of pages to copy
 	 *
-	 * Copy pages to system RAM.
+	 * Copy pages to system RAM. If the order of a @pagemap_addr entry
+	 * is greater than 0, the entry is populated but subsequent entries
+	 * within the range of that order are not populated.
 	 *
 	 * Return: 0 on success, a negative error code on failure.
 	 */

From ddeda6136038b4b313bb6f9b44f2d363e1233814 Mon Sep 17 00:00:00 2001
From: Francois Dugast <francois.dugast@intel.com>
Date: Tue, 5 Aug 2025 15:59:05 +0200
Subject: [PATCH 0297/1455] drm/pagemap: Allocate folios when possible
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If the order is greater than zero, allocate a folio when populating the
RAM PFNs instead of allocating individual pages one after the other. For
example if 2MB folios are used instead of 4KB pages, this reduces the
number of calls to the allocation API by 512.

v2:
- Use page order instead of extra argument (Matthew Brost)
- Allocate with folio_alloc() (Matthew Brost)
- Loop for mpages and free_pages based on order (Matthew Brost)

v3:
- Fix loops in drm_pagemap_migrate_populate_ram_pfn() (Matthew Brost)

v4:
- Use folio_trylock(), set local variable to NULL (Matthew Brost)

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Acked-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/20250805140028.599361-5-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 drivers/gpu/drm/drm_pagemap.c | 58 +++++++++++++++++++++++++----------
 1 file changed, 42 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/drm_pagemap.c b/drivers/gpu/drm/drm_pagemap.c
index 0312edb8d4a8..22c44807e3fe 100644
--- a/drivers/gpu/drm/drm_pagemap.c
+++ b/drivers/gpu/drm/drm_pagemap.c
@@ -460,54 +460,80 @@ static int drm_pagemap_migrate_populate_ram_pfn(struct vm_area_struct *vas,
 {
 	unsigned long i;
 
-	for (i = 0; i < npages; ++i, addr += PAGE_SIZE) {
-		struct page *page, *src_page;
+	for (i = 0; i < npages;) {
+		struct page *page = NULL, *src_page;
+		struct folio *folio;
+		unsigned int order = 0;
 
 		if (!(src_mpfn[i] & MIGRATE_PFN_MIGRATE))
-			continue;
+			goto next;
 
 		src_page = migrate_pfn_to_page(src_mpfn[i]);
 		if (!src_page)
-			continue;
+			goto next;
 
 		if (fault_page) {
 			if (src_page->zone_device_data !=
 			    fault_page->zone_device_data)
-				continue;
+				goto next;
 		}
 
-		if (vas)
-			page = alloc_page_vma(GFP_HIGHUSER, vas, addr);
-		else
-			page = alloc_page(GFP_HIGHUSER);
+		order = folio_order(page_folio(src_page));
 
-		if (!page)
+		/* TODO: Support fallback to single pages if THP allocation fails */
+		if (vas)
+			folio = vma_alloc_folio(GFP_HIGHUSER, order, vas, addr);
+		else
+			folio = folio_alloc(GFP_HIGHUSER, order);
+
+		if (!folio)
 			goto free_pages;
 
+		page = folio_page(folio, 0);
 		mpfn[i] = migrate_pfn(page_to_pfn(page));
+
+next:
+		if (page)
+			addr += page_size(page);
+		else
+			addr += PAGE_SIZE;
+
+		i += NR_PAGES(order);
 	}
 
-	for (i = 0; i < npages; ++i) {
+	for (i = 0; i < npages;) {
 		struct page *page = migrate_pfn_to_page(mpfn[i]);
+		unsigned int order = 0;
 
 		if (!page)
-			continue;
+			goto next_lock;
 
-		WARN_ON_ONCE(!trylock_page(page));
-		++*mpages;
+		WARN_ON_ONCE(!folio_trylock(page_folio(page)));
+
+		order = folio_order(page_folio(page));
+		*mpages += NR_PAGES(order);
+
+next_lock:
+		i += NR_PAGES(order);
 	}
 
 	return 0;
 
 free_pages:
-	for (i = 0; i < npages; ++i) {
+	for (i = 0; i < npages;) {
 		struct page *page = migrate_pfn_to_page(mpfn[i]);
+		unsigned int order = 0;
 
 		if (!page)
-			continue;
+			goto next_put;
 
 		put_page(page);
 		mpfn[i] = 0;
+
+		order = folio_order(page_folio(page));
+
+next_put:
+		i += NR_PAGES(order);
 	}
 	return -ENOMEM;
 }

From 321d42032567e6ccbcec971b5547ae8ee1d4b861 Mon Sep 17 00:00:00 2001
From: Francois Dugast <francois.dugast@intel.com>
Date: Tue, 5 Aug 2025 15:59:06 +0200
Subject: [PATCH 0298/1455] drm/xe/migrate: Populate struct drm_pagemap_addr
 array

Workaround to ensure all addresses are populated in the array as
this is expected when creating the copy batch. This is required
because the migrate layer does not support 2MB GPU pages yet. A
proper fix will come in a follow-up.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250805140028.599361-6-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 drivers/gpu/drm/xe/xe_migrate.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 6a0f04411ae0..0f9636a06083 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -1802,6 +1802,7 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m,
 	unsigned int pitch = len >= PAGE_SIZE && !(len & ~PAGE_MASK) ?
 		PAGE_SIZE : 4;
 	int err;
+	unsigned long i, j;
 
 	if (drm_WARN_ON(&xe->drm, (len & XE_CACHELINE_MASK) ||
 			(sram_offset | vram_addr) & XE_CACHELINE_MASK))
@@ -1818,6 +1819,24 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m,
 		return ERR_PTR(err);
 	}
 
+	/*
+	 * If the order of a struct drm_pagemap_addr entry is greater than 0,
+	 * the entry is populated by GPU pagemap but subsequent entries within
+	 * the range of that order are not populated.
+	 * build_pt_update_batch_sram() expects a fully populated array of
+	 * struct drm_pagemap_addr. Ensure this is the case even with higher
+	 * orders.
+	 */
+	for (i = 0; i < npages;) {
+		unsigned int order = sram_addr[i].order;
+
+		for (j = 1; j < NR_PAGES(order) && i + j < npages; j++)
+			if (!sram_addr[i + j].addr)
+				sram_addr[i + j].addr = sram_addr[i].addr + j * PAGE_SIZE;
+
+		i += NR_PAGES(order);
+	}
+
 	build_pt_update_batch_sram(m, bb, pt_slot * XE_PAGE_SIZE,
 				   sram_addr, len + sram_offset);
 

From 979f61633d069d761da606bef5efec42d33509e2 Mon Sep 17 00:00:00 2001
From: Francois Dugast <francois.dugast@intel.com>
Date: Tue, 5 Aug 2025 15:59:07 +0200
Subject: [PATCH 0299/1455] drm/xe/svm: Migrate folios when possible

The DMA mapping can now correspond to a folio (order > 0), so move
the iterator by the number of pages in the folio in order to migrate
all pages at once.

This requires forcing contiguous memory for SVM BOs, which greatly
simplifies the code and enables 2MB device page support, allowing a
major performance improvement. Negative effects like extra eviction
are unlikely as SVM BOs have a maximal size of 2MB.

v2:
- Improve commit message (Matthew Brost)
- Fix increment, chunk, assert match (Matthew Brost)

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250805140028.599361-7-francois.dugast@intel.com
Signed-off-by: Francois Dugast <francois.dugast@intel.com>
---
 drivers/gpu/drm/xe/xe_bo.c  | 2 ++
 drivers/gpu/drm/xe/xe_svm.c | 8 ++++++++
 2 files changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 1a72561f987c..6fea39842e1e 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -200,6 +200,8 @@ static bool force_contiguous(u32 bo_flags)
 	else if (bo_flags & XE_BO_FLAG_PINNED &&
 		 !(bo_flags & XE_BO_FLAG_PINNED_LATE_RESTORE))
 		return true; /* needs vmap */
+	else if (bo_flags & XE_BO_FLAG_CPU_ADDR_MIRROR)
+		return true;
 
 	/*
 	 * For eviction / restore on suspend / resume objects pinned in VRAM
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 1d097e76aabc..e35c6d4def20 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -383,6 +383,14 @@ static int xe_svm_copy(struct page **pages,
 			}
 
 			match = vram_addr + PAGE_SIZE * (i - pos) == __vram_addr;
+			/* Expected with contiguous memory */
+			xe_assert(vr->xe, match);
+
+			if (pagemap_addr[i].order) {
+				i += NR_PAGES(pagemap_addr[i].order) - 1;
+				chunk = (i - pos) == (XE_MIGRATE_CHUNK_SIZE / PAGE_SIZE);
+				last = (i + 1) == npages;
+			}
 		}
 
 		/*

From d2b48f2b30f25997a1ae1ad0cefac68c25f8c330 Mon Sep 17 00:00:00 2001
From: Lizhi Hou <lizhi.hou@amd.com>
Date: Sun, 3 Aug 2025 12:14:50 -0700
Subject: [PATCH 0300/1455] accel/amdxdna: Unify pm and rpm suspend and resume
 callbacks

The suspend and resume callbacks for pm and runtime pm should be same.
During suspending, it needs to stop all hardware contexts first. And
the hardware contexts will be restarted after the device is resumed.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250803191450.1568851-1-lizhi.hou@amd.com
---
 drivers/accel/amdxdna/aie2_ctx.c        | 59 ++++++++++----------
 drivers/accel/amdxdna/aie2_pci.c        | 37 +++++++++++--
 drivers/accel/amdxdna/aie2_pci.h        |  5 +-
 drivers/accel/amdxdna/amdxdna_ctx.c     | 26 ---------
 drivers/accel/amdxdna/amdxdna_ctx.h     |  2 -
 drivers/accel/amdxdna/amdxdna_pci_drv.c | 74 +++----------------------
 drivers/accel/amdxdna/amdxdna_pci_drv.h |  4 +-
 7 files changed, 73 insertions(+), 134 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index 2cff5419bd2f..910ffb7051f4 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -46,6 +46,17 @@ static void aie2_job_put(struct amdxdna_sched_job *job)
 	kref_put(&job->refcnt, aie2_job_release);
 }
 
+static void aie2_hwctx_status_shift_stop(struct amdxdna_hwctx *hwctx)
+{
+	 hwctx->old_status = hwctx->status;
+	 hwctx->status = HWCTX_STAT_STOP;
+}
+
+static void aie2_hwctx_status_restore(struct amdxdna_hwctx *hwctx)
+{
+	hwctx->status = hwctx->old_status;
+}
+
 /* The bad_job is used in aie2_sched_job_timedout, otherwise, set it to NULL */
 static void aie2_hwctx_stop(struct amdxdna_dev *xdna, struct amdxdna_hwctx *hwctx,
 			    struct drm_sched_job *bad_job)
@@ -89,25 +100,6 @@ out:
 	return ret;
 }
 
-void aie2_restart_ctx(struct amdxdna_client *client)
-{
-	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
-
-	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	mutex_lock(&client->hwctx_lock);
-	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
-		if (hwctx->status != HWCTX_STAT_STOP)
-			continue;
-
-		hwctx->status = hwctx->old_status;
-		XDNA_DBG(xdna, "Resetting %s", hwctx->name);
-		aie2_hwctx_restart(xdna, hwctx);
-	}
-	mutex_unlock(&client->hwctx_lock);
-}
-
 static struct dma_fence *aie2_cmd_get_out_fence(struct amdxdna_hwctx *hwctx, u64 seq)
 {
 	struct dma_fence *fence, *out_fence = NULL;
@@ -141,9 +133,11 @@ static void aie2_hwctx_wait_for_idle(struct amdxdna_hwctx *hwctx)
 	dma_fence_put(fence);
 }
 
-void aie2_hwctx_suspend(struct amdxdna_hwctx *hwctx)
+void aie2_hwctx_suspend(struct amdxdna_client *client)
 {
-	struct amdxdna_dev *xdna = hwctx->client->xdna;
+	struct amdxdna_dev *xdna = client->xdna;
+	struct amdxdna_hwctx *hwctx;
+	unsigned long hwctx_id;
 
 	/*
 	 * Command timeout is unlikely. But if it happens, it doesn't
@@ -151,15 +145,19 @@ void aie2_hwctx_suspend(struct amdxdna_hwctx *hwctx)
 	 * and abort all commands.
 	 */
 	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	aie2_hwctx_wait_for_idle(hwctx);
-	aie2_hwctx_stop(xdna, hwctx, NULL);
-	hwctx->old_status = hwctx->status;
-	hwctx->status = HWCTX_STAT_STOP;
+	guard(mutex)(&client->hwctx_lock);
+	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
+		aie2_hwctx_wait_for_idle(hwctx);
+		aie2_hwctx_stop(xdna, hwctx, NULL);
+		aie2_hwctx_status_shift_stop(hwctx);
+	}
 }
 
-void aie2_hwctx_resume(struct amdxdna_hwctx *hwctx)
+void aie2_hwctx_resume(struct amdxdna_client *client)
 {
-	struct amdxdna_dev *xdna = hwctx->client->xdna;
+	struct amdxdna_dev *xdna = client->xdna;
+	struct amdxdna_hwctx *hwctx;
+	unsigned long hwctx_id;
 
 	/*
 	 * The resume path cannot guarantee that mailbox channel can be
@@ -167,8 +165,11 @@ void aie2_hwctx_resume(struct amdxdna_hwctx *hwctx)
 	 * mailbox channel, error will return.
 	 */
 	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	hwctx->status = hwctx->old_status;
-	aie2_hwctx_restart(xdna, hwctx);
+	guard(mutex)(&client->hwctx_lock);
+	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
+		aie2_hwctx_status_restore(hwctx);
+		aie2_hwctx_restart(xdna, hwctx);
+	}
 }
 
 static void
diff --git a/drivers/accel/amdxdna/aie2_pci.c b/drivers/accel/amdxdna/aie2_pci.c
index 3474a8d4e560..6fc3191c3097 100644
--- a/drivers/accel/amdxdna/aie2_pci.c
+++ b/drivers/accel/amdxdna/aie2_pci.c
@@ -440,6 +440,37 @@ disable_dev:
 	return ret;
 }
 
+static int aie2_hw_suspend(struct amdxdna_dev *xdna)
+{
+	struct amdxdna_client *client;
+
+	guard(mutex)(&xdna->dev_lock);
+	list_for_each_entry(client, &xdna->client_list, node)
+		aie2_hwctx_suspend(client);
+
+	aie2_hw_stop(xdna);
+
+	return 0;
+}
+
+static int aie2_hw_resume(struct amdxdna_dev *xdna)
+{
+	struct amdxdna_client *client;
+	int ret;
+
+	guard(mutex)(&xdna->dev_lock);
+	ret = aie2_hw_start(xdna);
+	if (ret) {
+		XDNA_ERR(xdna, "Start hardware failed, %d", ret);
+		return ret;
+	}
+
+	list_for_each_entry(client, &xdna->client_list, node)
+		aie2_hwctx_resume(client);
+
+	return ret;
+}
+
 static int aie2_init(struct amdxdna_dev *xdna)
 {
 	struct pci_dev *pdev = to_pci_dev(xdna->ddev.dev);
@@ -901,8 +932,8 @@ static int aie2_set_state(struct amdxdna_client *client,
 const struct amdxdna_dev_ops aie2_ops = {
 	.init           = aie2_init,
 	.fini           = aie2_fini,
-	.resume         = aie2_hw_start,
-	.suspend        = aie2_hw_stop,
+	.resume         = aie2_hw_resume,
+	.suspend        = aie2_hw_suspend,
 	.get_aie_info   = aie2_get_info,
 	.set_aie_state	= aie2_set_state,
 	.hwctx_init     = aie2_hwctx_init,
@@ -910,6 +941,4 @@ const struct amdxdna_dev_ops aie2_ops = {
 	.hwctx_config   = aie2_hwctx_config,
 	.cmd_submit     = aie2_cmd_submit,
 	.hmm_invalidate = aie2_hmm_invalidate,
-	.hwctx_suspend  = aie2_hwctx_suspend,
-	.hwctx_resume   = aie2_hwctx_resume,
 };
diff --git a/drivers/accel/amdxdna/aie2_pci.h b/drivers/accel/amdxdna/aie2_pci.h
index 385914840eaa..488d8ee568eb 100644
--- a/drivers/accel/amdxdna/aie2_pci.h
+++ b/drivers/accel/amdxdna/aie2_pci.h
@@ -288,10 +288,9 @@ int aie2_sync_bo(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job,
 int aie2_hwctx_init(struct amdxdna_hwctx *hwctx);
 void aie2_hwctx_fini(struct amdxdna_hwctx *hwctx);
 int aie2_hwctx_config(struct amdxdna_hwctx *hwctx, u32 type, u64 value, void *buf, u32 size);
-void aie2_hwctx_suspend(struct amdxdna_hwctx *hwctx);
-void aie2_hwctx_resume(struct amdxdna_hwctx *hwctx);
+void aie2_hwctx_suspend(struct amdxdna_client *client);
+void aie2_hwctx_resume(struct amdxdna_client *client);
 int aie2_cmd_submit(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, u64 *seq);
 void aie2_hmm_invalidate(struct amdxdna_gem_obj *abo, unsigned long cur_seq);
-void aie2_restart_ctx(struct amdxdna_client *client);
 
 #endif /* _AIE2_PCI_H_ */
diff --git a/drivers/accel/amdxdna/amdxdna_ctx.c b/drivers/accel/amdxdna/amdxdna_ctx.c
index be073224bd69..b47a7f8e9017 100644
--- a/drivers/accel/amdxdna/amdxdna_ctx.c
+++ b/drivers/accel/amdxdna/amdxdna_ctx.c
@@ -60,32 +60,6 @@ static struct dma_fence *amdxdna_fence_create(struct amdxdna_hwctx *hwctx)
 	return &fence->base;
 }
 
-void amdxdna_hwctx_suspend(struct amdxdna_client *client)
-{
-	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
-
-	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	mutex_lock(&client->hwctx_lock);
-	amdxdna_for_each_hwctx(client, hwctx_id, hwctx)
-		xdna->dev_info->ops->hwctx_suspend(hwctx);
-	mutex_unlock(&client->hwctx_lock);
-}
-
-void amdxdna_hwctx_resume(struct amdxdna_client *client)
-{
-	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
-
-	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	mutex_lock(&client->hwctx_lock);
-	amdxdna_for_each_hwctx(client, hwctx_id, hwctx)
-		xdna->dev_info->ops->hwctx_resume(hwctx);
-	mutex_unlock(&client->hwctx_lock);
-}
-
 static void amdxdna_hwctx_destroy_rcu(struct amdxdna_hwctx *hwctx,
 				      struct srcu_struct *ss)
 {
diff --git a/drivers/accel/amdxdna/amdxdna_ctx.h b/drivers/accel/amdxdna/amdxdna_ctx.h
index f0a4a8586d85..c652229547a3 100644
--- a/drivers/accel/amdxdna/amdxdna_ctx.h
+++ b/drivers/accel/amdxdna/amdxdna_ctx.h
@@ -147,8 +147,6 @@ static inline u32 amdxdna_hwctx_col_map(struct amdxdna_hwctx *hwctx)
 
 void amdxdna_sched_job_cleanup(struct amdxdna_sched_job *job);
 void amdxdna_hwctx_remove_all(struct amdxdna_client *client);
-void amdxdna_hwctx_suspend(struct amdxdna_client *client);
-void amdxdna_hwctx_resume(struct amdxdna_client *client);
 
 int amdxdna_cmd_submit(struct amdxdna_client *client,
 		       u32 cmd_bo_hdls, u32 *arg_bo_hdls, u32 arg_bo_cnt,
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.c b/drivers/accel/amdxdna/amdxdna_pci_drv.c
index f2bf1d374cc7..fbca94183f96 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.c
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.c
@@ -343,89 +343,29 @@ static void amdxdna_remove(struct pci_dev *pdev)
 	mutex_unlock(&xdna->dev_lock);
 }
 
-static int amdxdna_dev_suspend_nolock(struct amdxdna_dev *xdna)
-{
-	if (xdna->dev_info->ops->suspend)
-		xdna->dev_info->ops->suspend(xdna);
-
-	return 0;
-}
-
-static int amdxdna_dev_resume_nolock(struct amdxdna_dev *xdna)
-{
-	if (xdna->dev_info->ops->resume)
-		return xdna->dev_info->ops->resume(xdna);
-
-	return 0;
-}
-
 static int amdxdna_pmops_suspend(struct device *dev)
 {
 	struct amdxdna_dev *xdna = pci_get_drvdata(to_pci_dev(dev));
-	struct amdxdna_client *client;
 
-	mutex_lock(&xdna->dev_lock);
-	list_for_each_entry(client, &xdna->client_list, node)
-		amdxdna_hwctx_suspend(client);
+	if (!xdna->dev_info->ops->suspend)
+		return -EOPNOTSUPP;
 
-	amdxdna_dev_suspend_nolock(xdna);
-	mutex_unlock(&xdna->dev_lock);
-
-	return 0;
+	return xdna->dev_info->ops->suspend(xdna);
 }
 
 static int amdxdna_pmops_resume(struct device *dev)
 {
 	struct amdxdna_dev *xdna = pci_get_drvdata(to_pci_dev(dev));
-	struct amdxdna_client *client;
-	int ret;
 
-	XDNA_INFO(xdna, "firmware resuming...");
-	mutex_lock(&xdna->dev_lock);
-	ret = amdxdna_dev_resume_nolock(xdna);
-	if (ret) {
-		XDNA_ERR(xdna, "resume NPU firmware failed");
-		mutex_unlock(&xdna->dev_lock);
-		return ret;
-	}
+	if (!xdna->dev_info->ops->resume)
+		return -EOPNOTSUPP;
 
-	XDNA_INFO(xdna, "hardware context resuming...");
-	list_for_each_entry(client, &xdna->client_list, node)
-		amdxdna_hwctx_resume(client);
-	mutex_unlock(&xdna->dev_lock);
-
-	return 0;
-}
-
-static int amdxdna_rpmops_suspend(struct device *dev)
-{
-	struct amdxdna_dev *xdna = pci_get_drvdata(to_pci_dev(dev));
-	int ret;
-
-	mutex_lock(&xdna->dev_lock);
-	ret = amdxdna_dev_suspend_nolock(xdna);
-	mutex_unlock(&xdna->dev_lock);
-
-	XDNA_DBG(xdna, "Runtime suspend done ret: %d", ret);
-	return ret;
-}
-
-static int amdxdna_rpmops_resume(struct device *dev)
-{
-	struct amdxdna_dev *xdna = pci_get_drvdata(to_pci_dev(dev));
-	int ret;
-
-	mutex_lock(&xdna->dev_lock);
-	ret = amdxdna_dev_resume_nolock(xdna);
-	mutex_unlock(&xdna->dev_lock);
-
-	XDNA_DBG(xdna, "Runtime resume done ret: %d", ret);
-	return ret;
+	return xdna->dev_info->ops->resume(xdna);
 }
 
 static const struct dev_pm_ops amdxdna_pm_ops = {
 	SYSTEM_SLEEP_PM_OPS(amdxdna_pmops_suspend, amdxdna_pmops_resume)
-	RUNTIME_PM_OPS(amdxdna_rpmops_suspend, amdxdna_rpmops_resume, NULL)
+	RUNTIME_PM_OPS(amdxdna_pmops_suspend, amdxdna_pmops_resume, NULL)
 };
 
 static struct pci_driver amdxdna_pci_driver = {
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.h b/drivers/accel/amdxdna/amdxdna_pci_drv.h
index ab79600911aa..40bbb3c06320 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.h
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.h
@@ -50,13 +50,11 @@ struct amdxdna_dev_ops {
 	int (*init)(struct amdxdna_dev *xdna);
 	void (*fini)(struct amdxdna_dev *xdna);
 	int (*resume)(struct amdxdna_dev *xdna);
-	void (*suspend)(struct amdxdna_dev *xdna);
+	int (*suspend)(struct amdxdna_dev *xdna);
 	int (*hwctx_init)(struct amdxdna_hwctx *hwctx);
 	void (*hwctx_fini)(struct amdxdna_hwctx *hwctx);
 	int (*hwctx_config)(struct amdxdna_hwctx *hwctx, u32 type, u64 value, void *buf, u32 size);
 	void (*hmm_invalidate)(struct amdxdna_gem_obj *abo, unsigned long cur_seq);
-	void (*hwctx_suspend)(struct amdxdna_hwctx *hwctx);
-	void (*hwctx_resume)(struct amdxdna_hwctx *hwctx);
 	int (*cmd_submit)(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, u64 *seq);
 	int (*get_aie_info)(struct amdxdna_client *client, struct amdxdna_drm_get_info *args);
 	int (*set_aie_state)(struct amdxdna_client *client, struct amdxdna_drm_set_state *args);

From 1b392348de8ffc340545c62079645cb4695f5602 Mon Sep 17 00:00:00 2001
From: Alexandre Demers <alexandre.f.demers@gmail.com>
Date: Sat, 2 Aug 2025 22:27:31 -0400
Subject: [PATCH 0301/1455] Documentation: Remove VCE support from OLAND's
 features

OLAND doesn't support VCE at all, but it does support UVD (3 or 4,
depending of the sources).

Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/dgpu-asic-info-table.csv | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
index a3796daec9ec..bfd44c6e052a 100644
--- a/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/dgpu-asic-info-table.csv
@@ -1,7 +1,7 @@
 Product Name, Code Reference, DCN/DCE version, GC version, VCN version, SDMA version, MP0 version, MP1 version
 AMD Radeon (TM) HD 8500M/ 8600M /M200 /M320 /M330 /M335 Series, HAINAN, --,  6, --, --, --, 6
 AMD Radeon HD 7800 /7900 /FireGL Series, TAHITI, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
-AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
+AMD Radeon R7 (TM|HD) M265 /M370 /8500M /8600 /8700 /8700M, OLAND, DCE 6, 6, -- / UVD 3, --, --, 6
 AMD Radeon (TM) (HD|R7) 7800 /7970 /8800 /8970 /370/ Series, PITCAIRN, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
 AMD Radeon (TM|R7|R9|HD) E8860 /M360 /7700 /7800 /8800 /9000(M) /W4100 Series, VERDE, DCE 6, 6, VCE 1 / UVD 3, --, --, 6
 AMD Radeon HD M280X /M380 /7700 /8950 /W5100, BONAIRE, DCE 8, 7, VCE 2 / UVD 4.2, 1, --, 7

From 9ed3d7bdf2dcdf1a1196630fab89a124526e9cc2 Mon Sep 17 00:00:00 2001
From: Xaver Hugl <xaver.hugl@kde.org>
Date: Fri, 1 Aug 2025 00:49:51 +0200
Subject: [PATCH 0302/1455] amdgpu/amdgpu_discovery: increase timeout limit for
 IFWI init

With a timeout of only 1 second, my rx 5700XT fails to initialize,
so this increases the timeout to 2s.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3697
Signed-off-by: Xaver Hugl <xaver.hugl@kde.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 81b3443c8d7f..047a63496441 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -276,7 +276,7 @@ static int amdgpu_discovery_read_binary_from_mem(struct amdgpu_device *adev,
 	u32 msg;
 
 	if (!amdgpu_sriov_vf(adev)) {
-		/* It can take up to a second for IFWI init to complete on some dGPUs,
+		/* It can take up to two second for IFWI init to complete on some dGPUs,
 		 * but generally it should be in the 60-100ms range.  Normally this starts
 		 * as soon as the device gets power so by the time the OS loads this has long
 		 * completed.  However, when a card is hotplugged via e.g., USB4, we need to
@@ -284,7 +284,7 @@ static int amdgpu_discovery_read_binary_from_mem(struct amdgpu_device *adev,
 		 * continue.
 		 */
 
-		for (i = 0; i < 1000; i++) {
+		for (i = 0; i < 2000; i++) {
 			msg = RREG32(mmMP0_SMN_C2PMSG_33);
 			if (msg & 0x80000000)
 				break;

From 8802ec0eba9887cf2799710aa39d112312213f58 Mon Sep 17 00:00:00 2001
From: Ethan Carter Edwards <ethan@ethancedwards.com>
Date: Fri, 1 Aug 2025 21:41:42 -0400
Subject: [PATCH 0303/1455] drm/amdgpu/gfx10: remove redundant repeated nested
 0 check

The repeated checks on grbm_soft_reset are unnecessary. Remove them.

Signed-off-by: Ethan Carter Edwards <ethan@ethancedwards.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 7bd506f06eb1..264183ab24ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -7668,19 +7668,17 @@ static int gfx_v10_0_soft_reset(struct amdgpu_ip_block *ip_block)
 		/* Disable MEC parsing/prefetching */
 		gfx_v10_0_cp_compute_enable(adev, false);
 
-		if (grbm_soft_reset) {
-			tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-			tmp |= grbm_soft_reset;
-			dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
-			WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
 
-			udelay(50);
+		udelay(50);
 
-			tmp &= ~grbm_soft_reset;
-			WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-		}
+		tmp &= ~grbm_soft_reset;
+		WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
 
 		/* Wait a little for things to settle down */
 		udelay(50);

From 90e1d0324af6ed4750e6c86bca4e19956f61ca2b Mon Sep 17 00:00:00 2001
From: Ethan Carter Edwards <ethan@ethancedwards.com>
Date: Fri, 1 Aug 2025 21:45:41 -0400
Subject: [PATCH 0304/1455] drm/amdgpu/gfx9: remove redundant repeated nested 0
 check

The repeated checks on grbm_soft_reset are unnecessary. Remove them.

Signed-off-by: Ethan Carter Edwards <ethan@ethancedwards.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 6a9cf3587cc6..a6ff9a137a83 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -4175,19 +4175,17 @@ static int gfx_v9_0_soft_reset(struct amdgpu_ip_block *ip_block)
 		/* Disable MEC parsing/prefetching */
 		gfx_v9_0_cp_compute_enable(adev, false);
 
-		if (grbm_soft_reset) {
-			tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-			tmp |= grbm_soft_reset;
-			dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
-			WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
 
-			udelay(50);
+		udelay(50);
 
-			tmp &= ~grbm_soft_reset;
-			WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
-		}
+		tmp &= ~grbm_soft_reset;
+		WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, tmp);
+		tmp = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET);
 
 		/* Wait a little for things to settle down */
 		udelay(50);

From 66f92d1035db75c8da6b362b7eca3fe76704f61f Mon Sep 17 00:00:00 2001
From: Ethan Carter Edwards <ethan@ethancedwards.com>
Date: Fri, 1 Aug 2025 21:38:16 -0400
Subject: [PATCH 0305/1455] drm/amdgpu/gfx9.4.3: remove redundant repeated
 nested 0 check

The repeated checks on grbm_soft_reset are unnecessary. Remove them.

Signed-off-by: Ethan Carter Edwards <ethan@ethancedwards.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 51babf5c78c8..8ba66d4dfe86 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2461,19 +2461,17 @@ static int gfx_v9_4_3_soft_reset(struct amdgpu_ip_block *ip_block)
 		/* Disable MEC parsing/prefetching */
 		gfx_v9_4_3_xcc_cp_compute_enable(adev, false, 0);
 
-		if (grbm_soft_reset) {
-			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
-			tmp |= grbm_soft_reset;
-			dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
-			WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
+		tmp |= grbm_soft_reset;
+		dev_info(adev->dev, "GRBM_SOFT_RESET=0x%08X\n", tmp);
+		WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
 
-			udelay(50);
+		udelay(50);
 
-			tmp &= ~grbm_soft_reset;
-			WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp);
-			tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
-		}
+		tmp &= ~grbm_soft_reset;
+		WREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET, tmp);
+		tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_SOFT_RESET);
 
 		/* Wait a little for things to settle down */
 		udelay(50);

From 32f73741d6ee41fd5db8791c1163931e313d0fdc Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Fri, 18 Jul 2025 18:50:58 +0530
Subject: [PATCH 0306/1455] drm/amdgpu: Wait for bootloader after PSPv11 reset

Some PSPv11 SOCs take a longer time for PSP based mode-1 reset. Instead
of checking for C2PMSG_33 status, add the callback wait_for_bootloader.
Wait for bootloader to be back to steady state is already part of the
generic mode-1 reset flow. Increase the retry count for bootloader wait
and also fix the mask to prevent fake pass.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/psp_v11_0.c | 19 ++++---------------
 1 file changed, 4 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index 6cc05d36e359..64b240b51f1a 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -149,12 +149,12 @@ static int psp_v11_0_wait_for_bootloader(struct psp_context *psp)
 	int ret;
 	int retry_loop;
 
-	for (retry_loop = 0; retry_loop < 10; retry_loop++) {
+	for (retry_loop = 0; retry_loop < 20; retry_loop++) {
 		/* Wait for bootloader to signify that is
 		    ready having bit 31 of C2PMSG_35 set to 1 */
 		ret = psp_wait_for(
 			psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35),
-			0x80000000, 0x80000000, PSP_WAITREG_NOVERBOSE);
+			0x80000000, 0x8000FFFF, PSP_WAITREG_NOVERBOSE);
 
 		if (ret == 0)
 			return 0;
@@ -397,18 +397,6 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp)
 
 	msleep(500);
 
-	offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33);
-
-	ret = psp_wait_for(psp, offset, MBOX_TOS_RESP_FLAG, MBOX_TOS_RESP_MASK,
-			   0);
-
-	if (ret) {
-		DRM_INFO("psp mode 1 reset failed!\n");
-		return -EINVAL;
-	}
-
-	DRM_INFO("psp mode1 reset succeed \n");
-
 	return 0;
 }
 
@@ -665,7 +653,8 @@ static const struct psp_funcs psp_v11_0_funcs = {
 	.ring_get_wptr = psp_v11_0_ring_get_wptr,
 	.ring_set_wptr = psp_v11_0_ring_set_wptr,
 	.load_usbc_pd_fw = psp_v11_0_load_usbc_pd_fw,
-	.read_usbc_pd_fw = psp_v11_0_read_usbc_pd_fw
+	.read_usbc_pd_fw = psp_v11_0_read_usbc_pd_fw,
+	.wait_for_bootloader = psp_v11_0_wait_for_bootloader
 };
 
 void psp_v11_0_set_psp_funcs(struct psp_context *psp)

From 0333052d90683d88531558dcfdbf2525cc37c233 Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Thu, 31 Jul 2025 20:45:00 -0400
Subject: [PATCH 0307/1455] drm/amdkfd: Destroy KFD debugfs after destroy KFD
 wq

Since KFD proc content was moved to kernel debugfs, we can't destroy KFD
debugfs before kfd_process_destroy_wq. Move kfd_process_destroy_wq prior
to kfd_debugfs_fini to fix a kernel NULL pointer problem. It happens
when /sys/kernel/debug/kfd was already destroyed in kfd_debugfs_fini but
kfd_process_destroy_wq calls kfd_debugfs_remove_process. This line
    debugfs_remove_recursive(entry->proc_dentry);
tries to remove /sys/kernel/debug/kfd/proc/<pid> while
/sys/kernel/debug/kfd is already gone. It hangs the kernel by kernel
NULL pointer.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Eric Huang <jinhuieric.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_module.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
index aee2212e52f6..33aa23450b3f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
@@ -78,8 +78,8 @@ err_ioctl:
 static void kfd_exit(void)
 {
 	kfd_cleanup_processes();
-	kfd_debugfs_fini();
 	kfd_process_destroy_wq();
+	kfd_debugfs_fini();
 	kfd_procfs_shutdown();
 	kfd_topology_shutdown();
 	kfd_chardev_exit();

From 91c4fd416463a68dbc2c02ba7b960be9c2486183 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 24 Jul 2025 12:58:10 +0530
Subject: [PATCH 0308/1455] drm/amdgpu: Set dpc status appropriately

Set the dpc status based on hardware state. Also, clear the status before
reinitialization after a successful reset.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Ce Sun <cesun102@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 24 ++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 7c7eecdd68d2..1055f5cda53c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5828,6 +5828,7 @@ int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context)
 		amdgpu_set_init_level(tmp_adev, init_level);
 		if (full_reset) {
 			/* post card */
+			amdgpu_reset_set_dpc_status(tmp_adev, false);
 			amdgpu_ras_clear_err_state(tmp_adev);
 			r = amdgpu_device_asic_init(tmp_adev);
 			if (r) {
@@ -6883,11 +6884,6 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 
 	dev_info(adev->dev, "PCI error: detected callback!!\n");
 
-	if (!amdgpu_dpm_is_link_reset_supported(adev)) {
-		dev_warn(adev->dev, "No support for XGMI hive yet...\n");
-		return PCI_ERS_RESULT_DISCONNECT;
-	}
-
 	adev->pci_channel_state = state;
 
 	switch (state) {
@@ -6897,10 +6893,23 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 	case pci_channel_io_frozen:
 		/* Fatal error, prepare for slot reset */
 		dev_info(adev->dev, "pci_channel_io_frozen: state(%d)!!\n", state);
+		if (hive) {
+			/* Hive devices should be able to support FW based
+			 * link reset on other devices, if not return.
+			 */
+			if (!amdgpu_dpm_is_link_reset_supported(adev)) {
+				dev_warn(adev->dev,
+					 "No support for XGMI hive yet...\n");
+				return PCI_ERS_RESULT_DISCONNECT;
+			}
+			/* Set dpc status only if device is part of hive
+			 * Non-hive devices should be able to recover after
+			 * link reset.
+			 */
+			amdgpu_reset_set_dpc_status(adev, true);
 
-		if (hive)
 			mutex_lock(&hive->hive_lock);
-		amdgpu_reset_set_dpc_status(adev, true);
+		}
 		memset(&reset_context, 0, sizeof(reset_context));
 		INIT_LIST_HEAD(&device_list);
 
@@ -7063,7 +7072,6 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 	amdgpu_device_sched_resume(&device_list, NULL, NULL);
 	amdgpu_device_gpu_resume(adev, &device_list, false);
 	amdgpu_device_recovery_put_reset_lock(adev, &device_list);
-	amdgpu_reset_set_dpc_status(adev, false);
 
 	if (hive) {
 		mutex_unlock(&hive->hive_lock);

From 6ec7120dec937559135a914a3da02a590421bba3 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 24 Jul 2025 11:13:27 +0530
Subject: [PATCH 0309/1455] drm/amd/pm: Add priority messages for SMU v13.0.6

Certain messages will processed with high priority by PMFW even if it
hasn't responded to a previous message. Send the priority message
regardless of the success/fail status of the previous message. Add
support on SMUv13.0.6 and SMUv13.0.12

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h       |  1 +
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c  |  2 +-
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c   |  2 +-
 drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c             | 14 +++++++++-----
 4 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
index d7a9e41820fa..fba980fa5a01 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
@@ -469,6 +469,7 @@ enum smu_feature_mask {
 /* Message category flags */
 #define SMU_MSG_VF_FLAG			(1U << 0)
 #define SMU_MSG_RAS_PRI			(1U << 1)
+#define SMU_MSG_NO_PRECHECK		(1U << 2)
 
 /* Firmware capability flags */
 #define SMU_FW_CAP_RAS_PRI		(1U << 0)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 02a455a31c25..476b7f062a24 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -106,7 +106,7 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(GetDpmFreqByIndex,		     PPSMC_MSG_GetDpmFreqByIndex,		1),
 	MSG_MAP(SetPptLimit,			     PPSMC_MSG_SetPptLimit,			0),
 	MSG_MAP(GetPptLimit,			     PPSMC_MSG_GetPptLimit,			1),
-	MSG_MAP(GfxDeviceDriverReset,		     PPSMC_MSG_GfxDriverReset,			SMU_MSG_RAS_PRI),
+	MSG_MAP(GfxDeviceDriverReset,		     PPSMC_MSG_GfxDriverReset,			SMU_MSG_RAS_PRI | SMU_MSG_NO_PRECHECK),
 	MSG_MAP(DramLogSetDramAddrHigh,		     PPSMC_MSG_DramLogSetDramAddrHigh,		0),
 	MSG_MAP(DramLogSetDramAddrLow,		     PPSMC_MSG_DramLogSetDramAddrLow,		0),
 	MSG_MAP(DramLogSetDramSize,		     PPSMC_MSG_DramLogSetDramSize,		0),
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index a0ce108aab44..520bb7713f14 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -145,7 +145,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU
 	MSG_MAP(GetDpmFreqByIndex,		     PPSMC_MSG_GetDpmFreqByIndex,		1),
 	MSG_MAP(SetPptLimit,			     PPSMC_MSG_SetPptLimit,			0),
 	MSG_MAP(GetPptLimit,			     PPSMC_MSG_GetPptLimit,			1),
-	MSG_MAP(GfxDeviceDriverReset,		     PPSMC_MSG_GfxDriverReset,			SMU_MSG_RAS_PRI),
+	MSG_MAP(GfxDeviceDriverReset,		     PPSMC_MSG_GfxDriverReset,			SMU_MSG_RAS_PRI | SMU_MSG_NO_PRECHECK),
 	MSG_MAP(DramLogSetDramAddrHigh,		     PPSMC_MSG_DramLogSetDramAddrHigh,		0),
 	MSG_MAP(DramLogSetDramAddrLow,		     PPSMC_MSG_DramLogSetDramAddrLow,		0),
 	MSG_MAP(DramLogSetDramSize,		     PPSMC_MSG_DramLogSetDramSize,		0),
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
index 59f9abd0f7b8..f532f7c69259 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c
@@ -256,11 +256,12 @@ static int __smu_cmn_ras_filter_msg(struct smu_context *smu,
 {
 	struct amdgpu_device *adev = smu->adev;
 	uint32_t flags, resp;
-	bool fed_status;
+	bool fed_status, pri;
 
 	flags = __smu_cmn_get_msg_flags(smu, msg);
 	*poll = true;
 
+	pri = !!(flags & SMU_MSG_NO_PRECHECK);
 	/* When there is RAS fatal error, FW won't process non-RAS priority
 	 * messages. Don't allow any messages other than RAS priority messages.
 	 */
@@ -272,15 +273,18 @@ static int __smu_cmn_ras_filter_msg(struct smu_context *smu,
 				smu_get_message_name(smu, msg));
 			return -EACCES;
 		}
+	}
 
+	if (pri || fed_status) {
 		/* FW will ignore non-priority messages when a RAS fatal error
-		 * is detected. Hence it is possible that a previous message
-		 * wouldn't have got response. Allow to continue without polling
-		 * for response status for priority messages.
+		 * or reset condition is detected. Hence it is possible that a
+		 * previous message wouldn't have got response. Allow to
+		 * continue without polling for response status for priority
+		 * messages.
 		 */
 		resp = RREG32(smu->resp_reg);
 		dev_dbg(adev->dev,
-			"Sending RAS priority message %s response status: %x",
+			"Sending priority message %s response status: %x",
 			smu_get_message_name(smu, msg), resp);
 		if (resp == 0)
 			*poll = false;

From 5c15a05b52553ad99c68a661b832aff3eeb2a7b7 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 28 May 2025 11:51:18 -0400
Subject: [PATCH 0310/1455] drm/amdkfd: remove unused code

upages is assigned under cpages = 0, so it isn't really used in this function.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Philip.Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 79251f22b702..f0b690d4bb46 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -736,7 +736,6 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 	if (!cpages) {
 		pr_debug("failed collect migrate device pages [0x%lx 0x%lx]\n",
 			 prange->start, prange->last);
-		upages = svm_migrate_unsuccessful_pages(&migrate);
 		goto out_free;
 	}
 	if (cpages != npages)

From bd6093e2f1601c0c83906f5115a2efb6b93050b1 Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Wed, 28 May 2025 12:38:58 -0400
Subject: [PATCH 0311/1455] drm/amdkfd: return migration pages from copy
 function

dst MIGRATE_PFN_VALID bit and src MIGRATE_PFN_MIGRATE bit
should always be set when migration success. cpage includes
src MIGRATE_PFN_MIGRATE bit set and MIGRATE_PFN_VALID bit
unset pages for both ram and vram when memory is only allocated
without being populated before migration, those ram pages should
be counted as migrate pages and those vram pages should not be
counted as migrate pages. Here migration pages refer to how many
vram pages involved.

-v2 use dst to check MIGRATE_PFN_VALID bit (suggested-by Philip)
-v3 add warning when vram pages is less than migration pages
    return migration pages directly from copy function
-v4 correct comments and copy function return mpage (suggested-by Felix)

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 72 ++++++++++++------------
 1 file changed, 36 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index f0b690d4bb46..5d7eb0234002 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -260,20 +260,7 @@ static void svm_migrate_put_sys_page(unsigned long addr)
 	put_page(page);
 }
 
-static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate)
-{
-	unsigned long upages = 0;
-	unsigned long i;
-
-	for (i = 0; i < migrate->npages; i++) {
-		if (migrate->src[i] & MIGRATE_PFN_VALID &&
-		    !(migrate->src[i] & MIGRATE_PFN_MIGRATE))
-			upages++;
-	}
-	return upages;
-}
-
-static int
+static long
 svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 			 struct migrate_vma *migrate, struct dma_fence **mfence,
 			 dma_addr_t *scratch, uint64_t ttm_res_offset)
@@ -282,7 +269,7 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 	struct amdgpu_device *adev = node->adev;
 	struct device *dev = adev->dev;
 	struct amdgpu_res_cursor cursor;
-	uint64_t mpages = 0;
+	long mpages;
 	dma_addr_t *src;
 	uint64_t *dst;
 	uint64_t i, j;
@@ -296,6 +283,7 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 
 	amdgpu_res_first(prange->ttm_res, ttm_res_offset,
 			 npages << PAGE_SHIFT, &cursor);
+	mpages = 0;
 	for (i = j = 0; (i < npages) && (mpages < migrate->cpages); i++) {
 		struct page *spage;
 
@@ -356,13 +344,14 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 out_free_vram_pages:
 	if (r) {
 		pr_debug("failed %d to copy memory to vram\n", r);
-		for (i = 0; i < npages && mpages; i++) {
+		while (i-- && mpages) {
 			if (!dst[i])
 				continue;
 			svm_migrate_put_vram_page(adev, dst[i]);
 			migrate->dst[i] = 0;
 			mpages--;
 		}
+		mpages = r;
 	}
 
 #ifdef DEBUG_FORCE_MIXED_DOMAINS
@@ -380,7 +369,7 @@ out_free_vram_pages:
 	}
 #endif
 
-	return r;
+	return mpages;
 }
 
 static long
@@ -395,7 +384,7 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	struct dma_fence *mfence = NULL;
 	struct migrate_vma migrate = { 0 };
 	unsigned long cpages = 0;
-	unsigned long mpages = 0;
+	long mpages = 0;
 	dma_addr_t *scratch;
 	void *buf;
 	int r = -ENOMEM;
@@ -441,15 +430,17 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	else
 		pr_debug("0x%lx pages collected\n", cpages);
 
-	r = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset);
+	mpages = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset);
 	migrate_vma_pages(&migrate);
 
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
 
-	mpages = cpages - svm_migrate_unsuccessful_pages(&migrate);
-	pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
+	if (mpages >= 0)
+		pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n",
 			 mpages, cpages, migrate.npages);
+	else
+		r = mpages;
 
 	svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages);
 
@@ -459,14 +450,13 @@ out_free:
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
 				    0, node->id, trigger, r);
 out:
-	if (!r && mpages) {
+	if (!r && mpages > 0) {
 		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_in, pdd->page_in + mpages);
-
-		return mpages;
 	}
-	return r;
+
+	return r ? r : mpages;
 }
 
 /**
@@ -577,7 +567,7 @@ static void svm_migrate_page_free(struct page *page)
 	}
 }
 
-static int
+static long
 svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 			struct migrate_vma *migrate, struct dma_fence **mfence,
 			dma_addr_t *scratch, uint64_t npages)
@@ -586,6 +576,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 	uint64_t *src;
 	dma_addr_t *dst;
 	struct page *dpage;
+	long mpages;
 	uint64_t i = 0, j;
 	uint64_t addr;
 	int r = 0;
@@ -598,6 +589,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 	src = (uint64_t *)(scratch + npages);
 	dst = scratch;
 
+	mpages = 0;
 	for (i = 0, j = 0; i < npages; i++, addr += PAGE_SIZE) {
 		struct page *spage;
 
@@ -646,6 +638,7 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 				     dst[i] >> PAGE_SHIFT, page_to_pfn(dpage));
 
 		migrate->dst[i] = migrate_pfn(page_to_pfn(dpage));
+		mpages++;
 		j++;
 	}
 
@@ -655,13 +648,17 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 out_oom:
 	if (r) {
 		pr_debug("failed %d copy to ram\n", r);
-		while (i--) {
+		while (i-- && mpages) {
+			if (!migrate->dst[i])
+				continue;
 			svm_migrate_put_sys_page(dst[i]);
 			migrate->dst[i] = 0;
+			mpages--;
 		}
+		mpages = r;
 	}
 
-	return r;
+	return mpages;
 }
 
 /**
@@ -688,9 +685,8 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 {
 	struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms);
 	uint64_t npages = (end - start) >> PAGE_SHIFT;
-	unsigned long upages = npages;
 	unsigned long cpages = 0;
-	unsigned long mpages = 0;
+	long mpages = 0;
 	struct amdgpu_device *adev = node->adev;
 	struct kfd_process_device *pdd;
 	struct dma_fence *mfence = NULL;
@@ -744,13 +740,15 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 	else
 		pr_debug("0x%lx pages collected\n", cpages);
 
-	r = svm_migrate_copy_to_ram(adev, prange, &migrate, &mfence,
+	mpages = svm_migrate_copy_to_ram(adev, prange, &migrate, &mfence,
 				    scratch, npages);
 	migrate_vma_pages(&migrate);
 
-	upages = svm_migrate_unsuccessful_pages(&migrate);
-	pr_debug("unsuccessful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
-		 upages, cpages, migrate.npages);
+	if (mpages >= 0)
+		pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n",
+		 mpages, cpages, migrate.npages);
+	else
+		r = mpages;
 
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
@@ -763,8 +761,7 @@ out_free:
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
 				    node->id, 0, trigger, r);
 out:
-	if (!r && cpages) {
-		mpages = cpages - upages;
+	if (!r && mpages > 0) {
 		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_out, pdd->page_out + mpages);
@@ -847,6 +844,9 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 	}
 
 	if (r >= 0) {
+		WARN_ONCE(prange->vram_pages < mpages,
+			"Recorded vram pages(0x%llx) should not be less than migration pages(0x%lx).",
+			prange->vram_pages, mpages);
 		prange->vram_pages -= mpages;
 
 		/* prange does not have vram page set its actual_loc to system

From e9c840d4505d5049da1873acf93744d384b12a0b Mon Sep 17 00:00:00 2001
From: Aurabindo Pillai <aurabindo.pillai@amd.com>
Date: Tue, 5 Aug 2025 10:02:07 -0400
Subject: [PATCH 0312/1455] drm/amd/display: Fix vupdate_offload_work doc

Fix the following warning in struct documentation:

drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h:168: warning: expecting prototype for struct dm_vupdate_work. Prototype was for struct vupdate_offload_work instead

Fixes: c210b757b400 ("drm/amd/display: fix dmub access race condition")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 94f312bae9ac..cbd107493f8a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -154,7 +154,7 @@ struct idle_workqueue {
 };
 
 /**
- * struct dm_vupdate_work - Work data for periodic action in idle
+ * struct vupdate_offload_work - Work data for offloading task from vupdate handler
  * @work: Kernel work data for the work event
  * @adev: amdgpu_device back pointer
  * @stream: DC stream associated with the crtc

From 83953ec1fedf37b900bded05ced31e6b7858cca3 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Sat, 2 Aug 2025 01:48:09 +0800
Subject: [PATCH 0313/1455] drm/amd/pm: Add dpm interface for temp metrics

Add dpm interface to get gpuboard/baseboard temperature metrics

v2: Add temperature metrics support check(Lijo)

v3: Return error code in case of operation not supported(Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/include/kgd_pp_interface.h    | 81 +++++++++++++++++++
 drivers/gpu/drm/amd/pm/amdgpu_dpm.c           | 60 ++++++++++++++
 drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h       |  4 +
 3 files changed, 145 insertions(+)

diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
index e2b1ea7467b0..2f7e4b5bebf3 100644
--- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
@@ -30,6 +30,12 @@ extern const struct amdgpu_ip_block_version smu_v12_0_ip_block;
 extern const struct amdgpu_ip_block_version smu_v13_0_ip_block;
 extern const struct amdgpu_ip_block_version smu_v14_0_ip_block;
 
+enum smu_temp_metric_type {
+	SMU_TEMP_METRIC_BASEBOARD,
+	SMU_TEMP_METRIC_GPUBOARD,
+	SMU_TEMP_METRIC_MAX,
+};
+
 enum smu_event_type {
 	SMU_EVENT_RESET_COMPLETE = 0,
 };
@@ -496,6 +502,8 @@ struct amd_pm_funcs {
 	int (*set_df_cstate)(void *handle, enum pp_df_cstate state);
 	int (*set_xgmi_pstate)(void *handle, uint32_t pstate);
 	ssize_t (*get_gpu_metrics)(void *handle, void **table);
+	ssize_t (*get_temp_metrics)(void *handle, enum smu_temp_metric_type type, void *table);
+	bool (*temp_metrics_is_supported)(void *handle, enum smu_temp_metric_type type);
 	ssize_t (*get_xcp_metrics)(void *handle, int xcp_id, void *table);
 	ssize_t (*get_pm_metrics)(void *handle, void *pmmetrics, size_t size);
 	int (*set_watermarks_for_clock_ranges)(void *handle,
@@ -1595,6 +1603,79 @@ struct amdgpu_pm_metrics {
 	uint8_t data[];
 };
 
+enum amdgpu_vr_temp {
+	AMDGPU_VDDCR_VDD0_TEMP,
+	AMDGPU_VDDCR_VDD1_TEMP,
+	AMDGPU_VDDCR_VDD2_TEMP,
+	AMDGPU_VDDCR_VDD3_TEMP,
+	AMDGPU_VDDCR_SOC_A_TEMP,
+	AMDGPU_VDDCR_SOC_C_TEMP,
+	AMDGPU_VDDCR_SOCIO_A_TEMP,
+	AMDGPU_VDDCR_SOCIO_C_TEMP,
+	AMDGPU_VDD_085_HBM_TEMP,
+	AMDGPU_VDDCR_11_HBM_B_TEMP,
+	AMDGPU_VDDCR_11_HBM_D_TEMP,
+	AMDGPU_VDD_USR_TEMP,
+	AMDGPU_VDDIO_11_E32_TEMP,
+	AMDGPU_VR_MAX_TEMP_ENTRIES,
+};
+
+enum amdgpu_system_temp {
+	AMDGPU_UBB_FPGA_TEMP,
+	AMDGPU_UBB_FRONT_TEMP,
+	AMDGPU_UBB_BACK_TEMP,
+	AMDGPU_UBB_OAM7_TEMP,
+	AMDGPU_UBB_IBC_TEMP,
+	AMDGPU_UBB_UFPGA_TEMP,
+	AMDGPU_UBB_OAM1_TEMP,
+	AMDGPU_OAM_0_1_HSC_TEMP,
+	AMDGPU_OAM_2_3_HSC_TEMP,
+	AMDGPU_OAM_4_5_HSC_TEMP,
+	AMDGPU_OAM_6_7_HSC_TEMP,
+	AMDGPU_UBB_FPGA_0V72_VR_TEMP,
+	AMDGPU_UBB_FPGA_3V3_VR_TEMP,
+	AMDGPU_RETIMER_0_1_2_3_1V2_VR_TEMP,
+	AMDGPU_RETIMER_4_5_6_7_1V2_VR_TEMP,
+	AMDGPU_RETIMER_0_1_0V9_VR_TEMP,
+	AMDGPU_RETIMER_4_5_0V9_VR_TEMP,
+	AMDGPU_RETIMER_2_3_0V9_VR_TEMP,
+	AMDGPU_RETIMER_6_7_0V9_VR_TEMP,
+	AMDGPU_OAM_0_1_2_3_3V3_VR_TEMP,
+	AMDGPU_OAM_4_5_6_7_3V3_VR_TEMP,
+	AMDGPU_IBC_HSC_TEMP,
+	AMDGPU_IBC_TEMP,
+	AMDGPU_SYSTEM_MAX_TEMP_ENTRIES = 32,
+};
+
+enum amdgpu_node_temp {
+	AMDGPU_RETIMER_X_TEMP,
+	AMDGPU_OAM_X_IBC_TEMP,
+	AMDGPU_OAM_X_IBC_2_TEMP,
+	AMDGPU_OAM_X_VDD18_VR_TEMP,
+	AMDGPU_OAM_X_04_HBM_B_VR_TEMP,
+	AMDGPU_OAM_X_04_HBM_D_VR_TEMP,
+	AMDGPU_NODE_MAX_TEMP_ENTRIES = 12,
+};
+
+struct amdgpu_gpuboard_temp_metrics_v1_0 {
+	struct metrics_table_header common_header;
+	uint16_t label_version;
+	uint16_t node_id;
+	uint64_t accumulation_counter;
+	/* Encoded temperature in Celcius, 24:31 is sensor id 0:23 is temp value */
+	uint32_t node_temp[AMDGPU_NODE_MAX_TEMP_ENTRIES];
+	uint32_t vr_temp[AMDGPU_VR_MAX_TEMP_ENTRIES];
+};
+
+struct amdgpu_baseboard_temp_metrics_v1_0 {
+	struct metrics_table_header common_header;
+	uint16_t label_version;
+	uint16_t node_id;
+	uint64_t accumulation_counter;
+	/* Encoded temperature in Celcius, 24:31 is sensor id 0:23 is temp value */
+	uint32_t system_temp[AMDGPU_SYSTEM_MAX_TEMP_ENTRIES];
+};
+
 struct amdgpu_partition_metrics_v1_0 {
 	struct metrics_table_header common_header;
 	/* Current clocks (Mhz) */
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 71d986dd7a6e..8e19054578f7 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -2037,6 +2037,66 @@ int amdgpu_dpm_get_dpm_clock_table(struct amdgpu_device *adev,
 	return ret;
 }
 
+/**
+ * amdgpu_dpm_get_temp_metrics - Retrieve metrics for a specific compute
+ * partition
+ * @adev: Pointer to the device.
+ * @type: Identifier for the temperature type metrics to be fetched.
+ * @table: Pointer to a buffer where the metrics will be stored. If NULL, the
+ * function returns the size of the metrics structure.
+ *
+ * This function retrieves metrics for a specific temperature type, If the
+ * table parameter is NULL, the function returns the size of the metrics
+ * structure without populating it.
+ *
+ * Return: Size of the metrics structure on success, or a negative error code on failure.
+ */
+ssize_t amdgpu_dpm_get_temp_metrics(struct amdgpu_device *adev,
+				    enum smu_temp_metric_type type, void *table)
+{
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+	int ret;
+
+	if (!pp_funcs->get_temp_metrics ||
+	    !amdgpu_dpm_is_temp_metrics_supported(adev, type))
+		return -EOPNOTSUPP;
+
+	mutex_lock(&adev->pm.mutex);
+	ret = pp_funcs->get_temp_metrics(adev->powerplay.pp_handle, type, table);
+	mutex_unlock(&adev->pm.mutex);
+
+	return ret;
+}
+
+/**
+ * amdgpu_dpm_is_temp_metrics_supported - Return if specific temperature metrics support
+ * is available
+ * @adev: Pointer to the device.
+ * @type: Identifier for the temperature type metrics to be fetched.
+ *
+ * This function returns metrics if specific temperature metrics type is supported or not.
+ *
+ * Return: True in case of metrics type supported else false.
+ */
+bool amdgpu_dpm_is_temp_metrics_supported(struct amdgpu_device *adev,
+					  enum smu_temp_metric_type type)
+{
+	const struct amd_pm_funcs *pp_funcs = adev->powerplay.pp_funcs;
+	bool support_temp_metrics = false;
+
+	if (!pp_funcs->temp_metrics_is_supported)
+		return support_temp_metrics;
+
+	if (is_support_sw_smu(adev)) {
+		mutex_lock(&adev->pm.mutex);
+		support_temp_metrics =
+			pp_funcs->temp_metrics_is_supported(adev->powerplay.pp_handle, type);
+		mutex_unlock(&adev->pm.mutex);
+	}
+
+	return support_temp_metrics;
+}
+
 /**
  * amdgpu_dpm_get_xcp_metrics - Retrieve metrics for a specific compute
  * partition
diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
index 768317ee1486..09962db988d6 100644
--- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
@@ -526,6 +526,8 @@ int amdgpu_dpm_set_power_profile_mode(struct amdgpu_device *adev,
 int amdgpu_dpm_get_gpu_metrics(struct amdgpu_device *adev, void **table);
 ssize_t amdgpu_dpm_get_xcp_metrics(struct amdgpu_device *adev, int xcp_id,
 				   void *table);
+ssize_t amdgpu_dpm_get_temp_metrics(struct amdgpu_device *adev,
+				    enum smu_temp_metric_type type, void *table);
 
 /**
  * @get_pm_metrics: Get one snapshot of power management metrics from PMFW. The
@@ -613,5 +615,7 @@ ssize_t amdgpu_dpm_get_pm_policy_info(struct amdgpu_device *adev,
 int amdgpu_dpm_reset_sdma(struct amdgpu_device *adev, uint32_t inst_mask);
 bool amdgpu_dpm_reset_sdma_is_supported(struct amdgpu_device *adev);
 int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask);
+bool amdgpu_dpm_is_temp_metrics_supported(struct amdgpu_device *adev,
+					  enum smu_temp_metric_type type);
 
 #endif

From 775c7e8e4d57022dd9e6f16eefc31abfda090aba Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Sat, 2 Aug 2025 02:10:12 +0800
Subject: [PATCH 0314/1455] drm/amd/pm: Add smu interface for temp metrics

Add smu interface to get baseboard/gpuboard temperature metrics

v2: Rename is_support to is_supported(Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     | 29 +++++++++++++++++++
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h | 27 +++++++++++++++++
 2 files changed, 56 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index b47cb4a5f488..8b015107f761 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -3831,6 +3831,33 @@ int smu_set_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type,
 	return ret;
 }
 
+static ssize_t smu_sys_get_temp_metrics(void *handle, enum smu_temp_metric_type type, void *table)
+{
+	struct smu_context *smu = handle;
+
+	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
+		return -EOPNOTSUPP;
+
+	if (!smu->smu_temp.temp_funcs || !smu->smu_temp.temp_funcs->get_temp_metrics)
+		return -EOPNOTSUPP;
+
+	return smu->smu_temp.temp_funcs->get_temp_metrics(smu, type, table);
+}
+
+static bool smu_temp_metrics_is_supported(void *handle, enum smu_temp_metric_type type)
+{
+	struct smu_context *smu = handle;
+	bool ret = false;
+
+	if (!smu->pm_enabled)
+		return false;
+
+	if (smu->smu_temp.temp_funcs && smu->smu_temp.temp_funcs->temp_metrics_is_supported)
+		ret = smu->smu_temp.temp_funcs->temp_metrics_is_supported(smu, type);
+
+	return ret;
+}
+
 static ssize_t smu_sys_get_xcp_metrics(void *handle, int xcp_id, void *table)
 {
 	struct smu_context *smu = handle;
@@ -3903,6 +3930,8 @@ static const struct amd_pm_funcs swsmu_pm_funcs = {
 	.get_dpm_clock_table              = smu_get_dpm_clock_table,
 	.get_smu_prv_buf_details = smu_get_prv_buffer_details,
 	.get_xcp_metrics                  = smu_sys_get_xcp_metrics,
+	.get_temp_metrics             = smu_sys_get_temp_metrics,
+	.temp_metrics_is_supported      = smu_temp_metrics_is_supported,
 };
 
 int smu_wait_for_event(struct smu_context *smu, enum smu_event_type event,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index b52e194397e2..b490c39e313e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -396,6 +396,10 @@ struct smu_dpm_context {
 	struct smu_dpm_policy_ctxt *dpm_policies;
 };
 
+struct smu_temp_context {
+	const struct smu_temp_funcs      *temp_funcs;
+};
+
 struct smu_power_gate {
 	bool uvd_gated;
 	bool vce_gated;
@@ -529,6 +533,7 @@ struct smu_context {
 	struct smu_table_context	smu_table;
 	struct smu_dpm_context		smu_dpm;
 	struct smu_power_context	smu_power;
+	struct smu_temp_context		smu_temp;
 	struct smu_feature		smu_feature;
 	struct amd_pp_display_configuration  *display_config;
 	struct smu_baco_context		smu_baco;
@@ -623,6 +628,28 @@ struct smu_context {
 
 struct i2c_adapter;
 
+/**
+ * struct smu_temp_funcs - Callbacks used to get temperature data.
+ */
+struct smu_temp_funcs {
+	/**
+	 * @get_temp_metrics: Calibrate voltage/frequency curve to fit the system's
+	 *           power delivery and voltage margins. Required for adaptive
+	 * @type Temperature metrics type(baseboard/gpuboard)
+	 * Return: Size of &table
+	 */
+	ssize_t (*get_temp_metrics)(struct smu_context *smu,
+				    enum smu_temp_metric_type type, void *table);
+
+	/**
+	 * @temp_metrics_is_support: Get if specific temperature metrics is supported
+	 * @type Temperature metrics type(baseboard/gpuboard)
+	 * Return: true if supported else false
+	 */
+	bool (*temp_metrics_is_supported)(struct smu_context *smu, enum smu_temp_metric_type type);
+
+};
+
 /**
  * struct pptable_funcs - Callbacks used to interact with the SMU.
  */

From 793ff2bafeb10e2b584d77ef0689d50ac13fd4c6 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Sat, 2 Aug 2025 02:41:50 +0800
Subject: [PATCH 0315/1455] drm/amd/pm: Update pmfw header for smu_v13_0_12

Update pmfw header for smu_v13_0_12 with system temperature metrics
table

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h  | 74 ++++++++++++++++++-
 .../pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h |  6 +-
 2 files changed, 78 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h
index 0a2ca544f4e3..1c407a8e96ee 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h
@@ -135,7 +135,63 @@ typedef enum {
   GFX_DVM_MARGIN_COUNT
 } GFX_DVM_MARGIN_e;
 
-#define SMU_METRICS_TABLE_VERSION 0x13
+typedef enum{
+  SYSTEM_TEMP_UBB_FPGA,
+  SYSTEM_TEMP_UBB_FRONT,
+  SYSTEM_TEMP_UBB_BACK,
+  SYSTEM_TEMP_UBB_OAM7,
+  SYSTEM_TEMP_UBB_IBC,
+  SYSTEM_TEMP_UBB_UFPGA,
+  SYSTEM_TEMP_UBB_OAM1,
+  SYSTEM_TEMP_OAM_0_1_HSC,
+  SYSTEM_TEMP_OAM_2_3_HSC,
+  SYSTEM_TEMP_OAM_4_5_HSC,
+  SYSTEM_TEMP_OAM_6_7_HSC,
+  SYSTEM_TEMP_UBB_FPGA_0V72_VR,
+  SYSTEM_TEMP_UBB_FPGA_3V3_VR,
+  SYSTEM_TEMP_RETIMER_0_1_2_3_1V2_VR,
+  SYSTEM_TEMP_RETIMER_4_5_6_7_1V2_VR,
+  SYSTEM_TEMP_RETIMER_0_1_0V9_VR,
+  SYSTEM_TEMP_RETIMER_4_5_0V9_VR,
+  SYSTEM_TEMP_RETIMER_2_3_0V9_VR,
+  SYSTEM_TEMP_RETIMER_6_7_0V9_VR,
+  SYSTEM_TEMP_OAM_0_1_2_3_3V3_VR,
+  SYSTEM_TEMP_OAM_4_5_6_7_3V3_VR,
+  SYSTEM_TEMP_IBC_HSC,
+  SYSTEM_TEMP_IBC,
+  SYSTEM_TEMP_MAX_ENTRIES   = 32
+} SYSTEM_TEMP_e;
+
+typedef enum{
+  NODE_TEMP_RETIMER,
+  NODE_TEMP_IBC_TEMP,
+  NODE_TEMP_IBC_2_TEMP,
+  NODE_TEMP_VDD18_VR_TEMP,
+  NODE_TEMP_04_HBM_B_VR_TEMP,
+  NODE_TEMP_04_HBM_D_VR_TEMP,
+  NODE_TEMP_MAX_TEMP_ENTRIES    = 12
+} NODE_TEMP_e;
+
+typedef enum {
+  SVI_VDDCR_VDD0_TEMP,
+  SVI_VDDCR_VDD1_TEMP,
+  SVI_VDDCR_VDD2_TEMP,
+  SVI_VDDCR_VDD3_TEMP,
+  SVI_VDDCR_SOC_A_TEMP,
+  SVI_VDDCR_SOC_C_TEMP,
+  SVI_VDDCR_SOCIO_A_TEMP,
+  SVI_VDDCR_SOCIO_C_TEMP,
+  SVI_VDD_085_HBM_TEMP,
+  SVI_VDDCR_11_HBM_B_TEMP,
+  SVI_VDDCR_11_HBM_D_TEMP,
+  SVI_VDD_USR_TEMP,
+  SVI_VDDIO_11_E32_TEMP,
+  SVI_MAX_TEMP_ENTRIES,   // 13
+} SVI_TEMP_e;
+
+#define SMU_METRICS_TABLE_VERSION 0x14
+
+#define SMU_SYSTEM_METRICS_TABLE_VERSION 0x0
 
 typedef struct __attribute__((packed, aligned(4))) {
   uint64_t AccumulationCounter;
@@ -231,11 +287,27 @@ typedef struct __attribute__((packed, aligned(4))) {
   uint64_t GfxclkBelowHostLimitThmAcc[8];
   uint64_t GfxclkBelowHostLimitTotalAcc[8];
   uint64_t GfxclkLowUtilizationAcc[8];
+
+  uint32_t AidTemperature[4];
+  uint32_t XcdTemperature[8];
+  uint32_t HbmTemperature[8];
 } MetricsTable_t;
 
 #define SMU_VF_METRICS_TABLE_MASK (1 << 31)
 #define SMU_VF_METRICS_TABLE_VERSION (0x6 | SMU_VF_METRICS_TABLE_MASK)
 
+#pragma pack(push, 4)
+typedef struct {
+  uint64_t AccumulationCounter;                             // Last update timestamp
+  uint16_t LabelVersion;                                    // Defaults to 0.
+  uint16_t NodeIdentifier;                                  // Unique identifier to each node on system.
+  int16_t  SystemTemperatures[SYSTEM_TEMP_MAX_ENTRIES];     // Signed integer temperature value in Celsius, unused fields are set to 0xFFFF
+  int16_t  NodeTemperatures[NODE_TEMP_MAX_TEMP_ENTRIES];    // Signed integer temperature value in Celsius, unused fields are set to 0xFFFF
+  int16_t  VrTemperatures[SVI_MAX_TEMP_ENTRIES];            // Signed integer temperature value in Celsius
+  int16_t  spare[3];
+} SystemMetricsTable_t;
+#pragma pack(pop)
+
 typedef struct __attribute__((packed, aligned(4))) {
   uint32_t AccumulationCounter;
   uint32_t InstGfxclk_TargFreq;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h
index e1f490b6ce64..aff2776a8b6f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h
@@ -116,7 +116,11 @@
 #define PPSMC_MSG_DumpErrorRecord                   0x57
 #define PPSMC_MSG_EraseRasTable                     0x58
 #define PPSMC_MSG_GetStaticMetricsTable             0x59
-#define PPSMC_Message_Count                         0x5A
+#define PPSMC_MSG_ResetVfArbitersByIndex            0x5A
+#define PPSMC_MSG_GetBadPageSeverity                0x5B
+#define PPSMC_MSG_GetSystemMetricsTable             0x5C
+#define PPSMC_MSG_GetSystemMetricsVersion           0x5D
+#define PPSMC_Message_Count                         0x5E
 
 //PPSMC Reset Types for driver msg argument
 #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET        0x1

From 33074558ec8481a54b16e76dd79c9beff4f673bd Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Sat, 2 Aug 2025 04:26:13 +0800
Subject: [PATCH 0316/1455] drm/amd/pm: Fetch and fill temperature metrics

Fetch system metrics table to fill gpuboard/baseboard temperature
metrics data for smu_v13_0_12

v2: Remove unnecessary checks, used separate metrics time for
temperature metrics table(Lijo)

v3: Use cached values for back to back system metrics query(Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     |   1 +
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h |   2 +
 drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h  |   3 +-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 248 +++++++++++++++++-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  |   6 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h  |   3 +
 drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h        |  26 ++
 7 files changed, 287 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 8b015107f761..dc48a1dd8be4 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -766,6 +766,7 @@ static int smu_set_funcs(struct amdgpu_device *adev)
 	case IP_VERSION(13, 0, 14):
 	case IP_VERSION(13, 0, 12):
 		smu_v13_0_6_set_ppt_funcs(smu);
+		smu_v13_0_6_set_temp_funcs(smu);
 		/* Enable pp_od_clk_voltage node */
 		smu->od_enabled = true;
 		break;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index b490c39e313e..611b381b9147 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -257,6 +257,7 @@ struct smu_table {
 	void *cpu_addr;
 	struct amdgpu_bo *bo;
 	uint32_t version;
+	unsigned long  metrics_time;
 };
 
 enum smu_perf_level_designation {
@@ -322,6 +323,7 @@ enum smu_table_id {
 	SMU_TABLE_ECCINFO,
 	SMU_TABLE_COMBO_PPTABLE,
 	SMU_TABLE_WIFIBAND,
+	SMU_TABLE_TEMP_METRICS,
 	SMU_TABLE_COUNT,
 };
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
index fba980fa5a01..2256c77da636 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_types.h
@@ -278,7 +278,8 @@
 	__SMU_DUMMY_MAP(MALLPowerState), \
 	__SMU_DUMMY_MAP(ResetSDMA), \
 	__SMU_DUMMY_MAP(ResetVCN), \
-	__SMU_DUMMY_MAP(GetStaticMetricsTable),
+	__SMU_DUMMY_MAP(GetStaticMetricsTable), \
+	__SMU_DUMMY_MAP(GetSystemMetricsTable),
 
 #undef __SMU_DUMMY_MAP
 #define __SMU_DUMMY_MAP(type)	SMU_MSG_##type
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 476b7f062a24..920f60da9c5c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -138,6 +138,7 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(SetThrottlingPolicy,                 PPSMC_MSG_SetThrottlingPolicy,             0),
 	MSG_MAP(ResetSDMA,                           PPSMC_MSG_ResetSDMA,                       0),
 	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           1),
+	MSG_MAP(GetSystemMetricsTable,               PPSMC_MSG_GetSystemMetricsTable,           0),
 };
 
 static int smu_v13_0_12_get_enabled_mask(struct smu_context *smu,
@@ -184,7 +185,8 @@ static int smu_v13_0_12_fru_get_product_info(struct smu_context *smu,
 
 int smu_v13_0_12_get_max_metrics_size(void)
 {
-	return max(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t));
+	return max3(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t),
+		   sizeof(SystemMetricsTable_t));
 }
 
 static void smu_v13_0_12_init_xgmi_data(struct smu_context *smu,
@@ -359,6 +361,245 @@ int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
 	return 0;
 }
 
+static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu, void *metrics_table,
+						 bool bypass_cache)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size;
+	struct smu_table *table = &smu_table->driver_table;
+	int ret;
+
+	if (bypass_cache || !smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time ||
+	    time_after(jiffies,
+		       smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time +
+		       msecs_to_jiffies(1))) {
+		ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetSystemMetricsTable, NULL);
+		if (ret) {
+			dev_info(smu->adev->dev,
+				 "Failed to export system metrics table!\n");
+			return ret;
+		}
+
+		amdgpu_asic_invalidate_hdp(smu->adev, NULL);
+		memcpy(smu_table->metrics_table, table->cpu_addr, table_size);
+
+		smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time = jiffies;
+	}
+
+	if (metrics_table)
+		memcpy(metrics_table, smu_table->metrics_table, sizeof(SystemMetricsTable_t));
+
+	return 0;
+}
+
+static enum amdgpu_node_temp smu_v13_0_12_get_node_sensor_type(NODE_TEMP_e type)
+{
+	switch (type) {
+	case NODE_TEMP_RETIMER:
+		return AMDGPU_RETIMER_X_TEMP;
+	case NODE_TEMP_IBC_TEMP:
+		return AMDGPU_OAM_X_IBC_TEMP;
+	case NODE_TEMP_IBC_2_TEMP:
+		return AMDGPU_OAM_X_IBC_2_TEMP;
+	case NODE_TEMP_VDD18_VR_TEMP:
+		return AMDGPU_OAM_X_VDD18_VR_TEMP;
+	case NODE_TEMP_04_HBM_B_VR_TEMP:
+		return AMDGPU_OAM_X_04_HBM_B_VR_TEMP;
+	case NODE_TEMP_04_HBM_D_VR_TEMP:
+		return AMDGPU_OAM_X_04_HBM_D_VR_TEMP;
+	default:
+		return -EINVAL;
+	}
+}
+
+static enum amdgpu_vr_temp smu_v13_0_12_get_vr_sensor_type(SVI_TEMP_e type)
+{
+	switch (type) {
+	case SVI_VDDCR_VDD0_TEMP:
+		return AMDGPU_VDDCR_VDD0_TEMP;
+	case SVI_VDDCR_VDD1_TEMP:
+		return AMDGPU_VDDCR_VDD1_TEMP;
+	case SVI_VDDCR_VDD2_TEMP:
+		return AMDGPU_VDDCR_VDD2_TEMP;
+	case SVI_VDDCR_VDD3_TEMP:
+		return AMDGPU_VDDCR_VDD3_TEMP;
+	case SVI_VDDCR_SOC_A_TEMP:
+		return AMDGPU_VDDCR_SOC_A_TEMP;
+	case SVI_VDDCR_SOC_C_TEMP:
+		return AMDGPU_VDDCR_SOC_C_TEMP;
+	case SVI_VDDCR_SOCIO_A_TEMP:
+		return AMDGPU_VDDCR_SOCIO_A_TEMP;
+	case SVI_VDDCR_SOCIO_C_TEMP:
+		return AMDGPU_VDDCR_SOCIO_C_TEMP;
+	case SVI_VDD_085_HBM_TEMP:
+		return AMDGPU_VDD_085_HBM_TEMP;
+	case SVI_VDDCR_11_HBM_B_TEMP:
+		return AMDGPU_VDDCR_11_HBM_B_TEMP;
+	case SVI_VDDCR_11_HBM_D_TEMP:
+		return AMDGPU_VDDCR_11_HBM_D_TEMP;
+	case SVI_VDD_USR_TEMP:
+		return AMDGPU_VDD_USR_TEMP;
+	case SVI_VDDIO_11_E32_TEMP:
+		return AMDGPU_VDDIO_11_E32_TEMP;
+	default:
+		return -EINVAL;
+	}
+}
+
+static enum amdgpu_system_temp smu_v13_0_12_get_system_sensor_type(SYSTEM_TEMP_e type)
+{
+	switch (type) {
+	case SYSTEM_TEMP_UBB_FPGA:
+		return AMDGPU_UBB_FPGA_TEMP;
+	case SYSTEM_TEMP_UBB_FRONT:
+		return AMDGPU_UBB_FRONT_TEMP;
+	case SYSTEM_TEMP_UBB_BACK:
+		return AMDGPU_UBB_BACK_TEMP;
+	case SYSTEM_TEMP_UBB_OAM7:
+		return AMDGPU_UBB_OAM7_TEMP;
+	case SYSTEM_TEMP_UBB_IBC:
+		return AMDGPU_UBB_IBC_TEMP;
+	case SYSTEM_TEMP_UBB_UFPGA:
+		return AMDGPU_UBB_UFPGA_TEMP;
+	case SYSTEM_TEMP_UBB_OAM1:
+		return AMDGPU_UBB_OAM1_TEMP;
+	case SYSTEM_TEMP_OAM_0_1_HSC:
+		return AMDGPU_OAM_0_1_HSC_TEMP;
+	case SYSTEM_TEMP_OAM_2_3_HSC:
+		return AMDGPU_OAM_2_3_HSC_TEMP;
+	case SYSTEM_TEMP_OAM_4_5_HSC:
+		return AMDGPU_OAM_4_5_HSC_TEMP;
+	case SYSTEM_TEMP_OAM_6_7_HSC:
+		return AMDGPU_OAM_6_7_HSC_TEMP;
+	case SYSTEM_TEMP_UBB_FPGA_0V72_VR:
+		return AMDGPU_UBB_FPGA_0V72_VR_TEMP;
+	case SYSTEM_TEMP_UBB_FPGA_3V3_VR:
+		return AMDGPU_UBB_FPGA_3V3_VR_TEMP;
+	case SYSTEM_TEMP_RETIMER_0_1_2_3_1V2_VR:
+		return AMDGPU_RETIMER_0_1_2_3_1V2_VR_TEMP;
+	case SYSTEM_TEMP_RETIMER_4_5_6_7_1V2_VR:
+		return AMDGPU_RETIMER_4_5_6_7_1V2_VR_TEMP;
+	case SYSTEM_TEMP_RETIMER_0_1_0V9_VR:
+		return AMDGPU_RETIMER_0_1_0V9_VR_TEMP;
+	case SYSTEM_TEMP_RETIMER_4_5_0V9_VR:
+		return AMDGPU_RETIMER_4_5_0V9_VR_TEMP;
+	case SYSTEM_TEMP_RETIMER_2_3_0V9_VR:
+		return AMDGPU_RETIMER_2_3_0V9_VR_TEMP;
+	case SYSTEM_TEMP_RETIMER_6_7_0V9_VR:
+		return AMDGPU_RETIMER_6_7_0V9_VR_TEMP;
+	case SYSTEM_TEMP_OAM_0_1_2_3_3V3_VR:
+		return AMDGPU_OAM_0_1_2_3_3V3_VR_TEMP;
+	case SYSTEM_TEMP_OAM_4_5_6_7_3V3_VR:
+		return AMDGPU_OAM_4_5_6_7_3V3_VR_TEMP;
+	case SYSTEM_TEMP_IBC_HSC:
+		return AMDGPU_IBC_HSC_TEMP;
+	case SYSTEM_TEMP_IBC:
+		return AMDGPU_IBC_TEMP;
+	default:
+		return -EINVAL;
+	}
+}
+
+static bool smu_v13_0_12_is_temp_metrics_supported(struct smu_context *smu,
+						   enum smu_temp_metric_type type)
+{
+	switch (type) {
+	case SMU_TEMP_METRIC_BASEBOARD:
+		if (smu->adev->gmc.xgmi.physical_node_id == 0 &&
+		    smu->adev->gmc.xgmi.num_physical_nodes > 1 &&
+		    smu_v13_0_6_cap_supported(smu, SMU_CAP(TEMP_METRICS)))
+			return true;
+		break;
+	case SMU_TEMP_METRIC_GPUBOARD:
+		return smu_v13_0_6_cap_supported(smu, SMU_CAP(TEMP_METRICS));
+	default:
+		break;
+	}
+
+	return false;
+}
+
+static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
+					     enum smu_temp_metric_type type, void *table)
+{
+	struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics;
+	struct amdgpu_baseboard_temp_metrics_v1_0 *baseboard_temp_metrics;
+	SystemMetricsTable_t *metrics;
+	int ret, sensor_type;
+	u32 idx, sensors;
+	ssize_t size;
+
+	size = (type == SMU_TEMP_METRIC_GPUBOARD) ?
+		sizeof(*gpuboard_temp_metrics) : sizeof(*baseboard_temp_metrics);
+
+	if (!table)
+		goto out;
+	metrics = kzalloc(sizeof(SystemMetricsTable_t), GFP_KERNEL);
+	if (!metrics)
+		return -ENOMEM;
+	gpuboard_temp_metrics = (struct amdgpu_gpuboard_temp_metrics_v1_0 *)table;
+	baseboard_temp_metrics = (struct amdgpu_baseboard_temp_metrics_v1_0 *)table;
+	if (type  == SMU_TEMP_METRIC_GPUBOARD)
+		smu_cmn_init_gpuboard_temp_metrics(gpuboard_temp_metrics, 1, 0);
+	else if (type  == SMU_TEMP_METRIC_BASEBOARD)
+		smu_cmn_init_baseboard_temp_metrics(baseboard_temp_metrics, 1, 0);
+
+	ret = smu_v13_0_12_get_system_metrics_table(smu, metrics, false);
+	if (ret) {
+		kfree(metrics);
+		return ret;
+	}
+
+	if (type == SMU_TEMP_METRIC_GPUBOARD) {
+		gpuboard_temp_metrics->accumulation_counter = metrics->AccumulationCounter;
+		gpuboard_temp_metrics->label_version = metrics->LabelVersion;
+		gpuboard_temp_metrics->node_id = metrics->NodeIdentifier;
+
+		idx = 0;
+		for (sensors = 0; sensors < NODE_TEMP_MAX_TEMP_ENTRIES; sensors++) {
+			if (metrics->NodeTemperatures[sensors] != -1) {
+				sensor_type = smu_v13_0_12_get_node_sensor_type(sensors);
+				gpuboard_temp_metrics->node_temp[idx] =
+					((int)metrics->NodeTemperatures[sensors])  & 0xFFFFFF;
+				gpuboard_temp_metrics->node_temp[idx] |= (sensor_type << 24);
+				idx++;
+			}
+		}
+
+		idx = 0;
+
+		for (sensors = 0; sensors < SVI_MAX_TEMP_ENTRIES; sensors++) {
+			if (metrics->VrTemperatures[sensors] != -1) {
+				sensor_type = smu_v13_0_12_get_vr_sensor_type(sensors);
+				gpuboard_temp_metrics->vr_temp[idx] =
+					((int)metrics->VrTemperatures[sensors])  & 0xFFFFFF;
+				gpuboard_temp_metrics->vr_temp[idx] |= (sensor_type << 24);
+				idx++;
+			}
+		}
+	} else if (type == SMU_TEMP_METRIC_BASEBOARD) {
+		baseboard_temp_metrics->accumulation_counter = metrics->AccumulationCounter;
+		baseboard_temp_metrics->label_version = metrics->LabelVersion;
+		baseboard_temp_metrics->node_id = metrics->NodeIdentifier;
+
+		idx = 0;
+		for (sensors = 0; sensors < SYSTEM_TEMP_MAX_ENTRIES; sensors++) {
+			if (metrics->SystemTemperatures[sensors] != -1) {
+				sensor_type = smu_v13_0_12_get_system_sensor_type(sensors);
+				baseboard_temp_metrics->system_temp[idx] =
+					((int)metrics->SystemTemperatures[sensors])  & 0xFFFFFF;
+				baseboard_temp_metrics->system_temp[idx] |= (sensor_type << 24);
+				idx++;
+			}
+		}
+	}
+
+	kfree(metrics);
+
+out:
+	return size;
+}
+
 ssize_t smu_v13_0_12_get_xcp_metrics(struct smu_context *smu, struct amdgpu_xcp *xcp, void *table, void *smu_metrics)
 {
 	const u8 num_jpeg_rings = NUM_JPEG_RINGS_FW;
@@ -572,3 +813,8 @@ ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table, void
 
 	return sizeof(*gpu_metrics);
 }
+
+const struct smu_temp_funcs smu_v13_0_12_temp_funcs = {
+	.temp_metrics_is_supported = smu_v13_0_12_is_temp_metrics_supported,
+	.get_temp_metrics = smu_v13_0_12_get_temp_metrics,
+};
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 520bb7713f14..5f9f74b9109d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3871,3 +3871,9 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
 	amdgpu_mca_smu_init_funcs(smu->adev, &smu_v13_0_6_mca_smu_funcs);
 	amdgpu_aca_set_smu_funcs(smu->adev, &smu_v13_0_6_aca_smu_funcs);
 }
+
+void smu_v13_0_6_set_temp_funcs(struct smu_context *smu)
+{
+	smu->smu_temp.temp_funcs = (amdgpu_ip_version(smu->adev, MP1_HWIP, 0)
+			== IP_VERSION(13, 0, 12)) ? &smu_v13_0_12_temp_funcs : NULL;
+}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index 67b30674fd31..ece04ad724fb 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -68,10 +68,12 @@ enum smu_v13_0_6_caps {
 	SMU_CAP(HST_LIMIT_METRICS),
 	SMU_CAP(BOARD_VOLTAGE),
 	SMU_CAP(PLDM_VERSION),
+	SMU_CAP(TEMP_METRICS),
 	SMU_CAP(ALL),
 };
 
 extern void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu);
+extern void smu_v13_0_6_set_temp_funcs(struct smu_context *smu);
 bool smu_v13_0_6_cap_supported(struct smu_context *smu, enum smu_v13_0_6_caps cap);
 int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu);
 int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table,
@@ -88,4 +90,5 @@ ssize_t smu_v13_0_12_get_xcp_metrics(struct smu_context *smu,
 				     void *smu_metrics);
 extern const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[];
 extern const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[];
+extern const struct smu_temp_funcs smu_v13_0_12_temp_funcs;
 #endif
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
index a608cdbdada4..d588f74b98de 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
@@ -65,6 +65,32 @@
 		header->structure_size = sizeof(*tmp);                     \
 	} while (0)
 
+#define smu_cmn_init_baseboard_temp_metrics(ptr, fr, cr)                        \
+	do {                                                                    \
+		typecheck(struct amdgpu_baseboard_temp_metrics_v##fr##_##cr *,  \
+			  (ptr));                                               \
+		struct amdgpu_baseboard_temp_metrics_v##fr##_##cr *tmp = (ptr); \
+		struct metrics_table_header *header =                           \
+			(struct metrics_table_header *)tmp;                     \
+		memset(header, 0xFF, sizeof(*tmp));                             \
+		header->format_revision = fr;                                   \
+		header->content_revision = cr;                                  \
+		header->structure_size = sizeof(*tmp);                          \
+	} while (0)
+
+#define smu_cmn_init_gpuboard_temp_metrics(ptr, fr, cr)                         \
+	do {                                                                    \
+		typecheck(struct amdgpu_gpuboard_temp_metrics_v##fr##_##cr *,   \
+			  (ptr));                                               \
+		struct amdgpu_gpuboard_temp_metrics_v##fr##_##cr *tmp = (ptr);  \
+		struct metrics_table_header *header =                           \
+			(struct metrics_table_header *)tmp;                     \
+		memset(header, 0xFF, sizeof(*tmp));                             \
+		header->format_revision = fr;                                   \
+		header->content_revision = cr;                                  \
+		header->structure_size = sizeof(*tmp);                          \
+	} while (0)
+
 extern const int link_speed[];
 
 /* Helper to Convert from PCIE Gen 1/2/3/4/5/6 to 0.1 GT/s speed units */

From 25e82f2e2c598164a02873f330f04fd62423da8b Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Sat, 2 Aug 2025 04:29:06 +0800
Subject: [PATCH 0317/1455] drm/amd/pm: Add temperature metrics sysfs entry

Add temperature metrics sysfs entry to expose gpuboard/baseboard
temperature metrics

v2: Removed unused function, rename functions(Lijo)

v3: Remove unnecessary initialization

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 135 +++++++++++++++++++++++++++++
 1 file changed, 135 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 4b64851fdb42..30a504122a5a 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -2073,6 +2073,134 @@ static int pp_dpm_clk_default_attr_update(struct amdgpu_device *adev, struct amd
 	return 0;
 }
 
+/**
+ * DOC: board
+ *
+ * Certain SOCs can support various board attributes reporting. This is useful
+ * for user application to monitor various board reated attributes.
+ *
+ * The amdgpu driver provides a sysfs API for reporting board attributes. Presently,
+ * only two types of attributes are reported, baseboard temperature and
+ * gpu board temperature. Both of them are reported as binary files.
+ *
+ * * .. code-block:: console
+ *
+ *      hexdump /sys/bus/pci/devices/.../board/baseboard_temp
+ *
+ *      hexdump /sys/bus/pci/devices/.../board/gpuboard_temp
+ *
+ */
+
+/**
+ * DOC: baseboard_temp
+ *
+ * The amdgpu driver provides a sysfs API for retrieving current baseboard
+ * temperature metrics data. The file baseboard_temp is used for this.
+ * Reading the file will dump all the current baseboard temperature  metrics data.
+ */
+static ssize_t amdgpu_get_baseboard_temp_metrics(struct device *dev,
+						 struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	ssize_t size;
+	int ret;
+
+	ret = amdgpu_pm_get_access_if_active(adev);
+	if (ret)
+		return ret;
+
+	size = amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_BASEBOARD, NULL);
+	if (size <= 0)
+		goto out;
+	if (size >= PAGE_SIZE) {
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_BASEBOARD, buf);
+
+out:
+	amdgpu_pm_put_access(adev);
+
+	if (ret)
+		return ret;
+
+	return size;
+}
+
+/**
+ * DOC: gpuboard_temp
+ *
+ * The amdgpu driver provides a sysfs API for retrieving current gpuboard
+ * temperature metrics data. The file gpuboard_temp is used for this.
+ * Reading the file will dump all the current gpuboard temperature  metrics data.
+ */
+static ssize_t amdgpu_get_gpuboard_temp_metrics(struct device *dev,
+						struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	ssize_t size;
+	int ret;
+
+	ret = amdgpu_pm_get_access_if_active(adev);
+	if (ret)
+		return ret;
+
+	size = amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_GPUBOARD, NULL);
+	if (size <= 0)
+		goto out;
+	if (size >= PAGE_SIZE) {
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	amdgpu_dpm_get_temp_metrics(adev, SMU_TEMP_METRIC_GPUBOARD, buf);
+
+out:
+	amdgpu_pm_put_access(adev);
+
+	if (ret)
+		return ret;
+
+	return size;
+}
+
+static DEVICE_ATTR(baseboard_temp, 0444, amdgpu_get_baseboard_temp_metrics, NULL);
+static DEVICE_ATTR(gpuboard_temp, 0444, amdgpu_get_gpuboard_temp_metrics, NULL);
+
+static struct attribute *board_attrs[] = {
+	&dev_attr_baseboard_temp.attr,
+	&dev_attr_gpuboard_temp.attr,
+	NULL
+};
+
+static umode_t amdgpu_board_attr_visible(struct kobject *kobj, struct attribute *attr, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+
+	if (attr == &dev_attr_baseboard_temp.attr) {
+		if (!amdgpu_dpm_is_temp_metrics_supported(adev, SMU_TEMP_METRIC_BASEBOARD))
+			return 0;
+	}
+
+	if (attr == &dev_attr_gpuboard_temp.attr) {
+		if (!amdgpu_dpm_is_temp_metrics_supported(adev, SMU_TEMP_METRIC_GPUBOARD))
+			return 0;
+	}
+
+	return attr->mode;
+}
+
+const struct attribute_group amdgpu_board_attr_group = {
+	.name = "board",
+	.attrs = board_attrs,
+	.is_visible = amdgpu_board_attr_visible,
+};
+
 /* pm policy attributes */
 struct amdgpu_pm_policy_attr {
 	struct device_attribute dev_attr;
@@ -4461,6 +4589,13 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 			goto err_out0;
 	}
 
+	if (amdgpu_dpm_is_temp_metrics_supported(adev, SMU_TEMP_METRIC_GPUBOARD)) {
+		ret = devm_device_add_group(adev->dev,
+					    &amdgpu_board_attr_group);
+		if (ret)
+			goto err_out0;
+	}
+
 	adev->pm.sysfs_initialized = true;
 
 	return 0;

From d9f6a0704364589d3856da454a0f78be64cc9b8c Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Sat, 2 Aug 2025 04:32:29 +0800
Subject: [PATCH 0318/1455] drm/amd/pm: Enable temperature metrics caps

Enable temperature metrics caps for smu_v13_0_12

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 5f9f74b9109d..90e66c8f2f82 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -350,6 +350,11 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu)
 		smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE));
 		smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION));
 	}
+
+	if (fw_ver >= 0x04560700) {
+		if (!amdgpu_sriov_vf(smu->adev))
+			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));
+	}
 }
 
 static void smu_v13_0_6_init_caps(struct smu_context *smu)

From b3505c2c486b36ea70889375dd0a64920f505f1f Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Mon, 4 Aug 2025 22:46:30 +0800
Subject: [PATCH 0319/1455] drm/amdgpu: Generate BP threshold exceed CPER once
 threshold exceeded

The bad pages threshold exceed CPER should be generated once threshold
exceeded, no matter the bad_page_threshold setted or not.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 4 ++++
 drivers/gpu/drm/amd/pm/amdgpu_dpm.c            | 4 ----
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index c3c908cc0859..e3e6b32e4830 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -765,6 +765,10 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 		dev_warn(adev->dev,
 			"Saved bad pages %d reaches threshold value %d\n",
 			control->ras_num_bad_pages, ras->bad_page_cnt_threshold);
+
+		if (adev->cper.enabled && amdgpu_cper_generate_bp_threshold_record(adev))
+			dev_warn(adev->dev, "fail to generate bad page threshold cper records\n");
+
 		if ((amdgpu_bad_page_threshold != -1) &&
 		    (amdgpu_bad_page_threshold != -2)) {
 			control->tbl_hdr.header = RAS_TABLE_HDR_BAD;
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 8e19054578f7..6e0d711820ea 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -764,10 +764,6 @@ int amdgpu_dpm_send_rma_reason(struct amdgpu_device *adev)
 	ret = smu_send_rma_reason(smu);
 	mutex_unlock(&adev->pm.mutex);
 
-	if (adev->cper.enabled)
-		if (amdgpu_cper_generate_bp_threshold_record(adev))
-			dev_warn(adev->dev, "fail to generate bad page threshold cper records\n");
-
 	return ret;
 }
 

From c5c62160a540db7d7e746e498fea91c35507d775 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 23 Jul 2025 10:43:00 +0530
Subject: [PATCH 0320/1455] drm/amdgpu: Log reset source during recovery

To get more context, add reset source to identify the source of gpu
recovery - job timeout, RAS, HWS hang etc.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 1055f5cda53c..fe087328f6fd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -6438,8 +6438,9 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev,
 		emergency_restart();
 	}
 
-	dev_info(adev->dev, "GPU %s begin!\n",
-		need_emergency_restart ? "jobs stop":"reset");
+	dev_info(adev->dev, "GPU %s begin!. Source:  %d\n",
+		 need_emergency_restart ? "jobs stop" : "reset",
+		 reset_context->src);
 
 	if (!amdgpu_sriov_vf(adev))
 		hive = amdgpu_get_xgmi_hive(adev);

From 1a0e57eb96c3fca338665ffd7d9b59f351e5fea7 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 5 Aug 2025 17:40:09 +0530
Subject: [PATCH 0321/1455] drm/amdgpu/vcn: Fix double-free of vcn dump buffer

The buffer is already freed as part of amdgpu_vcn_reg_dump_fini(). The
issue is introduced by below patch series.

Fixes: de55cbff5ce9 ("drm/amdgpu/vcn: Add regdump helper functions")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 1 +
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c   | 1 -
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 2 --
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 2 --
 4 files changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index b497a6714138..050a5411aae5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -1549,6 +1549,7 @@ int amdgpu_vcn_reg_dump_init(struct amdgpu_device *adev,
 static void amdgpu_vcn_reg_dump_fini(struct amdgpu_device *adev)
 {
 	kfree(adev->vcn.ip_dump);
+	adev->vcn.ip_dump = NULL;
 	adev->vcn.reg_list = NULL;
 	adev->vcn.reg_count = 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index ff2a85619f23..95173156f956 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -354,7 +354,6 @@ static int vcn_v3_0_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	kfree(adev->vcn.ip_dump);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index c6450ed65c12..019bd362edb2 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -280,8 +280,6 @@ static int vcn_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	kfree(adev->vcn.ip_dump);
-
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index f785467370d9..75c884a8f556 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -275,8 +275,6 @@ static int vcn_v4_0_5_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	kfree(adev->vcn.ip_dump);
-
 	return 0;
 }
 

From 892bac995bab7379aeca607a5dd592ecc7d98021 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 24 Jul 2025 13:05:12 +0530
Subject: [PATCH 0322/1455] drm/amdgpu: Prevent hardware access in dpc state

Don't allow hardware access while in dpc state.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Ce Sun <cesun102@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ----
 drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h  | 1 +
 2 files changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index fe087328f6fd..962184d1e96e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -6256,11 +6256,7 @@ static int amdgpu_device_asic_reset(struct amdgpu_device *adev,
 
 retry:	/* Rest of adevs pre asic reset from XGMI hive. */
 	list_for_each_entry(tmp_adev, device_list, reset_list) {
-		if (amdgpu_reset_in_dpc(adev))
-			tmp_adev->no_hw_access = true;
 		r = amdgpu_device_pre_asic_reset(tmp_adev, reset_context);
-		if (amdgpu_reset_in_dpc(adev))
-			tmp_adev->no_hw_access = false;
 		/*TODO Should we stop ?*/
 		if (r) {
 			dev_err(tmp_adev->dev, "GPU pre asic reset failed with err, %d for drm dev, %s ",
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
index 3a806953338f..2f92b3be40f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
@@ -164,6 +164,7 @@ static inline void amdgpu_reset_set_dpc_status(struct amdgpu_device *adev,
 					       bool status)
 {
 	adev->pcie_reset_ctx.occurs_dpc = status;
+	adev->no_hw_access = status;
 }
 
 static inline bool amdgpu_reset_in_dpc(struct amdgpu_device *adev)

From 589ea8a1fd6cc4d194f5472f358dbdcee954a2dc Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 4 Aug 2025 13:13:06 +0530
Subject: [PATCH 0323/1455] drm/amdgpu: Add helpers to set/get unique ids

Add a struct to store unique id information for each type. Add helper
to fetch the unique id.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        | 20 ++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 71 ++++++++++++++++++++++
 2 files changed, 91 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index ef3af170dda4..c2771e32d0bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -819,6 +819,20 @@ struct amdgpu_ip_map_info {
 					uint32_t mask);
 };
 
+enum amdgpu_uid_type {
+	AMDGPU_UID_TYPE_XCD,
+	AMDGPU_UID_TYPE_AID,
+	AMDGPU_UID_TYPE_SOC,
+	AMDGPU_UID_TYPE_MAX
+};
+
+#define AMDGPU_UID_INST_MAX 8 /* max number of instances for each UID type */
+
+struct amdgpu_uid {
+	uint64_t uid[AMDGPU_UID_TYPE_MAX][AMDGPU_UID_INST_MAX];
+	struct amdgpu_device *adev;
+};
+
 struct amd_powerplay {
 	void *pp_handle;
 	const struct amd_pm_funcs *pp_funcs;
@@ -1302,6 +1316,7 @@ struct amdgpu_device {
 	struct list_head		userq_mgr_list;
 	struct mutex                    userq_mutex;
 	bool                            userq_halt_for_enforce_isolation;
+	struct amdgpu_uid *uid_info;
 };
 
 static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev,
@@ -1785,4 +1800,9 @@ static inline int amdgpu_device_bus_status_check(struct amdgpu_device *adev)
        return 0;
 }
 
+void amdgpu_device_set_uid(struct amdgpu_uid *uid_info,
+			   enum amdgpu_uid_type type, uint8_t inst,
+			   uint64_t uid);
+uint64_t amdgpu_device_get_uid(struct amdgpu_uid *uid_info,
+			       enum amdgpu_uid_type type, uint8_t inst);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 962184d1e96e..83fba13b1d71 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2675,6 +2675,24 @@ out:
 	return err;
 }
 
+static void amdgpu_uid_init(struct amdgpu_device *adev)
+{
+	/* Initialize the UID for the device */
+	adev->uid_info = kzalloc(sizeof(struct amdgpu_uid), GFP_KERNEL);
+	if (!adev->uid_info) {
+		dev_warn(adev->dev, "Failed to allocate memory for UID\n");
+		return;
+	}
+	adev->uid_info->adev = adev;
+}
+
+static void amdgpu_uid_fini(struct amdgpu_device *adev)
+{
+	/* Free the UID memory */
+	kfree(adev->uid_info);
+	adev->uid_info = NULL;
+}
+
 /**
  * amdgpu_device_ip_early_init - run early init for hardware IPs
  *
@@ -2858,6 +2876,8 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev)
 	if (adev->gmc.xgmi.supported)
 		amdgpu_xgmi_early_init(adev);
 
+	if (amdgpu_is_multi_aid(adev))
+		amdgpu_uid_init(adev);
 	ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GFX);
 	if (ip_block->status.valid != false)
 		amdgpu_amdkfd_device_probe(adev);
@@ -3648,6 +3668,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
 	}
 
 	amdgpu_ras_fini(adev);
+	amdgpu_uid_fini(adev);
 
 	return 0;
 }
@@ -7486,3 +7507,53 @@ ssize_t amdgpu_show_reset_mask(char *buf, uint32_t supported_reset)
 	size += sysfs_emit_at(buf, size, "\n");
 	return size;
 }
+
+void amdgpu_device_set_uid(struct amdgpu_uid *uid_info,
+			   enum amdgpu_uid_type type, uint8_t inst,
+			   uint64_t uid)
+{
+	if (!uid_info)
+		return;
+
+	if (type >= AMDGPU_UID_TYPE_MAX) {
+		dev_err_once(uid_info->adev->dev, "Invalid UID type %d\n",
+			     type);
+		return;
+	}
+
+	if (inst >= AMDGPU_UID_INST_MAX) {
+		dev_err_once(uid_info->adev->dev, "Invalid UID instance %d\n",
+			     inst);
+		return;
+	}
+
+	if (uid_info->uid[type][inst] != 0) {
+		dev_warn_once(
+			uid_info->adev->dev,
+			"Overwriting existing UID %llu for type %d instance %d\n",
+			uid_info->uid[type][inst], type, inst);
+	}
+
+	uid_info->uid[type][inst] = uid;
+}
+
+u64 amdgpu_device_get_uid(struct amdgpu_uid *uid_info,
+			  enum amdgpu_uid_type type, uint8_t inst)
+{
+	if (!uid_info)
+		return 0;
+
+	if (type >= AMDGPU_UID_TYPE_MAX) {
+		dev_err_once(uid_info->adev->dev, "Invalid UID type %d\n",
+			     type);
+		return 0;
+	}
+
+	if (inst >= AMDGPU_UID_INST_MAX) {
+		dev_err_once(uid_info->adev->dev, "Invalid UID instance %d\n",
+			     inst);
+		return 0;
+	}
+
+	return uid_info->uid[type][inst];
+}
\ No newline at end of file

From 10c12aae4ed6d2561bb5fcf0bf06656c7f07da5c Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 4 Aug 2025 13:29:05 +0530
Subject: [PATCH 0324/1455] drm/amd/pm: Add unique ids for SMUv13.0.6 SOCs

Fetch and store the unique ids for AIDs/XCDs in SMUv13.0.6 SOCs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 90e66c8f2f82..733c4b3cef7d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -822,7 +822,7 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu)
 	struct PPTable_t *pptable =
 		(struct PPTable_t *)smu_table->driver_pptable;
 	int version = smu_v13_0_6_get_metrics_version(smu);
-	int ret, i, retry = 100;
+	int ret, i, retry = 100, n;
 	uint32_t table_version;
 	uint16_t max_speed;
 	uint8_t max_width;
@@ -884,6 +884,23 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu)
 		pptable->PublicSerialNumber_AID =
 			GET_METRIC_FIELD(PublicSerialNumber_AID, version)[0];
 
+		amdgpu_device_set_uid(smu->adev->uid_info, AMDGPU_UID_TYPE_SOC,
+				      0, pptable->PublicSerialNumber_AID);
+		n = ARRAY_SIZE(metrics_v0->PublicSerialNumber_AID);
+		for (i = 0; i < n; i++) {
+			amdgpu_device_set_uid(
+				smu->adev->uid_info, AMDGPU_UID_TYPE_AID, i,
+				GET_METRIC_FIELD(PublicSerialNumber_AID,
+						 version)[i]);
+		}
+		n = ARRAY_SIZE(metrics_v0->PublicSerialNumber_XCD);
+		for (i = 0; i < n; i++) {
+			amdgpu_device_set_uid(
+				smu->adev->uid_info, AMDGPU_UID_TYPE_XCD, i,
+				GET_METRIC_FIELD(PublicSerialNumber_XCD,
+						 version)[i]);
+		}
+
 		pptable->Init = true;
 		if (smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) {
 			ret = smu_v13_0_6_get_static_metrics_table(smu);

From 615471b860d628f7e7d9ffdf4b92477493e0a31c Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 6 Aug 2025 11:49:59 +0530
Subject: [PATCH 0325/1455] drm/amd/pm: Remove cache logic from SMUv13.0.12

Remove caching logic of temperature metrics from SMUv13.0.12. The
caching logic needs to be moved to a higher level.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 30 +++++++------------
 1 file changed, 11 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 920f60da9c5c..aa427fa8ddd2 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -83,7 +83,6 @@ const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[SMU_FEATURE_COUNT] =
 	SMU_13_0_12_FEA_MAP(SMU_FEATURE_PIT_BIT,			FEATURE_PIT),
 };
 
-// clang-format off
 const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] = {
 	MSG_MAP(TestMessage,			     PPSMC_MSG_TestMessage,			0),
 	MSG_MAP(GetSmuVersion,			     PPSMC_MSG_GetSmuVersion,			1),
@@ -361,31 +360,24 @@ int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
 	return 0;
 }
 
-static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu, void *metrics_table,
-						 bool bypass_cache)
+static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu,
+						 void *metrics_table)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
 	uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size;
 	struct smu_table *table = &smu_table->driver_table;
 	int ret;
 
-	if (bypass_cache || !smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time ||
-	    time_after(jiffies,
-		       smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time +
-		       msecs_to_jiffies(1))) {
-		ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetSystemMetricsTable, NULL);
-		if (ret) {
-			dev_info(smu->adev->dev,
-				 "Failed to export system metrics table!\n");
-			return ret;
-		}
-
-		amdgpu_asic_invalidate_hdp(smu->adev, NULL);
-		memcpy(smu_table->metrics_table, table->cpu_addr, table_size);
-
-		smu_table->tables[SMU_TABLE_TEMP_METRICS].metrics_time = jiffies;
+	ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetSystemMetricsTable, NULL);
+	if (ret) {
+		dev_info(smu->adev->dev,
+			 "Failed to export system metrics table!\n");
+		return ret;
 	}
 
+	amdgpu_asic_invalidate_hdp(smu->adev, NULL);
+	memcpy(smu_table->metrics_table, table->cpu_addr, table_size);
+
 	if (metrics_table)
 		memcpy(metrics_table, smu_table->metrics_table, sizeof(SystemMetricsTable_t));
 
@@ -544,7 +536,7 @@ static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
 	else if (type  == SMU_TEMP_METRIC_BASEBOARD)
 		smu_cmn_init_baseboard_temp_metrics(baseboard_temp_metrics, 1, 0);
 
-	ret = smu_v13_0_12_get_system_metrics_table(smu, metrics, false);
+	ret = smu_v13_0_12_get_system_metrics_table(smu, metrics);
 	if (ret) {
 		kfree(metrics);
 		return ret;

From 476060020f86e2cb713ef0ed956d950b64544858 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 6 Aug 2025 12:52:47 +0530
Subject: [PATCH 0326/1455] drm/amd/pm: Add cache logic for temperature metric

Add caching logic for baseboard and gpuboard temperature metrics tables.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     | 18 +++++
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h | 78 ++++++++++++++++++-
 2 files changed, 94 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index dc48a1dd8be4..0a40ab817634 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -3835,6 +3835,9 @@ int smu_set_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type,
 static ssize_t smu_sys_get_temp_metrics(void *handle, enum smu_temp_metric_type type, void *table)
 {
 	struct smu_context *smu = handle;
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = smu_table->tables;
+	enum smu_table_id table_id;
 
 	if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled)
 		return -EOPNOTSUPP;
@@ -3842,6 +3845,21 @@ static ssize_t smu_sys_get_temp_metrics(void *handle, enum smu_temp_metric_type
 	if (!smu->smu_temp.temp_funcs || !smu->smu_temp.temp_funcs->get_temp_metrics)
 		return -EOPNOTSUPP;
 
+	table_id = smu_metrics_get_temp_table_id(type);
+
+	if (table_id == SMU_TABLE_COUNT)
+		return -EINVAL;
+
+	/* If the request is to get size alone, return the cached table size */
+	if (!table && tables[table_id].cache.size)
+		return tables[table_id].cache.size;
+
+	if (smu_table_cache_is_valid(&tables[table_id])) {
+		memcpy(table, tables[table_id].cache.buffer,
+		       tables[table_id].cache.size);
+		return tables[table_id].cache.size;
+	}
+
 	return smu->smu_temp.temp_funcs->get_temp_metrics(smu, type, table);
 }
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 611b381b9147..2edd867f203e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -249,6 +249,14 @@ struct smu_user_dpm_profile {
 		tables[table_id].domain = d;		\
 	} while (0)
 
+struct smu_table_cache {
+	void *buffer;
+	size_t size;
+	/* interval in ms*/
+	uint32_t interval;
+	unsigned long last_cache_time;
+};
+
 struct smu_table {
 	uint64_t size;
 	uint32_t align;
@@ -257,7 +265,7 @@ struct smu_table {
 	void *cpu_addr;
 	struct amdgpu_bo *bo;
 	uint32_t version;
-	unsigned long  metrics_time;
+	struct smu_table_cache cache;
 };
 
 enum smu_perf_level_designation {
@@ -323,7 +331,8 @@ enum smu_table_id {
 	SMU_TABLE_ECCINFO,
 	SMU_TABLE_COMBO_PPTABLE,
 	SMU_TABLE_WIFIBAND,
-	SMU_TABLE_TEMP_METRICS,
+	SMU_TABLE_GPUBOARD_TEMP_METRICS,
+	SMU_TABLE_BASEBOARD_TEMP_METRICS,
 	SMU_TABLE_COUNT,
 };
 
@@ -1651,6 +1660,71 @@ typedef struct {
 struct smu_dpm_policy *smu_get_pm_policy(struct smu_context *smu,
 					 enum pp_pm_policy p_type);
 
+static inline enum smu_table_id
+smu_metrics_get_temp_table_id(enum smu_temp_metric_type type)
+{
+	switch (type) {
+	case SMU_TEMP_METRIC_BASEBOARD:
+		return SMU_TABLE_BASEBOARD_TEMP_METRICS;
+	case SMU_TEMP_METRIC_GPUBOARD:
+		return SMU_TABLE_GPUBOARD_TEMP_METRICS;
+	default:
+		return SMU_TABLE_COUNT;
+	}
+
+	return SMU_TABLE_COUNT;
+}
+
+static inline void smu_table_cache_update_time(struct smu_table *table,
+					       unsigned long time)
+{
+	table->cache.last_cache_time = time;
+}
+
+static inline bool smu_table_cache_is_valid(struct smu_table *table)
+{
+	if (!table->cache.buffer || !table->cache.last_cache_time ||
+	    !table->cache.interval || !table->cache.size ||
+	    time_after(jiffies,
+		       table->cache.last_cache_time +
+			       msecs_to_jiffies(table->cache.interval)))
+		return false;
+
+	return true;
+}
+
+static inline int smu_table_cache_init(struct smu_context *smu,
+				       enum smu_table_id table_id, size_t size,
+				       uint32_t cache_interval)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = smu_table->tables;
+
+	tables[table_id].cache.buffer = kzalloc(size, GFP_KERNEL);
+	if (!tables[table_id].cache.buffer)
+		return -ENOMEM;
+
+	tables[table_id].cache.last_cache_time = 0;
+	tables[table_id].cache.interval = cache_interval;
+	tables[table_id].cache.size = size;
+
+	return 0;
+}
+
+static inline void smu_table_cache_fini(struct smu_context *smu,
+					enum smu_table_id table_id)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = smu_table->tables;
+
+	if (tables[table_id].cache.buffer) {
+		kfree(tables[table_id].cache.buffer);
+		tables[table_id].cache.buffer = NULL;
+		tables[table_id].cache.last_cache_time = 0;
+		tables[table_id].cache.interval = 0;
+	}
+}
+
 #if !defined(SWSMU_CODE_LAYER_L2) && !defined(SWSMU_CODE_LAYER_L3) && !defined(SWSMU_CODE_LAYER_L4)
 int smu_get_power_limit(void *handle,
 			uint32_t *limit,

From 5bf93e1d6efd801a5337142669c951ded83ad008 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 6 Aug 2025 14:29:41 +0530
Subject: [PATCH 0327/1455] drm/amd/pm: Add caching to SMUv13.0.12 temp metric

Add table caching logic to temperature metrics tables in SMUv13.0.12

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 87 ++++++++++++++-----
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 14 ++-
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h  |  2 +
 3 files changed, 79 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index aa427fa8ddd2..fd06dcc2b81d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -140,6 +140,42 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(GetSystemMetricsTable,               PPSMC_MSG_GetSystemMetricsTable,           0),
 };
 
+int smu_v13_0_12_tables_init(struct smu_context *smu)
+{
+	struct amdgpu_baseboard_temp_metrics_v1_0 *baseboard_temp_metrics;
+	struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics;
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct smu_table *tables = smu_table->tables;
+	struct smu_table_cache *cache;
+	int ret;
+
+	ret = smu_table_cache_init(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS,
+				   sizeof(*baseboard_temp_metrics), 50);
+	if (ret)
+		return ret;
+	/* Initialize base board temperature metrics */
+	cache = &(tables[SMU_TABLE_BASEBOARD_TEMP_METRICS].cache);
+	baseboard_temp_metrics =
+		(struct amdgpu_baseboard_temp_metrics_v1_0 *) cache->buffer;
+	smu_cmn_init_baseboard_temp_metrics(baseboard_temp_metrics, 1, 0);
+	/* Initialize GPU board temperature metrics */
+	ret = smu_table_cache_init(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS,
+				   sizeof(*gpuboard_temp_metrics), 50);
+	if (ret)
+		return ret;
+	cache = &(tables[SMU_TABLE_GPUBOARD_TEMP_METRICS].cache);
+	gpuboard_temp_metrics = (struct amdgpu_gpuboard_temp_metrics_v1_0 *)cache->buffer;
+	smu_cmn_init_gpuboard_temp_metrics(gpuboard_temp_metrics, 1, 0);
+
+	return 0;
+}
+
+void smu_v13_0_12_tables_fini(struct smu_context *smu)
+{
+	smu_table_cache_fini(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS);
+	smu_table_cache_fini(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS);
+}
+
 static int smu_v13_0_12_get_enabled_mask(struct smu_context *smu,
 					 uint64_t *feature_mask)
 {
@@ -514,34 +550,40 @@ static bool smu_v13_0_12_is_temp_metrics_supported(struct smu_context *smu,
 static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
 					     enum smu_temp_metric_type type, void *table)
 {
-	struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics;
 	struct amdgpu_baseboard_temp_metrics_v1_0 *baseboard_temp_metrics;
-	SystemMetricsTable_t *metrics;
+	struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics;
+	struct smu_table_context *smu_table = &smu->smu_table;
+	SystemMetricsTable_t *metrics =
+		(SystemMetricsTable_t *)smu_table->metrics_table;
+
+	struct smu_table *data_table;
 	int ret, sensor_type;
 	u32 idx, sensors;
 	ssize_t size;
 
-	size = (type == SMU_TEMP_METRIC_GPUBOARD) ?
-		sizeof(*gpuboard_temp_metrics) : sizeof(*baseboard_temp_metrics);
-
-	if (!table)
-		goto out;
-	metrics = kzalloc(sizeof(SystemMetricsTable_t), GFP_KERNEL);
-	if (!metrics)
-		return -ENOMEM;
-	gpuboard_temp_metrics = (struct amdgpu_gpuboard_temp_metrics_v1_0 *)table;
-	baseboard_temp_metrics = (struct amdgpu_baseboard_temp_metrics_v1_0 *)table;
-	if (type  == SMU_TEMP_METRIC_GPUBOARD)
-		smu_cmn_init_gpuboard_temp_metrics(gpuboard_temp_metrics, 1, 0);
-	else if (type  == SMU_TEMP_METRIC_BASEBOARD)
-		smu_cmn_init_baseboard_temp_metrics(baseboard_temp_metrics, 1, 0);
-
-	ret = smu_v13_0_12_get_system_metrics_table(smu, metrics);
-	if (ret) {
-		kfree(metrics);
-		return ret;
+	if (type == SMU_TEMP_METRIC_BASEBOARD) {
+		/* Initialize base board temperature metrics */
+		data_table =
+			&smu->smu_table.tables[SMU_TABLE_BASEBOARD_TEMP_METRICS];
+		baseboard_temp_metrics =
+			(struct amdgpu_baseboard_temp_metrics_v1_0 *)
+				data_table->cache.buffer;
+		size = sizeof(*baseboard_temp_metrics);
+	} else {
+		data_table =
+			&smu->smu_table.tables[SMU_TABLE_GPUBOARD_TEMP_METRICS];
+		gpuboard_temp_metrics =
+			(struct amdgpu_gpuboard_temp_metrics_v1_0 *)
+				data_table->cache.buffer;
+		size = sizeof(*baseboard_temp_metrics);
 	}
 
+	ret = smu_v13_0_12_get_system_metrics_table(smu, NULL);
+	if (ret)
+		return ret;
+
+	smu_table_cache_update_time(data_table, jiffies);
+
 	if (type == SMU_TEMP_METRIC_GPUBOARD) {
 		gpuboard_temp_metrics->accumulation_counter = metrics->AccumulationCounter;
 		gpuboard_temp_metrics->label_version = metrics->LabelVersion;
@@ -586,9 +628,8 @@ static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
 		}
 	}
 
-	kfree(metrics);
+	memcpy(table, data_table->cache.buffer, size);
 
-out:
 	return size;
 }
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 733c4b3cef7d..3b4bd7a34d04 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -354,6 +354,8 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu)
 	if (fw_ver >= 0x04560700) {
 		if (!amdgpu_sriov_vf(smu->adev))
 			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));
+	} else {
+		smu_v13_0_12_tables_fini(smu);
 	}
 }
 
@@ -568,6 +570,9 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 		return -ENOMEM;
 	}
 
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12))
+		return smu_v13_0_12_tables_init(smu);
+
 	return 0;
 }
 
@@ -696,6 +701,13 @@ static int smu_v13_0_6_init_smc_tables(struct smu_context *smu)
 	return ret;
 }
 
+static int smu_v13_0_6_fini_smc_tables(struct smu_context *smu)
+{
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12))
+		smu_v13_0_12_tables_fini(smu);
+	return smu_v13_0_fini_smc_tables(smu);
+}
+
 static int smu_v13_0_6_get_allowed_feature_mask(struct smu_context *smu,
 						uint32_t *feature_mask,
 						uint32_t num)
@@ -3833,7 +3845,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.init_microcode = smu_v13_0_6_init_microcode,
 	.fini_microcode = smu_v13_0_fini_microcode,
 	.init_smc_tables = smu_v13_0_6_init_smc_tables,
-	.fini_smc_tables = smu_v13_0_fini_smc_tables,
+	.fini_smc_tables = smu_v13_0_6_fini_smc_tables,
 	.init_power = smu_v13_0_init_power,
 	.fini_power = smu_v13_0_fini_power,
 	.check_fw_status = smu_v13_0_6_check_fw_status,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index ece04ad724fb..f39dbfdd7a2f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -88,6 +88,8 @@ ssize_t smu_v13_0_12_get_gpu_metrics(struct smu_context *smu, void **table, void
 ssize_t smu_v13_0_12_get_xcp_metrics(struct smu_context *smu,
 				     struct amdgpu_xcp *xcp, void *table,
 				     void *smu_metrics);
+int smu_v13_0_12_tables_init(struct smu_context *smu);
+void smu_v13_0_12_tables_fini(struct smu_context *smu);
 extern const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[];
 extern const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[];
 extern const struct smu_temp_funcs smu_v13_0_12_temp_funcs;

From 9d693917949122e2018d7d1377327557f708101b Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 6 Aug 2025 18:15:22 +0530
Subject: [PATCH 0328/1455] drm/amd/display: Add NULL check for stream before
 dereference in 'dm_vupdate_high_irq'

Add a NULL check for acrtc->dm_irq_params.stream before
accessing its members.

Fixes below:
drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:623
dm_vupdate_high_irq() warn: variable dereferenced before check
'acrtc->dm_irq_params.stream' (see line 615)

614	if (vrr_active) {
615		bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled;
				 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
616		bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled;
			      ^^^^^^^^^^^^^^^^^^^^^^^^^^^ New dereferences

617		bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state
618			== VRR_STATE_ACTIVE_VARIABLE;
619
620		amdgpu_dm_crtc_handle_vblank(acrtc);
621
622		/* BTR processing for pre-DCE12 ASICs */
623		if (acrtc->dm_irq_params.stream &&
		    ^^^^^^^^^^^^^^^^^^^^^^^^^^^ But the existing code assumed it could be NULL. Someone is wrong.

624		    adev->family < AMDGPU_FAMILY_AI) {
625			spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);

Fixes: 6d31602a9f57 ("drm/amd/display: more liberal vmin/vmax update for freesync")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Cc: Alex Hung <alex.hung@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Roman Li <roman.li@amd.com>
Cc: ChiaHsuan Chung <chiahsuan.chung@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Ray Wu <ray.wu@amd.com>
Cc: Daniel Wheeler <daniel.wheeler@amd.com>
Cc: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 31ea57edeb45..c0d31c26850d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -611,7 +611,7 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 		 * page-flip completion events that have been queued to us
 		 * if a pageflip happened inside front-porch.
 		 */
-		if (vrr_active) {
+		if (vrr_active && acrtc->dm_irq_params.stream) {
 			bool replay_en = acrtc->dm_irq_params.stream->link->replay_settings.replay_feature_enabled;
 			bool psr_en = acrtc->dm_irq_params.stream->link->psr_settings.psr_feature_enabled;
 			bool fs_active_var_en = acrtc->dm_irq_params.freesync_config.state
@@ -620,8 +620,7 @@ static void dm_vupdate_high_irq(void *interrupt_params)
 			amdgpu_dm_crtc_handle_vblank(acrtc);
 
 			/* BTR processing for pre-DCE12 ASICs */
-			if (acrtc->dm_irq_params.stream &&
-			    adev->family < AMDGPU_FAMILY_AI) {
+			if (adev->family < AMDGPU_FAMILY_AI) {
 				spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
 				mod_freesync_handle_v_update(
 				    adev->dm.freesync_module,

From 62eedd150fa11aefc2d377fc746633fdb1baeb55 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 30 Jul 2025 11:16:05 -0400
Subject: [PATCH 0329/1455] drm/amdgpu/discovery: fix fw based ip discovery

We only need the fw based discovery table for sysfs.  No
need to parse it.  Additionally parsing some of the board
specific tables may result in incorrect data on some boards.
just load the binary and don't parse it on those boards.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4441
Fixes: 80a0e8282933 ("drm/amdgpu/discovery: optionally use fw based ip discovery")
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c    |  5 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 72 ++++++++++---------
 2 files changed, 41 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 83fba13b1d71..828d981611b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2570,9 +2570,6 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev)
 
 	adev->firmware.gpu_info_fw = NULL;
 
-	if (adev->mman.discovery_bin)
-		return 0;
-
 	switch (adev->asic_type) {
 	default:
 		return 0;
@@ -2594,6 +2591,8 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev)
 		chip_name = "arcturus";
 		break;
 	case CHIP_NAVI12:
+		if (adev->mman.discovery_bin)
+			return 0;
 		chip_name = "navi12";
 		break;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 047a63496441..efe0058b48ca 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -2555,40 +2555,11 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 
 	switch (adev->asic_type) {
 	case CHIP_VEGA10:
-	case CHIP_VEGA12:
-	case CHIP_RAVEN:
-	case CHIP_VEGA20:
-	case CHIP_ARCTURUS:
-	case CHIP_ALDEBARAN:
-		/* this is not fatal.  We have a fallback below
-		 * if the new firmwares are not present. some of
-		 * this will be overridden below to keep things
-		 * consistent with the current behavior.
+		/* This is not fatal.  We only need the discovery
+		 * binary for sysfs.  We don't need it for a
+		 * functional system.
 		 */
-		r = amdgpu_discovery_reg_base_init(adev);
-		if (!r) {
-			amdgpu_discovery_harvest_ip(adev);
-			amdgpu_discovery_get_gfx_info(adev);
-			amdgpu_discovery_get_mall_info(adev);
-			amdgpu_discovery_get_vcn_info(adev);
-		}
-		break;
-	default:
-		r = amdgpu_discovery_reg_base_init(adev);
-		if (r) {
-			drm_err(&adev->ddev, "discovery failed: %d\n", r);
-			return r;
-		}
-
-		amdgpu_discovery_harvest_ip(adev);
-		amdgpu_discovery_get_gfx_info(adev);
-		amdgpu_discovery_get_mall_info(adev);
-		amdgpu_discovery_get_vcn_info(adev);
-		break;
-	}
-
-	switch (adev->asic_type) {
-	case CHIP_VEGA10:
+		amdgpu_discovery_init(adev);
 		vega10_reg_base_init(adev);
 		adev->sdma.num_instances = 2;
 		adev->gmc.num_umc = 4;
@@ -2611,6 +2582,11 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->ip_versions[DCI_HWIP][0] = IP_VERSION(12, 0, 0);
 		break;
 	case CHIP_VEGA12:
+		/* This is not fatal.  We only need the discovery
+		 * binary for sysfs.  We don't need it for a
+		 * functional system.
+		 */
+		amdgpu_discovery_init(adev);
 		vega10_reg_base_init(adev);
 		adev->sdma.num_instances = 2;
 		adev->gmc.num_umc = 4;
@@ -2633,6 +2609,11 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->ip_versions[DCI_HWIP][0] = IP_VERSION(12, 0, 1);
 		break;
 	case CHIP_RAVEN:
+		/* This is not fatal.  We only need the discovery
+		 * binary for sysfs.  We don't need it for a
+		 * functional system.
+		 */
+		amdgpu_discovery_init(adev);
 		vega10_reg_base_init(adev);
 		adev->sdma.num_instances = 1;
 		adev->vcn.num_vcn_inst = 1;
@@ -2674,6 +2655,11 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		}
 		break;
 	case CHIP_VEGA20:
+		/* This is not fatal.  We only need the discovery
+		 * binary for sysfs.  We don't need it for a
+		 * functional system.
+		 */
+		amdgpu_discovery_init(adev);
 		vega20_reg_base_init(adev);
 		adev->sdma.num_instances = 2;
 		adev->gmc.num_umc = 8;
@@ -2697,6 +2683,11 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->ip_versions[DCI_HWIP][0] = IP_VERSION(12, 1, 0);
 		break;
 	case CHIP_ARCTURUS:
+		/* This is not fatal.  We only need the discovery
+		 * binary for sysfs.  We don't need it for a
+		 * functional system.
+		 */
+		amdgpu_discovery_init(adev);
 		arct_reg_base_init(adev);
 		adev->sdma.num_instances = 8;
 		adev->vcn.num_vcn_inst = 2;
@@ -2725,6 +2716,11 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->ip_versions[UVD_HWIP][1] = IP_VERSION(2, 5, 0);
 		break;
 	case CHIP_ALDEBARAN:
+		/* This is not fatal.  We only need the discovery
+		 * binary for sysfs.  We don't need it for a
+		 * functional system.
+		 */
+		amdgpu_discovery_init(adev);
 		aldebaran_reg_base_init(adev);
 		adev->sdma.num_instances = 5;
 		adev->vcn.num_vcn_inst = 2;
@@ -2751,6 +2747,16 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->ip_versions[XGMI_HWIP][0] = IP_VERSION(6, 1, 0);
 		break;
 	default:
+		r = amdgpu_discovery_reg_base_init(adev);
+		if (r) {
+			drm_err(&adev->ddev, "discovery failed: %d\n", r);
+			return r;
+		}
+
+		amdgpu_discovery_harvest_ip(adev);
+		amdgpu_discovery_get_gfx_info(adev);
+		amdgpu_discovery_get_mall_info(adev);
+		amdgpu_discovery_get_vcn_info(adev);
 		break;
 	}
 

From aae94897b6661a2a4b1de2d328090fc388b3e0af Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 4 Aug 2025 11:40:20 -0400
Subject: [PATCH 0330/1455] drm/amdgpu: add missing vram lost check for LEGACY
 RESET

Legacy resets reset the memory controllers so VRAM contents
may be unreliable after reset.

Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 828d981611b7..5eed5c500190 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -3290,6 +3290,7 @@ static bool amdgpu_device_check_vram_lost(struct amdgpu_device *adev)
 	 * always assumed to be lost.
 	 */
 	switch (amdgpu_asic_reset_method(adev)) {
+	case AMD_RESET_METHOD_LEGACY:
 	case AMD_RESET_METHOD_LINK:
 	case AMD_RESET_METHOD_BACO:
 	case AMD_RESET_METHOD_MODE1:

From 6fa821685464e786f43a66d18756cc909a9cb961 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 4 Aug 2025 13:36:52 +0530
Subject: [PATCH 0331/1455] drm/amd/pm: Add unique ids for SMUv13.0.12 SOCs

Fetch and store the unique ids for AIDs/XCDs in SMUv13.0.12 SOCs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c  | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index fd06dcc2b81d..ea2682f9d579 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -257,7 +257,7 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu)
 	struct PPTable_t *pptable =
 		(struct PPTable_t *)smu_table->driver_pptable;
 	uint32_t table_version;
-	int ret, i;
+	int ret, i, n;
 
 	if (!pptable->Init) {
 		ret = smu_v13_0_6_get_static_metrics_table(smu);
@@ -296,6 +296,22 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu)
 		/* use AID0 serial number by default */
 		pptable->PublicSerialNumber_AID =
 			static_metrics->PublicSerialNumber_AID[0];
+
+		amdgpu_device_set_uid(smu->adev->uid_info, AMDGPU_UID_TYPE_SOC,
+				      0, pptable->PublicSerialNumber_AID);
+		n = ARRAY_SIZE(static_metrics->PublicSerialNumber_AID);
+		for (i = 0; i < n; i++) {
+			amdgpu_device_set_uid(
+				smu->adev->uid_info, AMDGPU_UID_TYPE_AID, i,
+				static_metrics->PublicSerialNumber_AID[i]);
+		}
+		n = ARRAY_SIZE(static_metrics->PublicSerialNumber_XCD);
+		for (i = 0; i < n; i++) {
+			amdgpu_device_set_uid(
+				smu->adev->uid_info, AMDGPU_UID_TYPE_XCD, i,
+				static_metrics->PublicSerialNumber_XCD[i]);
+		}
+
 		ret = smu_v13_0_12_fru_get_product_info(smu, static_metrics);
 		if (ret)
 			return ret;

From 0b4d79dafae6a5b626037b4a5dafaf9f09cc1945 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 4 Aug 2025 13:53:21 +0530
Subject: [PATCH 0332/1455] drm/amdgpu: Assign unique id to compute partition

Assign unique id to compute partition. This is the unique id of the
first XCD instance belonging to the partition.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c | 20 ++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h |  1 +
 2 files changed, 21 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
index 699acc1b46b5..1083db8cea2e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c
@@ -120,6 +120,25 @@ static void __amdgpu_xcp_add_block(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id,
 	xcp->valid = true;
 }
 
+static void __amdgpu_xcp_set_unique_id(struct amdgpu_xcp_mgr *xcp_mgr,
+				       int xcp_id)
+{
+	struct amdgpu_xcp *xcp = &xcp_mgr->xcp[xcp_id];
+	struct amdgpu_device *adev = xcp_mgr->adev;
+	uint32_t inst_mask;
+	uint64_t uid;
+	int i;
+
+	if (!amdgpu_xcp_get_inst_details(xcp, AMDGPU_XCP_GFX, &inst_mask) &&
+	    inst_mask) {
+		i = GET_INST(GC, (ffs(inst_mask) - 1));
+		uid = amdgpu_device_get_uid(xcp_mgr->adev->uid_info,
+					    AMDGPU_UID_TYPE_XCD, i);
+		if (uid)
+			xcp->unique_id = uid;
+	}
+}
+
 int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode)
 {
 	struct amdgpu_device *adev = xcp_mgr->adev;
@@ -158,6 +177,7 @@ int amdgpu_xcp_init(struct amdgpu_xcp_mgr *xcp_mgr, int num_xcps, int mode)
 			else
 				xcp_mgr->xcp[i].mem_id = mem_id;
 		}
+		__amdgpu_xcp_set_unique_id(xcp_mgr, i);
 	}
 
 	xcp_mgr->num_xcps = num_xcps;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
index 70a0f8400b57..1928d9e224fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h
@@ -112,6 +112,7 @@ struct amdgpu_xcp {
 	struct amdgpu_sched	gpu_sched[AMDGPU_HW_IP_NUM][AMDGPU_RING_PRIO_MAX];
 	struct amdgpu_xcp_mgr *xcp_mgr;
 	struct kobject kobj;
+	uint64_t unique_id;
 };
 
 struct amdgpu_xcp_mgr {

From 0e7581eda8c76d1ca4cf519631a4d4eb9f82b94c Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Tue, 5 Aug 2025 21:28:25 +0530
Subject: [PATCH 0333/1455] drm/amdgpu/jpeg: Hold pg_lock before jpeg poweroff

Acquire jpeg_pg_lock before changes to jpeg power state
and release it after power off from idle work handler.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index 82d58ac7afb0..5d5e9ee83a5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -121,10 +121,12 @@ static void amdgpu_jpeg_idle_work_handler(struct work_struct *work)
 			fences += amdgpu_fence_count_emitted(&adev->jpeg.inst[i].ring_dec[j]);
 	}
 
-	if (!fences && !atomic_read(&adev->jpeg.total_submission_cnt))
+	if (!fences && !atomic_read(&adev->jpeg.total_submission_cnt)) {
+		mutex_lock(&adev->jpeg.jpeg_pg_lock);
 		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_JPEG,
 						       AMD_PG_STATE_GATE);
-	else
+		mutex_unlock(&adev->jpeg.jpeg_pg_lock);
+	} else
 		schedule_delayed_work(&adev->jpeg.idle_work, JPEG_IDLE_TIMEOUT);
 }
 

From 111821e4b5a3105c42c7c99f4abd4d8af9f64248 Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Tue, 5 Aug 2025 21:35:10 +0530
Subject: [PATCH 0334/1455] drm/amdgpu/vcn: Hold pg_lock before vcn power off

Acquire vcn_pg_lock before changes to vcn power state
and release it after power off in idle work handler.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 050a5411aae5..9a76e11d1c18 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -443,7 +443,9 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 	fences += fence[i];
 
 	if (!fences && !atomic_read(&vcn_inst->total_submission_cnt)) {
+		mutex_lock(&vcn_inst->vcn_pg_lock);
 		vcn_inst->set_pg_state(vcn_inst, AMD_PG_STATE_GATE);
+		mutex_unlock(&vcn_inst->vcn_pg_lock);
 		mutex_lock(&adev->vcn.workload_profile_mutex);
 		if (adev->vcn.workload_profile_active) {
 			r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,

From ad69d62588cd6bf8cddaff5e3e2eb1b8dd876d35 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Tue, 5 Aug 2025 11:18:50 +0200
Subject: [PATCH 0335/1455] drm/xe/pf: Program LMTT directory pointer on all
 GTs within a tile
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Previously, the LMTT directory pointer was only programmed for primary GT
within a tile. However, to ensure correct Local Memory access by VFs,
the LMTT configuration must be programmed on all GTs within the tile.
Lets program the LMTT directory pointer on every GT of the tile
to guarantee proper LMEM access across all GTs on VFs.

HSD: 18042797646
Bspec: 67468
Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250805091850.1508240-1-piotr.piorkowski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_lmtt.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c
index a2000307d5bf..a78c9d474a6e 100644
--- a/drivers/gpu/drm/xe/xe_lmtt.c
+++ b/drivers/gpu/drm/xe/xe_lmtt.c
@@ -195,14 +195,17 @@ static void lmtt_setup_dir_ptr(struct xe_lmtt *lmtt)
 	struct xe_tile *tile = lmtt_to_tile(lmtt);
 	struct xe_device *xe = tile_to_xe(tile);
 	dma_addr_t offset = xe_bo_main_addr(lmtt->pd->bo, XE_PAGE_SIZE);
+	struct xe_gt *gt;
+	u8 id;
 
 	lmtt_debug(lmtt, "DIR offset %pad\n", &offset);
 	lmtt_assert(lmtt, xe_bo_is_vram(lmtt->pd->bo));
 	lmtt_assert(lmtt, IS_ALIGNED(offset, SZ_64K));
 
-	xe_mmio_write32(&tile->mmio,
-			GRAPHICS_VER(xe) >= 20 ? XE2_LMEM_CFG : LMEM_CFG,
-			LMEM_EN | REG_FIELD_PREP(LMTT_DIR_PTR, offset / SZ_64K));
+	for_each_gt_on_tile(gt, tile, id)
+		xe_mmio_write32(&gt->mmio,
+				GRAPHICS_VER(xe) >= 20 ? XE2_LMEM_CFG : LMEM_CFG,
+				LMEM_EN | REG_FIELD_PREP(LMTT_DIR_PTR, offset / SZ_64K));
 }
 
 /**

From 39896c529454763fb7d40ecb2657ad9f65c462ee Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Jul 2025 12:57:37 +0300
Subject: [PATCH 0336/1455] drm/i915/display: use drm->debugfs_root for
 creating debugfs files

Since commit 0b30d57acafc ("drm/debugfs: rework debugfs directory
creation v5") we should be using drm->debugfs_root instead of
minor->debugfs_root for creating debugfs files.

As a rule of thumb, use a local variable when there are two or more
uses, otherwise just have the single reference inline.

Drop drm/drm_file.h include where possible.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/e8268546ec2a2941a3dc43c2fdc60f678dc03fce.1753782998.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.c                | 4 +---
 drivers/gpu/drm/i915/display/intel_cdclk.c               | 4 +---
 drivers/gpu/drm/i915/display/intel_display_debugfs.c     | 6 +++---
 .../gpu/drm/i915/display/intel_display_debugfs_params.c  | 7 +++----
 drivers/gpu/drm/i915/display/intel_dmc.c                 | 4 +---
 drivers/gpu/drm/i915/display/intel_dp_test.c             | 4 +---
 drivers/gpu/drm/i915/display/intel_fbc.c                 | 3 +--
 drivers/gpu/drm/i915/display/intel_hotplug.c             | 8 ++++----
 drivers/gpu/drm/i915/display/intel_opregion.c            | 4 +---
 drivers/gpu/drm/i915/display/intel_psr.c                 | 6 +++---
 drivers/gpu/drm/i915/display/intel_wm.c                  | 9 ++++-----
 drivers/gpu/drm/i915/display/skl_watermark.c             | 7 +++----
 12 files changed, 26 insertions(+), 40 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 276ac29e4daf..8de332c627f8 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -3744,8 +3744,6 @@ DEFINE_SHOW_ATTRIBUTE(intel_bios_vbt);
 
 void intel_bios_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
-
-	debugfs_create_file("i915_vbt", 0444, minor->debugfs_root,
+	debugfs_create_file("i915_vbt", 0444, display->drm->debugfs_root,
 			    display, &intel_bios_vbt_fops);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index 59997dc95984..f7f76de0f944 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -3627,9 +3627,7 @@ DEFINE_SHOW_ATTRIBUTE(i915_cdclk_info);
 
 void intel_cdclk_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
-
-	debugfs_create_file("i915_cdclk_info", 0444, minor->debugfs_root,
+	debugfs_create_file("i915_cdclk_info", 0444, display->drm->debugfs_root,
 			    display, &i915_cdclk_info_fops);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
index ce3f9810c42d..10dddec3796f 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
@@ -820,14 +820,14 @@ static const struct drm_info_list intel_display_debugfs_list[] = {
 
 void intel_display_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
+	struct dentry *debugfs_root = display->drm->debugfs_root;
 
-	debugfs_create_file("i915_fifo_underrun_reset", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_fifo_underrun_reset", 0644, debugfs_root,
 			    display, &i915_fifo_underrun_reset_ops);
 
 	drm_debugfs_create_files(intel_display_debugfs_list,
 				 ARRAY_SIZE(intel_display_debugfs_list),
-				 minor->debugfs_root, minor);
+				 debugfs_root, display->drm->primary);
 
 	intel_bios_debugfs_register(display);
 	intel_cdclk_debugfs_register(display);
diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c b/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c
index 88914a1f3f62..de62b774272d 100644
--- a/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c
+++ b/drivers/gpu/drm/i915/display/intel_display_debugfs_params.c
@@ -7,7 +7,6 @@
 #include <linux/kernel.h>
 
 #include <drm/drm_drv.h>
-#include <drm/drm_file.h>
 
 #include "intel_display_core.h"
 #include "intel_display_debugfs_params.h"
@@ -154,14 +153,14 @@ intel_display_debugfs_create_uint(const char *name, umode_t mode,
 /* add a subdirectory with files for each intel display param */
 void intel_display_debugfs_params(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
+	struct dentry *debugfs_root = display->drm->debugfs_root;
 	struct dentry *dir;
 	char dirname[16];
 
 	snprintf(dirname, sizeof(dirname), "%s_params", display->drm->driver->name);
-	dir = debugfs_lookup(dirname, minor->debugfs_root);
+	dir = debugfs_lookup(dirname, debugfs_root);
 	if (!dir)
-		dir = debugfs_create_dir(dirname, minor->debugfs_root);
+		dir = debugfs_create_dir(dirname, debugfs_root);
 	if (IS_ERR(dir))
 		return;
 
diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c
index 744f51c0eab8..77a0199f9ea5 100644
--- a/drivers/gpu/drm/i915/display/intel_dmc.c
+++ b/drivers/gpu/drm/i915/display/intel_dmc.c
@@ -1603,9 +1603,7 @@ DEFINE_SHOW_ATTRIBUTE(intel_dmc_debugfs_status);
 
 void intel_dmc_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
-
-	debugfs_create_file("i915_dmc_info", 0444, minor->debugfs_root,
+	debugfs_create_file("i915_dmc_info", 0444, display->drm->debugfs_root,
 			    display, &intel_dmc_debugfs_status_fops);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_dp_test.c b/drivers/gpu/drm/i915/display/intel_dp_test.c
index 6ed5012c5fac..5cfa1dd411da 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_test.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_test.c
@@ -6,7 +6,6 @@
 #include <drm/display/drm_dp.h>
 #include <drm/display/drm_dp_helper.h>
 #include <drm/drm_edid.h>
-#include <drm/drm_file.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
 
@@ -753,13 +752,12 @@ static const struct {
 
 void intel_dp_test_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(intel_display_debugfs_files); i++) {
 		debugfs_create_file(intel_display_debugfs_files[i].name,
 				    0644,
-				    minor->debugfs_root,
+				    display->drm->debugfs_root,
 				    display,
 				    intel_display_debugfs_files[i].fops);
 	}
diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index f82d392f1e22..d4c5deff9cbe 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -2240,10 +2240,9 @@ void intel_fbc_crtc_debugfs_add(struct intel_crtc *crtc)
 /* FIXME: remove this once igt is on board with per-crtc stuff */
 void intel_fbc_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
 	struct intel_fbc *fbc;
 
 	fbc = display->fbc[INTEL_FBC_A];
 	if (fbc)
-		intel_fbc_debugfs_add(fbc, minor->debugfs_root);
+		intel_fbc_debugfs_add(fbc, display->drm->debugfs_root);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_hotplug.c b/drivers/gpu/drm/i915/display/intel_hotplug.c
index bd8851fd2ee9..4451a792600a 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug.c
@@ -1332,12 +1332,12 @@ static const struct file_operations i915_hpd_short_storm_ctl_fops = {
 
 void intel_hpd_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
+	struct dentry *debugfs_root = display->drm->debugfs_root;
 
-	debugfs_create_file("i915_hpd_storm_ctl", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_hpd_storm_ctl", 0644, debugfs_root,
 			    display, &i915_hpd_storm_ctl_fops);
-	debugfs_create_file("i915_hpd_short_storm_ctl", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_hpd_short_storm_ctl", 0644, debugfs_root,
 			    display, &i915_hpd_short_storm_ctl_fops);
-	debugfs_create_bool("i915_ignore_long_hpd", 0644, minor->debugfs_root,
+	debugfs_create_bool("i915_ignore_long_hpd", 0644, debugfs_root,
 			    &display->hotplug.ignore_long_hpd);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c
index 81efdb17fc0c..380314a3b4d9 100644
--- a/drivers/gpu/drm/i915/display/intel_opregion.c
+++ b/drivers/gpu/drm/i915/display/intel_opregion.c
@@ -1299,8 +1299,6 @@ DEFINE_SHOW_ATTRIBUTE(intel_opregion);
 
 void intel_opregion_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
-
-	debugfs_create_file("i915_opregion", 0444, minor->debugfs_root,
+	debugfs_create_file("i915_opregion", 0444, display->drm->debugfs_root,
 			    display, &intel_opregion_fops);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 6bd3454bb00e..1d70ae31b6dd 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -4132,12 +4132,12 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_edp_psr_debug_fops,
 
 void intel_psr_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
+	struct dentry *debugfs_root = display->drm->debugfs_root;
 
-	debugfs_create_file("i915_edp_psr_debug", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_edp_psr_debug", 0644, debugfs_root,
 			    display, &i915_edp_psr_debug_fops);
 
-	debugfs_create_file("i915_edp_psr_status", 0444, minor->debugfs_root,
+	debugfs_create_file("i915_edp_psr_status", 0444, debugfs_root,
 			    display, &i915_edp_psr_status_fops);
 }
 
diff --git a/drivers/gpu/drm/i915/display/intel_wm.c b/drivers/gpu/drm/i915/display/intel_wm.c
index bba82e888db2..f887a664fe22 100644
--- a/drivers/gpu/drm/i915/display/intel_wm.c
+++ b/drivers/gpu/drm/i915/display/intel_wm.c
@@ -5,7 +5,6 @@
 
 #include <linux/debugfs.h>
 
-#include <drm/drm_file.h>
 #include <drm/drm_print.h>
 
 #include "i9xx_wm.h"
@@ -390,15 +389,15 @@ static const struct file_operations i915_cur_wm_latency_fops = {
 
 void intel_wm_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
+	struct dentry *debugfs_root = display->drm->debugfs_root;
 
-	debugfs_create_file("i915_pri_wm_latency", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_pri_wm_latency", 0644, debugfs_root,
 			    display, &i915_pri_wm_latency_fops);
 
-	debugfs_create_file("i915_spr_wm_latency", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_spr_wm_latency", 0644, debugfs_root,
 			    display, &i915_spr_wm_latency_fops);
 
-	debugfs_create_file("i915_cur_wm_latency", 0644, minor->debugfs_root,
+	debugfs_create_file("i915_cur_wm_latency", 0644, debugfs_root,
 			    display, &i915_cur_wm_latency_fops);
 
 	skl_watermark_debugfs_register(display);
diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index 222c069fdadb..def5150231a4 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -6,7 +6,6 @@
 #include <linux/debugfs.h>
 
 #include <drm/drm_blend.h>
-#include <drm/drm_file.h>
 #include <drm/drm_print.h>
 
 #include "soc/intel_dram.h"
@@ -4033,14 +4032,14 @@ DEFINE_SHOW_ATTRIBUTE(intel_sagv_status);
 
 void skl_watermark_debugfs_register(struct intel_display *display)
 {
-	struct drm_minor *minor = display->drm->primary;
+	struct dentry *debugfs_root = display->drm->debugfs_root;
 
 	if (HAS_IPC(display))
-		debugfs_create_file("i915_ipc_status", 0644, minor->debugfs_root,
+		debugfs_create_file("i915_ipc_status", 0644, debugfs_root,
 				    display, &skl_watermark_ipc_status_fops);
 
 	if (HAS_SAGV(display))
-		debugfs_create_file("i915_sagv_status", 0444, minor->debugfs_root,
+		debugfs_create_file("i915_sagv_status", 0444, debugfs_root,
 				    display, &intel_sagv_status_fops);
 }
 

From c5b9077eca39c3615f5e73bbb8493eaef9ceef3a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Jul 2025 12:57:38 +0300
Subject: [PATCH 0337/1455] drm/i915/gvt: use drm->debugfs_root for creating
 debugfs files

Since commit 0b30d57acafc ("drm/debugfs: rework debugfs directory
creation v5") we should be using drm->debugfs_root instead of
minor->debugfs_root for creating debugfs files.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/482a3516e00b2885cd62f872ad09f51a9d8176b4.1753782998.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/gvt/debugfs.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/gvt/debugfs.c b/drivers/gpu/drm/i915/gvt/debugfs.c
index 673534f061ef..415422b5943c 100644
--- a/drivers/gpu/drm/i915/gvt/debugfs.c
+++ b/drivers/gpu/drm/i915/gvt/debugfs.c
@@ -194,9 +194,9 @@ void intel_gvt_debugfs_add_vgpu(struct intel_vgpu *vgpu)
 void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu)
 {
 	struct intel_gvt *gvt = vgpu->gvt;
-	struct drm_minor *minor = gvt->gt->i915->drm.primary;
+	struct dentry *debugfs_root = gvt->gt->i915->drm.debugfs_root;
 
-	if (minor->debugfs_root && gvt->debugfs_root) {
+	if (debugfs_root && gvt->debugfs_root) {
 		debugfs_remove_recursive(vgpu->debugfs);
 		vgpu->debugfs = NULL;
 	}
@@ -208,9 +208,9 @@ void intel_gvt_debugfs_remove_vgpu(struct intel_vgpu *vgpu)
  */
 void intel_gvt_debugfs_init(struct intel_gvt *gvt)
 {
-	struct drm_minor *minor = gvt->gt->i915->drm.primary;
+	struct dentry *debugfs_root = gvt->gt->i915->drm.debugfs_root;
 
-	gvt->debugfs_root = debugfs_create_dir("gvt", minor->debugfs_root);
+	gvt->debugfs_root = debugfs_create_dir("gvt", debugfs_root);
 
 	debugfs_create_ulong("num_tracked_mmio", 0444, gvt->debugfs_root,
 			     &gvt->mmio.num_tracked_mmio);
@@ -222,9 +222,9 @@ void intel_gvt_debugfs_init(struct intel_gvt *gvt)
  */
 void intel_gvt_debugfs_clean(struct intel_gvt *gvt)
 {
-	struct drm_minor *minor = gvt->gt->i915->drm.primary;
+	struct dentry *debugfs_root = gvt->gt->i915->drm.debugfs_root;
 
-	if (minor->debugfs_root) {
+	if (debugfs_root) {
 		debugfs_remove_recursive(gvt->debugfs_root);
 		gvt->debugfs_root = NULL;
 	}

From f9b5bf76ea719a115502f9c1b20ab534ccd8194b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Jul 2025 12:57:39 +0300
Subject: [PATCH 0338/1455] drm/i915: use drm->debugfs_root for creating
 debugfs files

Since commit 0b30d57acafc ("drm/debugfs: rework debugfs directory
creation v5") we should be using drm->debugfs_root instead of
minor->debugfs_root for creating debugfs files.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/ba8a2a7ec10e54b4d0a96926ef20c96e268c0b94.1753782998.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_gt_debugfs.c   |  5 +++--
 drivers/gpu/drm/i915/i915_debugfs.c          | 20 +++++++++-----------
 drivers/gpu/drm/i915/i915_debugfs_params.c   |  4 ++--
 drivers/gpu/drm/i915/i915_gpu_error.c        |  6 +++---
 drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c |  8 ++++----
 5 files changed, 21 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c b/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c
index 4dc23b8d3aa2..dcd40b30a96b 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_debugfs.c
@@ -82,14 +82,15 @@ static void gt_debugfs_register(struct intel_gt *gt, struct dentry *root)
 
 void intel_gt_debugfs_register(struct intel_gt *gt)
 {
+	struct dentry *debugfs_root = gt->i915->drm.debugfs_root;
 	struct dentry *root;
 	char gtname[4];
 
-	if (!gt->i915->drm.primary->debugfs_root)
+	if (!debugfs_root)
 		return;
 
 	snprintf(gtname, sizeof(gtname), "gt%u", gt->info.id);
-	root = debugfs_create_dir(gtname, gt->i915->drm.primary->debugfs_root);
+	root = debugfs_create_dir(gtname, debugfs_root);
 	if (IS_ERR(root))
 		return;
 
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 967c0501e91e..23fa098c4479 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -720,26 +720,24 @@ static const struct i915_debugfs_files {
 	{"i915_gem_drop_caches", &i915_drop_caches_fops},
 };
 
-void i915_debugfs_register(struct drm_i915_private *dev_priv)
+void i915_debugfs_register(struct drm_i915_private *i915)
 {
-	struct drm_minor *minor = dev_priv->drm.primary;
+	struct dentry *debugfs_root = i915->drm.debugfs_root;
 	int i;
 
-	i915_debugfs_params(dev_priv);
+	i915_debugfs_params(i915);
 
-	debugfs_create_file("i915_forcewake_user", S_IRUSR, minor->debugfs_root,
-			    to_i915(minor->dev), &i915_forcewake_fops);
+	debugfs_create_file("i915_forcewake_user", S_IRUSR, debugfs_root,
+			    i915, &i915_forcewake_fops);
 	for (i = 0; i < ARRAY_SIZE(i915_debugfs_files); i++) {
-		debugfs_create_file(i915_debugfs_files[i].name,
-				    S_IRUGO | S_IWUSR,
-				    minor->debugfs_root,
-				    to_i915(minor->dev),
+		debugfs_create_file(i915_debugfs_files[i].name, S_IRUGO | S_IWUSR,
+				    debugfs_root, i915,
 				    i915_debugfs_files[i].fops);
 	}
 
 	drm_debugfs_create_files(i915_debugfs_list,
 				 ARRAY_SIZE(i915_debugfs_list),
-				 minor->debugfs_root, minor);
+				 debugfs_root, i915->drm.primary);
 
-	i915_gpu_error_debugfs_register(dev_priv);
+	i915_gpu_error_debugfs_register(i915);
 }
diff --git a/drivers/gpu/drm/i915/i915_debugfs_params.c b/drivers/gpu/drm/i915/i915_debugfs_params.c
index 33d2dcb0de65..89ab5eb14779 100644
--- a/drivers/gpu/drm/i915/i915_debugfs_params.c
+++ b/drivers/gpu/drm/i915/i915_debugfs_params.c
@@ -248,11 +248,11 @@ i915_debugfs_create_charp(const char *name, umode_t mode,
 /* add a subdirectory with files for each i915 param */
 struct dentry *i915_debugfs_params(struct drm_i915_private *i915)
 {
-	struct drm_minor *minor = i915->drm.primary;
+	struct dentry *debugfs_root = i915->drm.debugfs_root;
 	struct i915_params *params = &i915->params;
 	struct dentry *dir;
 
-	dir = debugfs_create_dir("i915_params", minor->debugfs_root);
+	dir = debugfs_create_dir("i915_params", debugfs_root);
 	if (IS_ERR(dir))
 		return dir;
 
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index f434b6825fc2..41256fbd1ce3 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -2445,11 +2445,11 @@ static const struct file_operations i915_error_state_fops = {
 
 void i915_gpu_error_debugfs_register(struct drm_i915_private *i915)
 {
-	struct drm_minor *minor = i915->drm.primary;
+	struct dentry *debugfs_root = i915->drm.debugfs_root;
 
-	debugfs_create_file("i915_error_state", 0644, minor->debugfs_root, i915,
+	debugfs_create_file("i915_error_state", 0644, debugfs_root, i915,
 			    &i915_error_state_fops);
-	debugfs_create_file("i915_gpu_info", 0644, minor->debugfs_root, i915,
+	debugfs_create_file("i915_gpu_info", 0644, debugfs_root, i915,
 			    &i915_gpu_info_fops);
 }
 
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c
index e07c5b380789..545f79eb0cc5 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp_debugfs.c
@@ -69,17 +69,17 @@ DEFINE_SIMPLE_ATTRIBUTE(pxp_terminate_fops, pxp_terminate_get, pxp_terminate_set
 
 void intel_pxp_debugfs_register(struct intel_pxp *pxp)
 {
-	struct drm_minor *minor;
+	struct dentry *debugfs_root;
 	struct dentry *pxproot;
 
 	if (!intel_pxp_is_supported(pxp))
 		return;
 
-	minor = pxp->ctrl_gt->i915->drm.primary;
-	if (!minor->debugfs_root)
+	debugfs_root = pxp->ctrl_gt->i915->drm.debugfs_root;
+	if (!debugfs_root)
 		return;
 
-	pxproot = debugfs_create_dir("pxp", minor->debugfs_root);
+	pxproot = debugfs_create_dir("pxp", debugfs_root);
 	if (IS_ERR(pxproot))
 		return;
 

From 5fe8d1dba706ee5234acef29c4410ade89170193 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Thu, 7 Aug 2025 07:26:35 +0300
Subject: [PATCH 0339/1455] drm/i915/dsi: Fix overflow issue in pclk parsing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Parsed divider p will overflow and is considered being valid in case
pll_ctl == 0.

Fix this by checking divider p before decreasing it. Also small improvement
is made by using fls() instead of custom loop.

v2: use fls() and check parsed divider

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/20250807042635.2491537-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/vlv_dsi_pll.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
index b52463fdec47..83afe1315e96 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
@@ -142,11 +142,9 @@ static int vlv_dsi_pclk(struct intel_encoder *encoder,
 	pll_div &= DSI_PLL_M1_DIV_MASK;
 	pll_div = pll_div >> DSI_PLL_M1_DIV_SHIFT;
 
-	while (pll_ctl) {
-		pll_ctl = pll_ctl >> 1;
-		p++;
-	}
-	p--;
+	p = fls(pll_ctl);
+	if (p)
+		p--;
 
 	if (!p) {
 		drm_err(display->drm, "wrong P1 divisor\n");

From 46fb38cb20c0d185a6391ab524b23e0e0219c41f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 1 Aug 2025 09:29:05 +0300
Subject: [PATCH 0340/1455] drm/i915/psr: Do not trigger Frame Change events
 from frontbuffer flush
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We want to get rid of triggering "Frame Change" events from
frontbuffer flush calls. We are about to move using TRANS_PUSH
register for this on LunarLake and onwards. Touching TRANS_PUSH
register from fronbuffer flush would be problematic as it's written by
DSB as well.

Fix this by using intel_psr_exit when flush or invalidate is done on
LunarLake and onwards. This is not possible on AlderLake and
MeteorLake due to HW bug in PSR2 disable.

This patch is also fixing problems with cursor plane where cursor is
disappearing or duplicate cursor is seen on the screen.

v2: Commit message updated

Bspec: 68927, 68934, 66624
Reported-by: Janna Martl <janna.martl109@gmail.com>
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5522
Fixes: 411ad63877bb ("drm/i915/psr: Use SFF_CTL on invalidate/flush for LunarLake onwards")
Tested-by: Janna Martl <janna.martl109@gmail.com>
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250801062905.564453-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 1d70ae31b6dd..8bea2b8188a7 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -3254,7 +3254,9 @@ static void intel_psr_configure_full_frame_update(struct intel_dp *intel_dp)
 
 static void _psr_invalidate_handle(struct intel_dp *intel_dp)
 {
-	if (intel_dp->psr.psr2_sel_fetch_enabled) {
+	struct intel_display *display = to_intel_display(intel_dp);
+
+	if (DISPLAY_VER(display) < 20 && intel_dp->psr.psr2_sel_fetch_enabled) {
 		if (!intel_dp->psr.psr2_sel_fetch_cff_enabled) {
 			intel_dp->psr.psr2_sel_fetch_cff_enabled = true;
 			intel_psr_configure_full_frame_update(intel_dp);
@@ -3340,7 +3342,7 @@ static void _psr_flush_handle(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
 
-	if (intel_dp->psr.psr2_sel_fetch_enabled) {
+	if (DISPLAY_VER(display) < 20 && intel_dp->psr.psr2_sel_fetch_enabled) {
 		if (intel_dp->psr.psr2_sel_fetch_cff_enabled) {
 			/* can we turn CFF off? */
 			if (intel_dp->psr.busy_frontbuffer_bits == 0)
@@ -3357,11 +3359,13 @@ static void _psr_flush_handle(struct intel_dp *intel_dp)
 		 * existing SU configuration
 		 */
 		intel_psr_configure_full_frame_update(intel_dp);
+
+		intel_psr_force_update(intel_dp);
+	} else {
+		intel_psr_exit(intel_dp);
 	}
 
-	intel_psr_force_update(intel_dp);
-
-	if (!intel_dp->psr.psr2_sel_fetch_enabled && !intel_dp->psr.active &&
+	if ((!intel_dp->psr.psr2_sel_fetch_enabled || DISPLAY_VER(display) >= 20) &&
 	    !intel_dp->psr.busy_frontbuffer_bits)
 		queue_work(display->wq.unordered, &intel_dp->psr.work);
 }

From 38b34e928a08ba594c4bbf7118aa3aadacd62fff Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 31 Jul 2025 10:38:09 +0100
Subject: [PATCH 0341/1455] drm/xe/migrate: prevent infinite recursion

If the buf + offset is not aligned to XE_CAHELINE_BYTES we fallback to
using a bounce buffer. However the bounce buffer here is allocated on
the stack, and the only alignment requirement here is that it's
naturally aligned to u8, and not XE_CACHELINE_BYTES. If the bounce
buffer is also misaligned we then recurse back into the function again,
however the new bounce buffer might also not be aligned, and might never
be until we eventually blow through the stack, as we keep recursing.

Instead of using the stack use kmalloc, which should respect the
power-of-two alignment request here. Fixes a kernel panic when
triggering this path through eudebug.

v2 (Stuart):
 - Add build bug check for power-of-two restriction
 - s/EINVAL/ENOMEM/

Fixes: 270172f64b11 ("drm/xe: Update xe_ttm_access_memory to use GPU for non-visible access")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Maciej Patelczyk <maciej.patelczyk@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250731093807.207572-6-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c | 29 +++++++++++++++++------------
 1 file changed, 17 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 0f9636a06083..1e9ff18656db 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -2017,15 +2017,19 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 	if (!IS_ALIGNED(len, XE_CACHELINE_BYTES) ||
 	    !IS_ALIGNED((unsigned long)buf + offset, XE_CACHELINE_BYTES)) {
 		int buf_offset = 0;
+		void *bounce;
+		int err;
+
+		BUILD_BUG_ON(!is_power_of_2(XE_CACHELINE_BYTES));
+		bounce = kmalloc(XE_CACHELINE_BYTES, GFP_KERNEL);
+		if (!bounce)
+			return -ENOMEM;
 
 		/*
 		 * Less than ideal for large unaligned access but this should be
 		 * fairly rare, can fixup if this becomes common.
 		 */
 		do {
-			u8 bounce[XE_CACHELINE_BYTES];
-			void *ptr = (void *)bounce;
-			int err;
 			int copy_bytes = min_t(int, bytes_left,
 					       XE_CACHELINE_BYTES -
 					       (offset & XE_CACHELINE_MASK));
@@ -2034,22 +2038,22 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 			err = xe_migrate_access_memory(m, bo,
 						       offset &
 						       ~XE_CACHELINE_MASK,
-						       (void *)ptr,
-						       sizeof(bounce), 0);
+						       bounce,
+						       XE_CACHELINE_BYTES, 0);
 			if (err)
-				return err;
+				break;
 
 			if (write) {
-				memcpy(ptr + ptr_offset, buf + buf_offset, copy_bytes);
+				memcpy(bounce + ptr_offset, buf + buf_offset, copy_bytes);
 
 				err = xe_migrate_access_memory(m, bo,
 							       offset & ~XE_CACHELINE_MASK,
-							       (void *)ptr,
-							       sizeof(bounce), write);
+							       bounce,
+							       XE_CACHELINE_BYTES, write);
 				if (err)
-					return err;
+					break;
 			} else {
-				memcpy(buf + buf_offset, ptr + ptr_offset,
+				memcpy(buf + buf_offset, bounce + ptr_offset,
 				       copy_bytes);
 			}
 
@@ -2058,7 +2062,8 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 			offset += copy_bytes;
 		} while (bytes_left);
 
-		return 0;
+		kfree(bounce);
+		return err;
 	}
 
 	pagemap_addr = xe_migrate_dma_map(xe, buf, len + page_offset, write);

From 8c2d61e0e916e077fda7e7b8e67f25ffe0f361fc Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 31 Jul 2025 10:38:10 +0100
Subject: [PATCH 0342/1455] drm/xe/migrate: don't overflow max copy size

With non-page aligned copy, we need to use 4 byte aligned pitch, however
the size itself might still be close to our maximum of ~8M, and so the
dimensions of the copy can easily exceed the S16_MAX limit of the copy
command leading to the following assert:

xe 0000:03:00.0: [drm] Assertion `size / pitch <= ((s16)(((u16)~0U) >> 1))` failed!
platform: BATTLEMAGE subplatform: 1
graphics: Xe2_HPG 20.01 step A0
media: Xe2_HPM 13.01 step A1
tile: 0 VRAM 10.0 GiB
GT: 0 type 1

WARNING: CPU: 23 PID: 10605 at drivers/gpu/drm/xe/xe_migrate.c:673 emit_copy+0x4b5/0x4e0 [xe]

To fix this account for the pitch when calculating the number of current
bytes to copy.

Fixes: 270172f64b11 ("drm/xe: Update xe_ttm_access_memory to use GPU for non-visible access")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Maciej Patelczyk <maciej.patelczyk@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250731093807.207572-7-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 1e9ff18656db..b36f5997706c 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -2084,6 +2084,12 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 		else
 			current_bytes = min_t(int, bytes_left, cursor.size);
 
+		if (current_bytes & ~PAGE_MASK) {
+			int pitch = 4;
+
+			current_bytes = min_t(int, current_bytes, S16_MAX * pitch);
+		}
+
 		if (fence)
 			dma_fence_put(fence);
 

From 9b7ca35ed28fe5fad86e9d9c24ebd1271e4c9c3e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 31 Jul 2025 10:38:11 +0100
Subject: [PATCH 0343/1455] drm/xe/migrate: prevent potential UAF

If we hit the error path, the previous fence (if there is one) has
already been put() prior to this, so doing a fence_wait could lead to
UAF. Tweak the flow to do to the put() until after we do the wait.

Fixes: 270172f64b11 ("drm/xe: Update xe_ttm_access_memory to use GPU for non-visible access")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Maciej Patelczyk <maciej.patelczyk@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250731093807.207572-8-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index b36f5997706c..4effe10d815b 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -2090,9 +2090,6 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 			current_bytes = min_t(int, current_bytes, S16_MAX * pitch);
 		}
 
-		if (fence)
-			dma_fence_put(fence);
-
 		__fence = xe_migrate_vram(m, current_bytes,
 					  (unsigned long)buf & ~PAGE_MASK,
 					  &pagemap_addr[current_page],
@@ -2100,11 +2097,15 @@ int xe_migrate_access_memory(struct xe_migrate *m, struct xe_bo *bo,
 					  XE_MIGRATE_COPY_TO_VRAM :
 					  XE_MIGRATE_COPY_TO_SRAM);
 		if (IS_ERR(__fence)) {
-			if (fence)
+			if (fence) {
 				dma_fence_wait(fence, false);
+				dma_fence_put(fence);
+			}
 			fence = __fence;
 			goto out_err;
 		}
+
+		dma_fence_put(fence);
 		fence = __fence;
 
 		buf += current_bytes;

From 80944d334182ce5eb27d00e2bf20a88bfc32dea1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Tue, 5 Aug 2025 09:48:42 +0200
Subject: [PATCH 0344/1455] drm/xe: Defer buffer object shrinker write-backs
 and GPU waits
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When the xe buffer-object shrinker allows GPU waits and write-back,
(typically from kswapd), perform multiple passes, skipping
subsequent passes if the shrinker number of scanned objects target
is reached.

1) Without GPU waits and write-back
2) Without write-back
3) With both GPU-waits and write-back

This is to avoid stalls and costly write- and readbacks unless they
are really necessary.

v2:
- Don't test for scan completion twice. (Stuart Summers)
- Update tags.

Reported-by: melvyn <melvyn2@dnsense.pub>
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5557
Cc: Summers Stuart <stuart.summers@intel.com>
Fixes: 00c8efc3180f ("drm/xe: Add a shrinker for xe bos")
Cc: <stable@vger.kernel.org> # v6.15+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250805074842.11359-1-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_shrinker.c | 51 +++++++++++++++++++++++++++++---
 1 file changed, 47 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_shrinker.c b/drivers/gpu/drm/xe/xe_shrinker.c
index 125c836e0ee4..967448f683eb 100644
--- a/drivers/gpu/drm/xe/xe_shrinker.c
+++ b/drivers/gpu/drm/xe/xe_shrinker.c
@@ -54,10 +54,10 @@ xe_shrinker_mod_pages(struct xe_shrinker *shrinker, long shrinkable, long purgea
 	write_unlock(&shrinker->lock);
 }
 
-static s64 xe_shrinker_walk(struct xe_device *xe,
-			    struct ttm_operation_ctx *ctx,
-			    const struct xe_bo_shrink_flags flags,
-			    unsigned long to_scan, unsigned long *scanned)
+static s64 __xe_shrinker_walk(struct xe_device *xe,
+			      struct ttm_operation_ctx *ctx,
+			      const struct xe_bo_shrink_flags flags,
+			      unsigned long to_scan, unsigned long *scanned)
 {
 	unsigned int mem_type;
 	s64 freed = 0, lret;
@@ -87,6 +87,48 @@ static s64 xe_shrinker_walk(struct xe_device *xe,
 	return freed;
 }
 
+/*
+ * Try shrinking idle objects without writeback first, then if not sufficient,
+ * try also non-idle objects and finally if that's not sufficient either,
+ * add writeback. This avoids stalls and explicit writebacks with light or
+ * moderate memory pressure.
+ */
+static s64 xe_shrinker_walk(struct xe_device *xe,
+			    struct ttm_operation_ctx *ctx,
+			    const struct xe_bo_shrink_flags flags,
+			    unsigned long to_scan, unsigned long *scanned)
+{
+	bool no_wait_gpu = true;
+	struct xe_bo_shrink_flags save_flags = flags;
+	s64 lret, freed;
+
+	swap(no_wait_gpu, ctx->no_wait_gpu);
+	save_flags.writeback = false;
+	lret = __xe_shrinker_walk(xe, ctx, save_flags, to_scan, scanned);
+	swap(no_wait_gpu, ctx->no_wait_gpu);
+	if (lret < 0 || *scanned >= to_scan)
+		return lret;
+
+	freed = lret;
+	if (!ctx->no_wait_gpu) {
+		lret = __xe_shrinker_walk(xe, ctx, save_flags, to_scan, scanned);
+		if (lret < 0)
+			return lret;
+		freed += lret;
+		if (*scanned >= to_scan)
+			return freed;
+	}
+
+	if (flags.writeback) {
+		lret = __xe_shrinker_walk(xe, ctx, flags, to_scan, scanned);
+		if (lret < 0)
+			return lret;
+		freed += lret;
+	}
+
+	return freed;
+}
+
 static unsigned long
 xe_shrinker_count(struct shrinker *shrink, struct shrink_control *sc)
 {
@@ -193,6 +235,7 @@ static unsigned long xe_shrinker_scan(struct shrinker *shrink, struct shrink_con
 		runtime_pm = xe_shrinker_runtime_pm_get(shrinker, true, 0, can_backup);
 
 	shrink_flags.purge = false;
+
 	lret = xe_shrinker_walk(shrinker->xe, &ctx, shrink_flags,
 				nr_to_scan, &nr_scanned);
 	if (lret >= 0)

From 4d5c98eb77fe518944465d96e69de8aa6e9a5c99 Mon Sep 17 00:00:00 2001
From: Matt Atwood <matthew.s.atwood@intel.com>
Date: Thu, 7 Aug 2025 14:42:23 -0700
Subject: [PATCH 0345/1455] drm/xe: rename XE_WA to XE_GT_WA

Now that there are two types of wa tables and infrastructure, be more
concise in the naming of GT wa macros.

v2: update the documentation

Signed-off-by: Matt Atwood <matthew.s.atwood@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250807214224.32728-1-matthew.s.atwood@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/display/intel_fbdev_fb.c   |  2 +-
 drivers/gpu/drm/xe/display/xe_display_wa.c    |  2 +-
 drivers/gpu/drm/xe/display/xe_plane_initial.c |  2 +-
 drivers/gpu/drm/xe/xe_device.c                |  6 +++---
 drivers/gpu/drm/xe/xe_eu_stall.c              |  4 ++--
 drivers/gpu/drm/xe/xe_ggtt.c                  |  8 ++++----
 drivers/gpu/drm/xe/xe_gsc.c                   |  6 +++---
 drivers/gpu/drm/xe/xe_gt.c                    |  6 +++---
 drivers/gpu/drm/xe/xe_gt_topology.c           |  2 +-
 drivers/gpu/drm/xe/xe_gt_types.h              |  2 +-
 drivers/gpu/drm/xe/xe_guc.c                   | 14 ++++++-------
 drivers/gpu/drm/xe/xe_guc_ads.c               | 20 +++++++++----------
 drivers/gpu/drm/xe/xe_guc_pc.c                | 10 +++++-----
 drivers/gpu/drm/xe/xe_hw_engine.c             |  2 +-
 drivers/gpu/drm/xe/xe_lrc.c                   |  8 ++++----
 drivers/gpu/drm/xe/xe_oa.c                    |  8 ++++----
 drivers/gpu/drm/xe/xe_query.c                 |  4 ++--
 drivers/gpu/drm/xe/xe_ring_ops.c              |  6 +++---
 drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c        |  2 +-
 drivers/gpu/drm/xe/xe_vm.c                    |  2 +-
 drivers/gpu/drm/xe/xe_wa.c                    |  2 +-
 drivers/gpu/drm/xe/xe_wa.h                    |  4 ++--
 22 files changed, 61 insertions(+), 61 deletions(-)

diff --git a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
index b28a94df824f..70dbdf106fb6 100644
--- a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
+++ b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
@@ -41,7 +41,7 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 	size = PAGE_ALIGN(size);
 	obj = ERR_PTR(-ENODEV);
 
-	if (!IS_DGFX(xe) && !XE_WA(xe_root_mmio_gt(xe), 22019338487_display)) {
+	if (!IS_DGFX(xe) && !XE_GT_WA(xe_root_mmio_gt(xe), 22019338487_display)) {
 		obj = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe),
 					   NULL, size,
 					   ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
diff --git a/drivers/gpu/drm/xe/display/xe_display_wa.c b/drivers/gpu/drm/xe/display/xe_display_wa.c
index 68d1387d81a0..8ada1cbcb16c 100644
--- a/drivers/gpu/drm/xe/display/xe_display_wa.c
+++ b/drivers/gpu/drm/xe/display/xe_display_wa.c
@@ -14,5 +14,5 @@ bool intel_display_needs_wa_16023588340(struct intel_display *display)
 {
 	struct xe_device *xe = to_xe_device(display->drm);
 
-	return XE_WA(xe_root_mmio_gt(xe), 16023588340);
+	return XE_GT_WA(xe_root_mmio_gt(xe), 16023588340);
 }
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index 54ff70ea2d25..1fe06f4f9aca 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -122,7 +122,7 @@ initial_plane_bo(struct xe_device *xe,
 		phys_base = base;
 		flags |= XE_BO_FLAG_STOLEN;
 
-		if (XE_WA(xe_root_mmio_gt(xe), 22019338487_display))
+		if (XE_GT_WA(xe_root_mmio_gt(xe), 22019338487_display))
 			return NULL;
 
 		/*
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 57edbc63da6f..3e0402dff423 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -883,7 +883,7 @@ int xe_device_probe(struct xe_device *xe)
 	}
 
 	if (xe->tiles->media_gt &&
-	    XE_WA(xe->tiles->media_gt, 15015404425_disable))
+	    XE_GT_WA(xe->tiles->media_gt, 15015404425_disable))
 		XE_DEVICE_WA_DISABLE(xe, 15015404425);
 
 	err = xe_devcoredump_init(xe);
@@ -1043,7 +1043,7 @@ void xe_device_l2_flush(struct xe_device *xe)
 
 	gt = xe_root_mmio_gt(xe);
 
-	if (!XE_WA(gt, 16023588340))
+	if (!XE_GT_WA(gt, 16023588340))
 		return;
 
 	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
@@ -1087,7 +1087,7 @@ void xe_device_td_flush(struct xe_device *xe)
 		return;
 
 	root_gt = xe_root_mmio_gt(xe);
-	if (XE_WA(root_gt, 16023588340)) {
+	if (XE_GT_WA(root_gt, 16023588340)) {
 		/* A transient flush is not sufficient: flush the L2 */
 		xe_device_l2_flush(xe);
 	} else {
diff --git a/drivers/gpu/drm/xe/xe_eu_stall.c b/drivers/gpu/drm/xe/xe_eu_stall.c
index af7916315ac6..fdd514fec5ef 100644
--- a/drivers/gpu/drm/xe/xe_eu_stall.c
+++ b/drivers/gpu/drm/xe/xe_eu_stall.c
@@ -649,7 +649,7 @@ static int xe_eu_stall_stream_enable(struct xe_eu_stall_data_stream *stream)
 		return -ETIMEDOUT;
 	}
 
-	if (XE_WA(gt, 22016596838))
+	if (XE_GT_WA(gt, 22016596838))
 		xe_gt_mcr_multicast_write(gt, ROW_CHICKEN2,
 					  _MASKED_BIT_ENABLE(DISABLE_DOP_GATING));
 
@@ -805,7 +805,7 @@ static int xe_eu_stall_disable_locked(struct xe_eu_stall_data_stream *stream)
 
 	cancel_delayed_work_sync(&stream->buf_poll_work);
 
-	if (XE_WA(gt, 22016596838))
+	if (XE_GT_WA(gt, 22016596838))
 		xe_gt_mcr_multicast_write(gt, ROW_CHICKEN2,
 					  _MASKED_BIT_DISABLE(DISABLE_DOP_GATING));
 
diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index 249713139f69..e03222f5ac5a 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -106,10 +106,10 @@ static unsigned int probe_gsm_size(struct pci_dev *pdev)
 static void ggtt_update_access_counter(struct xe_ggtt *ggtt)
 {
 	struct xe_tile *tile = ggtt->tile;
-	struct xe_gt *affected_gt = XE_WA(tile->primary_gt, 22019338487) ?
+	struct xe_gt *affected_gt = XE_GT_WA(tile->primary_gt, 22019338487) ?
 		tile->primary_gt : tile->media_gt;
 	struct xe_mmio *mmio = &affected_gt->mmio;
-	u32 max_gtt_writes = XE_WA(ggtt->tile->primary_gt, 22019338487) ? 1100 : 63;
+	u32 max_gtt_writes = XE_GT_WA(ggtt->tile->primary_gt, 22019338487) ? 1100 : 63;
 	/*
 	 * Wa_22019338487: GMD_ID is a RO register, a dummy write forces gunit
 	 * to wait for completion of prior GTT writes before letting this through.
@@ -284,8 +284,8 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt)
 
 	if (GRAPHICS_VERx100(xe) >= 1270)
 		ggtt->pt_ops = (ggtt->tile->media_gt &&
-			       XE_WA(ggtt->tile->media_gt, 22019338487)) ||
-			       XE_WA(ggtt->tile->primary_gt, 22019338487) ?
+			       XE_GT_WA(ggtt->tile->media_gt, 22019338487)) ||
+			       XE_GT_WA(ggtt->tile->primary_gt, 22019338487) ?
 			       &xelpg_pt_wa_ops : &xelpg_pt_ops;
 	else
 		ggtt->pt_ops = &xelp_pt_ops;
diff --git a/drivers/gpu/drm/xe/xe_gsc.c b/drivers/gpu/drm/xe/xe_gsc.c
index 1d84bf2f2cef..f5ae28af60d4 100644
--- a/drivers/gpu/drm/xe/xe_gsc.c
+++ b/drivers/gpu/drm/xe/xe_gsc.c
@@ -266,7 +266,7 @@ static int gsc_upload_and_init(struct xe_gsc *gsc)
 	unsigned int fw_ref;
 	int ret;
 
-	if (XE_WA(tile->primary_gt, 14018094691)) {
+	if (XE_GT_WA(tile->primary_gt, 14018094691)) {
 		fw_ref = xe_force_wake_get(gt_to_fw(tile->primary_gt), XE_FORCEWAKE_ALL);
 
 		/*
@@ -281,7 +281,7 @@ static int gsc_upload_and_init(struct xe_gsc *gsc)
 
 	ret = gsc_upload(gsc);
 
-	if (XE_WA(tile->primary_gt, 14018094691))
+	if (XE_GT_WA(tile->primary_gt, 14018094691))
 		xe_force_wake_put(gt_to_fw(tile->primary_gt), fw_ref);
 
 	if (ret)
@@ -593,7 +593,7 @@ void xe_gsc_wa_14015076503(struct xe_gt *gt, bool prep)
 	u32 gs1_clr = prep ? 0 : HECI_H_GS1_ER_PREP;
 
 	/* WA only applies if the GSC is loaded */
-	if (!XE_WA(gt, 14015076503) || !gsc_fw_is_loaded(gt))
+	if (!XE_GT_WA(gt, 14015076503) || !gsc_fw_is_loaded(gt))
 		return;
 
 	xe_mmio_rmw32(&gt->mmio, HECI_H_GS1(MTL_GSC_HECI2_BASE), gs1_clr, gs1_set);
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 390394bbaadc..43f2da27d9d9 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -106,7 +106,7 @@ static void xe_gt_enable_host_l2_vram(struct xe_gt *gt)
 	unsigned int fw_ref;
 	u32 reg;
 
-	if (!XE_WA(gt, 16023588340))
+	if (!XE_GT_WA(gt, 16023588340))
 		return;
 
 	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
@@ -128,7 +128,7 @@ static void xe_gt_disable_host_l2_vram(struct xe_gt *gt)
 	unsigned int fw_ref;
 	u32 reg;
 
-	if (!XE_WA(gt, 16023588340))
+	if (!XE_GT_WA(gt, 16023588340))
 		return;
 
 	if (xe_gt_is_media_type(gt))
@@ -966,7 +966,7 @@ int xe_gt_sanitize_freq(struct xe_gt *gt)
 	if ((!xe_uc_fw_is_available(&gt->uc.gsc.fw) ||
 	     xe_uc_fw_is_loaded(&gt->uc.gsc.fw) ||
 	     xe_uc_fw_is_in_error_state(&gt->uc.gsc.fw)) &&
-	    XE_WA(gt, 22019338487))
+	    XE_GT_WA(gt, 22019338487))
 		ret = xe_guc_pc_restore_stashed_freq(&gt->uc.guc.pc);
 
 	return ret;
diff --git a/drivers/gpu/drm/xe/xe_gt_topology.c b/drivers/gpu/drm/xe/xe_gt_topology.c
index 8c63e3263643..a0baa560dd71 100644
--- a/drivers/gpu/drm/xe/xe_gt_topology.c
+++ b/drivers/gpu/drm/xe/xe_gt_topology.c
@@ -138,7 +138,7 @@ load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask)
 	 * but there's no tracking number assigned yet so we use a custom
 	 * OOB workaround descriptor.
 	 */
-	if (XE_WA(gt, no_media_l3))
+	if (XE_GT_WA(gt, no_media_l3))
 		return;
 
 	if (GRAPHICS_VER(xe) >= 30) {
diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
index dfd4a16da5f0..ef0f2eecfa29 100644
--- a/drivers/gpu/drm/xe/xe_gt_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_types.h
@@ -413,7 +413,7 @@ struct xe_gt {
 		unsigned long *oob;
 		/**
 		 * @wa_active.oob_initialized: mark oob as initialized to help
-		 * detecting misuse of XE_WA() - it can only be called on
+		 * detecting misuse of XE_GT_WA() - it can only be called on
 		 * initialization after OOB WAs have being processed
 		 */
 		bool oob_initialized;
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 9e34401e4489..433abc787f7b 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -157,7 +157,7 @@ static bool needs_wa_dual_queue(struct xe_gt *gt)
 	 * on RCS and CCSes with different address spaces, which on DG2 is
 	 * required as a WA for an HW bug.
 	 */
-	if (XE_WA(gt, 22011391025))
+	if (XE_GT_WA(gt, 22011391025))
 		return true;
 
 	/*
@@ -184,10 +184,10 @@ static u32 guc_ctl_wa_flags(struct xe_guc *guc)
 	struct xe_gt *gt = guc_to_gt(guc);
 	u32 flags = 0;
 
-	if (XE_WA(gt, 22012773006))
+	if (XE_GT_WA(gt, 22012773006))
 		flags |= GUC_WA_POLLCS;
 
-	if (XE_WA(gt, 14014475959))
+	if (XE_GT_WA(gt, 14014475959))
 		flags |= GUC_WA_HOLD_CCS_SWITCHOUT;
 
 	if (needs_wa_dual_queue(gt))
@@ -201,17 +201,17 @@ static u32 guc_ctl_wa_flags(struct xe_guc *guc)
 	if (GRAPHICS_VERx100(xe) < 1270)
 		flags |= GUC_WA_PRE_PARSER;
 
-	if (XE_WA(gt, 22012727170) || XE_WA(gt, 22012727685))
+	if (XE_GT_WA(gt, 22012727170) || XE_GT_WA(gt, 22012727685))
 		flags |= GUC_WA_CONTEXT_ISOLATION;
 
-	if (XE_WA(gt, 18020744125) &&
+	if (XE_GT_WA(gt, 18020744125) &&
 	    !xe_hw_engine_mask_per_class(gt, XE_ENGINE_CLASS_RENDER))
 		flags |= GUC_WA_RCS_REGS_IN_CCS_REGS_LIST;
 
-	if (XE_WA(gt, 1509372804))
+	if (XE_GT_WA(gt, 1509372804))
 		flags |= GUC_WA_RENDER_RST_RC6_EXIT;
 
-	if (XE_WA(gt, 14018913170))
+	if (XE_GT_WA(gt, 14018913170))
 		flags |= GUC_WA_ENABLE_TSC_CHECK_ON_RC6;
 
 	return flags;
diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
index 2ceaa197cb2f..d7da67637079 100644
--- a/drivers/gpu/drm/xe/xe_guc_ads.c
+++ b/drivers/gpu/drm/xe/xe_guc_ads.c
@@ -247,7 +247,7 @@ static size_t calculate_regset_size(struct xe_gt *gt)
 
 	count += ADS_REGSET_EXTRA_MAX * XE_NUM_HW_ENGINES;
 
-	if (XE_WA(gt, 1607983814))
+	if (XE_GT_WA(gt, 1607983814))
 		count += LNCFCMOCS_REG_COUNT;
 
 	return count * sizeof(struct guc_mmio_reg);
@@ -317,17 +317,17 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 	offset = guc_ads_waklv_offset(ads);
 	remain = guc_ads_waklv_size(ads);
 
-	if (XE_WA(gt, 14019882105) || XE_WA(gt, 16021333562))
+	if (XE_GT_WA(gt, 14019882105) || XE_GT_WA(gt, 16021333562))
 		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
 				 GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED);
-	if (XE_WA(gt, 18024947630))
+	if (XE_GT_WA(gt, 18024947630))
 		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
 				 GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING);
-	if (XE_WA(gt, 16022287689))
+	if (XE_GT_WA(gt, 16022287689))
 		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
 				 GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE);
 
-	if (XE_WA(gt, 14022866841))
+	if (XE_GT_WA(gt, 14022866841))
 		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
 				 GUC_WA_KLV_WAKE_POWER_DOMAINS_FOR_OUTBOUND_MMIO);
 
@@ -336,21 +336,21 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 	 * the default value for this register is determined to be 0xC40. This could change in the
 	 * future, so GuC depends on KMD to send it the correct value.
 	 */
-	if (XE_WA(gt, 13011645652)) {
+	if (XE_GT_WA(gt, 13011645652)) {
 		u32 data = 0xC40;
 
 		guc_waklv_enable(ads, &data, sizeof(data) / sizeof(u32), &offset, &remain,
 				 GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE);
 	}
 
-	if (XE_WA(gt, 14022293748) || XE_WA(gt, 22019794406))
+	if (XE_GT_WA(gt, 14022293748) || XE_GT_WA(gt, 22019794406))
 		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
 				 GUC_WORKAROUND_KLV_ID_BACK_TO_BACK_RCS_ENGINE_RESET);
 
-	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_WA(gt, 16026508708))
+	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 44, 0) && XE_GT_WA(gt, 16026508708))
 		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
 				 GUC_WA_KLV_RESET_BB_STACK_PTR_ON_VF_SWITCH);
-	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 47, 0) && XE_WA(gt, 16026007364)) {
+	if (GUC_FIRMWARE_VER(&gt->uc.guc) >= MAKE_GUC_VER(70, 47, 0) && XE_GT_WA(gt, 16026007364)) {
 		u32 data[] = {
 			0x0,
 			0xF,
@@ -761,7 +761,7 @@ static unsigned int guc_mmio_regset_write(struct xe_guc_ads *ads,
 		guc_mmio_regset_write_one(ads, regset_map, e->reg, count++);
 	}
 
-	if (XE_WA(hwe->gt, 1607983814) && hwe->class == XE_ENGINE_CLASS_RENDER) {
+	if (XE_GT_WA(hwe->gt, 1607983814) && hwe->class == XE_ENGINE_CLASS_RENDER) {
 		for (i = 0; i < LNCFCMOCS_REG_COUNT; i++) {
 			guc_mmio_regset_write_one(ads, regset_map,
 						  XELP_LNCFCMOCS(i), count++);
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c
index 68b192fe3b32..0f8430acd16d 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.c
+++ b/drivers/gpu/drm/xe/xe_guc_pc.c
@@ -722,7 +722,7 @@ static int xe_guc_pc_set_max_freq_locked(struct xe_guc_pc *pc, u32 freq)
  */
 int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq)
 {
-	if (XE_WA(pc_to_gt(pc), 22019338487)) {
+	if (XE_GT_WA(pc_to_gt(pc), 22019338487)) {
 		if (wait_for_flush_complete(pc) != 0)
 			return -EAGAIN;
 	}
@@ -835,7 +835,7 @@ static u32 pc_max_freq_cap(struct xe_guc_pc *pc)
 {
 	struct xe_gt *gt = pc_to_gt(pc);
 
-	if (XE_WA(gt, 22019338487)) {
+	if (XE_GT_WA(gt, 22019338487)) {
 		if (xe_gt_is_media_type(gt))
 			return min(LNL_MERT_FREQ_CAP, pc->rp0_freq);
 		else
@@ -899,7 +899,7 @@ static int pc_adjust_freq_bounds(struct xe_guc_pc *pc)
 	if (pc_get_min_freq(pc) > pc->rp0_freq)
 		ret = pc_set_min_freq(pc, pc->rp0_freq);
 
-	if (XE_WA(tile->primary_gt, 14022085890))
+	if (XE_GT_WA(tile->primary_gt, 14022085890))
 		ret = pc_set_min_freq(pc, max(BMG_MIN_FREQ, pc_get_min_freq(pc)));
 
 out:
@@ -931,7 +931,7 @@ static bool needs_flush_freq_limit(struct xe_guc_pc *pc)
 {
 	struct xe_gt *gt = pc_to_gt(pc);
 
-	return  XE_WA(gt, 22019338487) &&
+	return  XE_GT_WA(gt, 22019338487) &&
 		pc->rp0_freq > BMG_MERT_FLUSH_FREQ_CAP;
 }
 
@@ -1017,7 +1017,7 @@ static int pc_set_mert_freq_cap(struct xe_guc_pc *pc)
 {
 	int ret;
 
-	if (!XE_WA(pc_to_gt(pc), 22019338487))
+	if (!XE_GT_WA(pc_to_gt(pc), 22019338487))
 		return 0;
 
 	guard(mutex)(&pc->freq_lock);
diff --git a/drivers/gpu/drm/xe/xe_hw_engine.c b/drivers/gpu/drm/xe/xe_hw_engine.c
index 796ba8c34a16..1cf623b4a5bc 100644
--- a/drivers/gpu/drm/xe/xe_hw_engine.c
+++ b/drivers/gpu/drm/xe/xe_hw_engine.c
@@ -576,7 +576,7 @@ static void adjust_idledly(struct xe_hw_engine *hwe)
 	u32 maxcnt_units_ns = 640;
 	bool inhibit_switch = 0;
 
-	if (!IS_SRIOV_VF(gt_to_xe(hwe->gt)) && XE_WA(gt, 16023105232)) {
+	if (!IS_SRIOV_VF(gt_to_xe(hwe->gt)) && XE_GT_WA(gt, 16023105232)) {
 		idledly = xe_mmio_read32(&gt->mmio, RING_IDLEDLY(hwe->mmio_base));
 		maxcnt = xe_mmio_read32(&gt->mmio, RING_PWRCTX_MAXCNT(hwe->mmio_base));
 
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 8760c4c2ca38..8f6c3ba47882 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -75,7 +75,7 @@ lrc_to_xe(struct xe_lrc *lrc)
 static bool
 gt_engine_needs_indirect_ctx(struct xe_gt *gt, enum xe_engine_class class)
 {
-	if (XE_WA(gt, 16010904313) &&
+	if (XE_GT_WA(gt, 16010904313) &&
 	    (class == XE_ENGINE_CLASS_RENDER ||
 	     class == XE_ENGINE_CLASS_COMPUTE))
 		return true;
@@ -1071,7 +1071,7 @@ static ssize_t setup_timestamp_wa(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 	const u32 ts_addr = __xe_lrc_ctx_timestamp_ggtt_addr(lrc);
 	u32 *cmd = batch;
 
-	if (!XE_WA(lrc->gt, 16010904313) ||
+	if (!XE_GT_WA(lrc->gt, 16010904313) ||
 	    !(hwe->class == XE_ENGINE_CLASS_RENDER ||
 	      hwe->class == XE_ENGINE_CLASS_COMPUTE ||
 	      hwe->class == XE_ENGINE_CLASS_COPY ||
@@ -1108,7 +1108,7 @@ static ssize_t setup_invalidate_state_cache_wa(struct xe_lrc *lrc,
 {
 	u32 *cmd = batch;
 
-	if (!XE_WA(lrc->gt, 18022495364) ||
+	if (!XE_GT_WA(lrc->gt, 18022495364) ||
 	    hwe->class != XE_ENGINE_CLASS_RENDER)
 		return 0;
 
@@ -2093,7 +2093,7 @@ u32 *xe_lrc_emit_hwe_state_instructions(struct xe_exec_queue *q, u32 *cs)
 	 * continue to emit all of the SVG state since it's best not to leak
 	 * any of the state between contexts, even if that leakage is harmless.
 	 */
-	if (XE_WA(gt, 14019789679) && q->hwe->class == XE_ENGINE_CLASS_RENDER) {
+	if (XE_GT_WA(gt, 14019789679) && q->hwe->class == XE_ENGINE_CLASS_RENDER) {
 		state_table = xe_hpg_svg_state;
 		state_table_size = ARRAY_SIZE(xe_hpg_svg_state);
 	}
diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c
index 5729e7d3e335..a188bad172ad 100644
--- a/drivers/gpu/drm/xe/xe_oa.c
+++ b/drivers/gpu/drm/xe/xe_oa.c
@@ -822,7 +822,7 @@ static void xe_oa_disable_metric_set(struct xe_oa_stream *stream)
 	u32 sqcnt1;
 
 	/* Enable thread stall DOP gating and EU DOP gating. */
-	if (XE_WA(stream->gt, 1508761755)) {
+	if (XE_GT_WA(stream->gt, 1508761755)) {
 		xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN,
 					  _MASKED_BIT_DISABLE(STALL_DOP_GATING_DISABLE));
 		xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN2,
@@ -1079,7 +1079,7 @@ static int xe_oa_enable_metric_set(struct xe_oa_stream *stream)
 	 * EU NOA signals behave incorrectly if EU clock gating is enabled.
 	 * Disable thread stall DOP gating and EU DOP gating.
 	 */
-	if (XE_WA(stream->gt, 1508761755)) {
+	if (XE_GT_WA(stream->gt, 1508761755)) {
 		xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN,
 					  _MASKED_BIT_ENABLE(STALL_DOP_GATING_DISABLE));
 		xe_gt_mcr_multicast_write(stream->gt, ROW_CHICKEN2,
@@ -1754,7 +1754,7 @@ static int xe_oa_stream_init(struct xe_oa_stream *stream,
 	 * GuC reset of engines causes OA to lose configuration
 	 * state. Prevent this by overriding GUCRC mode.
 	 */
-	if (XE_WA(stream->gt, 1509372804)) {
+	if (XE_GT_WA(stream->gt, 1509372804)) {
 		ret = xe_guc_pc_override_gucrc_mode(&gt->uc.guc.pc,
 						    SLPC_GUCRC_MODE_GUCRC_NO_RC6);
 		if (ret)
@@ -1886,7 +1886,7 @@ u32 xe_oa_timestamp_frequency(struct xe_gt *gt)
 {
 	u32 reg, shift;
 
-	if (XE_WA(gt, 18013179988) || XE_WA(gt, 14015568240)) {
+	if (XE_GT_WA(gt, 18013179988) || XE_GT_WA(gt, 14015568240)) {
 		xe_pm_runtime_get(gt_to_xe(gt));
 		reg = xe_mmio_read32(&gt->mmio, RPM_CONFIG0);
 		xe_pm_runtime_put(gt_to_xe(gt));
diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
index 44d44bbc71dc..4dbe5732cb7f 100644
--- a/drivers/gpu/drm/xe/xe_query.c
+++ b/drivers/gpu/drm/xe/xe_query.c
@@ -477,7 +477,7 @@ static size_t calc_topo_query_size(struct xe_device *xe)
 			sizeof_field(struct xe_gt, fuse_topo.eu_mask_per_dss);
 
 		/* L3bank mask may not be available for some GTs */
-		if (!XE_WA(gt, no_media_l3))
+		if (!XE_GT_WA(gt, no_media_l3))
 			query_size += sizeof(struct drm_xe_query_topology_mask) +
 				sizeof_field(struct xe_gt, fuse_topo.l3_bank_mask);
 	}
@@ -540,7 +540,7 @@ static int query_gt_topology(struct xe_device *xe,
 		 * mask, then it's better to omit L3 from the query rather than
 		 * reporting bogus or zeroed information to userspace.
 		 */
-		if (!XE_WA(gt, no_media_l3)) {
+		if (!XE_GT_WA(gt, no_media_l3)) {
 			topo.type = DRM_XE_TOPO_L3_BANK;
 			err = copy_mask(&query_ptr, &topo, gt->fuse_topo.l3_bank_mask,
 					sizeof(gt->fuse_topo.l3_bank_mask));
diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
index e8f22ec5f9af..5f15360d14bf 100644
--- a/drivers/gpu/drm/xe/xe_ring_ops.c
+++ b/drivers/gpu/drm/xe/xe_ring_ops.c
@@ -179,7 +179,7 @@ static int emit_render_cache_flush(struct xe_sched_job *job, u32 *dw, int i)
 	bool lacks_render = !(gt->info.engine_mask & XE_HW_ENGINE_RCS_MASK);
 	u32 flags;
 
-	if (XE_WA(gt, 14016712196))
+	if (XE_GT_WA(gt, 14016712196))
 		i = emit_pipe_control(dw, i, 0, PIPE_CONTROL_DEPTH_CACHE_FLUSH,
 				      LRC_PPHWSP_FLUSH_INVAL_SCRATCH_ADDR, 0);
 
@@ -190,7 +190,7 @@ static int emit_render_cache_flush(struct xe_sched_job *job, u32 *dw, int i)
 		 PIPE_CONTROL_DC_FLUSH_ENABLE |
 		 PIPE_CONTROL_FLUSH_ENABLE);
 
-	if (XE_WA(gt, 1409600907))
+	if (XE_GT_WA(gt, 1409600907))
 		flags |= PIPE_CONTROL_DEPTH_STALL;
 
 	if (lacks_render)
@@ -206,7 +206,7 @@ static int emit_pipe_control_to_ring_end(struct xe_hw_engine *hwe, u32 *dw, int
 	if (hwe->class != XE_ENGINE_CLASS_RENDER)
 		return i;
 
-	if (XE_WA(hwe->gt, 16020292621))
+	if (XE_GT_WA(hwe->gt, 16020292621))
 		i = emit_pipe_control(dw, i, 0, PIPE_CONTROL_LRI_POST_SYNC,
 				      RING_NOPID(hwe->mmio_base).addr, 0);
 
diff --git a/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c b/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c
index 9a9733447230..dc588255674d 100644
--- a/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c
+++ b/drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c
@@ -166,7 +166,7 @@ static u32 detect_bar2_integrated(struct xe_device *xe, struct xe_ttm_stolen_mgr
 
 	stolen_size -= wopcm_size;
 
-	if (media_gt && XE_WA(media_gt, 14019821291)) {
+	if (media_gt && XE_GT_WA(media_gt, 14019821291)) {
 		u64 gscpsmi_base = xe_mmio_read64_2x32(&media_gt->mmio, GSCPSMI_BASE)
 			& ~GENMASK_ULL(5, 0);
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 432ea325677d..148a2425006f 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2068,7 +2068,7 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data,
 	if (XE_IOCTL_DBG(xe, args->extensions))
 		return -EINVAL;
 
-	if (XE_WA(xe_root_mmio_gt(xe), 14016763929))
+	if (XE_GT_WA(xe_root_mmio_gt(xe), 14016763929))
 		args->flags |= DRM_XE_VM_CREATE_FLAG_SCRATCH_PAGE;
 
 	if (XE_IOCTL_DBG(xe, args->flags & DRM_XE_VM_CREATE_FLAG_FAULT_MODE &&
diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index 535067e7fb0c..0fc73af1bb4c 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -1090,6 +1090,6 @@ void xe_wa_apply_tile_workarounds(struct xe_tile *tile)
 	if (IS_SRIOV_VF(tile->xe))
 		return;
 
-	if (XE_WA(tile->primary_gt, 22010954014))
+	if (XE_GT_WA(tile->primary_gt, 22010954014))
 		xe_mmio_rmw32(mmio, XEHP_CLOCK_GATE_DIS, 0, SGSI_SIDECLK_DIS);
 }
diff --git a/drivers/gpu/drm/xe/xe_wa.h b/drivers/gpu/drm/xe/xe_wa.h
index f3880c65cb8d..105497c1d7d7 100644
--- a/drivers/gpu/drm/xe/xe_wa.h
+++ b/drivers/gpu/drm/xe/xe_wa.h
@@ -25,11 +25,11 @@ void xe_wa_device_dump(struct xe_device *xe, struct drm_printer *p);
 void xe_wa_dump(struct xe_gt *gt, struct drm_printer *p);
 
 /**
- * XE_WA - Out-of-band workarounds, to be queried and called as needed.
+ * XE_GT_WA - Out-of-band GT workarounds, to be queried and called as needed.
  * @gt__: gt instance
  * @id__: XE_OOB_<id__>, as generated by build system in generated/xe_wa_oob.h
  */
-#define XE_WA(gt__, id__) ({						\
+#define XE_GT_WA(gt__, id__) ({						\
 	xe_gt_assert(gt__, (gt__)->wa_active.oob_initialized);		\
 	test_bit(XE_WA_OOB_ ## id__, (gt__)->wa_active.oob);		\
 })

From 342d1f84320994ab72a6ca96e18f2985868e1cbd Mon Sep 17 00:00:00 2001
From: Matt Atwood <matthew.s.atwood@intel.com>
Date: Thu, 7 Aug 2025 14:42:24 -0700
Subject: [PATCH 0346/1455] drm/xe: Update function names for GT specific
 workarounds

Now that there distinctly different OOB functions, update the names to
reflect the IPs they interact with.

Signed-off-by: Matt Atwood <matthew.s.atwood@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250807214224.32728-2-matthew.s.atwood@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_gt.c |  4 ++--
 drivers/gpu/drm/xe/xe_wa.c | 10 +++++-----
 drivers/gpu/drm/xe/xe_wa.h |  4 ++--
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 43f2da27d9d9..a3397f04abcc 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -400,7 +400,7 @@ int xe_gt_init_early(struct xe_gt *gt)
 
 	xe_reg_sr_init(&gt->reg_sr, "GT", gt_to_xe(gt));
 
-	err = xe_wa_init(gt);
+	err = xe_wa_gt_init(gt);
 	if (err)
 		return err;
 
@@ -408,7 +408,7 @@ int xe_gt_init_early(struct xe_gt *gt)
 	if (err)
 		return err;
 
-	xe_wa_process_oob(gt);
+	xe_wa_process_gt_oob(gt);
 
 	xe_force_wake_init_gt(gt, gt_to_fw(gt));
 	spin_lock_init(&gt->global_invl_lock);
diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index 0fc73af1bb4c..52c7df4c3afd 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -916,13 +916,13 @@ void xe_wa_process_device_oob(struct xe_device *xe)
 }
 
 /**
- * xe_wa_process_oob - process OOB workaround table
+ * xe_wa_process_gt_oob - process GT OOB workaround table
  * @gt: GT instance to process workarounds for
  *
  * Process OOB workaround table for this platform, marking in @gt the
  * workarounds that are active.
  */
-void xe_wa_process_oob(struct xe_gt *gt)
+void xe_wa_process_gt_oob(struct xe_gt *gt)
 {
 	struct xe_rtp_process_ctx ctx = XE_RTP_PROCESS_CTX_INITIALIZER(gt);
 
@@ -1006,12 +1006,12 @@ int xe_wa_device_init(struct xe_device *xe)
 }
 
 /**
- * xe_wa_init - initialize gt with workaround bookkeeping
+ * xe_wa_gt_init - initialize gt with workaround bookkeeping
  * @gt: GT instance to initialize
  *
  * Returns 0 for success, negative error code otherwise.
  */
-int xe_wa_init(struct xe_gt *gt)
+int xe_wa_gt_init(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 	size_t n_oob, n_lrc, n_engine, n_gt, total;
@@ -1037,7 +1037,7 @@ int xe_wa_init(struct xe_gt *gt)
 
 	return 0;
 }
-ALLOW_ERROR_INJECTION(xe_wa_init, ERRNO); /* See xe_pci_probe() */
+ALLOW_ERROR_INJECTION(xe_wa_gt_init, ERRNO); /* See xe_pci_probe() */
 
 void xe_wa_device_dump(struct xe_device *xe, struct drm_printer *p)
 {
diff --git a/drivers/gpu/drm/xe/xe_wa.h b/drivers/gpu/drm/xe/xe_wa.h
index 105497c1d7d7..6a869b2de643 100644
--- a/drivers/gpu/drm/xe/xe_wa.h
+++ b/drivers/gpu/drm/xe/xe_wa.h
@@ -14,9 +14,9 @@ struct xe_hw_engine;
 struct xe_tile;
 
 int xe_wa_device_init(struct xe_device *xe);
-int xe_wa_init(struct xe_gt *gt);
+int xe_wa_gt_init(struct xe_gt *gt);
 void xe_wa_process_device_oob(struct xe_device *xe);
-void xe_wa_process_oob(struct xe_gt *gt);
+void xe_wa_process_gt_oob(struct xe_gt *gt);
 void xe_wa_process_gt(struct xe_gt *gt);
 void xe_wa_process_engine(struct xe_hw_engine *hwe);
 void xe_wa_process_lrc(struct xe_hw_engine *hwe);

From 9f8aa0bcd1f35baefe083adfa26bc6f9f68f652b Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Fri, 8 Aug 2025 13:06:28 +0530
Subject: [PATCH 0347/1455] drm/xe/vf: Refactor CCS save/restore to use default
 migration context

Previously, CCS save/restore operations created separate migration
contexts with new VM memory allocations, resulting in significant
overhead.

This commit eliminates redundant context creation reusing the default
migration context by registering new execution queues for CCS save and
restore on the existing migrate VM.

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250808073628.32745-2-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/xe_exec_queue.c         | 15 +++++++
 drivers/gpu/drm/xe/xe_exec_queue.h         |  1 +
 drivers/gpu/drm/xe/xe_migrate.c            | 21 +++++----
 drivers/gpu/drm/xe/xe_migrate.h            |  2 +-
 drivers/gpu/drm/xe/xe_pm.c                 |  3 ++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       | 51 ++++++++++++----------
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h |  6 +--
 7 files changed, 63 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 6c176183ed58..2d10a53f701d 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -789,6 +789,21 @@ int xe_exec_queue_get_property_ioctl(struct drm_device *dev, void *data,
 	return ret;
 }
 
+/**
+ * xe_exec_queue_lrc() - Get the LRC from exec queue.
+ * @q: The exec_queue.
+ *
+ * Retrieves the primary LRC for the exec queue. Note that this function
+ * returns only the first LRC instance, even when multiple parallel LRCs
+ * are configured.
+ *
+ * Return: Pointer to LRC on success, error on failure
+ */
+struct xe_lrc *xe_exec_queue_lrc(struct xe_exec_queue *q)
+{
+	return q->lrc[0];
+}
+
 /**
  * xe_exec_queue_is_lr() - Whether an exec_queue is long-running
  * @q: The exec_queue
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h
index 4d416f23001c..15ec852e7f7e 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue.h
@@ -94,4 +94,5 @@ int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch);
 
 void xe_exec_queue_jobs_ring_restore(struct xe_exec_queue *q);
 
+struct xe_lrc *xe_exec_queue_lrc(struct xe_exec_queue *q);
 #endif
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 4effe10d815b..ac9f4d4988d2 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -951,7 +951,7 @@ err_sync:
 }
 
 /**
- * xe_get_migrate_lrc() - Get the LRC from migrate context.
+ * xe_migrate_lrc() - Get the LRC from migrate context.
  * @migrate: Migrate context.
  *
  * Return: Pointer to LRC on success, error on failure
@@ -961,14 +961,15 @@ struct xe_lrc *xe_migrate_lrc(struct xe_migrate *migrate)
 	return migrate->q->lrc[0];
 }
 
-static int emit_flush_invalidate(struct xe_migrate *m, u32 *dw, int i,
+static int emit_flush_invalidate(struct xe_exec_queue *q, u32 *dw, int i,
 				 u32 flags)
 {
+	struct xe_lrc *lrc = xe_exec_queue_lrc(q);
 	dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW |
 		  MI_FLUSH_IMM_DW | flags;
-	dw[i++] = lower_32_bits(xe_lrc_start_seqno_ggtt_addr(xe_migrate_lrc(m))) |
+	dw[i++] = lower_32_bits(xe_lrc_start_seqno_ggtt_addr(lrc)) |
 		  MI_FLUSH_DW_USE_GTT;
-	dw[i++] = upper_32_bits(xe_lrc_start_seqno_ggtt_addr(xe_migrate_lrc(m)));
+	dw[i++] = upper_32_bits(xe_lrc_start_seqno_ggtt_addr(lrc));
 	dw[i++] = MI_NOOP;
 	dw[i++] = MI_NOOP;
 
@@ -977,7 +978,8 @@ static int emit_flush_invalidate(struct xe_migrate *m, u32 *dw, int i,
 
 /**
  * xe_migrate_ccs_rw_copy() - Copy content of TTM resources.
- * @m: The migration context.
+ * @tile: Tile whose migration context to be used.
+ * @q : Execution to be used along with migration context.
  * @src_bo: The buffer object @src is currently bound to.
  * @read_write : Creates BB commands for CCS read/write.
  *
@@ -988,7 +990,7 @@ static int emit_flush_invalidate(struct xe_migrate *m, u32 *dw, int i,
  *
  * Return: 0 if successful, negative error code on failure.
  */
-int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
+int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
 			   struct xe_bo *src_bo,
 			   enum xe_sriov_vf_ccs_rw_ctxs read_write)
 
@@ -996,7 +998,8 @@ int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
 	bool src_is_pltt = read_write == XE_SRIOV_VF_CCS_READ_CTX;
 	bool dst_is_pltt = read_write == XE_SRIOV_VF_CCS_WRITE_CTX;
 	struct ttm_resource *src = src_bo->ttm.resource;
-	struct xe_gt *gt = m->tile->primary_gt;
+	struct xe_migrate *m = tile->migrate;
+	struct xe_gt *gt = tile->primary_gt;
 	u32 batch_size, batch_size_allocated;
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_res_cursor src_it, ccs_it;
@@ -1079,11 +1082,11 @@ int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
 
 		emit_pte(m, bb, ccs_pt, false, false, &ccs_it, ccs_size, src);
 
-		bb->len = emit_flush_invalidate(m, bb->cs, bb->len, flush_flags);
+		bb->len = emit_flush_invalidate(q, bb->cs, bb->len, flush_flags);
 		flush_flags = xe_migrate_ccs_copy(m, bb, src_L0_ofs, src_is_pltt,
 						  src_L0_ofs, dst_is_pltt,
 						  src_L0, ccs_ofs, true);
-		bb->len = emit_flush_invalidate(m, bb->cs, bb->len, flush_flags);
+		bb->len = emit_flush_invalidate(q, bb->cs, bb->len, flush_flags);
 
 		size -= src_L0;
 	}
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index a3f3fa281e04..8978d2cc1a75 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -125,7 +125,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 				  struct ttm_resource *dst,
 				  bool copy_only_ccs);
 
-int xe_migrate_ccs_rw_copy(struct xe_migrate *m,
+int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
 			   struct xe_bo *src_bo,
 			   enum xe_sriov_vf_ccs_rw_ctxs read_write);
 
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index ed29791beb7c..cc489c41ae30 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -209,6 +209,9 @@ int xe_pm_resume(struct xe_device *xe)
 
 	xe_pxp_pm_resume(xe->pxp);
 
+	if (IS_SRIOV_VF(xe))
+		xe_sriov_vf_ccs_register_context(xe);
+
 	drm_dbg(&xe->drm, "Device resumed\n");
 	return 0;
 err:
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index f0ca2a9b2bb7..4872e43eb440 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -8,6 +8,7 @@
 #include "xe_bb.h"
 #include "xe_bo.h"
 #include "xe_device.h"
+#include "xe_exec_queue.h"
 #include "xe_exec_queue_types.h"
 #include "xe_guc_submit.h"
 #include "xe_lrc.h"
@@ -168,8 +169,8 @@ static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx)
 
 static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
 {
-	struct xe_lrc *lrc = xe_migrate_lrc(ctx->migrate);
 	u64 addr = xe_sa_manager_gpu_addr(ctx->mem.ccs_bb_pool);
+	struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
 	u32 dw[10], i = 0;
 
 	dw[i++] = MI_ARB_ON_OFF | MI_ARB_ENABLE;
@@ -183,13 +184,12 @@ static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
 	xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
 }
 
-static int register_save_restore_context(struct xe_migrate *m,
-					 enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
+static int register_save_restore_context(struct xe_tile_vf_ccs *ctx)
 {
 	int err = -EINVAL;
 	int ctx_type;
 
-	switch (ctx_id) {
+	switch (ctx->ctx_id) {
 	case XE_SRIOV_VF_CCS_READ_CTX:
 		ctx_type = GUC_CONTEXT_COMPRESSION_SAVE;
 		break;
@@ -200,7 +200,7 @@ static int register_save_restore_context(struct xe_migrate *m,
 		return err;
 	}
 
-	xe_guc_register_exec_queue(xe_migrate_exec_queue(m), ctx_type);
+	xe_guc_register_exec_queue(ctx->mig_q, ctx_type);
 	return 0;
 }
 
@@ -225,7 +225,7 @@ int xe_sriov_vf_ccs_register_context(struct xe_device *xe)
 
 	for_each_ccs_rw_ctx(ctx_id) {
 		ctx = &tile->sriov.vf.ccs[ctx_id];
-		err = register_save_restore_context(ctx->migrate, ctx_id);
+		err = register_save_restore_context(ctx);
 		if (err)
 			return err;
 	}
@@ -236,13 +236,14 @@ int xe_sriov_vf_ccs_register_context(struct xe_device *xe)
 static void xe_sriov_vf_ccs_fini(void *arg)
 {
 	struct xe_tile_vf_ccs *ctx = arg;
-	struct xe_lrc *lrc = xe_migrate_lrc(ctx->migrate);
+	struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
 
 	/*
 	 * Make TAIL = HEAD in the ring so that no issues are seen if Guc
 	 * submits this context to HW on VF pause after unbinding device.
 	 */
 	xe_lrc_set_ring_tail(lrc, xe_lrc_ring_head(lrc));
+	xe_exec_queue_put(ctx->mig_q);
 }
 
 /**
@@ -258,8 +259,9 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 {
 	struct xe_tile *tile = xe_device_get_root_tile(xe);
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-	struct xe_migrate *migrate;
 	struct xe_tile_vf_ccs *ctx;
+	struct xe_exec_queue *q;
+	u32 flags;
 	int err;
 
 	xe_assert(xe, IS_SRIOV_VF(xe));
@@ -270,37 +272,40 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 		ctx = &tile->sriov.vf.ccs[ctx_id];
 		ctx->ctx_id = ctx_id;
 
-		migrate = xe_migrate_alloc(tile);
-		if (!migrate) {
-			err = -ENOMEM;
+		flags = EXEC_QUEUE_FLAG_KERNEL |
+			EXEC_QUEUE_FLAG_PERMANENT |
+			EXEC_QUEUE_FLAG_MIGRATE;
+		q = xe_exec_queue_create_bind(xe, tile, flags, 0);
+		if (IS_ERR(q)) {
+			err = PTR_ERR(q);
 			goto err_ret;
 		}
-
-		err = xe_migrate_init(migrate);
-		if (err)
-			goto err_ret;
-
-		ctx->migrate = migrate;
+		ctx->mig_q = q;
 
 		err = alloc_bb_pool(tile, ctx);
 		if (err)
-			goto err_ret;
+			goto err_free_queue;
 
 		ccs_rw_update_ring(ctx);
 
-		err = register_save_restore_context(ctx->migrate, ctx_id);
+		err = register_save_restore_context(ctx);
 		if (err)
-			goto err_ret;
+			goto err_free_queue;
 
 		err = devm_add_action_or_reset(xe->drm.dev,
 					       xe_sriov_vf_ccs_fini,
 					       ctx);
+		if (err)
+			goto err_ret;
 	}
 
 	xe->sriov.vf.ccs.initialized = 1;
 
 	return 0;
 
+err_free_queue:
+	xe_exec_queue_put(q);
+
 err_ret:
 	return err;
 }
@@ -319,7 +324,7 @@ int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo)
 {
 	struct xe_device *xe = xe_bo_device(bo);
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-	struct xe_migrate *migrate;
+	struct xe_tile_vf_ccs *ctx;
 	struct xe_tile *tile;
 	struct xe_bb *bb;
 	int err = 0;
@@ -334,8 +339,8 @@ int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo)
 		/* bb should be NULL here. Assert if not NULL */
 		xe_assert(xe, !bb);
 
-		migrate = tile->sriov.vf.ccs[ctx_id].migrate;
-		err = xe_migrate_ccs_rw_copy(migrate, bo, ctx_id);
+		ctx = &tile->sriov.vf.ccs[ctx_id];
+		err = xe_migrate_ccs_rw_copy(tile, ctx->mig_q, bo, ctx_id);
 	}
 	return err;
 }
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
index e240f3fd18af..93435a6f4cb6 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -41,11 +41,11 @@ struct xe_sa_manager;
 struct xe_tile_vf_ccs {
 	/** @id: Id to which context it belongs to */
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-	/** @migrate: Migration helper for save/restore of CCS data */
-	struct xe_migrate *migrate;
+	/** @mig_q: exec queues used for migration */
+	struct xe_exec_queue *mig_q;
 
 	struct {
-		/** @ccs_rw_bb_pool: Pool from which batch buffers are allocated. */
+		/** @ccs_bb_pool: Pool from which batch buffers are allocated. */
 		struct xe_sa_manager *ccs_bb_pool;
 	} mem;
 };

From c4d0f59ded1efbf0004dfdfffb9050caba234993 Mon Sep 17 00:00:00 2001
From: Krzysztof Karas <krzysztof.karas@intel.com>
Date: Wed, 30 Jul 2025 07:39:23 +0000
Subject: [PATCH 0348/1455] drm/i915/selftests: Do not overwrite error code
 after intel_context_migrate_clear() call

Currently this function's error code is stored in err variable,
which, if a i915_request is present, will be immediately
overwritten by return from dma_resv_reserve_fences(). Call DMA
functions only if intel_context_migrate_clear() succeeded.

Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/c0348e052bcb2870acdc15d60f5389fbdf4ef886.1753859971.git.krzysztof.karas@intel.com
---
 drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
index 9c3f17e51885..c94b71a963b2 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
@@ -1228,7 +1228,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 					  i915_gem_object_is_lmem(obj),
 					  expand32(POISON_INUSE), &rq);
 	i915_gem_object_unpin_pages(obj);
-	if (rq) {
+	if (rq && !err) {
 		err = dma_resv_reserve_fences(obj->base.resv, 1);
 		if (!err)
 			dma_resv_add_fence(obj->base.resv, &rq->fence,

From f768ebbba9110846c9f986a96109d70154d60b5d Mon Sep 17 00:00:00 2001
From: Krzysztof Karas <krzysztof.karas@intel.com>
Date: Wed, 30 Jul 2025 07:40:18 +0000
Subject: [PATCH 0349/1455] drm/i915/selftests: Do not leak vm_area_struct on
 early return

This structure may be leaked on early failure paths, so include
vm_munmap() call in them to avoid that.

Suggested-by: Chris Wilson <chris.p.wilson@linux.intel.com>
Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/d8277e3af09c388fa5dfbf5c9bb3985ae0b191a2.1753859971.git.krzysztof.karas@intel.com
---
 .../drm/i915/gem/selftests/i915_gem_mman.c    | 68 +++++++++----------
 1 file changed, 31 insertions(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
index c94b71a963b2..78734c404a6d 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c
@@ -1096,32 +1096,20 @@ static int ___igt_mmap_migrate(struct drm_i915_private *i915,
 			       unsigned long addr,
 			       bool unfaultable)
 {
-	struct vm_area_struct *area;
-	int err = 0, i;
+	int i;
 
 	pr_info("igt_mmap(%s, %d) @ %lx\n",
 		obj->mm.region->name, I915_MMAP_TYPE_FIXED, addr);
 
-	mmap_read_lock(current->mm);
-	area = vma_lookup(current->mm, addr);
-	mmap_read_unlock(current->mm);
-	if (!area) {
-		pr_err("%s: Did not create a vm_area_struct for the mmap\n",
-		       obj->mm.region->name);
-		err = -EINVAL;
-		goto out_unmap;
-	}
-
 	for (i = 0; i < obj->base.size / sizeof(u32); i++) {
 		u32 __user *ux = u64_to_user_ptr((u64)(addr + i * sizeof(*ux)));
 		u32 x;
 
 		if (get_user(x, ux)) {
-			err = -EFAULT;
 			if (!unfaultable) {
 				pr_err("%s: Unable to read from mmap, offset:%zd\n",
 				       obj->mm.region->name, i * sizeof(x));
-				goto out_unmap;
+				return -EFAULT;
 			}
 
 			continue;
@@ -1130,37 +1118,29 @@ static int ___igt_mmap_migrate(struct drm_i915_private *i915,
 		if (unfaultable) {
 			pr_err("%s: Faulted unmappable memory\n",
 			       obj->mm.region->name);
-			err = -EINVAL;
-			goto out_unmap;
+			return -EINVAL;
 		}
 
 		if (x != expand32(POISON_INUSE)) {
 			pr_err("%s: Read incorrect value from mmap, offset:%zd, found:%x, expected:%x\n",
 			       obj->mm.region->name,
 			       i * sizeof(x), x, expand32(POISON_INUSE));
-			err = -EINVAL;
-			goto out_unmap;
+			return -EINVAL;
 		}
 
 		x = expand32(POISON_FREE);
 		if (put_user(x, ux)) {
 			pr_err("%s: Unable to write to mmap, offset:%zd\n",
 			       obj->mm.region->name, i * sizeof(x));
-			err = -EFAULT;
-			goto out_unmap;
+			return -EFAULT;
 		}
 	}
 
-	if (unfaultable) {
-		if (err == -EFAULT)
-			err = 0;
-	} else {
-		obj->flags &= ~I915_BO_ALLOC_GPU_ONLY;
-		err = wc_check(obj);
-	}
-out_unmap:
-	vm_munmap(addr, obj->base.size);
-	return err;
+	if (unfaultable)
+		return 0;
+
+	obj->flags &= ~I915_BO_ALLOC_GPU_ONLY;
+	return wc_check(obj);
 }
 
 #define IGT_MMAP_MIGRATE_TOPDOWN     (1 << 0)
@@ -1176,6 +1156,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 	struct drm_i915_private *i915 = placements[0]->i915;
 	struct drm_i915_gem_object *obj;
 	struct i915_request *rq = NULL;
+	struct vm_area_struct *area;
 	unsigned long addr;
 	LIST_HEAD(objects);
 	u64 offset;
@@ -1207,20 +1188,30 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 		goto out_put;
 	}
 
+	mmap_read_lock(current->mm);
+	area = vma_lookup(current->mm, addr);
+	mmap_read_unlock(current->mm);
+	if (!area) {
+		pr_err("%s: Did not create a vm_area_struct for the mmap\n",
+		       obj->mm.region->name);
+		err = -EINVAL;
+		goto out_addr;
+	}
+
 	if (flags & IGT_MMAP_MIGRATE_FILL) {
 		err = igt_fill_mappable(placements[0], &objects);
 		if (err)
-			goto out_put;
+			goto out_addr;
 	}
 
 	err = i915_gem_object_lock(obj, NULL);
 	if (err)
-		goto out_put;
+		goto out_addr;
 
 	err = i915_gem_object_pin_pages(obj);
 	if (err) {
 		i915_gem_object_unlock(obj);
-		goto out_put;
+		goto out_addr;
 	}
 
 	err = intel_context_migrate_clear(to_gt(i915)->migrate.context, NULL,
@@ -1237,7 +1228,7 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 	}
 	i915_gem_object_unlock(obj);
 	if (err)
-		goto out_put;
+		goto out_addr;
 
 	if (flags & IGT_MMAP_MIGRATE_EVICTABLE)
 		igt_make_evictable(&objects);
@@ -1245,16 +1236,16 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 	if (flags & IGT_MMAP_MIGRATE_FAIL_GPU) {
 		err = i915_gem_object_lock(obj, NULL);
 		if (err)
-			goto out_put;
+			goto out_addr;
 
 		/*
-		 * Ensure we only simulate the gpu failuire when faulting the
+		 * Ensure we only simulate the gpu failure when faulting the
 		 * pages.
 		 */
 		err = i915_gem_object_wait_moving_fence(obj, true);
 		i915_gem_object_unlock(obj);
 		if (err)
-			goto out_put;
+			goto out_addr;
 		i915_ttm_migrate_set_failure_modes(true, false);
 	}
 
@@ -1298,6 +1289,9 @@ static int __igt_mmap_migrate(struct intel_memory_region **placements,
 		}
 	}
 
+out_addr:
+	vm_munmap(addr, obj->base.size);
+
 out_put:
 	i915_gem_object_put(obj);
 	igt_close_objects(i915, &objects);

From 6510b62fe9303aaf48ff136ff69186bcfc32172d Mon Sep 17 00:00:00 2001
From: Seyediman Seyedarab <imandevel@gmail.com>
Date: Thu, 24 Jul 2025 15:59:13 -0400
Subject: [PATCH 0350/1455] drm/nouveau: replace snprintf() with scnprintf() in
 nvkm_snprintbf()

snprintf() returns the number of characters that *would* have been
written, which can overestimate how much you actually wrote to the
buffer in case of truncation. That leads to 'data += this' advancing
the pointer past the end of the buffer and size going negative.

Switching to scnprintf() prevents potential buffer overflows and ensures
consistent behavior when building the output string.

Signed-off-by: Seyediman Seyedarab <ImanDevel@gmail.com>
Link: https://lore.kernel.org/r/20250724195913.60742-1-ImanDevel@gmail.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/nouveau/nvkm/core/enum.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/core/enum.c b/drivers/gpu/drm/nouveau/nvkm/core/enum.c
index b9581feb24cc..a23b40b27b81 100644
--- a/drivers/gpu/drm/nouveau/nvkm/core/enum.c
+++ b/drivers/gpu/drm/nouveau/nvkm/core/enum.c
@@ -44,7 +44,7 @@ nvkm_snprintbf(char *data, int size, const struct nvkm_bitfield *bf, u32 value)
 	bool space = false;
 	while (size >= 1 && bf->name) {
 		if (value & bf->mask) {
-			int this = snprintf(data, size, "%s%s",
+			int this = scnprintf(data, size, "%s%s",
 					    space ? " " : "", bf->name);
 			size -= this;
 			data += this;

From 1bf30a45653d14b0d88b5ca3c72e0ecf030d9c88 Mon Sep 17 00:00:00 2001
From: Philipp Stanner <phasta@kernel.org>
Date: Fri, 1 Aug 2025 09:45:32 +0200
Subject: [PATCH 0351/1455] drm/nouveau: Remove surplus struct member

struct nouveau_channel contains the member 'accel_done' and a forgotten
TODO which hints at that mechanism being removed in the "near future".
Since that variable is read nowhere anymore, this "near future" is now.

Remove the variable and the TODO.

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250801074531.79237-2-phasta@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/nouveau/nouveau_chan.h | 2 --
 drivers/gpu/drm/nouveau/nouveau_dma.h  | 1 -
 2 files changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.h b/drivers/gpu/drm/nouveau/nouveau_chan.h
index 561877725aac..bb34b0a6082d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.h
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.h
@@ -31,8 +31,6 @@ struct nouveau_channel {
 		u64 addr;
 	} push;
 
-	/* TODO: this will be reworked in the near future */
-	bool accel_done;
 	void *fence;
 	struct {
 		int max;
diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.h b/drivers/gpu/drm/nouveau/nouveau_dma.h
index 0e27b76d1e1c..c25ef9a54b9f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dma.h
+++ b/drivers/gpu/drm/nouveau/nouveau_dma.h
@@ -90,7 +90,6 @@ FIRE_RING(struct nouveau_channel *chan)
 {
 	if (chan->dma.cur == chan->dma.put)
 		return;
-	chan->accel_done = true;
 
 	WRITE_PUT(chan->dma.cur);
 

From 317be9c64215f264740c6dffb0a44216960875b2 Mon Sep 17 00:00:00 2001
From: Krzysztof Karas <krzysztof.karas@intel.com>
Date: Tue, 5 Aug 2025 11:41:42 +0000
Subject: [PATCH 0352/1455] drm/i915/gt: Protect against overflow in
 active_engine()

It is unlikely, but possible for the first call to
intel_context_create() to fail with -ENOMEM, which would result
in entering the following code block and decrementing "count",
when it is set to 0 (initial condition in the for loop).

Protect from overflowing the variable by checking for 0 value
of "count" before entering the loop.

Signed-off-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/pogr74jktdqehrfap4tjky23ees4x7erh5dwgg5jb2n522cfkw@kpnxe4qzx4pj
---
 drivers/gpu/drm/i915/gt/selftest_hangcheck.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c
index f057c16410e7..619c70c54ef9 100644
--- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c
+++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c
@@ -904,6 +904,8 @@ static void active_engine(struct kthread_work *work)
 			arg->result = PTR_ERR(ce[count]);
 			pr_err("[%s] Create context #%ld failed: %d!\n",
 			       engine->name, count, arg->result);
+			if (!count)
+				return;
 			while (--count)
 				intel_context_put(ce[count]);
 			return;

From d301eb950da59f962bafe874cf5eb6d61a85b2c2 Mon Sep 17 00:00:00 2001
From: Karthik Poosa <karthik.poosa@intel.com>
Date: Sat, 9 Aug 2025 00:23:10 +0530
Subject: [PATCH 0353/1455] drm/xe/hwmon: Add SW clamp for power limits writes

Clamp writes to power limits powerX_crit/currX_crit, powerX_cap,
powerX_max, to the maximum supported by the pcode mailbox
when sysfs-provided values exceed this limit.
Although the pcode already performs clamping, values beyond the pcode
mailbox's supported range get truncated, leading to incorrect
critical power settings.
This patch ensures proper clamping to prevent such truncation.

v2:
 - Address below review comments. (Riana)
 - Split comments into multiple sentences.
 - Use local variables for readability.
 - Add a debug log.
 - Use u64 instead of unsigned long.

v3:
 - Change drm_dbg logs to drm_info. (Badal)

v4:
 - Rephrase the drm_info log. (Rodrigo, Riana)
 - Rename variable max_mbx_power_limit to max_supp_power_limit, as
   limit is same for platforms with and without mailbox power limit
   support.

Signed-off-by: Karthik Poosa <karthik.poosa@intel.com>
Fixes: 92d44a422d0d ("drm/xe/hwmon: Expose card reactive critical power")
Fixes: fb1b70607f73 ("drm/xe/hwmon: Expose power attributes")
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250808185310.3466529-1-karthik.poosa@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_hwmon.c | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c
index f08fc4377d25..c17ed1ae8649 100644
--- a/drivers/gpu/drm/xe/xe_hwmon.c
+++ b/drivers/gpu/drm/xe/xe_hwmon.c
@@ -332,6 +332,7 @@ static int xe_hwmon_power_max_write(struct xe_hwmon *hwmon, u32 attr, int channe
 	int ret = 0;
 	u32 reg_val, max;
 	struct xe_reg rapl_limit;
+	u64 max_supp_power_limit = 0;
 
 	mutex_lock(&hwmon->hwmon_lock);
 
@@ -356,6 +357,20 @@ static int xe_hwmon_power_max_write(struct xe_hwmon *hwmon, u32 attr, int channe
 		goto unlock;
 	}
 
+	/*
+	 * If the sysfs value exceeds the maximum pcode supported power limit value, clamp it to
+	 * the supported maximum (U12.3 format).
+	 * This is to avoid truncation during reg_val calculation below and ensure the valid
+	 * power limit is sent for pcode which would clamp it to card-supported value.
+	 */
+	max_supp_power_limit = ((PWR_LIM_VAL) >> hwmon->scl_shift_power) * SF_POWER;
+	if (value > max_supp_power_limit) {
+		value = max_supp_power_limit;
+		drm_info(&hwmon->xe->drm,
+			 "Power limit clamped as selected %s exceeds channel %d limit\n",
+			 PWR_ATTR_TO_STR(attr), channel);
+	}
+
 	/* Computation in 64-bits to avoid overflow. Round to nearest. */
 	reg_val = DIV_ROUND_CLOSEST_ULL((u64)value << hwmon->scl_shift_power, SF_POWER);
 
@@ -739,9 +754,23 @@ static int xe_hwmon_power_curr_crit_write(struct xe_hwmon *hwmon, int channel,
 {
 	int ret;
 	u32 uval;
+	u64 max_crit_power_curr = 0;
 
 	mutex_lock(&hwmon->hwmon_lock);
 
+	/*
+	 * If the sysfs value exceeds the pcode mailbox cmd POWER_SETUP_SUBCOMMAND_WRITE_I1
+	 * max supported value, clamp it to the command's max (U10.6 format).
+	 * This is to avoid truncation during uval calculation below and ensure the valid power
+	 * limit is sent for pcode which would clamp it to card-supported value.
+	 */
+	max_crit_power_curr = (POWER_SETUP_I1_DATA_MASK >> POWER_SETUP_I1_SHIFT) * scale_factor;
+	if (value > max_crit_power_curr) {
+		value = max_crit_power_curr;
+		drm_info(&hwmon->xe->drm,
+			 "Power limit clamped as selected exceeds channel %d limit\n",
+			 channel);
+	}
 	uval = DIV_ROUND_CLOSEST_ULL(value << POWER_SETUP_I1_SHIFT, scale_factor);
 	ret = xe_hwmon_pcode_write_i1(hwmon, uval);
 

From edb660ad79ffe81a982c2eca02360a6ffac83e46 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Fri, 8 Aug 2025 10:41:08 -0700
Subject: [PATCH 0354/1455] drm/intel/pciids: Add match on vendor/id only

All our PCI ID macros match on the PCI class besides the vendor and
devid, even for devices that may or may not have display. This may not
work going forward, so add a simple INTEL_PCI_DEVICE that matches only
on vendor/device IDs.

Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/20250808-intel-pci-device-v1-1-ce3545d86502@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 include/drm/intel/pciids.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/drm/intel/pciids.h b/include/drm/intel/pciids.h
index 76f8d26f9cc9..da6301a6fcea 100644
--- a/include/drm/intel/pciids.h
+++ b/include/drm/intel/pciids.h
@@ -26,6 +26,11 @@
 #define __PCIIDS_H__
 
 #ifdef __KERNEL__
+#define INTEL_PCI_DEVICE(_id, _info) { \
+	PCI_DEVICE(PCI_VENDOR_ID_INTEL, (_id)), \
+	.driver_data = (kernel_ulong_t)(_info), \
+}
+
 #define INTEL_VGA_DEVICE(_id, _info) { \
 	PCI_DEVICE(PCI_VENDOR_ID_INTEL, (_id)), \
 	.class = PCI_BASE_CLASS_DISPLAY << 16, .class_mask = 0xff << 16, \

From 1dd2fa0e00f17b909ff0fab92c1c48c2d7911c15 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Sat, 28 Jun 2025 10:50:25 +0530
Subject: [PATCH 0355/1455] drm/amdgpu: Save and restore switch state

During a DPC error kernel waits for the link to be active before
notifying downstream devices. On certain platforms with Broadcom switch
in synthetiic mode, switch responds with values even though the link is
not fully ready. The config space restoration done by pcie port driver
for SWUS/DS of dGPU is thus not effective as the switch is still doing
internal enumeration.

As a workaround, save state of SWUS/DS device in driver. Add additional
check to see if link is active and restore the values during DPC error
callbacks.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        |  3 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 85 ++++++++++++++++++++--
 2 files changed, 83 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index c2771e32d0bf..ddd472e56f69 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -910,6 +910,9 @@ struct amdgpu_pcie_reset_ctx {
 	bool in_link_reset;
 	bool occurs_dpc;
 	bool audio_suspended;
+	struct pci_dev *swus;
+	struct pci_saved_state *swus_pcistate;
+	struct pci_saved_state *swds_pcistate;
 };
 
 /*
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 5eed5c500190..9c373bdadb39 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -178,6 +178,8 @@ struct amdgpu_init_level amdgpu_init_minimal_xgmi = {
 		BIT(AMD_IP_BLOCK_TYPE_PSP)
 };
 
+static void amdgpu_device_load_switch_state(struct amdgpu_device *adev);
+
 static inline bool amdgpu_ip_member_of_hwini(struct amdgpu_device *adev,
 					     enum amd_ip_block_type block)
 {
@@ -5013,7 +5015,8 @@ void amdgpu_device_fini_sw(struct amdgpu_device *adev)
 	adev->reset_domain = NULL;
 
 	kfree(adev->pci_state);
-
+	kfree(adev->pcie_reset_ctx.swds_pcistate);
+	kfree(adev->pcie_reset_ctx.swus_pcistate);
 }
 
 /**
@@ -6986,16 +6989,34 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
 	struct amdgpu_device *tmp_adev;
 	struct amdgpu_hive_info *hive;
 	struct list_head device_list;
-	int r = 0, i;
+	struct pci_dev *link_dev;
+	int r = 0, i, timeout;
 	u32 memsize;
+	u16 status;
 
 	dev_info(adev->dev, "PCI error: slot reset callback!!\n");
 
 	memset(&reset_context, 0, sizeof(reset_context));
 
-	/* wait for asic to come out of reset */
-	msleep(700);
+	if (adev->pcie_reset_ctx.swus)
+		link_dev = adev->pcie_reset_ctx.swus;
+	else
+		link_dev = adev->pdev;
+	/* wait for asic to come out of reset, timeout = 10s */
+	timeout = 10000;
+	do {
+		usleep_range(10000, 10500);
+		r = pci_read_config_word(link_dev, PCI_VENDOR_ID, &status);
+		timeout -= 10;
+	} while (timeout > 0 && (status != PCI_VENDOR_ID_ATI) &&
+		 (status != PCI_VENDOR_ID_AMD));
 
+	if ((status != PCI_VENDOR_ID_ATI) && (status != PCI_VENDOR_ID_AMD)) {
+		r = -ETIME;
+		goto out;
+	}
+
+	amdgpu_device_load_switch_state(adev);
 	/* Restore PCI confspace */
 	amdgpu_device_load_pci_state(pdev);
 
@@ -7097,6 +7118,58 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
 	}
 }
 
+static void amdgpu_device_cache_switch_state(struct amdgpu_device *adev)
+{
+	struct pci_dev *parent = pci_upstream_bridge(adev->pdev);
+	int r;
+
+	if (parent->vendor != PCI_VENDOR_ID_ATI)
+		return;
+
+	/* If already saved, return */
+	if (adev->pcie_reset_ctx.swus)
+		return;
+	/* Upstream bridge is ATI, assume it's SWUS/DS architecture */
+	r = pci_save_state(parent);
+	if (r)
+		return;
+	adev->pcie_reset_ctx.swds_pcistate = pci_store_saved_state(parent);
+
+	parent = pci_upstream_bridge(parent);
+	r = pci_save_state(parent);
+	if (r)
+		return;
+	adev->pcie_reset_ctx.swus_pcistate = pci_store_saved_state(parent);
+
+	adev->pcie_reset_ctx.swus = parent;
+}
+
+static void amdgpu_device_load_switch_state(struct amdgpu_device *adev)
+{
+	struct pci_dev *pdev;
+	int r;
+
+	if (!adev->pcie_reset_ctx.swds_pcistate ||
+	    !adev->pcie_reset_ctx.swus_pcistate)
+		return;
+
+	pdev = adev->pcie_reset_ctx.swus;
+	r = pci_load_saved_state(pdev, adev->pcie_reset_ctx.swus_pcistate);
+	if (!r) {
+		pci_restore_state(pdev);
+	} else {
+		dev_warn(adev->dev, "Failed to load SWUS state, err:%d\n", r);
+		return;
+	}
+
+	pdev = pci_upstream_bridge(adev->pdev);
+	r = pci_load_saved_state(pdev, adev->pcie_reset_ctx.swds_pcistate);
+	if (!r)
+		pci_restore_state(pdev);
+	else
+		dev_warn(adev->dev, "Failed to load SWDS state, err:%d\n", r);
+}
+
 bool amdgpu_device_cache_pci_state(struct pci_dev *pdev)
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
@@ -7121,6 +7194,8 @@ bool amdgpu_device_cache_pci_state(struct pci_dev *pdev)
 		return false;
 	}
 
+	amdgpu_device_cache_switch_state(adev);
+
 	return true;
 }
 
@@ -7556,4 +7631,4 @@ u64 amdgpu_device_get_uid(struct amdgpu_uid *uid_info,
 	}
 
 	return uid_info->uid[type][inst];
-}
\ No newline at end of file
+}

From c31f486bc8dd6f481adcb9cca4a6e1837b8cf127 Mon Sep 17 00:00:00 2001
From: Vitaly Prosyak <vitaly.prosyak@amd.com>
Date: Thu, 7 Aug 2025 16:37:25 -0400
Subject: [PATCH 0356/1455] drm/amdgpu: add to custom amdgpu_drm_release
 drm_dev_enter/exit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

User queues are disabled before GEM objects are released
(protecting against user app crashes).
No races with PCI hot-unplug (because drm_dev_enter prevents cleanup
if iewdevice is being removed).

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Vitaly Prosyak <vitaly.prosyak@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 141ba1f100d4..17f754d1135d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2937,11 +2937,14 @@ static int amdgpu_drm_release(struct inode *inode, struct file *filp)
 {
 	struct drm_file *file_priv = filp->private_data;
 	struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
+	struct drm_device *dev = file_priv->minor->dev;
+	int idx;
 
-	if (fpriv) {
+	if (fpriv && drm_dev_enter(dev, &idx)) {
 		fpriv->evf_mgr.fd_closing = true;
 		amdgpu_eviction_fence_destroy(&fpriv->evf_mgr);
 		amdgpu_userq_mgr_fini(&fpriv->userq_mgr);
+		drm_dev_exit(idx);
 	}
 
 	return drm_release(inode, filp);

From e7496c15d830689cc4fc666b976c845ed2c5ed28 Mon Sep 17 00:00:00 2001
From: TungYu Lu <tungyu.lu@amd.com>
Date: Tue, 15 Jul 2025 16:56:59 +0800
Subject: [PATCH 0357/1455] drm/amd/display: Wait until OTG enable state is
 cleared

[Why]
Customer reported an issue that OS starts and stops device multiple times
during driver installation. Frequently disabling and enabling OTG may
prevent OTG from being safely disabled and cause incorrect configuration
upon the next enablement.

[How]
Add a wait until OTG_CURRENT_MASTER_EN_STATE is cleared as a short term
solution.

Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: TungYu Lu <tungyu.lu@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c
index ff79c38287df..5af13706e601 100644
--- a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c
@@ -226,6 +226,11 @@ bool optc401_disable_crtc(struct timing_generator *optc)
 	REG_UPDATE(CONTROL,
 			VTG0_ENABLE, 0);
 
+	// wait until CRTC_CURRENT_MASTER_EN_STATE == 0
+	REG_WAIT(OTG_CONTROL,
+			 OTG_CURRENT_MASTER_EN_STATE,
+			 0, 10, 15000);
+
 	/* CRTC disabled, so disable  clock. */
 	REG_WAIT(OTG_CLOCK_CONTROL,
 			OTG_BUSY, 0,

From ad335b5fc9ed1cdeb33fbe97d2969b3a2eedaf3e Mon Sep 17 00:00:00 2001
From: Danny Wang <Danny.Wang@amd.com>
Date: Thu, 24 Jul 2025 13:58:21 +0800
Subject: [PATCH 0358/1455] drm/amd/display: Reset
 apply_eamless_boot_optimization when dpms_off
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[WHY&HOW]
The user closed the lid while the system was powering on and opened it
again before the “apply_seamless_boot_optimization” was set to false,
resulting in the eDP remaining blank.
Reset the “apply_seamless_boot_optimization” to false when dpms off.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Danny Wang <Danny.Wang@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 242bcb30dd34..5bf97d8fb34a 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3391,7 +3391,7 @@ static void update_seamless_boot_flags(struct dc *dc,
 		int surface_count,
 		struct dc_stream_state *stream)
 {
-	if (get_seamless_boot_stream_count(context) > 0 && surface_count > 0) {
+	if (get_seamless_boot_stream_count(context) > 0 && (surface_count > 0 || stream->dpms_off)) {
 		/* Optimize seamless boot flag keeps clocks and watermarks high until
 		 * first flip. After first flip, optimization is required to lower
 		 * bandwidth. Important to note that it is expected UEFI will

From 7552bee9dd1f2f60ad0fb286b346ba6809e56f07 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Sun, 3 Aug 2025 16:12:50 -0400
Subject: [PATCH 0359/1455] drm/amd/display: [FW Promotion] Release 0.1.22.0

Add a new command for Panel Replay.

Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 69 +++++++++++++++++++
 1 file changed, 69 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 79b5b1bb9b93..52295efdba63 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -4019,6 +4019,10 @@ enum dmub_cmd_replay_type {
 	 * Set adaptive sync sdp enabled
 	 */
 	DMUB_CMD__REPLAY_DISABLED_ADAPTIVE_SYNC_SDP = 8,
+	/**
+	 * Set version
+	 */
+	DMUB_CMD__REPLAY_SET_VERSION = 9,
 	/**
 	 * Set Replay General command.
 	 */
@@ -4143,6 +4147,63 @@ struct dmub_cmd_replay_copy_settings_data {
 	uint8_t pad[2];
 };
 
+
+/**
+ * Replay versions.
+ */
+enum replay_version {
+	/**
+	 * FreeSync Replay
+	 */
+	REPLAY_VERSION_FREESYNC_REPLAY	= 0,
+	/**
+	 * Panel Replay
+	 */
+	REPLAY_VERSION_PANEL_REPLAY		= 1,
+	/**
+	 * Replay not supported.
+	 */
+	REPLAY_VERSION_UNSUPPORTED		= 0xFF,
+};
+
+/**
+ * Data passed from driver to FW in a DMUB_CMD___SET_REPLAY_VERSION command.
+ */
+struct dmub_cmd_replay_set_version_data {
+	/**
+	 * PSR version that FW should implement.
+	 */
+	enum replay_version version;
+	/**
+	 * PSR control version.
+	 */
+	uint8_t cmd_version;
+	/**
+	 * Panel Instance.
+	 * Panel instance to identify which psr_state to use
+	 * Currently the support is only for 0 or 1
+	 */
+	uint8_t panel_inst;
+	/**
+	 * Explicit padding to 4 byte boundary.
+	 */
+	uint8_t pad[2];
+};
+
+/**
+ * Definition of a DMUB_CMD__REPLAY_SET_VERSION command.
+ */
+struct dmub_rb_cmd_replay_set_version {
+	/**
+	 * Command header.
+	 */
+	struct dmub_cmd_header header;
+	/**
+	 * Data passed from driver to FW in a DMUB_CMD__REPLAY_SET_VERSION command.
+	 */
+	struct dmub_cmd_replay_set_version_data replay_set_version_data;
+};
+
 /**
  * Definition of a DMUB_CMD__REPLAY_COPY_SETTINGS command.
  */
@@ -4506,6 +4567,10 @@ union dmub_replay_cmd_set {
 	 * Definition of DMUB_CMD__REPLAY_DISABLED_ADAPTIVE_SYNC_SDP command data.
 	 */
 	struct dmub_cmd_replay_disabled_adaptive_sync_sdp_data disabled_adaptive_sync_sdp_data;
+	/**
+	 * Definition of DMUB_CMD__REPLAY_SET_VERSION command data.
+	 */
+	struct dmub_cmd_replay_set_version_data version_data;
 	/**
 	 * Definition of DMUB_CMD__REPLAY_SET_GENERAL_CMD command data.
 	 */
@@ -6265,6 +6330,10 @@ union dmub_rb_cmd {
 	 * Definition of a DMUB_CMD__IDLE_OPT_SET_DC_POWER_STATE command.
 	 */
 	struct dmub_rb_cmd_idle_opt_set_dc_power_state idle_opt_set_dc_power_state;
+	/**
+	 * Definition of a DMUB_CMD__REPLAY_SET_VERSION command.
+	 */
+	struct dmub_rb_cmd_replay_set_version replay_set_version;
 	/*
 	 * Definition of a DMUB_CMD__REPLAY_COPY_SETTINGS command.
 	 */

From 8d6593c192de6e51102ca7d797756a5003f08cd3 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Sun, 3 Aug 2025 18:38:31 -0500
Subject: [PATCH 0360/1455] drm/amd/display: Promote DC to 3.2.345

This version brings along following update:
-Fix close and open lid may cause eDP remaining blank
-Fix frequently disabling/enabling OTG may cause incorrect
 configuration of OTG

Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 18c68e531f71..1ab05eabbddb 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.344"
+#define DC_VER "3.2.345"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 388b68aef7319cfee7511b99826c46e951e2f1bd Mon Sep 17 00:00:00 2001
From: Cryolitia PukNgae <cryolitia@uniontech.com>
Date: Wed, 6 Aug 2025 11:34:18 +0800
Subject: [PATCH 0361/1455] drm/amdgpu: fix incorrect comment format
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Comments should not have a leading plus sign.

Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Cryolitia PukNgae <cryolitia@uniontech.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
index d5002ff931d8..860bc5cb03c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c
@@ -151,9 +151,9 @@ static void nbio_v7_4_sdma_doorbell_range(struct amdgpu_device *adev, int instan
 		 *   BIF_SDMA0_DOORBELL_RANGE:  0x3bc0
 		 *   BIF_SDMA1_DOORBELL_RANGE:  0x3bc4
 		 *   BIF_SDMA2_DOORBELL_RANGE:  0x3bd8
-+		 *   BIF_SDMA4_DOORBELL_RANGE:
-+		 *     ARCTURUS:  0x3be0
-+		 *     ALDEBARAN: 0x3be4
+		 *   BIF_SDMA4_DOORBELL_RANGE:
+		 *     ARCTURUS:  0x3be0
+		 *     ALDEBARAN: 0x3be4
 		 */
 		if (adev->asic_type == CHIP_ALDEBARAN && instance == 4)
 			reg = instance + 0x4 + 0x1 +

From 260dcf5b06d519bcf27a5dfdb5c626821a55c170 Mon Sep 17 00:00:00 2001
From: Brahmajit Das <listout@listout.xyz>
Date: Mon, 11 Aug 2025 14:51:25 +0530
Subject: [PATCH 0362/1455] drm/radeon/r600_cs: clean up of dead code in
 r600_cs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

GCC 16 enables -Werror=unused-but-set-variable= which results in build
error with the following message.

drivers/gpu/drm/radeon/r600_cs.c: In function ‘r600_texture_size’:
drivers/gpu/drm/radeon/r600_cs.c:1411:29: error: variable ‘level’ set but not used [-Werror=unused-but-set-variable=]
 1411 |         unsigned offset, i, level;
      |                             ^~~~~
cc1: all warnings being treated as errors
make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/radeon/r600_cs.o] Error 1

level although is set, but in never used in the function
r600_texture_size. Thus resulting in dead code and this error getting
triggered.

Fixes: 60b212f8ddcd ("drm/radeon: overhaul texture checking. (v3)")
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Brahmajit Das <listout@listout.xyz>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/r600_cs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index ac77d1246b94..811265648a58 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -1408,7 +1408,7 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned llevel,
 			      unsigned block_align, unsigned height_align, unsigned base_align,
 			      unsigned *l0_size, unsigned *mipmap_size)
 {
-	unsigned offset, i, level;
+	unsigned offset, i;
 	unsigned width, height, depth, size;
 	unsigned blocksize;
 	unsigned nbx, nby;
@@ -1420,7 +1420,7 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned llevel,
 	w0 = r600_mip_minify(w0, 0);
 	h0 = r600_mip_minify(h0, 0);
 	d0 = r600_mip_minify(d0, 0);
-	for(i = 0, offset = 0, level = blevel; i < nlevels; i++, level++) {
+	for (i = 0, offset = 0; i < nlevels; i++) {
 		width = r600_mip_minify(w0, i);
 		nbx = r600_fmt_get_nblocksx(format, width);
 

From fb6c2d16a46e6366cc154ade97dd466a9f595537 Mon Sep 17 00:00:00 2001
From: Ruben Wauters <rubenru09@aol.com>
Date: Tue, 22 Jul 2025 00:22:03 +0100
Subject: [PATCH 0363/1455] drm/gud: Remove unnecessary logging

The debug logging in gud_disconnect() adds zero detail and is
unnecessary, as it only prints the function name.

The same functionality can be achieved by using ftrace, and is
highlighted by checkpatch, stating the same.

This patch removes the debug log in the gud_disconnect() function.

Signed-off-by: Ruben Wauters <rubenru09@aol.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250721232210.12578-1-rubenru09@aol.com
---
 drivers/gpu/drm/gud/gud_drv.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/gud/gud_drv.c b/drivers/gpu/drm/gud/gud_drv.c
index 5385a2126e45..b52a12cbba3e 100644
--- a/drivers/gpu/drm/gud/gud_drv.c
+++ b/drivers/gpu/drm/gud/gud_drv.c
@@ -620,8 +620,6 @@ static void gud_disconnect(struct usb_interface *interface)
 	struct gud_device *gdrm = usb_get_intfdata(interface);
 	struct drm_device *drm = &gdrm->drm;
 
-	drm_dbg(drm, "%s:\n", __func__);
-
 	drm_kms_helper_poll_fini(drm);
 	drm_dev_unplug(drm);
 	drm_atomic_helper_shutdown(drm);

From 17593a69b75f098280ad88b625f2d8c5bfe4c6a1 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Fri, 8 Aug 2025 11:34:56 +0100
Subject: [PATCH 0364/1455] drm/xe: rework PDE PAT index selection

For non-leaf paging structures we end up selecting a random index
between [0, 3], depending on the first user if the page-table is shared,
since non-leaf structures only have two bits in the HW for encoding the
PAT index, and here we are just passing along the full user provided
index, which can be an index as large as ~31 on xe2+. The user provided
index is meant for the leaf node, which maps the actual BO pages where
we have more PAT bits, and not the non-leaf nodes which are only mapping
other paging structures, and so only needs a minimal PAT index range.
Also the chosen index might need to consider how the driver mapped the
paging structures on the host side, like wc vs wb, which is separate
from the user provided index.

With that move the PDE PAT index selection under driver control. For now
just use a coherent index on platforms with page-tables that are cached
on host side, and incoherent otherwise. Using a coherent index could
potentially be expensive, and would be overkill if we know the page-table
is always uncached on host side.

v2 (Stuart):
  - Add some documentation and split into separate helper.

BSpec: 59510
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250808103455.462424-2-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c  | 10 ++++------
 drivers/gpu/drm/xe/xe_pt.c       |  4 ++--
 drivers/gpu/drm/xe/xe_pt_types.h |  3 +--
 drivers/gpu/drm/xe/xe_vm.c       | 34 +++++++++++++++++++++++++++-----
 4 files changed, 36 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index ac9f4d4988d2..45cde272d45f 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -152,8 +152,7 @@ static void xe_migrate_program_identity(struct xe_device *xe, struct xe_vm *vm,
 	for (pos = dpa_base; pos < vram_limit;
 	     pos += SZ_1G, ofs += 8) {
 		if (pos + SZ_1G >= vram_limit) {
-			entry = vm->pt_ops->pde_encode_bo(bo, pt_2m_ofs,
-							  pat_index);
+			entry = vm->pt_ops->pde_encode_bo(bo, pt_2m_ofs);
 			xe_map_wr(xe, &bo->vmap, ofs, u64, entry);
 
 			flags = vm->pt_ops->pte_encode_addr(xe, 0,
@@ -207,7 +206,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 
 	/* PT30 & PT31 reserved for 2M identity map */
 	pt29_ofs = xe_bo_size(bo) - 3 * XE_PAGE_SIZE;
-	entry = vm->pt_ops->pde_encode_bo(bo, pt29_ofs, pat_index);
+	entry = vm->pt_ops->pde_encode_bo(bo, pt29_ofs);
 	xe_pt_write(xe, &vm->pt_root[id]->bo->vmap, 0, entry);
 
 	map_ofs = (num_entries - num_setup) * XE_PAGE_SIZE;
@@ -275,15 +274,14 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 			flags = XE_PDE_64K;
 
 		entry = vm->pt_ops->pde_encode_bo(bo, map_ofs + (u64)(level - 1) *
-						  XE_PAGE_SIZE, pat_index);
+						  XE_PAGE_SIZE);
 		xe_map_wr(xe, &bo->vmap, map_ofs + XE_PAGE_SIZE * level, u64,
 			  entry | flags);
 	}
 
 	/* Write PDE's that point to our BO. */
 	for (i = 0; i < map_ofs / PAGE_SIZE; i++) {
-		entry = vm->pt_ops->pde_encode_bo(bo, (u64)i * XE_PAGE_SIZE,
-						  pat_index);
+		entry = vm->pt_ops->pde_encode_bo(bo, (u64)i * XE_PAGE_SIZE);
 
 		xe_map_wr(xe, &bo->vmap, map_ofs + XE_PAGE_SIZE +
 			  (i + 1) * 8, u64, entry);
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 330cc0f54a3f..f3a39e734a90 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -69,7 +69,7 @@ static u64 __xe_pt_empty_pte(struct xe_tile *tile, struct xe_vm *vm,
 
 	if (level > MAX_HUGEPTE_LEVEL)
 		return vm->pt_ops->pde_encode_bo(vm->scratch_pt[id][level - 1]->bo,
-						 0, pat_index);
+						 0);
 
 	return vm->pt_ops->pte_encode_addr(xe, 0, pat_index, level, IS_DGFX(xe), 0) |
 		XE_PTE_NULL;
@@ -616,7 +616,7 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset,
 			xe_child->is_compact = true;
 		}
 
-		pte = vm->pt_ops->pde_encode_bo(xe_child->bo, 0, pat_index) | flags;
+		pte = vm->pt_ops->pde_encode_bo(xe_child->bo, 0) | flags;
 		ret = xe_pt_insert_entry(xe_walk, xe_parent, offset, xe_child,
 					 pte);
 	}
diff --git a/drivers/gpu/drm/xe/xe_pt_types.h b/drivers/gpu/drm/xe/xe_pt_types.h
index 69eab6f37cfe..17cdd7c7e9f5 100644
--- a/drivers/gpu/drm/xe/xe_pt_types.h
+++ b/drivers/gpu/drm/xe/xe_pt_types.h
@@ -45,8 +45,7 @@ struct xe_pt_ops {
 	u64 (*pte_encode_addr)(struct xe_device *xe, u64 addr,
 			       u16 pat_index,
 			       u32 pt_level, bool devmem, u64 flags);
-	u64 (*pde_encode_bo)(struct xe_bo *bo, u64 bo_offset,
-			     u16 pat_index);
+	u64 (*pde_encode_bo)(struct xe_bo *bo, u64 bo_offset);
 };
 
 struct xe_pt_entry {
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 148a2425006f..97913f496a46 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1512,14 +1512,39 @@ static u64 pte_encode_ps(u32 pt_level)
 	return 0;
 }
 
-static u64 xelp_pde_encode_bo(struct xe_bo *bo, u64 bo_offset,
-			      const u16 pat_index)
+static u16 pde_pat_index(struct xe_bo *bo)
+{
+	struct xe_device *xe = xe_bo_device(bo);
+	u16 pat_index;
+
+	/*
+	 * We only have two bits to encode the PAT index in non-leaf nodes, but
+	 * these only point to other paging structures so we only need a minimal
+	 * selection of options. The user PAT index is only for encoding leaf
+	 * nodes, where we have use of more bits to do the encoding. The
+	 * non-leaf nodes are instead under driver control so the chosen index
+	 * here should be distict from the user PAT index. Also the
+	 * corresponding coherency of the PAT index should be tied to the
+	 * allocation type of the page table (or at least we should pick
+	 * something which is always safe).
+	 */
+	if (!xe_bo_is_vram(bo) && bo->ttm.ttm->caching == ttm_cached)
+		pat_index = xe->pat.idx[XE_CACHE_WB];
+	else
+		pat_index = xe->pat.idx[XE_CACHE_NONE];
+
+	xe_assert(xe, pat_index <= 3);
+
+	return pat_index;
+}
+
+static u64 xelp_pde_encode_bo(struct xe_bo *bo, u64 bo_offset)
 {
 	u64 pde;
 
 	pde = xe_bo_addr(bo, bo_offset, XE_PAGE_SIZE);
 	pde |= XE_PAGE_PRESENT | XE_PAGE_RW;
-	pde |= pde_encode_pat_index(pat_index);
+	pde |= pde_encode_pat_index(pde_pat_index(bo));
 
 	return pde;
 }
@@ -2024,8 +2049,7 @@ struct xe_vm *xe_vm_lookup(struct xe_file *xef, u32 id)
 
 u64 xe_vm_pdp4_descriptor(struct xe_vm *vm, struct xe_tile *tile)
 {
-	return vm->pt_ops->pde_encode_bo(vm->pt_root[tile->id]->bo, 0,
-					 tile_to_xe(tile)->pat.idx[XE_CACHE_WB]);
+	return vm->pt_ops->pde_encode_bo(vm->pt_root[tile->id]->bo, 0);
 }
 
 static struct xe_exec_queue *

From c585c1b29a52e4c556565fc7c4bb767781894b57 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 27 Jun 2025 17:24:08 +0200
Subject: [PATCH 0365/1455] drm/repaper: Do not access GEM-DMA vaddr directly

Use DRM's shadow-plane helper to map and access the GEM object's buffer
within kernel address space. Encapsulates the vmap logic in the GEM-DMA
helpers.

The repaper driver currently reads the vaddr field from the GME buffer
object directly. This only works because GEM code 'automagically' sets
vaddr.

Shadow-plane helpers perform the same steps, but with correct abstraction
behind drm_gem_vmap(). The shadow-plane state provides the buffer address
in kernel address space and the format-conversion state.

v2:
- fix typo in commit description

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250627152422.8399-1-tzimmermann@suse.de
---
 drivers/gpu/drm/tiny/repaper.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/tiny/repaper.c b/drivers/gpu/drm/tiny/repaper.c
index 5c3b51eb0a97..4824f863fdba 100644
--- a/drivers/gpu/drm/tiny/repaper.c
+++ b/drivers/gpu/drm/tiny/repaper.c
@@ -510,13 +510,12 @@ static void repaper_get_temperature(struct repaper_epd *epd)
 	epd->factored_stage_time = epd->stage_time * factor10x / 10;
 }
 
-static int repaper_fb_dirty(struct drm_framebuffer *fb,
+static int repaper_fb_dirty(struct drm_framebuffer *fb, const struct iosys_map *vmap,
 			    struct drm_format_conv_state *fmtcnv_state)
 {
-	struct drm_gem_dma_object *dma_obj = drm_fb_dma_get_gem_obj(fb, 0);
 	struct repaper_epd *epd = drm_to_epd(fb->dev);
 	unsigned int dst_pitch = 0;
-	struct iosys_map dst, vmap;
+	struct iosys_map dst;
 	struct drm_rect clip;
 	int idx, ret = 0;
 	u8 *buf = NULL;
@@ -546,8 +545,7 @@ static int repaper_fb_dirty(struct drm_framebuffer *fb,
 		goto out_free;
 
 	iosys_map_set_vaddr(&dst, buf);
-	iosys_map_set_vaddr(&vmap, dma_obj->vaddr);
-	drm_fb_xrgb8888_to_mono(&dst, &dst_pitch, &vmap, fb, &clip, fmtcnv_state);
+	drm_fb_xrgb8888_to_mono(&dst, &dst_pitch, vmap, fb, &clip, fmtcnv_state);
 
 	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);
 
@@ -832,16 +830,15 @@ static void repaper_pipe_update(struct drm_simple_display_pipe *pipe,
 				struct drm_plane_state *old_state)
 {
 	struct drm_plane_state *state = pipe->plane.state;
-	struct drm_format_conv_state fmtcnv_state = DRM_FORMAT_CONV_STATE_INIT;
+	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(state);
 	struct drm_rect rect;
 
 	if (!pipe->crtc.state->active)
 		return;
 
 	if (drm_atomic_helper_damage_merged(old_state, state, &rect))
-		repaper_fb_dirty(state->fb, &fmtcnv_state);
-
-	drm_format_conv_state_release(&fmtcnv_state);
+		repaper_fb_dirty(state->fb, shadow_plane_state->data,
+				 &shadow_plane_state->fmtcnv_state);
 }
 
 static const struct drm_simple_display_pipe_funcs repaper_pipe_funcs = {
@@ -849,6 +846,7 @@ static const struct drm_simple_display_pipe_funcs repaper_pipe_funcs = {
 	.enable = repaper_pipe_enable,
 	.disable = repaper_pipe_disable,
 	.update = repaper_pipe_update,
+	DRM_GEM_SIMPLE_DISPLAY_PIPE_SHADOW_PLANE_FUNCS,
 };
 
 static int repaper_connector_get_modes(struct drm_connector *connector)

From 136c374d8c80378d2982a46b2adabfc007299641 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Fri, 27 Jun 2025 17:23:09 +0200
Subject: [PATCH 0366/1455] drm/sharp-memory: Do not access GEM-DMA vaddr
 directly

Use DRM's shadow-plane helper to map and access the GEM object's buffer
within kernel address space. Encapsulates the vmap logic in the GEM-DMA
helpers.

The sharp-memory driver currently reads the vaddr field from the GME
buffer object directly. This only works because GEM code 'automagically'
sets vaddr.

Shadow-plane helpers perform the same steps, but with correct abstraction
behind drm_gem_vmap(). The shadow-plane state provides the buffer address
in kernel address space and the format-conversion state.

v2:
- fix typo in commit description

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250627152327.8244-1-tzimmermann@suse.de
---
 drivers/gpu/drm/tiny/sharp-memory.c | 27 +++++++++++++--------------
 1 file changed, 13 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/tiny/sharp-memory.c b/drivers/gpu/drm/tiny/sharp-memory.c
index 03d2850310c4..64272cd0f6e2 100644
--- a/drivers/gpu/drm/tiny/sharp-memory.c
+++ b/drivers/gpu/drm/tiny/sharp-memory.c
@@ -126,28 +126,28 @@ static inline void sharp_memory_set_tx_buffer_addresses(u8 *buffer,
 
 static void sharp_memory_set_tx_buffer_data(u8 *buffer,
 					    struct drm_framebuffer *fb,
+					    const struct iosys_map *vmap,
 					    struct drm_rect clip,
 					    u32 pitch,
 					    struct drm_format_conv_state *fmtcnv_state)
 {
 	int ret;
-	struct iosys_map dst, vmap;
-	struct drm_gem_dma_object *dma_obj = drm_fb_dma_get_gem_obj(fb, 0);
+	struct iosys_map dst;
 
 	ret = drm_gem_fb_begin_cpu_access(fb, DMA_FROM_DEVICE);
 	if (ret)
 		return;
 
 	iosys_map_set_vaddr(&dst, buffer);
-	iosys_map_set_vaddr(&vmap, dma_obj->vaddr);
 
-	drm_fb_xrgb8888_to_mono(&dst, &pitch, &vmap, fb, &clip, fmtcnv_state);
+	drm_fb_xrgb8888_to_mono(&dst, &pitch, vmap, fb, &clip, fmtcnv_state);
 
 	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);
 }
 
 static int sharp_memory_update_display(struct sharp_memory_device *smd,
 				       struct drm_framebuffer *fb,
+				       const struct iosys_map *vmap,
 				       struct drm_rect clip,
 				       struct drm_format_conv_state *fmtcnv_state)
 {
@@ -163,7 +163,7 @@ static int sharp_memory_update_display(struct sharp_memory_device *smd,
 	sharp_memory_set_tx_buffer_mode(&tx_buffer[0],
 					SHARP_MEMORY_DISPLAY_UPDATE_MODE, vcom);
 	sharp_memory_set_tx_buffer_addresses(&tx_buffer[1], clip, pitch);
-	sharp_memory_set_tx_buffer_data(&tx_buffer[2], fb, clip, pitch, fmtcnv_state);
+	sharp_memory_set_tx_buffer_data(&tx_buffer[2], fb, vmap, clip, pitch, fmtcnv_state);
 
 	ret = sharp_memory_spi_write(smd->spi, tx_buffer, tx_buffer_size);
 
@@ -206,7 +206,8 @@ static int sharp_memory_clear_display(struct sharp_memory_device *smd)
 	return ret;
 }
 
-static void sharp_memory_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *rect,
+static void sharp_memory_fb_dirty(struct drm_framebuffer *fb, const struct iosys_map *vmap,
+				  struct drm_rect *rect,
 				  struct drm_format_conv_state *fmtconv_state)
 {
 	struct drm_rect clip;
@@ -218,7 +219,7 @@ static void sharp_memory_fb_dirty(struct drm_framebuffer *fb, struct drm_rect *r
 	clip.y1 = rect->y1;
 	clip.y2 = rect->y2;
 
-	sharp_memory_update_display(smd, fb, clip, fmtconv_state);
+	sharp_memory_update_display(smd, fb, vmap, clip, fmtconv_state);
 }
 
 static int sharp_memory_plane_atomic_check(struct drm_plane *plane,
@@ -242,7 +243,7 @@ static void sharp_memory_plane_atomic_update(struct drm_plane *plane,
 {
 	struct drm_plane_state *old_state = drm_atomic_get_old_plane_state(state, plane);
 	struct drm_plane_state *plane_state = plane->state;
-	struct drm_format_conv_state fmtcnv_state = DRM_FORMAT_CONV_STATE_INIT;
+	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
 	struct sharp_memory_device *smd;
 	struct drm_rect rect;
 
@@ -251,15 +252,15 @@ static void sharp_memory_plane_atomic_update(struct drm_plane *plane,
 		return;
 
 	if (drm_atomic_helper_damage_merged(old_state, plane_state, &rect))
-		sharp_memory_fb_dirty(plane_state->fb, &rect, &fmtcnv_state);
-
-	drm_format_conv_state_release(&fmtcnv_state);
+		sharp_memory_fb_dirty(plane_state->fb, shadow_plane_state->data,
+				      &rect, &shadow_plane_state->fmtcnv_state);
 }
 
 static const struct drm_plane_helper_funcs sharp_memory_plane_helper_funcs = {
 	.prepare_fb = drm_gem_plane_helper_prepare_fb,
 	.atomic_check = sharp_memory_plane_atomic_check,
 	.atomic_update = sharp_memory_plane_atomic_update,
+	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
 };
 
 static bool sharp_memory_format_mod_supported(struct drm_plane *plane,
@@ -273,9 +274,7 @@ static const struct drm_plane_funcs sharp_memory_plane_funcs = {
 	.update_plane = drm_atomic_helper_update_plane,
 	.disable_plane = drm_atomic_helper_disable_plane,
 	.destroy = drm_plane_cleanup,
-	.reset = drm_atomic_helper_plane_reset,
-	.atomic_duplicate_state	= drm_atomic_helper_plane_duplicate_state,
-	.atomic_destroy_state = drm_atomic_helper_plane_destroy_state,
+	DRM_GEM_SHADOW_PLANE_FUNCS,
 	.format_mod_supported = sharp_memory_format_mod_supported,
 };
 

From 754dbf164acd4d22dd7a5241b1880f54546d68f2 Mon Sep 17 00:00:00 2001
From: Christopher Orr <chris.orr@gmail.com>
Date: Tue, 5 Aug 2025 21:27:55 -0400
Subject: [PATCH 0367/1455] drm/panel-edp: Add SHP LQ134Z1 panel for Dell XPS
 9345

Introduce high-res OLED panel for the Dell XPS 9345

These timings were selected based on Alex Vinarkskis' commit,
(6b3815c6815f07acc7eeffa8ae734d1a1c0ee817) for the LQ134N1
and seem to work fine for the high-res OLED panel on the 9345.

The raw edid for this SHP panel is:

00 ff ff ff ff ff ff 00 4d 10 8f 15 00 00 00 00
2e 21 01 04 b5 1d 12 78 03 0f 95 ae 52 43 b0 26
0f 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 fd d7 00 a0 a0 40 fc 66 30 20
36 00 20 b4 10 00 00 18 00 00 00 fd 00 1e 78 cc
cc 38 01 0a 20 20 20 20 20 20 00 00 00 fe 00 43
37 31 4d 31 81 4c 51 31 33 34 5a 31 00 00 00 00
00 02 41 0c 32 01 01 00 00 0b 41 0a 20 20 01 ea

70 20 79 02 00 20 00 13 8c 52 19 8f 15 00 00 00
00 2e 17 07 4c 51 31 33 34 5a 31 21 00 1d 40 0b
08 07 00 0a 40 06 88 e1 fa 51 3d a4 b0 66 62 0f
02 45 54 d0 5f d0 5f 00 34 13 78 26 00 09 06 00
00 00 00 00 41 00 00 22 00 14 d9 6f 08 05 ff 09
9f 00 2f 00 1f 00 3f 06 5d 00 02 00 05 00 25 01
09 d9 6f 08 d9 6f 08 1e 78 80 81 00 0b e3 05 80
00 e6 06 05 01 6a 6a 39 00 00 00 00 00 00 58 90

Signed-off-by: Christopher Orr <chris.orr@gmail.com>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/aJKvm3SlhLGHW4qn@jander
---
 drivers/gpu/drm/panel/panel-edp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index 3e4c3a439988..feba520a0f45 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -2055,6 +2055,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('S', 'H', 'P', 0x1523, &delay_80_500_e50, "LQ140M1JW46"),
 	EDP_PANEL_ENTRY('S', 'H', 'P', 0x153a, &delay_200_500_e50, "LQ140T1JH01"),
 	EDP_PANEL_ENTRY('S', 'H', 'P', 0x154c, &delay_200_500_p2e100, "LQ116M1JW10"),
+	EDP_PANEL_ENTRY('S', 'H', 'P', 0x158f, &delay_200_500_p2e100, "LQ134Z1"),
 	EDP_PANEL_ENTRY('S', 'H', 'P', 0x1593, &delay_200_500_p2e100, "LQ134N1"),
 
 	EDP_PANEL_ENTRY('S', 'T', 'A', 0x0004, &delay_200_500_e200, "116KHD024006"),

From 27738c3003bf3b124527c9ed75e1e0d0c013c101 Mon Sep 17 00:00:00 2001
From: Timur Tabi <ttabi@nvidia.com>
Date: Fri, 8 Aug 2025 14:13:40 -0500
Subject: [PATCH 0368/1455] drm/nouveau: always set RMDevidCheckIgnore for
 GSP-RM

Always set the RMDevidCheckIgnore registry key for GSP-RM so that it
will continue support newer variants of already supported GPUs.

GSP-RM maintains an internal list of PCI IDs of GPUs that it supports,
and checks if the current GPU is on this list.  While the actual GPU
architecture (as specified in the BOOT_0/BOOT_42 registers) determines
how to enable the GPU, the PCI ID is used for the product name, e.g.
"NVIDIA GeForce RTX 5090".

Unfortunately, if there is no match, GSP-RM will refuse to initialize,
even if the device is fully supported.  Nouveau will get an error
return code, but by then it's too late.  This behavior may be corrected
in a future version of GSP-RM, but that does not help Nouveau today.

Fortunately, GSP-RM supports an undocumented registry key that tells it
to ignore the mismatch.  In such cases, the product name returned will
be a blank string, but otherwise GSP-RM will continue.

Unlike Nvidia's proprietary driver, Nouveau cannot update to newer
firmware versions to keep up with every new hardware release.  Instead,
we can permanently set this registry key, and GSP-RM will continue
to function the same with known hardware.

Signed-off-by: Timur Tabi <ttabi@nvidia.com>
Link: https://lore.kernel.org/r/20250808191340.1701983-1-ttabi@nvidia.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
index 588cb4ab85cb..32e6a065d6d7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/gsp.c
@@ -582,10 +582,13 @@ struct nv_gsp_registry_entries {
  * RMSecBusResetEnable - enables PCI secondary bus reset
  * RMForcePcieConfigSave - forces GSP-RM to preserve PCI configuration
  *   registers on any PCI reset.
+ * RMDevidCheckIgnore - allows GSP-RM to boot even if the PCI dev ID
+ *   is not found in the internal product name database.
  */
 static const struct nv_gsp_registry_entries r535_registry_entries[] = {
 	{ "RMSecBusResetEnable", 1 },
 	{ "RMForcePcieConfigSave", 1 },
+	{ "RMDevidCheckIgnore", 1 },
 };
 #define NV_GSP_REG_NUM_ENTRIES ARRAY_SIZE(r535_registry_entries)
 

From 0f85b6c1cb5746180b2bf674a436a48e6722e7ed Mon Sep 17 00:00:00 2001
From: Rhys Lloyd <krakow20@gmail.com>
Date: Fri, 18 Jul 2025 00:36:24 -0700
Subject: [PATCH 0369/1455] gpu: nova-core: vbios: use size_of instead of magic
 number

12 is identical to the value of `size_of::<BitHeader>()`, so use the
latter instead.

[acourbot@nvidia.com: remove `dead_code` expect to fix `unfulfilled_lint_expectations` lint]

Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Rhys Lloyd <krakow20@gmail.com>
Link: https://lore.kernel.org/r/20250718073633.194032-2-krakow20@gmail.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/vbios.rs | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/nova-core/vbios.rs b/drivers/gpu/nova-core/vbios.rs
index 5b5d9f38cbb3..b4bca1919d8f 100644
--- a/drivers/gpu/nova-core/vbios.rs
+++ b/drivers/gpu/nova-core/vbios.rs
@@ -345,7 +345,7 @@ impl PcirStruct {
 /// its header) is in the [`PciAtBiosImage`] and the falcon data it is pointing to is in the
 /// [`FwSecBiosImage`].
 #[derive(Debug, Clone, Copy)]
-#[expect(dead_code)]
+#[repr(C)]
 struct BitHeader {
     /// 0h: BIT Header Identifier (BMP=0x7FFF/BIT=0xB8FF)
     id: u16,
@@ -365,7 +365,7 @@ struct BitHeader {
 
 impl BitHeader {
     fn new(data: &[u8]) -> Result<Self> {
-        if data.len() < 12 {
+        if data.len() < core::mem::size_of::<Self>() {
             return Err(EINVAL);
         }
 

From 2872c8982c80e22f9a7fc4db7d8f0e8188ab4cd7 Mon Sep 17 00:00:00 2001
From: Rhys Lloyd <krakow20@gmail.com>
Date: Fri, 18 Jul 2025 00:36:25 -0700
Subject: [PATCH 0370/1455] gpu: nova-core: vbios: change PmuLookupTableEntry
 to use size_of

Annotate the PmuLookupTableEntry with an `#[repr(C, packed)]` attribute.
Removes another magic number by making the struct the same size as the
data it needs to read, allowing the use of
`size_of::<PmuLookupTableEntry>()`

[acourbot@nvidia.com: remove `dead_code` expect to fix `unfulfilled_lint_expectations` lint]

Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Rhys Lloyd <krakow20@gmail.com>
Link: https://lore.kernel.org/r/20250718073633.194032-3-krakow20@gmail.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/vbios.rs | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/nova-core/vbios.rs b/drivers/gpu/nova-core/vbios.rs
index b4bca1919d8f..50fbea69fc00 100644
--- a/drivers/gpu/nova-core/vbios.rs
+++ b/drivers/gpu/nova-core/vbios.rs
@@ -892,7 +892,7 @@ impl TryFrom<BiosImageBase> for PciAtBiosImage {
 /// The [`PmuLookupTableEntry`] structure is a single entry in the [`PmuLookupTable`].
 ///
 /// See the [`PmuLookupTable`] description for more information.
-#[expect(dead_code)]
+#[repr(C, packed)]
 struct PmuLookupTableEntry {
     application_id: u8,
     target_id: u8,
@@ -901,7 +901,7 @@ struct PmuLookupTableEntry {
 
 impl PmuLookupTableEntry {
     fn new(data: &[u8]) -> Result<Self> {
-        if data.len() < 6 {
+        if data.len() < core::mem::size_of::<Self>() {
             return Err(EINVAL);
         }
 

From 3b51739fa2001b047359e11afd755ae5e189b7db Mon Sep 17 00:00:00 2001
From: Philipp Stanner <phasta@kernel.org>
Date: Thu, 7 Aug 2025 09:07:57 +0200
Subject: [PATCH 0371/1455] MAINTAINERS: Add website of Nova GPU driver

The Nova GPU driver has a sub-website on the Rust-for-Linux website
which so far was missing from the respective section in MAINTAINERS.

Add the Nova website.

Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250807070756.103816-2-phasta@kernel.org
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 MAINTAINERS | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index fe168477caa4..936650c7eff9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7829,6 +7829,7 @@ M:	Danilo Krummrich <dakr@kernel.org>
 M:	Alexandre Courbot <acourbot@nvidia.com>
 L:	nouveau@lists.freedesktop.org
 S:	Supported
+W:	https://rust-for-linux.com/nova-gpu-driver
 Q:	https://patchwork.freedesktop.org/project/nouveau/
 B:	https://gitlab.freedesktop.org/drm/nova/-/issues
 C:	irc://irc.oftc.net/nouveau
@@ -7840,6 +7841,7 @@ DRM DRIVER FOR NVIDIA GPUS [RUST]
 M:	Danilo Krummrich <dakr@kernel.org>
 L:	nouveau@lists.freedesktop.org
 S:	Supported
+W:	https://rust-for-linux.com/nova-gpu-driver
 Q:	https://patchwork.freedesktop.org/project/nouveau/
 B:	https://gitlab.freedesktop.org/drm/nova/-/issues
 C:	irc://irc.oftc.net/nouveau

From 67e980f58dd71d9483d5af683558f5913b1c95ae Mon Sep 17 00:00:00 2001
From: Nemesa Garg <nemesa.garg@intel.com>
Date: Thu, 7 Aug 2025 17:08:55 +0530
Subject: [PATCH 0372/1455] drm/i915/scaler: Fix condition for WA_14011503117

As scaler_state can never be null so no need to
check this, only check if scaler_id is less
than 0 or not.

v2: Add scaler_id check [Jani]
v3: Modify commit message[Suraj]

Fixes: 73309ed9d598 ("drm/i915/display: WA_14011503117")
Signed-off-by: Nemesa Garg <nemesa.garg@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250807113855.3175435-1-nemesa.garg@intel.com
---
 drivers/gpu/drm/i915/display/skl_scaler.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
index cd7ebbeb9508..c6cccf170ff1 100644
--- a/drivers/gpu/drm/i915/display/skl_scaler.c
+++ b/drivers/gpu/drm/i915/display/skl_scaler.c
@@ -960,7 +960,7 @@ void adl_scaler_ecc_unmask(const struct intel_crtc_state *crtc_state)
 	const struct intel_crtc_scaler_state *scaler_state =
 		&crtc_state->scaler_state;
 
-	if (!scaler_state && scaler_state->scaler_id == -1)
+	if (scaler_state->scaler_id < 0)
 		return;
 
 	intel_de_write_fw(display,

From 21b137f651cf9d981e22d2c60a2a8105f50a6361 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Tue, 10 Jun 2025 16:58:25 -0700
Subject: [PATCH 0373/1455] drm/bridge: it6505: select REGMAP_I2C

Fix

  aarch64-linux-gnu-ld: drivers/gpu/drm/bridge/ite-it6505.o: in function `it6505_i2c_probe':
  ite-it6505.c:(.text+0x754): undefined reference to `__devm_regmap_init_i2c'

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Fixes: b5c84a9edcd4 ("drm/bridge: add it6505 driver")
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://patch.msgid.link/20250610235825.3113075-1-olvaffe@gmail.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
 drivers/gpu/drm/bridge/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 9cefbd15177a..6945029b3592 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -122,6 +122,7 @@ config DRM_ITE_IT6505
 	select EXTCON
 	select CRYPTO
 	select CRYPTO_HASH
+	select REGMAP_I2C
 	help
 	  ITE IT6505 DisplayPort bridge chip driver.
 

From fb357dbadbebc7a9ca3c5ef26f6c792b0e8e1278 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 15 Jul 2025 14:24:42 +0200
Subject: [PATCH 0374/1455] fbcon: Add necessary include statements and forward
 declarations

Make the header self contained for including.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Simona Vetter <simona.vetter@ffwll.ch>
Link: https://lore.kernel.org/r/20250715122643.137027-6-tzimmermann@suse.de
---
 include/linux/fbcon.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/include/linux/fbcon.h b/include/linux/fbcon.h
index 2382dec6d6ab..81f0e698acbf 100644
--- a/include/linux/fbcon.h
+++ b/include/linux/fbcon.h
@@ -1,6 +1,13 @@
 #ifndef _LINUX_FBCON_H
 #define _LINUX_FBCON_H
 
+#include <linux/compiler_types.h>
+
+struct fb_blit_caps;
+struct fb_info;
+struct fb_var_screeninfo;
+struct fb_videomode;
+
 #ifdef CONFIG_FRAMEBUFFER_CONSOLE
 void __init fb_console_init(void);
 void __exit fb_console_exit(void);

From f3c65a8e9b09f66042e418674283db8675a445a8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 11 Aug 2025 18:25:47 +0300
Subject: [PATCH 0375/1455] drm/i915/vbt: split up DSI VBT defs to a separate
 file

The DSI VBT definitions have ended up in intel_bios.h, because
intel_vbt_defs.h is supposed to be internal to intel_bios.c, but the DSI
VBT definitions are needed in more places.

Split out the DSI VBT definitions to intel_dsi_vbt_defs.h. This will
also help keep the definitions in sync with IGT.

Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/84417e0141f98ae8f8c7a66e9002c3e99c9ed3db.1754925923.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bios.h     | 174 -----------------
 .../drm/i915/display/intel_display_types.h    |   1 +
 .../gpu/drm/i915/display/intel_dsi_vbt_defs.h | 183 ++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_vbt_defs.h |   2 +-
 4 files changed, 185 insertions(+), 175 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h

diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h
index 6cd7a011b8c4..8fdde85f7939 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.h
+++ b/drivers/gpu/drm/i915/display/intel_bios.h
@@ -50,180 +50,6 @@ enum intel_backlight_type {
 	INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE,
 };
 
-/*
- * MIPI Sequence Block definitions
- *
- * Note the VBT spec has AssertReset / DeassertReset swapped from their
- * usual naming, we use the proper names here to avoid confusion when
- * reading the code.
- */
-enum mipi_seq {
-	MIPI_SEQ_END = 0,
-	MIPI_SEQ_DEASSERT_RESET,	/* Spec says MipiAssertResetPin */
-	MIPI_SEQ_INIT_OTP,
-	MIPI_SEQ_DISPLAY_ON,
-	MIPI_SEQ_DISPLAY_OFF,
-	MIPI_SEQ_ASSERT_RESET,		/* Spec says MipiDeassertResetPin */
-	MIPI_SEQ_BACKLIGHT_ON,		/* sequence block v2+ */
-	MIPI_SEQ_BACKLIGHT_OFF,		/* sequence block v2+ */
-	MIPI_SEQ_TEAR_ON,		/* sequence block v2+ */
-	MIPI_SEQ_TEAR_OFF,		/* sequence block v3+ */
-	MIPI_SEQ_POWER_ON,		/* sequence block v3+ */
-	MIPI_SEQ_POWER_OFF,		/* sequence block v3+ */
-	MIPI_SEQ_MAX
-};
-
-enum mipi_seq_element {
-	MIPI_SEQ_ELEM_END = 0,
-	MIPI_SEQ_ELEM_SEND_PKT,
-	MIPI_SEQ_ELEM_DELAY,
-	MIPI_SEQ_ELEM_GPIO,
-	MIPI_SEQ_ELEM_I2C,		/* sequence block v2+ */
-	MIPI_SEQ_ELEM_SPI,		/* sequence block v3+ */
-	MIPI_SEQ_ELEM_PMIC,		/* sequence block v3+ */
-	MIPI_SEQ_ELEM_MAX
-};
-
-#define MIPI_DSI_UNDEFINED_PANEL_ID	0
-#define MIPI_DSI_GENERIC_PANEL_ID	1
-
-struct mipi_config {
-	u16 panel_id;
-
-	/* General Params */
-	u32 enable_dithering:1;
-	u32 rsvd1:1;
-	u32 is_bridge:1;
-
-	u32 panel_arch_type:2;
-	u32 is_cmd_mode:1;
-
-#define NON_BURST_SYNC_PULSE	0x1
-#define NON_BURST_SYNC_EVENTS	0x2
-#define BURST_MODE		0x3
-	u32 video_transfer_mode:2;
-
-	u32 cabc_supported:1;
-#define PPS_BLC_PMIC   0
-#define PPS_BLC_SOC    1
-	u32 pwm_blc:1;
-
-	/* Bit 13:10 */
-#define PIXEL_FORMAT_RGB565			0x1
-#define PIXEL_FORMAT_RGB666			0x2
-#define PIXEL_FORMAT_RGB666_LOOSELY_PACKED	0x3
-#define PIXEL_FORMAT_RGB888			0x4
-	u32 videomode_color_format:4;
-
-	/* Bit 15:14 */
-#define ENABLE_ROTATION_0	0x0
-#define ENABLE_ROTATION_90	0x1
-#define ENABLE_ROTATION_180	0x2
-#define ENABLE_ROTATION_270	0x3
-	u32 rotation:2;
-	u32 bta_enabled:1;
-	u32 rsvd2:15;
-
-	/* 2 byte Port Description */
-#define DUAL_LINK_NOT_SUPPORTED	0
-#define DUAL_LINK_FRONT_BACK	1
-#define DUAL_LINK_PIXEL_ALT	2
-	u16 dual_link:2;
-	u16 lane_cnt:2;
-	u16 pixel_overlap:3;
-	u16 rgb_flip:1;
-#define DL_DCS_PORT_A			0x00
-#define DL_DCS_PORT_C			0x01
-#define DL_DCS_PORT_A_AND_C		0x02
-	u16 dl_dcs_cabc_ports:2;
-	u16 dl_dcs_backlight_ports:2;
-	u16 rsvd3:4;
-
-	u16 rsvd4;
-
-	u8 rsvd5;
-	u32 target_burst_mode_freq;
-	u32 dsi_ddr_clk;
-	u32 bridge_ref_clk;
-
-#define  BYTE_CLK_SEL_20MHZ		0
-#define  BYTE_CLK_SEL_10MHZ		1
-#define  BYTE_CLK_SEL_5MHZ		2
-	u8 byte_clk_sel:2;
-
-	u8 rsvd6:6;
-
-	/* DPHY Flags */
-	u16 dphy_param_valid:1;
-	u16 eot_pkt_disabled:1;
-	u16 enable_clk_stop:1;
-	u16 rsvd7:13;
-
-	u32 hs_tx_timeout;
-	u32 lp_rx_timeout;
-	u32 turn_around_timeout;
-	u32 device_reset_timer;
-	u32 master_init_timer;
-	u32 dbi_bw_timer;
-	u32 lp_byte_clk_val;
-
-	/*  4 byte Dphy Params */
-	u32 prepare_cnt:6;
-	u32 rsvd8:2;
-	u32 clk_zero_cnt:8;
-	u32 trail_cnt:5;
-	u32 rsvd9:3;
-	u32 exit_zero_cnt:6;
-	u32 rsvd10:2;
-
-	u32 clk_lane_switch_cnt;
-	u32 hl_switch_cnt;
-
-	u32 rsvd11[6];
-
-	/* timings based on dphy spec */
-	u8 tclk_miss;
-	u8 tclk_post;
-	u8 rsvd12;
-	u8 tclk_pre;
-	u8 tclk_prepare;
-	u8 tclk_settle;
-	u8 tclk_term_enable;
-	u8 tclk_trail;
-	u16 tclk_prepare_clkzero;
-	u8 rsvd13;
-	u8 td_term_enable;
-	u8 teot;
-	u8 ths_exit;
-	u8 ths_prepare;
-	u16 ths_prepare_hszero;
-	u8 rsvd14;
-	u8 ths_settle;
-	u8 ths_skip;
-	u8 ths_trail;
-	u8 tinit;
-	u8 tlpx;
-	u8 rsvd15[3];
-
-	/* GPIOs */
-	u8 panel_enable;
-	u8 bl_enable;
-	u8 pwm_enable;
-	u8 reset_r_n;
-	u8 pwr_down_r;
-	u8 stdby_r_n;
-
-} __packed;
-
-/* all delays have a unit of 100us */
-struct mipi_pps_data {
-	u16 panel_on_delay;
-	u16 bl_enable_delay;
-	u16 bl_disable_delay;
-	u16 panel_off_delay;
-	u16 panel_power_cycle_delay;
-} __packed;
-
 void intel_bios_init(struct intel_display *display);
 void intel_bios_init_panel_early(struct intel_display *display,
 				 struct intel_panel *panel,
diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 35596f3921e8..0d945d1fedd6 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -50,6 +50,7 @@
 #include "intel_display_limits.h"
 #include "intel_display_power.h"
 #include "intel_dpll_mgr.h"
+#include "intel_dsi_vbt_defs.h"
 #include "intel_wm_types.h"
 
 struct cec_notifier;
diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
new file mode 100644
index 000000000000..f83d42ed0c5a
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
@@ -0,0 +1,183 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __INTEL_DSI_VBT_DEFS_H__
+#define __INTEL_DSI_VBT_DEFS_H__
+
+#include <linux/types.h>
+
+/*
+ * MIPI Sequence Block definitions
+ *
+ * Note the VBT spec has AssertReset / DeassertReset swapped from their
+ * usual naming, we use the proper names here to avoid confusion when
+ * reading the code.
+ */
+enum mipi_seq {
+	MIPI_SEQ_END = 0,
+	MIPI_SEQ_DEASSERT_RESET,	/* Spec says MipiAssertResetPin */
+	MIPI_SEQ_INIT_OTP,
+	MIPI_SEQ_DISPLAY_ON,
+	MIPI_SEQ_DISPLAY_OFF,
+	MIPI_SEQ_ASSERT_RESET,		/* Spec says MipiDeassertResetPin */
+	MIPI_SEQ_BACKLIGHT_ON,		/* sequence block v2+ */
+	MIPI_SEQ_BACKLIGHT_OFF,		/* sequence block v2+ */
+	MIPI_SEQ_TEAR_ON,		/* sequence block v2+ */
+	MIPI_SEQ_TEAR_OFF,		/* sequence block v3+ */
+	MIPI_SEQ_POWER_ON,		/* sequence block v3+ */
+	MIPI_SEQ_POWER_OFF,		/* sequence block v3+ */
+	MIPI_SEQ_MAX
+};
+
+enum mipi_seq_element {
+	MIPI_SEQ_ELEM_END = 0,
+	MIPI_SEQ_ELEM_SEND_PKT,
+	MIPI_SEQ_ELEM_DELAY,
+	MIPI_SEQ_ELEM_GPIO,
+	MIPI_SEQ_ELEM_I2C,		/* sequence block v2+ */
+	MIPI_SEQ_ELEM_SPI,		/* sequence block v3+ */
+	MIPI_SEQ_ELEM_PMIC,		/* sequence block v3+ */
+	MIPI_SEQ_ELEM_MAX
+};
+
+#define MIPI_DSI_UNDEFINED_PANEL_ID	0
+#define MIPI_DSI_GENERIC_PANEL_ID	1
+
+struct mipi_config {
+	u16 panel_id;
+
+	/* General Params */
+	u32 enable_dithering:1;
+	u32 rsvd1:1;
+	u32 is_bridge:1;
+
+	u32 panel_arch_type:2;
+	u32 is_cmd_mode:1;
+
+#define NON_BURST_SYNC_PULSE	0x1
+#define NON_BURST_SYNC_EVENTS	0x2
+#define BURST_MODE		0x3
+	u32 video_transfer_mode:2;
+
+	u32 cabc_supported:1;
+#define PPS_BLC_PMIC   0
+#define PPS_BLC_SOC    1
+	u32 pwm_blc:1;
+
+	/* Bit 13:10 */
+#define PIXEL_FORMAT_RGB565			0x1
+#define PIXEL_FORMAT_RGB666			0x2
+#define PIXEL_FORMAT_RGB666_LOOSELY_PACKED	0x3
+#define PIXEL_FORMAT_RGB888			0x4
+	u32 videomode_color_format:4;
+
+	/* Bit 15:14 */
+#define ENABLE_ROTATION_0	0x0
+#define ENABLE_ROTATION_90	0x1
+#define ENABLE_ROTATION_180	0x2
+#define ENABLE_ROTATION_270	0x3
+	u32 rotation:2;
+	u32 bta_enabled:1;
+	u32 rsvd2:15;
+
+	/* 2 byte Port Description */
+#define DUAL_LINK_NOT_SUPPORTED	0
+#define DUAL_LINK_FRONT_BACK	1
+#define DUAL_LINK_PIXEL_ALT	2
+	u16 dual_link:2;
+	u16 lane_cnt:2;
+	u16 pixel_overlap:3;
+	u16 rgb_flip:1;
+#define DL_DCS_PORT_A			0x00
+#define DL_DCS_PORT_C			0x01
+#define DL_DCS_PORT_A_AND_C		0x02
+	u16 dl_dcs_cabc_ports:2;
+	u16 dl_dcs_backlight_ports:2;
+	u16 rsvd3:4;
+
+	u16 rsvd4;
+
+	u8 rsvd5;
+	u32 target_burst_mode_freq;
+	u32 dsi_ddr_clk;
+	u32 bridge_ref_clk;
+
+#define  BYTE_CLK_SEL_20MHZ		0
+#define  BYTE_CLK_SEL_10MHZ		1
+#define  BYTE_CLK_SEL_5MHZ		2
+	u8 byte_clk_sel:2;
+
+	u8 rsvd6:6;
+
+	/* DPHY Flags */
+	u16 dphy_param_valid:1;
+	u16 eot_pkt_disabled:1;
+	u16 enable_clk_stop:1;
+	u16 rsvd7:13;
+
+	u32 hs_tx_timeout;
+	u32 lp_rx_timeout;
+	u32 turn_around_timeout;
+	u32 device_reset_timer;
+	u32 master_init_timer;
+	u32 dbi_bw_timer;
+	u32 lp_byte_clk_val;
+
+	/*  4 byte Dphy Params */
+	u32 prepare_cnt:6;
+	u32 rsvd8:2;
+	u32 clk_zero_cnt:8;
+	u32 trail_cnt:5;
+	u32 rsvd9:3;
+	u32 exit_zero_cnt:6;
+	u32 rsvd10:2;
+
+	u32 clk_lane_switch_cnt;
+	u32 hl_switch_cnt;
+
+	u32 rsvd11[6];
+
+	/* timings based on dphy spec */
+	u8 tclk_miss;
+	u8 tclk_post;
+	u8 rsvd12;
+	u8 tclk_pre;
+	u8 tclk_prepare;
+	u8 tclk_settle;
+	u8 tclk_term_enable;
+	u8 tclk_trail;
+	u16 tclk_prepare_clkzero;
+	u8 rsvd13;
+	u8 td_term_enable;
+	u8 teot;
+	u8 ths_exit;
+	u8 ths_prepare;
+	u16 ths_prepare_hszero;
+	u8 rsvd14;
+	u8 ths_settle;
+	u8 ths_skip;
+	u8 ths_trail;
+	u8 tinit;
+	u8 tlpx;
+	u8 rsvd15[3];
+
+	/* GPIOs */
+	u8 panel_enable;
+	u8 bl_enable;
+	u8 pwm_enable;
+	u8 reset_r_n;
+	u8 pwr_down_r;
+	u8 stdby_r_n;
+
+} __packed;
+
+/* all delays have a unit of 100us */
+struct mipi_pps_data {
+	u16 panel_on_delay;
+	u16 bl_enable_delay;
+	u16 bl_disable_delay;
+	u16 panel_off_delay;
+	u16 panel_power_cycle_delay;
+} __packed;
+
+#endif /* __INTEL_DSI_VBT_DEFS_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
index 92c04811aa28..6612d3a4ec49 100644
--- a/drivers/gpu/drm/i915/display/intel_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
@@ -37,7 +37,7 @@
 #ifndef _INTEL_VBT_DEFS_H_
 #define _INTEL_VBT_DEFS_H_
 
-#include "intel_bios.h"
+#include "intel_dsi_vbt_defs.h"
 
 /* EDID derived structures */
 struct bdb_edid_pnp_id {

From aa0c02f04369b7080515a5be4c9b9a70c88306c2 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 11 Aug 2025 18:25:48 +0300
Subject: [PATCH 0376/1455] drm/i915/vbt: add anonymous structs to group DSI
 VBT defs

The grouping of DSI VBT definitions is hard to follow and match against
the spec. Use anonymous structs and add comments with the spec
description.

Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/f57ca596aefa3ef0b4ce1f36452410cf745acddd.1754925923.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../gpu/drm/i915/display/intel_dsi_vbt_defs.h | 85 ++++++++++---------
 1 file changed, 46 insertions(+), 39 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
index f83d42ed0c5a..7ac872dbba8d 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
@@ -47,53 +47,55 @@ struct mipi_config {
 	u16 panel_id;
 
 	/* General Params */
-	u32 enable_dithering:1;
-	u32 rsvd1:1;
-	u32 is_bridge:1;
+	struct {
+		u32 enable_dithering:1;
+		u32 rsvd1:1;
+		u32 is_bridge:1;
 
-	u32 panel_arch_type:2;
-	u32 is_cmd_mode:1;
+		u32 panel_arch_type:2;
+		u32 is_cmd_mode:1;
 
 #define NON_BURST_SYNC_PULSE	0x1
 #define NON_BURST_SYNC_EVENTS	0x2
 #define BURST_MODE		0x3
-	u32 video_transfer_mode:2;
+		u32 video_transfer_mode:2;
 
-	u32 cabc_supported:1;
+		u32 cabc_supported:1;
 #define PPS_BLC_PMIC   0
 #define PPS_BLC_SOC    1
-	u32 pwm_blc:1;
+		u32 pwm_blc:1;
 
-	/* Bit 13:10 */
 #define PIXEL_FORMAT_RGB565			0x1
 #define PIXEL_FORMAT_RGB666			0x2
 #define PIXEL_FORMAT_RGB666_LOOSELY_PACKED	0x3
 #define PIXEL_FORMAT_RGB888			0x4
-	u32 videomode_color_format:4;
+		u32 videomode_color_format:4;
 
-	/* Bit 15:14 */
 #define ENABLE_ROTATION_0	0x0
 #define ENABLE_ROTATION_90	0x1
 #define ENABLE_ROTATION_180	0x2
 #define ENABLE_ROTATION_270	0x3
-	u32 rotation:2;
-	u32 bta_enabled:1;
-	u32 rsvd2:15;
+		u32 rotation:2;
+		u32 bta_enabled:1;
+		u32 rsvd2:15;
+	} __packed;
 
-	/* 2 byte Port Description */
+	/* Port Desc */
+	struct {
 #define DUAL_LINK_NOT_SUPPORTED	0
 #define DUAL_LINK_FRONT_BACK	1
 #define DUAL_LINK_PIXEL_ALT	2
-	u16 dual_link:2;
-	u16 lane_cnt:2;
-	u16 pixel_overlap:3;
-	u16 rgb_flip:1;
+		u16 dual_link:2;
+		u16 lane_cnt:2;
+		u16 pixel_overlap:3;
+		u16 rgb_flip:1;
 #define DL_DCS_PORT_A			0x00
 #define DL_DCS_PORT_C			0x01
 #define DL_DCS_PORT_A_AND_C		0x02
-	u16 dl_dcs_cabc_ports:2;
-	u16 dl_dcs_backlight_ports:2;
-	u16 rsvd3:4;
+		u16 dl_dcs_cabc_ports:2;
+		u16 dl_dcs_backlight_ports:2;
+		u16 rsvd3:4;
+	} __packed;
 
 	u16 rsvd4;
 
@@ -102,18 +104,22 @@ struct mipi_config {
 	u32 dsi_ddr_clk;
 	u32 bridge_ref_clk;
 
+	/* LP Byte Clock */
+	struct {
 #define  BYTE_CLK_SEL_20MHZ		0
 #define  BYTE_CLK_SEL_10MHZ		1
 #define  BYTE_CLK_SEL_5MHZ		2
-	u8 byte_clk_sel:2;
+		u8 byte_clk_sel:2;
+		u8 rsvd6:6;
+	} __packed;
 
-	u8 rsvd6:6;
-
-	/* DPHY Flags */
-	u16 dphy_param_valid:1;
-	u16 eot_pkt_disabled:1;
-	u16 enable_clk_stop:1;
-	u16 rsvd7:13;
+	/* DPhy Flags */
+	struct {
+		u16 dphy_param_valid:1;
+		u16 eot_pkt_disabled:1;
+		u16 enable_clk_stop:1;
+		u16 rsvd7:13;
+	} __packed;
 
 	u32 hs_tx_timeout;
 	u32 lp_rx_timeout;
@@ -123,14 +129,16 @@ struct mipi_config {
 	u32 dbi_bw_timer;
 	u32 lp_byte_clk_val;
 
-	/*  4 byte Dphy Params */
-	u32 prepare_cnt:6;
-	u32 rsvd8:2;
-	u32 clk_zero_cnt:8;
-	u32 trail_cnt:5;
-	u32 rsvd9:3;
-	u32 exit_zero_cnt:6;
-	u32 rsvd10:2;
+	/*  DPhy Params */
+	struct {
+		u32 prepare_cnt:6;
+		u32 rsvd8:2;
+		u32 clk_zero_cnt:8;
+		u32 trail_cnt:5;
+		u32 rsvd9:3;
+		u32 exit_zero_cnt:6;
+		u32 rsvd10:2;
+	} __packed;
 
 	u32 clk_lane_switch_cnt;
 	u32 hl_switch_cnt;
@@ -168,7 +176,6 @@ struct mipi_config {
 	u8 reset_r_n;
 	u8 pwr_down_r;
 	u8 stdby_r_n;
-
 } __packed;
 
 /* all delays have a unit of 100us */

From 745b597a79301c999324a9f9b943d0cb9140f9cb Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 11 Aug 2025 18:25:49 +0300
Subject: [PATCH 0377/1455] drm/i915/vbt: flip bta_enabled to bta_disable

The meaning is disable, so flip the member name.

Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/180079eca346edc1671c164da2ca7f428c2ba1de.1754925923.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dsi_vbt.c      | 2 +-
 drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
index e6a851d276f8..23402408e172 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt.c
@@ -777,7 +777,7 @@ bool intel_dsi_vbt_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	intel_dsi->init_count = mipi_config->master_init_timer;
 	intel_dsi->bw_timer = mipi_config->dbi_bw_timer;
 	intel_dsi->video_frmt_cfg_bits =
-		mipi_config->bta_enabled ? DISABLE_VIDEO_BTA : 0;
+		mipi_config->bta_disable ? DISABLE_VIDEO_BTA : 0;
 	intel_dsi->bgr_enabled = mipi_config->rgb_flip;
 
 	/* Starting point, adjusted depending on dual link and burst mode */
diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
index 7ac872dbba8d..3f9b9ed6592c 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
@@ -76,7 +76,7 @@ struct mipi_config {
 #define ENABLE_ROTATION_180	0x2
 #define ENABLE_ROTATION_270	0x3
 		u32 rotation:2;
-		u32 bta_enabled:1;
+		u32 bta_disable:1;
 		u32 rsvd2:15;
 	} __packed;
 

From 906cbdc1fffa3b18992108c8bda66de058285daa Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 11 Aug 2025 18:25:50 +0300
Subject: [PATCH 0378/1455] drm/i915/vbt: add missing DSI VBT defs

Add some missing DSI VBT definitions.

Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/17e0f38391314aceff12619a04829c3e36fa26b7.1754925923.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
index 3f9b9ed6592c..edc7331dcca2 100644
--- a/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_dsi_vbt_defs.h
@@ -94,10 +94,15 @@ struct mipi_config {
 #define DL_DCS_PORT_A_AND_C		0x02
 		u16 dl_dcs_cabc_ports:2;
 		u16 dl_dcs_backlight_ports:2;
-		u16 rsvd3:4;
+		u16 port_sync:1;				/* 219-230 */
+		u16 rsvd3:3;
 	} __packed;
 
-	u16 rsvd4;
+	/* DSI Controller Parameters */
+	struct {
+		u16 dsi_usage:1;
+		u16 rsvd4:15;
+	} __packed;
 
 	u8 rsvd5;
 	u32 target_burst_mode_freq;
@@ -118,7 +123,9 @@ struct mipi_config {
 		u16 dphy_param_valid:1;
 		u16 eot_pkt_disabled:1;
 		u16 enable_clk_stop:1;
-		u16 rsvd7:13;
+		u16 blanking_packets_during_bllp:1;		/* 219+ */
+		u16 lp_clock_during_lpm:1;			/* 219+ */
+		u16 rsvd7:11;
 	} __packed;
 
 	u32 hs_tx_timeout;

From 89f4b196ee4b056e0e8c179b247b29d4a71a4e7e Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 11 Aug 2025 11:01:48 +0300
Subject: [PATCH 0379/1455] drm/i915/lnl+/tc: Fix handling of an
 enabled/disconnected dp-alt sink

The TypeC PHY HW readout during driver loading and system resume
determines which TypeC mode the PHY is in (legacy/DP-alt/TBT-alt) and
whether the PHY is connected, based on the PHY's Owned and Ready flags.
For the PHY to be in DP-alt or legacy mode and for the PHY to be in the
connected state in these modes, both the Owned (set by the BIOS/driver)
and the Ready (set by the HW) flags should be set.

On ICL-MTL the HW kept the PHY's Ready flag set after the driver
connected the PHY by acquiring the PHY ownership (by setting the Owned
flag), until the driver disconnected the PHY by releasing the PHY
ownership (by clearing the Owned flag). On LNL+ this has changed, in
that the HW clears the Ready flag as soon as the sink gets disconnected,
even if the PHY ownership was acquired already and hence the PHY is
being used by the display.

When inheriting the HW state from BIOS for a PHY connected in DP-alt
mode on which the sink got disconnected - i.e. in a case where the sink
was connected while BIOS/GOP was running and so the sink got enabled
connecting the PHY, but the user disconnected the sink by the time the
driver loaded - the PHY Owned but not Ready state must be accounted for
on LNL+ according to the above. Do that by assuming on LNL+ that the PHY
is connected in DP-alt mode whenever the PHY Owned flag is set,
regardless of the PHY Ready flag.

This fixes a problem on LNL+, where the PHY TypeC mode / connected state
was detected incorrectly for a DP-alt sink, which got connected and then
disconnected by the user in the above way.

v2: Rename tc_phy_in_legacy_or_dp_alt_mode() to tc_phy_owned_by_display().
    (Luca, Jani)

Cc: Jani Nikula <jani.nikula@intel.com>
Cc: stable@vger.kernel.org # v6.8+
Reported-by: Charlton Lin <charlton.lin@intel.com>
Tested-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
[Imre: Add one-liner function documentation for tc_phy_owned_by_display()]
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250811080152.906216-2-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 3bc57579fe53..8208539bfe66 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -1226,14 +1226,19 @@ static void tc_phy_get_hw_state(struct intel_tc_port *tc)
 	tc->phy_ops->get_hw_state(tc);
 }
 
-static bool tc_phy_is_ready_and_owned(struct intel_tc_port *tc,
-				      bool phy_is_ready, bool phy_is_owned)
+/* Is the PHY owned by display i.e. is it in legacy or DP-alt mode? */
+static bool tc_phy_owned_by_display(struct intel_tc_port *tc,
+				    bool phy_is_ready, bool phy_is_owned)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
 
-	drm_WARN_ON(display->drm, phy_is_owned && !phy_is_ready);
+	if (DISPLAY_VER(display) < 20) {
+		drm_WARN_ON(display->drm, phy_is_owned && !phy_is_ready);
 
-	return phy_is_ready && phy_is_owned;
+		return phy_is_ready && phy_is_owned;
+	} else {
+		return phy_is_owned;
+	}
 }
 
 static bool tc_phy_is_connected(struct intel_tc_port *tc,
@@ -1244,7 +1249,7 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc,
 	bool phy_is_owned = tc_phy_is_owned(tc);
 	bool is_connected;
 
-	if (tc_phy_is_ready_and_owned(tc, phy_is_ready, phy_is_owned))
+	if (tc_phy_owned_by_display(tc, phy_is_ready, phy_is_owned))
 		is_connected = port_pll_type == ICL_PORT_DPLL_MG_PHY;
 	else
 		is_connected = port_pll_type == ICL_PORT_DPLL_DEFAULT;
@@ -1352,7 +1357,7 @@ tc_phy_get_current_mode(struct intel_tc_port *tc)
 	phy_is_ready = tc_phy_is_ready(tc);
 	phy_is_owned = tc_phy_is_owned(tc);
 
-	if (!tc_phy_is_ready_and_owned(tc, phy_is_ready, phy_is_owned)) {
+	if (!tc_phy_owned_by_display(tc, phy_is_ready, phy_is_owned)) {
 		mode = get_tc_mode_in_phy_not_owned_state(tc, live_mode);
 	} else {
 		drm_WARN_ON(display->drm, live_mode == TC_PORT_TBT_ALT);

From 3e32438fc406761f81b1928d210b3d2a5e7501a0 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 11 Aug 2025 11:01:49 +0300
Subject: [PATCH 0380/1455] drm/i915/icl+/tc: Cache the max lane count value

The PHY's pin assignment value in the TCSS_DDI_STATUS register - as set
by the HW/FW based on the connected DP-alt sink's TypeC/PD pin
assignment negotiation - gets cleared by the HW/FW on LNL+ as soon as
the sink gets disconnected, even if the PHY ownership got acquired
already by the driver (and hence the PHY itself is still connected and
used by the display). This is similar to how the PHY Ready flag gets
cleared on LNL+ in the same register.

To be able to query the max lane count value on LNL+ - which is based on
the above pin assignment - at all times even after the sink gets
disconnected, the max lane count must be determined and cached during
the PHY's HW readout and connect sequences. Do that here, leaving the
actual use of the cached value to a follow-up change.

v2: Don't read out the pin configuration if the PHY is disconnected.

Cc: stable@vger.kernel.org # v6.8+
Reported-by: Charlton Lin <charlton.lin@intel.com>
Tested-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250811080152.906216-3-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 57 +++++++++++++++++++++----
 1 file changed, 48 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 8208539bfe66..34435c4fc280 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -66,6 +66,7 @@ struct intel_tc_port {
 	enum tc_port_mode init_mode;
 	enum phy_fia phy_fia;
 	u8 phy_fia_idx;
+	u8 max_lane_count;
 };
 
 static enum intel_display_power_domain
@@ -365,12 +366,12 @@ static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 	}
 }
 
-int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
+static int get_max_lane_count(struct intel_tc_port *tc)
 {
-	struct intel_display *display = to_intel_display(dig_port);
-	struct intel_tc_port *tc = to_tc_port(dig_port);
+	struct intel_display *display = to_intel_display(tc->dig_port);
+	struct intel_digital_port *dig_port = tc->dig_port;
 
-	if (!intel_encoder_is_tc(&dig_port->base) || tc->mode != TC_PORT_DP_ALT)
+	if (tc->mode != TC_PORT_DP_ALT)
 		return 4;
 
 	assert_tc_cold_blocked(tc);
@@ -384,6 +385,21 @@ int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 	return intel_tc_port_get_max_lane_count(dig_port);
 }
 
+static void read_pin_configuration(struct intel_tc_port *tc)
+{
+	tc->max_lane_count = get_max_lane_count(tc);
+}
+
+int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
+{
+	struct intel_tc_port *tc = to_tc_port(dig_port);
+
+	if (!intel_encoder_is_tc(&dig_port->base))
+		return 4;
+
+	return get_max_lane_count(tc);
+}
+
 void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,
 				      int required_lanes)
 {
@@ -596,9 +612,12 @@ static void icl_tc_phy_get_hw_state(struct intel_tc_port *tc)
 	tc_cold_wref = __tc_cold_block(tc, &domain);
 
 	tc->mode = tc_phy_get_current_mode(tc);
-	if (tc->mode != TC_PORT_DISCONNECTED)
+	if (tc->mode != TC_PORT_DISCONNECTED) {
 		tc->lock_wakeref = tc_cold_block(tc);
 
+		read_pin_configuration(tc);
+	}
+
 	__tc_cold_unblock(tc, domain, tc_cold_wref);
 }
 
@@ -656,8 +675,11 @@ static bool icl_tc_phy_connect(struct intel_tc_port *tc,
 
 	tc->lock_wakeref = tc_cold_block(tc);
 
-	if (tc->mode == TC_PORT_TBT_ALT)
+	if (tc->mode == TC_PORT_TBT_ALT) {
+		read_pin_configuration(tc);
+
 		return true;
+	}
 
 	if ((!tc_phy_is_ready(tc) ||
 	     !icl_tc_phy_take_ownership(tc, true)) &&
@@ -668,6 +690,7 @@ static bool icl_tc_phy_connect(struct intel_tc_port *tc,
 		goto out_unblock_tc_cold;
 	}
 
+	read_pin_configuration(tc);
 
 	if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes))
 		goto out_release_phy;
@@ -858,9 +881,12 @@ static void adlp_tc_phy_get_hw_state(struct intel_tc_port *tc)
 	port_wakeref = intel_display_power_get(display, port_power_domain);
 
 	tc->mode = tc_phy_get_current_mode(tc);
-	if (tc->mode != TC_PORT_DISCONNECTED)
+	if (tc->mode != TC_PORT_DISCONNECTED) {
 		tc->lock_wakeref = tc_cold_block(tc);
 
+		read_pin_configuration(tc);
+	}
+
 	intel_display_power_put(display, port_power_domain, port_wakeref);
 }
 
@@ -873,6 +899,9 @@ static bool adlp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 
 	if (tc->mode == TC_PORT_TBT_ALT) {
 		tc->lock_wakeref = tc_cold_block(tc);
+
+		read_pin_configuration(tc);
+
 		return true;
 	}
 
@@ -894,6 +923,8 @@ static bool adlp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 
 	tc->lock_wakeref = tc_cold_block(tc);
 
+	read_pin_configuration(tc);
+
 	if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes))
 		goto out_unblock_tc_cold;
 
@@ -1124,9 +1155,12 @@ static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc)
 	tc_cold_wref = __tc_cold_block(tc, &domain);
 
 	tc->mode = tc_phy_get_current_mode(tc);
-	if (tc->mode != TC_PORT_DISCONNECTED)
+	if (tc->mode != TC_PORT_DISCONNECTED) {
 		tc->lock_wakeref = tc_cold_block(tc);
 
+		read_pin_configuration(tc);
+	}
+
 	drm_WARN_ON(display->drm,
 		    (tc->mode == TC_PORT_DP_ALT || tc->mode == TC_PORT_LEGACY) &&
 		    !xelpdp_tc_phy_tcss_power_is_enabled(tc));
@@ -1138,14 +1172,19 @@ static bool xelpdp_tc_phy_connect(struct intel_tc_port *tc, int required_lanes)
 {
 	tc->lock_wakeref = tc_cold_block(tc);
 
-	if (tc->mode == TC_PORT_TBT_ALT)
+	if (tc->mode == TC_PORT_TBT_ALT) {
+		read_pin_configuration(tc);
+
 		return true;
+	}
 
 	if (!xelpdp_tc_phy_enable_tcss_power(tc, true))
 		goto out_unblock_tccold;
 
 	xelpdp_tc_phy_take_ownership(tc, true);
 
+	read_pin_configuration(tc);
+
 	if (!tc_phy_verify_legacy_or_dp_alt_mode(tc, required_lanes))
 		goto out_release_phy;
 

From 33cf70bc0fe760224f892bc1854a33665f27d482 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 11 Aug 2025 11:01:50 +0300
Subject: [PATCH 0381/1455] drm/i915/lnl+/tc: Fix max lane count HW readout

On LNL+ for a disconnected sink the pin assignment value gets cleared by
the HW/FW as soon as the sink gets disconnected, even if the PHY
ownership got acquired already by the BIOS/driver (and hence the PHY
itself is still connected and used by the display). During HW readout
this can result in detecting the PHY's max lane count as 0 - matching
the above cleared aka NONE pin assignment HW state. For a connected PHY
the driver in general (outside of intel_tc.c) expects the max lane count
value to be valid for the video mode enabled on the corresponding output
(1, 2 or 4). Ensure this by setting the max lane count to 4 in this
case. Note, that it doesn't matter if this lane count happened to be
more than the max lane count with which the PHY got connected and
enabled, since the only thing the driver can do with such an output -
where the DP-alt sink is disconnected - is to disable the output.

v2: Rebased on change reading out the pin configuration only if the PHY
    is connected.

Cc: stable@vger.kernel.org # v6.8+
Reported-by: Charlton Lin <charlton.lin@intel.com>
Tested-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250811080152.906216-4-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 34435c4fc280..3f9842040bb0 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -23,6 +23,7 @@
 #include "intel_modeset_lock.h"
 #include "intel_tc.h"
 
+#define DP_PIN_ASSIGNMENT_NONE	0x0
 #define DP_PIN_ASSIGNMENT_C	0x3
 #define DP_PIN_ASSIGNMENT_D	0x4
 #define DP_PIN_ASSIGNMENT_E	0x5
@@ -308,6 +309,8 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 		REG_FIELD_GET(TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK, val);
 
 	switch (pin_assignment) {
+	case DP_PIN_ASSIGNMENT_NONE:
+		return 0;
 	default:
 		MISSING_CASE(pin_assignment);
 		fallthrough;
@@ -1159,6 +1162,12 @@ static void xelpdp_tc_phy_get_hw_state(struct intel_tc_port *tc)
 		tc->lock_wakeref = tc_cold_block(tc);
 
 		read_pin_configuration(tc);
+		/*
+		 * Set a valid lane count value for a DP-alt sink which got
+		 * disconnected. The driver can only disable the output on this PHY.
+		 */
+		if (tc->max_lane_count == 0)
+			tc->max_lane_count = 4;
 	}
 
 	drm_WARN_ON(display->drm,

From afc4e84388079f4d5ba05271632b7a4d8d85165c Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 11 Aug 2025 11:01:51 +0300
Subject: [PATCH 0382/1455] drm/i915/lnl+/tc: Use the cached max lane count
 value

Use the cached max lane count value on LNL+, to account for scenarios
where this value is queried after the HW cleared the corresponding pin
assignment value in the TCSS_DDI_STATUS register after the sink got
disconnected.

For consistency, follow-up changes will use the cached max lane count
value on other platforms as well and will also cache the pin assignment
value in a similar way.

Cc: stable@vger.kernel.org # v6.8+
Reported-by: Charlton Lin <charlton.lin@intel.com>
Tested-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250811080152.906216-5-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 3f9842040bb0..6a2442a0649e 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -395,12 +395,16 @@ static void read_pin_configuration(struct intel_tc_port *tc)
 
 int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 {
+	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 
 	if (!intel_encoder_is_tc(&dig_port->base))
 		return 4;
 
-	return get_max_lane_count(tc);
+	if (DISPLAY_VER(display) < 20)
+		return get_max_lane_count(tc);
+
+	return tc->max_lane_count;
 }
 
 void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,

From 6cb52cba474b2bec1a3018d3dbf75292059a29a1 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Mon, 11 Aug 2025 11:01:52 +0300
Subject: [PATCH 0383/1455] drm/i915/icl+/tc: Convert AUX powered WARN to a
 debug message

The BIOS can leave the AUX power well enabled on an output, even if this
isn't required (on platforms where the AUX power is only needed for an
AUX access). This was observed at least on PTL. To avoid the WARN which
would be triggered by this during the HW readout, convert the WARN to a
debug message.

Cc: stable@vger.kernel.org # v6.8+
Reported-by: Charlton Lin <charlton.lin@intel.com>
Tested-by: Khaled Almahallawy <khaled.almahallawy@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250811080152.906216-6-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 6a2442a0649e..668ef139391b 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -1498,11 +1498,11 @@ static void intel_tc_port_reset_mode(struct intel_tc_port *tc,
 	intel_display_power_flush_work(display);
 	if (!intel_tc_cold_requires_aux_pw(dig_port)) {
 		enum intel_display_power_domain aux_domain;
-		bool aux_powered;
 
 		aux_domain = intel_aux_power_domain(dig_port);
-		aux_powered = intel_display_power_is_enabled(display, aux_domain);
-		drm_WARN_ON(display->drm, aux_powered);
+		if (intel_display_power_is_enabled(display, aux_domain))
+			drm_dbg_kms(display->drm, "Port %s: AUX unexpectedly powered\n",
+				    tc->port_name);
 	}
 
 	tc_phy_disconnect(tc);

From 9337166fa1d80f7bb7c7d3a8f901f21c348c0f2a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 11 Aug 2025 12:43:57 +0200
Subject: [PATCH 0384/1455] drm/xe: Assign ioctl xe file handler to vm in
 xe_vm_create
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In several code paths, such as xe_pt_create(), the vm->xef field is used
to determine whether a VM originates from userspace or the kernel.

Previously, this handler was only assigned in xe_vm_create_ioctl(),
after the VM was created by xe_vm_create(). However, xe_vm_create()
triggers page table creation, and that function assumes vm->xef should
be already set. This could lead to incorrect origin detection.

To fix this problem and ensure consistency in the initialization of
the VM object, let's move the assignment of this handler to
xe_vm_create.

v2:
 - take reference to the xe file object only when xef is not NULL
 - release the reference to the xe file object on the error path (Matthew)

Fixes: 7f387e6012b6 ("drm/xe: add XE_BO_FLAG_PINNED_LATE_RESTORE")
Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250811104358.2064150-2-piotr.piorkowski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_migrate.c    |  2 +-
 drivers/gpu/drm/xe/xe_pxp_submit.c |  2 +-
 drivers/gpu/drm/xe/xe_vm.c         | 11 ++++++-----
 drivers/gpu/drm/xe/xe_vm.h         |  2 +-
 4 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 45cde272d45f..ddfad7506a82 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -409,7 +409,7 @@ int xe_migrate_init(struct xe_migrate *m)
 
 	/* Special layout, prepared below.. */
 	vm = xe_vm_create(xe, XE_VM_FLAG_MIGRATION |
-			  XE_VM_FLAG_SET_TILE_ID(tile));
+			  XE_VM_FLAG_SET_TILE_ID(tile), NULL);
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c
index d92ec0f515b0..ca95f2a4d4ef 100644
--- a/drivers/gpu/drm/xe/xe_pxp_submit.c
+++ b/drivers/gpu/drm/xe/xe_pxp_submit.c
@@ -101,7 +101,7 @@ static int allocate_gsc_client_resources(struct xe_gt *gt,
 	xe_assert(xe, hwe);
 
 	/* PXP instructions must be issued from PPGTT */
-	vm = xe_vm_create(xe, XE_VM_FLAG_GSC);
+	vm = xe_vm_create(xe, XE_VM_FLAG_GSC, NULL);
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 97913f496a46..6c52a0a610c8 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1665,7 +1665,7 @@ static void xe_vm_free_scratch(struct xe_vm *vm)
 	}
 }
 
-struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags)
+struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 {
 	struct drm_gem_object *vm_resv_obj;
 	struct xe_vm *vm;
@@ -1686,9 +1686,10 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags)
 	vm->xe = xe;
 
 	vm->size = 1ull << xe->info.va_bits;
-
 	vm->flags = flags;
 
+	if (xef)
+		vm->xef = xe_file_get(xef);
 	/**
 	 * GSC VMs are kernel-owned, only used for PXP ops and can sometimes be
 	 * manipulated under the PXP mutex. However, the PXP mutex can be taken
@@ -1839,6 +1840,8 @@ err_no_resv:
 	for_each_tile(tile, xe, id)
 		xe_range_fence_tree_fini(&vm->rftree[id]);
 	ttm_lru_bulk_move_fini(&xe->ttm, &vm->lru_bulk_move);
+	if (vm->xef)
+		xe_file_put(vm->xef);
 	kfree(vm);
 	if (flags & XE_VM_FLAG_LR_MODE)
 		xe_pm_runtime_put(xe);
@@ -2121,7 +2124,7 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data,
 	if (args->flags & DRM_XE_VM_CREATE_FLAG_FAULT_MODE)
 		flags |= XE_VM_FLAG_FAULT_MODE;
 
-	vm = xe_vm_create(xe, flags);
+	vm = xe_vm_create(xe, flags, xef);
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
@@ -2137,8 +2140,6 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data,
 		vm->usm.asid = asid;
 	}
 
-	vm->xef = xe_file_get(xef);
-
 	/* Record BO memory for VM pagetable created against client */
 	for_each_tile(tile, xe, id)
 		if (vm->pt_root[id])
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 3475a118f666..2f213737c7e5 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -26,7 +26,7 @@ struct xe_sync_entry;
 struct xe_svm_range;
 struct drm_exec;
 
-struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags);
+struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef);
 
 struct xe_vm *xe_vm_lookup(struct xe_file *xef, u32 id);
 int xe_vma_cmp_vma_cb(const void *key, const struct rb_node *node);

From 30e0c3f43a414616e0b6ca76cf7f7b2cd387e1d4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Piotr=20Pi=C3=B3rkowski?= <piotr.piorkowski@intel.com>
Date: Mon, 11 Aug 2025 12:43:58 +0200
Subject: [PATCH 0385/1455] drm/xe: Move ASID allocation and user PT BO
 tracking into xe_vm_create
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, ASID assignment for user VMs and page-table BO accounting for
client memory tracking are performed in xe_vm_create_ioctl.
To consolidate VM object initialization, move this logic to
xe_vm_create.

v2:
 - removed unnecessary duplicate BO tracking code
 - using the local variable xef to verify whether the VM is being created
   by userspace

Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250811104358.2064150-3-piotr.piorkowski@intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c | 34 +++++++++++++++-------------------
 1 file changed, 15 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 6c52a0a610c8..d40d2d43c041 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1820,6 +1820,20 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 	if (number_tiles > 1)
 		vm->composite_fence_ctx = dma_fence_context_alloc(1);
 
+	if (xef && xe->info.has_asid) {
+		u32 asid;
+
+		down_write(&xe->usm.lock);
+		err = xa_alloc_cyclic(&xe->usm.asid_to_vm, &asid, vm,
+				      XA_LIMIT(1, XE_MAX_ASID - 1),
+				      &xe->usm.next_asid, GFP_KERNEL);
+		up_write(&xe->usm.lock);
+		if (err < 0)
+			goto err_unlock_close;
+
+		vm->usm.asid = asid;
+	}
+
 	trace_xe_vm_create(vm);
 
 	return vm;
@@ -2086,9 +2100,8 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data,
 	struct xe_device *xe = to_xe_device(dev);
 	struct xe_file *xef = to_xe_file(file);
 	struct drm_xe_vm_create *args = data;
-	struct xe_tile *tile;
 	struct xe_vm *vm;
-	u32 id, asid;
+	u32 id;
 	int err;
 	u32 flags = 0;
 
@@ -2128,23 +2141,6 @@ int xe_vm_create_ioctl(struct drm_device *dev, void *data,
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
-	if (xe->info.has_asid) {
-		down_write(&xe->usm.lock);
-		err = xa_alloc_cyclic(&xe->usm.asid_to_vm, &asid, vm,
-				      XA_LIMIT(1, XE_MAX_ASID - 1),
-				      &xe->usm.next_asid, GFP_KERNEL);
-		up_write(&xe->usm.lock);
-		if (err < 0)
-			goto err_close_and_put;
-
-		vm->usm.asid = asid;
-	}
-
-	/* Record BO memory for VM pagetable created against client */
-	for_each_tile(tile, xe, id)
-		if (vm->pt_root[id])
-			xe_drm_client_add_bo(vm->xef->client, vm->pt_root[id]->bo);
-
 #if IS_ENABLED(CONFIG_DRM_XE_DEBUG_MEM)
 	/* Warning: Security issue - never enable by default */
 	args->reserved[0] = xe_bo_main_addr(vm->pt_root[0]->bo, XE_PAGE_SIZE);

From 32a4d1b98e6663101fd0abfaf151c48feea7abb1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Micha=C5=82=20Winiarski?= <michal.winiarski@intel.com>
Date: Tue, 27 May 2025 14:06:37 +0200
Subject: [PATCH 0386/1455] drm/xe/pf: Set VF LMEM BAR size
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

LMEM is partitioned between multiple VFs and we expect that the more
VFs we have, the less LMEM is assigned to each VF.
This means that we can achieve full LMEM BAR access without the need to
attempt full VF LMEM BAR resize via pci_resize_resource().

Always try to set the largest possible BAR size that allows to fit the
number of enabled VFs and inform the user in case the resize attempt is
not successful.

Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20250527120637.665506-7-michal.winiarski@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_bars.h |  1 +
 drivers/gpu/drm/xe/xe_pci_sriov.c | 22 ++++++++++++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/xe/regs/xe_bars.h b/drivers/gpu/drm/xe/regs/xe_bars.h
index ce05b6ae832f..880140d6ccdc 100644
--- a/drivers/gpu/drm/xe/regs/xe_bars.h
+++ b/drivers/gpu/drm/xe/regs/xe_bars.h
@@ -7,5 +7,6 @@
 
 #define GTTMMADR_BAR			0 /* MMIO + GTT */
 #define LMEM_BAR			2 /* VRAM */
+#define VF_LMEM_BAR			9 /* VF VRAM */
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_pci_sriov.c b/drivers/gpu/drm/xe/xe_pci_sriov.c
index 447a7867eecb..af05db07162e 100644
--- a/drivers/gpu/drm/xe/xe_pci_sriov.c
+++ b/drivers/gpu/drm/xe/xe_pci_sriov.c
@@ -3,6 +3,10 @@
  * Copyright © 2023-2024 Intel Corporation
  */
 
+#include <linux/bitops.h>
+#include <linux/pci.h>
+
+#include "regs/xe_bars.h"
 #include "xe_assert.h"
 #include "xe_device.h"
 #include "xe_gt_sriov_pf_config.h"
@@ -128,6 +132,18 @@ static void pf_engine_activity_stats(struct xe_device *xe, unsigned int num_vfs,
 	}
 }
 
+static int resize_vf_vram_bar(struct xe_device *xe, int num_vfs)
+{
+	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
+	u32 sizes;
+
+	sizes = pci_iov_vf_bar_get_sizes(pdev, VF_LMEM_BAR, num_vfs);
+	if (!sizes)
+		return 0;
+
+	return pci_iov_vf_bar_set_size(pdev, VF_LMEM_BAR, __fls(sizes));
+}
+
 static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
 {
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
@@ -158,6 +174,12 @@ static int pf_enable_vfs(struct xe_device *xe, int num_vfs)
 	if (err < 0)
 		goto failed;
 
+	if (IS_DGFX(xe)) {
+		err = resize_vf_vram_bar(xe, num_vfs);
+		if (err)
+			xe_sriov_info(xe, "Failed to set VF LMEM BAR size: %d\n", err);
+	}
+
 	err = pci_enable_sriov(pdev, num_vfs);
 	if (err < 0)
 		goto failed;

From d543489aa1f141901a078dd84daaa16fcab0cf7f Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 17 Jul 2025 16:35:29 +0530
Subject: [PATCH 0387/1455] drm/amdgpu: Add description for partition commands

Add string description for partition commands.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 75911600d504..3d3815b9ff15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -666,6 +666,10 @@ static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id)
 		return "FB_FW_RESERV_ADDR";
 	case GFX_CMD_ID_FB_FW_RESERV_EXT_ADDR:
 		return "FB_FW_RESERV_EXT_ADDR";
+	case GFX_CMD_ID_SRIOV_SPATIAL_PART:
+		return "SPATIAL_PARTITION";
+	case GFX_CMD_ID_FB_NPS_MODE:
+		return "NPS_MODE_CHANGE";
 	default:
 		return "UNKNOWN CMD";
 	}

From 065e23170a1e09bc9104b761183e59562a029619 Mon Sep 17 00:00:00 2001
From: Frank Min <Frank.Min@amd.com>
Date: Tue, 5 Aug 2025 22:30:54 +0800
Subject: [PATCH 0388/1455] drm/amdgpu: Add PSP fw version check for fw reserve
 GFX command

The fw reserved GFX command is only supported starting from PSP fw
version 0x3a0e14 and 0x3b0e0d. Older versions do not support this command.

Add a version guard to ensure the command is only used when the running
PSP fw meets the minimum version requirement.

This ensures backward compatibility and safe operation across fw
revisions.

Fixes: a3b7f9c306e1 ("drm/amdgpu: reclaim psp fw reservation memory region")
Signed-off-by: Frank Min <Frank.Min@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 3d3815b9ff15..771a43681d72 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -1043,15 +1043,28 @@ int psp_update_fw_reservation(struct psp_context *psp)
 {
 	int ret;
 	uint64_t reserv_addr, reserv_addr_ext;
-	uint32_t reserv_size, reserv_size_ext;
+	uint32_t reserv_size, reserv_size_ext, mp0_ip_ver;
 	struct amdgpu_device *adev = psp->adev;
 
+	mp0_ip_ver = amdgpu_ip_version(adev, MP0_HWIP, 0);
+
 	if (amdgpu_sriov_vf(psp->adev))
 		return 0;
 
-	if ((amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(14, 0, 2)) &&
-	    (amdgpu_ip_version(adev, MP0_HWIP, 0) != IP_VERSION(14, 0, 3)))
+	switch (mp0_ip_ver) {
+	case IP_VERSION(14, 0, 2):
+		if (adev->psp.sos.fw_version < 0x3b0e0d)
+			return 0;
+		break;
+
+	case IP_VERSION(14, 0, 3):
+		if (adev->psp.sos.fw_version < 0x3a0e14)
+			return 0;
+		break;
+
+	default:
 		return 0;
+	}
 
 	ret = psp_get_fw_reservation_info(psp, GFX_CMD_ID_FB_FW_RESERV_ADDR, &reserv_addr, &reserv_size);
 	if (ret)

From 57af162bfc8c05332a28c4d458d246cc46d2746d Mon Sep 17 00:00:00 2001
From: Geoffrey McRae <geoffrey.mcrae@amd.com>
Date: Tue, 8 Jul 2025 13:53:40 +1000
Subject: [PATCH 0389/1455] drm/amdkfd: return -ENOTTY for unsupported IOCTLs

Some kfd ioctls may not be available depending on the kernel version the
user is running, as such we need to report -ENOTTY so userland can
determine the cause of the ioctl failure.

Signed-off-by: Geoffrey McRae <geoffrey.mcrae@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 828a9ceef1e7..79ed3be63d0d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -3252,8 +3252,10 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
 	int retcode = -EINVAL;
 	bool ptrace_attached = false;
 
-	if (nr >= AMDKFD_CORE_IOCTL_COUNT)
+	if (nr >= AMDKFD_CORE_IOCTL_COUNT) {
+		retcode = -ENOTTY;
 		goto err_i1;
+	}
 
 	if ((nr >= AMDKFD_COMMAND_START) && (nr < AMDKFD_COMMAND_END)) {
 		u32 amdkfd_size;
@@ -3266,8 +3268,10 @@ static long kfd_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
 			asize = amdkfd_size;
 
 		cmd = ioctl->cmd;
-	} else
+	} else {
+		retcode = -ENOTTY;
 		goto err_i1;
+	}
 
 	dev_dbg(kfd_device, "ioctl cmd 0x%x (#0x%x), arg 0x%lx\n", cmd, nr, arg);
 

From d38eaf27de1b8584f42d6fb3f717b7ec44b3a7a1 Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Tue, 12 Aug 2025 09:17:58 +0800
Subject: [PATCH 0390/1455] drm/amdgpu: fix vram reservation issue

The vram block allocation flag must be cleared
before making vram reservation, otherwise reserving
addresses within the currently freed memory range
will always fail.

Fixes: c9cad937c0c5 ("drm/amdgpu: add drm buddy support to amdgpu")
Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index 4bf3e99f47fe..40049b885667 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -685,9 +685,8 @@ static void amdgpu_vram_mgr_del(struct ttm_resource_manager *man,
 	list_for_each_entry(block, &vres->blocks, link)
 		vis_usage += amdgpu_vram_mgr_vis_size(adev, block);
 
-	amdgpu_vram_mgr_do_reserve(man);
-
 	drm_buddy_free_list(mm, &vres->blocks, vres->flags);
+	amdgpu_vram_mgr_do_reserve(man);
 	mutex_unlock(&mgr->lock);
 
 	atomic64_sub(vis_usage, &mgr->vis_usage);

From b08425fa77ad2f305fe57a33dceb456be03b653f Mon Sep 17 00:00:00 2001
From: Jack Xiao <Jack.Xiao@amd.com>
Date: Mon, 11 Aug 2025 15:20:55 +0800
Subject: [PATCH 0391/1455] drm/amdgpu: fix incorrect vm flags to map bo

It should use vm flags instead of pte flags
to specify bo vm attributes.

Fixes: 7946340fa389 ("drm/amdgpu: Move csa related code to separate file")
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Likun Gao <Likun.Gao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
index 02138aa55793..dfb6cfd83760 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -88,8 +88,8 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	}
 
 	r = amdgpu_vm_bo_map(adev, *bo_va, csa_addr, 0, size,
-			     AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
-			     AMDGPU_PTE_EXECUTABLE);
+			     AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE |
+			     AMDGPU_VM_PAGE_EXECUTABLE);
 
 	if (r) {
 		DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r);

From f101c13a8720c73e67f8f9d511fbbeda95bcedb1 Mon Sep 17 00:00:00 2001
From: Liu01 Tong <Tong.Liu01@amd.com>
Date: Mon, 11 Aug 2025 14:52:37 +0800
Subject: [PATCH 0392/1455] drm/amdgpu: fix task hang from failed job
 submission during process kill
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

During process kill, drm_sched_entity_flush() will kill the vm
entities. The following job submissions of this process will fail, and
the resources of these jobs have not been released, nor have the fences
been signalled, causing tasks to hang and timeout.

Fix by check entity status in amdgpu_vm_ready() and avoid submit jobs to
stopped entity.

v2: add amdgpu_vm_ready() check before amdgpu_vm_clear_freed() in
function amdgpu_cs_vm_handling().

Signed-off-by: Liu01 Tong <Tong.Liu01@amd.com>
Signed-off-by: Lin.Cao <lincao12@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c |  3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 15 +++++++++++----
 2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 27f8f316f6c2..2ac9729e4c86 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1139,6 +1139,9 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p)
 		}
 	}
 
+	if (!amdgpu_vm_ready(vm))
+		return -EINVAL;
+
 	r = amdgpu_vm_clear_freed(adev, vm, NULL);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 39b4250ede0f..bd12d8ff15a4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -654,11 +654,10 @@ int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm,
  * Check if all VM PDs/PTs are ready for updates
  *
  * Returns:
- * True if VM is not evicting.
+ * True if VM is not evicting and all VM entities are not stopped
  */
 bool amdgpu_vm_ready(struct amdgpu_vm *vm)
 {
-	bool empty;
 	bool ret;
 
 	amdgpu_vm_eviction_lock(vm);
@@ -666,10 +665,18 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm)
 	amdgpu_vm_eviction_unlock(vm);
 
 	spin_lock(&vm->status_lock);
-	empty = list_empty(&vm->evicted);
+	ret &= list_empty(&vm->evicted);
 	spin_unlock(&vm->status_lock);
 
-	return ret && empty;
+	spin_lock(&vm->immediate.lock);
+	ret &= !vm->immediate.stopped;
+	spin_unlock(&vm->immediate.lock);
+
+	spin_lock(&vm->delayed.lock);
+	ret &= !vm->delayed.stopped;
+	spin_unlock(&vm->delayed.lock);
+
+	return ret;
 }
 
 /**

From b4892169025f6753e4e1a72a94ab43fbc7dd4e47 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Jul 2025 14:17:08 +0300
Subject: [PATCH 0393/1455] drm/i915/display: add intel_dig_port_alloc()

Add a common allocator function for struct intel_digital_port, with some
member default initialization to deduplicate them from everywhere
else. This is similar to intel_connector_alloc().

At least for now, place this in intel_encoder.[ch]. We don't have a
dedicated file for dig port stuff, and there wouldn't be much to add
there anyway. A digital port is a sort of subclass of encoder, so the
location isn't far off the mark.

Reviewed-by: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Link: https://lore.kernel.org/r/4d2da1a40698f85014140f586405b19795437e81.1753787803.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/g4x_dp.c        |  7 +------
 drivers/gpu/drm/i915/display/g4x_hdmi.c      |  9 ++-------
 drivers/gpu/drm/i915/display/intel_ddi.c     |  8 +-------
 drivers/gpu/drm/i915/display/intel_encoder.c | 18 ++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_encoder.h |  3 +++
 5 files changed, 25 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_dp.c b/drivers/gpu/drm/i915/display/g4x_dp.c
index 846dbd8ae931..aa159f9ce12f 100644
--- a/drivers/gpu/drm/i915/display/g4x_dp.c
+++ b/drivers/gpu/drm/i915/display/g4x_dp.c
@@ -1300,12 +1300,10 @@ bool g4x_dp_init(struct intel_display *display,
 		drm_dbg_kms(display->drm, "No VBT child device for DP-%c\n",
 			    port_name(port));
 
-	dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL);
+	dig_port = intel_dig_port_alloc();
 	if (!dig_port)
 		return false;
 
-	dig_port->aux_ch = AUX_CH_NONE;
-
 	intel_connector = intel_connector_alloc();
 	if (!intel_connector)
 		goto err_connector_alloc;
@@ -1315,8 +1313,6 @@ bool g4x_dp_init(struct intel_display *display,
 
 	intel_encoder->devdata = devdata;
 
-	mutex_init(&dig_port->hdcp.mutex);
-
 	if (drm_encoder_init(display->drm, &intel_encoder->base,
 			     &intel_dp_enc_funcs, DRM_MODE_ENCODER_TMDS,
 			     "DP %c", port_name(port)))
@@ -1386,7 +1382,6 @@ bool g4x_dp_init(struct intel_display *display,
 	}
 
 	dig_port->dp.output_reg = output_reg;
-	dig_port->max_lanes = 4;
 
 	intel_encoder->type = INTEL_OUTPUT_DP;
 	intel_encoder->power_domain = intel_display_power_ddi_lanes_domain(display, port);
diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c
index 2610f5702fb9..108ebd97f9e4 100644
--- a/drivers/gpu/drm/i915/display/g4x_hdmi.c
+++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c
@@ -19,6 +19,7 @@
 #include "intel_display_types.h"
 #include "intel_dp_aux.h"
 #include "intel_dpio_phy.h"
+#include "intel_encoder.h"
 #include "intel_fdi.h"
 #include "intel_fifo_underrun.h"
 #include "intel_hdmi.h"
@@ -690,12 +691,10 @@ bool g4x_hdmi_init(struct intel_display *display,
 		drm_dbg_kms(display->drm, "No VBT child device for HDMI-%c\n",
 			    port_name(port));
 
-	dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL);
+	dig_port = intel_dig_port_alloc();
 	if (!dig_port)
 		return false;
 
-	dig_port->aux_ch = AUX_CH_NONE;
-
 	intel_connector = intel_connector_alloc();
 	if (!intel_connector)
 		goto err_connector_alloc;
@@ -704,8 +703,6 @@ bool g4x_hdmi_init(struct intel_display *display,
 
 	intel_encoder->devdata = devdata;
 
-	mutex_init(&dig_port->hdcp.mutex);
-
 	if (drm_encoder_init(display->drm, &intel_encoder->base,
 			     &intel_hdmi_enc_funcs, DRM_MODE_ENCODER_TMDS,
 			     "HDMI %c", port_name(port)))
@@ -767,8 +764,6 @@ bool g4x_hdmi_init(struct intel_display *display,
 		intel_encoder->cloneable |= BIT(INTEL_OUTPUT_HDMI);
 
 	dig_port->hdmi.hdmi_reg = hdmi_reg;
-	dig_port->dp.output_reg = INVALID_MMIO_REG;
-	dig_port->max_lanes = 4;
 
 	intel_infoframe_init(dig_port);
 
diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 866ed3e46664..18040f2c8c0d 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -5158,12 +5158,10 @@ void intel_ddi_init(struct intel_display *display,
 			    phy_name(phy));
 	}
 
-	dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL);
+	dig_port = intel_dig_port_alloc();
 	if (!dig_port)
 		return;
 
-	dig_port->aux_ch = AUX_CH_NONE;
-
 	encoder = &dig_port->base;
 	encoder->devdata = devdata;
 
@@ -5201,9 +5199,6 @@ void intel_ddi_init(struct intel_display *display,
 
 	intel_encoder_link_check_init(encoder, intel_ddi_link_check);
 
-	mutex_init(&dig_port->hdcp.mutex);
-	dig_port->hdcp.num_streams = 0;
-
 	encoder->hotplug = intel_ddi_hotplug;
 	encoder->compute_output_type = intel_ddi_compute_output_type;
 	encoder->compute_config = intel_ddi_compute_config;
@@ -5341,7 +5336,6 @@ void intel_ddi_init(struct intel_display *display,
 
 	dig_port->ddi_a_4_lanes = DISPLAY_VER(display) < 11 && ddi_buf_ctl & DDI_A_4_LANES;
 
-	dig_port->dp.output_reg = INVALID_MMIO_REG;
 	dig_port->max_lanes = intel_ddi_max_lanes(dig_port);
 
 	if (need_aux_ch(encoder, init_dp)) {
diff --git a/drivers/gpu/drm/i915/display/intel_encoder.c b/drivers/gpu/drm/i915/display/intel_encoder.c
index 4e2b77b87678..2ffe1f251ef8 100644
--- a/drivers/gpu/drm/i915/display/intel_encoder.c
+++ b/drivers/gpu/drm/i915/display/intel_encoder.c
@@ -103,3 +103,21 @@ void intel_encoder_shutdown_all(struct intel_display *display)
 		if (encoder->shutdown_complete)
 			encoder->shutdown_complete(encoder);
 }
+
+struct intel_digital_port *intel_dig_port_alloc(void)
+{
+	struct intel_digital_port *dig_port;
+
+	dig_port = kzalloc(sizeof(*dig_port), GFP_KERNEL);
+	if (!dig_port)
+		return NULL;
+
+	dig_port->hdmi.hdmi_reg = INVALID_MMIO_REG;
+	dig_port->dp.output_reg = INVALID_MMIO_REG;
+	dig_port->aux_ch = AUX_CH_NONE;
+	dig_port->max_lanes = 4;
+
+	mutex_init(&dig_port->hdcp.mutex);
+
+	return dig_port;
+}
diff --git a/drivers/gpu/drm/i915/display/intel_encoder.h b/drivers/gpu/drm/i915/display/intel_encoder.h
index e1d3aeab7c00..ace0fe1a8f27 100644
--- a/drivers/gpu/drm/i915/display/intel_encoder.h
+++ b/drivers/gpu/drm/i915/display/intel_encoder.h
@@ -6,6 +6,7 @@
 #ifndef __INTEL_ENCODER_H__
 #define __INTEL_ENCODER_H__
 
+struct intel_digital_port;
 struct intel_display;
 struct intel_encoder;
 
@@ -20,4 +21,6 @@ void intel_encoder_shutdown_all(struct intel_display *display);
 void intel_encoder_block_all_hpds(struct intel_display *display);
 void intel_encoder_unblock_all_hpds(struct intel_display *display);
 
+struct intel_digital_port *intel_dig_port_alloc(void);
+
 #endif /* __INTEL_ENCODER_H__ */

From a985ecc2bec87dba544f8487b84d6523dcbb321b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 29 Jul 2025 14:17:09 +0300
Subject: [PATCH 0394/1455] drm/i915/connector: make intel_connector_init()
 static

intel_connector_init() is only used in intel_connector.c. Make it
static.

Reviewed-by: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Link: https://lore.kernel.org/r/46443c16f9cbff039cd3c830871289ab17110905.1753787803.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_connector.c | 2 +-
 drivers/gpu/drm/i915/display/intel_connector.h | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_connector.c b/drivers/gpu/drm/i915/display/intel_connector.c
index 42c923f416b3..6a55854db5b6 100644
--- a/drivers/gpu/drm/i915/display/intel_connector.c
+++ b/drivers/gpu/drm/i915/display/intel_connector.c
@@ -77,7 +77,7 @@ void intel_connector_cancel_modeset_retry_work(struct intel_connector *connector
 		drm_connector_put(&connector->base);
 }
 
-int intel_connector_init(struct intel_connector *connector)
+static int intel_connector_init(struct intel_connector *connector)
 {
 	struct intel_digital_connector_state *conn_state;
 
diff --git a/drivers/gpu/drm/i915/display/intel_connector.h b/drivers/gpu/drm/i915/display/intel_connector.h
index aafb25a814fa..0aa86626e646 100644
--- a/drivers/gpu/drm/i915/display/intel_connector.h
+++ b/drivers/gpu/drm/i915/display/intel_connector.h
@@ -14,7 +14,6 @@ struct i2c_adapter;
 struct intel_connector;
 struct intel_encoder;
 
-int intel_connector_init(struct intel_connector *connector);
 struct intel_connector *intel_connector_alloc(void);
 void intel_connector_free(struct intel_connector *connector);
 void intel_connector_destroy(struct drm_connector *connector);

From e0ed674acbac5635494fc7b278abc9505eaa8e71 Mon Sep 17 00:00:00 2001
From: Mel Henning <mhenning@darkrefraction.com>
Date: Mon, 11 Aug 2025 17:32:31 -0400
Subject: [PATCH 0395/1455] drm/nouveau: Remove DRM_NOUVEAU_GSP_DEFAULT config

This option was originally intoduced because the GSP code path was
not well tested and we wanted to leave it up to distros which code path
they shipped by default. By now though, the GSP path is probably better
tested than the old firmware eg. Fedora ships GSP by default and we
generally run CTS on GSP. We've always been GSP-only on Ada and later.

So, this path removes the option and effectively sets the option to
always on. We still fall back to the old firmware if GSP is not found.
This change only affects Turing and Ampere.

Users can still set nouveau.config=NvGspRm=0 on the kernel command line
to force using the old firmware on Turing/Ampere.

Signed-off-by: Mel Henning <mhenning@darkrefraction.com>
Reviewed-by: Ben Skeggs <bskeggs@nvidia.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250811213843.4294-2-mhenning@darkrefraction.com
---
 drivers/gpu/drm/nouveau/Kconfig                 | 8 --------
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c | 6 +-----
 2 files changed, 1 insertion(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index d1587639ebb0..c88776d1e784 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -102,14 +102,6 @@ config DRM_NOUVEAU_SVM
 	  Say Y here if you want to enable experimental support for
 	  Shared Virtual Memory (SVM).
 
-config DRM_NOUVEAU_GSP_DEFAULT
-	bool "Use GSP firmware for Turing/Ampere (needs firmware installed)"
-	depends on DRM_NOUVEAU
-	default n
-	help
-	  Say Y here if you want to use the GSP codepaths by default on
-	  Turing and Ampere GPUs.
-
 config DRM_NOUVEAU_CH7006
 	tristate "Chrontel ch7006 TV encoder"
 	depends on DRM_NOUVEAU
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c
index 58e233bc53b1..81e56da0474a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/tu102.c
@@ -383,13 +383,9 @@ int
 tu102_gsp_load_rm(struct nvkm_gsp *gsp, const struct nvkm_gsp_fwif *fwif)
 {
 	struct nvkm_subdev *subdev = &gsp->subdev;
-	bool enable_gsp = fwif->enable;
 	int ret;
 
-#if IS_ENABLED(CONFIG_DRM_NOUVEAU_GSP_DEFAULT)
-	enable_gsp = true;
-#endif
-	if (!nvkm_boolopt(subdev->device->cfgopt, "NvGspRm", enable_gsp))
+	if (!nvkm_boolopt(subdev->device->cfgopt, "NvGspRm", true))
 		return -EINVAL;
 
 	ret = nvkm_gsp_load_fw(gsp, "gsp", fwif->ver, &gsp->fws.rm);

From 2e308a935fa1307ea802bf65d8cc8507244d3b23 Mon Sep 17 00:00:00 2001
From: Mel Henning <mhenning@darkrefraction.com>
Date: Mon, 11 Aug 2025 17:32:32 -0400
Subject: [PATCH 0396/1455] drm/nouveau: Remove nvkm_gsp_fwif.enable

This struct element is no longer used.

Signed-off-by: Mel Henning <mhenning@darkrefraction.com>
Reviewed-by: Ben Skeggs <bskeggs@nvidia.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250811213843.4294-3-mhenning@darkrefraction.com
---
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c | 4 ++--
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c | 2 +-
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c | 2 +-
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c | 2 +-
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h  | 1 -
 5 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c
index eb765da0876e..35d1fcef520b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/ad102.c
@@ -41,8 +41,8 @@ ad102_gsp = {
 
 static struct nvkm_gsp_fwif
 ad102_gsps[] = {
-	{ 1, tu102_gsp_load, &ad102_gsp, &r570_rm_ga102, "570.144", true },
-	{ 0, tu102_gsp_load, &ad102_gsp, &r535_rm_ga102, "535.113.01", true },
+	{ 1, tu102_gsp_load, &ad102_gsp, &r570_rm_ga102, "570.144" },
+	{ 0, tu102_gsp_load, &ad102_gsp, &r535_rm_ga102, "535.113.01" },
 	{}
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c
index 12a3f2c1ed82..1b3b31b95ce4 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb100.c
@@ -20,7 +20,7 @@ gb100_gsp = {
 
 static struct nvkm_gsp_fwif
 gb100_gsps[] = {
-	{ 0, gh100_gsp_load, &gb100_gsp, &r570_rm_gb10x, "570.144", true },
+	{ 0, gh100_gsp_load, &gb100_gsp, &r570_rm_gb10x, "570.144" },
 	{}
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c
index c1d718172ddf..51384c63148c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gb202.c
@@ -20,7 +20,7 @@ gb202_gsp = {
 
 static struct nvkm_gsp_fwif
 gb202_gsps[] = {
-	{ 0, gh100_gsp_load, &gb202_gsp, &r570_rm_gb20x, "570.144", true },
+	{ 0, gh100_gsp_load, &gb202_gsp, &r570_rm_gb20x, "570.144" },
 	{}
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c
index ce31e8248807..b0dd5fce7bad 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/gh100.c
@@ -344,7 +344,7 @@ done:
 
 static struct nvkm_gsp_fwif
 gh100_gsps[] = {
-	{ 0, gh100_gsp_load, &gh100_gsp, &r570_rm_gh100, "570.144", true },
+	{ 0, gh100_gsp_load, &gh100_gsp, &r570_rm_gh100, "570.144" },
 	{}
 };
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h
index 4f14e85fc69e..c3494b7ac572 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/priv.h
@@ -14,7 +14,6 @@ struct nvkm_gsp_fwif {
 	const struct nvkm_gsp_func *func;
 	const struct nvkm_rm_impl *rm;
 	const char *ver;
-	bool enable;
 };
 
 int nvkm_gsp_load_fw(struct nvkm_gsp *, const char *name, const char *ver,

From a3daf184bd85d7c08ce948a79bb0e4cac2203923 Mon Sep 17 00:00:00 2001
From: Mel Henning <mhenning@darkrefraction.com>
Date: Mon, 11 Aug 2025 17:32:33 -0400
Subject: [PATCH 0397/1455] drm/nouveau: Improve message for missing firmware

This is inteded to address concerns that users might get cryptic error
messages or a failure to boot if they set nouveau.config=NvGspRm=0 on
the kernel command line and their gpu requires gsp (Ada or newer).
With this patch, that configuration results in error messages like this:

nouveau 0000:01:00.0: gsp: Failed to load required firmware for device.
nouveau 0000:01:00.0: gsp ctor failed: -22
nouveau 0000:01:00.0: probe with driver nouveau failed with error -22

When nouveau fails to load like this, we still fall back to the generic
framebuffer device, so users will still have limited graphical output.

Signed-off-by: Mel Henning <mhenning@darkrefraction.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250811213843.4294-4-mhenning@darkrefraction.com
---
 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c
index d23243a83a4c..7ccb41761066 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/gsp/base.c
@@ -138,8 +138,10 @@ nvkm_gsp_new_(const struct nvkm_gsp_fwif *fwif, struct nvkm_device *device,
 	nvkm_subdev_ctor(&nvkm_gsp, device, type, inst, &gsp->subdev);
 
 	fwif = nvkm_firmware_load(&gsp->subdev, fwif, "Gsp", gsp);
-	if (IS_ERR(fwif))
+	if (IS_ERR(fwif)) {
+		nvkm_error(&gsp->subdev, "Failed to load required firmware for device.");
 		return PTR_ERR(fwif);
+	}
 
 	gsp->func = fwif->func;
 

From 04864af849d9ae0dd020798f5b3632d9cf26fa03 Mon Sep 17 00:00:00 2001
From: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Date: Wed, 23 Jul 2025 13:05:07 +0300
Subject: [PATCH 0398/1455] drm/bridge: cdns-dsi: Fix the _atomic_check()

Use the "adjusted_mode" for the dsi configuration check, as that is the
more appropriate display_mode for validation, and later bridge enable.

Also, fix the mode_valid_check parameter from false to true, as the dsi
configuration check is taking place during the check-phase, and the
crtc_* mode values are not expected to be populated yet.

Fixes: a53d987756ea ("drm/bridge: cdns-dsi: Move DSI mode check to _atomic_check()")
Signed-off-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-1-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index a57ca8c3bdae..695b6246b280 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -997,10 +997,10 @@ static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
 	struct cdns_dsi_bridge_state *dsi_state = to_cdns_dsi_bridge_state(bridge_state);
-	const struct drm_display_mode *mode = &crtc_state->mode;
+	const struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
 	struct cdns_dsi_cfg *dsi_cfg = &dsi_state->dsi_cfg;
 
-	return cdns_dsi_check_conf(dsi, mode, dsi_cfg, false);
+	return cdns_dsi_check_conf(dsi, adjusted_mode, dsi_cfg, true);
 }
 
 static struct drm_bridge_state *

From f7e831db2d021a6e27ab6a89d4ca42613db2dc8b Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:08 +0300
Subject: [PATCH 0399/1455] drm/tidss: Fix missing includes and struct decls

Fix missing includes and struct declarations. Even if these don't cause
any compile issues at the moment, it's good to have them correct.

Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-2-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.h       | 3 +++
 drivers/gpu/drm/tidss/tidss_drv.h         | 2 ++
 drivers/gpu/drm/tidss/tidss_plane.h       | 2 ++
 drivers/gpu/drm/tidss/tidss_scale_coefs.h | 2 ++
 4 files changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.h b/drivers/gpu/drm/tidss/tidss_dispc.h
index b8614f62186c..60c1b400eb89 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.h
+++ b/drivers/gpu/drm/tidss/tidss_dispc.h
@@ -7,11 +7,14 @@
 #ifndef __TIDSS_DISPC_H__
 #define __TIDSS_DISPC_H__
 
+#include <drm/drm_color_mgmt.h>
+
 #include "tidss_drv.h"
 
 struct dispc_device;
 
 struct drm_crtc_state;
+struct drm_plane_state;
 
 enum tidss_gamma_type { TIDSS_GAMMA_8BIT, TIDSS_GAMMA_10BIT };
 
diff --git a/drivers/gpu/drm/tidss/tidss_drv.h b/drivers/gpu/drm/tidss/tidss_drv.h
index d14d5d28f0a3..84454a4855d1 100644
--- a/drivers/gpu/drm/tidss/tidss_drv.h
+++ b/drivers/gpu/drm/tidss/tidss_drv.h
@@ -9,6 +9,8 @@
 
 #include <linux/spinlock.h>
 
+#include <drm/drm_device.h>
+
 #define TIDSS_MAX_PORTS 4
 #define TIDSS_MAX_PLANES 4
 #define TIDSS_MAX_OLDI_TXES 2
diff --git a/drivers/gpu/drm/tidss/tidss_plane.h b/drivers/gpu/drm/tidss/tidss_plane.h
index aecaf2728406..92c560c3a621 100644
--- a/drivers/gpu/drm/tidss/tidss_plane.h
+++ b/drivers/gpu/drm/tidss/tidss_plane.h
@@ -7,6 +7,8 @@
 #ifndef __TIDSS_PLANE_H__
 #define __TIDSS_PLANE_H__
 
+#include <drm/drm_plane.h>
+
 #define to_tidss_plane(p) container_of((p), struct tidss_plane, plane)
 
 struct tidss_device;
diff --git a/drivers/gpu/drm/tidss/tidss_scale_coefs.h b/drivers/gpu/drm/tidss/tidss_scale_coefs.h
index 9c560d0fdac0..9824d02d9d1f 100644
--- a/drivers/gpu/drm/tidss/tidss_scale_coefs.h
+++ b/drivers/gpu/drm/tidss/tidss_scale_coefs.h
@@ -9,6 +9,8 @@
 
 #include <linux/types.h>
 
+struct device;
+
 struct tidss_scale_coefs {
 	s16 c2[16];
 	s16 c1[16];

From 478306edc23eec4f0ec24a46222485910c66212d Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:09 +0300
Subject: [PATCH 0400/1455] drm/tidss: Use the crtc_* timings when programming
 the HW

Use the crtc_* fields from drm_display_mode, instead of the "logical"
fields. This shouldn't change anything in practice, but afaiu the crtc_*
fields are the correct ones to use here.

Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-3-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_crtc.c  |  2 +-
 drivers/gpu/drm/tidss/tidss_dispc.c | 16 ++++++++--------
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_crtc.c b/drivers/gpu/drm/tidss/tidss_crtc.c
index a2f40a5c7703..17efd77ce7f2 100644
--- a/drivers/gpu/drm/tidss/tidss_crtc.c
+++ b/drivers/gpu/drm/tidss/tidss_crtc.c
@@ -225,7 +225,7 @@ static void tidss_crtc_atomic_enable(struct drm_crtc *crtc,
 	tidss_runtime_get(tidss);
 
 	r = dispc_vp_set_clk_rate(tidss->dispc, tcrtc->hw_videoport,
-				  mode->clock * 1000);
+				  mode->crtc_clock * 1000);
 	if (r != 0)
 		return;
 
diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index c0277fa36425..3f6cff2ab1b2 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -1215,13 +1215,13 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 
 	dispc_set_num_datalines(dispc, hw_videoport, fmt->data_width);
 
-	hfp = mode->hsync_start - mode->hdisplay;
-	hsw = mode->hsync_end - mode->hsync_start;
-	hbp = mode->htotal - mode->hsync_end;
+	hfp = mode->crtc_hsync_start - mode->crtc_hdisplay;
+	hsw = mode->crtc_hsync_end - mode->crtc_hsync_start;
+	hbp = mode->crtc_htotal - mode->crtc_hsync_end;
 
-	vfp = mode->vsync_start - mode->vdisplay;
-	vsw = mode->vsync_end - mode->vsync_start;
-	vbp = mode->vtotal - mode->vsync_end;
+	vfp = mode->crtc_vsync_start - mode->crtc_vdisplay;
+	vsw = mode->crtc_vsync_end - mode->crtc_vsync_start;
+	vbp = mode->crtc_vtotal - mode->crtc_vsync_end;
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_TIMING_H,
 		       FLD_VAL(hsw - 1, 7, 0) |
@@ -1263,8 +1263,8 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 		       FLD_VAL(ivs, 12, 12));
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_SIZE_SCREEN,
-		       FLD_VAL(mode->hdisplay - 1, 11, 0) |
-		       FLD_VAL(mode->vdisplay - 1, 27, 16));
+		       FLD_VAL(mode->crtc_hdisplay - 1, 11, 0) |
+		       FLD_VAL(mode->crtc_vdisplay - 1, 27, 16));
 
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1, 0, 0);
 }

From 661598d90ca430832e42d81d1a031dc255e85068 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:10 +0300
Subject: [PATCH 0401/1455] drm/bridge: cdns-dsi: Remove extra line at the end
 of the file

Remove extra line at the end of the file.

Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-4-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 695b6246b280..0c7ad05b6e53 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -1442,4 +1442,3 @@ MODULE_AUTHOR("Boris Brezillon <boris.brezillon@bootlin.com>");
 MODULE_DESCRIPTION("Cadence DSI driver");
 MODULE_LICENSE("GPL");
 MODULE_ALIAS("platform:cdns-dsi");
-

From 2325e509ae3c688ad6fb2b84ec4cef2861e89542 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:11 +0300
Subject: [PATCH 0402/1455] drm/bridge: cdns-dsi: Drop crtc_* code

With recent change the cdns_dsi_check_conf() is always called with
mode_valid_check = true. We can thus remove all the code related to the
"false" paths.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-5-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 .../gpu/drm/bridge/cadence/cdns-dsi-core.c    | 60 ++++++-------------
 1 file changed, 17 insertions(+), 43 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 0c7ad05b6e53..eae9469ef431 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -452,15 +452,6 @@ bridge_to_cdns_dsi_input(struct drm_bridge *bridge)
 	return container_of(bridge, struct cdns_dsi_input, bridge);
 }
 
-static unsigned int mode_to_dpi_hfp(const struct drm_display_mode *mode,
-				    bool mode_valid_check)
-{
-	if (mode_valid_check)
-		return mode->hsync_start - mode->hdisplay;
-
-	return mode->crtc_hsync_start - mode->crtc_hdisplay;
-}
-
 static unsigned int dpi_to_dsi_timing(unsigned int dpi_timing,
 				      unsigned int dpi_bpp,
 				      unsigned int dsi_pkt_overhead)
@@ -477,8 +468,7 @@ static unsigned int dpi_to_dsi_timing(unsigned int dpi_timing,
 
 static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 			     const struct drm_display_mode *mode,
-			     struct cdns_dsi_cfg *dsi_cfg,
-			     bool mode_valid_check)
+			     struct cdns_dsi_cfg *dsi_cfg)
 {
 	struct cdns_dsi_output *output = &dsi->output;
 	unsigned int tmp;
@@ -492,30 +482,20 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 
 	bpp = mipi_dsi_pixel_format_to_bpp(output->dev->format);
 
-	if (mode_valid_check)
-		tmp = mode->htotal -
-		      (sync_pulse ? mode->hsync_end : mode->hsync_start);
-	else
-		tmp = mode->crtc_htotal -
-		      (sync_pulse ?
-		       mode->crtc_hsync_end : mode->crtc_hsync_start);
+	tmp = mode->htotal -
+		(sync_pulse ? mode->hsync_end : mode->hsync_start);
 
 	dsi_cfg->hbp = dpi_to_dsi_timing(tmp, bpp, DSI_HBP_FRAME_OVERHEAD);
 
 	if (sync_pulse) {
-		if (mode_valid_check)
-			tmp = mode->hsync_end - mode->hsync_start;
-		else
-			tmp = mode->crtc_hsync_end - mode->crtc_hsync_start;
+		tmp = mode->hsync_end - mode->hsync_start;
 
 		dsi_cfg->hsa = dpi_to_dsi_timing(tmp, bpp,
 						 DSI_HSA_FRAME_OVERHEAD);
 	}
 
-	dsi_cfg->hact = dpi_to_dsi_timing(mode_valid_check ?
-					  mode->hdisplay : mode->crtc_hdisplay,
-					  bpp, 0);
-	dsi_cfg->hfp = dpi_to_dsi_timing(mode_to_dpi_hfp(mode, mode_valid_check),
+	dsi_cfg->hact = dpi_to_dsi_timing(mode->hdisplay, bpp, 0);
+	dsi_cfg->hfp = dpi_to_dsi_timing(mode->hsync_start - mode->hdisplay,
 					 bpp, DSI_HFP_FRAME_OVERHEAD);
 
 	return 0;
@@ -524,14 +504,12 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 static int cdns_dsi_adjust_phy_config(struct cdns_dsi *dsi,
 			      struct cdns_dsi_cfg *dsi_cfg,
 			      struct phy_configure_opts_mipi_dphy *phy_cfg,
-			      const struct drm_display_mode *mode,
-			      bool mode_valid_check)
+			      const struct drm_display_mode *mode)
 {
 	struct cdns_dsi_output *output = &dsi->output;
 	unsigned long long dlane_bps;
 	unsigned long adj_dsi_htotal;
 	unsigned long dsi_htotal;
-	unsigned long dpi_htotal;
 	unsigned long dpi_hz;
 	unsigned int dsi_hfp_ext;
 	unsigned int lanes = output->dev->lanes;
@@ -552,12 +530,11 @@ static int cdns_dsi_adjust_phy_config(struct cdns_dsi *dsi,
 	if (dsi_htotal % lanes)
 		adj_dsi_htotal += lanes - (dsi_htotal % lanes);
 
-	dpi_hz = (mode_valid_check ? mode->clock : mode->crtc_clock) * 1000;
+	dpi_hz = mode->clock * 1000;
 	dlane_bps = (unsigned long long)dpi_hz * adj_dsi_htotal;
 
 	/* data rate in bytes/sec is not an integer, refuse the mode. */
-	dpi_htotal = mode_valid_check ? mode->htotal : mode->crtc_htotal;
-	if (do_div(dlane_bps, lanes * dpi_htotal))
+	if (do_div(dlane_bps, lanes * mode->htotal))
 		return -EINVAL;
 
 	/* data rate was in bytes/sec, convert to bits/sec. */
@@ -572,27 +549,25 @@ static int cdns_dsi_adjust_phy_config(struct cdns_dsi *dsi,
 
 static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 			       const struct drm_display_mode *mode,
-			       struct cdns_dsi_cfg *dsi_cfg,
-			       bool mode_valid_check)
+			       struct cdns_dsi_cfg *dsi_cfg)
 {
 	struct cdns_dsi_output *output = &dsi->output;
 	struct phy_configure_opts_mipi_dphy *phy_cfg = &output->phy_opts.mipi_dphy;
 	unsigned long dsi_hss_hsa_hse_hbp;
 	unsigned int nlanes = output->dev->lanes;
-	int mode_clock = (mode_valid_check ? mode->clock : mode->crtc_clock);
 	int ret;
 
-	ret = cdns_dsi_mode2cfg(dsi, mode, dsi_cfg, mode_valid_check);
+	ret = cdns_dsi_mode2cfg(dsi, mode, dsi_cfg);
 	if (ret)
 		return ret;
 
-	ret = phy_mipi_dphy_get_default_config(mode_clock * 1000,
+	ret = phy_mipi_dphy_get_default_config(mode->clock * 1000,
 					       mipi_dsi_pixel_format_to_bpp(output->dev->format),
 					       nlanes, phy_cfg);
 	if (ret)
 		return ret;
 
-	ret = cdns_dsi_adjust_phy_config(dsi, dsi_cfg, phy_cfg, mode, mode_valid_check);
+	ret = cdns_dsi_adjust_phy_config(dsi, dsi_cfg, phy_cfg, mode);
 	if (ret)
 		return ret;
 
@@ -610,9 +585,8 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 	 * interface.
 	 */
 	if ((u64)phy_cfg->hs_clk_rate *
-	    mode_to_dpi_hfp(mode, mode_valid_check) * nlanes <
-	    (u64)dsi_hss_hsa_hse_hbp *
-	    (mode_valid_check ? mode->clock : mode->crtc_clock) * 1000)
+	    (mode->hsync_start - mode->hdisplay) * nlanes <
+	    (u64)dsi_hss_hsa_hse_hbp * mode->clock  * 1000)
 		return -EINVAL;
 
 	return 0;
@@ -663,7 +637,7 @@ cdns_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 	if ((mode->hdisplay * bpp) % 32)
 		return MODE_H_ILLEGAL;
 
-	ret = cdns_dsi_check_conf(dsi, mode, &dsi_cfg, true);
+	ret = cdns_dsi_check_conf(dsi, mode, &dsi_cfg);
 	if (ret)
 		return MODE_BAD;
 
@@ -1000,7 +974,7 @@ static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 	const struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
 	struct cdns_dsi_cfg *dsi_cfg = &dsi_state->dsi_cfg;
 
-	return cdns_dsi_check_conf(dsi, adjusted_mode, dsi_cfg, true);
+	return cdns_dsi_check_conf(dsi, adjusted_mode, dsi_cfg);
 }
 
 static struct drm_bridge_state *

From 644696a4c48b27d99ff68fa6fee4efcf3910eef5 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:12 +0300
Subject: [PATCH 0403/1455] drm/bridge: cdns-dsi: Remove broken fifo emptying
 check

The driver checks if "DPI(HFP) > DSI(HSS+HSA+HSE+HBP)", and rejects the
mode if not.

However, testing shows that this doesn't hold at all. I can set the hfp
to very small values, with no errors. The feedback from the HW team also
was that the check is not right, although it's not clear if there's a
way to validate the FIFO emptying.

The check rejects quite a lot of modes, apparently for no good reason,
so drop the check.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-6-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 15 ---------------
 1 file changed, 15 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index eae9469ef431..000c5a2367fe 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -553,7 +553,6 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 {
 	struct cdns_dsi_output *output = &dsi->output;
 	struct phy_configure_opts_mipi_dphy *phy_cfg = &output->phy_opts.mipi_dphy;
-	unsigned long dsi_hss_hsa_hse_hbp;
 	unsigned int nlanes = output->dev->lanes;
 	int ret;
 
@@ -575,20 +574,6 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 	if (ret)
 		return ret;
 
-	dsi_hss_hsa_hse_hbp = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
-	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
-		dsi_hss_hsa_hse_hbp += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
-
-	/*
-	 * Make sure DPI(HFP) > DSI(HSS+HSA+HSE+HBP) to guarantee that the FIFO
-	 * is empty before we start a receiving a new line on the DPI
-	 * interface.
-	 */
-	if ((u64)phy_cfg->hs_clk_rate *
-	    (mode->hsync_start - mode->hdisplay) * nlanes <
-	    (u64)dsi_hss_hsa_hse_hbp * mode->clock  * 1000)
-		return -EINVAL;
-
 	return 0;
 }
 

From bcd6808aff4029a7df690cf4c9911e5cc826098d Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:13 +0300
Subject: [PATCH 0404/1455] drm/bridge: cdns-dsi: Drop checks that shouldn't be
 in .mode_valid()

The docs say about mode_valid():

"it is not allowed to look at anything else but the passed-in mode, and
validate it against configuration-invariant hardware constraints"

We're doing a lot more than just looking at the mode. The main issue
here is that we're doing checks based on the pixel clock, before we know
what the pixel clock from the crtc actually is.

So, drop the cdns_dsi_check_conf() call from .mode_valid().

Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-7-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 000c5a2367fe..b2b6529b1c70 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -603,8 +603,7 @@ cdns_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
 	struct cdns_dsi_output *output = &dsi->output;
-	struct cdns_dsi_cfg dsi_cfg;
-	int bpp, ret;
+	int bpp;
 
 	/*
 	 * VFP_DSI should be less than VFP_DPI and VFP_DSI should be at
@@ -622,10 +621,6 @@ cdns_dsi_bridge_mode_valid(struct drm_bridge *bridge,
 	if ((mode->hdisplay * bpp) % 32)
 		return MODE_H_ILLEGAL;
 
-	ret = cdns_dsi_check_conf(dsi, mode, &dsi_cfg);
-	if (ret)
-		return MODE_BAD;
-
 	return MODE_OK;
 }
 

From 3d97dc02a0869d2c8cff8ebb480c44ceff188157 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:14 +0300
Subject: [PATCH 0405/1455] drm/bridge: cdns-dsi: Update htotal in
 cdns_dsi_mode2cfg()

cdns_dsi_mode2cfg() calculates the dsi timings, but for some reason
doesn't set the htotal based on those timings. It is set only later, in
cdns_dsi_adjust_phy_config().

As cdns_dsi_mode2cfg() is the logical place to calculate it, let's move
it there. Especially as the following patch will remove
cdns_dsi_adjust_phy_config().

Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-8-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index b2b6529b1c70..7103878df1e7 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -498,6 +498,13 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 	dsi_cfg->hfp = dpi_to_dsi_timing(mode->hsync_start - mode->hdisplay,
 					 bpp, DSI_HFP_FRAME_OVERHEAD);
 
+	dsi_cfg->htotal = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
+	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
+		dsi_cfg->htotal += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
+
+	dsi_cfg->htotal += dsi_cfg->hact;
+	dsi_cfg->htotal += dsi_cfg->hfp + DSI_HFP_FRAME_OVERHEAD;
+
 	return 0;
 }
 
@@ -514,12 +521,7 @@ static int cdns_dsi_adjust_phy_config(struct cdns_dsi *dsi,
 	unsigned int dsi_hfp_ext;
 	unsigned int lanes = output->dev->lanes;
 
-	dsi_htotal = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
-	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
-		dsi_htotal += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
-
-	dsi_htotal += dsi_cfg->hact;
-	dsi_htotal += dsi_cfg->hfp + DSI_HFP_FRAME_OVERHEAD;
+	dsi_htotal = dsi_cfg->htotal;
 
 	/*
 	 * Make sure DSI htotal is aligned on a lane boundary when calculating

From c92765d52da9d2d1a4f867433b0ee5faf82815e4 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:15 +0300
Subject: [PATCH 0406/1455] drm/bridge: cdns-dsi: Drop
 cdns_dsi_adjust_phy_config()

cdns_dsi_adjust_phy_config() is called from cdns_dsi_check_conf(), which
is called from .atomic_check(). It checks the DSI htotal and adjusts it
to align on the DSI lane boundary by changing hfp and then recalculating
htotal and HS clock rate.

This has a few problems.

First is the fact that the whole thing is not needed: we do not need to
align on the lane boundary. The whole frame is sent in HS mode, and it
is fine if the line's last byte clock tick fills, say, only 2 of the 4
lanes. The next line will just continue from there. Assuming the
DSI timing values have been calculated to match the incoming DPI stream,
and the HS clock is compatible with the DPI pixel clock, the "uneven"
DSI lines will even out when multiple lines are being sent.

But we could do the align, aligning is not a problem as such. However,
adding more bytes to the hfp, as the function currently does, makes the
DSI line time longer, so the function then adjusts the HS clock rate.
This is where things fail: we don't know what rates we can get from the
HS clock, and at least in TI K3 SoC case the rates are quite coarsely
grained. Thus small adjustment to hfp will lead to a big change in HS
clock rate, and things break down.

We could do a loop here, adjusting hfp, adjusting clock, checking clock
rate, adjusting hfp again, etc., but considering that the whole
adjustment shouldn't be needed at all, it's easier to just remove the
function.

Something like this function should be added back later, when adding
burst mode support, but that's a bigger change and I don't think this
function would help that work in any way.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-9-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 .../gpu/drm/bridge/cadence/cdns-dsi-core.c    | 45 -------------------
 1 file changed, 45 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 7103878df1e7..f7d7d277367e 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -508,47 +508,6 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 	return 0;
 }
 
-static int cdns_dsi_adjust_phy_config(struct cdns_dsi *dsi,
-			      struct cdns_dsi_cfg *dsi_cfg,
-			      struct phy_configure_opts_mipi_dphy *phy_cfg,
-			      const struct drm_display_mode *mode)
-{
-	struct cdns_dsi_output *output = &dsi->output;
-	unsigned long long dlane_bps;
-	unsigned long adj_dsi_htotal;
-	unsigned long dsi_htotal;
-	unsigned long dpi_hz;
-	unsigned int dsi_hfp_ext;
-	unsigned int lanes = output->dev->lanes;
-
-	dsi_htotal = dsi_cfg->htotal;
-
-	/*
-	 * Make sure DSI htotal is aligned on a lane boundary when calculating
-	 * the expected data rate. This is done by extending HFP in case of
-	 * misalignment.
-	 */
-	adj_dsi_htotal = dsi_htotal;
-	if (dsi_htotal % lanes)
-		adj_dsi_htotal += lanes - (dsi_htotal % lanes);
-
-	dpi_hz = mode->clock * 1000;
-	dlane_bps = (unsigned long long)dpi_hz * adj_dsi_htotal;
-
-	/* data rate in bytes/sec is not an integer, refuse the mode. */
-	if (do_div(dlane_bps, lanes * mode->htotal))
-		return -EINVAL;
-
-	/* data rate was in bytes/sec, convert to bits/sec. */
-	phy_cfg->hs_clk_rate = dlane_bps * 8;
-
-	dsi_hfp_ext = adj_dsi_htotal - dsi_htotal;
-	dsi_cfg->hfp += dsi_hfp_ext;
-	dsi_cfg->htotal = dsi_htotal + dsi_hfp_ext;
-
-	return 0;
-}
-
 static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 			       const struct drm_display_mode *mode,
 			       struct cdns_dsi_cfg *dsi_cfg)
@@ -568,10 +527,6 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 	if (ret)
 		return ret;
 
-	ret = cdns_dsi_adjust_phy_config(dsi, dsi_cfg, phy_cfg, mode);
-	if (ret)
-		return ret;
-
 	ret = phy_validate(dsi->dphy, PHY_MODE_MIPI_DPHY, 0, &output->phy_opts);
 	if (ret)
 		return ret;

From e948bcaf7dd5719d17a0d3843250ed00a6b27ba5 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:16 +0300
Subject: [PATCH 0407/1455] drm/bridge: cdns-dsi: Adjust mode to negative syncs

The Cadence DSI requires negative syncs from the incoming video signal,
but at the moment that requirement is not expressed in any way. If the
crtc decides to use positive syncs, things break down.

Use the adjusted_mode in atomic_check to set the sync flags to negative
ones.

Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-10-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index f7d7d277367e..d49b4789a074 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -908,9 +908,13 @@ static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 	struct cdns_dsi_input *input = bridge_to_cdns_dsi_input(bridge);
 	struct cdns_dsi *dsi = input_to_dsi(input);
 	struct cdns_dsi_bridge_state *dsi_state = to_cdns_dsi_bridge_state(bridge_state);
-	const struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
+	struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
 	struct cdns_dsi_cfg *dsi_cfg = &dsi_state->dsi_cfg;
 
+	/* cdns-dsi requires negative syncs */
+	adjusted_mode->flags &= ~(DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
+	adjusted_mode->flags |= DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC;
+
 	return cdns_dsi_check_conf(dsi, adjusted_mode, dsi_cfg);
 }
 

From eea4f89b6461294ed6bea1d3285bb3f79c09a041 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:17 +0300
Subject: [PATCH 0408/1455] drm/bridge: cdns-dsi: Fix REG_WAKEUP_TIME value

The driver tries to calculate the value for REG_WAKEUP_TIME. However,
the calculation itself is not correct, and to add on it, the resulting
value is almost always larger than the field's size, so the actual
result is more or less random.

According to the docs, figuring out the value for REG_WAKEUP_TIME
requires HW characterization and there's no way to have a generic
algorithm to come up with the value. That doesn't help at all...

However, we know that the value must be smaller than the line time, and,
at least in my understanding, the proper value for it is quite small.
Testing shows that setting it to 1/10 of the line time seems to work
well. All video modes from my HDMI monitor work with this algorithm.

Hopefully we'll get more information on how to calculate the value, and
we can then update this.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-11-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index d49b4789a074..6bc0a0d00d69 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -793,7 +793,13 @@ static void cdns_dsi_bridge_atomic_pre_enable(struct drm_bridge *bridge,
 
 	tx_byte_period = DIV_ROUND_DOWN_ULL((u64)NSEC_PER_SEC * 8,
 					    phy_cfg->hs_clk_rate);
-	reg_wakeup = (phy_cfg->hs_prepare + phy_cfg->hs_zero) / tx_byte_period;
+
+	/*
+	 * Estimated time [in clock cycles] to perform LP->HS on D-PHY.
+	 * It is not clear how to calculate this, so for now,
+	 * set it to 1/10 of the total number of clocks in a line.
+	 */
+	reg_wakeup = dsi_cfg.htotal / nlanes / 10;
 	writel(REG_WAKEUP_TIME(reg_wakeup) | REG_LINE_DURATION(tmp),
 	       dsi->regs + VID_DPHY_TIME);
 

From ce4bc5ca7c1de554c3d8c95aa2dddce6cbcef14e Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:18 +0300
Subject: [PATCH 0409/1455] drm/bridge: cdns-dsi: Use video mode and clean up
 cdns_dsi_mode2cfg()

The driver does all the calculations and programming with video timings
(hftp, hbp, etc.) instead of the modeline values (hsync_start, ...).
Thus it makes sense to use struct videomode instead of struct
drm_display_mode internally.

Switch to videomode and do some cleanups in cdns_dsi_mode2cfg() along
the way.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Reviewed-by: Aradhya Bhatia <aradhya.bhatia@linux.dev>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-12-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 .../gpu/drm/bridge/cadence/cdns-dsi-core.c    | 45 ++++++++++---------
 1 file changed, 24 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 6bc0a0d00d69..07f8d5f5c2aa 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -9,6 +9,7 @@
 #include <drm/drm_drv.h>
 #include <drm/drm_probe_helper.h>
 #include <video/mipi_display.h>
+#include <video/videomode.h>
 
 #include <linux/clk.h>
 #include <linux/interrupt.h>
@@ -467,36 +468,35 @@ static unsigned int dpi_to_dsi_timing(unsigned int dpi_timing,
 }
 
 static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
-			     const struct drm_display_mode *mode,
+			     const struct videomode *vm,
 			     struct cdns_dsi_cfg *dsi_cfg)
 {
 	struct cdns_dsi_output *output = &dsi->output;
-	unsigned int tmp;
-	bool sync_pulse = false;
+	u32 dpi_hsa, dpi_hbp, dpi_hfp, dpi_hact;
+	bool sync_pulse;
 	int bpp;
 
+	dpi_hsa = vm->hsync_len;
+	dpi_hbp = vm->hback_porch;
+	dpi_hfp = vm->hfront_porch;
+	dpi_hact = vm->hactive;
+
 	memset(dsi_cfg, 0, sizeof(*dsi_cfg));
 
-	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
-		sync_pulse = true;
+	sync_pulse = output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE;
 
 	bpp = mipi_dsi_pixel_format_to_bpp(output->dev->format);
 
-	tmp = mode->htotal -
-		(sync_pulse ? mode->hsync_end : mode->hsync_start);
+	dsi_cfg->hbp = dpi_to_dsi_timing(dpi_hbp + (sync_pulse ? 0 : dpi_hsa),
+					 bpp, DSI_HBP_FRAME_OVERHEAD);
 
-	dsi_cfg->hbp = dpi_to_dsi_timing(tmp, bpp, DSI_HBP_FRAME_OVERHEAD);
+	if (sync_pulse)
+		dsi_cfg->hsa =
+			dpi_to_dsi_timing(dpi_hsa, bpp, DSI_HSA_FRAME_OVERHEAD);
 
-	if (sync_pulse) {
-		tmp = mode->hsync_end - mode->hsync_start;
+	dsi_cfg->hact = dpi_to_dsi_timing(dpi_hact, bpp, 0);
 
-		dsi_cfg->hsa = dpi_to_dsi_timing(tmp, bpp,
-						 DSI_HSA_FRAME_OVERHEAD);
-	}
-
-	dsi_cfg->hact = dpi_to_dsi_timing(mode->hdisplay, bpp, 0);
-	dsi_cfg->hfp = dpi_to_dsi_timing(mode->hsync_start - mode->hdisplay,
-					 bpp, DSI_HFP_FRAME_OVERHEAD);
+	dsi_cfg->hfp = dpi_to_dsi_timing(dpi_hfp, bpp, DSI_HFP_FRAME_OVERHEAD);
 
 	dsi_cfg->htotal = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
 	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
@@ -509,7 +509,7 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 }
 
 static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
-			       const struct drm_display_mode *mode,
+			       const struct videomode *vm,
 			       struct cdns_dsi_cfg *dsi_cfg)
 {
 	struct cdns_dsi_output *output = &dsi->output;
@@ -517,11 +517,11 @@ static int cdns_dsi_check_conf(struct cdns_dsi *dsi,
 	unsigned int nlanes = output->dev->lanes;
 	int ret;
 
-	ret = cdns_dsi_mode2cfg(dsi, mode, dsi_cfg);
+	ret = cdns_dsi_mode2cfg(dsi, vm, dsi_cfg);
 	if (ret)
 		return ret;
 
-	ret = phy_mipi_dphy_get_default_config(mode->clock * 1000,
+	ret = phy_mipi_dphy_get_default_config(vm->pixelclock,
 					       mipi_dsi_pixel_format_to_bpp(output->dev->format),
 					       nlanes, phy_cfg);
 	if (ret)
@@ -916,12 +916,15 @@ static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 	struct cdns_dsi_bridge_state *dsi_state = to_cdns_dsi_bridge_state(bridge_state);
 	struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
 	struct cdns_dsi_cfg *dsi_cfg = &dsi_state->dsi_cfg;
+	struct videomode vm;
 
 	/* cdns-dsi requires negative syncs */
 	adjusted_mode->flags &= ~(DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
 	adjusted_mode->flags |= DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC;
 
-	return cdns_dsi_check_conf(dsi, adjusted_mode, dsi_cfg);
+	drm_display_mode_to_videomode(adjusted_mode, &vm);
+
+	return cdns_dsi_check_conf(dsi, &vm, dsi_cfg);
 }
 
 static struct drm_bridge_state *

From ca3e6fc131836c4956de454996df52feb6c19af8 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:19 +0300
Subject: [PATCH 0410/1455] drm/bridge: cdns-dsi: Fix event mode

The timings calculation gets it wrong for DSI event mode, resulting in
too large hbp value. Fix the issue by taking into account the
pulse/event mode difference.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Reviewed-by: Jayesh Choudhary <j-choudhary@ti.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-13-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 .../gpu/drm/bridge/cadence/cdns-dsi-core.c    | 31 ++++++++++++-------
 1 file changed, 20 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 07f8d5f5c2aa..3bc4d011b4c6 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -418,7 +418,8 @@
 #define DSI_OUTPUT_PORT			0
 #define DSI_INPUT_PORT(inputid)		(1 + (inputid))
 
-#define DSI_HBP_FRAME_OVERHEAD		12
+#define DSI_HBP_FRAME_PULSE_OVERHEAD	12
+#define DSI_HBP_FRAME_EVENT_OVERHEAD	16
 #define DSI_HSA_FRAME_OVERHEAD		14
 #define DSI_HFP_FRAME_OVERHEAD		6
 #define DSI_HSS_VSS_VSE_FRAME_OVERHEAD	4
@@ -487,23 +488,31 @@ static int cdns_dsi_mode2cfg(struct cdns_dsi *dsi,
 
 	bpp = mipi_dsi_pixel_format_to_bpp(output->dev->format);
 
-	dsi_cfg->hbp = dpi_to_dsi_timing(dpi_hbp + (sync_pulse ? 0 : dpi_hsa),
-					 bpp, DSI_HBP_FRAME_OVERHEAD);
+	if (sync_pulse) {
+		dsi_cfg->hbp = dpi_to_dsi_timing(dpi_hbp, bpp,
+						 DSI_HBP_FRAME_PULSE_OVERHEAD);
 
-	if (sync_pulse)
-		dsi_cfg->hsa =
-			dpi_to_dsi_timing(dpi_hsa, bpp, DSI_HSA_FRAME_OVERHEAD);
+		dsi_cfg->hsa = dpi_to_dsi_timing(dpi_hsa, bpp,
+						 DSI_HSA_FRAME_OVERHEAD);
+	} else {
+		dsi_cfg->hbp = dpi_to_dsi_timing(dpi_hbp + dpi_hsa, bpp,
+						 DSI_HBP_FRAME_EVENT_OVERHEAD);
+
+		dsi_cfg->hsa = 0;
+	}
 
 	dsi_cfg->hact = dpi_to_dsi_timing(dpi_hact, bpp, 0);
 
 	dsi_cfg->hfp = dpi_to_dsi_timing(dpi_hfp, bpp, DSI_HFP_FRAME_OVERHEAD);
 
-	dsi_cfg->htotal = dsi_cfg->hbp + DSI_HBP_FRAME_OVERHEAD;
-	if (output->dev->mode_flags & MIPI_DSI_MODE_VIDEO_SYNC_PULSE)
-		dsi_cfg->htotal += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
+	dsi_cfg->htotal = dsi_cfg->hact + dsi_cfg->hfp + DSI_HFP_FRAME_OVERHEAD;
 
-	dsi_cfg->htotal += dsi_cfg->hact;
-	dsi_cfg->htotal += dsi_cfg->hfp + DSI_HFP_FRAME_OVERHEAD;
+	if (sync_pulse) {
+		dsi_cfg->htotal += dsi_cfg->hbp + DSI_HBP_FRAME_PULSE_OVERHEAD;
+		dsi_cfg->htotal += dsi_cfg->hsa + DSI_HSA_FRAME_OVERHEAD;
+	} else {
+		dsi_cfg->htotal += dsi_cfg->hbp + DSI_HBP_FRAME_EVENT_OVERHEAD;
+	}
 
 	return 0;
 }

From 2f73f8983280782077dfa9247e978003efeb5af5 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:20 +0300
Subject: [PATCH 0411/1455] drm/bridge: cdns-dsi: Tune adjusted_mode->clock
 according to dsi needs

The driver currently expects the pixel clock and the HS clock to be
compatible, but the DPHY PLL doesn't give very finely grained rates.
This often leads to the situation where the pipeline just fails, as the
resulting HS clock is just too off.

We could change the driver to do a better job on adjusting the DSI
blanking values, hopefully getting a working pipeline even if the pclk
and HS clocks are not exactly compatible. But that is a bigger work.

What we can do easily is to see in .atomic_check() what HS clock rate we
can get, based on the pixel clock rate, and then convert the HS clock
rate back to pixel clock rate and ask that rate from the crtc. If the
crtc has a good PLL (which is the case for TI K3 SoCs), this will fix
any issues wrt. the clock rates.

If the crtc cannot provide the requested clock, well, we're no worse off
with this patch than what we have at the moment.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-14-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 .../gpu/drm/bridge/cadence/cdns-dsi-core.c    | 37 +++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 3bc4d011b4c6..114d883c65dc 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -915,6 +915,28 @@ static u32 *cdns_dsi_bridge_get_input_bus_fmts(struct drm_bridge *bridge,
 	return input_fmts;
 }
 
+static long cdns_dsi_round_pclk(struct cdns_dsi *dsi, unsigned long pclk)
+{
+	struct cdns_dsi_output *output = &dsi->output;
+	unsigned int nlanes = output->dev->lanes;
+	union phy_configure_opts phy_opts = { 0 };
+	u32 bitspp;
+	int ret;
+
+	bitspp = mipi_dsi_pixel_format_to_bpp(output->dev->format);
+
+	ret = phy_mipi_dphy_get_default_config(pclk, bitspp, nlanes,
+					       &phy_opts.mipi_dphy);
+	if (ret)
+		return ret;
+
+	ret = phy_validate(dsi->dphy, PHY_MODE_MIPI_DPHY, 0, &phy_opts);
+	if (ret)
+		return ret;
+
+	return div_u64((u64)phy_opts.mipi_dphy.hs_clk_rate * nlanes, bitspp);
+}
+
 static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 					struct drm_bridge_state *bridge_state,
 					struct drm_crtc_state *crtc_state,
@@ -926,11 +948,26 @@ static int cdns_dsi_bridge_atomic_check(struct drm_bridge *bridge,
 	struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
 	struct cdns_dsi_cfg *dsi_cfg = &dsi_state->dsi_cfg;
 	struct videomode vm;
+	long pclk;
 
 	/* cdns-dsi requires negative syncs */
 	adjusted_mode->flags &= ~(DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC);
 	adjusted_mode->flags |= DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC;
 
+	/*
+	 * The DPHY PLL has quite a coarsely grained clock rate options. See
+	 * what hsclk rate we can achieve based on the pixel clock, convert it
+	 * back to pixel clock, set that to the adjusted_mode->clock. This is
+	 * all in hopes that the CRTC will be able to provide us the requested
+	 * clock, as otherwise the DPI and DSI clocks will be out of sync.
+	 */
+
+	pclk = cdns_dsi_round_pclk(dsi, adjusted_mode->clock * 1000);
+	if (pclk < 0)
+		return (int)pclk;
+
+	adjusted_mode->clock = pclk / 1000;
+
 	drm_display_mode_to_videomode(adjusted_mode, &vm);
 
 	return cdns_dsi_check_conf(dsi, &vm, dsi_cfg);

From 7070f55f294745c5a3c033623b76309f3512be67 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 23 Jul 2025 13:05:21 +0300
Subject: [PATCH 0412/1455] drm/bridge: cdns-dsi: Don't fail on
 MIPI_DSI_MODE_VIDEO_BURST

While the cdns-dsi does not support DSI burst mode, the burst mode is
essentially DSI event mode with more versatile clocking and timings.
Thus cdns-dsi doesn't need to fail if the DSI peripheral driver requests
MIPI_DSI_MODE_VIDEO_BURST.

In my particular use case, this allows the use of ti-sn65dsi83 driver.

Tested-by: Parth Pancholi <parth.pancholi@toradex.com>
Tested-by: Jayesh Choudhary <j-choudhary@ti.com>
Reviewed-by: Devarsh Thakkar <devarsht@ti.com>
Link: https://lore.kernel.org/r/20250723-cdns-dsi-impro-v5-15-e61cc06074c2@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
index 114d883c65dc..09b289f0fcbf 100644
--- a/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
+++ b/drivers/gpu/drm/bridge/cadence/cdns-dsi-core.c
@@ -1052,10 +1052,6 @@ static int cdns_dsi_attach(struct mipi_dsi_host *host,
 	if (output->dev)
 		return -EBUSY;
 
-	/* We do not support burst mode yet. */
-	if (dev->mode_flags & MIPI_DSI_MODE_VIDEO_BURST)
-		return -ENOTSUPP;
-
 	/*
 	 * The host <-> device link might be described using an OF-graph
 	 * representation, in this case we extract the device of_node from

From cfb29225db20c56432a8525366321c0c09edfb2e Mon Sep 17 00:00:00 2001
From: Jayesh Choudhary <j-choudhary@ti.com>
Date: Tue, 24 Jun 2025 13:34:02 +0530
Subject: [PATCH 0413/1455] drm/tidss: Set crtc modesetting parameters with
 adjusted mode

TIDSS uses crtc_* fields to propagate its registers and set the
clock rates. So set the CRTC modesetting timing parameters with
the adjusted mode when needed, to set correct values.

Cc: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Signed-off-by: Jayesh Choudhary <j-choudhary@ti.com>
Link: https://lore.kernel.org/r/20250624080402.302526-1-j-choudhary@ti.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_crtc.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/tidss/tidss_crtc.c b/drivers/gpu/drm/tidss/tidss_crtc.c
index 17efd77ce7f2..da89fd01c337 100644
--- a/drivers/gpu/drm/tidss/tidss_crtc.c
+++ b/drivers/gpu/drm/tidss/tidss_crtc.c
@@ -91,7 +91,7 @@ static int tidss_crtc_atomic_check(struct drm_crtc *crtc,
 	struct dispc_device *dispc = tidss->dispc;
 	struct tidss_crtc *tcrtc = to_tidss_crtc(crtc);
 	u32 hw_videoport = tcrtc->hw_videoport;
-	const struct drm_display_mode *mode;
+	struct drm_display_mode *mode;
 	enum drm_mode_status ok;
 
 	dev_dbg(ddev->dev, "%s\n", __func__);
@@ -108,6 +108,9 @@ static int tidss_crtc_atomic_check(struct drm_crtc *crtc,
 		return -EINVAL;
 	}
 
+	if (drm_atomic_crtc_needs_modeset(crtc_state))
+		drm_mode_set_crtcinfo(mode, 0);
+
 	return dispc_vp_bus_check(dispc, hw_videoport, crtc_state);
 }
 

From 852bb7c08c134162b7063d1fdf9b1cc1ba839d0c Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Wed, 2 Jul 2025 09:48:44 +0100
Subject: [PATCH 0414/1455] drm/tidss: remove redundant assignment to variable
 ret

The assignment of zero to variable is redundant as the following
continue statement loops back to the start of the loop where
ret is assigned a new value from the return to the call to
get_parent_dss_vp. Remove assignment.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20250702084844.966199-1-colin.i.king@gmail.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_oldi.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/tidss/tidss_oldi.c b/drivers/gpu/drm/tidss/tidss_oldi.c
index 8f25159d0666..7688251beba2 100644
--- a/drivers/gpu/drm/tidss/tidss_oldi.c
+++ b/drivers/gpu/drm/tidss/tidss_oldi.c
@@ -464,7 +464,6 @@ int tidss_oldi_init(struct tidss_device *tidss)
 				 * which may still be connected.
 				 * Continue to search for that.
 				 */
-				ret = 0;
 				continue;
 			}
 			goto err_put_node;

From 942e54a372b44da3ffb0191b4d289d476256c861 Mon Sep 17 00:00:00 2001
From: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Date: Wed, 16 Apr 2025 14:06:30 +0300
Subject: [PATCH 0415/1455] drm/tidss: Remove early fb

Add a call to drm_aperture_remove_framebuffers() to drop the possible
early fb (simplefb).

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250416-tidss-splash-v1-2-4ff396eb5008@ideasonboard.com
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_drv.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/tidss/tidss_drv.c b/drivers/gpu/drm/tidss/tidss_drv.c
index a1b12e52aca4..27d9a8fd541f 100644
--- a/drivers/gpu/drm/tidss/tidss_drv.c
+++ b/drivers/gpu/drm/tidss/tidss_drv.c
@@ -8,6 +8,7 @@
 #include <linux/of.h>
 #include <linux/module.h>
 #include <linux/pm_runtime.h>
+#include <linux/aperture.h>
 
 #include <drm/clients/drm_client_setup.h>
 #include <drm/drm_atomic.h>
@@ -192,12 +193,20 @@ static int tidss_probe(struct platform_device *pdev)
 		goto err_irq_uninstall;
 	}
 
+	/* Remove possible early fb before setting up the fbdev */
+	ret = aperture_remove_all_conflicting_devices(tidss_driver.name);
+	if (ret)
+		goto err_drm_dev_unreg;
+
 	drm_client_setup(ddev, NULL);
 
 	dev_dbg(dev, "%s done\n", __func__);
 
 	return 0;
 
+err_drm_dev_unreg:
+	drm_dev_unregister(ddev);
+
 err_irq_uninstall:
 	tidss_irq_uninstall(ddev);
 

From 1ebc27248ea0b81f0023ca28894ac40183b86b7c Mon Sep 17 00:00:00 2001
From: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Date: Thu, 7 Aug 2025 13:54:02 +0530
Subject: [PATCH 0416/1455] drm/i915/display: Optimize panel power-on wait time

The current wait_panel_status() uses intel_de_wait(),
which internally on Xe platforms calls  xe_mmio_wait32().
xe_mmio_wait32() increases poll interval exponentially.

This exponential poll interval increase causes unnessory delays
during resume or power-on when the panel becomes ready earlier,
but polling is delayed due to backoff.

Replace intel_de_wait() with read_poll_timeout() +
intel_de_read() to actively poll the register at a fixed 10ms interval
up to a 5 second timeout. This allows poll to exit
early  when panel is ready.

Changes in v2:
Replaced  two-phase intel_de_wait() with  read_poll_timeout()
 + intel_de_read()
Changes in v3:
 - Add poll_interval_ms argument  'wait_panel_status' function.
 - Modify 'wait_panel_status' callers with proper poll interval
Changes in v4:
 - Change 'wait_panel_off' poll interval to 10ms
Changes in v5:
 - Dropped  poll_interval_ms parameter,use fixed polling
   interval of 10ms (Jani Nikula)
Changes in v6:
 - Removed goto in error path

Signed-off-by: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Link: https://lore.kernel.org/r/20250807082402.79018-1-dibin.moolakadan.subrahmanian@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_pps.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
index b64d0b30f5b1..b84eb43bd2d0 100644
--- a/drivers/gpu/drm/i915/display/intel_pps.c
+++ b/drivers/gpu/drm/i915/display/intel_pps.c
@@ -4,6 +4,7 @@
  */
 
 #include <linux/debugfs.h>
+#include <linux/iopoll.h>
 
 #include <drm/drm_print.h>
 
@@ -608,6 +609,8 @@ static void wait_panel_status(struct intel_dp *intel_dp,
 	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
 	i915_reg_t pp_stat_reg, pp_ctrl_reg;
+	int ret;
+	u32 val;
 
 	lockdep_assert_held(&display->pps.mutex);
 
@@ -624,13 +627,20 @@ static void wait_panel_status(struct intel_dp *intel_dp,
 		    intel_de_read(display, pp_stat_reg),
 		    intel_de_read(display, pp_ctrl_reg));
 
-	if (intel_de_wait(display, pp_stat_reg, mask, value, 5000))
+	ret = read_poll_timeout(intel_de_read, val,
+				(val & mask) == value,
+				10 * 1000, 5000 * 1000, true,
+				display, pp_stat_reg);
+
+	if (ret) {
 		drm_err(display->drm,
 			"[ENCODER:%d:%s] %s panel status timeout: PP_STATUS: 0x%08x PP_CONTROL: 0x%08x\n",
 			dig_port->base.base.base.id, dig_port->base.base.name,
 			pps_name(intel_dp),
 			intel_de_read(display, pp_stat_reg),
 			intel_de_read(display, pp_ctrl_reg));
+		return;
+	}
 
 	drm_dbg_kms(display->drm, "Wait complete\n");
 }

From 6b12afac75fc3dbe735c05716cf99a965927a8fa Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 12 Aug 2025 10:10:32 +0200
Subject: [PATCH 0417/1455] drm/panel: panel-summit: Include <linux/property.h>
 and <linux/mod_devicetable.h>

Include <linux/property.h> to declare device_property_read_u32() and
<linux/mod_devicetable.h> to declare struct of_device_id. Avoids the
dependency on the backlight header to include it.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250812081118.221103-1-tzimmermann@suse.de
---
 drivers/gpu/drm/panel/panel-summit.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-summit.c b/drivers/gpu/drm/panel/panel-summit.c
index 4854437e2899..6d40b9ddfe02 100644
--- a/drivers/gpu/drm/panel/panel-summit.c
+++ b/drivers/gpu/drm/panel/panel-summit.c
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0-only
 
 #include <linux/backlight.h>
+#include <linux/mod_devicetable.h>
+#include <linux/property.h>
 #include <drm/drm_device.h>
 #include <drm/drm_mipi_dsi.h>
 #include <drm/drm_mode.h>

From aaf01f66e0ee688f0df7eb941914c78fdecf1edd Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:47 +0300
Subject: [PATCH 0418/1455] drm/i915/tc: Use the cached max lane count value

Use the PHY's cached max lane count value on all platforms similarly to
LNL+. On LNL+ using the cached value is mandatory - since the
corresponding HW register field can get cleared by the time the value is
queried - on earlier platforms there isn't a problem with using the HW
register instead. Having a uniform way to query the value still makes
sense and it's also a bit more efficient, so do that.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-7-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 668ef139391b..f00fb6fc94d8 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -395,15 +395,11 @@ static void read_pin_configuration(struct intel_tc_port *tc)
 
 int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 {
-	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 
 	if (!intel_encoder_is_tc(&dig_port->base))
 		return 4;
 
-	if (DISPLAY_VER(display) < 20)
-		return get_max_lane_count(tc);
-
 	return tc->max_lane_count;
 }
 

From 45d424856a074ea58c5a853e11bd0388a56a6951 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:48 +0300
Subject: [PATCH 0419/1455] drm/i915/tc: Move getting the power domain before
 reading DFLEX registers

Move getting the required display power domain right before reading the
PORT_TX_DFLEXDPSP and PORT_TX_DFLEXPA1 registers, similarly to how this
is done while reading the other TCSS_DDI_STATUS PHY register.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-8-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index f00fb6fc94d8..f311d403eef2 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -269,9 +269,11 @@ static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
+	intel_wakeref_t wakeref;
 	u32 lane_mask;
 
-	lane_mask = intel_de_read(display, PORT_TX_DFLEXDPSP(tc->phy_fia));
+	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
+		lane_mask = intel_de_read(display, PORT_TX_DFLEXDPSP(tc->phy_fia));
 
 	drm_WARN_ON(display->drm, lane_mask == 0xffffffff);
 	assert_tc_cold_blocked(tc);
@@ -284,9 +286,11 @@ u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
+	intel_wakeref_t wakeref;
 	u32 pin_mask;
 
-	pin_mask = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia));
+	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
+		pin_mask = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia));
 
 	drm_WARN_ON(display->drm, pin_mask == 0xffffffff);
 	assert_tc_cold_blocked(tc);
@@ -324,12 +328,9 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 
 static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
-	struct intel_display *display = to_intel_display(dig_port);
-	intel_wakeref_t wakeref;
 	u32 pin_mask;
 
-	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
-		pin_mask = intel_tc_port_get_pin_assignment_mask(dig_port);
+	pin_mask = intel_tc_port_get_pin_assignment_mask(dig_port);
 
 	switch (pin_mask) {
 	default:
@@ -345,12 +346,9 @@ static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 
 static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
-	struct intel_display *display = to_intel_display(dig_port);
-	intel_wakeref_t wakeref;
 	u32 lane_mask = 0;
 
-	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
-		lane_mask = intel_tc_port_get_lane_mask(dig_port);
+	lane_mask = intel_tc_port_get_lane_mask(dig_port);
 
 	switch (lane_mask) {
 	default:

From f1ecaa7a6ef06e2843f7c6638db5dbd12a405866 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:49 +0300
Subject: [PATCH 0420/1455] drm/i915/tc: Move asserting the power state after
 reading TCSS_DDI_STATUS

Move asserting the expected TC cold power state and the read out
register value right after reading the TCSS_DDI_STATUS register,
similarly to how this is done with the other PORT_TX_DFLEXDPSP and
PORT_TX_DFLEXPA1 PHY registers.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-9-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index f311d403eef2..41bd82089180 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -303,12 +303,16 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
 	enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base);
+	struct intel_tc_port *tc = to_tc_port(dig_port);
 	intel_wakeref_t wakeref;
 	u32 val, pin_assignment;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
 		val = intel_de_read(display, TCSS_DDI_STATUS(tc_port));
 
+	drm_WARN_ON(display->drm, val == 0xffffffff);
+	assert_tc_cold_blocked(tc);
+
 	pin_assignment =
 		REG_FIELD_GET(TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK, val);
 
@@ -375,8 +379,6 @@ static int get_max_lane_count(struct intel_tc_port *tc)
 	if (tc->mode != TC_PORT_DP_ALT)
 		return 4;
 
-	assert_tc_cold_blocked(tc);
-
 	if (DISPLAY_VER(display) >= 20)
 		return lnl_tc_port_get_max_lane_count(dig_port);
 

From 6eaec941df17d3690b0be455f69f2cf64dea70a4 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:50 +0300
Subject: [PATCH 0421/1455] drm/i915/tc: Add an enum for the TypeC pin
 assignment

Add an enum for the TypeC pin assignment, which is a better way to pass
its value around than a plain integer. While at it add a description for
each pin assignment, based on the DP and DP Alt mode Standards, opting
for more details to ease any future debugging related to a given pin
assignment and the cables / sink types used.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
[Imre: s/deined/defined in pin assignment enum documentation.]
Link: https://lore.kernel.org/r/20250805073700.642107-10-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 .../gpu/drm/i915/display/intel_display_regs.h |  2 +
 drivers/gpu/drm/i915/display/intel_tc.c       | 19 ++---
 drivers/gpu/drm/i915/display/intel_tc.h       | 69 +++++++++++++++++++
 3 files changed, 78 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_regs.h b/drivers/gpu/drm/i915/display/intel_display_regs.h
index 7bd09d981cd2..9d71e26a4fa2 100644
--- a/drivers/gpu/drm/i915/display/intel_display_regs.h
+++ b/drivers/gpu/drm/i915/display/intel_display_regs.h
@@ -2890,6 +2890,7 @@ enum skl_power_gate {
 #define   DP_PIN_ASSIGNMENT_SHIFT(idx)		((idx) * 4)
 #define   DP_PIN_ASSIGNMENT_MASK(idx)		(0xf << ((idx) * 4))
 #define   DP_PIN_ASSIGNMENT(idx, x)		((x) << ((idx) * 4))
+/* See enum intel_tc_pin_assignment for the pin assignment field values. */
 
 #define _TCSS_DDI_STATUS_1			0x161500
 #define _TCSS_DDI_STATUS_2			0x161504
@@ -2897,6 +2898,7 @@ enum skl_power_gate {
 								 _TCSS_DDI_STATUS_1, \
 								 _TCSS_DDI_STATUS_2))
 #define  TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK	REG_GENMASK(28, 25)
+/* See enum intel_tc_pin_assignment for the pin assignment field values. */
 #define  TCSS_DDI_STATUS_READY			REG_BIT(2)
 #define  TCSS_DDI_STATUS_HPD_LIVE_STATUS_TBT	REG_BIT(1)
 #define  TCSS_DDI_STATUS_HPD_LIVE_STATUS_ALT	REG_BIT(0)
diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 41bd82089180..ffead1cbf10f 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -23,11 +23,6 @@
 #include "intel_modeset_lock.h"
 #include "intel_tc.h"
 
-#define DP_PIN_ASSIGNMENT_NONE	0x0
-#define DP_PIN_ASSIGNMENT_C	0x3
-#define DP_PIN_ASSIGNMENT_D	0x4
-#define DP_PIN_ASSIGNMENT_E	0x5
-
 enum tc_port_mode {
 	TC_PORT_DISCONNECTED,
 	TC_PORT_TBT_ALT,
@@ -317,15 +312,15 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 		REG_FIELD_GET(TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK, val);
 
 	switch (pin_assignment) {
-	case DP_PIN_ASSIGNMENT_NONE:
+	case INTEL_TC_PIN_ASSIGNMENT_NONE:
 		return 0;
 	default:
 		MISSING_CASE(pin_assignment);
 		fallthrough;
-	case DP_PIN_ASSIGNMENT_D:
+	case INTEL_TC_PIN_ASSIGNMENT_D:
 		return 2;
-	case DP_PIN_ASSIGNMENT_C:
-	case DP_PIN_ASSIGNMENT_E:
+	case INTEL_TC_PIN_ASSIGNMENT_C:
+	case INTEL_TC_PIN_ASSIGNMENT_E:
 		return 4;
 	}
 }
@@ -340,10 +335,10 @@ static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 	default:
 		MISSING_CASE(pin_mask);
 		fallthrough;
-	case DP_PIN_ASSIGNMENT_D:
+	case INTEL_TC_PIN_ASSIGNMENT_D:
 		return 2;
-	case DP_PIN_ASSIGNMENT_C:
-	case DP_PIN_ASSIGNMENT_E:
+	case INTEL_TC_PIN_ASSIGNMENT_C:
+	case INTEL_TC_PIN_ASSIGNMENT_E:
 		return 4;
 	}
 }
diff --git a/drivers/gpu/drm/i915/display/intel_tc.h b/drivers/gpu/drm/i915/display/intel_tc.h
index 26c4265368c1..7ea7e8123971 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.h
+++ b/drivers/gpu/drm/i915/display/intel_tc.h
@@ -12,6 +12,75 @@ struct intel_crtc_state;
 struct intel_digital_port;
 struct intel_encoder;
 
+/*
+ * The following enum values must stay fixed, as they match the corresponding
+ * pin assignment fields in the PORT_TX_DFLEXPA1 and TCSS_DDI_STATUS registers.
+ */
+enum intel_tc_pin_assignment {            /* Lanes (a)   Signal/   Cable   Notes   */
+					  /* DP    USB   Rate (b)  type            */
+	INTEL_TC_PIN_ASSIGNMENT_NONE = 0, /* 4     -     -         -       (c)     */
+	INTEL_TC_PIN_ASSIGNMENT_A,        /* 2/4   0     GEN2      TC->TC  (d,e)   */
+	INTEL_TC_PIN_ASSIGNMENT_B,        /* 1/2   1     GEN2      TC->TC  (d,f,g) */
+	INTEL_TC_PIN_ASSIGNMENT_C,        /* 4     0     DP2       TC->TC  (h)     */
+	INTEL_TC_PIN_ASSIGNMENT_D,        /* 2     1     DP2       TC->TC  (h,g)   */
+	INTEL_TC_PIN_ASSIGNMENT_E,        /* 4     0     DP2       TC->DP          */
+	INTEL_TC_PIN_ASSIGNMENT_F,        /* 2     1     GEN1/DP1  TC->DP  (d,g,i) */
+	/*
+	 * (a) - DP unidirectional lanes, each lane using 1 differential signal
+	 *       pair.
+	 *     - USB SuperSpeed bidirectional lane, using 2 differential (TX and
+	 *       RX) signal pairs.
+	 *     - USB 2.0 (HighSpeed) unidirectional lane, using 1 differential
+	 *       signal pair. Not indicated, this lane is always present on pin
+	 *       assignments A-D and never present on pin assignments E/F.
+	 * (b) - GEN1: USB 3.1 GEN1 bit rate (5 Gbps) and signaling. This
+	 *             is used for transferring only a USB stream.
+	 *     - GEN2: USB 3.1 GEN2 bit rate (10 Gbps) and signaling. This
+	 *             allows transferring an HBR3 (8.1 Gbps) DP stream.
+	 *     - DP1:  Display Port signaling defined by the DP v1.3 Standard,
+	 *             with a maximum bit rate of HBR3.
+	 *     - DP2:  Display Port signaling defined by the DP v2.1 Standard,
+	 *             with a maximum bit rate defined by the DP Alt Mode
+	 *             v2.1a Standard depending on the cable type as follows:
+	 *             - Passive (Full-Featured) USB 3.2 GEN1
+	 *               TC->TC cables (CC3G1-X)                        : UHBR10
+	 *             - Passive (Full-Featured) USB 3.2/4 GEN2 and
+	 *               Thunderbolt Alt Mode GEN2
+	 *               TC->TC cables (CC3G2-X)                    all : UHBR10
+	 *                                                    DP54 logo : UHBR13.5
+	 *             - Passive (Full-Featured) USB4 GEN3+ and
+	 *               Thunderbolt Alt Mode GEN3+
+	 *               TC->TC cables (CC4G3-X)                    all : UHBR13.5
+	 *                                                    DP80 logo : UHBR20
+	 *             - Active Re-Timed or
+	 *               Active Linear Re-driven (LRD)
+	 *               USB3.2 GEN1/2 and USB4 GEN2+
+	 *               TC->TC cables                              all : HBR3
+	 *                                               with DP_BR CTS : UHBR10
+	 *                                                    DP54 logo : UHBR13.5
+	 *                                                    DP80 logo : UHBR20
+	 *             - Passive/Active Re-Timed or
+	 *               Active Linear Re-driven (LRD)
+	 *               TC->DP cables         with DP_BR CTS/DP8K logo : HBR3
+	 *                                               with DP_BR CTS : UHBR10
+	 *                                                    DP54 logo : UHBR13.5
+	 *                                                    DP80 logo : UHBR20
+	 * (c) Used in TBT-alt/legacy modes and on LNL+ after the sink
+	 *     disconnected in DP-alt mode.
+	 * (d) Only defined by the DP Alt Standard v1.0a, deprecated by v1.0b,
+	 *     only supported on ICL.
+	 * (e) GEN2 passive 1 m cable: 4 DP lanes, GEN2 active cable: 2 DP lanes.
+	 * (f) GEN2 passive 1 m cable: 2 DP lanes, GEN2 active cable: 1 DP lane.
+	 * (g) These pin assignments are also referred to as (USB/DP)
+	 *     multifunction or Multifunction Display Port (MFD) modes.
+	 * (h) Also used where one end of the cable is a captive connector,
+	 *     attached to a DP->HDMI/DVI/VGA converter.
+	 * (i) The DP end of the cable is a captive connector attached to a
+	 *     (DP/USB) multifunction dock as defined by the DockPort v1.0a
+	 *     specification.
+	 */
+};
+
 bool intel_tc_port_in_tbt_alt_mode(struct intel_digital_port *dig_port);
 bool intel_tc_port_in_dp_alt_mode(struct intel_digital_port *dig_port);
 bool intel_tc_port_in_legacy_mode(struct intel_digital_port *dig_port);

From 4ae33f0a5d1cef2830c364fd7dd6260ebfda9abd Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:51 +0300
Subject: [PATCH 0422/1455] drm/i915/tc: Pass pin assignment value around using
 the pin assignment enum

Pass around the pin assignment value via the corresponding enum instead
of a plain integer.

While at it rename intel_tc_port_get_pin_assignment_mask() to
intel_tc_port_get_pin_assignment(), since the value returned is not a
mask.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-11-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_ddi.c | 19 ++++++++++---------
 drivers/gpu/drm/i915/display/intel_tc.c  | 14 ++++++++------
 drivers/gpu/drm/i915/display/intel_tc.h  |  3 ++-
 3 files changed, 20 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 18040f2c8c0d..4e4ea3a0ff83 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2166,7 +2166,8 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port,
 {
 	struct intel_display *display = to_intel_display(crtc_state);
 	enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base);
-	u32 ln0, ln1, pin_assignment;
+	enum intel_tc_pin_assignment pin_assignment;
+	u32 ln0, ln1;
 	u8 width;
 
 	if (DISPLAY_VER(display) >= 14)
@@ -2188,11 +2189,11 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port,
 	ln1 &= ~(MG_DP_MODE_CFG_DP_X1_MODE | MG_DP_MODE_CFG_DP_X2_MODE);
 
 	/* DPPATC */
-	pin_assignment = intel_tc_port_get_pin_assignment_mask(dig_port);
+	pin_assignment = intel_tc_port_get_pin_assignment(dig_port);
 	width = crtc_state->lane_count;
 
 	switch (pin_assignment) {
-	case 0x0:
+	case INTEL_TC_PIN_ASSIGNMENT_NONE:
 		drm_WARN_ON(display->drm,
 			    !intel_tc_port_in_legacy_mode(dig_port));
 		if (width == 1) {
@@ -2202,20 +2203,20 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port,
 			ln1 |= MG_DP_MODE_CFG_DP_X2_MODE;
 		}
 		break;
-	case 0x1:
+	case INTEL_TC_PIN_ASSIGNMENT_A:
 		if (width == 4) {
 			ln0 |= MG_DP_MODE_CFG_DP_X2_MODE;
 			ln1 |= MG_DP_MODE_CFG_DP_X2_MODE;
 		}
 		break;
-	case 0x2:
+	case INTEL_TC_PIN_ASSIGNMENT_B:
 		if (width == 2) {
 			ln0 |= MG_DP_MODE_CFG_DP_X2_MODE;
 			ln1 |= MG_DP_MODE_CFG_DP_X2_MODE;
 		}
 		break;
-	case 0x3:
-	case 0x5:
+	case INTEL_TC_PIN_ASSIGNMENT_C:
+	case INTEL_TC_PIN_ASSIGNMENT_E:
 		if (width == 1) {
 			ln0 |= MG_DP_MODE_CFG_DP_X1_MODE;
 			ln1 |= MG_DP_MODE_CFG_DP_X1_MODE;
@@ -2224,8 +2225,8 @@ icl_program_mg_dp_mode(struct intel_digital_port *dig_port,
 			ln1 |= MG_DP_MODE_CFG_DP_X2_MODE;
 		}
 		break;
-	case 0x4:
-	case 0x6:
+	case INTEL_TC_PIN_ASSIGNMENT_D:
+	case INTEL_TC_PIN_ASSIGNMENT_F:
 		if (width == 1) {
 			ln0 |= MG_DP_MODE_CFG_DP_X1_MODE;
 			ln1 |= MG_DP_MODE_CFG_DP_X1_MODE;
diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index ffead1cbf10f..31663029642a 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -277,7 +277,8 @@ static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port)
 	return lane_mask >> DP_LANE_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
 }
 
-u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port)
+enum intel_tc_pin_assignment
+intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
@@ -299,8 +300,9 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 	struct intel_display *display = to_intel_display(dig_port);
 	enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
+	enum intel_tc_pin_assignment pin_assignment;
 	intel_wakeref_t wakeref;
-	u32 val, pin_assignment;
+	u32 val;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
 		val = intel_de_read(display, TCSS_DDI_STATUS(tc_port));
@@ -327,13 +329,13 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 
 static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
-	u32 pin_mask;
+	enum intel_tc_pin_assignment pin_assignment;
 
-	pin_mask = intel_tc_port_get_pin_assignment_mask(dig_port);
+	pin_assignment = intel_tc_port_get_pin_assignment(dig_port);
 
-	switch (pin_mask) {
+	switch (pin_assignment) {
 	default:
-		MISSING_CASE(pin_mask);
+		MISSING_CASE(pin_assignment);
 		fallthrough;
 	case INTEL_TC_PIN_ASSIGNMENT_D:
 		return 2;
diff --git a/drivers/gpu/drm/i915/display/intel_tc.h b/drivers/gpu/drm/i915/display/intel_tc.h
index 7ea7e8123971..fff8b96e4972 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.h
+++ b/drivers/gpu/drm/i915/display/intel_tc.h
@@ -88,7 +88,8 @@ bool intel_tc_port_handles_hpd_glitches(struct intel_digital_port *dig_port);
 
 bool intel_tc_port_connected(struct intel_encoder *encoder);
 
-u32 intel_tc_port_get_pin_assignment_mask(struct intel_digital_port *dig_port);
+enum intel_tc_pin_assignment
+intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port);
 int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port);
 void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,
 				      int required_lanes);

From 08d4fb9548d777fa1a7099beeade8888598f1323 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:52 +0300
Subject: [PATCH 0423/1455] drm/i915/tc: Handle pin assignment NONE on all
 platforms

For consistency, handle pin assignment NONE on all platforms similarly
to LNL+. On earlier platforms the driver doesn't actually see this pin
assignment - as it's not valid on a connected DP-alt PHY - however it's
a valid HW setting even on those platforms, for instance in legacy mode.
Handle this pin assignment on earlier platforms as well, so that the way
to query the pin assignment can be unified by a follow-up change.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-12-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 31663029642a..d1f17d2f236c 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -334,6 +334,8 @@ static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 	pin_assignment = intel_tc_port_get_pin_assignment(dig_port);
 
 	switch (pin_assignment) {
+	case INTEL_TC_PIN_ASSIGNMENT_NONE:
+		return 0;
 	default:
 		MISSING_CASE(pin_assignment);
 		fallthrough;

From 7b6503c500fbab7f47e59d78f4e5f268d4962438 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:53 +0300
Subject: [PATCH 0424/1455] drm/i915/tc: Validate the pin assignment on all
 platforms

Validate the pin assignment on ICL-TGL, similarly to how this is done on
MTL+. ICL supports all the pin assignments, while TGL+ supports only the
NONE, C, D, E pin assignments.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-13-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 28 ++++++++++++++++++++-----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index d1f17d2f236c..017096724163 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -282,17 +282,35 @@ intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
 {
 	struct intel_display *display = to_intel_display(dig_port);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
+	enum intel_tc_pin_assignment pin_assignment;
 	intel_wakeref_t wakeref;
-	u32 pin_mask;
+	u32 val;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
-		pin_mask = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia));
+		val = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia));
 
-	drm_WARN_ON(display->drm, pin_mask == 0xffffffff);
+	drm_WARN_ON(display->drm, val == 0xffffffff);
 	assert_tc_cold_blocked(tc);
 
-	return (pin_mask & DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx)) >>
-	       DP_PIN_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
+	pin_assignment = (val & DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx)) >>
+			 DP_PIN_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
+
+	switch (pin_assignment) {
+	case INTEL_TC_PIN_ASSIGNMENT_A:
+	case INTEL_TC_PIN_ASSIGNMENT_B:
+	case INTEL_TC_PIN_ASSIGNMENT_F:
+		drm_WARN_ON(display->drm, DISPLAY_VER(display) > 11);
+		break;
+	case INTEL_TC_PIN_ASSIGNMENT_NONE:
+	case INTEL_TC_PIN_ASSIGNMENT_C:
+	case INTEL_TC_PIN_ASSIGNMENT_D:
+	case INTEL_TC_PIN_ASSIGNMENT_E:
+		break;
+	default:
+		MISSING_CASE(pin_assignment);
+	}
+
+	return pin_assignment;
 }
 
 static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)

From faf4f2a0264dd5bf3845e3dcff742d93e42ac47a Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:54 +0300
Subject: [PATCH 0425/1455] drm/i915/tc: Unify the way to get the pin
 assignment on all platforms

Unify the way to get the pin assignment on all platforms. This removes
the duplication in the helper functions in this and a follow-up change.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-14-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 47 ++++++++++++++-----------
 1 file changed, 27 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 017096724163..62d60f02d0c6 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -277,23 +277,32 @@ static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port)
 	return lane_mask >> DP_LANE_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
 }
 
-enum intel_tc_pin_assignment
-intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
+static enum intel_tc_pin_assignment
+get_pin_assignment(struct intel_tc_port *tc)
 {
-	struct intel_display *display = to_intel_display(dig_port);
-	struct intel_tc_port *tc = to_tc_port(dig_port);
+	struct intel_display *display = to_intel_display(tc->dig_port);
+	enum tc_port tc_port = intel_encoder_to_tc(&tc->dig_port->base);
 	enum intel_tc_pin_assignment pin_assignment;
 	intel_wakeref_t wakeref;
+	i915_reg_t reg;
+	u32 mask;
 	u32 val;
 
+	if (DISPLAY_VER(display) >= 20) {
+		reg = TCSS_DDI_STATUS(tc_port);
+		mask = TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK;
+	} else {
+		reg = PORT_TX_DFLEXPA1(tc->phy_fia);
+		mask = DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx);
+	}
+
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
-		val = intel_de_read(display, PORT_TX_DFLEXPA1(tc->phy_fia));
+		val = intel_de_read(display, reg);
 
 	drm_WARN_ON(display->drm, val == 0xffffffff);
 	assert_tc_cold_blocked(tc);
 
-	pin_assignment = (val & DP_PIN_ASSIGNMENT_MASK(tc->phy_fia_idx)) >>
-			 DP_PIN_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
+	pin_assignment = (val & mask) >> (ffs(mask) - 1);
 
 	switch (pin_assignment) {
 	case INTEL_TC_PIN_ASSIGNMENT_A:
@@ -315,21 +324,10 @@ intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
 
 static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
-	struct intel_display *display = to_intel_display(dig_port);
-	enum tc_port tc_port = intel_encoder_to_tc(&dig_port->base);
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 	enum intel_tc_pin_assignment pin_assignment;
-	intel_wakeref_t wakeref;
-	u32 val;
 
-	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref)
-		val = intel_de_read(display, TCSS_DDI_STATUS(tc_port));
-
-	drm_WARN_ON(display->drm, val == 0xffffffff);
-	assert_tc_cold_blocked(tc);
-
-	pin_assignment =
-		REG_FIELD_GET(TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK, val);
+	pin_assignment = get_pin_assignment(tc);
 
 	switch (pin_assignment) {
 	case INTEL_TC_PIN_ASSIGNMENT_NONE:
@@ -347,9 +345,10 @@ static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 
 static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
+	struct intel_tc_port *tc = to_tc_port(dig_port);
 	enum intel_tc_pin_assignment pin_assignment;
 
-	pin_assignment = intel_tc_port_get_pin_assignment(dig_port);
+	pin_assignment = get_pin_assignment(tc);
 
 	switch (pin_assignment) {
 	case INTEL_TC_PIN_ASSIGNMENT_NONE:
@@ -420,6 +419,14 @@ int intel_tc_port_max_lane_count(struct intel_digital_port *dig_port)
 	return tc->max_lane_count;
 }
 
+enum intel_tc_pin_assignment
+intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
+{
+	struct intel_tc_port *tc = to_tc_port(dig_port);
+
+	return get_pin_assignment(tc);
+}
+
 void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,
 				      int required_lanes)
 {

From c96ba5883df7fabc5df34e1486df538d514f4c12 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:55 +0300
Subject: [PATCH 0426/1455] drm/i915/tc: Unify the way to get the max lane
 count value on MTL+

Unify the way to get the max lane count value on all MTL+ platforms,
reducing the code duplication.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-15-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 24 ------------------------
 1 file changed, 24 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 62d60f02d0c6..7248d60bbf0b 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -322,27 +322,6 @@ get_pin_assignment(struct intel_tc_port *tc)
 	return pin_assignment;
 }
 
-static int lnl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
-{
-	struct intel_tc_port *tc = to_tc_port(dig_port);
-	enum intel_tc_pin_assignment pin_assignment;
-
-	pin_assignment = get_pin_assignment(tc);
-
-	switch (pin_assignment) {
-	case INTEL_TC_PIN_ASSIGNMENT_NONE:
-		return 0;
-	default:
-		MISSING_CASE(pin_assignment);
-		fallthrough;
-	case INTEL_TC_PIN_ASSIGNMENT_D:
-		return 2;
-	case INTEL_TC_PIN_ASSIGNMENT_C:
-	case INTEL_TC_PIN_ASSIGNMENT_E:
-		return 4;
-	}
-}
-
 static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 {
 	struct intel_tc_port *tc = to_tc_port(dig_port);
@@ -395,9 +374,6 @@ static int get_max_lane_count(struct intel_tc_port *tc)
 	if (tc->mode != TC_PORT_DP_ALT)
 		return 4;
 
-	if (DISPLAY_VER(display) >= 20)
-		return lnl_tc_port_get_max_lane_count(dig_port);
-
 	if (DISPLAY_VER(display) >= 14)
 		return mtl_tc_port_get_max_lane_count(dig_port);
 

From 70435fce3225631011f36ec9b0dd1f8c9e90e986 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:56 +0300
Subject: [PATCH 0427/1455] drm/i915/tc: Handle non-TC encoders when getting
 the pin assignment

For consistency, handle the case where
intel_tc_port_get_pin_assignment() is called for a non-TypeC encoder,
returning the default NONE pin assignment value, similarly to how this
is done in intel_tc_port_max_lane_count().

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-16-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 7248d60bbf0b..169eb61e22ea 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -400,6 +400,9 @@ intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
 {
 	struct intel_tc_port *tc = to_tc_port(dig_port);
 
+	if (!intel_encoder_is_tc(&dig_port->base))
+		return INTEL_TC_PIN_ASSIGNMENT_NONE;
+
 	return get_pin_assignment(tc);
 }
 

From 76ed3b0226db373ce033291077c1c20e1eaf5d2d Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:57 +0300
Subject: [PATCH 0428/1455] drm/i915/tc: Pass intel_tc_port to internal lane
 mask/count helpers

Pass the intel_tc_port pointer instead of intel_digital_port to all lane
mask and count query helpers internal to intel_tc.c, to avoid the
redundant intel_digital_port -> intel_tc_port conversions.

While at it shorten the function names, keeping the intel_tc_port_
prefix only for exported functions and use the mtl_, icl_ prefixes
making it clear which platforms a given query function is specific for.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-17-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 169eb61e22ea..073c434fd378 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -260,10 +260,9 @@ assert_tc_port_power_enabled(struct intel_tc_port *tc)
 		    !intel_display_power_is_enabled(display, tc_port_power_domain(tc)));
 }
 
-static u32 intel_tc_port_get_lane_mask(struct intel_digital_port *dig_port)
+static u32 get_lane_mask(struct intel_tc_port *tc)
 {
-	struct intel_display *display = to_intel_display(dig_port);
-	struct intel_tc_port *tc = to_tc_port(dig_port);
+	struct intel_display *display = to_intel_display(tc->dig_port);
 	intel_wakeref_t wakeref;
 	u32 lane_mask;
 
@@ -322,9 +321,8 @@ get_pin_assignment(struct intel_tc_port *tc)
 	return pin_assignment;
 }
 
-static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
+static int mtl_get_max_lane_count(struct intel_tc_port *tc)
 {
-	struct intel_tc_port *tc = to_tc_port(dig_port);
 	enum intel_tc_pin_assignment pin_assignment;
 
 	pin_assignment = get_pin_assignment(tc);
@@ -343,11 +341,11 @@ static int mtl_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 	}
 }
 
-static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
+static int icl_get_max_lane_count(struct intel_tc_port *tc)
 {
 	u32 lane_mask = 0;
 
-	lane_mask = intel_tc_port_get_lane_mask(dig_port);
+	lane_mask = get_lane_mask(tc);
 
 	switch (lane_mask) {
 	default:
@@ -369,15 +367,14 @@ static int intel_tc_port_get_max_lane_count(struct intel_digital_port *dig_port)
 static int get_max_lane_count(struct intel_tc_port *tc)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
-	struct intel_digital_port *dig_port = tc->dig_port;
 
 	if (tc->mode != TC_PORT_DP_ALT)
 		return 4;
 
 	if (DISPLAY_VER(display) >= 14)
-		return mtl_tc_port_get_max_lane_count(dig_port);
+		return mtl_get_max_lane_count(tc);
 
-	return intel_tc_port_get_max_lane_count(dig_port);
+	return icl_get_max_lane_count(tc);
 }
 
 static void read_pin_configuration(struct intel_tc_port *tc)

From cca7c083a3f622164a3615ed243e6a5ccdd1dd10 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:58 +0300
Subject: [PATCH 0429/1455] dmc/i915/tc: Report pin assignment NONE in TBT-alt
 mode

The pin assignment is only relevant in case the PHY is owned by the
display, that is in legacy and DP-alt mode. In TBT-alt mode the PHY is
owned by the TBT FW/driver and so the pin assignment/configuration is
managed by those components. A follow-up change will cache the pin
assignment value in all the TypeC modes - querying this by calling
get_pin_assignment() - prepare for that here, by reporting pin
assignment NONE in the TBT-alt mode.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-18-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 073c434fd378..3e53332f804f 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -287,6 +287,9 @@ get_pin_assignment(struct intel_tc_port *tc)
 	u32 mask;
 	u32 val;
 
+	if (tc->mode == TC_PORT_TBT_ALT)
+		return INTEL_TC_PIN_ASSIGNMENT_NONE;
+
 	if (DISPLAY_VER(display) >= 20) {
 		reg = TCSS_DDI_STATUS(tc_port);
 		mask = TCSS_DDI_STATUS_PIN_ASSIGNMENT_MASK;

From 24d1cdacac11ddc3388714c5f82d31a480659ac7 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:36:59 +0300
Subject: [PATCH 0430/1455] drm/i915/tc: Cache the pin assignment value

Cache the pin assignment value. This is more consistent with the way the
max lane count value is tracked and a bit more efficient than reading
out the same value from HW each time it's queried.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-19-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 3e53332f804f..706f60a7d291 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -61,6 +61,7 @@ struct intel_tc_port {
 	enum tc_port_mode mode;
 	enum tc_port_mode init_mode;
 	enum phy_fia phy_fia;
+	enum intel_tc_pin_assignment pin_assignment;
 	u8 phy_fia_idx;
 	u8 max_lane_count;
 };
@@ -382,6 +383,7 @@ static int get_max_lane_count(struct intel_tc_port *tc)
 
 static void read_pin_configuration(struct intel_tc_port *tc)
 {
+	tc->pin_assignment = get_pin_assignment(tc);
 	tc->max_lane_count = get_max_lane_count(tc);
 }
 
@@ -403,7 +405,7 @@ intel_tc_port_get_pin_assignment(struct intel_digital_port *dig_port)
 	if (!intel_encoder_is_tc(&dig_port->base))
 		return INTEL_TC_PIN_ASSIGNMENT_NONE;
 
-	return get_pin_assignment(tc);
+	return tc->pin_assignment;
 }
 
 void intel_tc_port_set_fia_lane_count(struct intel_digital_port *dig_port,

From dcf101872d03a2394ed5f5aa5b2b036080285e3c Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Tue, 5 Aug 2025 10:37:00 +0300
Subject: [PATCH 0431/1455] drm/i915/tc: Debug print the pin assignment and max
 lane count

Debug print the TypeC pin assignment and max lane count value during HW
readout and after resetting the TypeC mode.

Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250805073700.642107-20-imre.deak@intel.com
Signed-off-by: Imre Deak <imre.deak@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 706f60a7d291..583fed5a386d 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -277,6 +277,14 @@ static u32 get_lane_mask(struct intel_tc_port *tc)
 	return lane_mask >> DP_LANE_ASSIGNMENT_SHIFT(tc->phy_fia_idx);
 }
 
+static char pin_assignment_name(enum intel_tc_pin_assignment pin_assignment)
+{
+	if (pin_assignment == INTEL_TC_PIN_ASSIGNMENT_NONE)
+		return '-';
+
+	return 'A' + pin_assignment - INTEL_TC_PIN_ASSIGNMENT_A;
+}
+
 static enum intel_tc_pin_assignment
 get_pin_assignment(struct intel_tc_port *tc)
 {
@@ -1510,10 +1518,13 @@ static void intel_tc_port_reset_mode(struct intel_tc_port *tc,
 	if (!force_disconnect)
 		tc_phy_connect(tc, required_lanes);
 
-	drm_dbg_kms(display->drm, "Port %s: TC port mode reset (%s -> %s)\n",
+	drm_dbg_kms(display->drm,
+		    "Port %s: TC port mode reset (%s -> %s) pin assignment: %c max lanes: %d\n",
 		    tc->port_name,
 		    tc_port_mode_name(old_tc_mode),
-		    tc_port_mode_name(tc->mode));
+		    tc_port_mode_name(tc->mode),
+		    pin_assignment_name(tc->pin_assignment),
+		    tc->max_lane_count);
 }
 
 static bool intel_tc_port_needs_reset(struct intel_tc_port *tc)
@@ -1668,9 +1679,11 @@ void intel_tc_port_sanitize_mode(struct intel_digital_port *dig_port,
 		__intel_tc_port_put_link(tc);
 	}
 
-	drm_dbg_kms(display->drm, "Port %s: sanitize mode (%s)\n",
+	drm_dbg_kms(display->drm, "Port %s: sanitize mode (%s) pin assignment: %c max lanes: %d\n",
 		    tc->port_name,
-		    tc_port_mode_name(tc->mode));
+		    tc_port_mode_name(tc->mode),
+		    pin_assignment_name(tc->pin_assignment),
+		    tc->max_lane_count);
 
 	mutex_unlock(&tc->lock);
 }

From 9465dd7c400d47439d3446de5b3f1ecfb5ea1bc6 Mon Sep 17 00:00:00 2001
From: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Date: Fri, 8 Aug 2025 13:49:30 +0530
Subject: [PATCH 0432/1455] drm/i915/display: Add power well mapping for WCL

WCL has 3 pipes and two TC ports, create power well mapping to reflect
HW. Rest remains similar to Xe3 power well configuration.

v2: Remove TC3/4 ports as they do not exist.

Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250808081931.4101388-1-chaitanya.kumar.borah@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 .../i915/display/intel_display_power_map.c    | 57 ++++++++++++++++++-
 1 file changed, 56 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_map.c b/drivers/gpu/drm/i915/display/intel_display_power_map.c
index 77268802b55e..39b71fffa2cd 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_map.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_map.c
@@ -1717,6 +1717,59 @@ static const struct i915_power_well_desc_list xe3lpd_power_wells[] = {
 	I915_PW_DESCRIPTORS(xe2lpd_power_wells_pica),
 };
 
+static const struct i915_power_well_desc wcl_power_wells_main[] = {
+	{
+		.instances = &I915_PW_INSTANCES(
+			I915_PW("PW_2", &xe3lpd_pwdoms_pw_2,
+				.hsw.idx = ICL_PW_CTL_IDX_PW_2,
+				.id = SKL_DISP_PW_2),
+		),
+		.ops = &hsw_power_well_ops,
+		.has_vga = true,
+		.has_fuses = true,
+	}, {
+		.instances = &I915_PW_INSTANCES(
+			I915_PW("PW_A", &xelpd_pwdoms_pw_a,
+				.hsw.idx = XELPD_PW_CTL_IDX_PW_A),
+		),
+		.ops = &hsw_power_well_ops,
+		.irq_pipe_mask = BIT(PIPE_A),
+		.has_fuses = true,
+	}, {
+		.instances = &I915_PW_INSTANCES(
+			I915_PW("PW_B", &xe3lpd_pwdoms_pw_b,
+				.hsw.idx = XELPD_PW_CTL_IDX_PW_B),
+		),
+		.ops = &hsw_power_well_ops,
+		.irq_pipe_mask = BIT(PIPE_B),
+		.has_fuses = true,
+	}, {
+		.instances = &I915_PW_INSTANCES(
+			I915_PW("PW_C", &xe3lpd_pwdoms_pw_c,
+				.hsw.idx = XELPD_PW_CTL_IDX_PW_C),
+		),
+		.ops = &hsw_power_well_ops,
+		.irq_pipe_mask = BIT(PIPE_C),
+		.has_fuses = true,
+	}, {
+		.instances = &I915_PW_INSTANCES(
+			I915_PW("AUX_A", &icl_pwdoms_aux_a, .xelpdp.aux_ch = AUX_CH_A),
+			I915_PW("AUX_B", &icl_pwdoms_aux_b, .xelpdp.aux_ch = AUX_CH_B),
+			I915_PW("AUX_TC1", &xelpdp_pwdoms_aux_tc1, .xelpdp.aux_ch = AUX_CH_USBC1),
+			I915_PW("AUX_TC2", &xelpdp_pwdoms_aux_tc2, .xelpdp.aux_ch = AUX_CH_USBC2),
+		),
+		.ops = &xelpdp_aux_power_well_ops,
+	},
+};
+
+static const struct i915_power_well_desc_list wcl_power_wells[] = {
+	I915_PW_DESCRIPTORS(i9xx_power_wells_always_on),
+	I915_PW_DESCRIPTORS(icl_power_wells_pw_1),
+	I915_PW_DESCRIPTORS(xe3lpd_power_wells_dcoff),
+	I915_PW_DESCRIPTORS(wcl_power_wells_main),
+	I915_PW_DESCRIPTORS(xe2lpd_power_wells_pica),
+};
+
 static void init_power_well_domains(const struct i915_power_well_instance *inst,
 				    struct i915_power_well *power_well)
 {
@@ -1824,7 +1877,9 @@ int intel_display_power_map_init(struct i915_power_domains *power_domains)
 		return 0;
 	}
 
-	if (DISPLAY_VER(display) >= 30)
+	if (DISPLAY_VERx100(display) == 3002)
+		return set_power_wells(power_domains, wcl_power_wells);
+	else if (DISPLAY_VER(display) >= 30)
 		return set_power_wells(power_domains, xe3lpd_power_wells);
 	else if (DISPLAY_VER(display) >= 20)
 		return set_power_wells(power_domains, xe2lpd_power_wells);

From 632633d095b6fc09ee38fd786778c02a003a8e67 Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Fri, 8 Aug 2025 13:49:31 +0530
Subject: [PATCH 0433/1455] drm/i915/wcl: Add display device info

Add device info for wildcat lake. WCL has 3 pipes and 2 TC ports.

Suggested-by: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250808081931.4101388-2-chaitanya.kumar.borah@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 .../gpu/drm/i915/display/intel_display_device.c   | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index 089cffabbad5..65f0efc35bb7 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -1354,6 +1354,19 @@ static const struct intel_display_device_info xe2_lpd_display = {
 	.__runtime_defaults.has_dbuf_overlap_detection = true,
 };
 
+static const struct intel_display_device_info wcl_display = {
+	XE_LPDP_FEATURES,
+
+	.__runtime_defaults.cpu_transcoder_mask =
+		BIT(TRANSCODER_A) | BIT(TRANSCODER_B) | BIT(TRANSCODER_C),
+	.__runtime_defaults.pipe_mask =
+		BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C),
+	.__runtime_defaults.fbc_mask =
+		BIT(INTEL_FBC_A) | BIT(INTEL_FBC_B) | BIT(INTEL_FBC_C),
+	.__runtime_defaults.port_mask =
+		BIT(PORT_A) | BIT(PORT_B) | BIT(PORT_TC1) | BIT(PORT_TC2),
+};
+
 static const struct intel_display_device_info xe2_hpd_display = {
 	XE_LPDP_FEATURES,
 	.__runtime_defaults.port_mask = BIT(PORT_A) |
@@ -1480,7 +1493,7 @@ static const struct {
 	{ 14,  1, &xe2_hpd_display },
 	{ 20,  0, &xe2_lpd_display },
 	{ 30,  0, &xe2_lpd_display },
-	{ 30,  2, &xe2_lpd_display },
+	{ 30,  2, &wcl_display },
 };
 
 static const struct intel_display_device_info *

From 46ab4278e9bbb16e3dc8cec52fcdcadc95936a89 Mon Sep 17 00:00:00 2001
From: Shankari Anand <shankari.ak0208@gmail.com>
Date: Wed, 16 Jul 2025 14:39:41 +0530
Subject: [PATCH 0434/1455] drm: nova: update ARef import from sync::aref

Update nova to import `ARef` from `sync::aref` instead of `types`.

This aligns with the ongoing effort to move `ARef` and
`AlwaysRefCounted` to sync.

Suggested-by: Benno Lossin <lossin@kernel.org>
Link: https://github.com/Rust-for-Linux/linux/issues/1173
Signed-off-by: Shankari Anand <shankari.ak0208@gmail.com>
Link: https://lore.kernel.org/r/20250716090941.811418-1-shankari.ak0208@gmail.com
[ Alter subject and commit message to be nova specific. - Danilo ]
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/nova/driver.rs | 4 +++-
 drivers/gpu/drm/nova/gem.rs    | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/nova/driver.rs b/drivers/gpu/drm/nova/driver.rs
index b28b2e05cc15..91b7380f83ab 100644
--- a/drivers/gpu/drm/nova/driver.rs
+++ b/drivers/gpu/drm/nova/driver.rs
@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: GPL-2.0
 
-use kernel::{auxiliary, c_str, device::Core, drm, drm::gem, drm::ioctl, prelude::*, types::ARef};
+use kernel::{
+    auxiliary, c_str, device::Core, drm, drm::gem, drm::ioctl, prelude::*, sync::aref::ARef,
+};
 
 use crate::file::File;
 use crate::gem::NovaObject;
diff --git a/drivers/gpu/drm/nova/gem.rs b/drivers/gpu/drm/nova/gem.rs
index 33b62d21400c..cd82773dab92 100644
--- a/drivers/gpu/drm/nova/gem.rs
+++ b/drivers/gpu/drm/nova/gem.rs
@@ -4,7 +4,7 @@ use kernel::{
     drm,
     drm::{gem, gem::BaseObject},
     prelude::*,
-    types::ARef,
+    sync::aref::ARef,
 };
 
 use crate::{

From ed07a76be7baa0bb164b152116486e4d9fed50dc Mon Sep 17 00:00:00 2001
From: "Rob Herring (Arm)" <robh@kernel.org>
Date: Thu, 3 Jul 2025 13:34:47 -0500
Subject: [PATCH 0435/1455] drm/simpledrm: Use
 of_reserved_mem_region_to_resource() for "memory-region"

Use the newly added of_reserved_mem_region_to_resource() function to
handle "memory-region" properties.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250703183447.2073902-1-robh@kernel.org
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
---
 drivers/gpu/drm/sysfb/simpledrm.c | 15 +++++----------
 1 file changed, 5 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/simpledrm.c b/drivers/gpu/drm/sysfb/simpledrm.c
index 8530a3ef8a7a..0358164a623c 100644
--- a/drivers/gpu/drm/sysfb/simpledrm.c
+++ b/drivers/gpu/drm/sysfb/simpledrm.c
@@ -4,7 +4,7 @@
 #include <linux/clk.h>
 #include <linux/of_clk.h>
 #include <linux/minmax.h>
-#include <linux/of_address.h>
+#include <linux/of_reserved_mem.h>
 #include <linux/platform_data/simplefb.h>
 #include <linux/platform_device.h>
 #include <linux/pm_domain.h>
@@ -179,22 +179,17 @@ simplefb_get_format_of(struct drm_device *dev, struct device_node *of_node)
 static struct resource *
 simplefb_get_memory_of(struct drm_device *dev, struct device_node *of_node)
 {
-	struct device_node *np;
-	struct resource *res;
+	struct resource r, *res;
 	int err;
 
-	np = of_parse_phandle(of_node, "memory-region", 0);
-	if (!np)
+	err = of_reserved_mem_region_to_resource(of_node, 0, &r);
+	if (err)
 		return NULL;
 
-	res = devm_kzalloc(dev->dev, sizeof(*res), GFP_KERNEL);
+	res = devm_kmemdup(dev->dev, &r, sizeof(r), GFP_KERNEL);
 	if (!res)
 		return ERR_PTR(-ENOMEM);
 
-	err = of_address_to_resource(np, 0, res);
-	if (err)
-		return ERR_PTR(err);
-
 	if (of_property_present(of_node, "reg"))
 		drm_warn(dev, "preferring \"memory-region\" over \"reg\" property\n");
 

From 0070851095d2954567510777976e40275f81ca1b Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Wed, 9 Jul 2025 20:43:25 +0200
Subject: [PATCH 0436/1455] drm: renesas: rz-du: mipi_dsi: Convert to
 RUNTIME_PM_OPS()

Convert the Renesas RZ/G2L MIPI DSI Encoder driver from
SET_RUNTIME_PM_OPS() to RUNTIME_PM_OPS() and pm_ptr().  This lets us
drop the __maybe_unused annotations from its runtime suspend and resume
callbacks, and reduces kernel size in case CONFIG_PM is disabled.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com>
Signed-off-by: Biju Das <biju.das.jz@bp.renesas.com>
Link: https://lore.kernel.org/r/cdfc1b8ec9e62553654639b9e9026bfed8dd07d1.1752086582.git.geert+renesas@glider.be
---
 drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c b/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c
index f87337c3cbb5..3b52dfc0ea1e 100644
--- a/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c
+++ b/drivers/gpu/drm/renesas/rz-du/rzg2l_mipi_dsi.c
@@ -913,7 +913,7 @@ static const struct mipi_dsi_host_ops rzg2l_mipi_dsi_host_ops = {
  * Power Management
  */
 
-static int __maybe_unused rzg2l_mipi_pm_runtime_suspend(struct device *dev)
+static int rzg2l_mipi_pm_runtime_suspend(struct device *dev)
 {
 	struct rzg2l_mipi_dsi *dsi = dev_get_drvdata(dev);
 
@@ -923,7 +923,7 @@ static int __maybe_unused rzg2l_mipi_pm_runtime_suspend(struct device *dev)
 	return 0;
 }
 
-static int __maybe_unused rzg2l_mipi_pm_runtime_resume(struct device *dev)
+static int rzg2l_mipi_pm_runtime_resume(struct device *dev)
 {
 	struct rzg2l_mipi_dsi *dsi = dev_get_drvdata(dev);
 	int ret;
@@ -940,7 +940,7 @@ static int __maybe_unused rzg2l_mipi_pm_runtime_resume(struct device *dev)
 }
 
 static const struct dev_pm_ops rzg2l_mipi_pm_ops = {
-	SET_RUNTIME_PM_OPS(rzg2l_mipi_pm_runtime_suspend, rzg2l_mipi_pm_runtime_resume, NULL)
+	RUNTIME_PM_OPS(rzg2l_mipi_pm_runtime_suspend, rzg2l_mipi_pm_runtime_resume, NULL)
 };
 
 /* -----------------------------------------------------------------------------
@@ -1072,7 +1072,7 @@ static struct platform_driver rzg2l_mipi_dsi_platform_driver = {
 	.remove = rzg2l_mipi_dsi_remove,
 	.driver	= {
 		.name = "rzg2l-mipi-dsi",
-		.pm = &rzg2l_mipi_pm_ops,
+		.pm = pm_ptr(&rzg2l_mipi_pm_ops),
 		.of_match_table = rzg2l_mipi_dsi_of_table,
 	},
 };

From cbd3baeffbc08052ce7dc53f11bf5524b4411056 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 5 Aug 2025 14:56:56 +0300
Subject: [PATCH 0437/1455] drm/i915: silence rpm wakeref asserts on
 GEN11_GU_MISC_IIR access
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit 8d9908e8fe9c ("drm/i915/display: remove small micro-optimizations
in irq handling") not only removed the optimizations, it also enabled
wakeref asserts for the GEN11_GU_MISC_IIR access. Silence the asserts by
wrapping the access inside intel_display_rpm_assert_{block,unblock}().

Reported-by: "Jason A. Donenfeld" <Jason@zx2c4.com>
Closes: https://lore.kernel.org/r/aG0tWkfmxWtxl_xc@zx2c4.com
Fixes: 8d9908e8fe9c ("drm/i915/display: remove small micro-optimizations in irq handling")
Cc: stable@vger.kernel.org # v6.13+
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/20250805115656.832235-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_irq.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index fb25ec8adae3..68157f177b6a 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -1506,10 +1506,14 @@ u32 gen11_gu_misc_irq_ack(struct intel_display *display, const u32 master_ctl)
 	if (!(master_ctl & GEN11_GU_MISC_IRQ))
 		return 0;
 
+	intel_display_rpm_assert_block(display);
+
 	iir = intel_de_read(display, GEN11_GU_MISC_IIR);
 	if (likely(iir))
 		intel_de_write(display, GEN11_GU_MISC_IIR, iir);
 
+	intel_display_rpm_assert_unblock(display);
+
 	return iir;
 }
 

From 21ecdf45f1cdf0267fb4260333494fed26683e9d Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:02 +0300
Subject: [PATCH 0438/1455] drm/i915/display: pass display to HAS_PCH_*()
 macros

Convert some leftover i915 usages to struct intel_display.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/c521c9fa8b7d2c9a04d472f6c541a41822f76327.1754499174.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_irq.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c
index 68157f177b6a..123e054affbe 100644
--- a/drivers/gpu/drm/i915/display/intel_display_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_display_irq.c
@@ -1990,20 +1990,17 @@ void vlv_display_irq_postinstall(struct intel_display *display)
 
 void ibx_display_irq_reset(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
-
-	if (HAS_PCH_NOP(i915))
+	if (HAS_PCH_NOP(display))
 		return;
 
 	gen2_irq_reset(to_intel_uncore(display->drm), SDE_IRQ_REGS);
 
-	if (HAS_PCH_CPT(i915) || HAS_PCH_LPT(i915))
+	if (HAS_PCH_CPT(display) || HAS_PCH_LPT(display))
 		intel_de_write(display, SERR_INT, 0xffffffff);
 }
 
 void gen8_display_irq_reset(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	enum pipe pipe;
 
 	if (!HAS_DISPLAY(display))
@@ -2020,7 +2017,7 @@ void gen8_display_irq_reset(struct intel_display *display)
 	intel_display_irq_regs_reset(display, GEN8_DE_PORT_IRQ_REGS);
 	intel_display_irq_regs_reset(display, GEN8_DE_MISC_IRQ_REGS);
 
-	if (HAS_PCH_SPLIT(i915))
+	if (HAS_PCH_SPLIT(display))
 		ibx_display_irq_reset(display);
 }
 

From 5064edaa3446abe8bc649e0c206cc1c4897fd6a9 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:03 +0300
Subject: [PATCH 0439/1455] drm/i915/fb: pass display to HAS_GMCH() and
 DISPLAY_VER()

Convert some leftover i915 usages to struct intel_display.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/8a209e3f60039986794123e6b1631185c9234f8f.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_fb_pin.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb_pin.c b/drivers/gpu/drm/i915/display/intel_fb_pin.c
index d598a005f847..45af04cb0fb2 100644
--- a/drivers/gpu/drm/i915/display/intel_fb_pin.c
+++ b/drivers/gpu/drm/i915/display/intel_fb_pin.c
@@ -152,7 +152,7 @@ intel_fb_pin_to_ggtt(const struct drm_framebuffer *fb,
 	 * happy to scanout from anywhere within its global aperture.
 	 */
 	pinctl = 0;
-	if (HAS_GMCH(dev_priv))
+	if (HAS_GMCH(display))
 		pinctl |= PIN_MAPPABLE;
 
 	i915_gem_ww_ctx_init(&ww, true);
@@ -193,7 +193,7 @@ retry:
 		 * mode that matches the user configuration.
 		 */
 		ret = i915_vma_pin_fence(vma);
-		if (ret != 0 && DISPLAY_VER(dev_priv) < 4) {
+		if (ret != 0 && DISPLAY_VER(display) < 4) {
 			i915_vma_unpin(vma);
 			goto err_unpin;
 		}

From 3a28f5c19d4d555363aeb99986bb3761d1cc29a4 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:04 +0300
Subject: [PATCH 0440/1455] drm/i915/clockgating: pass display to
 for_each_pipe()

Underneath, for_each_pipe() really expects struct intel_display. Switch
to it in preparation for removing the transitional __to_intel_display()
macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/1abcd9b4429a25eb40c30691c876d3db089b1cf2.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/intel_clock_gating.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c
index f86a3629ae9e..b4dddd03eaf3 100644
--- a/drivers/gpu/drm/i915/intel_clock_gating.c
+++ b/drivers/gpu/drm/i915/intel_clock_gating.c
@@ -132,9 +132,10 @@ static void ibx_init_clock_gating(struct drm_i915_private *i915)
 
 static void g4x_disable_trickle_feed(struct drm_i915_private *dev_priv)
 {
+	struct intel_display *display = dev_priv->display;
 	enum pipe pipe;
 
-	for_each_pipe(dev_priv, pipe) {
+	for_each_pipe(display, pipe) {
 		intel_uncore_rmw(&dev_priv->uncore, DSPCNTR(dev_priv, pipe),
 				 0, DISP_TRICKLE_FEED_DISABLE);
 
@@ -218,7 +219,7 @@ static void cpt_init_clock_gating(struct drm_i915_private *i915)
 	/* The below fixes the weird display corruption, a few pixels shifted
 	 * downward, on (only) LVDS of some HP laptops with IVY.
 	 */
-	for_each_pipe(i915, pipe) {
+	for_each_pipe(display, pipe) {
 		val = intel_uncore_read(&i915->uncore, TRANS_CHICKEN2(pipe));
 		val |= TRANS_CHICKEN2_TIMING_OVERRIDE;
 		val &= ~TRANS_CHICKEN2_FDI_POLARITY_REVERSED;
@@ -229,7 +230,7 @@ static void cpt_init_clock_gating(struct drm_i915_private *i915)
 		intel_uncore_write(&i915->uncore, TRANS_CHICKEN2(pipe), val);
 	}
 	/* WADP0ClockGatingDisable */
-	for_each_pipe(i915, pipe) {
+	for_each_pipe(display, pipe) {
 		intel_uncore_write(&i915->uncore, TRANS_CHICKEN1(pipe),
 				   TRANS_CHICKEN1_DP0UNIT_GC_DISABLE);
 	}
@@ -421,6 +422,7 @@ static void skl_init_clock_gating(struct drm_i915_private *i915)
 
 static void bdw_init_clock_gating(struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
 	enum pipe pipe;
 
 	/* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */
@@ -432,7 +434,7 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915)
 	/* WaPsrDPAMaskVBlankInSRD:bdw */
 	intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD);
 
-	for_each_pipe(i915, pipe) {
+	for_each_pipe(display, pipe) {
 		/* WaPsrDPRSUnmaskVBlankInSRD:bdw */
 		intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe),
 				 0, BDW_UNMASK_VBL_TO_REGS_IN_SRD);
@@ -468,6 +470,7 @@ static void bdw_init_clock_gating(struct drm_i915_private *i915)
 
 static void hsw_init_clock_gating(struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
 	enum pipe pipe;
 
 	/* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */
@@ -476,7 +479,7 @@ static void hsw_init_clock_gating(struct drm_i915_private *i915)
 	/* WaPsrDPAMaskVBlankInSRD:hsw */
 	intel_uncore_rmw(&i915->uncore, CHICKEN_PAR1_1, 0, HSW_MASK_VBL_TO_PIPE_IN_SRD);
 
-	for_each_pipe(i915, pipe) {
+	for_each_pipe(display, pipe) {
 		/* WaPsrDPRSUnmaskVBlankInSRD:hsw */
 		intel_uncore_rmw(&i915->uncore, CHICKEN_PIPESL_1(pipe),
 				 0, HSW_UNMASK_VBL_TO_REGS_IN_SRD);

From 077ad673d986c041f43aadf57e5ca5829a36ffc9 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:05 +0300
Subject: [PATCH 0441/1455] drm/i915/clockgating: pass display to HAS_PCH_*()
 macros

Underneath, the HAS_PCH_*() macros expect struct intel_display. Switch
to it in preparation for removing the transitional __to_intel_display()
macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/824750abaf4587501d61dce964b36542d93931f7.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/intel_clock_gating.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c
index b4dddd03eaf3..bd919fa3bee5 100644
--- a/drivers/gpu/drm/i915/intel_clock_gating.c
+++ b/drivers/gpu/drm/i915/intel_clock_gating.c
@@ -308,11 +308,13 @@ static void gen6_init_clock_gating(struct drm_i915_private *i915)
 
 static void lpt_init_clock_gating(struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
+
 	/*
 	 * TODO: this bit should only be enabled when really needed, then
 	 * disabled when not needed anymore in order to save power.
 	 */
-	if (HAS_PCH_LPT_LP(i915))
+	if (HAS_PCH_LPT_LP(display))
 		intel_uncore_rmw(&i915->uncore, SOUTH_DSPCLK_GATE_D,
 				 0, PCH_LP_PARTITION_LEVEL_DISABLE);
 
@@ -356,7 +358,9 @@ static void dg2_init_clock_gating(struct drm_i915_private *i915)
 
 static void cnp_init_clock_gating(struct drm_i915_private *i915)
 {
-	if (!HAS_PCH_CNP(i915))
+	struct intel_display *display = i915->display;
+
+	if (!HAS_PCH_CNP(display))
 		return;
 
 	/* Display WA #1181 WaSouthDisplayDisablePWMCGEGating: cnp */
@@ -497,6 +501,8 @@ static void hsw_init_clock_gating(struct drm_i915_private *i915)
 
 static void ivb_init_clock_gating(struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
+
 	intel_uncore_write(&i915->uncore, ILK_DSPCLK_GATE_D, ILK_VRHUNIT_CLOCK_GATE_DISABLE);
 
 	/* WaFbcAsynchFlipDisableFbcQueue:ivb */
@@ -534,7 +540,7 @@ static void ivb_init_clock_gating(struct drm_i915_private *i915)
 	intel_uncore_rmw(&i915->uncore, GEN6_MBCUNIT_SNPCR, GEN6_MBC_SNPCR_MASK,
 			 GEN6_MBC_SNPCR_MED);
 
-	if (!HAS_PCH_NOP(i915))
+	if (!HAS_PCH_NOP(display))
 		cpt_init_clock_gating(i915);
 
 	gen6_check_mch_setup(i915);

From 0418dfd86144662bae9360e8c7899cf928bbf04e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:06 +0300
Subject: [PATCH 0442/1455] drm/i915/clockgating: pass display to DSPCNTR and
 DSPSURF register macros

Underneath, the macros expect struct intel_display. Switch to it in
preparation for removing the transitional __to_intel_display() macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/014410b59845b401950faff992540ea9b32ac213.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/intel_clock_gating.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c
index bd919fa3bee5..e501f4937510 100644
--- a/drivers/gpu/drm/i915/intel_clock_gating.c
+++ b/drivers/gpu/drm/i915/intel_clock_gating.c
@@ -136,13 +136,13 @@ static void g4x_disable_trickle_feed(struct drm_i915_private *dev_priv)
 	enum pipe pipe;
 
 	for_each_pipe(display, pipe) {
-		intel_uncore_rmw(&dev_priv->uncore, DSPCNTR(dev_priv, pipe),
+		intel_uncore_rmw(&dev_priv->uncore, DSPCNTR(display, pipe),
 				 0, DISP_TRICKLE_FEED_DISABLE);
 
-		intel_uncore_rmw(&dev_priv->uncore, DSPSURF(dev_priv, pipe),
+		intel_uncore_rmw(&dev_priv->uncore, DSPSURF(display, pipe),
 				 0, 0);
 		intel_uncore_posting_read(&dev_priv->uncore,
-					  DSPSURF(dev_priv, pipe));
+					  DSPSURF(display, pipe));
 	}
 }
 

From 8f7528425ca5f5d0d126ca32a4e793a84fee8afd Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:07 +0300
Subject: [PATCH 0443/1455] drm/i915/irq: pass display to macros that expect
 display

Underneath, the HAS_PCH_NOP(), DISPLAY_VER(), HAS_FBC(), and
HAS_HOTPLUG() macros really expect a struct intel_display. Switch to it
in preparation for removing the transitional __to_intel_display() macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/de3302dd9ebc21226a9dadcbcdeeaf01e57186be.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/i915_irq.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 191ed8bb1d9c..a5fa40ab5de2 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -439,7 +439,7 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg)
 	 * able to process them after we restore SDEIER (as soon as we restore
 	 * it, we'll get an interrupt if SDEIIR still has something to process
 	 * due to its back queue). */
-	if (!HAS_PCH_NOP(i915)) {
+	if (!HAS_PCH_NOP(display)) {
 		sde_ier = raw_reg_read(regs, SDEIER);
 		raw_reg_write(regs, SDEIER, 0);
 	}
@@ -459,7 +459,7 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg)
 	de_iir = raw_reg_read(regs, DEIIR);
 	if (de_iir) {
 		raw_reg_write(regs, DEIIR, de_iir);
-		if (DISPLAY_VER(i915) >= 7)
+		if (DISPLAY_VER(display) >= 7)
 			ivb_display_irq_handler(display, de_iir);
 		else
 			ilk_display_irq_handler(display, de_iir);
@@ -834,6 +834,7 @@ static void cherryview_irq_postinstall(struct drm_i915_private *dev_priv)
 
 static u32 i9xx_error_mask(struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
 	/*
 	 * On gen2/3 FBC generates (seemingly spurious)
 	 * display INVALID_GTT/INVALID_GTT_PTE table errors.
@@ -846,7 +847,7 @@ static u32 i9xx_error_mask(struct drm_i915_private *i915)
 	 * Unfortunately we can't mask off individual PGTBL_ER bits,
 	 * so we just have to mask off all page table errors via EMR.
 	 */
-	if (HAS_FBC(i915))
+	if (HAS_FBC(display))
 		return I915_ERROR_MEMORY_REFRESH;
 	else
 		return I915_ERROR_PAGE_TABLE |
@@ -924,12 +925,12 @@ static void i915_irq_postinstall(struct drm_i915_private *dev_priv)
 		I915_MASTER_ERROR_INTERRUPT |
 		I915_USER_INTERRUPT;
 
-	if (DISPLAY_VER(dev_priv) >= 3) {
+	if (DISPLAY_VER(display) >= 3) {
 		dev_priv->irq_mask &= ~I915_ASLE_INTERRUPT;
 		enable_mask |= I915_ASLE_INTERRUPT;
 	}
 
-	if (HAS_HOTPLUG(dev_priv)) {
+	if (HAS_HOTPLUG(display)) {
 		dev_priv->irq_mask &= ~I915_DISPLAY_PORT_INTERRUPT;
 		enable_mask |= I915_DISPLAY_PORT_INTERRUPT;
 	}
@@ -963,7 +964,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg)
 
 		ret = IRQ_HANDLED;
 
-		if (HAS_HOTPLUG(dev_priv) &&
+		if (HAS_HOTPLUG(display) &&
 		    iir & I915_DISPLAY_PORT_INTERRUPT)
 			hotplug_status = i9xx_hpd_irq_ack(display);
 

From 7eb949663bbc09c057dff987908c55297725d426 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:08 +0300
Subject: [PATCH 0444/1455] drm/i915/dram: pass display to macros that expect
 display

Underneath, the HAS_DISPLAY() and DISPLAY_VER() macros really expect
struct intel_display. Switch to it in preparation for removing the
transitional __to_intel_display() macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/ee8bba57924629f97945b9cf05727ae7c0dfb8fd.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 3eeaabdf59e8..b3c407cc200f 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -710,13 +710,14 @@ static int xelpdp_get_dram_info(struct drm_i915_private *i915, struct dram_info
 
 int intel_dram_detect(struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
 	struct dram_info *dram_info;
 	int ret;
 
 	detect_fsb_freq(i915);
 	detect_mem_freq(i915);
 
-	if (GRAPHICS_VER(i915) < 9 || IS_DG2(i915) || !HAS_DISPLAY(i915))
+	if (GRAPHICS_VER(i915) < 9 || IS_DG2(i915) || !HAS_DISPLAY(display))
 		return 0;
 
 	dram_info = drmm_kzalloc(&i915->drm, sizeof(*dram_info), GFP_KERNEL);
@@ -731,7 +732,7 @@ int intel_dram_detect(struct drm_i915_private *i915)
 	 */
 	dram_info->wm_lv_0_adjust_needed = !IS_BROXTON(i915) && !IS_GEMINILAKE(i915);
 
-	if (DISPLAY_VER(i915) >= 14)
+	if (DISPLAY_VER(display) >= 14)
 		ret = xelpdp_get_dram_info(i915, dram_info);
 	else if (GRAPHICS_VER(i915) >= 12)
 		ret = gen12_get_dram_info(i915, dram_info);

From e3929f84a1b3a6e65855c38c53001eb7b34e3955 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:09 +0300
Subject: [PATCH 0445/1455] drm/i915/gmch: pass display to DISPLAY_VER()

Underneath, DISPLAY_VER() really expects struct intel_display. Switch to
it in preparation for removing the transitional __to_intel_display()
macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/1f951102b3d073fd58c5211f399750aa80b51548.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_gmch.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_gmch.c b/drivers/gpu/drm/i915/soc/intel_gmch.c
index 5346b8dda79a..f210c9655b53 100644
--- a/drivers/gpu/drm/i915/soc/intel_gmch.c
+++ b/drivers/gpu/drm/i915/soc/intel_gmch.c
@@ -148,7 +148,8 @@ void intel_gmch_bar_teardown(struct drm_i915_private *i915)
 
 int intel_gmch_vga_set_state(struct drm_i915_private *i915, bool enable_decode)
 {
-	unsigned int reg = DISPLAY_VER(i915) >= 6 ? SNB_GMCH_CTRL : INTEL_GMCH_CTRL;
+	struct intel_display *display = i915->display;
+	unsigned int reg = DISPLAY_VER(display) >= 6 ? SNB_GMCH_CTRL : INTEL_GMCH_CTRL;
 	u16 gmch_ctrl;
 
 	if (pci_read_config_word(i915->gmch.pdev, reg, &gmch_ctrl)) {

From a1a8c68cdac0ec0d2c0f29d1155bff686bb55ac3 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:10 +0300
Subject: [PATCH 0446/1455] drm/i915/gem: pass display to HAS_DISPLAY()

Underneath, HAS_DISPLAY() really expects a struct intel_display. Switch
to it in preparation for removing the transitional __to_intel_display()
macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/79c87f75771a1d120cdd35dcfd6b380620ce99f1.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
index 86d9d2fcb6a6..e747f5ed195e 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
@@ -110,6 +110,7 @@ struct tiled_blits {
 
 static bool fastblit_supports_x_tiling(const struct drm_i915_private *i915)
 {
+	struct intel_display *display = i915->display;
 	int gen = GRAPHICS_VER(i915);
 
 	/* XY_FAST_COPY_BLT does not exist on pre-gen9 platforms */
@@ -121,7 +122,7 @@ static bool fastblit_supports_x_tiling(const struct drm_i915_private *i915)
 	if (GRAPHICS_VER_FULL(i915) < IP_VER(12, 55))
 		return false;
 
-	return HAS_DISPLAY(i915);
+	return HAS_DISPLAY(display);
 }
 
 static bool fast_blit_ok(const struct blit_buffer *buf)

From d2e184f8e16ab85876ddcc0779358fbfe5162414 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:11 +0300
Subject: [PATCH 0447/1455] drm/i915/switcheroo: pass display to HAS_DISPLAY()

Underneath, HAS_DISPLAY() really expects a struct intel_display. Switch
to it in preparation for removing the transitional __to_intel_display()
macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/c9828991eb8870c45271141bf0912880a3a65c44.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/i915_switcheroo.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_switcheroo.c b/drivers/gpu/drm/i915/i915_switcheroo.c
index 4c02a04be681..231d27497706 100644
--- a/drivers/gpu/drm/i915/i915_switcheroo.c
+++ b/drivers/gpu/drm/i915/i915_switcheroo.c
@@ -15,13 +15,14 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev,
 				      enum vga_switcheroo_state state)
 {
 	struct drm_i915_private *i915 = pdev_to_i915(pdev);
+	struct intel_display *display = i915->display;
 	pm_message_t pmm = { .event = PM_EVENT_SUSPEND };
 
 	if (!i915) {
 		dev_err(&pdev->dev, "DRM not initialized, aborting switch.\n");
 		return;
 	}
-	if (!HAS_DISPLAY(i915)) {
+	if (!HAS_DISPLAY(display)) {
 		dev_err(&pdev->dev, "Device state not initialized, aborting switch.\n");
 		return;
 	}
@@ -44,13 +45,14 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev,
 static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_i915_private *i915 = pdev_to_i915(pdev);
+	struct intel_display *display = i915->display;
 
 	/*
 	 * FIXME: open_count is protected by drm_global_mutex but that would lead to
 	 * locking inversion with the driver load path. And the access here is
 	 * completely racy anyway. So don't bother with locking for now.
 	 */
-	return i915 && HAS_DISPLAY(i915) && atomic_read(&i915->drm.open_count) == 0;
+	return i915 && HAS_DISPLAY(display) && atomic_read(&i915->drm.open_count) == 0;
 }
 
 static const struct vga_switcheroo_client_ops i915_switcheroo_ops = {

From 8d588c01d1fd4e0e9345f43eb9182634f6b08792 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:12 +0300
Subject: [PATCH 0448/1455] drm/i915/drv: pass display to HAS_DISPLAY()

Underneath, HAS_DISPLAY() really expects a struct intel_display. Switch
to it in preparation for removing the transitional __to_intel_display()
macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/e1789e82d876eabef550abf537ca5dd2b71cbe6c.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/i915_driver.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index c6263c6d3384..70f042ce8705 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -977,7 +977,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915)
 	intel_power_domains_disable(display);
 
 	drm_client_dev_suspend(&i915->drm, false);
-	if (HAS_DISPLAY(i915)) {
+	if (HAS_DISPLAY(display)) {
 		drm_kms_helper_poll_disable(&i915->drm);
 		intel_display_driver_disable_user_access(display);
 
@@ -989,7 +989,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915)
 	intel_irq_suspend(i915);
 	intel_hpd_cancel_work(display);
 
-	if (HAS_DISPLAY(i915))
+	if (HAS_DISPLAY(display))
 		intel_display_driver_suspend_access(display);
 
 	intel_encoder_suspend_all(display);
@@ -1060,7 +1060,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 	 * properly. */
 	intel_power_domains_disable(display);
 	drm_client_dev_suspend(dev, false);
-	if (HAS_DISPLAY(dev_priv)) {
+	if (HAS_DISPLAY(display)) {
 		drm_kms_helper_poll_disable(dev);
 		intel_display_driver_disable_user_access(display);
 	}
@@ -1072,7 +1072,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 	intel_irq_suspend(dev_priv);
 	intel_hpd_cancel_work(display);
 
-	if (HAS_DISPLAY(dev_priv))
+	if (HAS_DISPLAY(display))
 		intel_display_driver_suspend_access(display);
 
 	intel_encoder_suspend_all(display);
@@ -1219,7 +1219,7 @@ static int i915_drm_resume(struct drm_device *dev)
 	 */
 	intel_irq_resume(dev_priv);
 
-	if (HAS_DISPLAY(dev_priv))
+	if (HAS_DISPLAY(display))
 		drm_mode_config_reset(dev);
 
 	i915_gem_resume(dev_priv);
@@ -1228,14 +1228,14 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	intel_clock_gating_init(dev_priv);
 
-	if (HAS_DISPLAY(dev_priv))
+	if (HAS_DISPLAY(display))
 		intel_display_driver_resume_access(display);
 
 	intel_hpd_init(display);
 
 	intel_display_driver_resume(display);
 
-	if (HAS_DISPLAY(dev_priv)) {
+	if (HAS_DISPLAY(display)) {
 		intel_display_driver_enable_user_access(display);
 		drm_kms_helper_poll_enable(dev);
 	}

From 240e3c640cd13988ce90b913e3d8643385feb6fa Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:13 +0300
Subject: [PATCH 0449/1455] drm/i915/uncore: pass display to
 HAS_FPGA_DBG_UNCLAIMED()

Underneath, HAS_FPGA_DBG_UNCLAIMED() really expects struct
intel_display. Switch to it in preparation for removing the transitional
__to_intel_display() macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/e8cda779e192095dfa35d618f88bf7bfa526ac06.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/intel_uncore.c           | 3 ++-
 drivers/gpu/drm/i915/selftests/intel_uncore.c | 8 +++++---
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index c8e29fd72290..4ccba7c8ffb3 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -2502,6 +2502,7 @@ static int sanity_check_mmio_access(struct intel_uncore *uncore)
 int intel_uncore_init_mmio(struct intel_uncore *uncore)
 {
 	struct drm_i915_private *i915 = uncore->i915;
+	struct intel_display *display = i915->display;
 	int ret;
 
 	ret = sanity_check_mmio_access(uncore);
@@ -2536,7 +2537,7 @@ int intel_uncore_init_mmio(struct intel_uncore *uncore)
 	GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.read_fw_domains);
 	GEM_BUG_ON(intel_uncore_has_forcewake(uncore) != !!uncore->funcs.write_fw_domains);
 
-	if (HAS_FPGA_DBG_UNCLAIMED(i915))
+	if (HAS_FPGA_DBG_UNCLAIMED(display))
 		uncore->flags |= UNCORE_HAS_FPGA_DBG_UNCLAIMED;
 
 	if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915))
diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c
index 41eaa9b7f67d..58bcbdcef563 100644
--- a/drivers/gpu/drm/i915/selftests/intel_uncore.c
+++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c
@@ -277,13 +277,15 @@ static int live_forcewake_domains(void *arg)
 #define FW_RANGE 0x40000
 	struct intel_gt *gt = arg;
 	struct intel_uncore *uncore = gt->uncore;
+	struct drm_i915_private *i915 = gt->i915;
+	struct intel_display *display = i915->display;
 	unsigned long *valid;
 	u32 offset;
 	int err;
 
-	if (!HAS_FPGA_DBG_UNCLAIMED(gt->i915) &&
-	    !IS_VALLEYVIEW(gt->i915) &&
-	    !IS_CHERRYVIEW(gt->i915))
+	if (!HAS_FPGA_DBG_UNCLAIMED(display) &&
+	    !IS_VALLEYVIEW(i915) &&
+	    !IS_CHERRYVIEW(i915))
 		return 0;
 
 	/*

From 85cac9ff650dd6d20d7589c1577aeaa7ef62d944 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:14 +0300
Subject: [PATCH 0450/1455] drm/i915/gvt: convert mmio table to struct
 intel_display

Underneath, the register macros really expect a struct
intel_display. Switch to it in preparation for removing the transitional
__to_intel_display() macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/ef50b08c9bf6066201c86dc7e0f08f008c7c47d5.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/intel_gvt_mmio_table.c | 266 ++++++++++----------
 1 file changed, 134 insertions(+), 132 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c
index 87ac4446d306..ca57a3dd3148 100644
--- a/drivers/gpu/drm/i915/intel_gvt_mmio_table.c
+++ b/drivers/gpu/drm/i915/intel_gvt_mmio_table.c
@@ -62,6 +62,7 @@
 static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter)
 {
 	struct drm_i915_private *dev_priv = iter->i915;
+	struct intel_display *display = dev_priv->display;
 
 	MMIO_RING_D(RING_IMR);
 	MMIO_D(SDEIMR);
@@ -133,38 +134,38 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter)
 	MMIO_D(_MMIO(0x650b4));
 	MMIO_D(_MMIO(0xc4040));
 	MMIO_D(DERRMR);
-	MMIO_D(PIPEDSL(dev_priv, PIPE_A));
-	MMIO_D(PIPEDSL(dev_priv, PIPE_B));
-	MMIO_D(PIPEDSL(dev_priv, PIPE_C));
-	MMIO_D(PIPEDSL(dev_priv, _PIPE_EDP));
-	MMIO_D(TRANSCONF(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANSCONF(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANSCONF(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANSCONF(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPESTAT(dev_priv, PIPE_A));
-	MMIO_D(PIPESTAT(dev_priv, PIPE_B));
-	MMIO_D(PIPESTAT(dev_priv, PIPE_C));
-	MMIO_D(PIPESTAT(dev_priv, _PIPE_EDP));
-	MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, PIPE_A));
-	MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, PIPE_B));
-	MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, PIPE_C));
-	MMIO_D(PIPE_FLIPCOUNT_G4X(dev_priv, _PIPE_EDP));
-	MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, PIPE_A));
-	MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, PIPE_B));
-	MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, PIPE_C));
-	MMIO_D(PIPE_FRMCOUNT_G4X(dev_priv, _PIPE_EDP));
-	MMIO_D(CURCNTR(dev_priv, PIPE_A));
-	MMIO_D(CURCNTR(dev_priv, PIPE_B));
-	MMIO_D(CURCNTR(dev_priv, PIPE_C));
-	MMIO_D(CURPOS(dev_priv, PIPE_A));
-	MMIO_D(CURPOS(dev_priv, PIPE_B));
-	MMIO_D(CURPOS(dev_priv, PIPE_C));
-	MMIO_D(CURBASE(dev_priv, PIPE_A));
-	MMIO_D(CURBASE(dev_priv, PIPE_B));
-	MMIO_D(CURBASE(dev_priv, PIPE_C));
-	MMIO_D(CUR_FBC_CTL(dev_priv, PIPE_A));
-	MMIO_D(CUR_FBC_CTL(dev_priv, PIPE_B));
-	MMIO_D(CUR_FBC_CTL(dev_priv, PIPE_C));
+	MMIO_D(PIPEDSL(display, PIPE_A));
+	MMIO_D(PIPEDSL(display, PIPE_B));
+	MMIO_D(PIPEDSL(display, PIPE_C));
+	MMIO_D(PIPEDSL(display, _PIPE_EDP));
+	MMIO_D(TRANSCONF(display, TRANSCODER_A));
+	MMIO_D(TRANSCONF(display, TRANSCODER_B));
+	MMIO_D(TRANSCONF(display, TRANSCODER_C));
+	MMIO_D(TRANSCONF(display, TRANSCODER_EDP));
+	MMIO_D(PIPESTAT(display, PIPE_A));
+	MMIO_D(PIPESTAT(display, PIPE_B));
+	MMIO_D(PIPESTAT(display, PIPE_C));
+	MMIO_D(PIPESTAT(display, _PIPE_EDP));
+	MMIO_D(PIPE_FLIPCOUNT_G4X(display, PIPE_A));
+	MMIO_D(PIPE_FLIPCOUNT_G4X(display, PIPE_B));
+	MMIO_D(PIPE_FLIPCOUNT_G4X(display, PIPE_C));
+	MMIO_D(PIPE_FLIPCOUNT_G4X(display, _PIPE_EDP));
+	MMIO_D(PIPE_FRMCOUNT_G4X(display, PIPE_A));
+	MMIO_D(PIPE_FRMCOUNT_G4X(display, PIPE_B));
+	MMIO_D(PIPE_FRMCOUNT_G4X(display, PIPE_C));
+	MMIO_D(PIPE_FRMCOUNT_G4X(display, _PIPE_EDP));
+	MMIO_D(CURCNTR(display, PIPE_A));
+	MMIO_D(CURCNTR(display, PIPE_B));
+	MMIO_D(CURCNTR(display, PIPE_C));
+	MMIO_D(CURPOS(display, PIPE_A));
+	MMIO_D(CURPOS(display, PIPE_B));
+	MMIO_D(CURPOS(display, PIPE_C));
+	MMIO_D(CURBASE(display, PIPE_A));
+	MMIO_D(CURBASE(display, PIPE_B));
+	MMIO_D(CURBASE(display, PIPE_C));
+	MMIO_D(CUR_FBC_CTL(display, PIPE_A));
+	MMIO_D(CUR_FBC_CTL(display, PIPE_B));
+	MMIO_D(CUR_FBC_CTL(display, PIPE_C));
 	MMIO_D(_MMIO(0x700ac));
 	MMIO_D(_MMIO(0x710ac));
 	MMIO_D(_MMIO(0x720ac));
@@ -172,32 +173,32 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter)
 	MMIO_D(_MMIO(0x70094));
 	MMIO_D(_MMIO(0x70098));
 	MMIO_D(_MMIO(0x7009c));
-	MMIO_D(DSPCNTR(dev_priv, PIPE_A));
-	MMIO_D(DSPADDR(dev_priv, PIPE_A));
-	MMIO_D(DSPSTRIDE(dev_priv, PIPE_A));
-	MMIO_D(DSPPOS(dev_priv, PIPE_A));
-	MMIO_D(DSPSIZE(dev_priv, PIPE_A));
-	MMIO_D(DSPSURF(dev_priv, PIPE_A));
-	MMIO_D(DSPOFFSET(dev_priv, PIPE_A));
-	MMIO_D(DSPSURFLIVE(dev_priv, PIPE_A));
+	MMIO_D(DSPCNTR(display, PIPE_A));
+	MMIO_D(DSPADDR(display, PIPE_A));
+	MMIO_D(DSPSTRIDE(display, PIPE_A));
+	MMIO_D(DSPPOS(display, PIPE_A));
+	MMIO_D(DSPSIZE(display, PIPE_A));
+	MMIO_D(DSPSURF(display, PIPE_A));
+	MMIO_D(DSPOFFSET(display, PIPE_A));
+	MMIO_D(DSPSURFLIVE(display, PIPE_A));
 	MMIO_D(REG_50080(PIPE_A, PLANE_PRIMARY));
-	MMIO_D(DSPCNTR(dev_priv, PIPE_B));
-	MMIO_D(DSPADDR(dev_priv, PIPE_B));
-	MMIO_D(DSPSTRIDE(dev_priv, PIPE_B));
-	MMIO_D(DSPPOS(dev_priv, PIPE_B));
-	MMIO_D(DSPSIZE(dev_priv, PIPE_B));
-	MMIO_D(DSPSURF(dev_priv, PIPE_B));
-	MMIO_D(DSPOFFSET(dev_priv, PIPE_B));
-	MMIO_D(DSPSURFLIVE(dev_priv, PIPE_B));
+	MMIO_D(DSPCNTR(display, PIPE_B));
+	MMIO_D(DSPADDR(display, PIPE_B));
+	MMIO_D(DSPSTRIDE(display, PIPE_B));
+	MMIO_D(DSPPOS(display, PIPE_B));
+	MMIO_D(DSPSIZE(display, PIPE_B));
+	MMIO_D(DSPSURF(display, PIPE_B));
+	MMIO_D(DSPOFFSET(display, PIPE_B));
+	MMIO_D(DSPSURFLIVE(display, PIPE_B));
 	MMIO_D(REG_50080(PIPE_B, PLANE_PRIMARY));
-	MMIO_D(DSPCNTR(dev_priv, PIPE_C));
-	MMIO_D(DSPADDR(dev_priv, PIPE_C));
-	MMIO_D(DSPSTRIDE(dev_priv, PIPE_C));
-	MMIO_D(DSPPOS(dev_priv, PIPE_C));
-	MMIO_D(DSPSIZE(dev_priv, PIPE_C));
-	MMIO_D(DSPSURF(dev_priv, PIPE_C));
-	MMIO_D(DSPOFFSET(dev_priv, PIPE_C));
-	MMIO_D(DSPSURFLIVE(dev_priv, PIPE_C));
+	MMIO_D(DSPCNTR(display, PIPE_C));
+	MMIO_D(DSPADDR(display, PIPE_C));
+	MMIO_D(DSPSTRIDE(display, PIPE_C));
+	MMIO_D(DSPPOS(display, PIPE_C));
+	MMIO_D(DSPSIZE(display, PIPE_C));
+	MMIO_D(DSPSURF(display, PIPE_C));
+	MMIO_D(DSPOFFSET(display, PIPE_C));
+	MMIO_D(DSPSURFLIVE(display, PIPE_C));
 	MMIO_D(REG_50080(PIPE_C, PLANE_PRIMARY));
 	MMIO_D(SPRCTL(PIPE_A));
 	MMIO_D(SPRLINOFF(PIPE_A));
@@ -238,73 +239,73 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter)
 	MMIO_D(SPRSCALE(PIPE_C));
 	MMIO_D(SPRSURFLIVE(PIPE_C));
 	MMIO_D(REG_50080(PIPE_C, PLANE_SPRITE0));
-	MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_A));
-	MMIO_D(BCLRPAT(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPESRC(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_B));
-	MMIO_D(BCLRPAT(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPESRC(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_C));
-	MMIO_D(BCLRPAT(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPESRC(dev_priv, TRANSCODER_C));
-	MMIO_D(TRANS_HTOTAL(dev_priv, TRANSCODER_EDP));
-	MMIO_D(TRANS_HBLANK(dev_priv, TRANSCODER_EDP));
-	MMIO_D(TRANS_HSYNC(dev_priv, TRANSCODER_EDP));
-	MMIO_D(TRANS_VTOTAL(dev_priv, TRANSCODER_EDP));
-	MMIO_D(TRANS_VBLANK(dev_priv, TRANSCODER_EDP));
-	MMIO_D(TRANS_VSYNC(dev_priv, TRANSCODER_EDP));
-	MMIO_D(BCLRPAT(dev_priv, TRANSCODER_EDP));
-	MMIO_D(TRANS_VSYNCSHIFT(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_A));
-	MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_B));
-	MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_C));
-	MMIO_D(PIPE_DATA_M1(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_DATA_N1(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_DATA_M2(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_DATA_N2(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_LINK_M1(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_LINK_N1(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_LINK_M2(dev_priv, TRANSCODER_EDP));
-	MMIO_D(PIPE_LINK_N2(dev_priv, TRANSCODER_EDP));
+	MMIO_D(TRANS_HTOTAL(display, TRANSCODER_A));
+	MMIO_D(TRANS_HBLANK(display, TRANSCODER_A));
+	MMIO_D(TRANS_HSYNC(display, TRANSCODER_A));
+	MMIO_D(TRANS_VTOTAL(display, TRANSCODER_A));
+	MMIO_D(TRANS_VBLANK(display, TRANSCODER_A));
+	MMIO_D(TRANS_VSYNC(display, TRANSCODER_A));
+	MMIO_D(BCLRPAT(display, TRANSCODER_A));
+	MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_A));
+	MMIO_D(PIPESRC(display, TRANSCODER_A));
+	MMIO_D(TRANS_HTOTAL(display, TRANSCODER_B));
+	MMIO_D(TRANS_HBLANK(display, TRANSCODER_B));
+	MMIO_D(TRANS_HSYNC(display, TRANSCODER_B));
+	MMIO_D(TRANS_VTOTAL(display, TRANSCODER_B));
+	MMIO_D(TRANS_VBLANK(display, TRANSCODER_B));
+	MMIO_D(TRANS_VSYNC(display, TRANSCODER_B));
+	MMIO_D(BCLRPAT(display, TRANSCODER_B));
+	MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_B));
+	MMIO_D(PIPESRC(display, TRANSCODER_B));
+	MMIO_D(TRANS_HTOTAL(display, TRANSCODER_C));
+	MMIO_D(TRANS_HBLANK(display, TRANSCODER_C));
+	MMIO_D(TRANS_HSYNC(display, TRANSCODER_C));
+	MMIO_D(TRANS_VTOTAL(display, TRANSCODER_C));
+	MMIO_D(TRANS_VBLANK(display, TRANSCODER_C));
+	MMIO_D(TRANS_VSYNC(display, TRANSCODER_C));
+	MMIO_D(BCLRPAT(display, TRANSCODER_C));
+	MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_C));
+	MMIO_D(PIPESRC(display, TRANSCODER_C));
+	MMIO_D(TRANS_HTOTAL(display, TRANSCODER_EDP));
+	MMIO_D(TRANS_HBLANK(display, TRANSCODER_EDP));
+	MMIO_D(TRANS_HSYNC(display, TRANSCODER_EDP));
+	MMIO_D(TRANS_VTOTAL(display, TRANSCODER_EDP));
+	MMIO_D(TRANS_VBLANK(display, TRANSCODER_EDP));
+	MMIO_D(TRANS_VSYNC(display, TRANSCODER_EDP));
+	MMIO_D(BCLRPAT(display, TRANSCODER_EDP));
+	MMIO_D(TRANS_VSYNCSHIFT(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_DATA_M1(display, TRANSCODER_A));
+	MMIO_D(PIPE_DATA_N1(display, TRANSCODER_A));
+	MMIO_D(PIPE_DATA_M2(display, TRANSCODER_A));
+	MMIO_D(PIPE_DATA_N2(display, TRANSCODER_A));
+	MMIO_D(PIPE_LINK_M1(display, TRANSCODER_A));
+	MMIO_D(PIPE_LINK_N1(display, TRANSCODER_A));
+	MMIO_D(PIPE_LINK_M2(display, TRANSCODER_A));
+	MMIO_D(PIPE_LINK_N2(display, TRANSCODER_A));
+	MMIO_D(PIPE_DATA_M1(display, TRANSCODER_B));
+	MMIO_D(PIPE_DATA_N1(display, TRANSCODER_B));
+	MMIO_D(PIPE_DATA_M2(display, TRANSCODER_B));
+	MMIO_D(PIPE_DATA_N2(display, TRANSCODER_B));
+	MMIO_D(PIPE_LINK_M1(display, TRANSCODER_B));
+	MMIO_D(PIPE_LINK_N1(display, TRANSCODER_B));
+	MMIO_D(PIPE_LINK_M2(display, TRANSCODER_B));
+	MMIO_D(PIPE_LINK_N2(display, TRANSCODER_B));
+	MMIO_D(PIPE_DATA_M1(display, TRANSCODER_C));
+	MMIO_D(PIPE_DATA_N1(display, TRANSCODER_C));
+	MMIO_D(PIPE_DATA_M2(display, TRANSCODER_C));
+	MMIO_D(PIPE_DATA_N2(display, TRANSCODER_C));
+	MMIO_D(PIPE_LINK_M1(display, TRANSCODER_C));
+	MMIO_D(PIPE_LINK_N1(display, TRANSCODER_C));
+	MMIO_D(PIPE_LINK_M2(display, TRANSCODER_C));
+	MMIO_D(PIPE_LINK_N2(display, TRANSCODER_C));
+	MMIO_D(PIPE_DATA_M1(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_DATA_N1(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_DATA_M2(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_DATA_N2(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_LINK_M1(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_LINK_N1(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_LINK_M2(display, TRANSCODER_EDP));
+	MMIO_D(PIPE_LINK_N2(display, TRANSCODER_EDP));
 	MMIO_D(PF_CTL(PIPE_A));
 	MMIO_D(PF_WIN_SZ(PIPE_A));
 	MMIO_D(PF_WIN_POS(PIPE_A));
@@ -513,12 +514,12 @@ static int iterate_generic_mmio(struct intel_gvt_mmio_table_iter *iter)
 	MMIO_D(GAMMA_MODE(PIPE_A));
 	MMIO_D(GAMMA_MODE(PIPE_B));
 	MMIO_D(GAMMA_MODE(PIPE_C));
-	MMIO_D(TRANS_MULT(dev_priv, TRANSCODER_A));
-	MMIO_D(TRANS_MULT(dev_priv, TRANSCODER_B));
-	MMIO_D(TRANS_MULT(dev_priv, TRANSCODER_C));
-	MMIO_D(HSW_TVIDEO_DIP_CTL(dev_priv, TRANSCODER_A));
-	MMIO_D(HSW_TVIDEO_DIP_CTL(dev_priv, TRANSCODER_B));
-	MMIO_D(HSW_TVIDEO_DIP_CTL(dev_priv, TRANSCODER_C));
+	MMIO_D(TRANS_MULT(display, TRANSCODER_A));
+	MMIO_D(TRANS_MULT(display, TRANSCODER_B));
+	MMIO_D(TRANS_MULT(display, TRANSCODER_C));
+	MMIO_D(HSW_TVIDEO_DIP_CTL(display, TRANSCODER_A));
+	MMIO_D(HSW_TVIDEO_DIP_CTL(display, TRANSCODER_B));
+	MMIO_D(HSW_TVIDEO_DIP_CTL(display, TRANSCODER_C));
 	MMIO_D(SFUSE_STRAP);
 	MMIO_D(SBI_ADDR);
 	MMIO_D(SBI_DATA);
@@ -1111,6 +1112,7 @@ static int iterate_skl_plus_mmio(struct intel_gvt_mmio_table_iter *iter)
 static int iterate_bxt_mmio(struct intel_gvt_mmio_table_iter *iter)
 {
 	struct drm_i915_private *dev_priv = iter->i915;
+	struct intel_display *display = dev_priv->display;
 
 	MMIO_F(_MMIO(0x80000), 0x3000);
 	MMIO_D(GEN7_SAMPLER_INSTDONE);
@@ -1242,9 +1244,9 @@ static int iterate_bxt_mmio(struct intel_gvt_mmio_table_iter *iter)
 	MMIO_D(BXT_DSI_PLL_ENABLE);
 	MMIO_D(GEN9_CLKGATE_DIS_0);
 	MMIO_D(GEN9_CLKGATE_DIS_4);
-	MMIO_D(HSW_TVIDEO_DIP_GCP(dev_priv, TRANSCODER_A));
-	MMIO_D(HSW_TVIDEO_DIP_GCP(dev_priv, TRANSCODER_B));
-	MMIO_D(HSW_TVIDEO_DIP_GCP(dev_priv, TRANSCODER_C));
+	MMIO_D(HSW_TVIDEO_DIP_GCP(display, TRANSCODER_A));
+	MMIO_D(HSW_TVIDEO_DIP_GCP(display, TRANSCODER_B));
+	MMIO_D(HSW_TVIDEO_DIP_GCP(display, TRANSCODER_C));
 	MMIO_D(RC6_CTX_BASE);
 	MMIO_D(GEN8_PUSHBUS_CONTROL);
 	MMIO_D(GEN8_PUSHBUS_ENABLE);

From 7e757092373ccf147e23f46d1167718680db8ec6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:15 +0300
Subject: [PATCH 0451/1455] drm/i915/reg: separate VLV_DSPCLK_GATE_D from
 DSPCLK_GATE_D

All the places that use DSPCLK_GATE_D are specific to certain platforms,
and the parametrization of it to support VLV/CHV MMIO display base isn't
really buying us anything. Add a separate macro for VLV_DSPCLK_GATE_D
and use it.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/ac16d9d5192595944bf9bcf70aa721b504bc90c0.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_power_well.c | 2 +-
 drivers/gpu/drm/i915/display/intel_gmbus.c              | 2 +-
 drivers/gpu/drm/i915/display/intel_overlay.c            | 5 ++---
 drivers/gpu/drm/i915/display/vlv_dsi.c                  | 4 ++--
 drivers/gpu/drm/i915/i915_reg.h                         | 3 ++-
 drivers/gpu/drm/i915/intel_clock_gating.c               | 4 ++--
 6 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 6efe5524cbbf..31c2a07bb188 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -1209,7 +1209,7 @@ static void vlv_init_display_clock_gating(struct intel_display *display)
 	 * (and never recovering) in this case. intel_dsi_post_disable() will
 	 * clear it when we turn off the display.
 	 */
-	intel_de_rmw(display, DSPCLK_GATE_D(display),
+	intel_de_rmw(display, VLV_DSPCLK_GATE_D,
 		     ~DPOUNIT_CLOCK_GATE_DISABLE, VRHUNIT_CLOCK_GATE_DISABLE);
 
 	/*
diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index 6a74805570e1..063335053d13 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -219,7 +219,7 @@ static void pnv_gmbus_clock_gating(struct intel_display *display,
 				   bool enable)
 {
 	/* When using bit bashing for I2C, this bit needs to be set to 1 */
-	intel_de_rmw(display, DSPCLK_GATE_D(display),
+	intel_de_rmw(display, DSPCLK_GATE_D,
 		     PNV_GMBUSUNIT_CLOCK_GATE_DISABLE,
 		     !enable ? PNV_GMBUSUNIT_CLOCK_GATE_DISABLE : 0);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c
index 159a5f998ea0..272f9e7af4d4 100644
--- a/drivers/gpu/drm/i915/display/intel_overlay.c
+++ b/drivers/gpu/drm/i915/display/intel_overlay.c
@@ -217,10 +217,9 @@ static void i830_overlay_clock_gating(struct intel_display *display,
 
 	/* WA_OVERLAY_CLKGATE:alm */
 	if (enable)
-		intel_de_write(display, DSPCLK_GATE_D(display), 0);
+		intel_de_write(display, DSPCLK_GATE_D, 0);
 	else
-		intel_de_write(display, DSPCLK_GATE_D(display),
-			       OVRUNIT_CLOCK_GATE_DISABLE);
+		intel_de_write(display, DSPCLK_GATE_D, OVRUNIT_CLOCK_GATE_DISABLE);
 
 	/* WA_DISABLE_L2CACHE_CLOCK_GATING:alm */
 	pci_bus_read_config_byte(pdev->bus,
diff --git a/drivers/gpu/drm/i915/display/vlv_dsi.c b/drivers/gpu/drm/i915/display/vlv_dsi.c
index 6d9f3312de7e..c9a53fde79c4 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi.c
@@ -761,7 +761,7 @@ static void intel_dsi_pre_enable(struct intel_atomic_state *state,
 
 	if (display->platform.valleyview || display->platform.cherryview) {
 		/* Disable DPOunit clock gating, can stall pipe */
-		intel_de_rmw(display, DSPCLK_GATE_D(display),
+		intel_de_rmw(display, VLV_DSPCLK_GATE_D,
 			     0, DPOUNIT_CLOCK_GATE_DISABLE);
 	}
 
@@ -918,7 +918,7 @@ static void intel_dsi_post_disable(struct intel_atomic_state *state,
 	} else {
 		vlv_dsi_pll_disable(encoder);
 
-		intel_de_rmw(display, DSPCLK_GATE_D(display),
+		intel_de_rmw(display, VLV_DSPCLK_GATE_D,
 			     DPOUNIT_CLOCK_GATE_DISABLE, 0);
 	}
 
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 03b895897f60..b283b25d8368 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -613,7 +613,8 @@
 #define  DSTATE_GFX_CLOCK_GATING		(1 << 1)
 #define  DSTATE_DOT_CLOCK_GATING		(1 << 0)
 
-#define DSPCLK_GATE_D(__i915)		_MMIO(DISPLAY_MMIO_BASE(__i915) + 0x6200)
+#define DSPCLK_GATE_D			_MMIO(0x6200)
+#define VLV_DSPCLK_GATE_D		_MMIO(VLV_DISPLAY_BASE + 0x6200)
 # define DPUNIT_B_CLOCK_GATE_DISABLE		(1 << 30) /* 965 */
 # define VSUNIT_CLOCK_GATE_DISABLE		(1 << 29) /* 965 */
 # define VRHUNIT_CLOCK_GATE_DISABLE		(1 << 28) /* 965 */
diff --git a/drivers/gpu/drm/i915/intel_clock_gating.c b/drivers/gpu/drm/i915/intel_clock_gating.c
index e501f4937510..467740969431 100644
--- a/drivers/gpu/drm/i915/intel_clock_gating.c
+++ b/drivers/gpu/drm/i915/intel_clock_gating.c
@@ -620,7 +620,7 @@ static void g4x_init_clock_gating(struct drm_i915_private *i915)
 		OVCUNIT_CLOCK_GATE_DISABLE;
 	if (IS_GM45(i915))
 		dspclk_gate |= DSSUNIT_CLOCK_GATE_DISABLE;
-	intel_uncore_write(&i915->uncore, DSPCLK_GATE_D(i915), dspclk_gate);
+	intel_uncore_write(&i915->uncore, DSPCLK_GATE_D, dspclk_gate);
 
 	g4x_disable_trickle_feed(i915);
 }
@@ -631,7 +631,7 @@ static void i965gm_init_clock_gating(struct drm_i915_private *i915)
 
 	intel_uncore_write(uncore, RENCLK_GATE_D1, I965_RCC_CLOCK_GATE_DISABLE);
 	intel_uncore_write(uncore, RENCLK_GATE_D2, 0);
-	intel_uncore_write(uncore, DSPCLK_GATE_D(i915), 0);
+	intel_uncore_write(uncore, DSPCLK_GATE_D, 0);
 	intel_uncore_write(uncore, RAMCLK_GATE_D, 0);
 	intel_uncore_write16(uncore, DEUC, 0);
 	intel_uncore_write(uncore,

From ada8faab6395fa1df9febda8f8ff17576aa177c2 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 6 Aug 2025 19:55:16 +0300
Subject: [PATCH 0452/1455] drm/i915/display: drop __to_intel_display() usage

All the places that use __to_intel_display() now get passed a struct
intel_display pointer, and the transitional adaptation macro has served
its purpose. Remove the macro.

Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/d2c4a5aab3d69cec6f0640eed157a97531bd72f2.1754499175.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../gpu/drm/i915/display/intel_display_conversion.c  |  2 +-
 .../gpu/drm/i915/display/intel_display_conversion.h  | 12 ------------
 drivers/gpu/drm/i915/display/intel_display_device.h  |  7 +++----
 drivers/gpu/drm/i915/display/intel_pch.h             |  4 +---
 4 files changed, 5 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_conversion.c b/drivers/gpu/drm/i915/display/intel_display_conversion.c
index 4d565935e2cc..d56065f22655 100644
--- a/drivers/gpu/drm/i915/display/intel_display_conversion.c
+++ b/drivers/gpu/drm/i915/display/intel_display_conversion.c
@@ -4,7 +4,7 @@
 #include "i915_drv.h"
 #include "intel_display_conversion.h"
 
-struct intel_display *__i915_to_display(struct drm_i915_private *i915)
+static struct intel_display *__i915_to_display(struct drm_i915_private *i915)
 {
 	return i915->display;
 }
diff --git a/drivers/gpu/drm/i915/display/intel_display_conversion.h b/drivers/gpu/drm/i915/display/intel_display_conversion.h
index 46c7208d42ba..d497bc58a73f 100644
--- a/drivers/gpu/drm/i915/display/intel_display_conversion.h
+++ b/drivers/gpu/drm/i915/display/intel_display_conversion.h
@@ -9,20 +9,8 @@
 #define __INTEL_DISPLAY_CONVERSION__
 
 struct drm_device;
-struct drm_i915_private;
 struct intel_display;
 
-struct intel_display *__i915_to_display(struct drm_i915_private *i915);
 struct intel_display *__drm_to_display(struct drm_device *drm);
-/*
- * Transitional macro to optionally convert struct drm_i915_private * to struct
- * intel_display *, also accepting the latter.
- */
-#define __to_intel_display(p)						\
-	_Generic(p,							\
-		 const struct drm_i915_private *: __i915_to_display((struct drm_i915_private *)(p)), \
-		 struct drm_i915_private *: __i915_to_display((struct drm_i915_private *)(p)), \
-		 const struct intel_display *: (p),			\
-		 struct intel_display *: (p))
 
 #endif /* __INTEL_DISPLAY_CONVERSION__ */
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 4308822f0415..6e87b763fe7c 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -9,7 +9,6 @@
 #include <linux/bitops.h>
 #include <linux/types.h>
 
-#include "intel_display_conversion.h"
 #include "intel_display_limits.h"
 
 struct drm_printer;
@@ -224,8 +223,8 @@ struct intel_display_platforms {
 	(IS_DISPLAY_VERx100((__display), (ipver), (ipver)) && \
 	 IS_DISPLAY_STEP((__display), (from), (until)))
 
-#define DISPLAY_INFO(__display)		(__to_intel_display(__display)->info.__device_info)
-#define DISPLAY_RUNTIME_INFO(__display)	(&__to_intel_display(__display)->info.__runtime_info)
+#define DISPLAY_INFO(__display)		((__display)->info.__device_info)
+#define DISPLAY_RUNTIME_INFO(__display)	(&(__display)->info.__runtime_info)
 
 #define DISPLAY_VER(__display)		(DISPLAY_RUNTIME_INFO(__display)->ip.ver)
 #define DISPLAY_VERx100(__display)	(DISPLAY_RUNTIME_INFO(__display)->ip.ver * 100 + \
@@ -236,7 +235,7 @@ struct intel_display_platforms {
 #define INTEL_DISPLAY_STEP(__display)	(DISPLAY_RUNTIME_INFO(__display)->step)
 
 #define IS_DISPLAY_STEP(__display, since, until) \
-	(drm_WARN_ON(__to_intel_display(__display)->drm, INTEL_DISPLAY_STEP(__display) == STEP_NONE), \
+	(drm_WARN_ON((__display)->drm, INTEL_DISPLAY_STEP(__display) == STEP_NONE), \
 	 INTEL_DISPLAY_STEP(__display) >= (since) && INTEL_DISPLAY_STEP(__display) < (until))
 
 #define ARLS_HOST_BRIDGE_PCI_ID1 0x7D1C
diff --git a/drivers/gpu/drm/i915/display/intel_pch.h b/drivers/gpu/drm/i915/display/intel_pch.h
index cf4dab1b98bf..19cac7412d0a 100644
--- a/drivers/gpu/drm/i915/display/intel_pch.h
+++ b/drivers/gpu/drm/i915/display/intel_pch.h
@@ -6,8 +6,6 @@
 #ifndef __INTEL_PCH__
 #define __INTEL_PCH__
 
-#include "intel_display_conversion.h"
-
 struct intel_display;
 
 /*
@@ -36,7 +34,7 @@ enum intel_pch {
 	PCH_LNL,
 };
 
-#define INTEL_PCH_TYPE(_display)		(__to_intel_display(_display)->pch_type)
+#define INTEL_PCH_TYPE(_display)		((_display)->pch_type)
 #define HAS_PCH_DG2(display)			(INTEL_PCH_TYPE(display) == PCH_DG2)
 #define HAS_PCH_ADP(display)			(INTEL_PCH_TYPE(display) == PCH_ADP)
 #define HAS_PCH_DG1(display)			(INTEL_PCH_TYPE(display) == PCH_DG1)

From ffc23a204a5f2e763a8cc8a8cfefe0027a6f0ec3 Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Wed, 30 Jul 2025 21:23:42 -0600
Subject: [PATCH 0453/1455] drm: Add MIPI read_multi func and two write macros

Create mipi_dsi_dcs_read_multi(), which accepts a mipi_dsi_multi_context
struct for improved error handling and cleaner panel driver code.

Create mipi_dsi_dcs_write_var_seq_multi() and
mipi_dsi_generic_write_var_seq_multi() macros which allow MIPI panel
drivers to write non-constant data to display controllers.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731032343.1258366-3-me@brighamcampbell.com
---
 drivers/gpu/drm/drm_mipi_dsi.c | 37 ++++++++++++++++++++++++++++++++++
 include/drm/drm_mipi_dsi.h     | 35 ++++++++++++++++++++++++++++++++
 2 files changed, 72 insertions(+)

diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index fe03915d0095..a712e177b350 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -1096,6 +1096,43 @@ ssize_t mipi_dsi_dcs_read(struct mipi_dsi_device *dsi, u8 cmd, void *data,
 }
 EXPORT_SYMBOL(mipi_dsi_dcs_read);
 
+/**
+ * mipi_dsi_dcs_read_multi() - mipi_dsi_dcs_read() w/ accum_err
+ * @ctx: Context for multiple DSI transactions
+ * @cmd: DCS command
+ * @data: buffer in which to receive data
+ * @len: size of receive buffer
+ *
+ * Like mipi_dsi_dcs_read() but deals with errors in a way that makes it
+ * convenient to make several calls in a row.
+ */
+void mipi_dsi_dcs_read_multi(struct mipi_dsi_multi_context *ctx, u8 cmd,
+			     void *data, size_t len)
+{
+	struct mipi_dsi_device *dsi = ctx->dsi;
+	struct device *dev = &dsi->dev;
+	struct mipi_dsi_msg msg = {
+		.channel = dsi->channel,
+		.type = MIPI_DSI_DCS_READ,
+		.tx_buf = &cmd,
+		.tx_len = 1,
+		.rx_buf = data,
+		.rx_len = len
+	};
+	ssize_t ret;
+
+	if (ctx->accum_err)
+		return;
+
+	ret = mipi_dsi_device_transfer(dsi, &msg);
+	if (ret < 0) {
+		ctx->accum_err = ret;
+		dev_err(dev, "dcs read with command %#x failed: %d\n", cmd,
+			ctx->accum_err);
+	}
+}
+EXPORT_SYMBOL(mipi_dsi_dcs_read_multi);
+
 /**
  * mipi_dsi_dcs_nop() - send DCS nop packet
  * @dsi: DSI peripheral device
diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h
index 16eeb9552064..3aba7b380c8d 100644
--- a/include/drm/drm_mipi_dsi.h
+++ b/include/drm/drm_mipi_dsi.h
@@ -342,6 +342,8 @@ ssize_t mipi_dsi_dcs_write(struct mipi_dsi_device *dsi, u8 cmd,
 			   const void *data, size_t len);
 ssize_t mipi_dsi_dcs_read(struct mipi_dsi_device *dsi, u8 cmd, void *data,
 			  size_t len);
+void mipi_dsi_dcs_read_multi(struct mipi_dsi_multi_context *ctx, u8 cmd,
+			     void *data, size_t len);
 int mipi_dsi_dcs_nop(struct mipi_dsi_device *dsi);
 int mipi_dsi_dcs_soft_reset(struct mipi_dsi_device *dsi);
 int mipi_dsi_dcs_get_power_mode(struct mipi_dsi_device *dsi, u8 *mode);
@@ -403,6 +405,22 @@ void mipi_dsi_dcs_set_tear_off_multi(struct mipi_dsi_multi_context *ctx);
 		mipi_dsi_generic_write_multi(ctx, d, ARRAY_SIZE(d)); \
 	} while (0)
 
+/**
+ * mipi_dsi_generic_write_var_seq_multi - transmit non-constant data using a
+ * generic write packet
+ *
+ * This macro will print errors for you and error handling is optimized for
+ * callers that call this multiple times in a row.
+ *
+ * @ctx: Context for multiple DSI transactions
+ * @seq: buffer containing the payload
+ */
+#define mipi_dsi_generic_write_var_seq_multi(ctx, seq...)	     \
+	do {							     \
+		const u8 d[] = { seq };				     \
+		mipi_dsi_generic_write_multi(ctx, d, ARRAY_SIZE(d)); \
+	} while (0)
+
 /**
  * mipi_dsi_dcs_write_seq_multi - transmit a DCS command with payload
  *
@@ -419,6 +437,23 @@ void mipi_dsi_dcs_set_tear_off_multi(struct mipi_dsi_multi_context *ctx);
 		mipi_dsi_dcs_write_buffer_multi(ctx, d, ARRAY_SIZE(d)); \
 	} while (0)
 
+/**
+ * mipi_dsi_dcs_write_var_seq_multi - transmit a DCS command with non-constant
+ * payload
+ *
+ * This macro will print errors for you and error handling is optimized for
+ * callers that call this multiple times in a row.
+ *
+ * @ctx: Context for multiple DSI transactions
+ * @cmd: Command
+ * @seq: buffer containing data to be transmitted
+ */
+#define mipi_dsi_dcs_write_var_seq_multi(ctx, cmd, seq...)		\
+	do {								\
+		const u8 d[] = { cmd, seq };				\
+		mipi_dsi_dcs_write_buffer_multi(ctx, d, ARRAY_SIZE(d));	\
+	} while (0)
+
 /**
  * mipi_dsi_dual - send the same MIPI DSI command to two interfaces
  *

From 5fbc0dbb92d639d10a5386ad0ae87a54680a04be Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Wed, 30 Jul 2025 21:23:43 -0600
Subject: [PATCH 0454/1455] drm/panel: novatek-nt35560: Clean up driver

Update driver to use the "multi" variants of MIPI functions which
facilitate improved error handling and cleaner driver code.

Remove information from a comment which was made obsolete by commit
994ea402c767 ("drm/panel: Rename Sony ACX424 to Novatek NT35560"), which
determined that this driver supports the Novatek NT35560 panel
controller.

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250731032343.1258366-4-me@brighamcampbell.com
---
 drivers/gpu/drm/panel/panel-novatek-nt35560.c | 198 ++++++------------
 1 file changed, 60 insertions(+), 138 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-novatek-nt35560.c b/drivers/gpu/drm/panel/panel-novatek-nt35560.c
index 17898a29efe8..561e6643dcbb 100644
--- a/drivers/gpu/drm/panel/panel-novatek-nt35560.c
+++ b/drivers/gpu/drm/panel/panel-novatek-nt35560.c
@@ -148,24 +148,20 @@ static inline struct nt35560 *panel_to_nt35560(struct drm_panel *panel)
 static int nt35560_set_brightness(struct backlight_device *bl)
 {
 	struct nt35560 *nt = bl_get_data(bl);
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev);
-	int period_ns = 1023;
+	struct mipi_dsi_multi_context dsi_ctx = {
+		.dsi = to_mipi_dsi_device(nt->dev)
+	};
 	int duty_ns = bl->props.brightness;
+	int period_ns = 1023;
 	u8 pwm_ratio;
 	u8 pwm_div;
-	u8 par;
-	int ret;
 
 	if (backlight_is_blank(bl)) {
 		/* Disable backlight */
-		par = 0x00;
-		ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY,
-					 &par, 1);
-		if (ret < 0) {
-			dev_err(nt->dev, "failed to disable display backlight (%d)\n", ret);
-			return ret;
-		}
-		return 0;
+		mipi_dsi_dcs_write_seq_multi(&dsi_ctx,
+					     MIPI_DCS_WRITE_CONTROL_DISPLAY,
+					     0x00);
+		return dsi_ctx.accum_err;
 	}
 
 	/* Calculate the PWM duty cycle in n/256's */
@@ -176,12 +172,6 @@ static int nt35560_set_brightness(struct backlight_device *bl)
 
 	/* Set up PWM dutycycle ONE byte (differs from the standard) */
 	dev_dbg(nt->dev, "calculated duty cycle %02x\n", pwm_ratio);
-	ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_SET_DISPLAY_BRIGHTNESS,
-				 &pwm_ratio, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to set display PWM ratio (%d)\n", ret);
-		return ret;
-	}
 
 	/*
 	 * Sequence to write PWMDIV:
@@ -192,46 +182,23 @@ static int nt35560_set_brightness(struct backlight_device *bl)
 	 *	0x22		PWMDIV
 	 *	0x7F		0xAA   CMD2 page 1 lock
 	 */
-	par = 0xaa;
-	ret = mipi_dsi_dcs_write(dsi, 0xf3, &par, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to unlock CMD 2 (%d)\n", ret);
-		return ret;
-	}
-	par = 0x01;
-	ret = mipi_dsi_dcs_write(dsi, 0x00, &par, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to enter page 1 (%d)\n", ret);
-		return ret;
-	}
-	par = 0x01;
-	ret = mipi_dsi_dcs_write(dsi, 0x7d, &par, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to disable MTP reload (%d)\n", ret);
-		return ret;
-	}
-	ret = mipi_dsi_dcs_write(dsi, 0x22, &pwm_div, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to set PWM divisor (%d)\n", ret);
-		return ret;
-	}
-	par = 0xaa;
-	ret = mipi_dsi_dcs_write(dsi, 0x7f, &par, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to lock CMD 2 (%d)\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_write_var_seq_multi(&dsi_ctx,
+					 MIPI_DCS_SET_DISPLAY_BRIGHTNESS,
+					 pwm_ratio);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0xf3, 0xaa);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x00, 0x01);
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x7d, 0x01);
+
+	mipi_dsi_dcs_write_var_seq_multi(&dsi_ctx, 0x22, pwm_div);
+
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, 0x7f, 0xaa);
 
 	/* Enable backlight */
-	par = 0x24;
-	ret = mipi_dsi_dcs_write(dsi, MIPI_DCS_WRITE_CONTROL_DISPLAY,
-				 &par, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to enable display backlight (%d)\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, MIPI_DCS_WRITE_CONTROL_DISPLAY,
+				     0x24);
 
-	return 0;
+	return dsi_ctx.accum_err;
 }
 
 static const struct backlight_ops nt35560_bl_ops = {
@@ -244,32 +211,23 @@ static const struct backlight_properties nt35560_bl_props = {
 	.max_brightness = 1023,
 };
 
-static int nt35560_read_id(struct nt35560 *nt)
+static void nt35560_read_id(struct mipi_dsi_multi_context *dsi_ctx)
 {
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev);
+	struct device dev = dsi_ctx->dsi->dev;
 	u8 vendor, version, panel;
 	u16 val;
-	int ret;
 
-	ret = mipi_dsi_dcs_read(dsi, NT35560_DCS_READ_ID1, &vendor, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "could not vendor ID byte\n");
-		return ret;
-	}
-	ret = mipi_dsi_dcs_read(dsi, NT35560_DCS_READ_ID2, &version, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "could not read device version byte\n");
-		return ret;
-	}
-	ret = mipi_dsi_dcs_read(dsi, NT35560_DCS_READ_ID3, &panel, 1);
-	if (ret < 0) {
-		dev_err(nt->dev, "could not read panel ID byte\n");
-		return ret;
-	}
+	mipi_dsi_dcs_read_multi(dsi_ctx, NT35560_DCS_READ_ID1, &vendor, 1);
+	mipi_dsi_dcs_read_multi(dsi_ctx, NT35560_DCS_READ_ID2, &version, 1);
+	mipi_dsi_dcs_read_multi(dsi_ctx, NT35560_DCS_READ_ID3, &panel, 1);
+
+	if (dsi_ctx->accum_err < 0)
+		return;
 
 	if (vendor == 0x00) {
-		dev_err(nt->dev, "device vendor ID is zero\n");
-		return -ENODEV;
+		dev_err(&dev, "device vendor ID is zero\n");
+		dsi_ctx->accum_err = -ENODEV;
+		return;
 	}
 
 	val = (vendor << 8) | panel;
@@ -278,16 +236,16 @@ static int nt35560_read_id(struct nt35560 *nt)
 	case DISPLAY_SONY_ACX424AKP_ID2:
 	case DISPLAY_SONY_ACX424AKP_ID3:
 	case DISPLAY_SONY_ACX424AKP_ID4:
-		dev_info(nt->dev, "MTP vendor: %02x, version: %02x, panel: %02x\n",
+		dev_info(&dev,
+			 "MTP vendor: %02x, version: %02x, panel: %02x\n",
 			 vendor, version, panel);
 		break;
 	default:
-		dev_info(nt->dev, "unknown vendor: %02x, version: %02x, panel: %02x\n",
+		dev_info(&dev,
+			 "unknown vendor: %02x, version: %02x, panel: %02x\n",
 			 vendor, version, panel);
 		break;
 	}
-
-	return 0;
 }
 
 static int nt35560_power_on(struct nt35560 *nt)
@@ -322,92 +280,56 @@ static void nt35560_power_off(struct nt35560 *nt)
 static int nt35560_prepare(struct drm_panel *panel)
 {
 	struct nt35560 *nt = panel_to_nt35560(panel);
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev);
-	const u8 mddi = 3;
+	struct mipi_dsi_multi_context dsi_ctx = {
+		.dsi = to_mipi_dsi_device(nt->dev)
+	};
 	int ret;
 
 	ret = nt35560_power_on(nt);
 	if (ret)
 		return ret;
 
-	ret = nt35560_read_id(nt);
-	if (ret) {
-		dev_err(nt->dev, "failed to read panel ID (%d)\n", ret);
-		goto err_power_off;
-	}
+	nt35560_read_id(&dsi_ctx);
 
-	/* Enabe tearing mode: send TE (tearing effect) at VBLANK */
-	ret = mipi_dsi_dcs_set_tear_on(dsi,
+	/* Enable tearing mode: send TE (tearing effect) at VBLANK */
+	mipi_dsi_dcs_set_tear_on_multi(&dsi_ctx,
 				       MIPI_DSI_DCS_TEAR_MODE_VBLANK);
-	if (ret) {
-		dev_err(nt->dev, "failed to enable vblank TE (%d)\n", ret);
-		goto err_power_off;
-	}
 
 	/*
 	 * Set MDDI
 	 *
 	 * This presumably deactivates the Qualcomm MDDI interface and
 	 * selects DSI, similar code is found in other drivers such as the
-	 * Sharp LS043T1LE01 which makes us suspect that this panel may be
-	 * using a Novatek NT35565 or similar display driver chip that shares
-	 * this command. Due to the lack of documentation we cannot know for
-	 * sure.
+	 * Sharp LS043T1LE01.
 	 */
-	ret = mipi_dsi_dcs_write(dsi, NT35560_DCS_SET_MDDI,
-				 &mddi, sizeof(mddi));
-	if (ret < 0) {
-		dev_err(nt->dev, "failed to set MDDI (%d)\n", ret);
-		goto err_power_off;
-	}
+	mipi_dsi_dcs_write_seq_multi(&dsi_ctx, NT35560_DCS_SET_MDDI, 3);
 
-	/* Exit sleep mode */
-	ret = mipi_dsi_dcs_exit_sleep_mode(dsi);
-	if (ret) {
-		dev_err(nt->dev, "failed to exit sleep mode (%d)\n", ret);
-		goto err_power_off;
-	}
-	msleep(140);
+	mipi_dsi_dcs_exit_sleep_mode_multi(&dsi_ctx);
+	mipi_dsi_msleep(&dsi_ctx, 140);
 
-	ret = mipi_dsi_dcs_set_display_on(dsi);
-	if (ret) {
-		dev_err(nt->dev, "failed to turn display on (%d)\n", ret);
-		goto err_power_off;
-	}
+	mipi_dsi_dcs_set_display_on_multi(&dsi_ctx);
 	if (nt->video_mode) {
-		/* In video mode turn peripheral on */
-		ret = mipi_dsi_turn_on_peripheral(dsi);
-		if (ret) {
-			dev_err(nt->dev, "failed to turn on peripheral\n");
-			goto err_power_off;
-		}
+		mipi_dsi_turn_on_peripheral_multi(&dsi_ctx);
 	}
 
-	return 0;
-
-err_power_off:
-	nt35560_power_off(nt);
-	return ret;
+	if (dsi_ctx.accum_err < 0)
+		nt35560_power_off(nt);
+	return dsi_ctx.accum_err;
 }
 
 static int nt35560_unprepare(struct drm_panel *panel)
 {
 	struct nt35560 *nt = panel_to_nt35560(panel);
-	struct mipi_dsi_device *dsi = to_mipi_dsi_device(nt->dev);
-	int ret;
+	struct mipi_dsi_multi_context dsi_ctx = {
+		.dsi = to_mipi_dsi_device(nt->dev)
+	};
 
-	ret = mipi_dsi_dcs_set_display_off(dsi);
-	if (ret) {
-		dev_err(nt->dev, "failed to turn display off (%d)\n", ret);
-		return ret;
-	}
+	mipi_dsi_dcs_set_display_off_multi(&dsi_ctx);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&dsi_ctx);
+
+	if (dsi_ctx.accum_err < 0)
+		return dsi_ctx.accum_err;
 
-	/* Enter sleep mode */
-	ret = mipi_dsi_dcs_enter_sleep_mode(dsi);
-	if (ret) {
-		dev_err(nt->dev, "failed to enter sleep mode (%d)\n", ret);
-		return ret;
-	}
 	msleep(85);
 
 	nt35560_power_off(nt);

From 5e04ff729f93fff16ec760e9cd7b72499963663f Mon Sep 17 00:00:00 2001
From: John Hubbard <jhubbard@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:06 +0900
Subject: [PATCH 0455/1455] gpu: nova-core: register: minor grammar and
 spelling fixes

There is only one top-level macro in this file at the moment, but the
"macros.rs" file name allows for more. Change the wording so that it
will remain valid even if additional macros are added to the file.

Fix a couple of spelling errors and grammatical errors, and break up a
run-on sentence, for clarity.

Cc: Alexandre Courbot <acourbot@nvidia.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-1-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index a3e6de1779d4..6d4080a65ee4 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -1,17 +1,17 @@
 // SPDX-License-Identifier: GPL-2.0
 
-//! Macro to define register layout and accessors.
+//! `register!` macro to define register layout and accessors.
 //!
 //! A single register typically includes several fields, which are accessed through a combination
 //! of bit-shift and mask operations that introduce a class of potential mistakes, notably because
 //! not all possible field values are necessarily valid.
 //!
-//! The macro in this module allow to define, using an intruitive and readable syntax, a dedicated
-//! type for each register with its own field accessors that can return an error is a field's value
-//! is invalid.
+//! The `register!` macro in this module provides an intuitive and readable syntax for defining a
+//! dedicated type for each register. Each such type comes with its own field accessors that can
+//! return an error if a field's value is invalid.
 
-/// Defines a dedicated type for a register with an absolute offset, alongside with getter and
-/// setter methods for its fields and methods to read and write it from an `Io` region.
+/// Defines a dedicated type for a register with an absolute offset, including getter and setter
+/// methods for its fields and methods to read and write it from an `Io` region.
 ///
 /// Example:
 ///
@@ -24,7 +24,7 @@
 /// ```
 ///
 /// This defines a `BOOT_0` type which can be read or written from offset `0x100` of an `Io`
-/// region. It is composed of 3 fields, for instance `minor_revision` is made of the 4 less
+/// region. It is composed of 3 fields, for instance `minor_revision` is made of the 4 least
 /// significant bits of the register. Each field can be accessed and modified using accessor
 /// methods:
 ///

From be3536a4bdda53ff5a91b7e542b167d12bddb317 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:07 +0900
Subject: [PATCH 0456/1455] gpu: nova-core: register: add missing space in
 register!()

A space was missing between arguments in this invocation.

[acourbot@nvidia.com: use more descriptive commit title]

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link:
https://lore.kernel.org/r/20250718-nova-regs-v2-2-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 6d4080a65ee4..9cc612f41f12 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -116,7 +116,7 @@ macro_rules! register {
     ) => {
         register!(@common $name @ $offset $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
-        register!(@io$name @ + $offset);
+        register!(@io $name @ + $offset);
     };
 
     // Creates a alias register of relative offset register `alias` with its own fields.

From c5aeb264b6b27c52fc6c9ef3b50eaaebff5d9b60 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:08 +0900
Subject: [PATCH 0457/1455] gpu: nova-core: register: allow fields named
 `offset`

`offset` is a common field name, yet using it triggers a build error due
to the conflict between the uppercased field constant (which becomes
`OFFSET` in this case) containing the bitrange of the field, and the
`OFFSET` constant constaining the offset of the register.

Fix this by adding `_RANGE` the field's range constant to avoid the
name collision.

[acourbot@nvidia.com: fix merge conflict due to switch from `as u32` to
`u32::from`.]

Reported-by: Timur Tabi <ttabi@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-3-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs.rs        | 5 +++--
 drivers/gpu/nova-core/regs/macros.rs | 2 +-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index d49fddf6a3c6..c8f8adb24f6e 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -28,7 +28,7 @@ impl NV_PMC_BOOT_0 {
     /// Combines `architecture_0` and `architecture_1` to obtain the architecture of the chip.
     pub(crate) fn architecture(self) -> Result<Architecture> {
         Architecture::try_from(
-            self.architecture_0() | (self.architecture_1() << Self::ARCHITECTURE_0.len()),
+            self.architecture_0() | (self.architecture_1() << Self::ARCHITECTURE_0_RANGE.len()),
         )
     }
 
@@ -36,7 +36,8 @@ impl NV_PMC_BOOT_0 {
     pub(crate) fn chipset(self) -> Result<Chipset> {
         self.architecture()
             .map(|arch| {
-                ((arch as u32) << Self::IMPLEMENTATION.len()) | u32::from(self.implementation())
+                ((arch as u32) << Self::IMPLEMENTATION_RANGE.len())
+                    | u32::from(self.implementation())
             })
             .and_then(Chipset::try_from)
     }
diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 9cc612f41f12..023748685412 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -278,7 +278,7 @@ macro_rules! register {
             { $process:expr } $to_type:ty => $res_type:ty $(, $comment:literal)?;
     ) => {
         ::kernel::macros::paste!(
-        const [<$field:upper>]: ::core::ops::RangeInclusive<u8> = $lo..=$hi;
+        const [<$field:upper _RANGE>]: ::core::ops::RangeInclusive<u8> = $lo..=$hi;
         const [<$field:upper _MASK>]: u32 = ((((1 << $hi) - 1) << 1) + 1) - ((1 << $lo) - 1);
         const [<$field:upper _SHIFT>]: u32 = Self::[<$field:upper _MASK>].trailing_zeros();
         );

From cb2607f3b27087120c15c13f57f1a950251445c9 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:09 +0900
Subject: [PATCH 0458/1455] gpu: nova-core: register: improve documentation for
 basic registers

Reword parts of the documentation that were a bit heavy to read, and
harmonize/fix the examples.

The relative registers section is about to be redesigned and its
documentation rewritten, so do not touch this part.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-4-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 023748685412..31df1827de77 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -33,25 +33,25 @@
 /// let boot0 = BOOT_0::read(&bar);
 /// pr_info!("chip revision: {}.{}", boot0.major_revision(), boot0.minor_revision());
 ///
-/// // `Chipset::try_from` will be called with the value of the field and returns an error if the
-/// // value is invalid.
+/// // `Chipset::try_from` is called with the value of the `chipset` field and returns an
+/// // error if it is invalid.
 /// let chipset = boot0.chipset()?;
 ///
 /// // Update some fields and write the value back.
 /// boot0.set_major_revision(3).set_minor_revision(10).write(&bar);
 ///
-/// // Or just read and update the register in a single step:
+/// // Or, just read and update the register in a single step:
 /// BOOT_0::alter(&bar, |r| r.set_major_revision(3).set_minor_revision(10));
 /// ```
 ///
-/// Fields can be defined as follows:
+/// Fields are defined as follows:
 ///
-/// - `as <type>` simply returns the field value casted as the requested integer type, typically
-///   `u32`, `u16`, `u8` or `bool`. Note that `bool` fields must have a range of 1 bit.
+/// - `as <type>` simply returns the field value casted to <type>, typically `u32`, `u16`, `u8` or
+///   `bool`. Note that `bool` fields must have a range of 1 bit.
 /// - `as <type> => <into_type>` calls `<into_type>`'s `From::<<type>>` implementation and returns
 ///   the result.
 /// - `as <type> ?=> <try_into_type>` calls `<try_into_type>`'s `TryFrom::<<type>>` implementation
-///   and returns the result. This is useful on fields for which not all values are value.
+///   and returns the result. This is useful with fields for which not all values are valid.
 ///
 /// The documentation strings are optional. If present, they will be added to the type's
 /// definition, or the field getter and setter methods they are attached to.
@@ -76,15 +76,17 @@
 /// for cases where a register's interpretation depends on the context:
 ///
 /// ```no_run
-/// register!(SCRATCH_0 @ 0x0000100, "Scratch register 0" {
+/// register!(SCRATCH @ 0x00000200, "Scratch register" {
 ///    31:0     value as u32, "Raw value";
+/// });
 ///
-/// register!(SCRATCH_0_BOOT_STATUS => SCRATCH_0, "Boot status of the firmware" {
+/// register!(SCRATCH_BOOT_STATUS => SCRATCH, "Boot status of the firmware" {
 ///     0:0     completed as bool, "Whether the firmware has completed booting";
+/// });
 /// ```
 ///
-/// In this example, `SCRATCH_0_BOOT_STATUS` uses the same I/O address as `SCRATCH_0`, while also
-/// providing its own `completed` method.
+/// In this example, `SCRATCH_0_BOOT_STATUS` uses the same I/O address as `SCRATCH`, while also
+/// providing its own `completed` field.
 macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
     (

From 4f7f8f847d4a9c2460ecd43a702e51ac31919e2f Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:10 +0900
Subject: [PATCH 0459/1455] gpu: nova-core: register: simplify @leaf_accessor
 rule

The `$type` metavariable is not used in the @leaf_accessor rule, so
remove it.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-5-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 31df1827de77..44e2a12dce52 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -230,7 +230,7 @@ macro_rules! register {
             $(, $comment:literal)?;
     ) => {
         register!(
-            @leaf_accessor $name $hi:$lo $field as bool
+            @leaf_accessor $name $hi:$lo $field
             { |f| <$into_type>::from(if f != 0 { true } else { false }) }
             $into_type => $into_type $(, $comment)?;
         );
@@ -248,7 +248,7 @@ macro_rules! register {
         @field_accessor $name:ident $hi:tt:$lo:tt $field:ident as $type:tt ?=> $try_into_type:ty
             $(, $comment:literal)?;
     ) => {
-        register!(@leaf_accessor $name $hi:$lo $field as $type
+        register!(@leaf_accessor $name $hi:$lo $field
             { |f| <$try_into_type>::try_from(f as $type) } $try_into_type =>
             ::core::result::Result<
                 $try_into_type,
@@ -262,7 +262,7 @@ macro_rules! register {
         @field_accessor $name:ident $hi:tt:$lo:tt $field:ident as $type:tt => $into_type:ty
             $(, $comment:literal)?;
     ) => {
-        register!(@leaf_accessor $name $hi:$lo $field as $type
+        register!(@leaf_accessor $name $hi:$lo $field
             { |f| <$into_type>::from(f as $type) } $into_type => $into_type $(, $comment)?;);
     };
 
@@ -276,7 +276,7 @@ macro_rules! register {
 
     // Generates the accessor methods for a single field.
     (
-        @leaf_accessor $name:ident $hi:tt:$lo:tt $field:ident as $type:ty
+        @leaf_accessor $name:ident $hi:tt:$lo:tt $field:ident
             { $process:expr } $to_type:ty => $res_type:ty $(, $comment:literal)?;
     ) => {
         ::kernel::macros::paste!(

From b567daf4ea4e4565118436aca51d12adf8ab3e12 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:11 +0900
Subject: [PATCH 0460/1455] gpu: nova-core: register: remove `try_` accessors
 for relative registers

Relative registers are always accessed using a literal base, meaning
their validity can always be checked at compile-time. Thus remove the
`try_` accessors that have no purpose.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-6-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 38 +---------------------------
 1 file changed, 1 insertion(+), 37 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 44e2a12dce52..0567b3aae4d2 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -57,9 +57,7 @@
 /// definition, or the field getter and setter methods they are attached to.
 ///
 /// Putting a `+` before the address of the register makes it relative to a base: the `read` and
-/// `write` methods take a `base` argument that is added to the specified address before access,
-/// and `try_read` and `try_write` methods are also created, allowing access with offsets unknown
-/// at compile-time:
+/// `write` methods take a `base` argument that is added to the specified address before access:
 ///
 /// ```no_run
 /// register!(CPU_CTL @ +0x0000010, "CPU core control" {
@@ -386,40 +384,6 @@ macro_rules! register {
                 let reg = f(Self::read(io, base));
                 reg.write(io, base);
             }
-
-            #[inline]
-            pub(crate) fn try_read<const SIZE: usize, T>(
-                io: &T,
-                base: usize,
-            ) -> ::kernel::error::Result<Self> where
-                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
-            {
-                io.try_read32(base + $offset).map(Self)
-            }
-
-            #[inline]
-            pub(crate) fn try_write<const SIZE: usize, T>(
-                self,
-                io: &T,
-                base: usize,
-            ) -> ::kernel::error::Result<()> where
-                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
-            {
-                io.try_write32(self.0, base + $offset)
-            }
-
-            #[inline]
-            pub(crate) fn try_alter<const SIZE: usize, T, F>(
-                io: &T,
-                base: usize,
-                f: F,
-            ) -> ::kernel::error::Result<()> where
-                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
-                F: ::core::ops::FnOnce(Self) -> Self,
-            {
-                let reg = f(Self::try_read(io, base)?);
-                reg.try_write(io, base)
-            }
         }
     };
 }

From 036c5fa25020d8b1619a96dabbcf9b156f4329bb Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:12 +0900
Subject: [PATCH 0461/1455] gpu: nova-core: register: move OFFSET declaration
 to I/O impl block

The OFFSET const is an I/O property, and having to pass it to the
@common rule makes it impossible to make I/O optional, as we want to get
to eventually.

Thus, move OFFSET to the I/O impl block so it is not needed by the
@common rule anymore.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-7-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 0567b3aae4d2..3d4476cc2421 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -92,7 +92,7 @@ macro_rules! register {
             $($fields:tt)*
         }
     ) => {
-        register!(@common $name @ $offset $(, $comment)?);
+        register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ $offset);
     };
@@ -103,7 +103,7 @@ macro_rules! register {
             $($fields:tt)*
         }
     ) => {
-        register!(@common $name @ $alias::OFFSET $(, $comment)?);
+        register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ $alias::OFFSET);
     };
@@ -114,7 +114,7 @@ macro_rules! register {
             $($fields:tt)*
         }
     ) => {
-        register!(@common $name @ $offset $(, $comment)?);
+        register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ + $offset);
     };
@@ -125,7 +125,7 @@ macro_rules! register {
             $($fields:tt)*
         }
     ) => {
-        register!(@common $name @ $alias::OFFSET $(, $comment)?);
+        register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ + $alias::OFFSET);
     };
@@ -134,7 +134,7 @@ macro_rules! register {
 
     // Defines the wrapper `$name` type, as well as its relevant implementations (`Debug`, `BitOr`,
     // and conversion to regular `u32`).
-    (@common $name:ident @ $offset:expr $(, $comment:literal)?) => {
+    (@common $name:ident $(, $comment:literal)?) => {
         $(
         #[doc=$comment]
         )?
@@ -142,11 +142,6 @@ macro_rules! register {
         #[derive(Clone, Copy, Default)]
         pub(crate) struct $name(u32);
 
-        #[allow(dead_code)]
-        impl $name {
-            pub(crate) const OFFSET: usize = $offset;
-        }
-
         // TODO[REGA]: display the raw hex value, then the value of all the fields. This requires
         // matching the fields, which will complexify the syntax considerably...
         impl ::core::fmt::Debug for $name {
@@ -319,6 +314,8 @@ macro_rules! register {
     (@io $name:ident @ $offset:expr) => {
         #[allow(dead_code)]
         impl $name {
+            pub(crate) const OFFSET: usize = $offset;
+
             #[inline]
             pub(crate) fn read<const SIZE: usize, T>(io: &T) -> Self where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
@@ -351,6 +348,8 @@ macro_rules! register {
     (@io $name:ident @ + $offset:literal) => {
         #[allow(dead_code)]
         impl $name {
+            pub(crate) const OFFSET: usize = $offset;
+
             #[inline]
             pub(crate) fn read<const SIZE: usize, T>(
                 io: &T,

From e40d2b26167200b99fa4fb1df15d4c870489c82e Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:13 +0900
Subject: [PATCH 0462/1455] gpu: nova-core: register: fix documentation and
 indentation

Fix a few documentation inconsistencies, and harmonize indentation where
possible.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-8-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 34 ++++++++--------------------
 1 file changed, 9 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 3d4476cc2421..5b3a320de1d3 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -87,44 +87,28 @@
 /// providing its own `completed` field.
 macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
-    (
-        $name:ident @ $offset:literal $(, $comment:literal)? {
-            $($fields:tt)*
-        }
-    ) => {
+    ($name:ident @ $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ $offset);
     };
 
-    // Creates a alias register of fixed offset register `alias` with its own fields.
-    (
-        $name:ident => $alias:ident $(, $comment:literal)? {
-            $($fields:tt)*
-        }
-    ) => {
+    // Creates an alias register of fixed offset register `alias` with its own fields.
+    ($name:ident => $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ $alias::OFFSET);
     };
 
     // Creates a register at a relative offset from a base address.
-    (
-        $name:ident @ + $offset:literal $(, $comment:literal)? {
-            $($fields:tt)*
-        }
-    ) => {
+    ($name:ident @ + $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ + $offset);
     };
 
-    // Creates a alias register of relative offset register `alias` with its own fields.
-    (
-        $name:ident => + $alias:ident $(, $comment:literal)? {
-            $($fields:tt)*
-        }
-    ) => {
+    // Creates an alias register of relative offset register `alias` with its own fields.
+    ($name:ident => + $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@common $name $(, $comment)?);
         register!(@field_accessors $name { $($fields)* });
         register!(@io $name @ + $alias::OFFSET);
@@ -259,7 +243,7 @@ macro_rules! register {
             { |f| <$into_type>::from(f as $type) } $into_type => $into_type $(, $comment)?;);
     };
 
-    // Shortcut for fields defined as non-`bool` without the `=>` or `?=>` syntax.
+    // Shortcut for non-boolean fields defined without the `=>` or `?=>` syntax.
     (
         @field_accessor $name:ident $hi:tt:$lo:tt $field:ident as $type:tt
             $(, $comment:literal)?;
@@ -310,7 +294,7 @@ macro_rules! register {
         );
     };
 
-    // Creates the IO accessors for a fixed offset register.
+    // Generates the IO accessors for a fixed offset register.
     (@io $name:ident @ $offset:expr) => {
         #[allow(dead_code)]
         impl $name {
@@ -344,7 +328,7 @@ macro_rules! register {
         }
     };
 
-    // Create the IO accessors for a relative offset register.
+    // Generates the IO accessors for a relative offset register.
     (@io $name:ident @ + $offset:literal) => {
         #[allow(dead_code)]
         impl $name {

From 9b2379f02423801d372bc588d38b62a1fa6cd05f Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:14 +0900
Subject: [PATCH 0463/1455] gpu: nova-core: register: add missing doccomments
 for fixed registers I/O accessors

Add the missing doccomments for these accessors, as having a bit of
inline documentation is always helpful.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-9-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 5b3a320de1d3..d5ce7cb67f21 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -300,6 +300,7 @@ macro_rules! register {
         impl $name {
             pub(crate) const OFFSET: usize = $offset;
 
+            /// Read the register from its address in `io`.
             #[inline]
             pub(crate) fn read<const SIZE: usize, T>(io: &T) -> Self where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
@@ -307,6 +308,7 @@ macro_rules! register {
                 Self(io.read32($offset))
             }
 
+            /// Write the value contained in `self` to the register address in `io`.
             #[inline]
             pub(crate) fn write<const SIZE: usize, T>(self, io: &T) where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
@@ -314,6 +316,8 @@ macro_rules! register {
                 io.write32(self.0, $offset)
             }
 
+            /// Read the register from its address in `io` and run `f` on its value to obtain a new
+            /// value to write back.
             #[inline]
             pub(crate) fn alter<const SIZE: usize, T, F>(
                 io: &T,

From 7a9cb3dfb07d3e29c0eeebb26b11a34a19a889a4 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:15 +0900
Subject: [PATCH 0464/1455] gpu: nova-core: register: add fields dispatcher
 internal rule

Fields are complex and cumbersome to match in a rule, and were only
captured in order to generate the field accessors. However, there are
other places (like the `Debug` and `Default` implementations) where we
would benefit from having access to at least some of the field
information, but refrained from doing so because it would have meant
matching the whole fields in a rule more complex than we need.

Introduce a new `@fields_dispatcher` internal rule that captures all the
field information and passes it to `@field_accessors`. It does not
provide any functional change in itself, but allows us to reuse the
captured field information partially to provide better `Debug` and
`Default` implementations in following patches.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-10-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 42 +++++++++++++++++++++-------
 1 file changed, 32 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index d5ce7cb67f21..ff2deca6b1f3 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -88,37 +88,33 @@
 macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
     ($name:ident @ $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
-        register!(@common $name $(, $comment)?);
-        register!(@field_accessors $name { $($fields)* });
+        register!(@core $name $(, $comment)? { $($fields)* } );
         register!(@io $name @ $offset);
     };
 
     // Creates an alias register of fixed offset register `alias` with its own fields.
     ($name:ident => $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
-        register!(@common $name $(, $comment)?);
-        register!(@field_accessors $name { $($fields)* });
+        register!(@core $name $(, $comment)? { $($fields)* } );
         register!(@io $name @ $alias::OFFSET);
     };
 
     // Creates a register at a relative offset from a base address.
     ($name:ident @ + $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
-        register!(@common $name $(, $comment)?);
-        register!(@field_accessors $name { $($fields)* });
+        register!(@core $name $(, $comment)? { $($fields)* } );
         register!(@io $name @ + $offset);
     };
 
     // Creates an alias register of relative offset register `alias` with its own fields.
     ($name:ident => + $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
-        register!(@common $name $(, $comment)?);
-        register!(@field_accessors $name { $($fields)* });
+        register!(@core $name $(, $comment)? { $($fields)* } );
         register!(@io $name @ + $alias::OFFSET);
     };
 
     // All rules below are helpers.
 
     // Defines the wrapper `$name` type, as well as its relevant implementations (`Debug`, `BitOr`,
-    // and conversion to regular `u32`).
-    (@common $name:ident $(, $comment:literal)?) => {
+    // and conversion to the value type) and field accessor methods.
+    (@core $name:ident $(, $comment:literal)? { $($fields:tt)* }) => {
         $(
         #[doc=$comment]
         )?
@@ -149,6 +145,32 @@ macro_rules! register {
                 reg.0
             }
         }
+
+        register!(@fields_dispatcher $name { $($fields)* });
+    };
+
+    // Captures the fields and passes them to all the implementers that require field information.
+    //
+    // Used to simplify the matching rules for implementers, so they don't need to match the entire
+    // complex fields rule even though they only make use of part of it.
+    (@fields_dispatcher $name:ident {
+        $($hi:tt:$lo:tt $field:ident as $type:tt
+            $(?=> $try_into_type:ty)?
+            $(=> $into_type:ty)?
+            $(, $comment:literal)?
+        ;
+        )*
+    }
+    ) => {
+        register!(@field_accessors $name {
+            $(
+                $hi:$lo $field as $type
+                $(?=> $try_into_type)?
+                $(=> $into_type)?
+                $(, $comment)?
+            ;
+            )*
+        });
     };
 
     // Defines all the field getter/methods methods for `$name`.

From 6ecd6b73e0845e2a1ba0d4da273c7cc7c21c88db Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:16 +0900
Subject: [PATCH 0465/1455] gpu: nova-core: register: improve `Debug`
 implementation

Now that we have an internal rule to dispatch field information where
needed, use it to generate a better `Debug` implementation where the raw
hexadecimal value of the register is displayed, as well as the `Debug`
values of its individual fields.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-11-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index ff2deca6b1f3..c966e58b775d 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -122,16 +122,6 @@ macro_rules! register {
         #[derive(Clone, Copy, Default)]
         pub(crate) struct $name(u32);
 
-        // TODO[REGA]: display the raw hex value, then the value of all the fields. This requires
-        // matching the fields, which will complexify the syntax considerably...
-        impl ::core::fmt::Debug for $name {
-            fn fmt(&self, f: &mut ::core::fmt::Formatter<'_>) -> ::core::fmt::Result {
-                f.debug_tuple(stringify!($name))
-                    .field(&format_args!("0x{0:x}", &self.0))
-                    .finish()
-            }
-        }
-
         impl ::core::ops::BitOr for $name {
             type Output = Self;
 
@@ -171,6 +161,7 @@ macro_rules! register {
             ;
             )*
         });
+        register!(@debug $name { $($field;)* });
     };
 
     // Defines all the field getter/methods methods for `$name`.
@@ -316,6 +307,20 @@ macro_rules! register {
         );
     };
 
+    // Generates the `Debug` implementation for `$name`.
+    (@debug $name:ident { $($field:ident;)* }) => {
+        impl ::core::fmt::Debug for $name {
+            fn fmt(&self, f: &mut ::core::fmt::Formatter<'_>) -> ::core::fmt::Result {
+                f.debug_struct(stringify!($name))
+                    .field("<raw>", &format_args!("{:#x}", &self.0))
+                $(
+                    .field(stringify!($field), &self.$field())
+                )*
+                    .finish()
+            }
+        }
+    };
+
     // Generates the IO accessors for a fixed offset register.
     (@io $name:ident @ $offset:expr) => {
         #[allow(dead_code)]

From 3fa145bef533f899ed6de641fe99c70028e28481 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:17 +0900
Subject: [PATCH 0466/1455] gpu: nova-core: register: generate correct
 `Default` implementation

The `Default` implementation of a register should be the aggregate of
the default values of all its fields, and not simply be zeroed.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-12-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 26 +++++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index c966e58b775d..e43e1ab7ae2e 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -112,14 +112,14 @@ macro_rules! register {
 
     // All rules below are helpers.
 
-    // Defines the wrapper `$name` type, as well as its relevant implementations (`Debug`, `BitOr`,
-    // and conversion to the value type) and field accessor methods.
+    // Defines the wrapper `$name` type, as well as its relevant implementations (`Debug`,
+    // `Default`, `BitOr`, and conversion to the value type) and field accessor methods.
     (@core $name:ident $(, $comment:literal)? { $($fields:tt)* }) => {
         $(
         #[doc=$comment]
         )?
         #[repr(transparent)]
-        #[derive(Clone, Copy, Default)]
+        #[derive(Clone, Copy)]
         pub(crate) struct $name(u32);
 
         impl ::core::ops::BitOr for $name {
@@ -162,6 +162,7 @@ macro_rules! register {
             )*
         });
         register!(@debug $name { $($field;)* });
+        register!(@default $name { $($field;)* });
     };
 
     // Defines all the field getter/methods methods for `$name`.
@@ -321,6 +322,25 @@ macro_rules! register {
         }
     };
 
+    // Generates the `Default` implementation for `$name`.
+    (@default $name:ident { $($field:ident;)* }) => {
+        /// Returns a value for the register where all fields are set to their default value.
+        impl ::core::default::Default for $name {
+            fn default() -> Self {
+                #[allow(unused_mut)]
+                let mut value = Self(Default::default());
+
+                ::kernel::macros::paste!(
+                $(
+                value.[<set_ $field>](Default::default());
+                )*
+                );
+
+                value
+            }
+        }
+    };
+
     // Generates the IO accessors for a fixed offset register.
     (@io $name:ident @ $offset:expr) => {
         #[allow(dead_code)]

From fcdce54d645a0779892faaed1209105350e15d92 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:18 +0900
Subject: [PATCH 0467/1455] gpu: nova-core: register: split @io rule into fixed
 and relative versions

We used the same @io rule with different patterns to define both the
fixed and relative I/O accessors. This can be confusing as the matching
rules are very similar.

Since all call sites know which version they want to call, split @io
into @io_fixed and @io_relative to remove any ambiguity.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-13-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index e43e1ab7ae2e..432af02b3378 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -89,25 +89,25 @@ macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
     ($name:ident @ $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@core $name $(, $comment)? { $($fields)* } );
-        register!(@io $name @ $offset);
+        register!(@io_fixed $name @ $offset);
     };
 
     // Creates an alias register of fixed offset register `alias` with its own fields.
     ($name:ident => $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@core $name $(, $comment)? { $($fields)* } );
-        register!(@io $name @ $alias::OFFSET);
+        register!(@io_fixed $name @ $alias::OFFSET);
     };
 
     // Creates a register at a relative offset from a base address.
     ($name:ident @ + $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@core $name $(, $comment)? { $($fields)* } );
-        register!(@io $name @ + $offset);
+        register!(@io_relative $name @ + $offset);
     };
 
     // Creates an alias register of relative offset register `alias` with its own fields.
     ($name:ident => + $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@core $name $(, $comment)? { $($fields)* } );
-        register!(@io $name @ + $alias::OFFSET);
+        register!(@io_relative $name @ + $alias::OFFSET);
     };
 
     // All rules below are helpers.
@@ -342,7 +342,7 @@ macro_rules! register {
     };
 
     // Generates the IO accessors for a fixed offset register.
-    (@io $name:ident @ $offset:expr) => {
+    (@io_fixed $name:ident @ $offset:expr) => {
         #[allow(dead_code)]
         impl $name {
             pub(crate) const OFFSET: usize = $offset;
@@ -380,7 +380,7 @@ macro_rules! register {
     };
 
     // Generates the IO accessors for a relative offset register.
-    (@io $name:ident @ + $offset:literal) => {
+    (@io_relative $name:ident @ + $offset:literal) => {
         #[allow(dead_code)]
         impl $name {
             pub(crate) const OFFSET: usize = $offset;

From c6bc4225279d9a6eff8aafc94347183b2babc52a Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:19 +0900
Subject: [PATCH 0468/1455] gpu: nova-core: register: use #[inline(always)] for
 all methods

These methods should always be inlined, so use the strongest compiler
hint that exists to maximize the chance they will indeed be.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-14-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/regs/macros.rs | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 432af02b3378..200fbe847571 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -280,7 +280,7 @@ macro_rules! register {
         #[doc="Returns the value of this field:"]
         #[doc=$comment]
         )?
-        #[inline]
+        #[inline(always)]
         pub(crate) fn $field(self) -> $res_type {
             ::kernel::macros::paste!(
             const MASK: u32 = $name::[<$field:upper _MASK>];
@@ -296,7 +296,7 @@ macro_rules! register {
         #[doc="Sets the value of this field:"]
         #[doc=$comment]
         )?
-        #[inline]
+        #[inline(always)]
         pub(crate) fn [<set_ $field>](mut self, value: $to_type) -> Self {
             const MASK: u32 = $name::[<$field:upper _MASK>];
             const SHIFT: u32 = $name::[<$field:upper _SHIFT>];
@@ -348,7 +348,7 @@ macro_rules! register {
             pub(crate) const OFFSET: usize = $offset;
 
             /// Read the register from its address in `io`.
-            #[inline]
+            #[inline(always)]
             pub(crate) fn read<const SIZE: usize, T>(io: &T) -> Self where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
             {
@@ -356,7 +356,7 @@ macro_rules! register {
             }
 
             /// Write the value contained in `self` to the register address in `io`.
-            #[inline]
+            #[inline(always)]
             pub(crate) fn write<const SIZE: usize, T>(self, io: &T) where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
             {
@@ -365,7 +365,7 @@ macro_rules! register {
 
             /// Read the register from its address in `io` and run `f` on its value to obtain a new
             /// value to write back.
-            #[inline]
+            #[inline(always)]
             pub(crate) fn alter<const SIZE: usize, T, F>(
                 io: &T,
                 f: F,
@@ -385,7 +385,7 @@ macro_rules! register {
         impl $name {
             pub(crate) const OFFSET: usize = $offset;
 
-            #[inline]
+            #[inline(always)]
             pub(crate) fn read<const SIZE: usize, T>(
                 io: &T,
                 base: usize,
@@ -395,7 +395,7 @@ macro_rules! register {
                 Self(io.read32(base + $offset))
             }
 
-            #[inline]
+            #[inline(always)]
             pub(crate) fn write<const SIZE: usize, T>(
                 self,
                 io: &T,
@@ -406,7 +406,7 @@ macro_rules! register {
                 io.write32(self.0, base + $offset)
             }
 
-            #[inline]
+            #[inline(always)]
             pub(crate) fn alter<const SIZE: usize, T, F>(
                 io: &T,
                 base: usize,

From af10924fc471d1c693b8689249f53ea10b0519b7 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:20 +0900
Subject: [PATCH 0469/1455] gpu: nova-core: register: redesign relative
 registers

The relative registers are currently very unsafe to use: callers can
specify any constant as the base address for access, meaning they can
effectively interpret any I/O address as any relative register.

Ideally, valid base addresses for a family of registers should be
explicitly defined in the code, and could only be used with the relevant
registers

This patch changes the relative register declaration from e.g.:

    register!(CPU_CTL @ +0x0000010, "CPU core control" {
        0:0     start as bool, "Start the CPU core";
    });

into:

    register!(CPU_CTL @ CpuCtlBase[0x10], "CPU core control" {
        0:0     start as bool, "Start the CPU core";
    });

Where `CpuCtlBase` is the name of a ZST used as a parameter of the
`RegisterBase<>` trait to define a trait unique to a class of register.
This specialized trait is then implemented for every type that provides
a valid base address, enabling said types to be passed as the base
address provider for the register's I/O accessor methods.

This design thus makes it impossible to pass an unexpected base address
to a relative register, and, since the valid bases are all known at
compile-time, also guarantees that all I/O accesses are done within the
valid bounds of the I/O range.

[acourbot@nvidia.com: add example in the commit log.]

Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-15-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 Documentation/gpu/nova/core/todo.rst      |   1 -
 drivers/gpu/nova-core/falcon.rs           |  67 +++++-----
 drivers/gpu/nova-core/falcon/gsp.rs       |  12 +-
 drivers/gpu/nova-core/falcon/hal/ga102.rs |  14 +-
 drivers/gpu/nova-core/falcon/sec2.rs      |   9 +-
 drivers/gpu/nova-core/regs.rs             |  50 +++----
 drivers/gpu/nova-core/regs/macros.rs      | 156 ++++++++++++++++++----
 7 files changed, 212 insertions(+), 97 deletions(-)

diff --git a/Documentation/gpu/nova/core/todo.rst b/Documentation/gpu/nova/core/todo.rst
index 894a1e9c3741..a1d12c1b289d 100644
--- a/Documentation/gpu/nova/core/todo.rst
+++ b/Documentation/gpu/nova/core/todo.rst
@@ -131,7 +131,6 @@ crate so it can be used by other components as well.
 
 Features desired before this happens:
 
-* Relative register with build-time base address validation,
 * Arrays of registers with build-time index validation,
 * Make I/O optional I/O (for field values that are not registers),
 * Support other sizes than `u32`,
diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index 50437c67c14a..67265a0b5d7b 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -14,6 +14,7 @@ use crate::dma::DmaObject;
 use crate::driver::Bar0;
 use crate::gpu::Chipset;
 use crate::regs;
+use crate::regs::macros::RegisterBase;
 use crate::util;
 
 pub(crate) mod gsp;
@@ -274,10 +275,16 @@ impl From<bool> for FalconFbifMemType {
     }
 }
 
-/// Trait defining the parameters of a given Falcon instance.
-pub(crate) trait FalconEngine: Sync {
-    /// Base I/O address for the falcon, relative from which its registers are accessed.
-    const BASE: usize;
+/// Type used to represent the `PFALCON` registers address base for a given falcon engine.
+pub(crate) struct PFalconBase(());
+
+/// Trait defining the parameters of a given Falcon engine.
+///
+/// Each engine provides one base for `PFALCON` and `PFALCON2` registers. The `ID` constant is used
+/// to identify a given Falcon instance with register I/O methods.
+pub(crate) trait FalconEngine: Sync + RegisterBase<PFalconBase> + Sized {
+    /// Singleton of the engine, used to identify it with register I/O methods.
+    const ID: Self;
 }
 
 /// Represents a portion of the firmware to be loaded into a particular memory (e.g. IMEM or DMEM).
@@ -343,13 +350,13 @@ impl<E: FalconEngine + 'static> Falcon<E> {
         bar: &Bar0,
         need_riscv: bool,
     ) -> Result<Self> {
-        let hwcfg1 = regs::NV_PFALCON_FALCON_HWCFG1::read(bar, E::BASE);
+        let hwcfg1 = regs::NV_PFALCON_FALCON_HWCFG1::read(bar, &E::ID);
         // Check that the revision and security model contain valid values.
         let _ = hwcfg1.core_rev()?;
         let _ = hwcfg1.security_model()?;
 
         if need_riscv {
-            let hwcfg2 = regs::NV_PFALCON_FALCON_HWCFG2::read(bar, E::BASE);
+            let hwcfg2 = regs::NV_PFALCON_FALCON_HWCFG2::read(bar, &E::ID);
             if !hwcfg2.riscv() {
                 dev_err!(
                     dev,
@@ -369,7 +376,7 @@ impl<E: FalconEngine + 'static> Falcon<E> {
     fn reset_wait_mem_scrubbing(&self, bar: &Bar0) -> Result {
         // TIMEOUT: memory scrubbing should complete in less than 20ms.
         util::wait_on(Delta::from_millis(20), || {
-            if regs::NV_PFALCON_FALCON_HWCFG2::read(bar, E::BASE).mem_scrubbing_done() {
+            if regs::NV_PFALCON_FALCON_HWCFG2::read(bar, &E::ID).mem_scrubbing_done() {
                 Some(())
             } else {
                 None
@@ -379,12 +386,12 @@ impl<E: FalconEngine + 'static> Falcon<E> {
 
     /// Reset the falcon engine.
     fn reset_eng(&self, bar: &Bar0) -> Result {
-        let _ = regs::NV_PFALCON_FALCON_HWCFG2::read(bar, E::BASE);
+        let _ = regs::NV_PFALCON_FALCON_HWCFG2::read(bar, &E::ID);
 
         // According to OpenRM's `kflcnPreResetWait_GA102` documentation, HW sometimes does not set
         // RESET_READY so a non-failing timeout is used.
         let _ = util::wait_on(Delta::from_micros(150), || {
-            let r = regs::NV_PFALCON_FALCON_HWCFG2::read(bar, E::BASE);
+            let r = regs::NV_PFALCON_FALCON_HWCFG2::read(bar, &E::ID);
             if r.reset_ready() {
                 Some(())
             } else {
@@ -392,13 +399,13 @@ impl<E: FalconEngine + 'static> Falcon<E> {
             }
         });
 
-        regs::NV_PFALCON_FALCON_ENGINE::alter(bar, E::BASE, |v| v.set_reset(true));
+        regs::NV_PFALCON_FALCON_ENGINE::alter(bar, &E::ID, |v| v.set_reset(true));
 
         // TODO[DLAY]: replace with udelay() or equivalent once available.
         // TIMEOUT: falcon engine should not take more than 10us to reset.
         let _: Result = util::wait_on(Delta::from_micros(10), || None);
 
-        regs::NV_PFALCON_FALCON_ENGINE::alter(bar, E::BASE, |v| v.set_reset(false));
+        regs::NV_PFALCON_FALCON_ENGINE::alter(bar, &E::ID, |v| v.set_reset(false));
 
         self.reset_wait_mem_scrubbing(bar)?;
 
@@ -413,7 +420,7 @@ impl<E: FalconEngine + 'static> Falcon<E> {
 
         regs::NV_PFALCON_FALCON_RM::default()
             .set_value(regs::NV_PMC_BOOT_0::read(bar).into())
-            .write(bar, E::BASE);
+            .write(bar, &E::ID);
 
         Ok(())
     }
@@ -464,10 +471,10 @@ impl<E: FalconEngine + 'static> Falcon<E> {
 
         regs::NV_PFALCON_FALCON_DMATRFBASE::default()
             .set_base((dma_start >> 8) as u32)
-            .write(bar, E::BASE);
+            .write(bar, &E::ID);
         regs::NV_PFALCON_FALCON_DMATRFBASE1::default()
             .set_base((dma_start >> 40) as u16)
-            .write(bar, E::BASE);
+            .write(bar, &E::ID);
 
         let cmd = regs::NV_PFALCON_FALCON_DMATRFCMD::default()
             .set_size(DmaTrfCmdSize::Size256B)
@@ -478,17 +485,17 @@ impl<E: FalconEngine + 'static> Falcon<E> {
             // Perform a transfer of size `DMA_LEN`.
             regs::NV_PFALCON_FALCON_DMATRFMOFFS::default()
                 .set_offs(load_offsets.dst_start + pos)
-                .write(bar, E::BASE);
+                .write(bar, &E::ID);
             regs::NV_PFALCON_FALCON_DMATRFFBOFFS::default()
                 .set_offs(src_start + pos)
-                .write(bar, E::BASE);
-            cmd.write(bar, E::BASE);
+                .write(bar, &E::ID);
+            cmd.write(bar, &E::ID);
 
             // Wait for the transfer to complete.
             // TIMEOUT: arbitrarily large value, no DMA transfer to the falcon's small memories
             // should ever take that long.
             util::wait_on(Delta::from_secs(2), || {
-                let r = regs::NV_PFALCON_FALCON_DMATRFCMD::read(bar, E::BASE);
+                let r = regs::NV_PFALCON_FALCON_DMATRFCMD::read(bar, &E::ID);
                 if r.idle() {
                     Some(())
                 } else {
@@ -502,9 +509,9 @@ impl<E: FalconEngine + 'static> Falcon<E> {
 
     /// Perform a DMA load into `IMEM` and `DMEM` of `fw`, and prepare the falcon to run it.
     pub(crate) fn dma_load<F: FalconFirmware<Target = E>>(&self, bar: &Bar0, fw: &F) -> Result {
-        regs::NV_PFALCON_FBIF_CTL::alter(bar, E::BASE, |v| v.set_allow_phys_no_ctx(true));
-        regs::NV_PFALCON_FALCON_DMACTL::default().write(bar, E::BASE);
-        regs::NV_PFALCON_FBIF_TRANSCFG::alter(bar, E::BASE, |v| {
+        regs::NV_PFALCON_FBIF_CTL::alter(bar, &E::ID, |v| v.set_allow_phys_no_ctx(true));
+        regs::NV_PFALCON_FALCON_DMACTL::default().write(bar, &E::ID);
+        regs::NV_PFALCON_FBIF_TRANSCFG::alter(bar, &E::ID, |v| {
             v.set_target(FalconFbifTarget::CoherentSysmem)
                 .set_mem_type(FalconFbifMemType::Physical)
         });
@@ -517,7 +524,7 @@ impl<E: FalconEngine + 'static> Falcon<E> {
         // Set `BootVec` to start of non-secure code.
         regs::NV_PFALCON_FALCON_BOOTVEC::default()
             .set_value(fw.boot_addr())
-            .write(bar, E::BASE);
+            .write(bar, &E::ID);
 
         Ok(())
     }
@@ -538,27 +545,27 @@ impl<E: FalconEngine + 'static> Falcon<E> {
         if let Some(mbox0) = mbox0 {
             regs::NV_PFALCON_FALCON_MAILBOX0::default()
                 .set_value(mbox0)
-                .write(bar, E::BASE);
+                .write(bar, &E::ID);
         }
 
         if let Some(mbox1) = mbox1 {
             regs::NV_PFALCON_FALCON_MAILBOX1::default()
                 .set_value(mbox1)
-                .write(bar, E::BASE);
+                .write(bar, &E::ID);
         }
 
-        match regs::NV_PFALCON_FALCON_CPUCTL::read(bar, E::BASE).alias_en() {
+        match regs::NV_PFALCON_FALCON_CPUCTL::read(bar, &E::ID).alias_en() {
             true => regs::NV_PFALCON_FALCON_CPUCTL_ALIAS::default()
                 .set_startcpu(true)
-                .write(bar, E::BASE),
+                .write(bar, &E::ID),
             false => regs::NV_PFALCON_FALCON_CPUCTL::default()
                 .set_startcpu(true)
-                .write(bar, E::BASE),
+                .write(bar, &E::ID),
         }
 
         // TIMEOUT: arbitrarily large value, firmwares should complete in less than 2 seconds.
         util::wait_on(Delta::from_secs(2), || {
-            let r = regs::NV_PFALCON_FALCON_CPUCTL::read(bar, E::BASE);
+            let r = regs::NV_PFALCON_FALCON_CPUCTL::read(bar, &E::ID);
             if r.halted() {
                 Some(())
             } else {
@@ -567,8 +574,8 @@ impl<E: FalconEngine + 'static> Falcon<E> {
         })?;
 
         let (mbox0, mbox1) = (
-            regs::NV_PFALCON_FALCON_MAILBOX0::read(bar, E::BASE).value(),
-            regs::NV_PFALCON_FALCON_MAILBOX1::read(bar, E::BASE).value(),
+            regs::NV_PFALCON_FALCON_MAILBOX0::read(bar, &E::ID).value(),
+            regs::NV_PFALCON_FALCON_MAILBOX1::read(bar, &E::ID).value(),
         );
 
         Ok((mbox0, mbox1))
diff --git a/drivers/gpu/nova-core/falcon/gsp.rs b/drivers/gpu/nova-core/falcon/gsp.rs
index d622e9a64470..0db9f94036a6 100644
--- a/drivers/gpu/nova-core/falcon/gsp.rs
+++ b/drivers/gpu/nova-core/falcon/gsp.rs
@@ -2,23 +2,27 @@
 
 use crate::{
     driver::Bar0,
-    falcon::{Falcon, FalconEngine},
-    regs,
+    falcon::{Falcon, FalconEngine, PFalconBase},
+    regs::{self, macros::RegisterBase},
 };
 
 /// Type specifying the `Gsp` falcon engine. Cannot be instantiated.
 pub(crate) struct Gsp(());
 
-impl FalconEngine for Gsp {
+impl RegisterBase<PFalconBase> for Gsp {
     const BASE: usize = 0x00110000;
 }
 
+impl FalconEngine for Gsp {
+    const ID: Self = Gsp(());
+}
+
 impl Falcon<Gsp> {
     /// Clears the SWGEN0 bit in the Falcon's IRQ status clear register to
     /// allow GSP to signal CPU for processing new messages in message queue.
     pub(crate) fn clear_swgen0_intr(&self, bar: &Bar0) {
         regs::NV_PFALCON_FALCON_IRQSCLR::default()
             .set_swgen0(true)
-            .write(bar, Gsp::BASE);
+            .write(bar, &Gsp::ID);
     }
 }
diff --git a/drivers/gpu/nova-core/falcon/hal/ga102.rs b/drivers/gpu/nova-core/falcon/hal/ga102.rs
index 52c33d3f22a8..3fdacd19322d 100644
--- a/drivers/gpu/nova-core/falcon/hal/ga102.rs
+++ b/drivers/gpu/nova-core/falcon/hal/ga102.rs
@@ -16,15 +16,15 @@ use crate::util;
 use super::FalconHal;
 
 fn select_core_ga102<E: FalconEngine>(bar: &Bar0) -> Result {
-    let bcr_ctrl = regs::NV_PRISCV_RISCV_BCR_CTRL::read(bar, E::BASE);
+    let bcr_ctrl = regs::NV_PRISCV_RISCV_BCR_CTRL::read(bar, &E::ID);
     if bcr_ctrl.core_select() != PeregrineCoreSelect::Falcon {
         regs::NV_PRISCV_RISCV_BCR_CTRL::default()
             .set_core_select(PeregrineCoreSelect::Falcon)
-            .write(bar, E::BASE);
+            .write(bar, &E::ID);
 
         // TIMEOUT: falcon core should take less than 10ms to report being enabled.
         util::wait_on(Delta::from_millis(10), || {
-            let r = regs::NV_PRISCV_RISCV_BCR_CTRL::read(bar, E::BASE);
+            let r = regs::NV_PRISCV_RISCV_BCR_CTRL::read(bar, &E::ID);
             if r.valid() {
                 Some(())
             } else {
@@ -76,16 +76,16 @@ fn signature_reg_fuse_version_ga102(
 fn program_brom_ga102<E: FalconEngine>(bar: &Bar0, params: &FalconBromParams) -> Result {
     regs::NV_PFALCON2_FALCON_BROM_PARAADDR::default()
         .set_value(params.pkc_data_offset)
-        .write(bar, E::BASE);
+        .write(bar, &E::ID);
     regs::NV_PFALCON2_FALCON_BROM_ENGIDMASK::default()
         .set_value(u32::from(params.engine_id_mask))
-        .write(bar, E::BASE);
+        .write(bar, &E::ID);
     regs::NV_PFALCON2_FALCON_BROM_CURR_UCODE_ID::default()
         .set_ucode_id(params.ucode_id)
-        .write(bar, E::BASE);
+        .write(bar, &E::ID);
     regs::NV_PFALCON2_FALCON_MOD_SEL::default()
         .set_algo(FalconModSelAlgo::Rsa3k)
-        .write(bar, E::BASE);
+        .write(bar, &E::ID);
 
     Ok(())
 }
diff --git a/drivers/gpu/nova-core/falcon/sec2.rs b/drivers/gpu/nova-core/falcon/sec2.rs
index 5147d9e2a7fe..dbc486a712ff 100644
--- a/drivers/gpu/nova-core/falcon/sec2.rs
+++ b/drivers/gpu/nova-core/falcon/sec2.rs
@@ -1,10 +1,15 @@
 // SPDX-License-Identifier: GPL-2.0
 
-use crate::falcon::FalconEngine;
+use crate::falcon::{FalconEngine, PFalconBase};
+use crate::regs::macros::RegisterBase;
 
 /// Type specifying the `Sec2` falcon engine. Cannot be instantiated.
 pub(crate) struct Sec2(());
 
-impl FalconEngine for Sec2 {
+impl RegisterBase<PFalconBase> for Sec2 {
     const BASE: usize = 0x00840000;
 }
+
+impl FalconEngine for Sec2 {
+    const ID: Self = Sec2(());
+}
diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index c8f8adb24f6e..7730c5a475d5 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -5,11 +5,11 @@
 #![allow(non_camel_case_types)]
 
 #[macro_use]
-mod macros;
+pub(crate) mod macros;
 
 use crate::falcon::{
     DmaTrfCmdSize, FalconCoreRev, FalconCoreRevSubversion, FalconFbifMemType, FalconFbifTarget,
-    FalconModSelAlgo, FalconSecurityModel, PeregrineCoreSelect,
+    FalconModSelAlgo, FalconSecurityModel, PFalconBase, PeregrineCoreSelect,
 };
 use crate::gpu::{Architecture, Chipset};
 use kernel::prelude::*;
@@ -195,24 +195,24 @@ register!(NV_FUSE_OPT_FPF_GSP_UCODE1_VERSION @ 0x008241c0 {
 
 // PFALCON
 
-register!(NV_PFALCON_FALCON_IRQSCLR @ +0x00000004 {
+register!(NV_PFALCON_FALCON_IRQSCLR @ PFalconBase[0x00000004] {
     4:4     halt as bool;
     6:6     swgen0 as bool;
 });
 
-register!(NV_PFALCON_FALCON_MAILBOX0 @ +0x00000040 {
+register!(NV_PFALCON_FALCON_MAILBOX0 @ PFalconBase[0x00000040] {
     31:0    value as u32;
 });
 
-register!(NV_PFALCON_FALCON_MAILBOX1 @ +0x00000044 {
+register!(NV_PFALCON_FALCON_MAILBOX1 @ PFalconBase[0x00000044] {
     31:0    value as u32;
 });
 
-register!(NV_PFALCON_FALCON_RM @ +0x00000084 {
+register!(NV_PFALCON_FALCON_RM @ PFalconBase[0x00000084] {
     31:0    value as u32;
 });
 
-register!(NV_PFALCON_FALCON_HWCFG2 @ +0x000000f4 {
+register!(NV_PFALCON_FALCON_HWCFG2 @ PFalconBase[0x000000f4] {
     10:10   riscv as bool;
     12:12   mem_scrubbing as bool, "Set to 0 after memory scrubbing is completed";
     31:31   reset_ready as bool, "Signal indicating that reset is completed (GA102+)";
@@ -225,17 +225,17 @@ impl NV_PFALCON_FALCON_HWCFG2 {
     }
 }
 
-register!(NV_PFALCON_FALCON_CPUCTL @ +0x00000100 {
+register!(NV_PFALCON_FALCON_CPUCTL @ PFalconBase[0x00000100] {
     1:1     startcpu as bool;
     4:4     halted as bool;
     6:6     alias_en as bool;
 });
 
-register!(NV_PFALCON_FALCON_BOOTVEC @ +0x00000104 {
+register!(NV_PFALCON_FALCON_BOOTVEC @ PFalconBase[0x00000104] {
     31:0    value as u32;
 });
 
-register!(NV_PFALCON_FALCON_DMACTL @ +0x0000010c {
+register!(NV_PFALCON_FALCON_DMACTL @ PFalconBase[0x0000010c] {
     0:0     require_ctx as bool;
     1:1     dmem_scrubbing as bool;
     2:2     imem_scrubbing as bool;
@@ -243,15 +243,15 @@ register!(NV_PFALCON_FALCON_DMACTL @ +0x0000010c {
     7:7     secure_stat as bool;
 });
 
-register!(NV_PFALCON_FALCON_DMATRFBASE @ +0x00000110 {
+register!(NV_PFALCON_FALCON_DMATRFBASE @ PFalconBase[0x00000110] {
     31:0    base as u32;
 });
 
-register!(NV_PFALCON_FALCON_DMATRFMOFFS @ +0x00000114 {
+register!(NV_PFALCON_FALCON_DMATRFMOFFS @ PFalconBase[0x00000114] {
     23:0    offs as u32;
 });
 
-register!(NV_PFALCON_FALCON_DMATRFCMD @ +0x00000118 {
+register!(NV_PFALCON_FALCON_DMATRFCMD @ PFalconBase[0x00000118] {
     0:0     full as bool;
     1:1     idle as bool;
     3:2     sec as u8;
@@ -262,60 +262,60 @@ register!(NV_PFALCON_FALCON_DMATRFCMD @ +0x00000118 {
     16:16   set_dmtag as u8;
 });
 
-register!(NV_PFALCON_FALCON_DMATRFFBOFFS @ +0x0000011c {
+register!(NV_PFALCON_FALCON_DMATRFFBOFFS @ PFalconBase[0x0000011c] {
     31:0    offs as u32;
 });
 
-register!(NV_PFALCON_FALCON_DMATRFBASE1 @ +0x00000128 {
+register!(NV_PFALCON_FALCON_DMATRFBASE1 @ PFalconBase[0x00000128] {
     8:0     base as u16;
 });
 
-register!(NV_PFALCON_FALCON_HWCFG1 @ +0x0000012c {
+register!(NV_PFALCON_FALCON_HWCFG1 @ PFalconBase[0x0000012c] {
     3:0     core_rev as u8 ?=> FalconCoreRev, "Core revision";
     5:4     security_model as u8 ?=> FalconSecurityModel, "Security model";
     7:6     core_rev_subversion as u8 ?=> FalconCoreRevSubversion, "Core revision subversion";
 });
 
-register!(NV_PFALCON_FALCON_CPUCTL_ALIAS @ +0x00000130 {
+register!(NV_PFALCON_FALCON_CPUCTL_ALIAS @ PFalconBase[0x00000130] {
     1:1     startcpu as bool;
 });
 
 // Actually known as `NV_PSEC_FALCON_ENGINE` and `NV_PGSP_FALCON_ENGINE` depending on the falcon
 // instance.
-register!(NV_PFALCON_FALCON_ENGINE @ +0x000003c0 {
+register!(NV_PFALCON_FALCON_ENGINE @ PFalconBase[0x000003c0] {
     0:0     reset as bool;
 });
 
 // TODO[REGA]: this is an array of registers.
-register!(NV_PFALCON_FBIF_TRANSCFG @ +0x00000600 {
+register!(NV_PFALCON_FBIF_TRANSCFG @ PFalconBase[0x00000600] {
     1:0     target as u8 ?=> FalconFbifTarget;
     2:2     mem_type as bool => FalconFbifMemType;
 });
 
-register!(NV_PFALCON_FBIF_CTL @ +0x00000624 {
+register!(NV_PFALCON_FBIF_CTL @ PFalconBase[0x00000624] {
     7:7     allow_phys_no_ctx as bool;
 });
 
-register!(NV_PFALCON2_FALCON_MOD_SEL @ +0x00001180 {
+register!(NV_PFALCON2_FALCON_MOD_SEL @ PFalconBase[0x00001180] {
     7:0     algo as u8 ?=> FalconModSelAlgo;
 });
 
-register!(NV_PFALCON2_FALCON_BROM_CURR_UCODE_ID @ +0x00001198 {
+register!(NV_PFALCON2_FALCON_BROM_CURR_UCODE_ID @ PFalconBase[0x00001198] {
     7:0    ucode_id as u8;
 });
 
-register!(NV_PFALCON2_FALCON_BROM_ENGIDMASK @ +0x0000119c {
+register!(NV_PFALCON2_FALCON_BROM_ENGIDMASK @ PFalconBase[0x0000119c] {
     31:0    value as u32;
 });
 
 // TODO[REGA]: this is an array of registers.
-register!(NV_PFALCON2_FALCON_BROM_PARAADDR @ +0x00001210 {
+register!(NV_PFALCON2_FALCON_BROM_PARAADDR @ PFalconBase[0x00001210] {
     31:0    value as u32;
 });
 
 // PRISCV
 
-register!(NV_PRISCV_RISCV_BCR_CTRL @ +0x00001668 {
+register!(NV_PRISCV_RISCV_BCR_CTRL @ PFalconBase[0x00001668] {
     0:0     valid as bool;
     4:4     core_select as bool => PeregrineCoreSelect;
     8:8     br_fetch as bool;
diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 200fbe847571..005ee7afe279 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -10,6 +10,16 @@
 //! dedicated type for each register. Each such type comes with its own field accessors that can
 //! return an error if a field's value is invalid.
 
+/// Trait providing a base address to be added to the offset of a relative register to obtain
+/// its actual offset.
+///
+/// The `T` generic argument is used to distinguish which base to use, in case a type provides
+/// several bases. It is given to the `register!` macro to restrict the use of the register to
+/// implementors of this particular variant.
+pub(crate) trait RegisterBase<T> {
+    const BASE: usize;
+}
+
 /// Defines a dedicated type for a register with an absolute offset, including getter and setter
 /// methods for its fields and methods to read and write it from an `Io` region.
 ///
@@ -56,20 +66,6 @@
 /// The documentation strings are optional. If present, they will be added to the type's
 /// definition, or the field getter and setter methods they are attached to.
 ///
-/// Putting a `+` before the address of the register makes it relative to a base: the `read` and
-/// `write` methods take a `base` argument that is added to the specified address before access:
-///
-/// ```no_run
-/// register!(CPU_CTL @ +0x0000010, "CPU core control" {
-///    0:0     start as bool, "Start the CPU core";
-/// });
-///
-/// // Flip the `start` switch for the CPU core which base address is at `CPU_BASE`.
-/// let cpuctl = CPU_CTL::read(&bar, CPU_BASE);
-/// pr_info!("CPU CTL: {:#x}", cpuctl);
-/// cpuctl.set_start(true).write(&bar, CPU_BASE);
-/// ```
-///
 /// It is also possible to create a alias register by using the `=> ALIAS` syntax. This is useful
 /// for cases where a register's interpretation depends on the context:
 ///
@@ -85,6 +81,87 @@
 ///
 /// In this example, `SCRATCH_0_BOOT_STATUS` uses the same I/O address as `SCRATCH`, while also
 /// providing its own `completed` field.
+///
+/// ## Relative registers
+///
+/// A register can be defined as being accessible from a fixed offset of a provided base. For
+/// instance, imagine the following I/O space:
+///
+/// ```text
+///           +-----------------------------+
+///           |             ...             |
+///           |                             |
+///  0x100--->+------------CPU0-------------+
+///           |                             |
+///  0x110--->+-----------------------------+
+///           |           CPU_CTL           |
+///           +-----------------------------+
+///           |             ...             |
+///           |                             |
+///           |                             |
+///  0x200--->+------------CPU1-------------+
+///           |                             |
+///  0x210--->+-----------------------------+
+///           |           CPU_CTL           |
+///           +-----------------------------+
+///           |             ...             |
+///           +-----------------------------+
+/// ```
+///
+/// `CPU0` and `CPU1` both have a `CPU_CTL` register that starts at offset `0x10` of their I/O
+/// space segment. Since both instances of `CPU_CTL` share the same layout, we don't want to define
+/// them twice and would prefer a way to select which one to use from a single definition
+///
+/// This can be done using the `Base[Offset]` syntax when specifying the register's address.
+///
+/// `Base` is an arbitrary type (typically a ZST) to be used as a generic parameter of the
+/// [`RegisterBase`] trait to provide the base as a constant, i.e. each type providing a base for
+/// this register needs to implement `RegisterBase<Base>`. Here is the above example translated
+/// into code:
+///
+/// ```no_run
+/// // Type used to identify the base.
+/// pub(crate) struct CpuCtlBase;
+///
+/// // ZST describing `CPU0`.
+/// struct Cpu0;
+/// impl RegisterBase<CpuCtlBase> for Cpu0 {
+///     const BASE: usize = 0x100;
+/// }
+/// // Singleton of `CPU0` used to identify it.
+/// const CPU0: Cpu0 = Cpu0;
+///
+/// // ZST describing `CPU1`.
+/// struct Cpu1;
+/// impl RegisterBase<CpuCtlBase> for Cpu1 {
+///     const BASE: usize = 0x200;
+/// }
+/// // Singleton of `CPU1` used to identify it.
+/// const CPU1: Cpu1 = Cpu1;
+///
+/// // This makes `CPU_CTL` accessible from all implementors of `RegisterBase<CpuCtlBase>`.
+/// register!(CPU_CTL @ CpuCtlBase[0x10], "CPU core control" {
+///     0:0     start as bool, "Start the CPU core";
+/// });
+///
+/// // The `read`, `write` and `alter` methods of relative registers take an extra `base` argument
+/// // that is used to resolve its final address by adding its `BASE` to the offset of the
+/// // register.
+///
+/// // Start `CPU0`.
+/// CPU_CTL::alter(bar, &CPU0, |r| r.set_start(true));
+///
+/// // Start `CPU1`.
+/// CPU_CTL::alter(bar, &CPU1, |r| r.set_start(true));
+///
+/// // Aliases can also be defined for relative register.
+/// register!(CPU_CTL_ALIAS => CpuCtlBase[CPU_CTL], "Alias to CPU core control" {
+///     1:1     alias_start as bool, "Start the aliased CPU core";
+/// });
+///
+/// // Start the aliased `CPU0`.
+/// CPU_CTL_ALIAS::alter(bar, &CPU0, |r| r.set_alias_start(true));
+/// ```
 macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
     ($name:ident @ $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
@@ -98,16 +175,16 @@ macro_rules! register {
         register!(@io_fixed $name @ $alias::OFFSET);
     };
 
-    // Creates a register at a relative offset from a base address.
-    ($name:ident @ + $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
+    // Creates a register at a relative offset from a base address provider.
+    ($name:ident @ $base:ty [ $offset:literal ] $(, $comment:literal)? { $($fields:tt)* } ) => {
         register!(@core $name $(, $comment)? { $($fields)* } );
-        register!(@io_relative $name @ + $offset);
+        register!(@io_relative $name @ $base [ $offset ]);
     };
 
     // Creates an alias register of relative offset register `alias` with its own fields.
-    ($name:ident => + $alias:ident $(, $comment:literal)? { $($fields:tt)* } ) => {
+    ($name:ident => $base:ty [ $alias:ident ] $(, $comment:literal)? { $($fields:tt)* }) => {
         register!(@core $name $(, $comment)? { $($fields)* } );
-        register!(@io_relative $name @ + $alias::OFFSET);
+        register!(@io_relative $name @ $base [ $alias::OFFSET ]);
     };
 
     // All rules below are helpers.
@@ -380,39 +457,62 @@ macro_rules! register {
     };
 
     // Generates the IO accessors for a relative offset register.
-    (@io_relative $name:ident @ + $offset:literal) => {
+    (@io_relative $name:ident @ $base:ty [ $offset:expr ]) => {
         #[allow(dead_code)]
         impl $name {
             pub(crate) const OFFSET: usize = $offset;
 
+            /// Read the register from `io`, using the base address provided by `base` and adding
+            /// the register's offset to it.
             #[inline(always)]
-            pub(crate) fn read<const SIZE: usize, T>(
+            pub(crate) fn read<const SIZE: usize, T, B>(
                 io: &T,
-                base: usize,
+                #[allow(unused_variables)]
+                base: &B,
             ) -> Self where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
             {
-                Self(io.read32(base + $offset))
+                const OFFSET: usize = $name::OFFSET;
+
+                let value = io.read32(
+                    <B as crate::regs::macros::RegisterBase<$base>>::BASE + OFFSET
+                );
+
+                Self(value)
             }
 
+            /// Write the value contained in `self` to `io`, using the base address provided by
+            /// `base` and adding the register's offset to it.
             #[inline(always)]
-            pub(crate) fn write<const SIZE: usize, T>(
+            pub(crate) fn write<const SIZE: usize, T, B>(
                 self,
                 io: &T,
-                base: usize,
+                #[allow(unused_variables)]
+                base: &B,
             ) where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
             {
-                io.write32(self.0, base + $offset)
+                const OFFSET: usize = $name::OFFSET;
+
+                io.write32(
+                    self.0,
+                    <B as crate::regs::macros::RegisterBase<$base>>::BASE + OFFSET
+                );
             }
 
+            /// Read the register from `io`, using the base address provided by `base` and adding
+            /// the register's offset to it, then run `f` on its value to obtain a new value to
+            /// write back.
             #[inline(always)]
-            pub(crate) fn alter<const SIZE: usize, T, F>(
+            pub(crate) fn alter<const SIZE: usize, T, B, F>(
                 io: &T,
-                base: usize,
+                base: &B,
                 f: F,
             ) where
                 T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
                 F: ::core::ops::FnOnce(Self) -> Self,
             {
                 let reg = f(Self::read(io, base));

From e617f3a3702ca1ca2a498edbad0b12ceadc27321 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:21 +0900
Subject: [PATCH 0470/1455] gpu: nova-core: falcon: add distinct base address
 for PFALCON2

Falcon engines have two distinct register bases: `PFALCON` and
`PFALCON2`. So far we assumed that `PFALCON2` was located at `PFALCON +
0x1000` because that is the case of most engines, but there are
exceptions (NVDEC uses `0x1c00`).

Fix this shortcoming by leveraging the redesigned relative registers
definitions to assign a distinct `PFalcon2Base` base address to each
falcon engine.

Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-16-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/falcon.rs      |  7 ++++++-
 drivers/gpu/nova-core/falcon/gsp.rs  |  6 +++++-
 drivers/gpu/nova-core/falcon/sec2.rs |  6 +++++-
 drivers/gpu/nova-core/regs.rs        | 12 +++++++-----
 4 files changed, 23 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index 67265a0b5d7b..2ecdcc6b9b9d 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -278,11 +278,16 @@ impl From<bool> for FalconFbifMemType {
 /// Type used to represent the `PFALCON` registers address base for a given falcon engine.
 pub(crate) struct PFalconBase(());
 
+/// Type used to represent the `PFALCON2` registers address base for a given falcon engine.
+pub(crate) struct PFalcon2Base(());
+
 /// Trait defining the parameters of a given Falcon engine.
 ///
 /// Each engine provides one base for `PFALCON` and `PFALCON2` registers. The `ID` constant is used
 /// to identify a given Falcon instance with register I/O methods.
-pub(crate) trait FalconEngine: Sync + RegisterBase<PFalconBase> + Sized {
+pub(crate) trait FalconEngine:
+    Sync + RegisterBase<PFalconBase> + RegisterBase<PFalcon2Base> + Sized
+{
     /// Singleton of the engine, used to identify it with register I/O methods.
     const ID: Self;
 }
diff --git a/drivers/gpu/nova-core/falcon/gsp.rs b/drivers/gpu/nova-core/falcon/gsp.rs
index 0db9f94036a6..f17599cb49fa 100644
--- a/drivers/gpu/nova-core/falcon/gsp.rs
+++ b/drivers/gpu/nova-core/falcon/gsp.rs
@@ -2,7 +2,7 @@
 
 use crate::{
     driver::Bar0,
-    falcon::{Falcon, FalconEngine, PFalconBase},
+    falcon::{Falcon, FalconEngine, PFalcon2Base, PFalconBase},
     regs::{self, macros::RegisterBase},
 };
 
@@ -13,6 +13,10 @@ impl RegisterBase<PFalconBase> for Gsp {
     const BASE: usize = 0x00110000;
 }
 
+impl RegisterBase<PFalcon2Base> for Gsp {
+    const BASE: usize = 0x00111000;
+}
+
 impl FalconEngine for Gsp {
     const ID: Self = Gsp(());
 }
diff --git a/drivers/gpu/nova-core/falcon/sec2.rs b/drivers/gpu/nova-core/falcon/sec2.rs
index dbc486a712ff..815786c8480d 100644
--- a/drivers/gpu/nova-core/falcon/sec2.rs
+++ b/drivers/gpu/nova-core/falcon/sec2.rs
@@ -1,6 +1,6 @@
 // SPDX-License-Identifier: GPL-2.0
 
-use crate::falcon::{FalconEngine, PFalconBase};
+use crate::falcon::{FalconEngine, PFalcon2Base, PFalconBase};
 use crate::regs::macros::RegisterBase;
 
 /// Type specifying the `Sec2` falcon engine. Cannot be instantiated.
@@ -10,6 +10,10 @@ impl RegisterBase<PFalconBase> for Sec2 {
     const BASE: usize = 0x00840000;
 }
 
+impl RegisterBase<PFalcon2Base> for Sec2 {
+    const BASE: usize = 0x00841000;
+}
+
 impl FalconEngine for Sec2 {
     const ID: Self = Sec2(());
 }
diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index 7730c5a475d5..e55525916bad 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -9,7 +9,7 @@ pub(crate) mod macros;
 
 use crate::falcon::{
     DmaTrfCmdSize, FalconCoreRev, FalconCoreRevSubversion, FalconFbifMemType, FalconFbifTarget,
-    FalconModSelAlgo, FalconSecurityModel, PFalconBase, PeregrineCoreSelect,
+    FalconModSelAlgo, FalconSecurityModel, PFalcon2Base, PFalconBase, PeregrineCoreSelect,
 };
 use crate::gpu::{Architecture, Chipset};
 use kernel::prelude::*;
@@ -296,20 +296,22 @@ register!(NV_PFALCON_FBIF_CTL @ PFalconBase[0x00000624] {
     7:7     allow_phys_no_ctx as bool;
 });
 
-register!(NV_PFALCON2_FALCON_MOD_SEL @ PFalconBase[0x00001180] {
+/* PFALCON2 */
+
+register!(NV_PFALCON2_FALCON_MOD_SEL @ PFalcon2Base[0x00000180] {
     7:0     algo as u8 ?=> FalconModSelAlgo;
 });
 
-register!(NV_PFALCON2_FALCON_BROM_CURR_UCODE_ID @ PFalconBase[0x00001198] {
+register!(NV_PFALCON2_FALCON_BROM_CURR_UCODE_ID @ PFalcon2Base[0x00000198] {
     7:0    ucode_id as u8;
 });
 
-register!(NV_PFALCON2_FALCON_BROM_ENGIDMASK @ PFalconBase[0x0000119c] {
+register!(NV_PFALCON2_FALCON_BROM_ENGIDMASK @ PFalcon2Base[0x0000019c] {
     31:0    value as u32;
 });
 
 // TODO[REGA]: this is an array of registers.
-register!(NV_PFALCON2_FALCON_BROM_PARAADDR @ PFalconBase[0x00001210] {
+register!(NV_PFALCON2_FALCON_BROM_PARAADDR @ PFalcon2Base[0x00000210] {
     31:0    value as u32;
 });
 

From 20ed4a8695b277c296e3a30306ef8551903e1e04 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:22 +0900
Subject: [PATCH 0471/1455] gpu: nova-core: register: add support for register
 arrays

Having registers that can be interpreted identically in a contiguous I/O
area (or at least, following a given stride) is a common way to organize
registers, and is used by NVIDIA hardware. Thus, add a way to simply and
safely declare such a layout using the register!() macro.

Build-time bound-checking is effective for array accesses performed with
a constant. For cases where the index cannot be known at compile time,
`try_` variants of the accessors are also made available that return
`EINVAL` if the access is out-of-bounds.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-17-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/gpu.rs         |   2 +-
 drivers/gpu/nova-core/regs.rs        |  15 ++-
 drivers/gpu/nova-core/regs/macros.rs | 195 +++++++++++++++++++++++++++
 3 files changed, 204 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index b5c9786619a9..8caecaf7dfb4 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -221,7 +221,7 @@ impl Gpu {
         fwsec_frts.run(dev, falcon, bar)?;
 
         // SCRATCH_E contains the error code for FWSEC-FRTS.
-        let frts_status = regs::NV_PBUS_SW_SCRATCH_0E::read(bar).frts_err_code();
+        let frts_status = regs::NV_PBUS_SW_SCRATCH_0E_FRTS_ERR::read(bar).frts_err_code();
         if frts_status != 0 {
             dev_err!(
                 dev,
diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index e55525916bad..c45de208055d 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -45,8 +45,10 @@ impl NV_PMC_BOOT_0 {
 
 // PBUS
 
-// TODO[REGA]: this is an array of registers.
-register!(NV_PBUS_SW_SCRATCH_0E@0x00001438  {
+register!(NV_PBUS_SW_SCRATCH @ 0x00001400[64]  {});
+
+register!(NV_PBUS_SW_SCRATCH_0E_FRTS_ERR => NV_PBUS_SW_SCRATCH[0xe],
+    "scratch register 0xe used as FRTS firmware error code" {
     31:16   frts_err_code as u16;
 });
 
@@ -124,13 +126,12 @@ register!(NV_PGC6_AON_SECURE_SCRATCH_GROUP_05_PRIV_LEVEL_MASK @ 0x00118128,
     0:0     read_protection_level0 as bool, "Set after FWSEC lowers its protection level";
 });
 
-// TODO[REGA]: This is an array of registers.
-register!(NV_PGC6_AON_SECURE_SCRATCH_GROUP_05 @ 0x00118234 {
-    31:0    value as u32;
-});
+// OpenRM defines this as a register array, but doesn't specify its size and only uses its first
+// element. Be conservative until we know the actual size or need to use more registers.
+register!(NV_PGC6_AON_SECURE_SCRATCH_GROUP_05 @ 0x00118234[1] {});
 
 register!(
-    NV_PGC6_AON_SECURE_SCRATCH_GROUP_05_0_GFW_BOOT => NV_PGC6_AON_SECURE_SCRATCH_GROUP_05,
+    NV_PGC6_AON_SECURE_SCRATCH_GROUP_05_0_GFW_BOOT => NV_PGC6_AON_SECURE_SCRATCH_GROUP_05[0],
     "Scratch group 05 register 0 used as GFW boot progress indicator" {
         7:0    progress as u8, "Progress of GFW boot (0xff means completed)";
     }
diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index 005ee7afe279..f38177065c25 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -162,6 +162,57 @@ pub(crate) trait RegisterBase<T> {
 /// // Start the aliased `CPU0`.
 /// CPU_CTL_ALIAS::alter(bar, &CPU0, |r| r.set_alias_start(true));
 /// ```
+///
+/// ## Arrays of registers
+///
+/// Some I/O areas contain consecutive values that can be interpreted in the same way. These areas
+/// can be defined as an array of identical registers, allowing them to be accessed by index with
+/// compile-time or runtime bound checking. Simply define their address as `Address[Size]`, and add
+/// an `idx` parameter to their `read`, `write` and `alter` methods:
+///
+/// ```no_run
+/// # fn no_run() -> Result<(), Error> {
+/// # fn get_scratch_idx() -> usize {
+/// #   0x15
+/// # }
+/// // Array of 64 consecutive registers with the same layout starting at offset `0x80`.
+/// register!(SCRATCH @ 0x00000080[64], "Scratch registers" {
+///     31:0    value as u32;
+/// });
+///
+/// // Read scratch register 0, i.e. I/O address `0x80`.
+/// let scratch_0 = SCRATCH::read(bar, 0).value();
+/// // Read scratch register 15, i.e. I/O address `0x80 + (15 * 4)`.
+/// let scratch_15 = SCRATCH::read(bar, 15).value();
+///
+/// // This is out of bounds and won't build.
+/// // let scratch_128 = SCRATCH::read(bar, 128).value();
+///
+/// // Runtime-obtained array index.
+/// let scratch_idx = get_scratch_idx();
+/// // Access on a runtime index returns an error if it is out-of-bounds.
+/// let some_scratch = SCRATCH::try_read(bar, scratch_idx)?.value();
+///
+/// // Alias to a particular register in an array.
+/// // Here `SCRATCH[8]` is used to convey the firmware exit code.
+/// register!(FIRMWARE_STATUS => SCRATCH[8], "Firmware exit status code" {
+///     7:0     status as u8;
+/// });
+///
+/// let status = FIRMWARE_STATUS::read(bar).status();
+///
+/// // Non-contiguous register arrays can be defined by adding a stride parameter.
+/// // Here, each of the 16 registers of the array are separated by 8 bytes, meaning that the
+/// // registers of the two declarations below are interleaved.
+/// register!(SCRATCH_INTERLEAVED_0 @ 0x000000c0[16 ; 8], "Scratch registers bank 0" {
+///     31:0    value as u32;
+/// });
+/// register!(SCRATCH_INTERLEAVED_1 @ 0x000000c4[16 ; 8], "Scratch registers bank 1" {
+///     31:0    value as u32;
+/// });
+/// # Ok(())
+/// # }
+/// ```
 macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
     ($name:ident @ $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
@@ -187,6 +238,35 @@ macro_rules! register {
         register!(@io_relative $name @ $base [ $alias::OFFSET ]);
     };
 
+    // Creates an array of registers at a fixed offset of the MMIO space.
+    (
+        $name:ident @ $offset:literal [ $size:expr ; $stride:expr ] $(, $comment:literal)? {
+            $($fields:tt)*
+        }
+    ) => {
+        static_assert!(::core::mem::size_of::<u32>() <= $stride);
+        register!(@core $name $(, $comment)? { $($fields)* } );
+        register!(@io_array $name @ $offset [ $size ; $stride ]);
+    };
+
+    // Shortcut for contiguous array of registers (stride == size of element).
+    (
+        $name:ident @ $offset:literal [ $size:expr ] $(, $comment:literal)? {
+            $($fields:tt)*
+        }
+    ) => {
+        register!($name @ $offset [ $size ; ::core::mem::size_of::<u32>() ] $(, $comment)? {
+            $($fields)*
+        } );
+    };
+
+    // Creates an alias of register `idx` of array of registers `alias` with its own fields.
+    ($name:ident => $alias:ident [ $idx:expr ] $(, $comment:literal)? { $($fields:tt)* }) => {
+        static_assert!($idx < $alias::SIZE);
+        register!(@core $name $(, $comment)? { $($fields)* } );
+        register!(@io_fixed $name @ $alias::OFFSET + $idx * $alias::STRIDE );
+    };
+
     // All rules below are helpers.
 
     // Defines the wrapper `$name` type, as well as its relevant implementations (`Debug`,
@@ -520,4 +600,119 @@ macro_rules! register {
             }
         }
     };
+
+    // Generates the IO accessors for an array of registers.
+    (@io_array $name:ident @ $offset:literal [ $size:expr ; $stride:expr ]) => {
+        #[allow(dead_code)]
+        impl $name {
+            pub(crate) const OFFSET: usize = $offset;
+            pub(crate) const SIZE: usize = $size;
+            pub(crate) const STRIDE: usize = $stride;
+
+            /// Read the array register at index `idx` from its address in `io`.
+            #[inline(always)]
+            pub(crate) fn read<const SIZE: usize, T>(
+                io: &T,
+                idx: usize,
+            ) -> Self where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+            {
+                build_assert!(idx < Self::SIZE);
+
+                let offset = Self::OFFSET + (idx * Self::STRIDE);
+                let value = io.read32(offset);
+
+                Self(value)
+            }
+
+            /// Write the value contained in `self` to the array register with index `idx` in `io`.
+            #[inline(always)]
+            pub(crate) fn write<const SIZE: usize, T>(
+                self,
+                io: &T,
+                idx: usize
+            ) where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+            {
+                build_assert!(idx < Self::SIZE);
+
+                let offset = Self::OFFSET + (idx * Self::STRIDE);
+
+                io.write32(self.0, offset);
+            }
+
+            /// Read the array register at index `idx` in `io` and run `f` on its value to obtain a
+            /// new value to write back.
+            #[inline(always)]
+            pub(crate) fn alter<const SIZE: usize, T, F>(
+                io: &T,
+                idx: usize,
+                f: F,
+            ) where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                F: ::core::ops::FnOnce(Self) -> Self,
+            {
+                let reg = f(Self::read(io, idx));
+                reg.write(io, idx);
+            }
+
+            /// Read the array register at index `idx` from its address in `io`.
+            ///
+            /// The validity of `idx` is checked at run-time, and `EINVAL` is returned is the
+            /// access was out-of-bounds.
+            #[inline(always)]
+            pub(crate) fn try_read<const SIZE: usize, T>(
+                io: &T,
+                idx: usize,
+            ) -> ::kernel::error::Result<Self> where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+            {
+                if idx < Self::SIZE {
+                    Ok(Self::read(io, idx))
+                } else {
+                    Err(EINVAL)
+                }
+            }
+
+            /// Write the value contained in `self` to the array register with index `idx` in `io`.
+            ///
+            /// The validity of `idx` is checked at run-time, and `EINVAL` is returned is the
+            /// access was out-of-bounds.
+            #[inline(always)]
+            pub(crate) fn try_write<const SIZE: usize, T>(
+                self,
+                io: &T,
+                idx: usize,
+            ) -> ::kernel::error::Result where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+            {
+                if idx < Self::SIZE {
+                    Ok(self.write(io, idx))
+                } else {
+                    Err(EINVAL)
+                }
+            }
+
+            /// Read the array register at index `idx` in `io` and run `f` on its value to obtain a
+            /// new value to write back.
+            ///
+            /// The validity of `idx` is checked at run-time, and `EINVAL` is returned is the
+            /// access was out-of-bounds.
+            #[inline(always)]
+            pub(crate) fn try_alter<const SIZE: usize, T, F>(
+                io: &T,
+                idx: usize,
+                f: F,
+            ) -> ::kernel::error::Result where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                F: ::core::ops::FnOnce(Self) -> Self,
+            {
+                if idx < Self::SIZE {
+                    Ok(Self::alter(io, idx, f))
+                } else {
+                    Err(EINVAL)
+                }
+            }
+        }
+    };
 }

From ec2f6c81d2a16a93d882488062e8e09e2f1d6865 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:23 +0900
Subject: [PATCH 0472/1455] gpu: nova-core: falcon: use register arrays for
 FUSE registers

FUSE registers are an array of 16 consecutive registers. Use the
newly available register array feature to define them properly and
improve the code using them.

Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-18-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/falcon/hal/ga102.rs | 33 +++++++++++------------
 drivers/gpu/nova-core/regs.rs             |  8 +++---
 2 files changed, 20 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/nova-core/falcon/hal/ga102.rs b/drivers/gpu/nova-core/falcon/hal/ga102.rs
index 3fdacd19322d..13c945fd6d6b 100644
--- a/drivers/gpu/nova-core/falcon/hal/ga102.rs
+++ b/drivers/gpu/nova-core/falcon/hal/ga102.rs
@@ -42,35 +42,32 @@ fn signature_reg_fuse_version_ga102(
     engine_id_mask: u16,
     ucode_id: u8,
 ) -> Result<u32> {
-    // TODO[REGA]: The ucode fuse versions are contained in the
-    // FUSE_OPT_FPF_<ENGINE>_UCODE<X>_VERSION registers, which are an array. Our register
-    // definition macros do not allow us to manage them properly, so we need to hardcode their
-    // addresses for now. Clean this up once we support register arrays.
+    const NV_FUSE_OPT_FPF_SIZE: u8 = regs::NV_FUSE_OPT_FPF_SIZE as u8;
 
     // Each engine has 16 ucode version registers numbered from 1 to 16.
-    if ucode_id == 0 || ucode_id > 16 {
-        dev_err!(dev, "invalid ucode id {:#x}", ucode_id);
-        return Err(EINVAL);
-    }
+    let ucode_idx = match ucode_id {
+        1..=NV_FUSE_OPT_FPF_SIZE => (ucode_id - 1) as usize,
+        _ => {
+            dev_err!(dev, "invalid ucode id {:#x}", ucode_id);
+            return Err(EINVAL);
+        }
+    };
 
-    // Base address of the FUSE registers array corresponding to the engine.
-    let reg_fuse_base = if engine_id_mask & 0x0001 != 0 {
-        regs::NV_FUSE_OPT_FPF_SEC2_UCODE1_VERSION::OFFSET
+    // `ucode_idx` is guaranteed to be in the range [0..15], making the `read` calls provable valid
+    // at build-time.
+    let reg_fuse_version = if engine_id_mask & 0x0001 != 0 {
+        regs::NV_FUSE_OPT_FPF_SEC2_UCODE1_VERSION::read(bar, ucode_idx).data()
     } else if engine_id_mask & 0x0004 != 0 {
-        regs::NV_FUSE_OPT_FPF_NVDEC_UCODE1_VERSION::OFFSET
+        regs::NV_FUSE_OPT_FPF_NVDEC_UCODE1_VERSION::read(bar, ucode_idx).data()
     } else if engine_id_mask & 0x0400 != 0 {
-        regs::NV_FUSE_OPT_FPF_GSP_UCODE1_VERSION::OFFSET
+        regs::NV_FUSE_OPT_FPF_GSP_UCODE1_VERSION::read(bar, ucode_idx).data()
     } else {
         dev_err!(dev, "unexpected engine_id_mask {:#x}", engine_id_mask);
         return Err(EINVAL);
     };
 
-    // Read `reg_fuse_base[ucode_id - 1]`.
-    let reg_fuse_version =
-        bar.read32(reg_fuse_base + ((ucode_id - 1) as usize * core::mem::size_of::<u32>()));
-
     // TODO[NUMM]: replace with `last_set_bit` once it lands.
-    Ok(u32::BITS - reg_fuse_version.leading_zeros())
+    Ok(u16::BITS - reg_fuse_version.leading_zeros())
 }
 
 fn program_brom_ga102<E: FalconEngine>(bar: &Bar0, params: &FalconBromParams) -> Result {
diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index c45de208055d..4ae2a91bcce8 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -182,15 +182,17 @@ impl NV_PDISP_VGA_WORKSPACE_BASE {
 
 // FUSE
 
-register!(NV_FUSE_OPT_FPF_NVDEC_UCODE1_VERSION @ 0x00824100 {
+pub(crate) const NV_FUSE_OPT_FPF_SIZE: usize = 16;
+
+register!(NV_FUSE_OPT_FPF_NVDEC_UCODE1_VERSION @ 0x00824100[NV_FUSE_OPT_FPF_SIZE] {
     15:0    data as u16;
 });
 
-register!(NV_FUSE_OPT_FPF_SEC2_UCODE1_VERSION @ 0x00824140 {
+register!(NV_FUSE_OPT_FPF_SEC2_UCODE1_VERSION @ 0x00824140[NV_FUSE_OPT_FPF_SIZE] {
     15:0    data as u16;
 });
 
-register!(NV_FUSE_OPT_FPF_GSP_UCODE1_VERSION @ 0x008241c0 {
+register!(NV_FUSE_OPT_FPF_GSP_UCODE1_VERSION @ 0x008241c0[NV_FUSE_OPT_FPF_SIZE] {
     15:0    data as u16;
 });
 

From 0988099646cfc6c72a4448cad39d4ee22ad457a7 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 18 Jul 2025 16:26:24 +0900
Subject: [PATCH 0473/1455] gpu: nova-core: register: add support for relative
 array registers

Add support for declaring arrays of registers available from a variable
base. This is effectively a combination of the relative and array
registers features.

nova-core does not make much use of this yet, but it will become helpful
to have for GSP boot.

Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250718-nova-regs-v2-19-7b6a762aa1cd@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 Documentation/gpu/nova/core/todo.rst      |   1 -
 drivers/gpu/nova-core/falcon.rs           |   2 +-
 drivers/gpu/nova-core/falcon/hal/ga102.rs |   2 +-
 drivers/gpu/nova-core/regs.rs             |   8 +-
 drivers/gpu/nova-core/regs/macros.rs      | 242 ++++++++++++++++++++++
 5 files changed, 248 insertions(+), 7 deletions(-)

diff --git a/Documentation/gpu/nova/core/todo.rst b/Documentation/gpu/nova/core/todo.rst
index a1d12c1b289d..48b20656dcb1 100644
--- a/Documentation/gpu/nova/core/todo.rst
+++ b/Documentation/gpu/nova/core/todo.rst
@@ -131,7 +131,6 @@ crate so it can be used by other components as well.
 
 Features desired before this happens:
 
-* Arrays of registers with build-time index validation,
 * Make I/O optional I/O (for field values that are not registers),
 * Support other sizes than `u32`,
 * Allow visibility control for registers and individual fields,
diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index 2ecdcc6b9b9d..d235a6f9efca 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -516,7 +516,7 @@ impl<E: FalconEngine + 'static> Falcon<E> {
     pub(crate) fn dma_load<F: FalconFirmware<Target = E>>(&self, bar: &Bar0, fw: &F) -> Result {
         regs::NV_PFALCON_FBIF_CTL::alter(bar, &E::ID, |v| v.set_allow_phys_no_ctx(true));
         regs::NV_PFALCON_FALCON_DMACTL::default().write(bar, &E::ID);
-        regs::NV_PFALCON_FBIF_TRANSCFG::alter(bar, &E::ID, |v| {
+        regs::NV_PFALCON_FBIF_TRANSCFG::alter(bar, &E::ID, 0, |v| {
             v.set_target(FalconFbifTarget::CoherentSysmem)
                 .set_mem_type(FalconFbifMemType::Physical)
         });
diff --git a/drivers/gpu/nova-core/falcon/hal/ga102.rs b/drivers/gpu/nova-core/falcon/hal/ga102.rs
index 13c945fd6d6b..0b1cbe7853b3 100644
--- a/drivers/gpu/nova-core/falcon/hal/ga102.rs
+++ b/drivers/gpu/nova-core/falcon/hal/ga102.rs
@@ -73,7 +73,7 @@ fn signature_reg_fuse_version_ga102(
 fn program_brom_ga102<E: FalconEngine>(bar: &Bar0, params: &FalconBromParams) -> Result {
     regs::NV_PFALCON2_FALCON_BROM_PARAADDR::default()
         .set_value(params.pkc_data_offset)
-        .write(bar, &E::ID);
+        .write(bar, &E::ID, 0);
     regs::NV_PFALCON2_FALCON_BROM_ENGIDMASK::default()
         .set_value(u32::from(params.engine_id_mask))
         .write(bar, &E::ID);
diff --git a/drivers/gpu/nova-core/regs.rs b/drivers/gpu/nova-core/regs.rs
index 4ae2a91bcce8..206dab2e1335 100644
--- a/drivers/gpu/nova-core/regs.rs
+++ b/drivers/gpu/nova-core/regs.rs
@@ -289,8 +289,7 @@ register!(NV_PFALCON_FALCON_ENGINE @ PFalconBase[0x000003c0] {
     0:0     reset as bool;
 });
 
-// TODO[REGA]: this is an array of registers.
-register!(NV_PFALCON_FBIF_TRANSCFG @ PFalconBase[0x00000600] {
+register!(NV_PFALCON_FBIF_TRANSCFG @ PFalconBase[0x00000600[8]] {
     1:0     target as u8 ?=> FalconFbifTarget;
     2:2     mem_type as bool => FalconFbifMemType;
 });
@@ -313,8 +312,9 @@ register!(NV_PFALCON2_FALCON_BROM_ENGIDMASK @ PFalcon2Base[0x0000019c] {
     31:0    value as u32;
 });
 
-// TODO[REGA]: this is an array of registers.
-register!(NV_PFALCON2_FALCON_BROM_PARAADDR @ PFalcon2Base[0x00000210] {
+// OpenRM defines this as a register array, but doesn't specify its size and only uses its first
+// element. Be conservative until we know the actual size or need to use more registers.
+register!(NV_PFALCON2_FALCON_BROM_PARAADDR @ PFalcon2Base[0x00000210[1]] {
     31:0    value as u32;
 });
 
diff --git a/drivers/gpu/nova-core/regs/macros.rs b/drivers/gpu/nova-core/regs/macros.rs
index f38177065c25..754c14ee7f40 100644
--- a/drivers/gpu/nova-core/regs/macros.rs
+++ b/drivers/gpu/nova-core/regs/macros.rs
@@ -213,6 +213,74 @@ pub(crate) trait RegisterBase<T> {
 /// # Ok(())
 /// # }
 /// ```
+///
+/// ## Relative arrays of registers
+///
+/// Combining the two features described in the sections above, arrays of registers accessible from
+/// a base can also be defined:
+///
+/// ```no_run
+/// # fn no_run() -> Result<(), Error> {
+/// # fn get_scratch_idx() -> usize {
+/// #   0x15
+/// # }
+/// // Type used as parameter of `RegisterBase` to specify the base.
+/// pub(crate) struct CpuCtlBase;
+///
+/// // ZST describing `CPU0`.
+/// struct Cpu0;
+/// impl RegisterBase<CpuCtlBase> for Cpu0 {
+///     const BASE: usize = 0x100;
+/// }
+/// // Singleton of `CPU0` used to identify it.
+/// const CPU0: Cpu0 = Cpu0;
+///
+/// // ZST describing `CPU1`.
+/// struct Cpu1;
+/// impl RegisterBase<CpuCtlBase> for Cpu1 {
+///     const BASE: usize = 0x200;
+/// }
+/// // Singleton of `CPU1` used to identify it.
+/// const CPU1: Cpu1 = Cpu1;
+///
+/// // 64 per-cpu scratch registers, arranged as an contiguous array.
+/// register!(CPU_SCRATCH @ CpuCtlBase[0x00000080[64]], "Per-CPU scratch registers" {
+///     31:0    value as u32;
+/// });
+///
+/// let cpu0_scratch_0 = CPU_SCRATCH::read(bar, &Cpu0, 0).value();
+/// let cpu1_scratch_15 = CPU_SCRATCH::read(bar, &Cpu1, 15).value();
+///
+/// // This won't build.
+/// // let cpu0_scratch_128 = CPU_SCRATCH::read(bar, &Cpu0, 128).value();
+///
+/// // Runtime-obtained array index.
+/// let scratch_idx = get_scratch_idx();
+/// // Access on a runtime value returns an error if it is out-of-bounds.
+/// let cpu0_some_scratch = CPU_SCRATCH::try_read(bar, &Cpu0, scratch_idx)?.value();
+///
+/// // `SCRATCH[8]` is used to convey the firmware exit code.
+/// register!(CPU_FIRMWARE_STATUS => CpuCtlBase[CPU_SCRATCH[8]],
+///     "Per-CPU firmware exit status code" {
+///     7:0     status as u8;
+/// });
+///
+/// let cpu0_status = CPU_FIRMWARE_STATUS::read(bar, &Cpu0).status();
+///
+/// // Non-contiguous register arrays can be defined by adding a stride parameter.
+/// // Here, each of the 16 registers of the array are separated by 8 bytes, meaning that the
+/// // registers of the two declarations below are interleaved.
+/// register!(CPU_SCRATCH_INTERLEAVED_0 @ CpuCtlBase[0x00000d00[16 ; 8]],
+///           "Scratch registers bank 0" {
+///     31:0    value as u32;
+/// });
+/// register!(CPU_SCRATCH_INTERLEAVED_1 @ CpuCtlBase[0x00000d04[16 ; 8]],
+///           "Scratch registers bank 1" {
+///     31:0    value as u32;
+/// });
+/// # Ok(())
+/// # }
+/// ```
 macro_rules! register {
     // Creates a register at a fixed offset of the MMIO space.
     ($name:ident @ $offset:literal $(, $comment:literal)? { $($fields:tt)* } ) => {
@@ -260,7 +328,41 @@ macro_rules! register {
         } );
     };
 
+    // Creates an array of registers at a relative offset from a base address provider.
+    (
+        $name:ident @ $base:ty [ $offset:literal [ $size:expr ; $stride:expr ] ]
+            $(, $comment:literal)? { $($fields:tt)* }
+    ) => {
+        static_assert!(::core::mem::size_of::<u32>() <= $stride);
+        register!(@core $name $(, $comment)? { $($fields)* } );
+        register!(@io_relative_array $name @ $base [ $offset [ $size ; $stride ] ]);
+    };
+
+    // Shortcut for contiguous array of relative registers (stride == size of element).
+    (
+        $name:ident @ $base:ty [ $offset:literal [ $size:expr ] ] $(, $comment:literal)? {
+            $($fields:tt)*
+        }
+    ) => {
+        register!($name @ $base [ $offset [ $size ; ::core::mem::size_of::<u32>() ] ]
+            $(, $comment)? { $($fields)* } );
+    };
+
+    // Creates an alias of register `idx` of relative array of registers `alias` with its own
+    // fields.
+    (
+        $name:ident => $base:ty [ $alias:ident [ $idx:expr ] ] $(, $comment:literal)? {
+            $($fields:tt)*
+        }
+    ) => {
+        static_assert!($idx < $alias::SIZE);
+        register!(@core $name $(, $comment)? { $($fields)* } );
+        register!(@io_relative $name @ $base [ $alias::OFFSET + $idx * $alias::STRIDE ] );
+    };
+
     // Creates an alias of register `idx` of array of registers `alias` with its own fields.
+    // This rule belongs to the (non-relative) register arrays set, but needs to be put last
+    // to avoid it being interpreted in place of the relative register array alias rule.
     ($name:ident => $alias:ident [ $idx:expr ] $(, $comment:literal)? { $($fields:tt)* }) => {
         static_assert!($idx < $alias::SIZE);
         register!(@core $name $(, $comment)? { $($fields)* } );
@@ -715,4 +817,144 @@ macro_rules! register {
             }
         }
     };
+
+    // Generates the IO accessors for an array of relative registers.
+    (
+        @io_relative_array $name:ident @ $base:ty
+            [ $offset:literal [ $size:expr ; $stride:expr ] ]
+    ) => {
+        #[allow(dead_code)]
+        impl $name {
+            pub(crate) const OFFSET: usize = $offset;
+            pub(crate) const SIZE: usize = $size;
+            pub(crate) const STRIDE: usize = $stride;
+
+            /// Read the array register at index `idx` from `io`, using the base address provided
+            /// by `base` and adding the register's offset to it.
+            #[inline(always)]
+            pub(crate) fn read<const SIZE: usize, T, B>(
+                io: &T,
+                #[allow(unused_variables)]
+                base: &B,
+                idx: usize,
+            ) -> Self where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
+            {
+                build_assert!(idx < Self::SIZE);
+
+                let offset = <B as crate::regs::macros::RegisterBase<$base>>::BASE +
+                    Self::OFFSET + (idx * Self::STRIDE);
+                let value = io.read32(offset);
+
+                Self(value)
+            }
+
+            /// Write the value contained in `self` to `io`, using the base address provided by
+            /// `base` and adding the offset of array register `idx` to it.
+            #[inline(always)]
+            pub(crate) fn write<const SIZE: usize, T, B>(
+                self,
+                io: &T,
+                #[allow(unused_variables)]
+                base: &B,
+                idx: usize
+            ) where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
+            {
+                build_assert!(idx < Self::SIZE);
+
+                let offset = <B as crate::regs::macros::RegisterBase<$base>>::BASE +
+                    Self::OFFSET + (idx * Self::STRIDE);
+
+                io.write32(self.0, offset);
+            }
+
+            /// Read the array register at index `idx` from `io`, using the base address provided
+            /// by `base` and adding the register's offset to it, then run `f` on its value to
+            /// obtain a new value to write back.
+            #[inline(always)]
+            pub(crate) fn alter<const SIZE: usize, T, B, F>(
+                io: &T,
+                base: &B,
+                idx: usize,
+                f: F,
+            ) where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
+                F: ::core::ops::FnOnce(Self) -> Self,
+            {
+                let reg = f(Self::read(io, base, idx));
+                reg.write(io, base, idx);
+            }
+
+            /// Read the array register at index `idx` from `io`, using the base address provided
+            /// by `base` and adding the register's offset to it.
+            ///
+            /// The validity of `idx` is checked at run-time, and `EINVAL` is returned is the
+            /// access was out-of-bounds.
+            #[inline(always)]
+            pub(crate) fn try_read<const SIZE: usize, T, B>(
+                io: &T,
+                base: &B,
+                idx: usize,
+            ) -> ::kernel::error::Result<Self> where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
+            {
+                if idx < Self::SIZE {
+                    Ok(Self::read(io, base, idx))
+                } else {
+                    Err(EINVAL)
+                }
+            }
+
+            /// Write the value contained in `self` to `io`, using the base address provided by
+            /// `base` and adding the offset of array register `idx` to it.
+            ///
+            /// The validity of `idx` is checked at run-time, and `EINVAL` is returned is the
+            /// access was out-of-bounds.
+            #[inline(always)]
+            pub(crate) fn try_write<const SIZE: usize, T, B>(
+                self,
+                io: &T,
+                base: &B,
+                idx: usize,
+            ) -> ::kernel::error::Result where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
+            {
+                if idx < Self::SIZE {
+                    Ok(self.write(io, base, idx))
+                } else {
+                    Err(EINVAL)
+                }
+            }
+
+            /// Read the array register at index `idx` from `io`, using the base address provided
+            /// by `base` and adding the register's offset to it, then run `f` on its value to
+            /// obtain a new value to write back.
+            ///
+            /// The validity of `idx` is checked at run-time, and `EINVAL` is returned is the
+            /// access was out-of-bounds.
+            #[inline(always)]
+            pub(crate) fn try_alter<const SIZE: usize, T, B, F>(
+                io: &T,
+                base: &B,
+                idx: usize,
+                f: F,
+            ) -> ::kernel::error::Result where
+                T: ::core::ops::Deref<Target = ::kernel::io::Io<SIZE>>,
+                B: crate::regs::macros::RegisterBase<$base>,
+                F: ::core::ops::FnOnce(Self) -> Self,
+            {
+                if idx < Self::SIZE {
+                    Ok(Self::alter(io, base, idx, f))
+                } else {
+                    Err(EINVAL)
+                }
+            }
+        }
+    };
 }

From 60b006b7cf0dd5e8ee99bd77fb4573e1a3956035 Mon Sep 17 00:00:00 2001
From: Uros Bizjak <ubizjak@gmail.com>
Date: Thu, 14 Aug 2025 07:43:26 +0100
Subject: [PATCH 0474/1455] drm/i915/active: Use try_cmpxchg64() in
 __active_lookup()

Replace this pattern in __active_lookup():

    cmpxchg64(*ptr, old, new) == old

... with the simpler and faster:

    try_cmpxchg64(*ptr, &old, new)

The x86 CMPXCHG instruction returns success in the ZF flag,
so this change saves a compare after the CMPXCHG.

The patch also improves the explanation of what the code really
does. cmpxchg64() will *succeed* for the winner of the race and
try_cmpxchg64() nicely documents this fact.

No functional change intended.

Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Tvrtko Ursulin <tursulin@ursulin.net>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Tvrtko Ursulin <tursulin@ursulin.net>
Link: https://lore.kernel.org/r/20250814064326.95519-1-tvrtko.ursulin@igalia.com
---
 drivers/gpu/drm/i915/i915_active.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c
index 0dbc4e289300..6b0c1162505a 100644
--- a/drivers/gpu/drm/i915/i915_active.c
+++ b/drivers/gpu/drm/i915/i915_active.c
@@ -257,10 +257,9 @@ static struct active_node *__active_lookup(struct i915_active *ref, u64 idx)
 		 * claimed the cache and we know that is does not match our
 		 * idx. If, and only if, the timeline is currently zero is it
 		 * worth competing to claim it atomically for ourselves (for
-		 * only the winner of that race will cmpxchg return the old
-		 * value of 0).
+		 * only the winner of that race will cmpxchg succeed).
 		 */
-		if (!cached && !cmpxchg64(&it->timeline, 0, idx))
+		if (!cached && try_cmpxchg64(&it->timeline, &cached, idx))
 			return it;
 	}
 

From 7c79b8aa91413f6ce2e6b18d02689db615085c95 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 1 Aug 2025 15:28:32 +0300
Subject: [PATCH 0475/1455] drm/i915/audio: drop irq enabled check from LPE
 audio setup

There shouldn't be anything requiring irqs to be enabled at the point of
LPE audio setup. Regardless, we've never hit the warning, as irqs are
always enabled at the time LPE audio is initialized. Drop the
superfluous warning, and the dependency on i915_drv.h.

Fix style a bit while at it.

Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Link: https://lore.kernel.org/r/20250801122832.249985-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_lpe_audio.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_lpe_audio.c b/drivers/gpu/drm/i915/display/intel_lpe_audio.c
index 666148a14522..42284e9928f2 100644
--- a/drivers/gpu/drm/i915/display/intel_lpe_audio.c
+++ b/drivers/gpu/drm/i915/display/intel_lpe_audio.c
@@ -68,9 +68,9 @@
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 
+#include <drm/drm_print.h>
 #include <drm/intel/intel_lpe_audio.h>
 
-#include "i915_drv.h"
 #include "i915_irq.h"
 #include "intel_audio_regs.h"
 #include "intel_de.h"
@@ -170,14 +170,11 @@ static struct irq_chip lpe_audio_irqchip = {
 
 static int lpe_audio_irq_init(struct intel_display *display)
 {
-	struct drm_i915_private *dev_priv = to_i915(display->drm);
 	int irq = display->audio.lpe.irq;
 
-	drm_WARN_ON(display->drm, !intel_irqs_enabled(dev_priv));
-	irq_set_chip_and_handler_name(irq,
-				&lpe_audio_irqchip,
-				handle_simple_irq,
-				"hdmi_lpe_audio_irq_handler");
+	irq_set_chip_and_handler_name(irq, &lpe_audio_irqchip,
+				      handle_simple_irq,
+				      "hdmi_lpe_audio_irq_handler");
 
 	return 0;
 }

From 94332524b5ccf9cfcb3c6184f1b9d6be65c6aad6 Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:27 +0100
Subject: [PATCH 0476/1455] drm/panthor: Add panthor_hw and move gpu_info
 initialization into it

This patch introduces panthor_hw and moves the initialization of the
gpu_info struct into panthor_hw.c in preparation of handling future GPU
register and naming changes.

Future GPU support can be added by extending panthor_gpu_info_init()
with the necessary register reads behind GPU architecture version guards
if the change is minor. For more complex changes, the function can be
forked and the appropriate function will need to be called based on the
GPU architecture version.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-2-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/Makefile         |   1 +
 drivers/gpu/drm/panthor/panthor_device.c |   5 +
 drivers/gpu/drm/panthor/panthor_gpu.c    |  95 -------------------
 drivers/gpu/drm/panthor/panthor_hw.c     | 113 +++++++++++++++++++++++
 drivers/gpu/drm/panthor/panthor_hw.h     |  11 +++
 5 files changed, 130 insertions(+), 95 deletions(-)
 create mode 100644 drivers/gpu/drm/panthor/panthor_hw.c
 create mode 100644 drivers/gpu/drm/panthor/panthor_hw.h

diff --git a/drivers/gpu/drm/panthor/Makefile b/drivers/gpu/drm/panthor/Makefile
index 15294719b09c..02db21748c12 100644
--- a/drivers/gpu/drm/panthor/Makefile
+++ b/drivers/gpu/drm/panthor/Makefile
@@ -8,6 +8,7 @@ panthor-y := \
 	panthor_gem.o \
 	panthor_gpu.o \
 	panthor_heap.o \
+	panthor_hw.o \
 	panthor_mmu.o \
 	panthor_sched.o
 
diff --git a/drivers/gpu/drm/panthor/panthor_device.c b/drivers/gpu/drm/panthor/panthor_device.c
index f0b2da5b2b96..81df49880bd8 100644
--- a/drivers/gpu/drm/panthor/panthor_device.c
+++ b/drivers/gpu/drm/panthor/panthor_device.c
@@ -18,6 +18,7 @@
 #include "panthor_device.h"
 #include "panthor_fw.h"
 #include "panthor_gpu.h"
+#include "panthor_hw.h"
 #include "panthor_mmu.h"
 #include "panthor_regs.h"
 #include "panthor_sched.h"
@@ -244,6 +245,10 @@ int panthor_device_init(struct panthor_device *ptdev)
 			return ret;
 	}
 
+	ret = panthor_hw_init(ptdev);
+	if (ret)
+		goto err_rpm_put;
+
 	ret = panthor_gpu_init(ptdev);
 	if (ret)
 		goto err_rpm_put;
diff --git a/drivers/gpu/drm/panthor/panthor_gpu.c b/drivers/gpu/drm/panthor/panthor_gpu.c
index cb7a335e07d7..5e2c3173ae27 100644
--- a/drivers/gpu/drm/panthor/panthor_gpu.c
+++ b/drivers/gpu/drm/panthor/panthor_gpu.c
@@ -37,40 +37,6 @@ struct panthor_gpu {
 	wait_queue_head_t reqs_acked;
 };
 
-/**
- * struct panthor_model - GPU model description
- */
-struct panthor_model {
-	/** @name: Model name. */
-	const char *name;
-
-	/** @arch_major: Major version number of architecture. */
-	u8 arch_major;
-
-	/** @product_major: Major version number of product. */
-	u8 product_major;
-};
-
-/**
- * GPU_MODEL() - Define a GPU model. A GPU product can be uniquely identified
- * by a combination of the major architecture version and the major product
- * version.
- * @_name: Name for the GPU model.
- * @_arch_major: Architecture major.
- * @_product_major: Product major.
- */
-#define GPU_MODEL(_name, _arch_major, _product_major) \
-{\
-	.name = __stringify(_name),				\
-	.arch_major = _arch_major,				\
-	.product_major = _product_major,			\
-}
-
-static const struct panthor_model gpu_models[] = {
-	GPU_MODEL(g610, 10, 7),
-	{},
-};
-
 #define GPU_INTERRUPTS_MASK	\
 	(GPU_IRQ_FAULT | \
 	 GPU_IRQ_PROTM_FAULT | \
@@ -83,66 +49,6 @@ static void panthor_gpu_coherency_set(struct panthor_device *ptdev)
 		ptdev->coherent ? GPU_COHERENCY_PROT_BIT(ACE_LITE) : GPU_COHERENCY_NONE);
 }
 
-static void panthor_gpu_init_info(struct panthor_device *ptdev)
-{
-	const struct panthor_model *model;
-	u32 arch_major, product_major;
-	u32 major, minor, status;
-	unsigned int i;
-
-	ptdev->gpu_info.gpu_id = gpu_read(ptdev, GPU_ID);
-	ptdev->gpu_info.csf_id = gpu_read(ptdev, GPU_CSF_ID);
-	ptdev->gpu_info.gpu_rev = gpu_read(ptdev, GPU_REVID);
-	ptdev->gpu_info.core_features = gpu_read(ptdev, GPU_CORE_FEATURES);
-	ptdev->gpu_info.l2_features = gpu_read(ptdev, GPU_L2_FEATURES);
-	ptdev->gpu_info.tiler_features = gpu_read(ptdev, GPU_TILER_FEATURES);
-	ptdev->gpu_info.mem_features = gpu_read(ptdev, GPU_MEM_FEATURES);
-	ptdev->gpu_info.mmu_features = gpu_read(ptdev, GPU_MMU_FEATURES);
-	ptdev->gpu_info.thread_features = gpu_read(ptdev, GPU_THREAD_FEATURES);
-	ptdev->gpu_info.max_threads = gpu_read(ptdev, GPU_THREAD_MAX_THREADS);
-	ptdev->gpu_info.thread_max_workgroup_size = gpu_read(ptdev, GPU_THREAD_MAX_WORKGROUP_SIZE);
-	ptdev->gpu_info.thread_max_barrier_size = gpu_read(ptdev, GPU_THREAD_MAX_BARRIER_SIZE);
-	ptdev->gpu_info.coherency_features = gpu_read(ptdev, GPU_COHERENCY_FEATURES);
-	for (i = 0; i < 4; i++)
-		ptdev->gpu_info.texture_features[i] = gpu_read(ptdev, GPU_TEXTURE_FEATURES(i));
-
-	ptdev->gpu_info.as_present = gpu_read(ptdev, GPU_AS_PRESENT);
-
-	ptdev->gpu_info.shader_present = gpu_read64(ptdev, GPU_SHADER_PRESENT);
-	ptdev->gpu_info.tiler_present = gpu_read64(ptdev, GPU_TILER_PRESENT);
-	ptdev->gpu_info.l2_present = gpu_read64(ptdev, GPU_L2_PRESENT);
-
-	arch_major = GPU_ARCH_MAJOR(ptdev->gpu_info.gpu_id);
-	product_major = GPU_PROD_MAJOR(ptdev->gpu_info.gpu_id);
-	major = GPU_VER_MAJOR(ptdev->gpu_info.gpu_id);
-	minor = GPU_VER_MINOR(ptdev->gpu_info.gpu_id);
-	status = GPU_VER_STATUS(ptdev->gpu_info.gpu_id);
-
-	for (model = gpu_models; model->name; model++) {
-		if (model->arch_major == arch_major &&
-		    model->product_major == product_major)
-			break;
-	}
-
-	drm_info(&ptdev->base,
-		 "mali-%s id 0x%x major 0x%x minor 0x%x status 0x%x",
-		 model->name ?: "unknown", ptdev->gpu_info.gpu_id >> 16,
-		 major, minor, status);
-
-	drm_info(&ptdev->base,
-		 "Features: L2:%#x Tiler:%#x Mem:%#x MMU:%#x AS:%#x",
-		 ptdev->gpu_info.l2_features,
-		 ptdev->gpu_info.tiler_features,
-		 ptdev->gpu_info.mem_features,
-		 ptdev->gpu_info.mmu_features,
-		 ptdev->gpu_info.as_present);
-
-	drm_info(&ptdev->base,
-		 "shader_present=0x%0llx l2_present=0x%0llx tiler_present=0x%0llx",
-		 ptdev->gpu_info.shader_present, ptdev->gpu_info.l2_present,
-		 ptdev->gpu_info.tiler_present);
-}
-
 static void panthor_gpu_irq_handler(struct panthor_device *ptdev, u32 status)
 {
 	gpu_write(ptdev, GPU_INT_CLEAR, status);
@@ -205,7 +111,6 @@ int panthor_gpu_init(struct panthor_device *ptdev)
 	spin_lock_init(&gpu->reqs_lock);
 	init_waitqueue_head(&gpu->reqs_acked);
 	ptdev->gpu = gpu;
-	panthor_gpu_init_info(ptdev);
 
 	dma_set_max_seg_size(ptdev->base.dev, UINT_MAX);
 	pa_bits = GPU_MMU_FEATURES_PA_BITS(ptdev->gpu_info.mmu_features);
diff --git a/drivers/gpu/drm/panthor/panthor_hw.c b/drivers/gpu/drm/panthor/panthor_hw.c
new file mode 100644
index 000000000000..94f170a8e5c5
--- /dev/null
+++ b/drivers/gpu/drm/panthor/panthor_hw.c
@@ -0,0 +1,113 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+/* Copyright 2025 ARM Limited. All rights reserved. */
+
+#include "panthor_device.h"
+#include "panthor_hw.h"
+#include "panthor_regs.h"
+
+/**
+ * struct panthor_model - GPU model description
+ */
+struct panthor_model {
+	/** @name: Model name. */
+	const char *name;
+
+	/** @arch_major: Major version number of architecture. */
+	u8 arch_major;
+
+	/** @product_major: Major version number of product. */
+	u8 product_major;
+};
+
+/**
+ * GPU_MODEL() - Define a GPU model. A GPU product can be uniquely identified
+ * by a combination of the major architecture version and the major product
+ * version.
+ * @_name: Name for the GPU model.
+ * @_arch_major: Architecture major.
+ * @_product_major: Product major.
+ */
+#define GPU_MODEL(_name, _arch_major, _product_major) \
+{\
+	.name = __stringify(_name),				\
+	.arch_major = _arch_major,				\
+	.product_major = _product_major,			\
+}
+
+static const struct panthor_model gpu_models[] = {
+	GPU_MODEL(g610, 10, 7),
+	{},
+};
+
+static void panthor_gpu_info_init(struct panthor_device *ptdev)
+{
+	unsigned int i;
+
+	ptdev->gpu_info.gpu_id = gpu_read(ptdev, GPU_ID);
+	ptdev->gpu_info.csf_id = gpu_read(ptdev, GPU_CSF_ID);
+	ptdev->gpu_info.gpu_rev = gpu_read(ptdev, GPU_REVID);
+	ptdev->gpu_info.core_features = gpu_read(ptdev, GPU_CORE_FEATURES);
+	ptdev->gpu_info.l2_features = gpu_read(ptdev, GPU_L2_FEATURES);
+	ptdev->gpu_info.tiler_features = gpu_read(ptdev, GPU_TILER_FEATURES);
+	ptdev->gpu_info.mem_features = gpu_read(ptdev, GPU_MEM_FEATURES);
+	ptdev->gpu_info.mmu_features = gpu_read(ptdev, GPU_MMU_FEATURES);
+	ptdev->gpu_info.thread_features = gpu_read(ptdev, GPU_THREAD_FEATURES);
+	ptdev->gpu_info.max_threads = gpu_read(ptdev, GPU_THREAD_MAX_THREADS);
+	ptdev->gpu_info.thread_max_workgroup_size = gpu_read(ptdev, GPU_THREAD_MAX_WORKGROUP_SIZE);
+	ptdev->gpu_info.thread_max_barrier_size = gpu_read(ptdev, GPU_THREAD_MAX_BARRIER_SIZE);
+	ptdev->gpu_info.coherency_features = gpu_read(ptdev, GPU_COHERENCY_FEATURES);
+	for (i = 0; i < 4; i++)
+		ptdev->gpu_info.texture_features[i] = gpu_read(ptdev, GPU_TEXTURE_FEATURES(i));
+
+	ptdev->gpu_info.as_present = gpu_read(ptdev, GPU_AS_PRESENT);
+
+	ptdev->gpu_info.shader_present = gpu_read64(ptdev, GPU_SHADER_PRESENT);
+	ptdev->gpu_info.tiler_present = gpu_read64(ptdev, GPU_TILER_PRESENT);
+	ptdev->gpu_info.l2_present = gpu_read64(ptdev, GPU_L2_PRESENT);
+}
+
+static void panthor_hw_info_init(struct panthor_device *ptdev)
+{
+	const struct panthor_model *model;
+	u32 arch_major, product_major;
+	u32 major, minor, status;
+
+	panthor_gpu_info_init(ptdev);
+
+	arch_major = GPU_ARCH_MAJOR(ptdev->gpu_info.gpu_id);
+	product_major = GPU_PROD_MAJOR(ptdev->gpu_info.gpu_id);
+	major = GPU_VER_MAJOR(ptdev->gpu_info.gpu_id);
+	minor = GPU_VER_MINOR(ptdev->gpu_info.gpu_id);
+	status = GPU_VER_STATUS(ptdev->gpu_info.gpu_id);
+
+	for (model = gpu_models; model->name; model++) {
+		if (model->arch_major == arch_major &&
+		    model->product_major == product_major)
+			break;
+	}
+
+	drm_info(&ptdev->base,
+		 "mali-%s id 0x%x major 0x%x minor 0x%x status 0x%x",
+		 model->name ?: "unknown", ptdev->gpu_info.gpu_id >> 16,
+		 major, minor, status);
+
+	drm_info(&ptdev->base,
+		 "Features: L2:%#x Tiler:%#x Mem:%#x MMU:%#x AS:%#x",
+		 ptdev->gpu_info.l2_features,
+		 ptdev->gpu_info.tiler_features,
+		 ptdev->gpu_info.mem_features,
+		 ptdev->gpu_info.mmu_features,
+		 ptdev->gpu_info.as_present);
+
+	drm_info(&ptdev->base,
+		 "shader_present=0x%0llx l2_present=0x%0llx tiler_present=0x%0llx",
+		 ptdev->gpu_info.shader_present, ptdev->gpu_info.l2_present,
+		 ptdev->gpu_info.tiler_present);
+}
+
+int panthor_hw_init(struct panthor_device *ptdev)
+{
+	panthor_hw_info_init(ptdev);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/panthor/panthor_hw.h b/drivers/gpu/drm/panthor/panthor_hw.h
new file mode 100644
index 000000000000..0af6acc6aa6a
--- /dev/null
+++ b/drivers/gpu/drm/panthor/panthor_hw.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: GPL-2.0 or MIT */
+/* Copyright 2025 ARM Limited. All rights reserved. */
+
+#ifndef __PANTHOR_HW_H__
+#define __PANTHOR_HW_H__
+
+struct panthor_device;
+
+int panthor_hw_init(struct panthor_device *ptdev);
+
+#endif /* __PANTHOR_HW_H__ */

From c012663b178c4e6fb0179c2816fb0a270f080b59 Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:28 +0100
Subject: [PATCH 0477/1455] drm/panthor: Simplify getting the GPU model name

This patch replaces the panthor_model structure with a simple switch
case based on the product_id which is in the format of:
        ((arch_major << 24) | product_major)

This simplifies comparison and allows extending of the function to
accommodate naming differences based on supported GPU features.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-3-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/panthor_hw.c | 55 +++++++---------------------
 1 file changed, 14 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_hw.c b/drivers/gpu/drm/panthor/panthor_hw.c
index 94f170a8e5c5..6bff69fd247e 100644
--- a/drivers/gpu/drm/panthor/panthor_hw.c
+++ b/drivers/gpu/drm/panthor/panthor_hw.c
@@ -5,40 +5,23 @@
 #include "panthor_hw.h"
 #include "panthor_regs.h"
 
-/**
- * struct panthor_model - GPU model description
- */
-struct panthor_model {
-	/** @name: Model name. */
-	const char *name;
+#define GPU_PROD_ID_MAKE(arch_major, prod_major) \
+	(((arch_major) << 24) | (prod_major))
 
-	/** @arch_major: Major version number of architecture. */
-	u8 arch_major;
+static char *get_gpu_model_name(struct panthor_device *ptdev)
+{
+	const u32 gpu_id = ptdev->gpu_info.gpu_id;
+	const u32 product_id = GPU_PROD_ID_MAKE(GPU_ARCH_MAJOR(gpu_id),
+						GPU_PROD_MAJOR(gpu_id));
 
-	/** @product_major: Major version number of product. */
-	u8 product_major;
-};
+	switch (product_id) {
+	case GPU_PROD_ID_MAKE(10, 7):
+		return "Mali-G610";
+	}
 
-/**
- * GPU_MODEL() - Define a GPU model. A GPU product can be uniquely identified
- * by a combination of the major architecture version and the major product
- * version.
- * @_name: Name for the GPU model.
- * @_arch_major: Architecture major.
- * @_product_major: Product major.
- */
-#define GPU_MODEL(_name, _arch_major, _product_major) \
-{\
-	.name = __stringify(_name),				\
-	.arch_major = _arch_major,				\
-	.product_major = _product_major,			\
+	return "(Unknown Mali GPU)";
 }
 
-static const struct panthor_model gpu_models[] = {
-	GPU_MODEL(g610, 10, 7),
-	{},
-};
-
 static void panthor_gpu_info_init(struct panthor_device *ptdev)
 {
 	unsigned int i;
@@ -68,27 +51,17 @@ static void panthor_gpu_info_init(struct panthor_device *ptdev)
 
 static void panthor_hw_info_init(struct panthor_device *ptdev)
 {
-	const struct panthor_model *model;
-	u32 arch_major, product_major;
 	u32 major, minor, status;
 
 	panthor_gpu_info_init(ptdev);
 
-	arch_major = GPU_ARCH_MAJOR(ptdev->gpu_info.gpu_id);
-	product_major = GPU_PROD_MAJOR(ptdev->gpu_info.gpu_id);
 	major = GPU_VER_MAJOR(ptdev->gpu_info.gpu_id);
 	minor = GPU_VER_MINOR(ptdev->gpu_info.gpu_id);
 	status = GPU_VER_STATUS(ptdev->gpu_info.gpu_id);
 
-	for (model = gpu_models; model->name; model++) {
-		if (model->arch_major == arch_major &&
-		    model->product_major == product_major)
-			break;
-	}
-
 	drm_info(&ptdev->base,
-		 "mali-%s id 0x%x major 0x%x minor 0x%x status 0x%x",
-		 model->name ?: "unknown", ptdev->gpu_info.gpu_id >> 16,
+		 "%s id 0x%x major 0x%x minor 0x%x status 0x%x",
+		 get_gpu_model_name(ptdev), ptdev->gpu_info.gpu_id >> 16,
 		 major, minor, status);
 
 	drm_info(&ptdev->base,

From 2c0d0a81c0ad5a7271177b016014360c331ad62c Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:29 +0100
Subject: [PATCH 0478/1455] drm/panthor: Add support for Mali-G710, Mali-G510
 and Mali-G310

This patch adds GPU model name and FW binary support for Mali-G710,
Mali-G510, and Mali-G310.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-4-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/panthor_fw.c | 2 ++
 drivers/gpu/drm/panthor/panthor_hw.c | 6 ++++++
 2 files changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c
index 36f1034839c2..b7b454d16f12 100644
--- a/drivers/gpu/drm/panthor/panthor_fw.c
+++ b/drivers/gpu/drm/panthor/panthor_fw.c
@@ -1402,3 +1402,5 @@ err_unplug_fw:
 }
 
 MODULE_FIRMWARE("arm/mali/arch10.8/mali_csffw.bin");
+MODULE_FIRMWARE("arm/mali/arch10.10/mali_csffw.bin");
+MODULE_FIRMWARE("arm/mali/arch10.12/mali_csffw.bin");
diff --git a/drivers/gpu/drm/panthor/panthor_hw.c b/drivers/gpu/drm/panthor/panthor_hw.c
index 6bff69fd247e..6d2ad3c8cb5e 100644
--- a/drivers/gpu/drm/panthor/panthor_hw.c
+++ b/drivers/gpu/drm/panthor/panthor_hw.c
@@ -15,6 +15,12 @@ static char *get_gpu_model_name(struct panthor_device *ptdev)
 						GPU_PROD_MAJOR(gpu_id));
 
 	switch (product_id) {
+	case GPU_PROD_ID_MAKE(10, 2):
+		return "Mali-G710";
+	case GPU_PROD_ID_MAKE(10, 3):
+		return "Mali-G510";
+	case GPU_PROD_ID_MAKE(10, 4):
+		return "Mali-G310";
 	case GPU_PROD_ID_MAKE(10, 7):
 		return "Mali-G610";
 	}

From 3b1dc21d6d800cb86c1ef0c97968f5c783343f2b Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:30 +0100
Subject: [PATCH 0479/1455] drm/panthor: Add support for Mali-Gx15 family of
 GPUs

Mali-Gx15 introduces a new GPU_FEATURES register that provides
information about GPU-wide supported features. The register value will
be passed on to userspace via gpu_info.

Additionally, Mali-Gx15 presents an 'Immortalis' naming variant
depending on the shader core count and presence of Ray Intersection
feature support.

This patch adds:
- support for correctly identifying the model names for Mali-Gx15 GPUs.
- arch 11.8 FW binary support

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-5-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/panthor_fw.c   |  1 +
 drivers/gpu/drm/panthor/panthor_hw.c   | 15 +++++++++++++++
 drivers/gpu/drm/panthor/panthor_regs.h |  3 +++
 include/uapi/drm/panthor_drm.h         |  3 +++
 4 files changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c
index b7b454d16f12..fa6e0b48a0b2 100644
--- a/drivers/gpu/drm/panthor/panthor_fw.c
+++ b/drivers/gpu/drm/panthor/panthor_fw.c
@@ -1404,3 +1404,4 @@ err_unplug_fw:
 MODULE_FIRMWARE("arm/mali/arch10.8/mali_csffw.bin");
 MODULE_FIRMWARE("arm/mali/arch10.10/mali_csffw.bin");
 MODULE_FIRMWARE("arm/mali/arch10.12/mali_csffw.bin");
+MODULE_FIRMWARE("arm/mali/arch11.8/mali_csffw.bin");
diff --git a/drivers/gpu/drm/panthor/panthor_hw.c b/drivers/gpu/drm/panthor/panthor_hw.c
index 6d2ad3c8cb5e..8c041e1074a1 100644
--- a/drivers/gpu/drm/panthor/panthor_hw.c
+++ b/drivers/gpu/drm/panthor/panthor_hw.c
@@ -13,6 +13,9 @@ static char *get_gpu_model_name(struct panthor_device *ptdev)
 	const u32 gpu_id = ptdev->gpu_info.gpu_id;
 	const u32 product_id = GPU_PROD_ID_MAKE(GPU_ARCH_MAJOR(gpu_id),
 						GPU_PROD_MAJOR(gpu_id));
+	const bool ray_intersection = !!(ptdev->gpu_info.gpu_features &
+					 GPU_FEATURES_RAY_INTERSECTION);
+	const u8 shader_core_count = hweight64(ptdev->gpu_info.shader_present);
 
 	switch (product_id) {
 	case GPU_PROD_ID_MAKE(10, 2):
@@ -23,6 +26,15 @@ static char *get_gpu_model_name(struct panthor_device *ptdev)
 		return "Mali-G310";
 	case GPU_PROD_ID_MAKE(10, 7):
 		return "Mali-G610";
+	case GPU_PROD_ID_MAKE(11, 2):
+		if (shader_core_count > 10 && ray_intersection)
+			return "Mali-G715-Immortalis";
+		else if (shader_core_count >= 7)
+			return "Mali-G715";
+
+		fallthrough;
+	case GPU_PROD_ID_MAKE(11, 3):
+		return "Mali-G615";
 	}
 
 	return "(Unknown Mali GPU)";
@@ -53,6 +65,9 @@ static void panthor_gpu_info_init(struct panthor_device *ptdev)
 	ptdev->gpu_info.shader_present = gpu_read64(ptdev, GPU_SHADER_PRESENT);
 	ptdev->gpu_info.tiler_present = gpu_read64(ptdev, GPU_TILER_PRESENT);
 	ptdev->gpu_info.l2_present = gpu_read64(ptdev, GPU_L2_PRESENT);
+
+	/* Introduced in arch 11.x */
+	ptdev->gpu_info.gpu_features = gpu_read64(ptdev, GPU_FEATURES);
 }
 
 static void panthor_hw_info_init(struct panthor_device *ptdev)
diff --git a/drivers/gpu/drm/panthor/panthor_regs.h b/drivers/gpu/drm/panthor/panthor_regs.h
index 48bbfd40138c..8bee76d01bf8 100644
--- a/drivers/gpu/drm/panthor/panthor_regs.h
+++ b/drivers/gpu/drm/panthor/panthor_regs.h
@@ -70,6 +70,9 @@
 #define GPU_PWR_OVERRIDE0				0x54
 #define GPU_PWR_OVERRIDE1				0x58
 
+#define GPU_FEATURES					0x60
+#define   GPU_FEATURES_RAY_INTERSECTION			BIT(2)
+
 #define GPU_TIMESTAMP_OFFSET				0x88
 #define GPU_CYCLE_COUNT					0x90
 #define GPU_TIMESTAMP					0x98
diff --git a/include/uapi/drm/panthor_drm.h b/include/uapi/drm/panthor_drm.h
index e1f43deb7eca..467d365ed7ba 100644
--- a/include/uapi/drm/panthor_drm.h
+++ b/include/uapi/drm/panthor_drm.h
@@ -327,6 +327,9 @@ struct drm_panthor_gpu_info {
 
 	/** @pad: MBZ. */
 	__u32 pad;
+
+	/** @gpu_features: Bitmask describing supported GPU-wide features */
+	__u64 gpu_features;
 };
 
 /**

From e322a4844811b54477b7072eb40dc9e402a1725d Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:31 +0100
Subject: [PATCH 0480/1455] drm/panthor: Serialize GPU cache flush operations

In certain scenarios, it is possible for multiple cache flushes to be
requested before the previous one completes. This patch introduces the
cache_flush_lock mutex to serialize these operations and ensure that
any requested cache flushes are completed instead of dropped.

Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Co-developed-by: Dennis Tsiang <dennis.tsiang@arm.com>
Signed-off-by: Dennis Tsiang <dennis.tsiang@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-6-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/panthor_gpu.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/panthor/panthor_gpu.c b/drivers/gpu/drm/panthor/panthor_gpu.c
index 5e2c3173ae27..db69449a5be0 100644
--- a/drivers/gpu/drm/panthor/panthor_gpu.c
+++ b/drivers/gpu/drm/panthor/panthor_gpu.c
@@ -35,6 +35,9 @@ struct panthor_gpu {
 
 	/** @reqs_acked: GPU request wait queue. */
 	wait_queue_head_t reqs_acked;
+
+	/** @cache_flush_lock: Lock to serialize cache flushes */
+	struct mutex cache_flush_lock;
 };
 
 #define GPU_INTERRUPTS_MASK	\
@@ -110,6 +113,7 @@ int panthor_gpu_init(struct panthor_device *ptdev)
 
 	spin_lock_init(&gpu->reqs_lock);
 	init_waitqueue_head(&gpu->reqs_acked);
+	mutex_init(&gpu->cache_flush_lock);
 	ptdev->gpu = gpu;
 
 	dma_set_max_seg_size(ptdev->base.dev, UINT_MAX);
@@ -258,6 +262,9 @@ int panthor_gpu_flush_caches(struct panthor_device *ptdev,
 	bool timedout = false;
 	unsigned long flags;
 
+	/* Serialize cache flush operations. */
+	guard(mutex)(&ptdev->gpu->cache_flush_lock);
+
 	spin_lock_irqsave(&ptdev->gpu->reqs_lock, flags);
 	if (!drm_WARN_ON(&ptdev->base,
 			 ptdev->gpu->pending_reqs & GPU_IRQ_CLEAN_CACHES_COMPLETED)) {

From dd11c7dec74b0e125b75def169a53f3c86f35d6d Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:32 +0100
Subject: [PATCH 0481/1455] drm/panthor: Make MMU cache maintenance use
 FLUSH_CACHES command

As the FLUSH_MEM and FLUSH_PT MMU_AS commands are deprecated in GPUs
from Mali-Gx20 onwards, this patch adds support for performing cache
maintenance via the FLUSH_CACHES command in GPU_COMMAND in place of
FLUSH_MEM and FLUSH_PT commands.

Mali-Gx10 and Mali-Gx15 GPUs also has support for the FLUSH_CACHES
command and will also use this by default going forward.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-7-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/panthor_mmu.c | 33 +++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index 4140f697ba5a..367c89aca558 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -29,6 +29,7 @@
 
 #include "panthor_device.h"
 #include "panthor_gem.h"
+#include "panthor_gpu.h"
 #include "panthor_heap.h"
 #include "panthor_mmu.h"
 #include "panthor_regs.h"
@@ -568,6 +569,35 @@ static void lock_region(struct panthor_device *ptdev, u32 as_nr,
 	write_cmd(ptdev, as_nr, AS_COMMAND_LOCK);
 }
 
+static int mmu_hw_do_flush_on_gpu_ctrl(struct panthor_device *ptdev, int as_nr,
+				       u32 op)
+{
+	const u32 l2_flush_op = CACHE_CLEAN | CACHE_INV;
+	u32 lsc_flush_op = 0;
+	int ret;
+
+	if (op == AS_COMMAND_FLUSH_MEM)
+		lsc_flush_op = CACHE_CLEAN | CACHE_INV;
+
+	ret = wait_ready(ptdev, as_nr);
+	if (ret)
+		return ret;
+
+	ret = panthor_gpu_flush_caches(ptdev, l2_flush_op, lsc_flush_op, 0);
+	if (ret)
+		return ret;
+
+	/*
+	 * Explicitly unlock the region as the AS is not unlocked automatically
+	 * at the end of the GPU_CONTROL cache flush command, unlike
+	 * AS_COMMAND_FLUSH_MEM or AS_COMMAND_FLUSH_PT.
+	 */
+	write_cmd(ptdev, as_nr, AS_COMMAND_UNLOCK);
+
+	/* Wait for the unlock command to complete */
+	return wait_ready(ptdev, as_nr);
+}
+
 static int mmu_hw_do_operation_locked(struct panthor_device *ptdev, int as_nr,
 				      u64 iova, u64 size, u32 op)
 {
@@ -585,6 +615,9 @@ static int mmu_hw_do_operation_locked(struct panthor_device *ptdev, int as_nr,
 	if (op != AS_COMMAND_UNLOCK)
 		lock_region(ptdev, as_nr, iova, size);
 
+	if (op == AS_COMMAND_FLUSH_MEM || op == AS_COMMAND_FLUSH_PT)
+		return mmu_hw_do_flush_on_gpu_ctrl(ptdev, as_nr, op);
+
 	/* Run the MMU operation */
 	write_cmd(ptdev, as_nr, op);
 

From f9c67b019bc3c0324ee42c0dbfbb2d55726d751e Mon Sep 17 00:00:00 2001
From: Karunika Choo <karunika.choo@arm.com>
Date: Thu, 7 Aug 2025 17:26:33 +0100
Subject: [PATCH 0482/1455] drm/panthor: Add support for Mali-Gx20 and
 Mali-Gx25 GPUs

This patch adds firmware binary and GPU model naming support for
Mali-Gx20 and Mali-Gx25 GPUs.

Reviewed-by: Steven Price <steven.price@arm.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Signed-off-by: Karunika Choo <karunika.choo@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250807162633.3666310-8-karunika.choo@arm.com
---
 drivers/gpu/drm/panthor/panthor_fw.c |  2 ++
 drivers/gpu/drm/panthor/panthor_hw.c | 18 ++++++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/drivers/gpu/drm/panthor/panthor_fw.c b/drivers/gpu/drm/panthor/panthor_fw.c
index fa6e0b48a0b2..9bf06e55eaee 100644
--- a/drivers/gpu/drm/panthor/panthor_fw.c
+++ b/drivers/gpu/drm/panthor/panthor_fw.c
@@ -1405,3 +1405,5 @@ MODULE_FIRMWARE("arm/mali/arch10.8/mali_csffw.bin");
 MODULE_FIRMWARE("arm/mali/arch10.10/mali_csffw.bin");
 MODULE_FIRMWARE("arm/mali/arch10.12/mali_csffw.bin");
 MODULE_FIRMWARE("arm/mali/arch11.8/mali_csffw.bin");
+MODULE_FIRMWARE("arm/mali/arch12.8/mali_csffw.bin");
+MODULE_FIRMWARE("arm/mali/arch13.8/mali_csffw.bin");
diff --git a/drivers/gpu/drm/panthor/panthor_hw.c b/drivers/gpu/drm/panthor/panthor_hw.c
index 8c041e1074a1..4f2858114e5e 100644
--- a/drivers/gpu/drm/panthor/panthor_hw.c
+++ b/drivers/gpu/drm/panthor/panthor_hw.c
@@ -35,6 +35,24 @@ static char *get_gpu_model_name(struct panthor_device *ptdev)
 		fallthrough;
 	case GPU_PROD_ID_MAKE(11, 3):
 		return "Mali-G615";
+	case GPU_PROD_ID_MAKE(12, 0):
+		if (shader_core_count >= 10 && ray_intersection)
+			return "Mali-G720-Immortalis";
+		else if (shader_core_count >= 6)
+			return "Mali-G720";
+
+		fallthrough;
+	case GPU_PROD_ID_MAKE(12, 1):
+		return "Mali-G620";
+	case GPU_PROD_ID_MAKE(13, 0):
+		if (shader_core_count >= 10 && ray_intersection)
+			return "Mali-G925-Immortalis";
+		else if (shader_core_count >= 6)
+			return "Mali-G725";
+
+		fallthrough;
+	case GPU_PROD_ID_MAKE(13, 1):
+		return "Mali-G625";
 	}
 
 	return "(Unknown Mali GPU)";

From 63e0b17ae79a4365e874d7e2ff3816db2e6d55b5 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 14 Aug 2025 16:24:45 +0300
Subject: [PATCH 0483/1455] drm/i915/bo: remove unnecessary include

intel_bo.c doesn't need intel_display_types.h for anything. Remove.

Reviewed-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/20250814132445.1573035-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_bo.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bo.c b/drivers/gpu/drm/i915/display/intel_bo.c
index 65d64f79a4bd..d29c1508ccb9 100644
--- a/drivers/gpu/drm/i915/display/intel_bo.c
+++ b/drivers/gpu/drm/i915/display/intel_bo.c
@@ -2,7 +2,7 @@
 /* Copyright © 2024 Intel Corporation */
 
 #include <drm/drm_panic.h>
-#include "display/intel_display_types.h"
+
 #include "gem/i915_gem_mman.h"
 #include "gem/i915_gem_object.h"
 #include "gem/i915_gem_object_frontbuffer.h"

From 646dc268ec25810e7d6aebef1a58ad837b4378d5 Mon Sep 17 00:00:00 2001
From: Sakari Ailus <sakari.ailus@linux.intel.com>
Date: Fri, 4 Jul 2025 10:54:11 +0300
Subject: [PATCH 0484/1455] drm/panfrost: Remove redundant
 pm_runtime_mark_last_busy() calls

pm_runtime_put_autosuspend(), pm_runtime_put_sync_autosuspend(),
pm_runtime_autosuspend() and pm_request_autosuspend() now include a call
to pm_runtime_mark_last_busy(). Remove the now-reduntant explicit call to
pm_runtime_mark_last_busy().

Signed-off-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250704075411.3218059-1-sakari.ailus@linux.intel.com
---
 drivers/gpu/drm/panfrost/panfrost_perfcnt.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
index 563f16bae543..0dd62e8b2fa7 100644
--- a/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
+++ b/drivers/gpu/drm/panfrost/panfrost_perfcnt.c
@@ -203,7 +203,6 @@ static int panfrost_perfcnt_disable_locked(struct panfrost_device *pfdev,
 	panfrost_mmu_as_put(pfdev, perfcnt->mapping->mmu);
 	panfrost_gem_mapping_put(perfcnt->mapping);
 	perfcnt->mapping = NULL;
-	pm_runtime_mark_last_busy(pfdev->dev);
 	pm_runtime_put_autosuspend(pfdev->dev);
 
 	return 0;
@@ -279,7 +278,6 @@ void panfrost_perfcnt_close(struct drm_file *file_priv)
 	if (perfcnt->user == pfile)
 		panfrost_perfcnt_disable_locked(pfdev, file_priv);
 	mutex_unlock(&perfcnt->lock);
-	pm_runtime_mark_last_busy(pfdev->dev);
 	pm_runtime_put_autosuspend(pfdev->dev);
 }
 

From cb640b2ca54617f4a9d4d6efd5ff2afd6be11f19 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Sat, 2 Aug 2025 13:40:35 +0300
Subject: [PATCH 0485/1455] drm/bridge: display-connector: don't set OP_DETECT
 for DisplayPorts

Detecting the monitor for DisplayPort targets is more complicated than
just reading the HPD pin level: it requires reading the DPCD in order to
check what kind of device is attached to the port and whether there is
an actual display attached.

In order to let DRM framework handle such configurations, disable
DRM_BRIDGE_OP_DETECT for dp-connector devices, letting the actual DP
driver perform detection. This still keeps DRM_BRIDGE_OP_HPD enabled, so
it is valid for the bridge to report HPD events.

Currently inside the kernel there are only two targets which list
hpd-gpios for dp-connector devices: arm64/qcom/qcs6490-rb3gen2 and
arm64/qcom/sa8295p-adp. Both should be fine with this change.

Cc: Bjorn Andersson <andersson@kernel.org>
Cc: Konrad Dybcio <konradybcio@kernel.org>
Cc: linux-arm-msm@vger.kernel.org
Acked-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Link: https://lore.kernel.org/r/20250802-dp-conn-no-detect-v1-1-2748c2b946da@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/display-connector.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c
index d7e1c2f8f53c..e9f16dbc9535 100644
--- a/drivers/gpu/drm/bridge/display-connector.c
+++ b/drivers/gpu/drm/bridge/display-connector.c
@@ -373,7 +373,8 @@ static int display_connector_probe(struct platform_device *pdev)
 	if (conn->bridge.ddc)
 		conn->bridge.ops |= DRM_BRIDGE_OP_EDID
 				 |  DRM_BRIDGE_OP_DETECT;
-	if (conn->hpd_gpio)
+	/* Detecting the monitor requires reading DPCD */
+	if (conn->hpd_gpio && type != DRM_MODE_CONNECTOR_DisplayPort)
 		conn->bridge.ops |= DRM_BRIDGE_OP_DETECT;
 	if (conn->hpd_irq >= 0)
 		conn->bridge.ops |= DRM_BRIDGE_OP_HPD;

From 3e03525ce15cb9fc633b3d0266866020e22ce5e6 Mon Sep 17 00:00:00 2001
From: Xichao Zhao <zhao.xichao@vivo.com>
Date: Tue, 12 Aug 2025 11:16:03 +0800
Subject: [PATCH 0486/1455] drm/radeon: replace min/max nesting with clamp()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The clamp() macro explicitly expresses the intent of constraining
a value within bounds.Therefore, replacing min(max(a, b), c) and
max(min(a,b),c) with clamp(val, lo, hi) can improve code readability.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Xichao Zhao <zhao.xichao@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_display.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index b4bf5dfeea2d..d66c1a30df95 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -926,10 +926,10 @@ static void avivo_get_fb_ref_div(unsigned nom, unsigned den, unsigned post_div,
 				 unsigned *fb_div, unsigned *ref_div)
 {
 	/* limit reference * post divider to a maximum */
-	ref_div_max = max(min(100 / post_div, ref_div_max), 1u);
+	ref_div_max = clamp(100 / post_div, 1u, ref_div_max);
 
 	/* get matching reference and feedback divider */
-	*ref_div = min(max(den/post_div, 1u), ref_div_max);
+	*ref_div = clamp(den / post_div, 1u, ref_div_max);
 	*fb_div = DIV_ROUND_CLOSEST(nom * *ref_div * post_div, den);
 
 	/* limit fb divider to its maximum */

From 90b810dd859c0df9db2290da1ac5842e5f031267 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Tue, 12 Aug 2025 15:50:14 +0800
Subject: [PATCH 0487/1455] drm/amd/display: Remove redundant semicolons

Remove unnecessary semicolons.

Fixes: dda4fb85e433 ("drm/amd/display: DML changes for DCN32/321")
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c    | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c
index 9ba6cb67655f..6c75aa82327a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_rq_dlg_calc_32.c
@@ -139,7 +139,6 @@ void dml32_rq_dlg_get_rq_reg(display_rq_regs_st *rq_regs,
 	if (dual_plane) {
 		unsigned int p1_pte_row_height_linear = get_dpte_row_height_linear_c(mode_lib, e2e_pipe_param,
 				num_pipes, pipe_idx);
-		;
 		if (src->sw_mode == dm_sw_linear)
 			ASSERT(p1_pte_row_height_linear >= 8);
 

From 5c8d5e2619f7d2985adfe45608dc942ca8151aa3 Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Tue, 12 Aug 2025 16:31:49 +0800
Subject: [PATCH 0488/1455] drm/amd/display: Use boolean context for pointer
 null checks

Replace "out == 0" with "!out" for pointer comparison to improve code
readability and conform to coding style.

Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c | 2 +-
 .../amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c | 2 +-
 .../amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c   | 2 +-
 .../amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c   | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c
index 28394de02885..640087e862f8 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_factory.c
@@ -10,7 +10,7 @@ bool dml2_core_create(enum dml2_project_id project_id, struct dml2_core_instance
 {
 	bool result = false;
 
-	if (out == 0)
+	if (!out)
 		return false;
 
 	memset(out, 0, sizeof(struct dml2_core_instance));
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c
index 3861bc6c9621..dfd01440737d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_dpmm/dml2_dpmm_factory.c
@@ -20,7 +20,7 @@ bool dml2_dpmm_create(enum dml2_project_id project_id, struct dml2_dpmm_instance
 {
 	bool result = false;
 
-	if (out == 0)
+	if (!out)
 		return false;
 
 	memset(out, 0, sizeof(struct dml2_dpmm_instance));
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c
index cd3fbc0591d8..c60b8fe90819 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_mcg/dml2_mcg_factory.c
@@ -15,7 +15,7 @@ bool dml2_mcg_create(enum dml2_project_id project_id, struct dml2_mcg_instance *
 {
 	bool result = false;
 
-	if (out == 0)
+	if (!out)
 		return false;
 
 	memset(out, 0, sizeof(struct dml2_mcg_instance));
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c
index 7ed0242a4b33..55d2464365d0 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_factory.c
@@ -26,7 +26,7 @@ bool dml2_pmo_create(enum dml2_project_id project_id, struct dml2_pmo_instance *
 {
 	bool result = false;
 
-	if (out == 0)
+	if (!out)
 		return false;
 
 	memset(out, 0, sizeof(struct dml2_pmo_instance));

From 3a75edf93aaec4a3178bdae64723fb41992398f2 Mon Sep 17 00:00:00 2001
From: Eric Huang <jinhuieric.huang@amd.com>
Date: Thu, 7 Aug 2025 14:23:11 -0400
Subject: [PATCH 0489/1455] drm/amdkfd: set uuid for each partition in topology

Currently each kfd compute partition/node is sharing
the same uuid of AID, which doen't meet the CUDA spec
for visible device, so corresponding XCD id for each
partition in smu has been assigned to xcp, and exposed
to kfd topology.

v2: add NULL check (Lijo)

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 4ec73f33535e..82dbd68d8c99 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -530,6 +530,8 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr,
 		sysfs_show_32bit_prop(buffer, offs, "sdma_fw_version",
 				      dev->gpu->kfd->sdma_fw_version);
 		sysfs_show_64bit_prop(buffer, offs, "unique_id",
+				      dev->gpu->xcp ?
+				      dev->gpu->xcp->unique_id :
 				      dev->gpu->adev->unique_id);
 		sysfs_show_32bit_prop(buffer, offs, "num_xcc",
 				      NUM_XCC(dev->gpu->xcc_mask));

From 859958a7faefe5b7742b7b8cdbc170713d4bf158 Mon Sep 17 00:00:00 2001
From: Heng Zhou <Heng.Zhou@amd.com>
Date: Wed, 13 Aug 2025 11:18:04 +0800
Subject: [PATCH 0490/1455] drm/amdgpu: fix nullptr err of vm_handle_moved

If a amdgpu_bo_va is fpriv->prt_va, the bo of this one is always NULL.
So, such kind of amdgpu_bo_va should be updated separately before
amdgpu_vm_handle_moved.

Signed-off-by: Heng Zhou <Heng.Zhou@amd.com>
Reviewed-by: Kasiviswanathan, Harish <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index 37d8a7034a7e..d478acb4568a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -2970,9 +2970,22 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence __rcu *
 		struct amdgpu_device *adev = amdgpu_ttm_adev(
 			peer_vm->root.bo->tbo.bdev);
 
+		struct amdgpu_fpriv *fpriv =
+			container_of(peer_vm, struct amdgpu_fpriv, vm);
+
+		ret = amdgpu_vm_bo_update(adev, fpriv->prt_va, false);
+		if (ret) {
+			dev_dbg(adev->dev,
+				"Memory eviction: handle PRT moved failed, pid %8d. Try again.\n",
+				pid_nr(process_info->pid));
+			goto validate_map_fail;
+		}
+
 		ret = amdgpu_vm_handle_moved(adev, peer_vm, &exec.ticket);
 		if (ret) {
-			pr_debug("Memory eviction: handle moved failed. Try again\n");
+			dev_dbg(adev->dev,
+				"Memory eviction: handle moved failed, pid %8d. Try again.\n",
+				pid_nr(process_info->pid));
 			goto validate_map_fail;
 		}
 	}

From 37b9257be7cdab4d84958202f499e3c8b66abeb8 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 13 Aug 2025 10:36:58 +0800
Subject: [PATCH 0491/1455] drm/amd/pm: Add VCN reset support check capability

This change introduces infrastructure to check whether VCN reset
is supported by the SMU firmware. Key changes include:

1. Added new functions to query VCN reset support:
   - amdgpu_dpm_reset_vcn_is_supported()
   - smu_reset_vcn_is_supported()
   - pptable_funcs.reset_vcn_is_supported callback

2. Implemented proper locking in the DPM layer with mutex protection

3. Maintained consistency with existing SDMA reset support checks

The new capability allows callers to check for VCN reset support
before attempting the operation, preventing unnecessary attempts
on unsupported platforms.

v2: clean up debug info(Alex)

Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_dpm.c           | 15 +++++++++++++++
 drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h       |  1 +
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     | 10 ++++++++++
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h |  5 +++++
 4 files changed, 31 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
index 6e0d711820ea..518d07afc7df 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm.c
@@ -820,6 +820,21 @@ int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask)
 	return ret;
 }
 
+bool amdgpu_dpm_reset_vcn_is_supported(struct amdgpu_device *adev)
+{
+	struct smu_context *smu = adev->powerplay.pp_handle;
+	bool ret;
+
+	if (!is_support_sw_smu(adev))
+		return false;
+
+	mutex_lock(&adev->pm.mutex);
+	ret = smu_reset_vcn_is_supported(smu);
+	mutex_unlock(&adev->pm.mutex);
+
+	return ret;
+}
+
 int amdgpu_dpm_get_dpm_freq_range(struct amdgpu_device *adev,
 				  enum pp_clock_type type,
 				  uint32_t *min,
diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
index 09962db988d6..9748744133d9 100644
--- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
@@ -615,6 +615,7 @@ ssize_t amdgpu_dpm_get_pm_policy_info(struct amdgpu_device *adev,
 int amdgpu_dpm_reset_sdma(struct amdgpu_device *adev, uint32_t inst_mask);
 bool amdgpu_dpm_reset_sdma_is_supported(struct amdgpu_device *adev);
 int amdgpu_dpm_reset_vcn(struct amdgpu_device *adev, uint32_t inst_mask);
+bool amdgpu_dpm_reset_vcn_is_supported(struct amdgpu_device *adev);
 bool amdgpu_dpm_is_temp_metrics_supported(struct amdgpu_device *adev,
 					  enum smu_temp_metric_type type);
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 0a40ab817634..c5965924e7c6 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -4124,6 +4124,16 @@ int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask)
 	return ret;
 }
 
+bool smu_reset_vcn_is_supported(struct smu_context *smu)
+{
+	bool ret = false;
+
+	if (smu->ppt_funcs && smu->ppt_funcs->reset_vcn_is_supported)
+		ret = smu->ppt_funcs->reset_vcn_is_supported(smu);
+
+	return ret;
+}
+
 int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask)
 {
 	if (smu->ppt_funcs && smu->ppt_funcs->dpm_reset_vcn)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 2edd867f203e..5dd49eca598d 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -1435,6 +1435,10 @@ struct pptable_funcs {
 	 * @reset_vcn: message SMU to soft reset vcn instance.
 	 */
 	int (*dpm_reset_vcn)(struct smu_context *smu, uint32_t inst_mask);
+	/**
+	 * @reset_vcn_is_supported: Check if support resets vcn.
+	 */
+	bool (*reset_vcn_is_supported)(struct smu_context *smu);
 
 	/**
 	 * @get_ecc_table:  message SMU to get ECC INFO table.
@@ -1776,6 +1780,7 @@ int smu_send_rma_reason(struct smu_context *smu);
 int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask);
 bool smu_reset_sdma_is_supported(struct smu_context *smu);
 int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask);
+bool smu_reset_vcn_is_supported(struct smu_context *smu);
 int smu_set_pm_policy(struct smu_context *smu, enum pp_pm_policy p_type,
 		      int level);
 ssize_t smu_get_pm_policy_info(struct smu_context *smu,

From 9d20f37a106f30f0d867e2f8521788f88dd9ed49 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 13 Aug 2025 10:40:18 +0800
Subject: [PATCH 0492/1455] drm/amd/pm: Add VCN reset support for SMU v13.0.6

This commit implements VCN reset capability for SMU v13.0.6 with the following changes:

1. Added new PPSMC message ID (0x5B) for VCN reset in SMU firmware interface
2. Extended SMU capabilities to include VCN_RESET support
3. Implemented VCN reset support check:
   - Added smu_v13_0_6_reset_vcn_is_supported() function
4. Updated SMU v13.0.6 PPT functions to include VCN reset operations

v2: clean up debug info (Alex)
v3: remove unsupported message and split smu v13.0.6 changes to a separate patch (Lijo)
v4: simply the function (smu_v13_0_6_reset_vcn_is_supported) (Lijo)

Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h | 4 ++--
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c     | 9 +++++++++
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h     | 1 +
 3 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
index 41f268313613..63a088ef7169 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_ppsmc.h
@@ -94,9 +94,9 @@
 #define PPSMC_MSG_RmaDueToBadPageThreshold          0x43
 #define PPSMC_MSG_SetThrottlingPolicy               0x44
 #define PPSMC_MSG_ResetSDMA                         0x4D
-#define PPSMC_MSG_ResetVCN                          0x4E
 #define PPSMC_MSG_GetStaticMetricsTable             0x59
-#define PPSMC_Message_Count                         0x5A
+#define PPSMC_MSG_ResetVCN                          0x5B
+#define PPSMC_Message_Count                         0x5C
 
 //PPSMC Reset Types for driver msg argument
 #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET        0x1
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 3b4bd7a34d04..cc59d884025f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -436,6 +436,9 @@ static void smu_v13_0_6_init_caps(struct smu_context *smu)
 	    ((pgm == 0) && (fw_ver >= 0x00557900)) ||
 	    ((pgm == 4) && (fw_ver >= 0x4557000)))
 		smu_v13_0_6_cap_set(smu, SMU_CAP(SDMA_RESET));
+
+	if ((pgm == 4) && (fw_ver >= 0x04557100))
+		smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET));
 }
 
 static void smu_v13_0_x_init_caps(struct smu_context *smu)
@@ -3200,6 +3203,11 @@ static int smu_v13_0_6_reset_sdma(struct smu_context *smu, uint32_t inst_mask)
 	return ret;
 }
 
+static bool smu_v13_0_6_reset_vcn_is_supported(struct smu_context *smu)
+{
+	return smu_v13_0_6_cap_supported(smu, SMU_CAP(VCN_RESET));
+}
+
 static int smu_v13_0_6_reset_vcn(struct smu_context *smu, uint32_t inst_mask)
 {
 	int ret = 0;
@@ -3888,6 +3896,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.reset_sdma = smu_v13_0_6_reset_sdma,
 	.reset_sdma_is_supported = smu_v13_0_6_reset_sdma_is_supported,
 	.dpm_reset_vcn = smu_v13_0_6_reset_vcn,
+	.reset_vcn_is_supported = smu_v13_0_6_reset_vcn_is_supported,
 };
 
 void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index f39dbfdd7a2f..bcb8246c0804 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -64,6 +64,7 @@ enum smu_v13_0_6_caps {
 	SMU_CAP(RMA_MSG),
 	SMU_CAP(ACA_SYND),
 	SMU_CAP(SDMA_RESET),
+	SMU_CAP(VCN_RESET),
 	SMU_CAP(STATIC_METRICS),
 	SMU_CAP(HST_LIMIT_METRICS),
 	SMU_CAP(BOARD_VOLTAGE),

From 0ed704d058cec7643a716a21888d58c7d03f2c3e Mon Sep 17 00:00:00 2001
From: Kent Russell <kent.russell@amd.com>
Date: Mon, 21 Jul 2025 14:06:36 -0400
Subject: [PATCH 0493/1455] drm/amdkfd: Handle lack of READ permissions in SVM
 mapping

HMM assumes that pages have READ permissions by default. Inside
svm_range_validate_and_map, we add READ permissions then add WRITE
permissions if the VMA isn't read-only. This will conflict with regions
that only have PROT_WRITE or have PROT_NONE. When that happens,
svm_range_restore_work will continue to retry, silently, giving the
impression of a hang if pr_debug isn't enabled to show the retries..

If pages don't have READ permissions, simply unmap them and continue. If
they weren't mapped in the first place, this would be a no-op. Since x86
doesn't support write-only, and PROT_NONE doesn't allow reads or writes
anyways, this will allow the svm range validation to continue without
getting stuck in a loop forever on mappings we can't use with HMM.

Signed-off-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index e23b5a0f31f2..521c14c7a789 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1713,6 +1713,29 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 
 			next = min(vma->vm_end, end);
 			npages = (next - addr) >> PAGE_SHIFT;
+			/* HMM requires at least READ permissions. If provided with PROT_NONE,
+			 * unmap the memory. If it's not already mapped, this is a no-op
+			 * If PROT_WRITE is provided without READ, warn first then unmap
+			 */
+			if (!(vma->vm_flags & VM_READ)) {
+				unsigned long e, s;
+
+				svm_range_lock(prange);
+				if (vma->vm_flags & VM_WRITE)
+					pr_debug("VM_WRITE without VM_READ is not supported");
+				s = max(start, prange->start);
+				e = min(end, prange->last);
+				if (e >= s)
+					r = svm_range_unmap_from_gpus(prange, s, e,
+						       KFD_SVM_UNMAP_TRIGGER_UNMAP_FROM_CPU);
+				svm_range_unlock(prange);
+				/* If unmap returns non-zero, we'll bail on the next for loop
+				 * iteration, so just leave r and continue
+				 */
+				addr = next;
+				continue;
+			}
+
 			WRITE_ONCE(p->svms.faulting_task, current);
 			r = amdgpu_hmm_range_get_pages(&prange->notifier, addr, npages,
 						       readonly, owner, NULL,

From 655d6403ad143feffb2d747b9eb9a0ce4e12b713 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 13 Aug 2025 10:55:44 +0800
Subject: [PATCH 0494/1455] drm/amd/vcn: Add late_init callback for VCN v4.0.3
 reset handling

This change reorganizes VCN reset capability detection by:

1. Moving reset mask configuration from sw_init to new late_init phase
2. Adding vcn_v4_0_3_late_init() to properly check for per-queue reset support
3. Only setting soft full reset mask as fallback when per-queue reset isn't supported
4. Removing TODO comment now that queue reset support is implemented

V2: Removed unrelated changes. Keep amdgpu_get_soft_full_reset_mask in place
    and remove TODO comment. (Alex)
v3: set the flags at one place (all in late_init) (Lijo)

Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 019bd362edb2..a63a1e3435ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -134,6 +134,19 @@ static int vcn_v4_0_3_early_init(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
+static int vcn_v4_0_3_late_init(struct amdgpu_ip_block *ip_block)
+{
+	struct amdgpu_device *adev = ip_block->adev;
+
+	adev->vcn.supported_reset =
+		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
+
+	if (amdgpu_dpm_reset_vcn_is_supported(adev))
+		adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
+
+	return 0;
+}
+
 static int vcn_v4_0_3_fw_shared_init(struct amdgpu_device *adev, int inst_idx)
 {
 	struct amdgpu_vcn4_fw_shared *fw_shared;
@@ -211,10 +224,6 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block)
 			adev->vcn.inst[i].pause_dpg_mode = vcn_v4_0_3_pause_dpg_mode;
 	}
 
-	/* TODO: Add queue reset mask when FW fully supports it */
-	adev->vcn.supported_reset =
-		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
-
 	if (amdgpu_sriov_vf(adev)) {
 		r = amdgpu_virt_alloc_mm_table(adev);
 		if (r)
@@ -1871,6 +1880,7 @@ static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev)
 static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = {
 	.name = "vcn_v4_0_3",
 	.early_init = vcn_v4_0_3_early_init,
+	.late_init = vcn_v4_0_3_late_init,
 	.sw_init = vcn_v4_0_3_sw_init,
 	.sw_fini = vcn_v4_0_3_sw_fini,
 	.hw_init = vcn_v4_0_3_hw_init,

From 8a358aaa5d171512c07e8eb87b4600ebc844e6c7 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 12 Aug 2025 13:21:49 +0530
Subject: [PATCH 0495/1455] drm/amd/pm: Free SMUv13.0.6 resources on failure

Free the resources allocated if smu_v13_0_12_tables_init fails.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Fixes: 5bf93e1d6efd ("drm/amd/pm: Add caching to SMUv13.0.12 temp metric")
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c    |  4 +++-
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 12 ++++++++++--
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index ea2682f9d579..32fd0be05cff 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -161,8 +161,10 @@ int smu_v13_0_12_tables_init(struct smu_context *smu)
 	/* Initialize GPU board temperature metrics */
 	ret = smu_table_cache_init(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS,
 				   sizeof(*gpuboard_temp_metrics), 50);
-	if (ret)
+	if (ret) {
+		smu_table_cache_fini(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS);
 		return ret;
+	}
 	cache = &(tables[SMU_TABLE_GPUBOARD_TEMP_METRICS].cache);
 	gpuboard_temp_metrics = (struct amdgpu_gpuboard_temp_metrics_v1_0 *)cache->buffer;
 	smu_cmn_init_gpuboard_temp_metrics(gpuboard_temp_metrics, 1, 0);
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index cc59d884025f..627a8188d868 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -537,6 +537,7 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 	struct smu_table *tables = smu_table->tables;
 	struct amdgpu_device *adev = smu->adev;
 	int gpu_metrcs_size = METRICS_TABLE_SIZE;
+	int ret;
 
 	if (!(adev->flags & AMD_IS_APU))
 		SMU_TABLE_INIT(tables, SMU_TABLE_PMSTATUSLOG, SMU13_TOOL_SIZE,
@@ -573,8 +574,15 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 		return -ENOMEM;
 	}
 
-	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12))
-		return smu_v13_0_12_tables_init(smu);
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) ==
+	    IP_VERSION(13, 0, 12)) {
+		ret = smu_v13_0_12_tables_init(smu);
+		if (ret) {
+			kfree(smu_table->metrics_table);
+			kfree(smu_table->gpu_metrics_table);
+			return ret;
+		}
+	}
 
 	return 0;
 }

From 54f7f3ca982af73512e5dfd541fe20dc72673180 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 8 Aug 2025 13:12:07 -0400
Subject: [PATCH 0496/1455] drm/amdgpu/swm14: Update power limit logic

Take into account the limits from the vbios.  Ported
from the SMU13 code.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4352
Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c  | 30 +++++++++++++++----
 1 file changed, 25 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
index 3aea32baea3d..f32474af90b3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
@@ -1697,9 +1697,11 @@ static int smu_v14_0_2_get_power_limit(struct smu_context *smu,
 				       uint32_t *min_power_limit)
 {
 	struct smu_table_context *table_context = &smu->smu_table;
+	struct smu_14_0_2_powerplay_table *powerplay_table =
+		table_context->power_play_table;
 	PPTable_t *pptable = table_context->driver_pptable;
 	CustomSkuTable_t *skutable = &pptable->CustomSkuTable;
-	uint32_t power_limit;
+	uint32_t power_limit, od_percent_upper = 0, od_percent_lower = 0;
 	uint32_t msg_limit = pptable->SkuTable.MsgLimits.Power[PPT_THROTTLER_PPT0][POWER_SOURCE_AC];
 
 	if (smu_v14_0_get_current_power_limit(smu, &power_limit))
@@ -1712,11 +1714,29 @@ static int smu_v14_0_2_get_power_limit(struct smu_context *smu,
 	if (default_power_limit)
 		*default_power_limit = power_limit;
 
-	if (max_power_limit)
-		*max_power_limit = msg_limit;
+	if (powerplay_table) {
+		if (smu->od_enabled &&
+		    smu_v14_0_2_is_od_feature_supported(smu, PP_OD_FEATURE_PPT_BIT)) {
+			od_percent_upper = pptable->SkuTable.OverDriveLimitsBasicMax.Ppt;
+			od_percent_lower = pptable->SkuTable.OverDriveLimitsBasicMin.Ppt;
+		} else if (smu_v14_0_2_is_od_feature_supported(smu, PP_OD_FEATURE_PPT_BIT)) {
+			od_percent_upper = 0;
+			od_percent_lower = pptable->SkuTable.OverDriveLimitsBasicMin.Ppt;
+		}
+	}
 
-	if (min_power_limit)
-		*min_power_limit = 0;
+	dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n",
+					od_percent_upper, od_percent_lower, power_limit);
+
+	if (max_power_limit) {
+		*max_power_limit = msg_limit * (100 + od_percent_upper);
+		*max_power_limit /= 100;
+	}
+
+	if (min_power_limit) {
+		*min_power_limit = power_limit * (100 + od_percent_lower);
+		*min_power_limit /= 100;
+	}
 
 	return 0;
 }

From 300d3e2d478d0d60dd60ab4800319b6257abf926 Mon Sep 17 00:00:00 2001
From: Alexandre Demers <alexandre.f.demers@gmail.com>
Date: Tue, 12 Aug 2025 20:46:25 -0400
Subject: [PATCH 0497/1455] drm/radeon: fix typos

Various small typos found around.

Signed-off-by: Alexandre Demers <alexandre.f.demers@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_device.c | 4 ++--
 drivers/gpu/drm/radeon/radeon_test.c   | 4 ++--
 drivers/gpu/drm/radeon/radeon_vce.c    | 6 +++---
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 7a3e510327b7..9e35b14e2bf0 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -554,7 +554,7 @@ int radeon_wb_init(struct radeon_device *rdev)
  * cover the whole aperture even if VRAM size is inferior to aperture size
  * Novell bug 204882 + along with lots of ubuntu ones
  *
- * Note 3: when limiting vram it's safe to overwritte real_vram_size because
+ * Note 3: when limiting vram it's safe to overwrite real_vram_size because
  * we are not in case where real_vram_size is inferior to mc_vram_size (ie
  * not affected by bogus hw of Novell bug 204882 + along with lots of ubuntu
  * ones)
@@ -562,7 +562,7 @@ int radeon_wb_init(struct radeon_device *rdev)
  * Note 4: IGP TOM addr should be the same as the aperture addr, we don't
  * explicitly check for that thought.
  *
- * FIXME: when reducing VRAM size align new size on power of 2.
+ * FIXME: when reducing VRAM size, align new size on power of 2.
  */
 void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *mc, u64 base)
 {
diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c
index c9fef9b61ced..818554e60537 100644
--- a/drivers/gpu/drm/radeon/radeon_test.c
+++ b/drivers/gpu/drm/radeon/radeon_test.c
@@ -455,7 +455,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
 
 	r = radeon_ring_lock(rdev, ringC, 64);
 	if (r) {
-		DRM_ERROR("Failed to lock ring B %p\n", ringC);
+		DRM_ERROR("Failed to lock ring C %p\n", ringC);
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
@@ -481,7 +481,7 @@ static void radeon_test_ring_sync2(struct radeon_device *rdev,
 
 	r = radeon_ring_lock(rdev, ringC, 64);
 	if (r) {
-		DRM_ERROR("Failed to lock ring B %p\n", ringC);
+		DRM_ERROR("Failed to lock ring C %p\n", ringC);
 		goto out_cleanup;
 	}
 	radeon_semaphore_emit_signal(rdev, ringC->idx, semaphore);
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index 2355a78e1b69..bdbc1bbe8a9b 100644
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -86,7 +86,7 @@ int radeon_vce_init(struct radeon_device *rdev)
 
 	r = request_firmware(&rdev->vce_fw, fw_name, rdev->dev);
 	if (r) {
-		dev_err(rdev->dev, "radeon_vce: Can't load firmware \"%s\"\n",
+		dev_err(rdev->dev, "radeon_vce: can't load firmware \"%s\"\n",
 			fw_name);
 		return r;
 	}
@@ -126,7 +126,7 @@ int radeon_vce_init(struct radeon_device *rdev)
 
 	rdev->vce.fw_version = (start << 24) | (mid << 16) | (end << 8);
 
-	/* we can only work with this fw version for now */
+	/* we can only work with these fw versions for now */
 	if ((rdev->vce.fw_version != ((40 << 24) | (2 << 16) | (2 << 8))) &&
 	    (rdev->vce.fw_version != ((50 << 24) | (0 << 16) | (1 << 8))) &&
 	    (rdev->vce.fw_version != ((50 << 24) | (1 << 16) | (2 << 8))))
@@ -281,7 +281,7 @@ static void radeon_vce_idle_work_handler(struct work_struct *work)
  *
  * @rdev: radeon_device pointer
  *
- * Make sure VCE is powerd up when we want to use it
+ * Make sure VCE is powered up when we want to use it
  */
 void radeon_vce_note_usage(struct radeon_device *rdev)
 {

From bf4e4b97d0fdc66f04fc19d807e24dd8421b8f11 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 13 Aug 2025 12:14:01 +0530
Subject: [PATCH 0498/1455] drm/amd/display: Add NULL pointer checks in
 dc_stream cursor attribute functions

The function dc_stream_set_cursor_attributes() currently dereferences
the `stream` pointer and nested members `stream->ctx->dc->current_state`
without checking for NULL.

All callers of these functions, such as in
`dcn30_apply_idle_power_optimizations()` and
`amdgpu_dm_plane_handle_cursor_update()`, already perform NULL checks
before calling these functions.

Fixes below:
drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_stream.c:336 dc_stream_program_cursor_attributes()
error: we previously assumed 'stream' could be null (see line 334)

drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_stream.c
    327 bool dc_stream_program_cursor_attributes(
    328         struct dc_stream_state *stream,
    329         const struct dc_cursor_attributes *attributes)
    330 {
    331         struct dc  *dc;
    332         bool reset_idle_optimizations = false;
    333
    334         dc = stream ? stream->ctx->dc : NULL;
                     ^^^^^^
The old code assumed stream could be NULL.

    335
--> 336         if (dc_stream_set_cursor_attributes(stream, attributes)) {
                                                    ^^^^^^
The refactor added an unchecked dereference.

drivers/gpu/drm/amd/amdgpu/../display/dc/core/dc_stream.c
   313  bool dc_stream_set_cursor_attributes(
   314          struct dc_stream_state *stream,
   315          const struct dc_cursor_attributes *attributes)
   316  {
   317          bool result = false;
   318
   319          if (dc_stream_check_cursor_attributes(stream, stream->ctx->dc->current_state, attributes)) {
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Here.
This function used to check for if stream as NULL and return false at
the start. Probably we should add that back.

Fixes: 4465dd0e41e8 ("drm/amd/display: Refactor SubVP cursor limiting logic")
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Cc: Alex Hung <alex.hung@amd.com>
Cc: Alvin Lee <alvin.lee2@amd.com>
Cc: Ray Wu <ray.wu@amd.com>
Cc: Dillon Varone <dillon.varone@amd.com>
Cc: Aurabindo Pillai <aurabindo.pillai@amd.com>
Cc: Roman Li <roman.li@amd.com>
Cc: ChiaHsuan Chung <chiahsuan.chung@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Cc: Daniel Wheeler <daniel.wheeler@amd.com>
Cc: Tom Chung <chiahsuan.chung@amd.com>
Cc: Wenjing Liu <wenjing.liu@amd.com>
Cc: Jun Lei <Jun.Lei@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Dillon Varone <Dillon.varone@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc_stream.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 4d6bc9fd4faa..9ac2d41f8fca 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -316,6 +316,9 @@ bool dc_stream_set_cursor_attributes(
 {
 	bool result = false;
 
+	if (!stream)
+		return false;
+
 	if (dc_stream_check_cursor_attributes(stream, stream->ctx->dc->current_state, attributes)) {
 		stream->cursor_attributes = *attributes;
 		result = true;
@@ -331,7 +334,10 @@ bool dc_stream_program_cursor_attributes(
 	struct dc  *dc;
 	bool reset_idle_optimizations = false;
 
-	dc = stream ? stream->ctx->dc : NULL;
+	if (!stream)
+		return false;
+
+	dc = stream->ctx->dc;
 
 	if (dc_stream_set_cursor_attributes(stream, attributes)) {
 		dc_z10_restore(dc);

From fc4e990a326e608eb8937eba737908c660b7a410 Mon Sep 17 00:00:00 2001
From: Qiang Liu <liuqiang@kylinos.cn>
Date: Tue, 12 Aug 2025 20:30:21 +0800
Subject: [PATCH 0499/1455] drm/amdgpu: remove duplicated argument wptr_va

The duplicate judgment of wptr_va could be removed to simplify the logic

Signed-off-by: Qiang Liu <liuqiang@kylinos.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index c3ace8030530..86cabb20bff1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -567,7 +567,6 @@ int amdgpu_userq_ioctl(struct drm_device *dev, void *data,
 		    args->in.queue_size ||
 		    args->in.rptr_va ||
 		    args->in.wptr_va ||
-		    args->in.wptr_va ||
 		    args->in.mqd ||
 		    args->in.mqd_size)
 			return -EINVAL;

From d2fa0ec6e0aea6ffbd41939d0c7671db16991ca4 Mon Sep 17 00:00:00 2001
From: Chenglei Xie <Chenglei.Xie@amd.com>
Date: Thu, 7 Aug 2025 16:52:34 -0400
Subject: [PATCH 0500/1455] drm/amdgpu: refactor bad_page_work for corner case
 handling

When a poison is consumed on the guest before the guest receives the host's poison creation msg, a corner case may occur to have poison_handler complete processing earlier than it should to cause the guest to hang waiting for the req_bad_pages reply during a VF FLR, resulting in the VM becoming inaccessible in stress tests.

To fix this issue, this patch refactored the mailbox sequence by seperating the bad_page_work into two parts req_bad_pages_work and handle_bad_pages_work.
Old sequence:
  1.Stop data exchange work
  2.Guest sends MB_REQ_RAS_BAD_PAGES to host and keep polling for IDH_RAS_BAD_PAGES_READY
  3.If the IDH_RAS_BAD_PAGES_READY arrives within timeout limit, re-init the data exchange region for updated bad page info
    else timeout with error message
New sequence:
req_bad_pages_work:
  1.Stop data exhange work
  2.Guest sends MB_REQ_RAS_BAD_PAGES to host
Once Guest receives IDH_RAS_BAD_PAGES_READY event
handle_bad_pages_work:
  3.re-init the data exchange region for updated bad page info

Signed-off-by: Chenglei Xie <Chenglei.Xie@amd.com>
Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h |  3 +-
 drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c    | 32 +++++++++++++++++++---
 drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c    | 35 +++++++++++++++++++-----
 drivers/gpu/drm/amd/amdgpu/soc15.c       |  1 -
 4 files changed, 58 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 3da3ebb1d9a1..58accf2259b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -267,7 +267,8 @@ struct amdgpu_virt {
 	struct amdgpu_irq_src		rcv_irq;
 
 	struct work_struct		flr_work;
-	struct work_struct		bad_pages_work;
+	struct work_struct		req_bad_pages_work;
+	struct work_struct		handle_bad_pages_work;
 
 	struct amdgpu_mm_table		mm_table;
 	const struct amdgpu_virt_ops	*ops;
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
index 48101a34e049..9a40107a0869 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c
@@ -292,14 +292,32 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work)
 	}
 }
 
-static void xgpu_ai_mailbox_bad_pages_work(struct work_struct *work)
+static void xgpu_ai_mailbox_req_bad_pages_work(struct work_struct *work)
 {
-	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work);
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, req_bad_pages_work);
 	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
 
 	if (down_read_trylock(&adev->reset_domain->sem)) {
 		amdgpu_virt_fini_data_exchange(adev);
 		amdgpu_virt_request_bad_pages(adev);
+		up_read(&adev->reset_domain->sem);
+	}
+}
+
+/**
+ * xgpu_ai_mailbox_handle_bad_pages_work - Reinitialize the data exchange region to get fresh bad page information
+ * @work: pointer to the work_struct
+ *
+ * This work handler is triggered when bad pages are ready, and it reinitializes
+ * the data exchange region to retrieve updated bad page information from the host.
+ */
+static void xgpu_ai_mailbox_handle_bad_pages_work(struct work_struct *work)
+{
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, handle_bad_pages_work);
+	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+
+	if (down_read_trylock(&adev->reset_domain->sem)) {
+		amdgpu_virt_fini_data_exchange(adev);
 		amdgpu_virt_init_data_exchange(adev);
 		up_read(&adev->reset_domain->sem);
 	}
@@ -327,10 +345,15 @@ static int xgpu_ai_mailbox_rcv_irq(struct amdgpu_device *adev,
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	switch (event) {
+	case IDH_RAS_BAD_PAGES_READY:
+		xgpu_ai_mailbox_send_ack(adev);
+		if (amdgpu_sriov_runtime(adev))
+			schedule_work(&adev->virt.handle_bad_pages_work);
+		break;
 	case IDH_RAS_BAD_PAGES_NOTIFICATION:
 		xgpu_ai_mailbox_send_ack(adev);
 		if (amdgpu_sriov_runtime(adev))
-			schedule_work(&adev->virt.bad_pages_work);
+			schedule_work(&adev->virt.req_bad_pages_work);
 		break;
 	case IDH_UNRECOV_ERR_NOTIFICATION:
 		xgpu_ai_mailbox_send_ack(adev);
@@ -415,7 +438,8 @@ int xgpu_ai_mailbox_get_irq(struct amdgpu_device *adev)
 	}
 
 	INIT_WORK(&adev->virt.flr_work, xgpu_ai_mailbox_flr_work);
-	INIT_WORK(&adev->virt.bad_pages_work, xgpu_ai_mailbox_bad_pages_work);
+	INIT_WORK(&adev->virt.req_bad_pages_work, xgpu_ai_mailbox_req_bad_pages_work);
+	INIT_WORK(&adev->virt.handle_bad_pages_work, xgpu_ai_mailbox_handle_bad_pages_work);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
index f6d8597452ed..457972aa5632 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
@@ -202,9 +202,6 @@ send_request:
 	case IDH_REQ_RAS_CPER_DUMP:
 		event = IDH_RAS_CPER_DUMP_READY;
 		break;
-	case IDH_REQ_RAS_BAD_PAGES:
-		event = IDH_RAS_BAD_PAGES_READY;
-		break;
 	default:
 		break;
 	}
@@ -359,14 +356,32 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 	}
 }
 
-static void xgpu_nv_mailbox_bad_pages_work(struct work_struct *work)
+static void xgpu_nv_mailbox_req_bad_pages_work(struct work_struct *work)
 {
-	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work);
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, req_bad_pages_work);
 	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
 
 	if (down_read_trylock(&adev->reset_domain->sem)) {
 		amdgpu_virt_fini_data_exchange(adev);
 		amdgpu_virt_request_bad_pages(adev);
+		up_read(&adev->reset_domain->sem);
+	}
+}
+
+/**
+ * xgpu_nv_mailbox_handle_bad_pages_work - Reinitialize the data exchange region to get fresh bad page information
+ * @work: pointer to the work_struct
+ *
+ * This work handler is triggered when bad pages are ready, and it reinitializes
+ * the data exchange region to retrieve updated bad page information from the host.
+ */
+static void xgpu_nv_mailbox_handle_bad_pages_work(struct work_struct *work)
+{
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, handle_bad_pages_work);
+	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+
+	if (down_read_trylock(&adev->reset_domain->sem)) {
+		amdgpu_virt_fini_data_exchange(adev);
 		amdgpu_virt_init_data_exchange(adev);
 		up_read(&adev->reset_domain->sem);
 	}
@@ -397,10 +412,15 @@ static int xgpu_nv_mailbox_rcv_irq(struct amdgpu_device *adev,
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 
 	switch (event) {
+	case IDH_RAS_BAD_PAGES_READY:
+		xgpu_nv_mailbox_send_ack(adev);
+		if (amdgpu_sriov_runtime(adev))
+			schedule_work(&adev->virt.handle_bad_pages_work);
+		break;
 	case IDH_RAS_BAD_PAGES_NOTIFICATION:
 		xgpu_nv_mailbox_send_ack(adev);
 		if (amdgpu_sriov_runtime(adev))
-			schedule_work(&adev->virt.bad_pages_work);
+			schedule_work(&adev->virt.req_bad_pages_work);
 		break;
 	case IDH_UNRECOV_ERR_NOTIFICATION:
 		xgpu_nv_mailbox_send_ack(adev);
@@ -485,7 +505,8 @@ int xgpu_nv_mailbox_get_irq(struct amdgpu_device *adev)
 	}
 
 	INIT_WORK(&adev->virt.flr_work, xgpu_nv_mailbox_flr_work);
-	INIT_WORK(&adev->virt.bad_pages_work, xgpu_nv_mailbox_bad_pages_work);
+	INIT_WORK(&adev->virt.req_bad_pages_work, xgpu_nv_mailbox_req_bad_pages_work);
+	INIT_WORK(&adev->virt.handle_bad_pages_work, xgpu_nv_mailbox_handle_bad_pages_work);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
index 9e74c9822e62..9785fada4fa7 100644
--- a/drivers/gpu/drm/amd/amdgpu/soc15.c
+++ b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -741,7 +741,6 @@ static void soc15_reg_base_init(struct amdgpu_device *adev)
 void soc15_set_virt_ops(struct amdgpu_device *adev)
 {
 	adev->virt.ops = &xgpu_ai_virt_ops;
-
 	/* init soc15 reg base early enough so we can
 	 * request request full access for sriov before
 	 * set_ip_blocks. */

From cc8b221172eff821c54d1012bc0c3414638b86cf Mon Sep 17 00:00:00 2001
From: Raag Jadav <raag.jadav@intel.com>
Date: Thu, 14 Aug 2025 14:48:45 +0530
Subject: [PATCH 0501/1455] drm/xe/i2c: Introduce xe_i2c_present()

In preparation of wider usecases which require checking for I2C
controller presence, introduce xe_i2c_present() helper.

v2: Doc fix (Riana)

Suggested-by: Riana Tauro <riana.tauro@intel.com>
Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250814091845.1356135-1-raag.jadav@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_i2c.c | 18 ++++++++++++++++--
 drivers/gpu/drm/xe/xe_i2c.h |  2 ++
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_i2c.c b/drivers/gpu/drm/xe/xe_i2c.c
index bc7dc2099470..044dda517b7c 100644
--- a/drivers/gpu/drm/xe/xe_i2c.c
+++ b/drivers/gpu/drm/xe/xe_i2c.c
@@ -146,6 +146,20 @@ static void xe_i2c_unregister_adapter(struct xe_i2c *i2c)
 	fwnode_remove_software_node(i2c->adapter_node);
 }
 
+/**
+ * xe_i2c_present - I2C controller is present and functional
+ * @xe: xe device instance
+ *
+ * Check whether the I2C controller is present and functioning with valid
+ * endpoint cookie.
+ *
+ * Return: %true if present, %false otherwise.
+ */
+bool xe_i2c_present(struct xe_device *xe)
+{
+	return xe->i2c && xe->i2c->ep.cookie == XE_I2C_EP_COOKIE_DEVICE;
+}
+
 /**
  * xe_i2c_irq_handler: Handler for I2C interrupts
  * @xe: xe device instance
@@ -230,7 +244,7 @@ void xe_i2c_pm_suspend(struct xe_device *xe)
 {
 	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
 
-	if (!xe->i2c || xe->i2c->ep.cookie != XE_I2C_EP_COOKIE_DEVICE)
+	if (!xe_i2c_present(xe))
 		return;
 
 	xe_mmio_rmw32(mmio, I2C_CONFIG_PMCSR, PCI_PM_CTRL_STATE_MASK, (__force u32)PCI_D3hot);
@@ -241,7 +255,7 @@ void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold)
 {
 	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
 
-	if (!xe->i2c || xe->i2c->ep.cookie != XE_I2C_EP_COOKIE_DEVICE)
+	if (!xe_i2c_present(xe))
 		return;
 
 	if (d3cold)
diff --git a/drivers/gpu/drm/xe/xe_i2c.h b/drivers/gpu/drm/xe/xe_i2c.h
index b767ed8ce52b..ecd5f10358e2 100644
--- a/drivers/gpu/drm/xe/xe_i2c.h
+++ b/drivers/gpu/drm/xe/xe_i2c.h
@@ -49,11 +49,13 @@ struct xe_i2c {
 
 #if IS_ENABLED(CONFIG_I2C)
 int xe_i2c_probe(struct xe_device *xe);
+bool xe_i2c_present(struct xe_device *xe);
 void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl);
 void xe_i2c_pm_suspend(struct xe_device *xe);
 void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold);
 #else
 static inline int xe_i2c_probe(struct xe_device *xe) { return 0; }
+static inline bool xe_i2c_present(struct xe_device *xe) { return false; }
 static inline void xe_i2c_irq_handler(struct xe_device *xe, u32 master_ctl) { }
 static inline void xe_i2c_pm_suspend(struct xe_device *xe) { }
 static inline void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold) { }

From 14fcd7361ed1c74dad4062313f2863b643409f6f Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Tue, 12 Aug 2025 22:06:13 +0530
Subject: [PATCH 0502/1455] drm/xe/pm: Disable RPM for SR-IOV VFs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

VFs without native PCIe Power Management (PM) capabilities inherit their
PF's power state as per PCIe specifications(§5.10.1 PCIe Base Spec 7.0).
Enabling Runtime Power Management (RPM) for these VFs trigger unnecessary
driver suspend/resume operations that ultimately perform no PCI-level power
transition.

Since VFs without PM capabilities cannot independently enter low-power
states, the existing RPM workflow becomes redundant:
1. Driver executes full suspend/resume sequence
2. PCI PM transition step becomes no-op
3. VF power state remains tied to PF's status

Disabling RPM for VFs eliminates this redundant processing while
maintaining proper power management through PF dependency. This
optimization ensures VFs follow their PF's power state without superfluous
runtime handling.

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Cc: Anshuman Gupta <anshuman.gupta@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250812163613.9954-1-satyanarayana.k.v.p@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_pm.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index cc489c41ae30..51ed7270994a 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -247,6 +247,10 @@ static void xe_pm_runtime_init(struct xe_device *xe)
 {
 	struct device *dev = xe->drm.dev;
 
+	/* Our current VFs do not support RPM. so, disable it */
+	if (IS_SRIOV_VF(xe))
+		return;
+
 	/*
 	 * Disable the system suspend direct complete optimization.
 	 * We need to ensure that the regular device suspend/resume functions
@@ -371,6 +375,10 @@ static void xe_pm_runtime_fini(struct xe_device *xe)
 {
 	struct device *dev = xe->drm.dev;
 
+	/* Our current VFs do not support RPM. so, disable it */
+	if (IS_SRIOV_VF(xe))
+		return;
+
 	pm_runtime_get_sync(dev);
 	pm_runtime_forbid(dev);
 }

From c8cea4371e5eca30cda8660aabb337747dabc51d Mon Sep 17 00:00:00 2001
From: Lizhi Hou <lizhi.hou@amd.com>
Date: Fri, 15 Aug 2025 10:16:34 -0700
Subject: [PATCH 0503/1455] accel/amdxdna: Add a function to walk hardware
 contexts

Walking hardware contexts created by a process is duplicated in multiple
spots. Add a function, amdxdna_hwctx_walk(), and replace all spots.

hwctx_srcu and dev_lock are good enough to protect hardware context list.
Remove hwctx_lock.

Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250815171634.3417487-1-lizhi.hou@amd.com
---
 drivers/accel/amdxdna/aie2_ctx.c        | 38 ++++++----
 drivers/accel/amdxdna/aie2_message.c    | 21 +++---
 drivers/accel/amdxdna/aie2_pci.c        | 97 ++++++++++++-------------
 drivers/accel/amdxdna/aie2_pci.h        |  2 +-
 drivers/accel/amdxdna/amdxdna_ctx.c     | 26 +++++--
 drivers/accel/amdxdna/amdxdna_ctx.h     |  8 +-
 drivers/accel/amdxdna/amdxdna_pci_drv.c |  7 +-
 drivers/accel/amdxdna/amdxdna_pci_drv.h |  2 -
 8 files changed, 104 insertions(+), 97 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index 910ffb7051f4..420467a5325c 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -133,11 +133,20 @@ static void aie2_hwctx_wait_for_idle(struct amdxdna_hwctx *hwctx)
 	dma_fence_put(fence);
 }
 
+static int aie2_hwctx_suspend_cb(struct amdxdna_hwctx *hwctx, void *arg)
+{
+	struct amdxdna_dev *xdna = hwctx->client->xdna;
+
+	aie2_hwctx_wait_for_idle(hwctx);
+	aie2_hwctx_stop(xdna, hwctx, NULL);
+	aie2_hwctx_status_shift_stop(hwctx);
+
+	return 0;
+}
+
 void aie2_hwctx_suspend(struct amdxdna_client *client)
 {
 	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
 
 	/*
 	 * Command timeout is unlikely. But if it happens, it doesn't
@@ -145,19 +154,20 @@ void aie2_hwctx_suspend(struct amdxdna_client *client)
 	 * and abort all commands.
 	 */
 	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	guard(mutex)(&client->hwctx_lock);
-	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
-		aie2_hwctx_wait_for_idle(hwctx);
-		aie2_hwctx_stop(xdna, hwctx, NULL);
-		aie2_hwctx_status_shift_stop(hwctx);
-	}
+	amdxdna_hwctx_walk(client, NULL, aie2_hwctx_suspend_cb);
 }
 
-void aie2_hwctx_resume(struct amdxdna_client *client)
+static int aie2_hwctx_resume_cb(struct amdxdna_hwctx *hwctx, void *arg)
+{
+	struct amdxdna_dev *xdna = hwctx->client->xdna;
+
+	aie2_hwctx_status_restore(hwctx);
+	return aie2_hwctx_restart(xdna, hwctx);
+}
+
+int aie2_hwctx_resume(struct amdxdna_client *client)
 {
 	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
 
 	/*
 	 * The resume path cannot guarantee that mailbox channel can be
@@ -165,11 +175,7 @@ void aie2_hwctx_resume(struct amdxdna_client *client)
 	 * mailbox channel, error will return.
 	 */
 	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
-	guard(mutex)(&client->hwctx_lock);
-	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
-		aie2_hwctx_status_restore(hwctx);
-		aie2_hwctx_restart(xdna, hwctx);
-	}
+	return amdxdna_hwctx_walk(client, NULL, aie2_hwctx_resume_cb);
 }
 
 static void
diff --git a/drivers/accel/amdxdna/aie2_message.c b/drivers/accel/amdxdna/aie2_message.c
index 82412eec9a4b..9caad083543d 100644
--- a/drivers/accel/amdxdna/aie2_message.c
+++ b/drivers/accel/amdxdna/aie2_message.c
@@ -290,18 +290,25 @@ int aie2_map_host_buf(struct amdxdna_dev_hdl *ndev, u32 context_id, u64 addr, u6
 	return 0;
 }
 
+static int amdxdna_hwctx_col_map(struct amdxdna_hwctx *hwctx, void *arg)
+{
+	u32 *bitmap = arg;
+
+	*bitmap |= GENMASK(hwctx->start_col + hwctx->num_col - 1, hwctx->start_col);
+
+	return 0;
+}
+
 int aie2_query_status(struct amdxdna_dev_hdl *ndev, char __user *buf,
 		      u32 size, u32 *cols_filled)
 {
 	DECLARE_AIE2_MSG(aie_column_info, MSG_OP_QUERY_COL_STATUS);
 	struct amdxdna_dev *xdna = ndev->xdna;
 	struct amdxdna_client *client;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
 	dma_addr_t dma_addr;
 	u32 aie_bitmap = 0;
 	u8 *buff_addr;
-	int ret, idx;
+	int ret;
 
 	buff_addr = dma_alloc_noncoherent(xdna->ddev.dev, size, &dma_addr,
 					  DMA_FROM_DEVICE, GFP_KERNEL);
@@ -309,12 +316,8 @@ int aie2_query_status(struct amdxdna_dev_hdl *ndev, char __user *buf,
 		return -ENOMEM;
 
 	/* Go through each hardware context and mark the AIE columns that are active */
-	list_for_each_entry(client, &xdna->client_list, node) {
-		idx = srcu_read_lock(&client->hwctx_srcu);
-		amdxdna_for_each_hwctx(client, hwctx_id, hwctx)
-			aie_bitmap |= amdxdna_hwctx_col_map(hwctx);
-		srcu_read_unlock(&client->hwctx_srcu, idx);
-	}
+	list_for_each_entry(client, &xdna->client_list, node)
+		amdxdna_hwctx_walk(client, &aie_bitmap, amdxdna_hwctx_col_map);
 
 	*cols_filled = 0;
 	req.dump_buff_addr = dma_addr;
diff --git a/drivers/accel/amdxdna/aie2_pci.c b/drivers/accel/amdxdna/aie2_pci.c
index 6fc3191c3097..16ac0cab4f44 100644
--- a/drivers/accel/amdxdna/aie2_pci.c
+++ b/drivers/accel/amdxdna/aie2_pci.c
@@ -10,6 +10,7 @@
 #include <drm/drm_managed.h>
 #include <drm/drm_print.h>
 #include <drm/gpu_scheduler.h>
+#include <linux/cleanup.h>
 #include <linux/errno.h>
 #include <linux/firmware.h>
 #include <linux/iommu.h>
@@ -465,8 +466,11 @@ static int aie2_hw_resume(struct amdxdna_dev *xdna)
 		return ret;
 	}
 
-	list_for_each_entry(client, &xdna->client_list, node)
-		aie2_hwctx_resume(client);
+	list_for_each_entry(client, &xdna->client_list, node) {
+		ret = aie2_hwctx_resume(client);
+		if (ret)
+			break;
+	}
 
 	return ret;
 }
@@ -779,65 +783,56 @@ static int aie2_get_clock_metadata(struct amdxdna_client *client,
 	return ret;
 }
 
-static int aie2_get_hwctx_status(struct amdxdna_client *client,
-				 struct amdxdna_drm_get_info *args)
+static int aie2_hwctx_status_cb(struct amdxdna_hwctx *hwctx, void *arg)
 {
-	struct amdxdna_drm_query_hwctx __user *buf;
-	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_drm_query_hwctx *tmp;
-	struct amdxdna_client *tmp_client;
-	struct amdxdna_hwctx *hwctx;
-	unsigned long hwctx_id;
-	bool overflow = false;
-	u32 req_bytes = 0;
-	u32 hw_i = 0;
-	int ret = 0;
-	int idx;
+	struct amdxdna_drm_query_hwctx __user *buf, *tmp __free(kfree) = NULL;
+	struct amdxdna_drm_get_info *get_info_args = arg;
 
-	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
+	if (get_info_args->buffer_size < sizeof(*tmp))
+		return -EINVAL;
 
 	tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
 	if (!tmp)
 		return -ENOMEM;
 
-	buf = u64_to_user_ptr(args->buffer);
+	tmp->pid = hwctx->client->pid;
+	tmp->context_id = hwctx->id;
+	tmp->start_col = hwctx->start_col;
+	tmp->num_col = hwctx->num_col;
+	tmp->command_submissions = hwctx->priv->seq;
+	tmp->command_completions = hwctx->priv->completed;
+
+	buf = u64_to_user_ptr(get_info_args->buffer);
+
+	if (copy_to_user(buf, tmp, sizeof(*tmp)))
+		return -EFAULT;
+
+	get_info_args->buffer += sizeof(*tmp);
+	get_info_args->buffer_size -= sizeof(*tmp);
+
+	return 0;
+}
+
+static int aie2_get_hwctx_status(struct amdxdna_client *client,
+				 struct amdxdna_drm_get_info *args)
+{
+	struct amdxdna_dev *xdna = client->xdna;
+	struct amdxdna_drm_get_info info_args;
+	struct amdxdna_client *tmp_client;
+	int ret;
+
+	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
+
+	info_args.buffer = args->buffer;
+	info_args.buffer_size = args->buffer_size;
+
 	list_for_each_entry(tmp_client, &xdna->client_list, node) {
-		idx = srcu_read_lock(&tmp_client->hwctx_srcu);
-		amdxdna_for_each_hwctx(tmp_client, hwctx_id, hwctx) {
-			req_bytes += sizeof(*tmp);
-			if (args->buffer_size < req_bytes) {
-				/* Continue iterating to get the required size */
-				overflow = true;
-				continue;
-			}
-
-			memset(tmp, 0, sizeof(*tmp));
-			tmp->pid = tmp_client->pid;
-			tmp->context_id = hwctx->id;
-			tmp->start_col = hwctx->start_col;
-			tmp->num_col = hwctx->num_col;
-			tmp->command_submissions = hwctx->priv->seq;
-			tmp->command_completions = hwctx->priv->completed;
-
-			if (copy_to_user(&buf[hw_i], tmp, sizeof(*tmp))) {
-				ret = -EFAULT;
-				srcu_read_unlock(&tmp_client->hwctx_srcu, idx);
-				goto out;
-			}
-			hw_i++;
-		}
-		srcu_read_unlock(&tmp_client->hwctx_srcu, idx);
+		ret = amdxdna_hwctx_walk(tmp_client, &info_args, aie2_hwctx_status_cb);
+		if (ret)
+			break;
 	}
 
-	if (overflow) {
-		XDNA_ERR(xdna, "Invalid buffer size. Given: %u Need: %u.",
-			 args->buffer_size, req_bytes);
-		ret = -EINVAL;
-	}
-
-out:
-	kfree(tmp);
-	args->buffer_size = req_bytes;
+	args->buffer_size = (u32)(info_args.buffer - args->buffer);
 	return ret;
 }
 
diff --git a/drivers/accel/amdxdna/aie2_pci.h b/drivers/accel/amdxdna/aie2_pci.h
index 488d8ee568eb..91a8e948f82a 100644
--- a/drivers/accel/amdxdna/aie2_pci.h
+++ b/drivers/accel/amdxdna/aie2_pci.h
@@ -289,7 +289,7 @@ int aie2_hwctx_init(struct amdxdna_hwctx *hwctx);
 void aie2_hwctx_fini(struct amdxdna_hwctx *hwctx);
 int aie2_hwctx_config(struct amdxdna_hwctx *hwctx, u32 type, u64 value, void *buf, u32 size);
 void aie2_hwctx_suspend(struct amdxdna_client *client);
-void aie2_hwctx_resume(struct amdxdna_client *client);
+int aie2_hwctx_resume(struct amdxdna_client *client);
 int aie2_cmd_submit(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, u64 *seq);
 void aie2_hmm_invalidate(struct amdxdna_gem_obj *abo, unsigned long cur_seq);
 
diff --git a/drivers/accel/amdxdna/amdxdna_ctx.c b/drivers/accel/amdxdna/amdxdna_ctx.c
index b47a7f8e9017..4bfe4ef20550 100644
--- a/drivers/accel/amdxdna/amdxdna_ctx.c
+++ b/drivers/accel/amdxdna/amdxdna_ctx.c
@@ -68,14 +68,30 @@ static void amdxdna_hwctx_destroy_rcu(struct amdxdna_hwctx *hwctx,
 	synchronize_srcu(ss);
 
 	/* At this point, user is not able to submit new commands */
-	mutex_lock(&xdna->dev_lock);
 	xdna->dev_info->ops->hwctx_fini(hwctx);
-	mutex_unlock(&xdna->dev_lock);
 
 	kfree(hwctx->name);
 	kfree(hwctx);
 }
 
+int amdxdna_hwctx_walk(struct amdxdna_client *client, void *arg,
+		       int (*walk)(struct amdxdna_hwctx *hwctx, void *arg))
+{
+	struct amdxdna_hwctx *hwctx;
+	unsigned long hwctx_id;
+	int ret = 0, idx;
+
+	idx = srcu_read_lock(&client->hwctx_srcu);
+	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
+		ret = walk(hwctx, arg);
+		if (ret)
+			break;
+	}
+	srcu_read_unlock(&client->hwctx_srcu, idx);
+
+	return ret;
+}
+
 void *amdxdna_cmd_get_payload(struct amdxdna_gem_obj *abo, u32 *size)
 {
 	struct amdxdna_cmd *cmd = abo->mem.kva;
@@ -126,16 +142,12 @@ void amdxdna_hwctx_remove_all(struct amdxdna_client *client)
 	struct amdxdna_hwctx *hwctx;
 	unsigned long hwctx_id;
 
-	mutex_lock(&client->hwctx_lock);
 	amdxdna_for_each_hwctx(client, hwctx_id, hwctx) {
 		XDNA_DBG(client->xdna, "PID %d close HW context %d",
 			 client->pid, hwctx->id);
 		xa_erase(&client->hwctx_xa, hwctx->id);
-		mutex_unlock(&client->hwctx_lock);
 		amdxdna_hwctx_destroy_rcu(hwctx, &client->hwctx_srcu);
-		mutex_lock(&client->hwctx_lock);
 	}
-	mutex_unlock(&client->hwctx_lock);
 }
 
 int amdxdna_drm_create_hwctx_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
@@ -225,6 +237,7 @@ int amdxdna_drm_destroy_hwctx_ioctl(struct drm_device *dev, void *data, struct d
 	if (!drm_dev_enter(dev, &idx))
 		return -ENODEV;
 
+	mutex_lock(&xdna->dev_lock);
 	hwctx = xa_erase(&client->hwctx_xa, args->handle);
 	if (!hwctx) {
 		ret = -EINVAL;
@@ -241,6 +254,7 @@ int amdxdna_drm_destroy_hwctx_ioctl(struct drm_device *dev, void *data, struct d
 
 	XDNA_DBG(xdna, "PID %d destroyed HW context %d", client->pid, args->handle);
 out:
+	mutex_unlock(&xdna->dev_lock);
 	drm_dev_exit(idx);
 	return ret;
 }
diff --git a/drivers/accel/amdxdna/amdxdna_ctx.h b/drivers/accel/amdxdna/amdxdna_ctx.h
index c652229547a3..7cd7a55936f0 100644
--- a/drivers/accel/amdxdna/amdxdna_ctx.h
+++ b/drivers/accel/amdxdna/amdxdna_ctx.h
@@ -139,14 +139,10 @@ amdxdna_cmd_get_state(struct amdxdna_gem_obj *abo)
 void *amdxdna_cmd_get_payload(struct amdxdna_gem_obj *abo, u32 *size);
 int amdxdna_cmd_get_cu_idx(struct amdxdna_gem_obj *abo);
 
-static inline u32 amdxdna_hwctx_col_map(struct amdxdna_hwctx *hwctx)
-{
-	return GENMASK(hwctx->start_col + hwctx->num_col - 1,
-		       hwctx->start_col);
-}
-
 void amdxdna_sched_job_cleanup(struct amdxdna_sched_job *job);
 void amdxdna_hwctx_remove_all(struct amdxdna_client *client);
+int amdxdna_hwctx_walk(struct amdxdna_client *client, void *arg,
+		       int (*walk)(struct amdxdna_hwctx *hwctx, void *arg));
 
 int amdxdna_cmd_submit(struct amdxdna_client *client,
 		       u32 cmd_bo_hdls, u32 *arg_bo_hdls, u32 arg_bo_cnt,
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.c b/drivers/accel/amdxdna/amdxdna_pci_drv.c
index fbca94183f96..8ef5e4f27f5e 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.c
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.c
@@ -81,7 +81,6 @@ static int amdxdna_drm_open(struct drm_device *ddev, struct drm_file *filp)
 		ret = -ENODEV;
 		goto unbind_sva;
 	}
-	mutex_init(&client->hwctx_lock);
 	init_srcu_struct(&client->hwctx_srcu);
 	xa_init_flags(&client->hwctx_xa, XA_FLAGS_ALLOC);
 	mutex_init(&client->mm_lock);
@@ -116,7 +115,6 @@ static void amdxdna_drm_close(struct drm_device *ddev, struct drm_file *filp)
 
 	xa_destroy(&client->hwctx_xa);
 	cleanup_srcu_struct(&client->hwctx_srcu);
-	mutex_destroy(&client->hwctx_lock);
 	mutex_destroy(&client->mm_lock);
 	if (client->dev_heap)
 		drm_gem_object_put(to_gobj(client->dev_heap));
@@ -142,8 +140,8 @@ static int amdxdna_flush(struct file *f, fl_owner_t id)
 
 	mutex_lock(&xdna->dev_lock);
 	list_del_init(&client->node);
-	mutex_unlock(&xdna->dev_lock);
 	amdxdna_hwctx_remove_all(client);
+	mutex_unlock(&xdna->dev_lock);
 
 	drm_dev_exit(idx);
 	return 0;
@@ -330,11 +328,8 @@ static void amdxdna_remove(struct pci_dev *pdev)
 					  struct amdxdna_client, node);
 	while (client) {
 		list_del_init(&client->node);
-		mutex_unlock(&xdna->dev_lock);
-
 		amdxdna_hwctx_remove_all(client);
 
-		mutex_lock(&xdna->dev_lock);
 		client = list_first_entry_or_null(&xdna->client_list,
 						  struct amdxdna_client, node);
 	}
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.h b/drivers/accel/amdxdna/amdxdna_pci_drv.h
index 40bbb3c06320..b6b3b424d1d5 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.h
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.h
@@ -116,8 +116,6 @@ struct amdxdna_device_id {
 struct amdxdna_client {
 	struct list_head		node;
 	pid_t				pid;
-	struct mutex			hwctx_lock; /* protect hwctx */
-	/* do NOT wait this srcu when hwctx_lock is held */
 	struct srcu_struct		hwctx_srcu;
 	struct xarray			hwctx_xa;
 	u32				next_hwctxid;

From a01b704527c28a2fd43a17a85f8996b75ec8492a Mon Sep 17 00:00:00 2001
From: Christoph Manszewski <christoph.manszewski@intel.com>
Date: Wed, 13 Aug 2025 12:12:30 +0200
Subject: [PATCH 0504/1455] drm/xe: Fix vm_bind_ioctl double free bug
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If the argument check during an array bind fails, the bind_ops are freed
twice as seen below. Fix this by setting bind_ops to NULL after freeing.

==================================================================
BUG: KASAN: double-free in xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
Free of addr ffff88813bb9b800 by task xe_vm/14198

CPU: 5 UID: 0 PID: 14198 Comm: xe_vm Not tainted 6.16.0-xe-eudebug-cmanszew+ #520 PREEMPT(full)
Hardware name: Intel Corporation Alder Lake Client Platform/AlderLake-P DDR5 RVP, BIOS ADLPFWI1.R00.2411.A02.2110081023 10/08/2021
Call Trace:
 <TASK>
 dump_stack_lvl+0x82/0xd0
 print_report+0xcb/0x610
 ? __virt_addr_valid+0x19a/0x300
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 kasan_report_invalid_free+0xc8/0xf0
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 check_slab_allocation+0x102/0x130
 kfree+0x10d/0x440
 ? should_fail_ex+0x57/0x2f0
 ? xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 xe_vm_bind_ioctl+0x1b2/0x21f0 [xe]
 ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe]
 ? __lock_acquire+0xab9/0x27f0
 ? lock_acquire+0x165/0x300
 ? drm_dev_enter+0x53/0xe0 [drm]
 ? find_held_lock+0x2b/0x80
 ? drm_dev_exit+0x30/0x50 [drm]
 ? drm_ioctl_kernel+0x128/0x1c0 [drm]
 drm_ioctl_kernel+0x128/0x1c0 [drm]
 ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe]
 ? find_held_lock+0x2b/0x80
 ? __pfx_drm_ioctl_kernel+0x10/0x10 [drm]
 ? should_fail_ex+0x57/0x2f0
 ? __pfx_xe_vm_bind_ioctl+0x10/0x10 [xe]
 drm_ioctl+0x352/0x620 [drm]
 ? __pfx_drm_ioctl+0x10/0x10 [drm]
 ? __pfx_rpm_resume+0x10/0x10
 ? do_raw_spin_lock+0x11a/0x1b0
 ? find_held_lock+0x2b/0x80
 ? __pm_runtime_resume+0x61/0xc0
 ? rcu_is_watching+0x20/0x50
 ? trace_irq_enable.constprop.0+0xac/0xe0
 xe_drm_ioctl+0x91/0xc0 [xe]
 __x64_sys_ioctl+0xb2/0x100
 ? rcu_is_watching+0x20/0x50
 do_syscall_64+0x68/0x2e0
 entry_SYSCALL_64_after_hwframe+0x76/0x7e
RIP: 0033:0x7fa9acb24ded

Fixes: b43e864af0d4 ("drm/xe/uapi: Add DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Christoph Manszewski <christoph.manszewski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250813101231.196632-2-christoph.manszewski@intel.com
---
 drivers/gpu/drm/xe/xe_vm.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index d40d2d43c041..6fd28b205453 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -3442,6 +3442,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
 free_bind_ops:
 	if (args->num_binds > 1)
 		kvfree(*bind_ops);
+	*bind_ops = NULL;
 	return err;
 }
 
@@ -3548,7 +3549,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	struct xe_exec_queue *q = NULL;
 	u32 num_syncs, num_ufence = 0;
 	struct xe_sync_entry *syncs = NULL;
-	struct drm_xe_vm_bind_op *bind_ops;
+	struct drm_xe_vm_bind_op *bind_ops = NULL;
 	struct xe_vma_ops vops;
 	struct dma_fence *fence;
 	int err;

From aa81d55a5997b190660e21e72177f1542c4e3d59 Mon Sep 17 00:00:00 2001
From: Christoph Manszewski <christoph.manszewski@intel.com>
Date: Wed, 13 Aug 2025 12:12:31 +0200
Subject: [PATCH 0505/1455] drm/xe: Untangle vm_bind_ioctl cleanup order

Change the resource cleanup order to align with reverse allocation order.

Cc: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Christoph Manszewski <christoph.manszewski@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250813101231.196632-3-christoph.manszewski@intel.com
---
 drivers/gpu/drm/xe/xe_vm.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 6fd28b205453..c86337e08a55 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -3567,7 +3567,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 		q = xe_exec_queue_lookup(xef, args->exec_queue_id);
 		if (XE_IOCTL_DBG(xe, !q)) {
 			err = -ENOENT;
-			goto put_vm;
+			goto free_bind_ops;
 		}
 
 		if (XE_IOCTL_DBG(xe, !(q->flags & EXEC_QUEUE_FLAG_VM))) {
@@ -3613,7 +3613,7 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 			       __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
 		if (!ops) {
 			err = -ENOMEM;
-			goto release_vm_lock;
+			goto free_bos;
 		}
 	}
 
@@ -3747,17 +3747,20 @@ free_syncs:
 put_obj:
 	for (i = 0; i < args->num_binds; ++i)
 		xe_bo_put(bos[i]);
+
+	kvfree(ops);
+free_bos:
+	kvfree(bos);
 release_vm_lock:
 	up_write(&vm->lock);
 put_exec_queue:
 	if (q)
 		xe_exec_queue_put(q);
-put_vm:
-	xe_vm_put(vm);
-	kvfree(bos);
-	kvfree(ops);
+free_bind_ops:
 	if (args->num_binds > 1)
 		kvfree(bind_ops);
+put_vm:
+	xe_vm_put(vm);
 	return err;
 }
 

From b85bb2d677153d990924d31be9416166d22382eb Mon Sep 17 00:00:00 2001
From: Simon Richter <Simon.Richter@hogyros.de>
Date: Mon, 18 Aug 2025 15:47:35 +0900
Subject: [PATCH 0506/1455] drm/xe: Make page size consistent in loop

If PAGE_SIZE != XE_PAGE_SIZE (which is currently locked behind
CONFIG_BROKEN), this would generate the wrong number of PDEs.

Since these PDEs are consumed by the GPU, the GPU page size needs to be
used.

Signed-off-by: Simon Richter <Simon.Richter@hogyros.de>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250818064806.2835-1-Simon.Richter@hogyros.de
---
 drivers/gpu/drm/xe/xe_migrate.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index ddfad7506a82..57e6d5a8ac39 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -280,7 +280,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 	}
 
 	/* Write PDE's that point to our BO. */
-	for (i = 0; i < map_ofs / PAGE_SIZE; i++) {
+	for (i = 0; i < map_ofs / XE_PAGE_SIZE; i++) {
 		entry = vm->pt_ops->pde_encode_bo(bo, (u64)i * XE_PAGE_SIZE);
 
 		xe_map_wr(xe, &bo->vmap, map_ofs + XE_PAGE_SIZE +

From f3358399ce7151179a898bf02c1f3b68f2c4ef36 Mon Sep 17 00:00:00 2001
From: Rafal Ostrowski <rostrows@amd.com>
Date: Fri, 11 Jul 2025 13:13:27 +0200
Subject: [PATCH 0507/1455] drm/amd/display: Add LSDMA Linear Sub Window Copy
 support

[WHAT]
Add support for LSDMA Linear Sub Window Copy command.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Rafal Ostrowski <rostrows@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 58 ++++++++++++++++++--
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h | 36 +++++++++++-
 2 files changed, 88 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 714c468c010d..55b362196612 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -2010,11 +2010,12 @@ bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv)
 	return result;
 }
 
-bool dmub_lsdma_send_linear_copy_packet(
+bool dmub_lsdma_send_linear_copy_command(
 	struct dc_dmub_srv *dc_dmub_srv,
 	uint64_t src_addr,
 	uint64_t dst_addr,
-	uint32_t count)
+	uint32_t count
+)
 {
 	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
 	union dmub_rb_cmd cmd;
@@ -2042,9 +2043,54 @@ bool dmub_lsdma_send_linear_copy_packet(
 	return result;
 }
 
+bool dmub_lsdma_send_linear_sub_window_copy_command(
+	struct dc_dmub_srv *dc_dmub_srv,
+	struct lsdma_linear_sub_window_copy_params copy_data
+)
+{
+	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
+	union dmub_rb_cmd cmd;
+	enum dm_dmub_wait_type wait_type;
+	struct dmub_cmd_lsdma_data *lsdma_data = &cmd.lsdma.lsdma_data;
+	bool result;
+
+	memset(&cmd, 0, sizeof(cmd));
+
+	cmd.cmd_common.header.type     = DMUB_CMD__LSDMA;
+	cmd.cmd_common.header.sub_type = DMUB_CMD__LSDMA_LINEAR_SUB_WINDOW_COPY;
+	wait_type                      = DM_DMUB_WAIT_TYPE_NO_WAIT;
+
+	lsdma_data->u.linear_sub_window_copy_data.tmz              = copy_data.tmz;
+	lsdma_data->u.linear_sub_window_copy_data.element_size     = copy_data.element_size;
+	lsdma_data->u.linear_sub_window_copy_data.src_lo           = copy_data.src_lo;
+	lsdma_data->u.linear_sub_window_copy_data.src_hi           = copy_data.src_hi;
+	lsdma_data->u.linear_sub_window_copy_data.src_x            = copy_data.src_x;
+	lsdma_data->u.linear_sub_window_copy_data.src_y            = copy_data.src_y;
+	lsdma_data->u.linear_sub_window_copy_data.src_pitch        = copy_data.src_pitch - 1;
+	lsdma_data->u.linear_sub_window_copy_data.src_slice_pitch  = copy_data.src_slice_pitch - 1;
+	lsdma_data->u.linear_sub_window_copy_data.dst_lo           = copy_data.dst_lo;
+	lsdma_data->u.linear_sub_window_copy_data.dst_hi           = copy_data.dst_hi;
+	lsdma_data->u.linear_sub_window_copy_data.dst_x            = copy_data.dst_x;
+	lsdma_data->u.linear_sub_window_copy_data.dst_y            = copy_data.dst_y;
+	lsdma_data->u.linear_sub_window_copy_data.dst_pitch        = copy_data.dst_pitch - 1;
+	lsdma_data->u.linear_sub_window_copy_data.dst_slice_pitch  = copy_data.dst_slice_pitch - 1;
+	lsdma_data->u.linear_sub_window_copy_data.rect_x           = copy_data.rect_x - 1;
+	lsdma_data->u.linear_sub_window_copy_data.rect_y           = copy_data.rect_y - 1;
+	lsdma_data->u.linear_sub_window_copy_data.src_cache_policy = copy_data.src_cache_policy;
+	lsdma_data->u.linear_sub_window_copy_data.dst_cache_policy = copy_data.dst_cache_policy;
+
+	result = dc_wake_and_execute_dmub_cmd(dc_ctx, &cmd, wait_type);
+
+	if (!result)
+		DC_ERROR("LSDMA Linear Sub Window Copy failed in DMUB");
+
+	return result;
+}
+
 bool dmub_lsdma_send_tiled_to_tiled_copy_command(
 	struct dc_dmub_srv *dc_dmub_srv,
-	struct lsdma_send_tiled_to_tiled_copy_command_params params)
+	struct lsdma_send_tiled_to_tiled_copy_command_params params
+)
 {
 	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
 	union dmub_rb_cmd cmd;
@@ -2097,7 +2143,8 @@ bool dmub_lsdma_send_pio_copy_command(
 	uint64_t src_addr,
 	uint64_t dst_addr,
 	uint32_t byte_count,
-	uint32_t overlap_disable)
+	uint32_t overlap_disable
+)
 {
 	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
 	union dmub_rb_cmd cmd;
@@ -2130,7 +2177,8 @@ bool dmub_lsdma_send_pio_constfill_command(
 	struct dc_dmub_srv *dc_dmub_srv,
 	uint64_t dst_addr,
 	uint32_t byte_count,
-	uint32_t data)
+	uint32_t data
+)
 {
 	struct dc_context *dc_ctx = dc_dmub_srv->ctx;
 	union dmub_rb_cmd cmd;
diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
index 8ea320f21269..7ef93444ef3c 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h
@@ -211,11 +211,45 @@ void dc_dmub_srv_fams2_passthrough_flip(
 		int surface_count);
 
 bool dmub_lsdma_init(struct dc_dmub_srv *dc_dmub_srv);
-bool dmub_lsdma_send_linear_copy_packet(
+bool dmub_lsdma_send_linear_copy_command(
 	struct dc_dmub_srv *dc_dmub_srv,
 	uint64_t src_addr,
 	uint64_t dst_addr,
 	uint32_t count);
+
+struct lsdma_linear_sub_window_copy_params {
+	uint32_t src_lo;
+	uint32_t src_hi;
+
+	uint32_t dst_lo;
+	uint32_t dst_hi;
+
+	uint32_t src_x        : 16;
+	uint32_t src_y        : 16;
+
+	uint32_t dst_x        : 16;
+	uint32_t dst_y        : 16;
+
+	uint32_t rect_x       : 16;
+	uint32_t rect_y       : 16;
+
+	uint32_t src_pitch    : 16;
+	uint32_t dst_pitch    : 16;
+
+	uint32_t src_slice_pitch;
+	uint32_t dst_slice_pitch;
+
+	uint32_t tmz              : 1;
+	uint32_t element_size     : 3;
+	uint32_t src_cache_policy : 3;
+	uint32_t dst_cache_policy : 3;
+	uint32_t padding          : 22;
+};
+
+bool dmub_lsdma_send_linear_sub_window_copy_command(
+	struct dc_dmub_srv *dc_dmub_srv,
+	struct lsdma_linear_sub_window_copy_params copy_data
+);
 bool dmub_lsdma_send_pio_copy_command(
 	struct dc_dmub_srv *dc_dmub_srv,
 	uint64_t src_addr,

From bdd34a647e842695f86294efa442f5c4d162d705 Mon Sep 17 00:00:00 2001
From: Lohita Mudimela <lohita.mudimela@amd.com>
Date: Thu, 17 Jul 2025 16:43:36 +0530
Subject: [PATCH 0508/1455] drm/amd/display: Refactor DPP enum for backwards
 compatibility

[WHY]
Conflict for enum type in DPP source files.

[HOW]
Refactor DPP source files to resolve the enum conflicts.

Reviewed-by: Ilya Bakoulin <ilya.bakoulin@amd.com>
Reviewed-by: Martin Leung <martin.leung@amd.com>
Signed-off-by: Lohita Mudimela <lohita.mudimela@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dpp/dcn401/dcn401_dpp.h    | 10 ++++++
 .../display/dc/dpp/dcn401/dcn401_dpp_dscl.c   | 36 +++++++------------
 2 files changed, 23 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h
index 5a6a861402b3..5f6b431ec398 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp.h
@@ -673,6 +673,16 @@ struct dcn401_dpp {
 	struct pwl_params pwl_data;
 };
 
+enum dcn401_dscl_mode_sel {
+	DCN401_DSCL_MODE_SCALING_444_BYPASS = 0,
+	DCN401_DSCL_MODE_SCALING_444_RGB_ENABLE = 1,
+	DCN401_DSCL_MODE_SCALING_444_YCBCR_ENABLE = 2,
+	DCN401_DSCL_MODE_SCALING_420_YCBCR_ENABLE = 3,
+	DCN401_DSCL_MODE_SCALING_420_LUMA_BYPASS = 4,
+	DCN401_DSCL_MODE_SCALING_420_CHROMA_BYPASS = 5,
+	DCN401_DSCL_MODE_DSCL_BYPASS = 6
+};
+
 bool dpp401_construct(struct dcn401_dpp *dpp401,
 	struct dc_context *ctx,
 	uint32_t inst,
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c
index 2f92e7d4981b..6df3419f825f 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn401/dcn401_dpp_dscl.c
@@ -78,16 +78,6 @@ enum dscl_autocal_mode {
 	AUTOCAL_MODE_AUTOREPLICATE = 3
 };
 
-enum dscl_mode_sel {
-	DSCL_MODE_SCALING_444_BYPASS = 0,
-	DSCL_MODE_SCALING_444_RGB_ENABLE = 1,
-	DSCL_MODE_SCALING_444_YCBCR_ENABLE = 2,
-	DSCL_MODE_SCALING_420_YCBCR_ENABLE = 3,
-	DSCL_MODE_SCALING_420_LUMA_BYPASS = 4,
-	DSCL_MODE_SCALING_420_CHROMA_BYPASS = 5,
-	DSCL_MODE_DSCL_BYPASS = 6
-};
-
 static int dpp401_dscl_get_pixel_depth_val(enum lb_pixel_depth depth)
 {
 	if (depth == LB_PIXEL_DEPTH_30BPP)
@@ -122,7 +112,7 @@ static bool dpp401_dscl_is_420_format(enum pixel_format format)
 		return false;
 }
 
-static enum dscl_mode_sel dpp401_dscl_get_dscl_mode(
+static enum dcn401_dscl_mode_sel dpp401_dscl_get_dscl_mode(
 		struct dpp *dpp_base,
 		const struct scaler_data *data,
 		bool dbg_always_scale)
@@ -132,7 +122,7 @@ static enum dscl_mode_sel dpp401_dscl_get_dscl_mode(
 	if (dpp_base->caps->dscl_data_proc_format == DSCL_DATA_PRCESSING_FIXED_FORMAT) {
 		/* DSCL is processing data in fixed format */
 		if (data->format == PIXEL_FORMAT_FP16)
-			return DSCL_MODE_DSCL_BYPASS;
+			return DCN401_DSCL_MODE_DSCL_BYPASS;
 	}
 
 	if (data->ratios.horz.value == one
@@ -140,20 +130,20 @@ static enum dscl_mode_sel dpp401_dscl_get_dscl_mode(
 			&& data->ratios.horz_c.value == one
 			&& data->ratios.vert_c.value == one
 			&& !dbg_always_scale)
-		return DSCL_MODE_SCALING_444_BYPASS;
+		return DCN401_DSCL_MODE_SCALING_444_BYPASS;
 
 	if (!dpp401_dscl_is_420_format(data->format)) {
 		if (dpp401_dscl_is_video_format(data->format))
-			return DSCL_MODE_SCALING_444_YCBCR_ENABLE;
+			return DCN401_DSCL_MODE_SCALING_444_YCBCR_ENABLE;
 		else
-			return DSCL_MODE_SCALING_444_RGB_ENABLE;
+			return DCN401_DSCL_MODE_SCALING_444_RGB_ENABLE;
 	}
 	if (data->ratios.horz.value == one && data->ratios.vert.value == one)
-		return DSCL_MODE_SCALING_420_LUMA_BYPASS;
+		return DCN401_DSCL_MODE_SCALING_420_LUMA_BYPASS;
 	if (data->ratios.horz_c.value == one && data->ratios.vert_c.value == one)
-		return DSCL_MODE_SCALING_420_CHROMA_BYPASS;
+		return DCN401_DSCL_MODE_SCALING_420_CHROMA_BYPASS;
 
-	return DSCL_MODE_SCALING_420_YCBCR_ENABLE;
+	return DCN401_DSCL_MODE_SCALING_420_YCBCR_ENABLE;
 }
 
 static void dpp401_power_on_dscl(
@@ -1071,7 +1061,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base,
 	uint32_t v_num_taps_c = scl_data->taps.v_taps_c - 1;
 	uint32_t h_num_taps = scl_data->taps.h_taps - 1;
 	uint32_t h_num_taps_c = scl_data->taps.h_taps_c - 1;
-	enum dscl_mode_sel dscl_mode = dpp401_dscl_get_dscl_mode(
+	enum dcn401_dscl_mode_sel dscl_mode = dpp401_dscl_get_dscl_mode(
 			dpp_base, scl_data, dpp_base->ctx->dc->debug.always_scale);
 	bool ycbcr = scl_data->format >= PIXEL_FORMAT_VIDEO_BEGIN
 				&& scl_data->format <= PIXEL_FORMAT_VIDEO_END;
@@ -1102,7 +1092,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base,
 	dpp->scl_data = *scl_data;
 
 	if ((dpp->base.ctx->dc->config.use_spl) && (!dpp->base.ctx->dc->debug.disable_spl)) {
-		dscl_mode = (enum dscl_mode_sel) scl_data->dscl_prog_data.dscl_mode;
+		dscl_mode = (enum dcn401_dscl_mode_sel) scl_data->dscl_prog_data.dscl_mode;
 		rect = (struct rect *)&scl_data->dscl_prog_data.recout;
 		mpc_width = scl_data->dscl_prog_data.mpc_size.width;
 		mpc_height = scl_data->dscl_prog_data.mpc_size.height;
@@ -1112,7 +1102,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base,
 		h_num_taps_c = scl_data->dscl_prog_data.taps.h_taps_c;
 	}
 	if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.dscl) {
-		if (dscl_mode != DSCL_MODE_DSCL_BYPASS)
+		if (dscl_mode != DCN401_DSCL_MODE_DSCL_BYPASS)
 			dpp401_power_on_dscl(dpp_base, true);
 	}
 
@@ -1139,7 +1129,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base,
 	/* SCL mode */
 	REG_UPDATE(SCL_MODE, DSCL_MODE, dscl_mode);
 
-	if (dscl_mode == DSCL_MODE_DSCL_BYPASS) {
+	if (dscl_mode == DCN401_DSCL_MODE_DSCL_BYPASS) {
 		if (dpp_base->ctx->dc->debug.enable_mem_low_power.bits.dscl)
 			dpp401_power_on_dscl(dpp_base, false);
 		return;
@@ -1149,7 +1139,7 @@ void dpp401_dscl_set_scaler_manual_scale(struct dpp *dpp_base,
 	lb_config =  dpp401_dscl_find_lb_memory_config(dpp, scl_data);
 	dpp401_dscl_set_lb(dpp, &scl_data->lb_params, lb_config);
 
-	if (dscl_mode == DSCL_MODE_SCALING_444_BYPASS) {
+	if (dscl_mode == DCN401_DSCL_MODE_SCALING_444_BYPASS) {
 		if (dpp->base.ctx->dc->config.prefer_easf)
 			dpp401_dscl_disable_easf(dpp_base, scl_data);
 		dpp401_dscl_program_isharp(dpp_base, scl_data, program_isharp_1dlut, &bs_coeffs_updated);

From 63a16df7eb58f0ef699c9f08f3276791261b853a Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Tue, 29 Jul 2025 18:08:24 +0530
Subject: [PATCH 0509/1455] drm/amd/display: Optimize
 amdgpu_dm_atomic_commit_tail()

[WHY]
The first two loops of for_each_oldnew_connector_in_state() both operate
on an HDCP queue.  If one isn't setup then each connector is iterated but
skipped TWICE.  This is wasteful for the majority of cases.

[HOW]
Combine the two HDCP related loops of for_each_oldnew_connector_in_state()
and check for the HDCP workqueue before even running either of them. This
should avoid running the functions in most cases, and if HDCP is setup only
run once.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 118 +++++++++---------
 1 file changed, 56 insertions(+), 62 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c0d31c26850d..e8cfae2bd2ae 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -10129,69 +10129,40 @@ static void dm_set_writeback(struct amdgpu_display_manager *dm,
 	drm_writeback_queue_job(wb_conn, new_con_state);
 }
 
-/**
- * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation.
- * @state: The atomic state to commit
- *
- * This will tell DC to commit the constructed DC state from atomic_check,
- * programming the hardware. Any failures here implies a hardware failure, since
- * atomic check should have filtered anything non-kosher.
- */
-static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
+static void amdgpu_dm_update_hdcp(struct drm_atomic_state *state)
 {
-	struct drm_device *dev = state->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_display_manager *dm = &adev->dm;
-	struct dm_atomic_state *dm_state;
-	struct dc_state *dc_state = NULL;
-	u32 i, j;
-	struct drm_crtc *crtc;
-	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
-	unsigned long flags;
-	bool wait_for_vblank = true;
-	struct drm_connector *connector;
 	struct drm_connector_state *old_con_state, *new_con_state;
-	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
-	int crtc_disable_count = 0;
+	struct drm_device *dev = state->dev;
+	struct drm_connector *connector;
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	int i;
 
-	trace_amdgpu_dm_atomic_commit_tail_begin(state);
-
-	drm_atomic_helper_update_legacy_modeset_state(dev, state);
-	drm_dp_mst_atomic_wait_for_dependencies(state);
-
-	dm_state = dm_atomic_get_new_state(state);
-	if (dm_state && dm_state->context) {
-		dc_state = dm_state->context;
-		amdgpu_dm_commit_streams(state, dc_state);
-	}
+	if (!adev->dm.hdcp_workqueue)
+		return;
 
 	for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) {
 		struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state);
 		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc);
+		struct drm_crtc_state *old_crtc_state, *new_crtc_state;
+		struct dm_crtc_state *dm_new_crtc_state;
 		struct amdgpu_dm_connector *aconnector;
 
-		if (connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK)
+		if (!connector || connector->connector_type == DRM_MODE_CONNECTOR_WRITEBACK)
 			continue;
 
 		aconnector = to_amdgpu_dm_connector(connector);
 
-		if (!adev->dm.hdcp_workqueue)
-			continue;
+		drm_dbg(dev, "[HDCP_DM] -------------- i : %x ----------\n", i);
 
-		pr_debug("[HDCP_DM] -------------- i : %x ----------\n", i);
-
-		if (!connector)
-			continue;
-
-		pr_debug("[HDCP_DM] connector->index: %x connect_status: %x dpms: %x\n",
+		drm_dbg(dev, "[HDCP_DM] connector->index: %x connect_status: %x dpms: %x\n",
 			connector->index, connector->status, connector->dpms);
-		pr_debug("[HDCP_DM] state protection old: %x new: %x\n",
+		drm_dbg(dev, "[HDCP_DM] state protection old: %x new: %x\n",
 			old_con_state->content_protection, new_con_state->content_protection);
 
 		if (aconnector->dc_sink) {
 			if (aconnector->dc_sink->sink_signal != SIGNAL_TYPE_VIRTUAL &&
 				aconnector->dc_sink->sink_signal != SIGNAL_TYPE_NONE) {
-				pr_debug("[HDCP_DM] pipe_ctx dispname=%s\n",
+				drm_dbg(dev, "[HDCP_DM] pipe_ctx dispname=%s\n",
 				aconnector->dc_sink->edid_caps.display_name);
 			}
 		}
@@ -10205,7 +10176,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 		}
 
 		if (old_crtc_state)
-			pr_debug("old crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n",
+			drm_dbg(dev, "old crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n",
 			old_crtc_state->enable,
 			old_crtc_state->active,
 			old_crtc_state->mode_changed,
@@ -10213,29 +10184,13 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 			old_crtc_state->connectors_changed);
 
 		if (new_crtc_state)
-			pr_debug("NEW crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n",
+			drm_dbg(dev, "NEW crtc en: %x a: %x m: %x a-chg: %x c-chg: %x\n",
 			new_crtc_state->enable,
 			new_crtc_state->active,
 			new_crtc_state->mode_changed,
 			new_crtc_state->active_changed,
 			new_crtc_state->connectors_changed);
-	}
 
-	for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) {
-		struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state);
-		struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc);
-		struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
-
-		if (!adev->dm.hdcp_workqueue)
-			continue;
-
-		new_crtc_state = NULL;
-		old_crtc_state = NULL;
-
-		if (acrtc) {
-			new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base);
-			old_crtc_state = drm_atomic_get_old_crtc_state(state, &acrtc->base);
-		}
 
 		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
 
@@ -10279,7 +10234,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 				new_con_state->content_protection >= DRM_MODE_CONTENT_PROTECTION_DESIRED)
 				enable_encryption = true;
 
-			drm_info(adev_to_drm(adev), "[HDCP_DM] hdcp_update_display enable_encryption = %x\n", enable_encryption);
+			drm_info(dev, "[HDCP_DM] hdcp_update_display enable_encryption = %x\n", enable_encryption);
 
 			if (aconnector->dc_link)
 				hdcp_update_display(
@@ -10287,6 +10242,45 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 					new_con_state->hdcp_content_type, enable_encryption);
 		}
 	}
+}
+
+/**
+ * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation.
+ * @state: The atomic state to commit
+ *
+ * This will tell DC to commit the constructed DC state from atomic_check,
+ * programming the hardware. Any failures here implies a hardware failure, since
+ * atomic check should have filtered anything non-kosher.
+ */
+static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
+{
+	struct drm_device *dev = state->dev;
+	struct amdgpu_device *adev = drm_to_adev(dev);
+	struct amdgpu_display_manager *dm = &adev->dm;
+	struct dm_atomic_state *dm_state;
+	struct dc_state *dc_state = NULL;
+	u32 i, j;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
+	unsigned long flags;
+	bool wait_for_vblank = true;
+	struct drm_connector *connector;
+	struct drm_connector_state *old_con_state, *new_con_state;
+	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
+	int crtc_disable_count = 0;
+
+	trace_amdgpu_dm_atomic_commit_tail_begin(state);
+
+	drm_atomic_helper_update_legacy_modeset_state(dev, state);
+	drm_dp_mst_atomic_wait_for_dependencies(state);
+
+	dm_state = dm_atomic_get_new_state(state);
+	if (dm_state && dm_state->context) {
+		dc_state = dm_state->context;
+		amdgpu_dm_commit_streams(state, dc_state);
+	}
+
+	amdgpu_dm_update_hdcp(state);
 
 	/* Handle connector state changes */
 	for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) {

From de0f5be79182c6ac53133d750cc965fe9621f57b Mon Sep 17 00:00:00 2001
From: Clay King <clayking@amd.com>
Date: Wed, 30 Jul 2025 10:23:19 -0400
Subject: [PATCH 0510/1455] drm/amd/display: Delete unused functions

[WHAT]
Removing unused code

Reviewed-by: Joshua Aberback <joshua.aberback@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h       | 15 ---------------
 .../drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c    |  8 --------
 .../drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h    |  5 -----
 3 files changed, 28 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
index 14f0304e3eb9..22960ee03dee 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mpc.h
@@ -1069,21 +1069,6 @@ struct mpc_funcs {
 	*/
 	void (*program_lut_mode)(struct mpc *mpc, const enum MCM_LUT_ID id, const enum MCM_LUT_XABLE xable,
 			bool lut_bank_a, int mpcc_id);
-	/**
-	* @program_3dlut_size:
-	*
-	* Program 3D LUT size.
-	*
-	* Parameters:
-	* - [in/out] mpc - MPC context.
-	* - [in] is_17x17x17 - is 3dlut 17x17x17
-	* - [in] mpcc_id
-	*
-	* Return:
-	*
-	* void
-	*/
-	void (*program_3dlut_size)(struct mpc *mpc, bool is_17x17x17, int mpcc_id);
 
 	/**
 	 * @mcm:
diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c
index f3fb3fe13757..e1a0308dee57 100644
--- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c
+++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.c
@@ -287,13 +287,6 @@ void mpc401_program_lut_read_write_control(struct mpc *mpc, const enum MCM_LUT_I
 	}
 }
 
-void mpc401_program_3dlut_size(struct mpc *mpc, bool is_17x17x17, int mpcc_id)
-{
-	struct dcn401_mpc *mpc401 = TO_DCN401_MPC(mpc);
-
-	REG_UPDATE(MPCC_MCM_3DLUT_MODE[mpcc_id], MPCC_MCM_3DLUT_SIZE, is_17x17x17 ? 0 : 1);
-}
-
 void mpc_program_gamut_remap(
 	struct mpc *mpc,
 	unsigned int mpcc_id,
@@ -611,7 +604,6 @@ static const struct mpc_funcs dcn401_mpc_funcs = {
 	.populate_lut = mpc401_populate_lut,
 	.program_lut_read_write_control = mpc401_program_lut_read_write_control,
 	.program_lut_mode = mpc401_program_lut_mode,
-	.program_3dlut_size = mpc401_program_3dlut_size,
 };
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h
index eb0c68d0b0c7..fdc42f8ab3ff 100644
--- a/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h
+++ b/drivers/gpu/drm/amd/display/dc/mpc/dcn401/dcn401_mpc.h
@@ -221,11 +221,6 @@ void mpc401_program_lut_read_write_control(
 		bool lut_bank_a,
 		int mpcc_id);
 
-void mpc401_program_3dlut_size(
-		struct mpc *mpc,
-		bool is_17x17x17,
-		int mpcc_id);
-
 void mpc401_set_gamut_remap(
 	struct mpc *mpc,
 	int mpcc_id,

From 421507ad489b90fe81ae3c5fef60f6cd98093eb3 Mon Sep 17 00:00:00 2001
From: Rafal Ostrowski <rostrows@amd.com>
Date: Tue, 5 Aug 2025 14:53:37 +0200
Subject: [PATCH 0511/1455] drm/amd/display: Align LSDMA commands fields

[WHY]
DC LSDMA functions had to remember to extract 1 from several fields
to be compliant with DMUB LSDMA commands interface.
Now this logic is moved to DMUB.

[HOW]
Moved extraction by 1 in several fields of LSDMA commands to DMUB.
Changed DC to not do it.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Rafal Ostrowski <rostrows@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c | 24 ++++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
index 55b362196612..53a088ebddef 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c
@@ -2066,16 +2066,16 @@ bool dmub_lsdma_send_linear_sub_window_copy_command(
 	lsdma_data->u.linear_sub_window_copy_data.src_hi           = copy_data.src_hi;
 	lsdma_data->u.linear_sub_window_copy_data.src_x            = copy_data.src_x;
 	lsdma_data->u.linear_sub_window_copy_data.src_y            = copy_data.src_y;
-	lsdma_data->u.linear_sub_window_copy_data.src_pitch        = copy_data.src_pitch - 1;
-	lsdma_data->u.linear_sub_window_copy_data.src_slice_pitch  = copy_data.src_slice_pitch - 1;
+	lsdma_data->u.linear_sub_window_copy_data.src_pitch        = copy_data.src_pitch;
+	lsdma_data->u.linear_sub_window_copy_data.src_slice_pitch  = copy_data.src_slice_pitch;
 	lsdma_data->u.linear_sub_window_copy_data.dst_lo           = copy_data.dst_lo;
 	lsdma_data->u.linear_sub_window_copy_data.dst_hi           = copy_data.dst_hi;
 	lsdma_data->u.linear_sub_window_copy_data.dst_x            = copy_data.dst_x;
 	lsdma_data->u.linear_sub_window_copy_data.dst_y            = copy_data.dst_y;
-	lsdma_data->u.linear_sub_window_copy_data.dst_pitch        = copy_data.dst_pitch - 1;
-	lsdma_data->u.linear_sub_window_copy_data.dst_slice_pitch  = copy_data.dst_slice_pitch - 1;
-	lsdma_data->u.linear_sub_window_copy_data.rect_x           = copy_data.rect_x - 1;
-	lsdma_data->u.linear_sub_window_copy_data.rect_y           = copy_data.rect_y - 1;
+	lsdma_data->u.linear_sub_window_copy_data.dst_pitch        = copy_data.dst_pitch;
+	lsdma_data->u.linear_sub_window_copy_data.dst_slice_pitch  = copy_data.dst_slice_pitch;
+	lsdma_data->u.linear_sub_window_copy_data.rect_x           = copy_data.rect_x;
+	lsdma_data->u.linear_sub_window_copy_data.rect_y           = copy_data.rect_y;
 	lsdma_data->u.linear_sub_window_copy_data.src_cache_policy = copy_data.src_cache_policy;
 	lsdma_data->u.linear_sub_window_copy_data.dst_cache_policy = copy_data.dst_cache_policy;
 
@@ -2112,20 +2112,20 @@ bool dmub_lsdma_send_tiled_to_tiled_copy_command(
 	lsdma_data->u.tiled_copy_data.src_y            = params.src_y;
 	lsdma_data->u.tiled_copy_data.dst_x            = params.dst_x;
 	lsdma_data->u.tiled_copy_data.dst_y            = params.dst_y;
-	lsdma_data->u.tiled_copy_data.src_width        = params.src_width - 1; // LSDMA controller expects width -1
-	lsdma_data->u.tiled_copy_data.dst_width        = params.dst_width - 1; // LSDMA controller expects width -1
+	lsdma_data->u.tiled_copy_data.src_width        = params.src_width;
+	lsdma_data->u.tiled_copy_data.dst_width        = params.dst_width;
 	lsdma_data->u.tiled_copy_data.src_swizzle_mode = params.swizzle_mode;
 	lsdma_data->u.tiled_copy_data.dst_swizzle_mode = params.swizzle_mode;
 	lsdma_data->u.tiled_copy_data.src_element_size = params.element_size;
 	lsdma_data->u.tiled_copy_data.dst_element_size = params.element_size;
-	lsdma_data->u.tiled_copy_data.rect_x           = params.rect_x - 1;
-	lsdma_data->u.tiled_copy_data.rect_y           = params.rect_y - 1;
+	lsdma_data->u.tiled_copy_data.rect_x           = params.rect_x;
+	lsdma_data->u.tiled_copy_data.rect_y           = params.rect_y;
 	lsdma_data->u.tiled_copy_data.dcc              = params.dcc;
 	lsdma_data->u.tiled_copy_data.tmz              = params.tmz;
 	lsdma_data->u.tiled_copy_data.read_compress    = params.read_compress;
 	lsdma_data->u.tiled_copy_data.write_compress   = params.write_compress;
-	lsdma_data->u.tiled_copy_data.src_height       = params.src_height - 1; // LSDMA controller expects height -1
-	lsdma_data->u.tiled_copy_data.dst_height       = params.dst_height - 1; // LSDMA controller expects height -1
+	lsdma_data->u.tiled_copy_data.src_height       = params.src_height;
+	lsdma_data->u.tiled_copy_data.dst_height       = params.dst_height;
 	lsdma_data->u.tiled_copy_data.data_format      = params.data_format;
 	lsdma_data->u.tiled_copy_data.max_com          = params.max_com;
 	lsdma_data->u.tiled_copy_data.max_uncom        = params.max_uncom;

From d728fd03e5f2117853d91b3626d434a97fe896d1 Mon Sep 17 00:00:00 2001
From: Austin Zheng <Austin.Zheng@amd.com>
Date: Tue, 5 Aug 2025 15:18:02 -0400
Subject: [PATCH 0512/1455] drm/amd/display: Setup Second Stutter Watermark
 Implementation

[WHY & HOW]
Setup initial changes required to program another set of watermarks
for a 2nd stutter mode. The 2nd stutter mode will be lower power but
have higher enter/exit latencies.

PMFW to choose which stutter mode to use based on stutter efficiences
to see if original stutter (LP1) or low power stutter (LP2) will result
in better power savings.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Austin Zheng <Austin.Zheng@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                  |  9 +++++++++
 .../display/dc/dml2/dml21/dml21_translation_helper.c |  2 ++
 .../dc/dml2/dml21/inc/dml_top_dchub_registers.h      |  2 ++
 .../dc/dml2/dml21/inc/dml_top_soc_parameter_types.h  |  2 ++
 .../amd/display/dc/dml2/dml21/inc/dml_top_types.h    |  2 ++
 .../dml21/src/dml2_core/dml2_core_shared_types.h     | 12 ++++++++++++
 6 files changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 1ab05eabbddb..29aaa38cc784 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -694,6 +694,15 @@ struct dc_clocks {
 	int idle_fclk_khz;
 	int subvp_prefetch_dramclk_khz;
 	int subvp_prefetch_fclk_khz;
+
+	/* Stutter efficiency is technically not clock values
+	 * but stored here so the values are part of the update_clocks call similar to num_ways
+	 * Efficiencies are stored as percentage (0-100)
+	 */
+	struct {
+		uint8_t base_efficiency; //LP1
+		uint8_t low_power_efficiency; //LP2
+	} stutter_efficiency;
 };
 
 struct dc_bw_validation_profile {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
index a06217a9eef6..23fdb17f851a 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
@@ -1165,6 +1165,8 @@ void dml21_copy_clocks_to_dc_state(struct dml2_context *in_ctx, struct dc_state
 	context->bw_ctx.bw.dcn.clk.socclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.socclk_khz;
 	context->bw_ctx.bw.dcn.clk.subvp_prefetch_dramclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.svp_prefetch_no_throttle.uclk_khz;
 	context->bw_ctx.bw.dcn.clk.subvp_prefetch_fclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.svp_prefetch_no_throttle.fclk_khz;
+	context->bw_ctx.bw.dcn.clk.stutter_efficiency.base_efficiency = in_ctx->v21.mode_programming.programming->stutter.base_percent_efficiency;
+	context->bw_ctx.bw.dcn.clk.stutter_efficiency.low_power_efficiency = in_ctx->v21.mode_programming.programming->stutter.low_power_percent_efficiency;
 }
 
 static struct dml2_dchub_watermark_regs *wm_set_index_to_dc_wm_set(union dcn_watermark_set *watermarks, const enum dml2_dchub_watermark_reg_set_index wm_index)
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h
index b05030926ce8..91955bbe24b8 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_dchub_registers.h
@@ -159,6 +159,8 @@ struct dml2_dchub_watermark_regs {
 	uint32_t sr_exit;
 	uint32_t sr_enter_z8;
 	uint32_t sr_exit_z8;
+	uint32_t sr_enter_low_power;
+	uint32_t sr_exit_low_power;
 	uint32_t uclk_pstate;
 	uint32_t fclk_pstate;
 	uint32_t temp_read_or_ppt;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h
index 8c9f414aa6bf..176f55947664 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_soc_parameter_types.h
@@ -96,6 +96,8 @@ struct dml2_soc_power_management_parameters {
 	double g7_temperature_read_blackout_us;
 	double stutter_enter_plus_exit_latency_us;
 	double stutter_exit_latency_us;
+	double low_power_stutter_enter_plus_exit_latency_us;
+	double low_power_stutter_exit_latency_us;
 	double z8_stutter_enter_plus_exit_latency_us;
 	double z8_stutter_exit_latency_us;
 	double z8_min_idle_time;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h
index 98c0234e2f47..7de10a95cfdb 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h
@@ -417,6 +417,8 @@ struct dml2_display_cfg_programming {
 
 	struct {
 		bool supported_in_blank; // Changing to configurations where this is false requires stutter to be disabled during the transition
+		uint8_t base_percent_efficiency; //LP1
+		uint8_t low_power_percent_efficiency; //LP2
 	} stutter;
 
 	struct {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h
index 28687565ac22..ffb8c09f37a5 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_shared_types.h
@@ -201,6 +201,8 @@ struct dml2_core_internal_watermarks {
 	double WritebackFCLKChangeWatermark;
 	double StutterExitWatermark;
 	double StutterEnterPlusExitWatermark;
+	double LowPowerStutterExitWatermark;
+	double LowPowerStutterEnterPlusExitWatermark;
 	double Z8StutterExitWatermark;
 	double Z8StutterEnterPlusExitWatermark;
 	double USRRetrainingWatermark;
@@ -877,6 +879,9 @@ struct dml2_core_internal_mode_program {
 	double Z8StutterEfficiency;
 	unsigned int Z8NumberOfStutterBurstsPerFrame;
 	double Z8StutterEfficiencyNotIncludingVBlank;
+	double LowPowerStutterEfficiency;
+	double LowPowerStutterEfficiencyNotIncludingVBlank;
+	unsigned int LowPowerNumberOfStutterBurstsPerFrame;
 	double StutterPeriod;
 	double Z8StutterEfficiencyBestCase;
 	unsigned int Z8NumberOfStutterBurstsPerFrameBestCase;
@@ -1016,6 +1021,8 @@ struct dml2_core_internal_SOCParametersList {
 	double FCLKChangeLatency;
 	double SRExitTime;
 	double SREnterPlusExitTime;
+	double SRExitTimeLowPower;
+	double SREnterPlusExitTimeLowPower;
 	double SRExitZ8Time;
 	double SREnterPlusExitZ8Time;
 	double USRRetrainingLatency;
@@ -1851,9 +1858,11 @@ struct dml2_core_calcs_CalculateStutterEfficiency_params {
 	unsigned int CompbufReservedSpaceZs;
 	bool hw_debug5;
 	double SRExitTime;
+	double SRExitTimeLowPower;
 	double SRExitZ8Time;
 	bool SynchronizeTimings;
 	double StutterEnterPlusExitWatermark;
+	double LowPowerStutterEnterPlusExitWatermark;
 	double Z8StutterEnterPlusExitWatermark;
 	bool ProgressiveToInterlaceUnitInOPP;
 	double *MinTTUVBlank;
@@ -1879,7 +1888,10 @@ struct dml2_core_calcs_CalculateStutterEfficiency_params {
 	// output
 	double *StutterEfficiencyNotIncludingVBlank;
 	double *StutterEfficiency;
+	double *LowPowerStutterEfficiencyNotIncludingVBlank;
+	double *LowPowerStutterEfficiency;
 	unsigned int *NumberOfStutterBurstsPerFrame;
+	unsigned int *LowPowerNumberOfStutterBurstsPerFrame;
 	double *Z8StutterEfficiencyNotIncludingVBlank;
 	double *Z8StutterEfficiency;
 	unsigned int *Z8NumberOfStutterBurstsPerFrame;

From a928f8d542ccd1e40e101cffc923ff33e36bf147 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 24 Jul 2025 13:01:09 -0500
Subject: [PATCH 0513/1455] drm/amd/display: Attach privacy screen to DRM
 connector

[WHY]
If a system has a privacy screen advertised by a driver it should
be included in the DRM connector for the eDP panel.

[HOW]
Detect statically declared privacy screens when creating eDP connector
and attach privacy screen DRM properties.

Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 25 ++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index e8cfae2bd2ae..176f420effd9 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -80,6 +80,7 @@
 #include <linux/component.h>
 #include <linux/sort.h>
 
+#include <drm/drm_privacy_screen_consumer.h>
 #include <drm/display/drm_dp_mst_helper.h>
 #include <drm/display/drm_hdmi_helper.h>
 #include <drm/drm_atomic.h>
@@ -7846,6 +7847,14 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn,
 	if (!crtc)
 		return 0;
 
+	if (new_con_state->privacy_screen_sw_state != old_con_state->privacy_screen_sw_state) {
+		new_crtc_state = drm_atomic_get_crtc_state(state, crtc);
+		if (IS_ERR(new_crtc_state))
+			return PTR_ERR(new_crtc_state);
+
+		new_crtc_state->mode_changed = true;
+	}
+
 	if (new_con_state->colorspace != old_con_state->colorspace) {
 		new_crtc_state = drm_atomic_get_crtc_state(state, crtc);
 		if (IS_ERR(new_crtc_state))
@@ -8541,6 +8550,18 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
 		if (adev->dm.hdcp_workqueue)
 			drm_connector_attach_content_protection_property(&aconnector->base, true);
 	}
+
+	if (connector_type == DRM_MODE_CONNECTOR_eDP) {
+		struct drm_privacy_screen *privacy_screen;
+
+		privacy_screen = drm_privacy_screen_get(adev_to_drm(adev)->dev, NULL);
+		if (!IS_ERR(privacy_screen)) {
+			drm_connector_attach_privacy_screen_provider(&aconnector->base,
+								     privacy_screen);
+		} else if (PTR_ERR(privacy_screen) != -ENODEV) {
+			drm_warn(adev_to_drm(adev), "Error getting privacy-screen\n");
+		}
+	}
 }
 
 static int amdgpu_dm_i2c_xfer(struct i2c_adapter *i2c_adap,
@@ -10265,7 +10286,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 	unsigned long flags;
 	bool wait_for_vblank = true;
 	struct drm_connector *connector;
-	struct drm_connector_state *old_con_state, *new_con_state;
+	struct drm_connector_state *old_con_state = NULL, *new_con_state = NULL;
 	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
 	int crtc_disable_count = 0;
 
@@ -10383,6 +10404,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
 					    &stream_update);
 		mutex_unlock(&dm->dc_lock);
 		kfree(dummy_updates);
+
+		drm_connector_update_privacy_screen(new_con_state);
 	}
 
 	/**

From 1e5e8d672fec9f2ab352be121be971877bff2af9 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 24 Jul 2025 15:00:43 -0500
Subject: [PATCH 0514/1455] drm/amd/display: Avoid a NULL pointer dereference

[WHY]
Although unlikely drm_atomic_get_new_connector_state() or
drm_atomic_get_old_connector_state() can return NULL.

[HOW]
Check returns before dereference.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 176f420effd9..b944abea306d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -7836,6 +7836,9 @@ amdgpu_dm_connector_atomic_check(struct drm_connector *conn,
 	struct amdgpu_dm_connector *aconn = to_amdgpu_dm_connector(conn);
 	int ret;
 
+	if (WARN_ON(unlikely(!old_con_state || !new_con_state)))
+		return -EINVAL;
+
 	trace_amdgpu_dm_connector_atomic_check(new_con_state);
 
 	if (conn->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {

From eac4c502119313e2c10a8a5dca099c973dc178bd Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 8 Aug 2025 17:25:15 -0400
Subject: [PATCH 0515/1455] drm/amd/display: [FW Promotion] Release 0.1.23.0

1. Fix loop counter.
2. Check whether rb->capacity is 0.

Acked-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dmub/inc/dmub_cmd.h   | 21 ++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 52295efdba63..d7008d84c1ec 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -6542,15 +6542,18 @@ static inline bool dmub_rb_full(struct dmub_rb *rb)
 static inline bool dmub_rb_push_front(struct dmub_rb *rb,
 				      const union dmub_rb_cmd *cmd)
 {
-	uint64_t volatile *dst = (uint64_t volatile *)((uint8_t *)(rb->base_address) + rb->wrpt);
-	const uint64_t *src = (const uint64_t *)cmd;
+	uint8_t *dst = (uint8_t *)(rb->base_address) + rb->wrpt;
+	const uint8_t *src = (const uint8_t *)cmd;
 	uint8_t i;
 
+	if (rb->capacity == 0)
+		return false;
+
 	if (dmub_rb_full(rb))
 		return false;
 
 	// copying data
-	for (i = 0; i < DMUB_RB_CMD_SIZE / sizeof(uint64_t); i++)
+	for (i = 0; i < DMUB_RB_CMD_SIZE; i++)
 		*dst++ = *src++;
 
 	rb->wrpt += DMUB_RB_CMD_SIZE;
@@ -6575,6 +6578,9 @@ static inline bool dmub_rb_out_push_front(struct dmub_rb *rb,
 	uint8_t *dst = (uint8_t *)(rb->base_address) + rb->wrpt;
 	const uint8_t *src = (const uint8_t *)cmd;
 
+	if (rb->capacity == 0)
+		return false;
+
 	if (dmub_rb_full(rb))
 		return false;
 
@@ -6620,6 +6626,9 @@ static inline void dmub_rb_get_rptr_with_offset(struct dmub_rb *rb,
 				  uint32_t num_cmds,
 				  uint32_t *next_rptr)
 {
+	if (rb->capacity == 0)
+		return;
+
 	*next_rptr = rb->rptr + DMUB_RB_CMD_SIZE * num_cmds;
 
 	if (*next_rptr >= rb->capacity)
@@ -6683,6 +6692,9 @@ static inline bool dmub_rb_out_front(struct dmub_rb *rb,
  */
 static inline bool dmub_rb_pop_front(struct dmub_rb *rb)
 {
+	if (rb->capacity == 0)
+		return false;
+
 	if (dmub_rb_empty(rb))
 		return false;
 
@@ -6707,6 +6719,9 @@ static inline void dmub_rb_flush_pending(const struct dmub_rb *rb)
 	uint32_t rptr = rb->rptr;
 	uint32_t wptr = rb->wrpt;
 
+	if (rb->capacity == 0)
+		return;
+
 	while (rptr != wptr) {
 		uint64_t *data = (uint64_t *)((uint8_t *)(rb->base_address) + rptr);
 		uint8_t i;

From 34d66bc7ff10e146a4cec76cf286979740a10954 Mon Sep 17 00:00:00 2001
From: Tom Chung <chiahsuan.chung@amd.com>
Date: Fri, 18 Jul 2025 18:25:08 +0800
Subject: [PATCH 0516/1455] drm/amd/display: Fix Xorg desktop unresponsive on
 Replay panel

[WHY & HOW]
IPS & self-fresh feature can cause vblank counter resets between
vblank disable and enable.
It may cause system stuck due to wait the vblank counter.

Call the drm_crtc_vblank_restore() during vblank enable to estimate
missed vblanks by using timestamps and update the vblank counter in
DRM.

It can make the vblank counter increase smoothly and resolve this issue.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/amdgpu_dm/amdgpu_dm_crtc.c    | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index 010172f930ae..45feb404b097 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -299,6 +299,25 @@ static inline int amdgpu_dm_crtc_set_vblank(struct drm_crtc *crtc, bool enable)
 	irq_type = amdgpu_display_crtc_idx_to_irq_type(adev, acrtc->crtc_id);
 
 	if (enable) {
+		struct dc *dc = adev->dm.dc;
+		struct drm_vblank_crtc *vblank = drm_crtc_vblank_crtc(crtc);
+		struct psr_settings *psr = &acrtc_state->stream->link->psr_settings;
+		struct replay_settings *pr = &acrtc_state->stream->link->replay_settings;
+		bool sr_supported = (psr->psr_version != DC_PSR_VERSION_UNSUPPORTED) ||
+								pr->config.replay_supported;
+
+		/*
+		 * IPS & self-refresh feature can cause vblank counter resets between
+		 * vblank disable and enable.
+		 * It may cause system stuck due to waiting for the vblank counter.
+		 * Call this function to estimate missed vblanks by using timestamps and
+		 * update the vblank counter in DRM.
+		 */
+		if (dc->caps.ips_support &&
+			dc->config.disable_ips != DMUB_IPS_DISABLE_ALL &&
+			sr_supported && vblank->config.disable_immediate)
+			drm_crtc_vblank_restore(crtc);
+
 		/* vblank irq on -> Only need vupdate irq in vrr mode */
 		if (amdgpu_dm_crtc_vrr_active(acrtc_state))
 			rc = amdgpu_dm_crtc_set_vupdate_irq(crtc, true);

From c9ec952216d3a520d4e066491a78743eabb3ce7c Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 8 Aug 2025 17:25:32 -0500
Subject: [PATCH 0517/1455] drm/amd/display: Promote DC to 3.2.346

This version brings along following updates:

- Fix Xorg desktop unresponsive on Replay panel
- [FW Promotion] Release 0.1.23.0
- Avoid a NULL pointer dereference
- Attach privacy screen to DRM connector
- Setup Second Stutter Watermark Implementation
- Align LSDMA commands fields
- Delete unused functions
- Optimize amdgpu_dm_atomic_commit_tail()
- Add primary plane to commits for correct VRR handling
- Refactor DPP enum for backwards compatibility.
- Add LSDMA Linear Sub Window Copy support

Acked-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 29aaa38cc784..eb7019ed92b2 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.345"
+#define DC_VER "3.2.346"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 5e43eb3cd731649c4f8b9134f857be62a416c893 Mon Sep 17 00:00:00 2001
From: Chenyuan Yang <chenyuan0y@gmail.com>
Date: Wed, 23 Jul 2025 21:36:41 -0500
Subject: [PATCH 0518/1455] drm/amd/display: Add null pointer check in
 mod_hdcp_hdcp1_create_session()

The function mod_hdcp_hdcp1_create_session() calls the function
get_first_active_display(), but does not check its return value.
The return value is a null pointer if the display list is empty.
This will lead to a null pointer dereference.

Add a null pointer check for get_first_active_display() and return
MOD_HDCP_STATUS_DISPLAY_NOT_FOUND if the function return null.

This is similar to the commit c3e9826a2202
("drm/amd/display: Add null pointer check for get_first_active_display()").

Fixes: 2deade5ede56 ("drm/amd/display: Remove hdcp display state with mst fix")
Signed-off-by: Chenyuan Yang <chenyuan0y@gmail.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c
index e58e7b93810b..6b7db8ec9a53 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_psp.c
@@ -260,6 +260,9 @@ enum mod_hdcp_status mod_hdcp_hdcp1_create_session(struct mod_hdcp *hdcp)
 		return MOD_HDCP_STATUS_FAILURE;
 	}
 
+	if (!display)
+		return MOD_HDCP_STATUS_DISPLAY_NOT_FOUND;
+
 	hdcp_cmd = (struct ta_hdcp_shared_memory *)psp->hdcp_context.context.mem_context.shared_buf;
 
 	mutex_lock(&psp->hdcp_context.mutex);

From 8a614ddfad1f3d2baa45b6520688b2032f280a7d Mon Sep 17 00:00:00 2001
From: Xichao Zhao <zhao.xichao@vivo.com>
Date: Thu, 31 Jul 2025 15:54:50 +0800
Subject: [PATCH 0519/1455] drm/amd/display: Clean up coding style

Adjust whitespace around operators to improve code readability
and comply with kernel coding style guidelines.

These changes are purely stylistic and introduce no
functional modifications.

Signed-off-by: Xichao Zhao <zhao.xichao@vivo.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/link/protocols/link_dp_capability.c    | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index e0c4416993d9..caddb7dfb133 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -1525,8 +1525,8 @@ bool read_is_mst_supported(struct dc_link *link)
 		return false;
 	}
 
-	rev.raw  = 0;
-	cap.raw  = 0;
+	rev.raw = 0;
+	cap.raw = 0;
 
 	st = core_link_read_dpcd(link, DP_DPCD_REV, &rev.raw,
 			sizeof(rev));
@@ -2125,13 +2125,13 @@ void detect_edp_sink_caps(struct dc_link *link)
 						&backlight_adj_cap, sizeof(backlight_adj_cap));
 
 	link->dpcd_caps.dynamic_backlight_capable_edp =
-				(backlight_adj_cap & DP_EDP_DYNAMIC_BACKLIGHT_CAP) ? true:false;
+				(backlight_adj_cap & DP_EDP_DYNAMIC_BACKLIGHT_CAP) ? true : false;
 
 	core_link_read_dpcd(link, DP_EDP_GENERAL_CAP_1,
 						&general_edp_cap, sizeof(general_edp_cap));
 
 	link->dpcd_caps.set_power_state_capable_edp =
-				(general_edp_cap & DP_EDP_SET_POWER_CAP) ? true:false;
+				(general_edp_cap & DP_EDP_SET_POWER_CAP) ? true : false;
 
 	set_default_brightness_aux(link);
 

From 5599b214c4afcb4bfe213e67336ce8e8dbb0d5ab Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 6 Aug 2025 20:41:20 +0800
Subject: [PATCH 0520/1455] drm/amd/display: Use swap() to simplify code

Replace the original swapping logic with swap() to improve readability and
remove temporary variables

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
index 5f1b49a50049..4cfe64aa8492 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_dc_resource_mgmt.c
@@ -473,7 +473,6 @@ static void sort_pipes_for_splitting(struct dc_plane_pipe_pool *pipes)
 {
 	bool sorted, swapped;
 	unsigned int cur_index;
-	unsigned int temp;
 	int odm_slice_index;
 
 	for (odm_slice_index = 0; odm_slice_index < pipes->num_pipes_assigned_to_plane_for_odm_combine; odm_slice_index++) {
@@ -489,9 +488,8 @@ static void sort_pipes_for_splitting(struct dc_plane_pipe_pool *pipes)
 		swapped = false;
 		while (!sorted) {
 			if (pipes->pipes_assigned_to_plane[odm_slice_index][cur_index] > pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1]) {
-				temp = pipes->pipes_assigned_to_plane[odm_slice_index][cur_index];
-				pipes->pipes_assigned_to_plane[odm_slice_index][cur_index] = pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1];
-				pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1] = temp;
+				swap(pipes->pipes_assigned_to_plane[odm_slice_index][cur_index + 1],
+					 pipes->pipes_assigned_to_plane[odm_slice_index][cur_index]);
 
 				swapped = true;
 			}

From 27b16364ea126dc74105f151e9f64ac9e06eb268 Mon Sep 17 00:00:00 2001
From: Xichao Zhao <zhao.xichao@vivo.com>
Date: Fri, 8 Aug 2025 10:52:09 +0800
Subject: [PATCH 0521/1455] drm/amd/display: replace min/max nesting with
 clamp()

The clamp() macro explicitly expresses the intent of constraining
a value within bounds.Therefore, replacing min(max(a, b), c) with
clamp(val, lo, hi) can improve code readability.

Signed-off-by: Xichao Zhao <zhao.xichao@vivo.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
index 2066a65c69bb..1aaa77265eed 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn301/dcn301_fpu.c
@@ -449,12 +449,12 @@ void dcn301_fpu_calculate_wm_and_dlg(struct dc *dc,
 						&context->bw_ctx.dml, pipes, pipe_cnt);
 	/* WM Set C */
 	table_entry = &bw_params->wm_table.entries[WM_C];
-	vlevel = min(max(vlevel_req, 2), vlevel_max);
+	vlevel = clamp(vlevel_req, 2, vlevel_max);
 	calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.c,
 						&context->bw_ctx.dml, pipes, pipe_cnt);
 	/* WM Set B */
 	table_entry = &bw_params->wm_table.entries[WM_B];
-	vlevel = min(max(vlevel_req, 1), vlevel_max);
+	vlevel = clamp(vlevel_req, 1, vlevel_max);
 	calculate_wm_set_for_vlevel(vlevel, table_entry, &context->bw_ctx.bw.dcn.watermarks.b,
 						&context->bw_ctx.dml, pipes, pipe_cnt);
 

From 427980c1cbd22bb256b9385f5ce73c0937562408 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:46 +0200
Subject: [PATCH 0522/1455] drm/amd/display: Don't overclock DCE 6 by 15%
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The extra 15% clock was added as a workaround for a Polaris issue
which uses DCE 11, and should not have been used on DCE 6 which
is already hardcoded to the highest possible display clock.
Unfortunately, the extra 15% was mistakenly copied and kept
even on code paths which don't affect Polaris.

This commit fixes that and also adds a check to make sure
not to exceed the maximum DCE 6 display clock.

Fixes: 8cd61c313d8b ("drm/amd/display: Raise dispclk value for Polaris")
Fixes: dc88b4a684d2 ("drm/amd/display: make clk mgr soc specific")
Fixes: 3ecb3b794e2c ("drm/amd/display: dc/clk_mgr: add support for SI parts (v2)")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c  | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
index 0267644717b2..cfd7309f2c6a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
@@ -123,11 +123,9 @@ static void dce60_update_clocks(struct clk_mgr *clk_mgr_base,
 {
 	struct clk_mgr_internal *clk_mgr_dce = TO_CLK_MGR_INTERNAL(clk_mgr_base);
 	struct dm_pp_power_level_change_request level_change_req;
-	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
-
-	/*TODO: W/A for dal3 linux, investigate why this works */
-	if (!clk_mgr_dce->dfs_bypass_active)
-		patched_disp_clk = patched_disp_clk * 115 / 100;
+	const int max_disp_clk =
+		clk_mgr_dce->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz;
+	int patched_disp_clk = MIN(max_disp_clk, context->bw_ctx.bw.dce.dispclk_khz);
 
 	level_change_req.power_level = dce_get_required_clocks_state(clk_mgr_base, context);
 	/* get max clock state from PPLIB */

From 1ae45b5d4f371af8ae51a3827d0ec9fe27eeb867 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:47 +0200
Subject: [PATCH 0523/1455] drm/amd/display: Adjust DCE 8-10 clock, don't
 overclock by 15%
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Adjust the nominal (and performance) clocks for DCE 8-10,
and set them to 625 MHz, which is the value used by the legacy
display code in amdgpu_atombios_get_clock_info.

This was tested with Hawaii, Tonga and Fiji.
These GPUs can output 4K 60Hz (10-bit depth) at 625 MHz.

The extra 15% clock was added as a workaround for a Polaris issue
which uses DCE 11, and should not have been used on DCE 8-10 which
are already hardcoded to the highest possible display clock.
Unfortunately, the extra 15% was mistakenly copied and kept
even on code paths which don't affect Polaris.

This commit fixes that and also	adds a check to	make sure
not to exceed the maximum DCE 8-10 display clock.

Fixes: 8cd61c313d8b ("drm/amd/display: Raise dispclk value for Polaris")
Fixes: dc88b4a684d2 ("drm/amd/display: make clk mgr soc specific")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c  | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
index f5ad0a177038..a324dbd4543c 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
@@ -72,9 +72,9 @@ static const struct state_dependent_clocks dce80_max_clks_by_state[] = {
 /* ClocksStateLow */
 { .display_clk_khz = 352000, .pixel_clk_khz = 330000},
 /* ClocksStateNominal */
-{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 },
+{ .display_clk_khz = 625000, .pixel_clk_khz = 400000 },
 /* ClocksStatePerformance */
-{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 } };
+{ .display_clk_khz = 625000, .pixel_clk_khz = 400000 } };
 
 int dentist_get_divider_from_did(int did)
 {
@@ -405,11 +405,9 @@ static void dce_update_clocks(struct clk_mgr *clk_mgr_base,
 {
 	struct clk_mgr_internal *clk_mgr_dce = TO_CLK_MGR_INTERNAL(clk_mgr_base);
 	struct dm_pp_power_level_change_request level_change_req;
-	int patched_disp_clk = context->bw_ctx.bw.dce.dispclk_khz;
-
-	/*TODO: W/A for dal3 linux, investigate why this works */
-	if (!clk_mgr_dce->dfs_bypass_active)
-		patched_disp_clk = patched_disp_clk * 115 / 100;
+	const int max_disp_clk =
+		clk_mgr_dce->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz;
+	int patched_disp_clk = MIN(max_disp_clk, context->bw_ctx.bw.dce.dispclk_khz);
 
 	level_change_req.power_level = dce_get_required_clocks_state(clk_mgr_base, context);
 	/* get max clock state from PPLIB */

From 4ab09785f8d5d03df052827af073d5c508ff5f63 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:48 +0200
Subject: [PATCH 0524/1455] drm/amd/display: Find first CRTC and its line time
 in dce110_fill_display_configs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

dce110_fill_display_configs is shared between DCE 6-11, and
finding the first CRTC and its line time is relevant to DCE 6 too.
Move the code to find it from DCE 11 specific code.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/clk_mgr/dce110/dce110_clk_mgr.c        | 30 ++++++++++++-------
 1 file changed, 20 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
index f8409453434c..baeac8f1c04f 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
@@ -120,9 +120,12 @@ void dce110_fill_display_configs(
 	const struct dc_state *context,
 	struct dm_pp_display_configuration *pp_display_cfg)
 {
+	struct dc *dc = context->clk_mgr->ctx->dc;
 	int j;
 	int num_cfgs = 0;
 
+	pp_display_cfg->crtc_index = dc->res_pool->res_cap->num_timing_generator;
+
 	for (j = 0; j < context->stream_count; j++) {
 		int k;
 
@@ -164,6 +167,23 @@ void dce110_fill_display_configs(
 		cfg->v_refresh /= stream->timing.h_total;
 		cfg->v_refresh = (cfg->v_refresh + stream->timing.v_total / 2)
 							/ stream->timing.v_total;
+
+		/* Find first CRTC index and calculate its line time.
+		 * This is necessary for DPM on SI GPUs.
+		 */
+		if (cfg->pipe_idx < pp_display_cfg->crtc_index) {
+			const struct dc_crtc_timing *timing =
+				&context->streams[0]->timing;
+
+			pp_display_cfg->crtc_index = cfg->pipe_idx;
+			pp_display_cfg->line_time_in_us =
+				timing->h_total * 10000 / timing->pix_clk_100hz;
+		}
+	}
+
+	if (!num_cfgs) {
+		pp_display_cfg->crtc_index = 0;
+		pp_display_cfg->line_time_in_us = 0;
 	}
 
 	pp_display_cfg->display_count = num_cfgs;
@@ -232,16 +252,6 @@ void dce11_pplib_apply_display_requirements(
 
 	dce110_fill_display_configs(context, pp_display_cfg);
 
-	/* TODO: is this still applicable?*/
-	if (pp_display_cfg->display_count == 1) {
-		const struct dc_crtc_timing *timing =
-			&context->streams[0]->timing;
-
-		pp_display_cfg->crtc_index =
-			pp_display_cfg->disp_configs[0].pipe_idx;
-		pp_display_cfg->line_time_in_us = timing->h_total * 10000 / timing->pix_clk_100hz;
-	}
-
 	if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) !=  0)
 		dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
 }

From 8107432dff37db26fcb641b6cebeae8981cd73a0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:49 +0200
Subject: [PATCH 0525/1455] drm/amd/display: Fill display clock and vblank time
 in dce110_fill_display_configs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Also needed by DCE 6.
This way the code that gathers this info can be shared between
different DCE versions and doesn't have to be repeated.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c    |  2 --
 .../drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c | 10 +++-------
 .../drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c   |  2 --
 3 files changed, 3 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
index a324dbd4543c..dbd6ef1b60a0 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
@@ -391,8 +391,6 @@ static void dce_pplib_apply_display_requirements(
 {
 	struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
 
-	pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context);
-
 	dce110_fill_display_configs(context, pp_display_cfg);
 
 	if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) !=  0)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
index baeac8f1c04f..13cf415e38e5 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
@@ -124,6 +124,9 @@ void dce110_fill_display_configs(
 	int j;
 	int num_cfgs = 0;
 
+	pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context);
+	pp_display_cfg->disp_clk_khz = dc->clk_mgr->clks.dispclk_khz;
+	pp_display_cfg->avail_mclk_switch_time_in_disp_active_us = 0;
 	pp_display_cfg->crtc_index = dc->res_pool->res_cap->num_timing_generator;
 
 	for (j = 0; j < context->stream_count; j++) {
@@ -243,13 +246,6 @@ void dce11_pplib_apply_display_requirements(
 	pp_display_cfg->min_engine_clock_deep_sleep_khz
 			= context->bw_ctx.bw.dce.sclk_deep_sleep_khz;
 
-	pp_display_cfg->avail_mclk_switch_time_us =
-						dce110_get_min_vblank_time_us(context);
-	/* TODO: dce11.2*/
-	pp_display_cfg->avail_mclk_switch_time_in_disp_active_us = 0;
-
-	pp_display_cfg->disp_clk_khz = dc->clk_mgr->clks.dispclk_khz;
-
 	dce110_fill_display_configs(context, pp_display_cfg);
 
 	if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) !=  0)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
index cfd7309f2c6a..7044b437fe9d 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
@@ -109,8 +109,6 @@ static void dce60_pplib_apply_display_requirements(
 {
 	struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
 
-	pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context);
-
 	dce110_fill_display_configs(context, pp_display_cfg);
 
 	if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) !=  0)

From 33e0227ee96e62d034781e91f215e32fd0b1d512 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:50 +0200
Subject: [PATCH 0526/1455] drm/amd/display: Don't warn when missing DCE
 encoder caps
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On some GPUs the VBIOS just doesn't have encoder caps,
or maybe not for every encoder.

This isn't really a problem and it's handled well,
so let's not litter the logs with it.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
index 4a9d07c31bc5..0c50fe266c8a 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
@@ -896,13 +896,13 @@ void dce110_link_encoder_construct(
 						enc110->base.id, &bp_cap_info);
 
 	/* Override features with DCE-specific values */
-	if (BP_RESULT_OK == result) {
+	if (result == BP_RESULT_OK) {
 		enc110->base.features.flags.bits.IS_HBR2_CAPABLE =
 				bp_cap_info.DP_HBR2_EN;
 		enc110->base.features.flags.bits.IS_HBR3_CAPABLE =
 				bp_cap_info.DP_HBR3_EN;
 		enc110->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN;
-	} else {
+	} else if (result != BP_RESULT_NORECORD) {
 		DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n",
 				__func__,
 				result);
@@ -1798,13 +1798,13 @@ void dce60_link_encoder_construct(
 						enc110->base.id, &bp_cap_info);
 
 	/* Override features with DCE-specific values */
-	if (BP_RESULT_OK == result) {
+	if (result == BP_RESULT_OK) {
 		enc110->base.features.flags.bits.IS_HBR2_CAPABLE =
 				bp_cap_info.DP_HBR2_EN;
 		enc110->base.features.flags.bits.IS_HBR3_CAPABLE =
 				bp_cap_info.DP_HBR3_EN;
 		enc110->base.features.flags.bits.HDMI_6GB_EN = bp_cap_info.HDMI_6GB_EN;
-	} else {
+	} else if (result != BP_RESULT_NORECORD) {
 		DC_LOG_WARNING("%s: Failed to get encoder_cap_info from VBIOS with error code %d!\n",
 				__func__,
 				result);

From 249d4bc5f1935f04bb45b3b63c0f8922565124f7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:51 +0200
Subject: [PATCH 0527/1455] drm/amd/display: Don't print errors for nonexistent
 connectors
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When getting the number of connectors, the VBIOS reports
the number of valid indices, but it doesn't say which indices
are valid, and not every valid index has an actual connector.
If we don't find a connector on an index, that is not an error.

Considering these are not actual errors, don't litter the logs.

Fixes: 60df5628144b ("drm/amd/display: handle invalid connector indices")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/bios/bios_parser.c |  5 +----
 drivers/gpu/drm/amd/display/dc/core/dc.c          | 15 ++++++++++++++-
 2 files changed, 15 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
index 67f08495b7e6..154fd2c18e88 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
@@ -174,11 +174,8 @@ static struct graphics_object_id bios_parser_get_connector_id(
 		return object_id;
 	}
 
-	if (tbl->ucNumberOfObjects <= i) {
-		dm_error("Can't find connector id %d in connector table of size %d.\n",
-			 i, tbl->ucNumberOfObjects);
+	if (tbl->ucNumberOfObjects <= i)
 		return object_id;
-	}
 
 	id = le16_to_cpu(tbl->asObjects[i].usObjectID);
 	object_id = object_id_from_bios_object_id(id);
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 5bf97d8fb34a..c30d9ee51c83 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -217,11 +217,24 @@ static bool create_links(
 		connectors_num,
 		num_virtual_links);
 
-	// condition loop on link_count to allow skipping invalid indices
+	/* When getting the number of connectors, the VBIOS reports the number of valid indices,
+	 * but it doesn't say which indices are valid, and not every index has an actual connector.
+	 * So, if we don't find a connector on an index, that is not an error.
+	 *
+	 * - There is no guarantee that the first N indices will be valid
+	 * - VBIOS may report a higher amount of valid indices than there are actual connectors
+	 * - Some VBIOS have valid configurations for more connectors than there actually are
+	 *   on the card. This may be because the manufacturer used the same VBIOS for different
+	 *   variants of the same card.
+	 */
 	for (i = 0; dc->link_count < connectors_num && i < MAX_LINKS; i++) {
+		struct graphics_object_id connector_id = bios->funcs->get_connector_id(bios, i);
 		struct link_init_data link_init_params = {0};
 		struct dc_link *link;
 
+		if (connector_id.id == CONNECTOR_ID_UNKNOWN)
+			continue;
+
 		DC_LOG_DC("BIOS object table - printing link object info for connector number: %d, link_index: %d", i, dc->link_count);
 
 		link_init_params.ctx = dc->ctx;

From 027e7acc7e17802ebf28e1edb88a404836ad50d6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 31 Jul 2025 11:43:52 +0200
Subject: [PATCH 0528/1455] drm/amd/display: Fix fractional fb divider in
 set_pixel_clock_v3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For later VBIOS versions, the fractional feedback divider is
calculated as the remainder of dividing the feedback divider by
a factor, which is set to 1000000. For reference, see:
- calculate_fb_and_fractional_fb_divider
- calc_pll_max_vco_construct

However, in case of old VBIOS versions that have
set_pixel_clock_v3, they only have 1 byte available for the
fractional feedback divider, and it's expected to be set to the
remainder from dividing the feedback divider by 10.
For reference see the legacy display code:
- amdgpu_pll_compute
- amdgpu_atombios_crtc_program_pll

This commit fixes set_pixel_clock_v3 by dividing the fractional
feedback divider passed to the function by 100000.

Fixes: 4562236b3bc0 ("drm/amd/dc: Add dc display driver (v2)")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/bios/command_table.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table.c b/drivers/gpu/drm/amd/display/dc/bios/command_table.c
index 2bcae0643e61..58e88778da7f 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/command_table.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/command_table.c
@@ -993,7 +993,7 @@ static enum bp_result set_pixel_clock_v3(
 	allocation.sPCLKInput.usFbDiv =
 			cpu_to_le16((uint16_t)bp_params->feedback_divider);
 	allocation.sPCLKInput.ucFracFbDiv =
-			(uint8_t)bp_params->fractional_feedback_divider;
+			(uint8_t)(bp_params->fractional_feedback_divider / 100000);
 	allocation.sPCLKInput.ucPostDiv =
 			(uint8_t)bp_params->pixel_clock_post_divider;
 

From 3856a53db6620f29dda1b42af678698c0ccc6504 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Sat, 16 Aug 2025 10:27:27 -0600
Subject: [PATCH 0529/1455] drm/amdgpu/vcn: Remove unnecessary check

The function amdgpu_vcn_sysfs_reset_mask_init already returns 0, which
makes the check of the result unnecessary in the vcn_v4_0_3_sw_init().
Just return the amdgpu_vcn_sysfs_reset_mask_init directly.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index a63a1e3435ab..7b93a275ec4f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -242,11 +242,7 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block)
 	if (r)
 		return r;
 
-	r = amdgpu_vcn_sysfs_reset_mask_init(adev);
-	if (r)
-		return r;
-
-	return 0;
+	return amdgpu_vcn_sysfs_reset_mask_init(adev);
 }
 
 /**

From e3bd536580fe553d5768673660b1745ff1bb3b91 Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Sat, 16 Aug 2025 22:37:51 +0800
Subject: [PATCH 0530/1455] drm/radeon: Use vmalloc_array and vcalloc to
 simplify code

Use vcalloc() and vmalloc_array() to simplify the functions
radeon_gart_init().

vmalloc_array() is also optimized better, resulting in less instructions
being used.

Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_gart.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c
index 4bb242437ff6..acd89a20f272 100644
--- a/drivers/gpu/drm/radeon/radeon_gart.c
+++ b/drivers/gpu/drm/radeon/radeon_gart.c
@@ -346,14 +346,14 @@ int radeon_gart_init(struct radeon_device *rdev)
 	DRM_INFO("GART: num cpu pages %u, num gpu pages %u\n",
 		 rdev->gart.num_cpu_pages, rdev->gart.num_gpu_pages);
 	/* Allocate pages table */
-	rdev->gart.pages = vzalloc(array_size(sizeof(void *),
-				   rdev->gart.num_cpu_pages));
+	rdev->gart.pages = vcalloc(rdev->gart.num_cpu_pages,
+				   sizeof(void *));
 	if (rdev->gart.pages == NULL) {
 		radeon_gart_fini(rdev);
 		return -ENOMEM;
 	}
-	rdev->gart.pages_entry = vmalloc(array_size(sizeof(uint64_t),
-						    rdev->gart.num_gpu_pages));
+	rdev->gart.pages_entry = vmalloc_array(rdev->gart.num_gpu_pages,
+					       sizeof(uint64_t));
 	if (rdev->gart.pages_entry == NULL) {
 		radeon_gart_fini(rdev);
 		return -ENOMEM;

From 645cc7863da5de700547d236697dffd6760cf051 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Sat, 2 Aug 2025 17:51:53 +0200
Subject: [PATCH 0531/1455] drm/amd/display: Fix DP audio DTO1 clock source on
 DCE 6.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On DCE 6, DP audio was not working. However, it worked when an
HDMI monitor was also plugged in.

Looking at dce_aud_wall_dto_setup it seems that the main
difference is that we use DTO1 when only DP is plugged in.

When programming DTO1, it uses audio_dto_source_clock_in_khz
which is set from get_dp_ref_freq_khz

The dce60_get_dp_ref_freq_khz implementation looks incorrect,
because DENTIST_DISPCLK_CNTL seems to be always zero on DCE 6,
so it isn't usable.
I compared dce60_get_dp_ref_freq_khz to the legacy display code,
specifically dce_v6_0_audio_set_dto, and it turns out that in
case of DCE 6, it needs to use the display clock. With that,
DP audio started working on Pitcairn, Oland and Cape Verde.

However, it still didn't work on Tahiti. Despite having the
same DCE version, Tahiti seems to have a different audio device.
After some trial and error I realized that it works with the
default display clock as reported by the VBIOS, not the current
display clock.

The patch was tested on all four SI GPUs:

* Pitcairn (DCE 6.0)
* Oland (DCE 6.4)
* Cape Verde (DCE 6.0)
* Tahiti (DCE 6.0 but different)

The testing was done on Samsung Odyssey G7 LS28BG700EPXEN on
each of the above GPUs, at the following settings:

* 4K 60 Hz
* 1080p 60 Hz
* 1080p 144 Hz

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dce60/dce60_clk_mgr.c  | 21 ++++++-------------
 1 file changed, 6 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
index 7044b437fe9d..a39641a0ff09 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
@@ -83,22 +83,13 @@ static const struct state_dependent_clocks dce60_max_clks_by_state[] = {
 static int dce60_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr_base)
 {
 	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
-	int dprefclk_wdivider;
-	int dp_ref_clk_khz;
-	int target_div;
+	struct dc_context *ctx = clk_mgr_base->ctx;
+	int dp_ref_clk_khz = 0;
 
-	/* DCE6 has no DPREFCLK_CNTL to read DP Reference Clock source */
-
-	/* Read the mmDENTIST_DISPCLK_CNTL to get the currently
-	 * programmed DID DENTIST_DPREFCLK_WDIVIDER*/
-	REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, &dprefclk_wdivider);
-
-	/* Convert DENTIST_DPREFCLK_WDIVIDERto actual divider*/
-	target_div = dentist_get_divider_from_did(dprefclk_wdivider);
-
-	/* Calculate the current DFS clock, in kHz.*/
-	dp_ref_clk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR
-		* clk_mgr->base.dentist_vco_freq_khz) / target_div;
+	if (ASIC_REV_IS_TAHITI_P(ctx->asic_id.hw_internal_rev))
+		dp_ref_clk_khz = ctx->dc_bios->fw_info.default_display_engine_pll_frequency;
+	else
+		dp_ref_clk_khz = clk_mgr_base->clks.dispclk_khz;
 
 	return dce_adjust_dp_ref_freq_for_ss(clk_mgr, dp_ref_clk_khz);
 }

From c9932f0d604e4c8f2c6018e598a322acb43c68a2 Mon Sep 17 00:00:00 2001
From: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Date: Mon, 11 Aug 2025 09:12:31 +0000
Subject: [PATCH 0532/1455] drm/i915/gt: Relocate compression repacking WA for
 JSL/EHL

CACHE_MODE_0 registers should be saved and restored as part of
the context, not during engine reset. Move the related workaround
(Disable Repacking for Compression) from rcs_engine_wa_init()
to icl_ctx_workarounds_init() for Jasper Lake and Elkhart
Lake platforms. This ensures the WA is applied during context
initialisation.

BSPEC: 11322

Fixes: 0ddae025ab6c ("drm/i915: Disable compression tricks on JSL")
Closes: Fixes: 0ddae025ab6c ("drm/i915: Disable compression tricks on JSL")
Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Cc: stable@vger.kernel.org # v6.13+
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/4feaa24094e019e000ceb6011d8cd419b0361b3f.1754902406.git.sebastian.brzezinka@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index b37e400f74e5..5a95f06900b5 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -634,6 +634,8 @@ static void cfl_ctx_workarounds_init(struct intel_engine_cs *engine,
 static void icl_ctx_workarounds_init(struct intel_engine_cs *engine,
 				     struct i915_wa_list *wal)
 {
+	struct drm_i915_private *i915 = engine->i915;
+
 	/* Wa_1406697149 (WaDisableBankHangMode:icl) */
 	wa_write(wal, GEN8_L3CNTLREG, GEN8_ERRDETBCTRL);
 
@@ -669,6 +671,15 @@ static void icl_ctx_workarounds_init(struct intel_engine_cs *engine,
 
 	/* Wa_1406306137:icl,ehl */
 	wa_mcr_masked_en(wal, GEN9_ROW_CHICKEN4, GEN11_DIS_PICK_2ND_EU);
+
+	if (IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) {
+		/*
+		 * Disable Repacking for Compression (masked R/W access)
+		 * before rendering compressed surfaces for display.
+		 */
+		wa_masked_en(wal, CACHE_MODE_0_GEN7,
+			     DISABLE_REPACKING_FOR_COMPRESSION);
+	}
 }
 
 /*
@@ -2306,15 +2317,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 			     GEN8_RC_SEMA_IDLE_MSG_DISABLE);
 	}
 
-	if (IS_JASPERLAKE(i915) || IS_ELKHARTLAKE(i915)) {
-		/*
-		 * "Disable Repacking for Compression (masked R/W access)
-		 *  before rendering compressed surfaces for display."
-		 */
-		wa_masked_en(wal, CACHE_MODE_0_GEN7,
-			     DISABLE_REPACKING_FOR_COMPRESSION);
-	}
-
 	if (GRAPHICS_VER(i915) == 11) {
 		/* This is not an Wa. Enable for better image quality */
 		wa_masked_en(wal,

From 77a16455fae43e304e6adaf83da5b2ba6f3ad1ad Mon Sep 17 00:00:00 2001
From: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Date: Mon, 11 Aug 2025 09:12:39 +0000
Subject: [PATCH 0533/1455] drm/i915/gt: Relocate Gen7 context-specific
 workarounds

CACHE_MODE_1 and CACHE_MODE_0 register should be saved and restored
as part of the context, not during engine reset. Move the related
workarounds (RC_OP_FLUSH_ENABLE, PIXEL_SUBSPAN_COLLECT_OPT_DISABLE)
from rcs_engine_wa_init() to gen7_ctx_workarounds_init() for
Gen7 platforms. This ensures the WA is applied during context
initialisation.

BSPEC: 11322, 11323

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/06cf152803ab0050e09c521ac2fc3637549860b3.1754902406.git.sebastian.brzezinka@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 23 ++++++++++-----------
 1 file changed, 11 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 5a95f06900b5..18f1ee529de9 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -343,6 +343,17 @@ static void gen7_ctx_workarounds_init(struct intel_engine_cs *engine,
 				      struct i915_wa_list *wal)
 {
 	wa_masked_en(wal, INSTPM, INSTPM_FORCE_ORDERING);
+	/* WaDisable_RenderCache_OperationalFlush:ivb,vlv,hsw */
+	wa_masked_dis(wal, CACHE_MODE_0_GEN7, RC_OP_FLUSH_ENABLE);
+
+	/*
+	 * BSpec says this must be set, even though
+	 * WaDisable4x2SubspanOptimization:ivb,hsw
+	 * WaDisable4x2SubspanOptimization isn't listed for VLV.
+	 */
+	wa_masked_en(wal,
+		     CACHE_MODE_1,
+		     PIXEL_SUBSPAN_COLLECT_OPT_DISABLE);
 }
 
 static void gen8_ctx_workarounds_init(struct intel_engine_cs *engine,
@@ -2567,18 +2578,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 			     RING_MODE_GEN7(RENDER_RING_BASE),
 			     GFX_TLB_INVALIDATE_EXPLICIT | GFX_REPLAY_MODE);
 
-		/* WaDisable_RenderCache_OperationalFlush:ivb,vlv,hsw */
-		wa_masked_dis(wal, CACHE_MODE_0_GEN7, RC_OP_FLUSH_ENABLE);
-
-		/*
-		 * BSpec says this must be set, even though
-		 * WaDisable4x2SubspanOptimization:ivb,hsw
-		 * WaDisable4x2SubspanOptimization isn't listed for VLV.
-		 */
-		wa_masked_en(wal,
-			     CACHE_MODE_1,
-			     PIXEL_SUBSPAN_COLLECT_OPT_DISABLE);
-
 		/*
 		 * BSpec recommends 8x4 when MSAA is used,
 		 * however in practice 16x4 seems fastest.

From b7a855f488c6825e3dc06e78c49326491bef6f98 Mon Sep 17 00:00:00 2001
From: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Date: Mon, 11 Aug 2025 09:12:45 +0000
Subject: [PATCH 0534/1455] drm/i915/gt: Relocate Gen6 context-specific
 workaround

CACHE_MODE_0 register should be saved and restored as part
of the context, not during engine reset. Move the related
workaround (RC_OP_FLUSH_ENABLE) from rcs_engine_wa_init() to
gen6_ctx_workarounds_init() for Gen6 platforms. This ensures the WA
is applied during context initialisation.

CM0_STC_EVICT_DISABLE_LRA_SNB is also Gen6-specific, but it does
not stick when applied in context, so it remains in engine init.

BSPEC: 11322

Signed-off-by: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/f493bab389e51b2faf7c9a439724e9ea9ca04053.1754902406.git.sebastian.brzezinka@intel.com
---
 drivers/gpu/drm/i915/gt/intel_workarounds.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c
index 18f1ee529de9..7d486dfa2fc1 100644
--- a/drivers/gpu/drm/i915/gt/intel_workarounds.c
+++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c
@@ -337,6 +337,9 @@ static void gen6_ctx_workarounds_init(struct intel_engine_cs *engine,
 				      struct i915_wa_list *wal)
 {
 	wa_masked_en(wal, INSTPM, INSTPM_FORCE_ORDERING);
+
+	/* WaDisable_RenderCache_OperationalFlush:snb */
+	wa_masked_dis(wal, CACHE_MODE_0, RC_OP_FLUSH_ENABLE);
 }
 
 static void gen7_ctx_workarounds_init(struct intel_engine_cs *engine,
@@ -2644,9 +2647,6 @@ rcs_engine_wa_init(struct intel_engine_cs *engine, struct i915_wa_list *wal)
 				    GEN6_WIZ_HASHING_MASK,
 				    GEN6_WIZ_HASHING_16x4);
 
-		/* WaDisable_RenderCache_OperationalFlush:snb */
-		wa_masked_dis(wal, CACHE_MODE_0, RC_OP_FLUSH_ENABLE);
-
 		/*
 		 * From the Sandybridge PRM, volume 1 part 3, page 24:
 		 * "If this bit is set, STCunit will have LRA as replacement

From 2e9da93d604ef66b2ed6c5df7b7bfda3a187e71c Mon Sep 17 00:00:00 2001
From: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Date: Wed, 30 Jul 2025 11:25:22 +0530
Subject: [PATCH 0535/1455] drm/i915/dp: Refactor intel_dp_in_hdr_mode() for
 broader reuse

The intel_dp_in_hdr_mode() helper was previously defined in
intel_dp_aux_backlight.c but is generally useful beyond that
context. Move the function to intel_dp.c and declare it in
intel_dp.h to make it accessible to other DP-related code
paths that need to check HDR metadata state.

This is a pure refactor with no functional change and
prepares for a follow-up patch that uses this helper.

Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250730055523.2214966-2-chaitanya.kumar.borah@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c             | 13 +++++++++++++
 drivers/gpu/drm/i915/display/intel_dp.h             |  1 +
 .../gpu/drm/i915/display/intel_dp_aux_backlight.c   | 13 -------------
 3 files changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 625036c47bdf..8a155bc958f0 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -2917,6 +2917,19 @@ static void intel_dp_compute_vsc_sdp(struct intel_dp *intel_dp,
 	}
 }
 
+bool
+intel_dp_in_hdr_mode(const struct drm_connector_state *conn_state)
+{
+	struct hdr_output_metadata *hdr_metadata;
+
+	if (!conn_state->hdr_output_metadata)
+		return false;
+
+	hdr_metadata = conn_state->hdr_output_metadata->data;
+
+	return hdr_metadata->hdmi_metadata_type1.eotf == HDMI_EOTF_SMPTE_ST2084;
+}
+
 static void
 intel_dp_compute_hdr_metadata_infoframe_sdp(struct intel_dp *intel_dp,
 					    struct intel_crtc_state *crtc_state,
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index 0657f5681196..5def589e3c0e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -214,5 +214,6 @@ int intel_dp_compute_min_hblank(struct intel_crtc_state *crtc_state,
 
 int intel_dp_dsc_bpp_step_x16(const struct intel_connector *connector);
 void intel_dp_dpcd_set_probe(struct intel_dp *intel_dp, bool force_on_external);
+bool intel_dp_in_hdr_mode(const struct drm_connector_state *conn_state);
 
 #endif /* __INTEL_DP_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
index 271b27c9de51..f665d845c809 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
@@ -225,19 +225,6 @@ intel_dp_aux_hdr_set_aux_backlight(const struct drm_connector_state *conn_state,
 			connector->base.base.id, connector->base.name);
 }
 
-static bool
-intel_dp_in_hdr_mode(const struct drm_connector_state *conn_state)
-{
-	struct hdr_output_metadata *hdr_metadata;
-
-	if (!conn_state->hdr_output_metadata)
-		return false;
-
-	hdr_metadata = conn_state->hdr_output_metadata->data;
-
-	return hdr_metadata->hdmi_metadata_type1.eotf == HDMI_EOTF_SMPTE_ST2084;
-}
-
 static void
 intel_dp_aux_hdr_set_backlight(const struct drm_connector_state *conn_state, u32 level)
 {

From ba49a4643cf53c6e9582b4a31142b67f5829cd82 Mon Sep 17 00:00:00 2001
From: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Date: Wed, 30 Jul 2025 11:25:23 +0530
Subject: [PATCH 0536/1455] drm/i915/dp: Set min_bpp limit to 30 in HDR mode

Update intel_dp_compute_config_limits() to use a minimum of
30 bits per pixel when the connector is in HDR mode
(specifically, when EOTF is SMPTE ST2084), aligning with HDR
display requirements.

To support this, the function now takes a drm_connector_state
instead of an intel_connector, and the required updates are
made in all call sites, including MST handling.

This ensures sufficient bitdepth for HDR content to avoid
banding.

If the required bandwidth for 30 bpp cannot be supported,
the driver will either fall back to DSC or reject the mode
during atomic check if DSC is not supported.

Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250730055523.2214966-3-chaitanya.kumar.borah@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c     | 11 +++++++----
 drivers/gpu/drm/i915/display/intel_dp.h     |  2 +-
 drivers/gpu/drm/i915/display/intel_dp_mst.c | 11 +++++++----
 3 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 8a155bc958f0..996edb8deded 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -2536,13 +2536,15 @@ intel_dp_dsc_compute_pipe_bpp_limits(struct intel_dp *intel_dp,
 
 bool
 intel_dp_compute_config_limits(struct intel_dp *intel_dp,
-			       struct intel_connector *connector,
+			       struct drm_connector_state *conn_state,
 			       struct intel_crtc_state *crtc_state,
 			       bool respect_downstream_limits,
 			       bool dsc,
 			       struct link_config_limits *limits)
 {
 	bool is_mst = intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DP_MST);
+	struct intel_connector *connector =
+		to_intel_connector(conn_state->connector);
 
 	limits->min_rate = intel_dp_min_link_rate(intel_dp);
 	limits->max_rate = intel_dp_max_link_rate(intel_dp);
@@ -2552,7 +2554,8 @@ intel_dp_compute_config_limits(struct intel_dp *intel_dp,
 	limits->min_lane_count = intel_dp_min_lane_count(intel_dp);
 	limits->max_lane_count = intel_dp_max_lane_count(intel_dp);
 
-	limits->pipe.min_bpp = intel_dp_min_bpp(crtc_state->output_format);
+	limits->pipe.min_bpp = intel_dp_in_hdr_mode(conn_state) ? 30 :
+				intel_dp_min_bpp(crtc_state->output_format);
 	if (is_mst) {
 		/*
 		 * FIXME: If all the streams can't fit into the link with their
@@ -2651,7 +2654,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 	joiner_needs_dsc = intel_dp_joiner_needs_dsc(display, num_joined_pipes);
 
 	dsc_needed = joiner_needs_dsc || intel_dp->force_dsc_en ||
-		     !intel_dp_compute_config_limits(intel_dp, connector, pipe_config,
+		     !intel_dp_compute_config_limits(intel_dp, conn_state, pipe_config,
 						     respect_downstream_limits,
 						     false,
 						     &limits);
@@ -2685,7 +2688,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder,
 			    str_yes_no(ret), str_yes_no(joiner_needs_dsc),
 			    str_yes_no(intel_dp->force_dsc_en));
 
-		if (!intel_dp_compute_config_limits(intel_dp, connector, pipe_config,
+		if (!intel_dp_compute_config_limits(intel_dp, conn_state, pipe_config,
 						    respect_downstream_limits,
 						    true,
 						    &limits))
diff --git a/drivers/gpu/drm/i915/display/intel_dp.h b/drivers/gpu/drm/i915/display/intel_dp.h
index 5def589e3c0e..f90cfd1dbbd0 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.h
+++ b/drivers/gpu/drm/i915/display/intel_dp.h
@@ -193,7 +193,7 @@ void intel_dp_wait_source_oui(struct intel_dp *intel_dp);
 int intel_dp_output_bpp(enum intel_output_format output_format, int bpp);
 
 bool intel_dp_compute_config_limits(struct intel_dp *intel_dp,
-				    struct intel_connector *connector,
+				    struct drm_connector_state *conn_state,
 				    struct intel_crtc_state *crtc_state,
 				    bool respect_downstream_limits,
 				    bool dsc,
diff --git a/drivers/gpu/drm/i915/display/intel_dp_mst.c b/drivers/gpu/drm/i915/display/intel_dp_mst.c
index 74497c9a0554..352f7ef29c28 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_mst.c
@@ -611,12 +611,15 @@ adjust_limits_for_dsc_hblank_expansion_quirk(struct intel_dp *intel_dp,
 
 static bool
 mst_stream_compute_config_limits(struct intel_dp *intel_dp,
-				 struct intel_connector *connector,
+				 struct drm_connector_state *conn_state,
 				 struct intel_crtc_state *crtc_state,
 				 bool dsc,
 				 struct link_config_limits *limits)
 {
-	if (!intel_dp_compute_config_limits(intel_dp, connector,
+	struct intel_connector *connector =
+		to_intel_connector(conn_state->connector);
+
+	if (!intel_dp_compute_config_limits(intel_dp, conn_state,
 					    crtc_state, false, dsc,
 					    limits))
 		return false;
@@ -665,7 +668,7 @@ static int mst_stream_compute_config(struct intel_encoder *encoder,
 	joiner_needs_dsc = intel_dp_joiner_needs_dsc(display, num_joined_pipes);
 
 	dsc_needed = joiner_needs_dsc || intel_dp->force_dsc_en ||
-		!mst_stream_compute_config_limits(intel_dp, connector,
+		!mst_stream_compute_config_limits(intel_dp, conn_state,
 						  pipe_config, false, &limits);
 
 	if (!dsc_needed) {
@@ -691,7 +694,7 @@ static int mst_stream_compute_config(struct intel_encoder *encoder,
 			    str_yes_no(intel_dp->force_dsc_en));
 
 
-		if (!mst_stream_compute_config_limits(intel_dp, connector,
+		if (!mst_stream_compute_config_limits(intel_dp, conn_state,
 						      pipe_config, true,
 						      &limits))
 			return -EINVAL;

From 1d409b0a77ee55ae2062d74fb8c0148cb663bfb8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 18 Aug 2025 10:16:05 +0300
Subject: [PATCH 0537/1455] drm/i915/switcheroo: check for NULL before
 dereferencing

Both i915_switcheroo_set_state() and i915_switcheroo_can_switch() check
for i915 == NULL. Commit d2e184f8e16a ("drm/i915/switcheroo: pass
display to HAS_DISPLAY()") started dereferencing it before the NULL
check. Fix it.

Fixes: d2e184f8e16a ("drm/i915/switcheroo: pass display to HAS_DISPLAY()")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/r/202508160035.hmzuKiww-lkp@intel.com/
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250818071605.2541523-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/i915_switcheroo.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_switcheroo.c b/drivers/gpu/drm/i915/i915_switcheroo.c
index 231d27497706..3a95a55b2e87 100644
--- a/drivers/gpu/drm/i915/i915_switcheroo.c
+++ b/drivers/gpu/drm/i915/i915_switcheroo.c
@@ -15,7 +15,7 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev,
 				      enum vga_switcheroo_state state)
 {
 	struct drm_i915_private *i915 = pdev_to_i915(pdev);
-	struct intel_display *display = i915->display;
+	struct intel_display *display = i915 ? i915->display : NULL;
 	pm_message_t pmm = { .event = PM_EVENT_SUSPEND };
 
 	if (!i915) {
@@ -45,7 +45,7 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev,
 static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
 {
 	struct drm_i915_private *i915 = pdev_to_i915(pdev);
-	struct intel_display *display = i915->display;
+	struct intel_display *display = i915 ? i915->display : NULL;
 
 	/*
 	 * FIXME: open_count is protected by drm_global_mutex but that would lead to

From e0b1bbed5e48efe353a2f1381cc37ea092eb98b7 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 18 Aug 2025 13:07:24 +0300
Subject: [PATCH 0538/1455] drm/i915/dram: add intel_fsb_freq() and use it

Add a more generic intel_fsb_freq() function instead of platform
specific ones.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/c5b77311c5f64b7163c86a042b7d023c07a685e2.1755511595.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cdclk.c     |  2 +-
 drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c |  2 +-
 drivers/gpu/drm/i915/soc/intel_dram.c          | 14 ++++++++++----
 drivers/gpu/drm/i915/soc/intel_dram.h          |  2 +-
 4 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index f7f76de0f944..d7ba3970e1e9 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -3574,7 +3574,7 @@ static int i9xx_hrawclk(struct intel_display *display)
 	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	/* hrawclock is 1/4 the FSB frequency */
-	return DIV_ROUND_CLOSEST(i9xx_fsb_freq(i915), 4);
+	return DIV_ROUND_CLOSEST(intel_fsb_freq(i915), 4);
 }
 
 /**
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c
index 6c499692d61e..88b147fa5cb1 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c
@@ -148,7 +148,7 @@ static u32 gen4_read_clock_frequency(struct intel_uncore *uncore)
 	 *
 	 * Testing on actual hardware has shown there is no /16.
 	 */
-	return DIV_ROUND_CLOSEST(i9xx_fsb_freq(uncore->i915), 4) * 1000;
+	return DIV_ROUND_CLOSEST(intel_fsb_freq(uncore->i915), 4) * 1000;
 }
 
 static u32 read_clock_frequency(struct intel_uncore *uncore)
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index b3c407cc200f..366891cda6b9 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -154,7 +154,7 @@ static void detect_mem_freq(struct drm_i915_private *i915)
 		drm_dbg(&i915->drm, "DDR speed: %d kHz\n", i915->mem_freq);
 }
 
-unsigned int i9xx_fsb_freq(struct drm_i915_private *i915)
+static unsigned int i9xx_fsb_freq(struct drm_i915_private *i915)
 {
 	u32 fsb;
 
@@ -236,13 +236,19 @@ static unsigned int ilk_fsb_freq(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void detect_fsb_freq(struct drm_i915_private *i915)
+unsigned int intel_fsb_freq(struct drm_i915_private *i915)
 {
 	if (GRAPHICS_VER(i915) == 5)
-		i915->fsb_freq = ilk_fsb_freq(i915);
+		return ilk_fsb_freq(i915);
 	else if (GRAPHICS_VER(i915) == 3 || GRAPHICS_VER(i915) == 4)
-		i915->fsb_freq = i9xx_fsb_freq(i915);
+		return i9xx_fsb_freq(i915);
+	else
+		return 0;
+}
 
+static void detect_fsb_freq(struct drm_i915_private *i915)
+{
+	i915->fsb_freq = intel_fsb_freq(i915);
 	if (i915->fsb_freq)
 		drm_dbg(&i915->drm, "FSB frequency: %d kHz\n", i915->fsb_freq);
 }
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 2a696e03aad4..09a7a581d949 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -33,7 +33,7 @@ struct dram_info {
 
 void intel_dram_edram_detect(struct drm_i915_private *i915);
 int intel_dram_detect(struct drm_i915_private *i915);
-unsigned int i9xx_fsb_freq(struct drm_i915_private *i915);
+unsigned int intel_fsb_freq(struct drm_i915_private *i915);
 const struct dram_info *intel_dram_info(struct drm_device *drm);
 
 #endif /* __INTEL_DRAM_H__ */

From 6e977fe781a1d1cc003415ba32de18191e255150 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 18 Aug 2025 13:07:25 +0300
Subject: [PATCH 0539/1455] drm/i915/dram: add intel_mem_freq()

Add a more generic intel_mem_freq() function instead of platform
specific ones. Expose it for future use outside of intel_dram.c.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/602103b290a92ba26d581eeb595ba5e707eb5bc4.1755511595.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 17 ++++++++++++-----
 drivers/gpu/drm/i915/soc/intel_dram.h |  1 +
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 366891cda6b9..02b8fa784525 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -136,16 +136,23 @@ static unsigned int vlv_mem_freq(struct drm_i915_private *i915)
 	return 0;
 }
 
-static void detect_mem_freq(struct drm_i915_private *i915)
+unsigned int intel_mem_freq(struct drm_i915_private *i915)
 {
 	if (IS_PINEVIEW(i915))
-		i915->mem_freq = pnv_mem_freq(i915);
+		return pnv_mem_freq(i915);
 	else if (GRAPHICS_VER(i915) == 5)
-		i915->mem_freq = ilk_mem_freq(i915);
+		return ilk_mem_freq(i915);
 	else if (IS_CHERRYVIEW(i915))
-		i915->mem_freq = chv_mem_freq(i915);
+		return chv_mem_freq(i915);
 	else if (IS_VALLEYVIEW(i915))
-		i915->mem_freq = vlv_mem_freq(i915);
+		return vlv_mem_freq(i915);
+	else
+		return 0;
+}
+
+static void detect_mem_freq(struct drm_i915_private *i915)
+{
+	i915->mem_freq = intel_mem_freq(i915);
 
 	if (IS_PINEVIEW(i915))
 		i915->is_ddr3 = pnv_is_ddr3(i915);
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 09a7a581d949..5ba75e279e84 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -34,6 +34,7 @@ struct dram_info {
 void intel_dram_edram_detect(struct drm_i915_private *i915);
 int intel_dram_detect(struct drm_i915_private *i915);
 unsigned int intel_fsb_freq(struct drm_i915_private *i915);
+unsigned int intel_mem_freq(struct drm_i915_private *i915);
 const struct dram_info *intel_dram_info(struct drm_device *drm);
 
 #endif /* __INTEL_DRAM_H__ */

From 6c9f41d8fa808f9020fbd64f74cf872080f2cd74 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 18 Aug 2025 13:07:26 +0300
Subject: [PATCH 0540/1455] drm/i915/rps: use intel_fsb_freq() and
 intel_mem_freq()

The rps init only happens once, so it's not important to use the cached
versions, and we can drop the dependency on them.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/6f3b703f7cb5605bf139cbe27697c1d4ffe7e719.1755511595.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_rps.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index 0b35fdd461d4..006042e0b229 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -9,6 +9,7 @@
 
 #include "display/intel_display.h"
 #include "display/intel_display_rps.h"
+#include "soc/intel_dram.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
@@ -276,20 +277,24 @@ static void gen5_rps_init(struct intel_rps *rps)
 {
 	struct drm_i915_private *i915 = rps_to_i915(rps);
 	struct intel_uncore *uncore = rps_to_uncore(rps);
+	unsigned int fsb_freq, mem_freq;
 	u8 fmax, fmin, fstart;
 	u32 rgvmodectl;
 	int c_m, i;
 
-	if (i915->fsb_freq <= 3200000)
+	fsb_freq = intel_fsb_freq(i915);
+	mem_freq = intel_mem_freq(i915);
+
+	if (fsb_freq <= 3200000)
 		c_m = 0;
-	else if (i915->fsb_freq <= 4800000)
+	else if (fsb_freq <= 4800000)
 		c_m = 1;
 	else
 		c_m = 2;
 
 	for (i = 0; i < ARRAY_SIZE(cparams); i++) {
 		if (cparams[i].i == c_m &&
-		    cparams[i].t == DIV_ROUND_CLOSEST(i915->mem_freq, 1000)) {
+		    cparams[i].t == DIV_ROUND_CLOSEST(mem_freq, 1000)) {
 			rps->ips.m = cparams[i].m;
 			rps->ips.c = cparams[i].c;
 			break;

From 10e656f8c4a6b99f905bf024712c4d56882bb6a9 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 18 Aug 2025 13:07:27 +0300
Subject: [PATCH 0541/1455] drm/i915/dram: bypass fsb/mem freq detection on dg2
 and no display

Non-display now calls the intel_fsb_freq() and intel_mem_freq()
functions, so we don't have to have the frequencies initialized for dg2
or non-display cases.

This is in preparation for unifying the pre-gen9 handling in dram info.

DG2 remains a special case as described in commit 5eb6bf0b44e7
("drm/i915/dg2: Don't read DRAM info").

v2: Rebase

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/7bfed06d431354f3918ea73d43a2ec8ed9426a76.1755511595.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 02b8fa784525..e087e8d205f7 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -727,10 +727,13 @@ int intel_dram_detect(struct drm_i915_private *i915)
 	struct dram_info *dram_info;
 	int ret;
 
+	if (IS_DG2(i915) || !HAS_DISPLAY(display))
+		return 0;
+
 	detect_fsb_freq(i915);
 	detect_mem_freq(i915);
 
-	if (GRAPHICS_VER(i915) < 9 || IS_DG2(i915) || !HAS_DISPLAY(display))
+	if (GRAPHICS_VER(i915) < 9)
 		return 0;
 
 	dram_info = drmm_kzalloc(&i915->drm, sizeof(*dram_info), GFP_KERNEL);

From 0492e13eb819008e0ce7570a7e152559e9a89a4e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Mon, 18 Aug 2025 13:07:28 +0300
Subject: [PATCH 0542/1455] drm/i915/dram: move fsb_freq and mem_freq to dram
 info

Store fsb_freq and mem_freq in dram info the same way we do for other
memory info on later platforms for a slightly more unified approach.

This allows us to remove fsb_freq, mem_freq and is_ddr3 members from
struct drm_i915_private and struct xe_device.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/a38c4b105ba9098fa0b128cb86cd4eb63bcc27e8.1755511595.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/i9xx_wm.c | 13 +++++----
 drivers/gpu/drm/i915/i915_drv.h        |  2 --
 drivers/gpu/drm/i915/soc/intel_dram.c  | 38 +++++++++++---------------
 drivers/gpu/drm/i915/soc/intel_dram.h  |  2 ++
 drivers/gpu/drm/xe/xe_device_types.h   |  1 -
 5 files changed, 26 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c
index 1f9db5118777..591acce2a4b1 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.c
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.c
@@ -3,6 +3,8 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include "soc/intel_dram.h"
+
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "i9xx_wm.h"
@@ -85,7 +87,8 @@ static const struct cxsr_latency cxsr_latency_table[] = {
 
 static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
+	const struct dram_info *dram_info = intel_dram_info(display->drm);
+	bool is_ddr3 = dram_info->type == INTEL_DRAM_DDR3;
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(cxsr_latency_table); i++) {
@@ -93,15 +96,15 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *dis
 		bool is_desktop = !display->platform.mobile;
 
 		if (is_desktop == latency->is_desktop &&
-		    i915->is_ddr3 == latency->is_ddr3 &&
-		    DIV_ROUND_CLOSEST(i915->fsb_freq, 1000) == latency->fsb_freq &&
-		    DIV_ROUND_CLOSEST(i915->mem_freq, 1000) == latency->mem_freq)
+		    is_ddr3 == latency->is_ddr3 &&
+		    DIV_ROUND_CLOSEST(dram_info->fsb_freq, 1000) == latency->fsb_freq &&
+		    DIV_ROUND_CLOSEST(dram_info->mem_freq, 1000) == latency->mem_freq)
 			return latency;
 	}
 
 	drm_dbg_kms(display->drm,
 		    "Could not find CxSR latency for DDR%s, FSB %u kHz, MEM %u kHz\n",
-		    i915->is_ddr3 ? "3" : "2", i915->fsb_freq, i915->mem_freq);
+		    is_ddr3 ? "3" : "2", dram_info->fsb_freq, dram_info->mem_freq);
 
 	return NULL;
 }
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 5e4c49f0d5d4..f76d138df0af 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -239,8 +239,6 @@ struct drm_i915_private {
 
 	bool preserve_bios_swizzle;
 
-	unsigned int fsb_freq, mem_freq, is_ddr3;
-
 	unsigned int hpll_freq;
 	unsigned int czclk_freq;
 
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index e087e8d205f7..3eb748ab44d9 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -150,17 +150,6 @@ unsigned int intel_mem_freq(struct drm_i915_private *i915)
 		return 0;
 }
 
-static void detect_mem_freq(struct drm_i915_private *i915)
-{
-	i915->mem_freq = intel_mem_freq(i915);
-
-	if (IS_PINEVIEW(i915))
-		i915->is_ddr3 = pnv_is_ddr3(i915);
-
-	if (i915->mem_freq)
-		drm_dbg(&i915->drm, "DDR speed: %d kHz\n", i915->mem_freq);
-}
-
 static unsigned int i9xx_fsb_freq(struct drm_i915_private *i915)
 {
 	u32 fsb;
@@ -253,11 +242,20 @@ unsigned int intel_fsb_freq(struct drm_i915_private *i915)
 		return 0;
 }
 
-static void detect_fsb_freq(struct drm_i915_private *i915)
+static int i915_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info)
 {
-	i915->fsb_freq = intel_fsb_freq(i915);
-	if (i915->fsb_freq)
-		drm_dbg(&i915->drm, "FSB frequency: %d kHz\n", i915->fsb_freq);
+	dram_info->fsb_freq = intel_fsb_freq(i915);
+	if (dram_info->fsb_freq)
+		drm_dbg(&i915->drm, "FSB frequency: %d kHz\n", dram_info->fsb_freq);
+
+	dram_info->mem_freq = intel_mem_freq(i915);
+	if (dram_info->mem_freq)
+		drm_dbg(&i915->drm, "DDR speed: %d kHz\n", dram_info->mem_freq);
+
+	if (IS_PINEVIEW(i915) && pnv_is_ddr3(i915))
+		dram_info->type = INTEL_DRAM_DDR3;
+
+	return 0;
 }
 
 static int intel_dimm_num_devices(const struct dram_dimm_info *dimm)
@@ -730,12 +728,6 @@ int intel_dram_detect(struct drm_i915_private *i915)
 	if (IS_DG2(i915) || !HAS_DISPLAY(display))
 		return 0;
 
-	detect_fsb_freq(i915);
-	detect_mem_freq(i915);
-
-	if (GRAPHICS_VER(i915) < 9)
-		return 0;
-
 	dram_info = drmm_kzalloc(&i915->drm, sizeof(*dram_info), GFP_KERNEL);
 	if (!dram_info)
 		return -ENOMEM;
@@ -756,8 +748,10 @@ int intel_dram_detect(struct drm_i915_private *i915)
 		ret = gen11_get_dram_info(i915, dram_info);
 	else if (IS_BROXTON(i915) || IS_GEMINILAKE(i915))
 		ret = bxt_get_dram_info(i915, dram_info);
-	else
+	else if (GRAPHICS_VER(i915) >= 9)
 		ret = skl_get_dram_info(i915, dram_info);
+	else
+		ret = i915_get_dram_info(i915, dram_info);
 
 	drm_dbg_kms(&i915->drm, "DRAM type: %s\n",
 		    intel_dram_type_str(dram_info->type));
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 5ba75e279e84..97d21894abdc 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -29,6 +29,8 @@ struct dram_info {
 	} type;
 	u8 num_qgv_points;
 	u8 num_psf_gv_points;
+	unsigned int fsb_freq;
+	unsigned int mem_freq;
 };
 
 void intel_dram_edram_detect(struct drm_i915_private *i915);
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 4f795157f711..7cba3413323d 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -621,7 +621,6 @@ struct xe_device {
 	struct {
 		unsigned int hpll_freq;
 		unsigned int czclk_freq;
-		unsigned int fsb_freq, mem_freq, is_ddr3;
 	};
 #endif
 };

From f1f2a22b8683d7ac38821d4508d4549a2f0c0a0a Mon Sep 17 00:00:00 2001
From: Shankari Anand <shankari.ak0208@gmail.com>
Date: Fri, 15 Aug 2025 21:47:06 +0530
Subject: [PATCH 0543/1455] rust: drm: update ARef and AlwaysRefCounted imports
 from sync::aref

Update call sites in drm to import `ARef` and
`AlwaysRefCounted` from `sync::aref` instead of `types`.

This aligns with the ongoing effort to move `ARef` and
`AlwaysRefCounted` to sync.

Suggested-by: Benno Lossin <lossin@kernel.org>
Link: https://github.com/Rust-for-Linux/linux/issues/1173
Signed-off-by: Shankari Anand <shankari.ak0208@gmail.com>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Reviewed-by: Elle Rhumsaa <elle@weathered-steel.dev>
Link: https://lore.kernel.org/r/20250815161706.1324860-1-shankari.ak0208@gmail.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/drm/device.rs  | 3 ++-
 rust/kernel/drm/driver.rs  | 2 +-
 rust/kernel/drm/gem/mod.rs | 3 ++-
 3 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/rust/kernel/drm/device.rs b/rust/kernel/drm/device.rs
index 3bb7c83966cf..4a62f9fd88b7 100644
--- a/rust/kernel/drm/device.rs
+++ b/rust/kernel/drm/device.rs
@@ -10,7 +10,8 @@ use crate::{
     error::from_err_ptr,
     error::Result,
     prelude::*,
-    types::{ARef, AlwaysRefCounted, Opaque},
+    sync::aref::{ARef, AlwaysRefCounted},
+    types::Opaque,
 };
 use core::{mem, ops::Deref, ptr, ptr::NonNull};
 
diff --git a/rust/kernel/drm/driver.rs b/rust/kernel/drm/driver.rs
index fe7e8d06961a..8fefae41bcc6 100644
--- a/rust/kernel/drm/driver.rs
+++ b/rust/kernel/drm/driver.rs
@@ -8,7 +8,7 @@ use crate::{
     bindings, device, devres, drm,
     error::{to_result, Result},
     prelude::*,
-    types::ARef,
+    sync::aref::ARef,
 };
 use macros::vtable;
 
diff --git a/rust/kernel/drm/gem/mod.rs b/rust/kernel/drm/gem/mod.rs
index b71821cfb5ea..a822aedee949 100644
--- a/rust/kernel/drm/gem/mod.rs
+++ b/rust/kernel/drm/gem/mod.rs
@@ -10,7 +10,8 @@ use crate::{
     drm::driver::{AllocImpl, AllocOps},
     error::{to_result, Result},
     prelude::*,
-    types::{ARef, AlwaysRefCounted, Opaque},
+    sync::aref::{ARef, AlwaysRefCounted},
+    types::Opaque,
 };
 use core::{mem, ops::Deref, ptr::NonNull};
 

From 000a45dce7adc13e45b2925b383e39f32e5f3004 Mon Sep 17 00:00:00 2001
From: Boris Brezillon <boris.brezillon@collabora.com>
Date: Tue, 19 Aug 2025 21:50:55 +0530
Subject: [PATCH 0544/1455] drm/gpuvm: Pass map arguments through a struct

We are about to pass more arguments to drm_gpuvm_sm_map[_ops_create](),
so, before we do that, let's pass arguments through a struct instead
of changing each call site every time a new optional argument is added.

Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Brendan King <Brendan.King@imgtec.com>
Cc: Matt Coster <matt.coster@imgtec.com>
Cc: Boris Brezillon <bbrezillon@kernel.org>
Cc: Caterina Shablia <caterina.shablia@collabora.com>
Cc: Rob Clark <robin.clark@oss.qualcomm.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: <dri-devel@lists.freedesktop.org>
Co-developed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Matt Coster <matt.coster@imgtec.com> # imagination/pvr_vm.c
Acked-by: Matt Coster <matt.coster@imgtec.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-2-himal.prasad.ghimiray@intel.com
---
 drivers/gpu/drm/drm_gpuvm.c            | 74 +++++++++++---------------
 drivers/gpu/drm/imagination/pvr_vm.c   | 15 ++++--
 drivers/gpu/drm/msm/msm_gem_vma.c      | 27 +++++++---
 drivers/gpu/drm/nouveau/nouveau_uvmm.c | 11 ++--
 drivers/gpu/drm/panthor/panthor_mmu.c  | 13 +++--
 drivers/gpu/drm/xe/xe_vm.c             | 13 +++--
 include/drm/drm_gpuvm.h                | 20 ++++---
 7 files changed, 102 insertions(+), 71 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpuvm.c b/drivers/gpu/drm/drm_gpuvm.c
index bbc7fecb6f4a..6c18cec70f11 100644
--- a/drivers/gpu/drm/drm_gpuvm.c
+++ b/drivers/gpu/drm/drm_gpuvm.c
@@ -486,13 +486,18 @@
  *				  u64 addr, u64 range,
  *				  struct drm_gem_object *obj, u64 offset)
  *	{
+ *		struct drm_gpuvm_map_req map_req = {
+ *		        .map.va.addr = addr,
+ *	                .map.va.range = range,
+ *	                .map.gem.obj = obj,
+ *	                .map.gem.offset = offset,
+ *	           };
  *		struct drm_gpuva_ops *ops;
  *		struct drm_gpuva_op *op
  *		struct drm_gpuvm_bo *vm_bo;
  *
  *		driver_lock_va_space();
- *		ops = drm_gpuvm_sm_map_ops_create(gpuvm, addr, range,
- *						  obj, offset);
+ *		ops = drm_gpuvm_sm_map_ops_create(gpuvm, &map_req);
  *		if (IS_ERR(ops))
  *			return PTR_ERR(ops);
  *
@@ -2054,16 +2059,15 @@ EXPORT_SYMBOL_GPL(drm_gpuva_unmap);
 
 static int
 op_map_cb(const struct drm_gpuvm_ops *fn, void *priv,
-	  u64 addr, u64 range,
-	  struct drm_gem_object *obj, u64 offset)
+	  const struct drm_gpuvm_map_req *req)
 {
 	struct drm_gpuva_op op = {};
 
 	op.op = DRM_GPUVA_OP_MAP;
-	op.map.va.addr = addr;
-	op.map.va.range = range;
-	op.map.gem.obj = obj;
-	op.map.gem.offset = offset;
+	op.map.va.addr = req->map.va.addr;
+	op.map.va.range = req->map.va.range;
+	op.map.gem.obj = req->map.gem.obj;
+	op.map.gem.offset = req->map.gem.offset;
 
 	return fn->sm_step_map(&op, priv);
 }
@@ -2102,10 +2106,14 @@ op_unmap_cb(const struct drm_gpuvm_ops *fn, void *priv,
 static int
 __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 		   const struct drm_gpuvm_ops *ops, void *priv,
-		   u64 req_addr, u64 req_range,
-		   struct drm_gem_object *req_obj, u64 req_offset)
+		   const struct drm_gpuvm_map_req *req)
 {
+	struct drm_gem_object *req_obj = req->map.gem.obj;
 	struct drm_gpuva *va, *next;
+
+	u64 req_offset = req->map.gem.offset;
+	u64 req_range = req->map.va.range;
+	u64 req_addr = req->map.va.addr;
 	u64 req_end = req_addr + req_range;
 	int ret;
 
@@ -2236,9 +2244,7 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 		}
 	}
 
-	return op_map_cb(ops, priv,
-			 req_addr, req_range,
-			 req_obj, req_offset);
+	return op_map_cb(ops, priv, req);
 }
 
 static int
@@ -2303,10 +2309,7 @@ __drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm,
  * drm_gpuvm_sm_map() - calls the &drm_gpuva_op split/merge steps
  * @gpuvm: the &drm_gpuvm representing the GPU VA space
  * @priv: pointer to a driver private data structure
- * @req_addr: the start address of the new mapping
- * @req_range: the range of the new mapping
- * @req_obj: the &drm_gem_object to map
- * @req_offset: the offset within the &drm_gem_object
+ * @req: ptr to struct drm_gpuvm_map_req
  *
  * This function iterates the given range of the GPU VA space. It utilizes the
  * &drm_gpuvm_ops to call back into the driver providing the split and merge
@@ -2333,8 +2336,7 @@ __drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm,
  */
 int
 drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv,
-		 u64 req_addr, u64 req_range,
-		 struct drm_gem_object *req_obj, u64 req_offset)
+		 const struct drm_gpuvm_map_req *req)
 {
 	const struct drm_gpuvm_ops *ops = gpuvm->ops;
 
@@ -2343,9 +2345,7 @@ drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv,
 		       ops->sm_step_unmap)))
 		return -EINVAL;
 
-	return __drm_gpuvm_sm_map(gpuvm, ops, priv,
-				  req_addr, req_range,
-				  req_obj, req_offset);
+	return __drm_gpuvm_sm_map(gpuvm, ops, priv, req);
 }
 EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map);
 
@@ -2421,10 +2421,7 @@ static const struct drm_gpuvm_ops lock_ops = {
  * @gpuvm: the &drm_gpuvm representing the GPU VA space
  * @exec: the &drm_exec locking context
  * @num_fences: for newly mapped objects, the # of fences to reserve
- * @req_addr: the start address of the range to unmap
- * @req_range: the range of the mappings to unmap
- * @req_obj: the &drm_gem_object to map
- * @req_offset: the offset within the &drm_gem_object
+ * @req: ptr to drm_gpuvm_map_req struct
  *
  * This function locks (drm_exec_lock_obj()) objects that will be unmapped/
  * remapped, and locks+prepares (drm_exec_prepare_object()) objects that
@@ -2445,9 +2442,7 @@ static const struct drm_gpuvm_ops lock_ops = {
  *                    ret = drm_gpuvm_sm_unmap_exec_lock(gpuvm, &exec, op->addr, op->range);
  *                    break;
  *                case DRIVER_OP_MAP:
- *                    ret = drm_gpuvm_sm_map_exec_lock(gpuvm, &exec, num_fences,
- *                                                     op->addr, op->range,
- *                                                     obj, op->obj_offset);
+ *                    ret = drm_gpuvm_sm_map_exec_lock(gpuvm, &exec, num_fences, &req);
  *                    break;
  *                }
  *
@@ -2478,18 +2473,17 @@ static const struct drm_gpuvm_ops lock_ops = {
 int
 drm_gpuvm_sm_map_exec_lock(struct drm_gpuvm *gpuvm,
 			   struct drm_exec *exec, unsigned int num_fences,
-			   u64 req_addr, u64 req_range,
-			   struct drm_gem_object *req_obj, u64 req_offset)
+			   struct drm_gpuvm_map_req *req)
 {
+	struct drm_gem_object *req_obj = req->map.gem.obj;
+
 	if (req_obj) {
 		int ret = drm_exec_prepare_obj(exec, req_obj, num_fences);
 		if (ret)
 			return ret;
 	}
 
-	return __drm_gpuvm_sm_map(gpuvm, &lock_ops, exec,
-				  req_addr, req_range,
-				  req_obj, req_offset);
+	return __drm_gpuvm_sm_map(gpuvm, &lock_ops, exec, req);
 
 }
 EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map_exec_lock);
@@ -2611,10 +2605,7 @@ static const struct drm_gpuvm_ops gpuvm_list_ops = {
 /**
  * drm_gpuvm_sm_map_ops_create() - creates the &drm_gpuva_ops to split and merge
  * @gpuvm: the &drm_gpuvm representing the GPU VA space
- * @req_addr: the start address of the new mapping
- * @req_range: the range of the new mapping
- * @req_obj: the &drm_gem_object to map
- * @req_offset: the offset within the &drm_gem_object
+ * @req: map request arguments
  *
  * This function creates a list of operations to perform splitting and merging
  * of existent mapping(s) with the newly requested one.
@@ -2642,8 +2633,7 @@ static const struct drm_gpuvm_ops gpuvm_list_ops = {
  */
 struct drm_gpuva_ops *
 drm_gpuvm_sm_map_ops_create(struct drm_gpuvm *gpuvm,
-			    u64 req_addr, u64 req_range,
-			    struct drm_gem_object *req_obj, u64 req_offset)
+			    const struct drm_gpuvm_map_req *req)
 {
 	struct drm_gpuva_ops *ops;
 	struct {
@@ -2661,9 +2651,7 @@ drm_gpuvm_sm_map_ops_create(struct drm_gpuvm *gpuvm,
 	args.vm = gpuvm;
 	args.ops = ops;
 
-	ret = __drm_gpuvm_sm_map(gpuvm, &gpuvm_list_ops, &args,
-				 req_addr, req_range,
-				 req_obj, req_offset);
+	ret = __drm_gpuvm_sm_map(gpuvm, &gpuvm_list_ops, &args, req);
 	if (ret)
 		goto err_free_ops;
 
diff --git a/drivers/gpu/drm/imagination/pvr_vm.c b/drivers/gpu/drm/imagination/pvr_vm.c
index 2896fa7501b1..3d97990170bf 100644
--- a/drivers/gpu/drm/imagination/pvr_vm.c
+++ b/drivers/gpu/drm/imagination/pvr_vm.c
@@ -185,12 +185,17 @@ struct pvr_vm_bind_op {
 static int pvr_vm_bind_op_exec(struct pvr_vm_bind_op *bind_op)
 {
 	switch (bind_op->type) {
-	case PVR_VM_BIND_TYPE_MAP:
+	case PVR_VM_BIND_TYPE_MAP: {
+		const struct drm_gpuvm_map_req map_req = {
+			.map.va.addr = bind_op->device_addr,
+			.map.va.range = bind_op->size,
+			.map.gem.obj = gem_from_pvr_gem(bind_op->pvr_obj),
+			.map.gem.offset = bind_op->offset,
+		};
+
 		return drm_gpuvm_sm_map(&bind_op->vm_ctx->gpuvm_mgr,
-					bind_op, bind_op->device_addr,
-					bind_op->size,
-					gem_from_pvr_gem(bind_op->pvr_obj),
-					bind_op->offset);
+					bind_op, &map_req);
+	}
 
 	case PVR_VM_BIND_TYPE_UNMAP:
 		return drm_gpuvm_sm_unmap(&bind_op->vm_ctx->gpuvm_mgr,
diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 3cd8562a5109..792b35433666 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -1171,11 +1171,17 @@ vm_bind_job_lock_objects(struct msm_vm_bind_job *job, struct drm_exec *exec)
 							      op->obj_offset);
 				break;
 			case MSM_VM_BIND_OP_MAP:
-			case MSM_VM_BIND_OP_MAP_NULL:
-				ret = drm_gpuvm_sm_map_exec_lock(job->vm, exec, 1,
-							    op->iova, op->range,
-							    op->obj, op->obj_offset);
+			case MSM_VM_BIND_OP_MAP_NULL: {
+				struct drm_gpuvm_map_req map_req = {
+					.map.va.addr = op->iova,
+					.map.va.range = op->range,
+					.map.gem.obj = op->obj,
+					.map.gem.offset = op->obj_offset,
+				};
+
+				ret = drm_gpuvm_sm_map_exec_lock(job->vm, exec, 1, &map_req);
 				break;
+			}
 			default:
 				/*
 				 * lookup_op() should have already thrown an error for
@@ -1282,10 +1288,17 @@ vm_bind_job_prepare(struct msm_vm_bind_job *job)
 			if (op->flags & MSM_VM_BIND_OP_DUMP)
 				arg.flags |= MSM_VMA_DUMP;
 			fallthrough;
-		case MSM_VM_BIND_OP_MAP_NULL:
-			ret = drm_gpuvm_sm_map(job->vm, &arg, op->iova,
-					       op->range, op->obj, op->obj_offset);
+		case MSM_VM_BIND_OP_MAP_NULL: {
+			struct drm_gpuvm_map_req map_req = {
+				.map.va.addr = op->iova,
+				.map.va.range = op->range,
+				.map.gem.obj = op->obj,
+				.map.gem.offset = op->obj_offset,
+			};
+
+			ret = drm_gpuvm_sm_map(job->vm, &arg, &map_req);
 			break;
+		}
 		default:
 			/*
 			 * lookup_op() should have already thrown an error for
diff --git a/drivers/gpu/drm/nouveau/nouveau_uvmm.c b/drivers/gpu/drm/nouveau/nouveau_uvmm.c
index ddfc46bc1b3e..d94a85509176 100644
--- a/drivers/gpu/drm/nouveau/nouveau_uvmm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_uvmm.c
@@ -1276,6 +1276,12 @@ nouveau_uvmm_bind_job_submit(struct nouveau_job *job,
 			break;
 		case OP_MAP: {
 			struct nouveau_uvma_region *reg;
+			struct drm_gpuvm_map_req map_req = {
+				.map.va.addr = op->va.addr,
+				.map.va.range = op->va.range,
+				.map.gem.obj = op->gem.obj,
+				.map.gem.offset = op->gem.offset,
+			};
 
 			reg = nouveau_uvma_region_find_first(uvmm,
 							     op->va.addr,
@@ -1301,10 +1307,7 @@ nouveau_uvmm_bind_job_submit(struct nouveau_job *job,
 			}
 
 			op->ops = drm_gpuvm_sm_map_ops_create(&uvmm->base,
-							      op->va.addr,
-							      op->va.range,
-							      op->gem.obj,
-							      op->gem.offset);
+							      &map_req);
 			if (IS_ERR(op->ops)) {
 				ret = PTR_ERR(op->ops);
 				goto unwind_continue;
diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index 367c89aca558..2003b91a8409 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -2202,15 +2202,22 @@ panthor_vm_exec_op(struct panthor_vm *vm, struct panthor_vm_op_ctx *op,
 	mutex_lock(&vm->op_lock);
 	vm->op_ctx = op;
 	switch (op_type) {
-	case DRM_PANTHOR_VM_BIND_OP_TYPE_MAP:
+	case DRM_PANTHOR_VM_BIND_OP_TYPE_MAP: {
+		const struct drm_gpuvm_map_req map_req = {
+			.map.va.addr = op->va.addr,
+			.map.va.range = op->va.range,
+			.map.gem.obj = op->map.vm_bo->obj,
+			.map.gem.offset = op->map.bo_offset,
+		};
+
 		if (vm->unusable) {
 			ret = -EINVAL;
 			break;
 		}
 
-		ret = drm_gpuvm_sm_map(&vm->base, vm, op->va.addr, op->va.range,
-				       op->map.vm_bo->obj, op->map.bo_offset);
+		ret = drm_gpuvm_sm_map(&vm->base, vm, &map_req);
 		break;
+	}
 
 	case DRM_PANTHOR_VM_BIND_OP_TYPE_UNMAP:
 		ret = drm_gpuvm_sm_unmap(&vm->base, vm, op->va.addr, op->va.range);
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 2035604121e6..86842247e9d8 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2316,10 +2316,17 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops,
 
 	switch (operation) {
 	case DRM_XE_VM_BIND_OP_MAP:
-	case DRM_XE_VM_BIND_OP_MAP_USERPTR:
-		ops = drm_gpuvm_sm_map_ops_create(&vm->gpuvm, addr, range,
-						  obj, bo_offset_or_userptr);
+	case DRM_XE_VM_BIND_OP_MAP_USERPTR: {
+		struct drm_gpuvm_map_req map_req = {
+			.map.va.addr = addr,
+			.map.va.range = range,
+			.map.gem.obj = obj,
+			.map.gem.offset = bo_offset_or_userptr,
+		};
+
+		ops = drm_gpuvm_sm_map_ops_create(&vm->gpuvm, &map_req);
 		break;
+	}
 	case DRM_XE_VM_BIND_OP_UNMAP:
 		ops = drm_gpuvm_sm_unmap_ops_create(&vm->gpuvm, addr, range);
 		break;
diff --git a/include/drm/drm_gpuvm.h b/include/drm/drm_gpuvm.h
index 274532facfd6..a9fa44148e0c 100644
--- a/include/drm/drm_gpuvm.h
+++ b/include/drm/drm_gpuvm.h
@@ -1058,10 +1058,20 @@ struct drm_gpuva_ops {
  */
 #define drm_gpuva_next_op(op) list_next_entry(op, entry)
 
+/**
+ * struct drm_gpuvm_map_req - arguments passed to drm_gpuvm_sm_map[_ops_create]()
+ */
+struct drm_gpuvm_map_req {
+	/**
+	 * @op_map: struct drm_gpuva_op_map
+	 */
+	struct drm_gpuva_op_map map;
+};
+
 struct drm_gpuva_ops *
 drm_gpuvm_sm_map_ops_create(struct drm_gpuvm *gpuvm,
-			    u64 addr, u64 range,
-			    struct drm_gem_object *obj, u64 offset);
+			    const struct drm_gpuvm_map_req *req);
+
 struct drm_gpuva_ops *
 drm_gpuvm_sm_unmap_ops_create(struct drm_gpuvm *gpuvm,
 			      u64 addr, u64 range);
@@ -1205,16 +1215,14 @@ struct drm_gpuvm_ops {
 };
 
 int drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv,
-		     u64 addr, u64 range,
-		     struct drm_gem_object *obj, u64 offset);
+		     const struct drm_gpuvm_map_req *req);
 
 int drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm, void *priv,
 		       u64 addr, u64 range);
 
 int drm_gpuvm_sm_map_exec_lock(struct drm_gpuvm *gpuvm,
 			  struct drm_exec *exec, unsigned int num_fences,
-			  u64 req_addr, u64 req_range,
-			  struct drm_gem_object *obj, u64 offset);
+			  struct drm_gpuvm_map_req *req);
 
 int drm_gpuvm_sm_unmap_exec_lock(struct drm_gpuvm *gpuvm, struct drm_exec *exec,
 				 u64 req_addr, u64 req_range);

From 3309323241fbb3c1da885e6b84bdf95e9708e4bb Mon Sep 17 00:00:00 2001
From: Boris Brezillon <boris.brezillon@collabora.com>
Date: Tue, 19 Aug 2025 21:50:56 +0530
Subject: [PATCH 0545/1455] drm/gpuvm: Kill drm_gpuva_init()

drm_gpuva_init() only has one internal user, and given we are about to
add new optional fields, it only add maintenance burden for no real
benefit, so let's kill the thing now.

Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Rob Clark <robin.clark@oss.qualcomm.com>
Signed-off-by: Boris Brezillon <boris.brezillon@collabora.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-3-himal.prasad.ghimiray@intel.com
---
 drivers/gpu/drm/msm/msm_gem_vma.c |  8 +++++++-
 include/drm/drm_gpuvm.h           | 15 ++++-----------
 2 files changed, 11 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 792b35433666..210604181c05 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -371,6 +371,12 @@ struct drm_gpuva *
 msm_gem_vma_new(struct drm_gpuvm *gpuvm, struct drm_gem_object *obj,
 		u64 offset, u64 range_start, u64 range_end)
 {
+	struct drm_gpuva_op_map op_map = {
+		.va.addr = range_start,
+		.va.range = range_end - range_start,
+		.gem.obj = obj,
+		.gem.offset = offset,
+	};
 	struct msm_gem_vm *vm = to_msm_vm(gpuvm);
 	struct drm_gpuvm_bo *vm_bo;
 	struct msm_gem_vma *vma;
@@ -399,7 +405,7 @@ msm_gem_vma_new(struct drm_gpuvm *gpuvm, struct drm_gem_object *obj,
 	if (obj)
 		GEM_WARN_ON((range_end - range_start) > obj->size);
 
-	drm_gpuva_init(&vma->base, range_start, range_end - range_start, obj, offset);
+	drm_gpuva_init_from_op(&vma->base, &op_map);
 	vma->mapped = false;
 
 	ret = drm_gpuva_insert(&vm->base, &vma->base);
diff --git a/include/drm/drm_gpuvm.h b/include/drm/drm_gpuvm.h
index a9fa44148e0c..05347ac6cc73 100644
--- a/include/drm/drm_gpuvm.h
+++ b/include/drm/drm_gpuvm.h
@@ -160,15 +160,6 @@ struct drm_gpuva *drm_gpuva_find_first(struct drm_gpuvm *gpuvm,
 struct drm_gpuva *drm_gpuva_find_prev(struct drm_gpuvm *gpuvm, u64 start);
 struct drm_gpuva *drm_gpuva_find_next(struct drm_gpuvm *gpuvm, u64 end);
 
-static inline void drm_gpuva_init(struct drm_gpuva *va, u64 addr, u64 range,
-				  struct drm_gem_object *obj, u64 offset)
-{
-	va->va.addr = addr;
-	va->va.range = range;
-	va->gem.obj = obj;
-	va->gem.offset = offset;
-}
-
 /**
  * drm_gpuva_invalidate() - sets whether the backing GEM of this &drm_gpuva is
  * invalidated
@@ -1089,8 +1080,10 @@ void drm_gpuva_ops_free(struct drm_gpuvm *gpuvm,
 static inline void drm_gpuva_init_from_op(struct drm_gpuva *va,
 					  struct drm_gpuva_op_map *op)
 {
-	drm_gpuva_init(va, op->va.addr, op->va.range,
-		       op->gem.obj, op->gem.offset);
+	va->va.addr = op->va.addr;
+	va->va.range = op->va.range;
+	va->gem.obj = op->gem.obj;
+	va->gem.offset = op->gem.offset;
 }
 
 /**

From baf1638c095686ad970aecee4ca9446c1de18dad Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Tue, 19 Aug 2025 21:50:57 +0530
Subject: [PATCH 0546/1455] drm/gpuvm: Introduce drm_gpuvm_madvise_ops_create

This ops is used to iterate over GPUVA's in the user-provided range
and split the existing sparse VMA's if the start or end of the input
range lies within it. The operations can create up to 2 REMAPS and 2 MAPs.

The primary use case is for drivers to assign attributes to GPU VAs in
the specified range without performing unmaps or merging mappings,
supporting fine-grained control over sparse va's.

Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Boris Brezillon <bbrezillon@kernel.org>
Cc: <dri-devel@lists.freedesktop.org>
Signed-off-by: Himal Prasad Ghimiray<himal.prasad.ghimiray@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-4-himal.prasad.ghimiray@intel.com
---
 drivers/gpu/drm/drm_gpuvm.c | 225 ++++++++++++++++++++++++++++++------
 include/drm/drm_gpuvm.h     |   3 +
 2 files changed, 191 insertions(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpuvm.c b/drivers/gpu/drm/drm_gpuvm.c
index 6c18cec70f11..d6bea8a4fffd 100644
--- a/drivers/gpu/drm/drm_gpuvm.c
+++ b/drivers/gpu/drm/drm_gpuvm.c
@@ -420,6 +420,71 @@
  *	 new: |-----------|-----| (b.bo_offset=m,a.bo_offset=n+2)
  */
 
+/**
+ * DOC: Madvise Logic - Splitting and Traversal
+ *
+ * This logic handles GPU VA range updates by generating remap and map operations
+ * without performing unmaps or merging existing mappings.
+ *
+ * 1) The requested range lies entirely within a single drm_gpuva. The logic splits
+ * the existing mapping at the start and end boundaries and inserts a new map.
+ *
+ * ::
+ *              a      start    end     b
+ *         pre: |-----------------------|
+ *                     drm_gpuva1
+ *
+ *              a      start    end     b
+ *         new: |-----|=========|-------|
+ *               remap   map      remap
+ *
+ * one REMAP and one MAP : Same behaviour as SPLIT and MERGE
+ *
+ * 2) The requested range spans multiple drm_gpuva regions. The logic traverses
+ * across boundaries, remapping the start and end segments, and inserting two
+ * map operations to cover the full range.
+ *
+ * ::           a       start      b              c        end       d
+ *         pre: |------------------|--------------|------------------|
+ *                    drm_gpuva1      drm_gpuva2         drm_gpuva3
+ *
+ *              a       start      b              c        end       d
+ *         new: |-------|==========|--------------|========|---------|
+ *                remap1   map1       drm_gpuva2    map2     remap2
+ *
+ * two REMAPS and two MAPS
+ *
+ * 3) Either start or end lies within a drm_gpuva. A single remap and map operation
+ * are generated to update the affected portion.
+ *
+ *
+ * ::           a/start            b              c        end       d
+ *         pre: |------------------|--------------|------------------|
+ *                    drm_gpuva1      drm_gpuva2         drm_gpuva3
+ *
+ *              a/start            b              c        end       d
+ *         new: |------------------|--------------|========|---------|
+ *                drm_gpuva1         drm_gpuva2     map1     remap1
+ *
+ * ::           a       start      b              c/end              d
+ *         pre: |------------------|--------------|------------------|
+ *                    drm_gpuva1      drm_gpuva2         drm_gpuva3
+ *
+ *              a       start      b              c/end              d
+ *         new: |-------|==========|--------------|------------------|
+ *                remap1   map1       drm_gpuva2        drm_gpuva3
+ *
+ * one REMAP and one MAP
+ *
+ * 4) Both start and end align with existing drm_gpuva boundaries. No operations
+ * are needed as the range is already covered.
+ *
+ * 5) No existing drm_gpuvas. No operations.
+ *
+ * Unlike drm_gpuvm_sm_map_ops_create, this logic avoids unmaps and merging,
+ * focusing solely on remap and map operations for efficient traversal and update.
+ */
+
 /**
  * DOC: Locking
  *
@@ -2063,6 +2128,9 @@ op_map_cb(const struct drm_gpuvm_ops *fn, void *priv,
 {
 	struct drm_gpuva_op op = {};
 
+	if (!req)
+		return 0;
+
 	op.op = DRM_GPUVA_OP_MAP;
 	op.map.va.addr = req->map.va.addr;
 	op.map.va.range = req->map.va.range;
@@ -2092,10 +2160,13 @@ op_remap_cb(const struct drm_gpuvm_ops *fn, void *priv,
 
 static int
 op_unmap_cb(const struct drm_gpuvm_ops *fn, void *priv,
-	    struct drm_gpuva *va, bool merge)
+	    struct drm_gpuva *va, bool merge, bool madvise)
 {
 	struct drm_gpuva_op op = {};
 
+	if (madvise)
+		return 0;
+
 	op.op = DRM_GPUVA_OP_UNMAP;
 	op.unmap.va = va;
 	op.unmap.keep = merge;
@@ -2106,11 +2177,12 @@ op_unmap_cb(const struct drm_gpuvm_ops *fn, void *priv,
 static int
 __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 		   const struct drm_gpuvm_ops *ops, void *priv,
-		   const struct drm_gpuvm_map_req *req)
+		   const struct drm_gpuvm_map_req *req,
+		   bool madvise)
 {
 	struct drm_gem_object *req_obj = req->map.gem.obj;
+	const struct drm_gpuvm_map_req *op_map = madvise ? NULL : req;
 	struct drm_gpuva *va, *next;
-
 	u64 req_offset = req->map.gem.offset;
 	u64 req_range = req->map.va.range;
 	u64 req_addr = req->map.va.addr;
@@ -2128,19 +2200,22 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 		u64 end = addr + range;
 		bool merge = !!va->gem.obj;
 
+		if (madvise && obj)
+			continue;
+
 		if (addr == req_addr) {
 			merge &= obj == req_obj &&
 				 offset == req_offset;
 
 			if (end == req_end) {
-				ret = op_unmap_cb(ops, priv, va, merge);
+				ret = op_unmap_cb(ops, priv, va, merge, madvise);
 				if (ret)
 					return ret;
 				break;
 			}
 
 			if (end < req_end) {
-				ret = op_unmap_cb(ops, priv, va, merge);
+				ret = op_unmap_cb(ops, priv, va, merge, madvise);
 				if (ret)
 					return ret;
 				continue;
@@ -2161,6 +2236,9 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 				ret = op_remap_cb(ops, priv, NULL, &n, &u);
 				if (ret)
 					return ret;
+
+				if (madvise)
+					op_map = req;
 				break;
 			}
 		} else if (addr < req_addr) {
@@ -2181,6 +2259,9 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 				ret = op_remap_cb(ops, priv, &p, NULL, &u);
 				if (ret)
 					return ret;
+
+				if (madvise)
+					op_map = req;
 				break;
 			}
 
@@ -2188,6 +2269,18 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 				ret = op_remap_cb(ops, priv, &p, NULL, &u);
 				if (ret)
 					return ret;
+
+				if (madvise) {
+					struct drm_gpuvm_map_req map_req = {
+						.map.va.addr =  req_addr,
+						.map.va.range = end - req_addr,
+					};
+
+					ret = op_map_cb(ops, priv, &map_req);
+					if (ret)
+						return ret;
+				}
+
 				continue;
 			}
 
@@ -2203,6 +2296,9 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 				ret = op_remap_cb(ops, priv, &p, &n, &u);
 				if (ret)
 					return ret;
+
+				if (madvise)
+					op_map = req;
 				break;
 			}
 		} else if (addr > req_addr) {
@@ -2211,16 +2307,18 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 					   (addr - req_addr);
 
 			if (end == req_end) {
-				ret = op_unmap_cb(ops, priv, va, merge);
+				ret = op_unmap_cb(ops, priv, va, merge, madvise);
 				if (ret)
 					return ret;
+
 				break;
 			}
 
 			if (end < req_end) {
-				ret = op_unmap_cb(ops, priv, va, merge);
+				ret = op_unmap_cb(ops, priv, va, merge, madvise);
 				if (ret)
 					return ret;
+
 				continue;
 			}
 
@@ -2239,12 +2337,20 @@ __drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm,
 				ret = op_remap_cb(ops, priv, NULL, &n, &u);
 				if (ret)
 					return ret;
+
+				if (madvise) {
+					struct drm_gpuvm_map_req map_req = {
+						.map.va.addr =  addr,
+						.map.va.range = req_end - addr,
+					};
+
+					return op_map_cb(ops, priv, &map_req);
+				}
 				break;
 			}
 		}
 	}
-
-	return op_map_cb(ops, priv, req);
+	return op_map_cb(ops, priv, op_map);
 }
 
 static int
@@ -2296,7 +2402,7 @@ __drm_gpuvm_sm_unmap(struct drm_gpuvm *gpuvm,
 			if (ret)
 				return ret;
 		} else {
-			ret = op_unmap_cb(ops, priv, va, false);
+			ret = op_unmap_cb(ops, priv, va, false, false);
 			if (ret)
 				return ret;
 		}
@@ -2345,7 +2451,7 @@ drm_gpuvm_sm_map(struct drm_gpuvm *gpuvm, void *priv,
 		       ops->sm_step_unmap)))
 		return -EINVAL;
 
-	return __drm_gpuvm_sm_map(gpuvm, ops, priv, req);
+	return __drm_gpuvm_sm_map(gpuvm, ops, priv, req, false);
 }
 EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map);
 
@@ -2483,7 +2589,7 @@ drm_gpuvm_sm_map_exec_lock(struct drm_gpuvm *gpuvm,
 			return ret;
 	}
 
-	return __drm_gpuvm_sm_map(gpuvm, &lock_ops, exec, req);
+	return __drm_gpuvm_sm_map(gpuvm, &lock_ops, exec, req, false);
 
 }
 EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map_exec_lock);
@@ -2602,6 +2708,38 @@ static const struct drm_gpuvm_ops gpuvm_list_ops = {
 	.sm_step_unmap = drm_gpuva_sm_step,
 };
 
+static struct drm_gpuva_ops *
+__drm_gpuvm_sm_map_ops_create(struct drm_gpuvm *gpuvm,
+			      const struct drm_gpuvm_map_req *req,
+			      bool madvise)
+{
+	struct drm_gpuva_ops *ops;
+	struct {
+		struct drm_gpuvm *vm;
+		struct drm_gpuva_ops *ops;
+	} args;
+	int ret;
+
+	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
+	if (unlikely(!ops))
+		return ERR_PTR(-ENOMEM);
+
+	INIT_LIST_HEAD(&ops->list);
+
+	args.vm = gpuvm;
+	args.ops = ops;
+
+	ret = __drm_gpuvm_sm_map(gpuvm, &gpuvm_list_ops, &args, req, madvise);
+	if (ret)
+		goto err_free_ops;
+
+	return ops;
+
+err_free_ops:
+	drm_gpuva_ops_free(gpuvm, ops);
+	return ERR_PTR(ret);
+}
+
 /**
  * drm_gpuvm_sm_map_ops_create() - creates the &drm_gpuva_ops to split and merge
  * @gpuvm: the &drm_gpuvm representing the GPU VA space
@@ -2635,34 +2773,47 @@ struct drm_gpuva_ops *
 drm_gpuvm_sm_map_ops_create(struct drm_gpuvm *gpuvm,
 			    const struct drm_gpuvm_map_req *req)
 {
-	struct drm_gpuva_ops *ops;
-	struct {
-		struct drm_gpuvm *vm;
-		struct drm_gpuva_ops *ops;
-	} args;
-	int ret;
-
-	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
-	if (unlikely(!ops))
-		return ERR_PTR(-ENOMEM);
-
-	INIT_LIST_HEAD(&ops->list);
-
-	args.vm = gpuvm;
-	args.ops = ops;
-
-	ret = __drm_gpuvm_sm_map(gpuvm, &gpuvm_list_ops, &args, req);
-	if (ret)
-		goto err_free_ops;
-
-	return ops;
-
-err_free_ops:
-	drm_gpuva_ops_free(gpuvm, ops);
-	return ERR_PTR(ret);
+	return __drm_gpuvm_sm_map_ops_create(gpuvm, req, false);
 }
 EXPORT_SYMBOL_GPL(drm_gpuvm_sm_map_ops_create);
 
+/**
+ * drm_gpuvm_madvise_ops_create() - creates the &drm_gpuva_ops to split
+ * @gpuvm: the &drm_gpuvm representing the GPU VA space
+ * @req: map request arguments
+ *
+ * This function creates a list of operations to perform splitting
+ * of existent mapping(s) at start or end, based on the request map.
+ *
+ * The list can be iterated with &drm_gpuva_for_each_op and must be processed
+ * in the given order. It can contain map and remap operations, but it
+ * also can be empty if no operation is required, e.g. if the requested mapping
+ * already exists is the exact same way.
+ *
+ * There will be no unmap operations, a maximum of two remap operations and two
+ * map operations. The two map operations correspond to: one from start to the
+ * end of drm_gpuvaX, and another from the start of drm_gpuvaY to end.
+ *
+ * Note that before calling this function again with another mapping request it
+ * is necessary to update the &drm_gpuvm's view of the GPU VA space. The
+ * previously obtained operations must be either processed or abandoned. To
+ * update the &drm_gpuvm's view of the GPU VA space drm_gpuva_insert(),
+ * drm_gpuva_destroy_locked() and/or drm_gpuva_destroy_unlocked() should be
+ * used.
+ *
+ * After the caller finished processing the returned &drm_gpuva_ops, they must
+ * be freed with &drm_gpuva_ops_free.
+ *
+ * Returns: a pointer to the &drm_gpuva_ops on success, an ERR_PTR on failure
+ */
+struct drm_gpuva_ops *
+drm_gpuvm_madvise_ops_create(struct drm_gpuvm *gpuvm,
+			     const struct drm_gpuvm_map_req *req)
+{
+	return __drm_gpuvm_sm_map_ops_create(gpuvm, req, true);
+}
+EXPORT_SYMBOL_GPL(drm_gpuvm_madvise_ops_create);
+
 /**
  * drm_gpuvm_sm_unmap_ops_create() - creates the &drm_gpuva_ops to split on
  * unmap
diff --git a/include/drm/drm_gpuvm.h b/include/drm/drm_gpuvm.h
index 05347ac6cc73..4a22b9d848f7 100644
--- a/include/drm/drm_gpuvm.h
+++ b/include/drm/drm_gpuvm.h
@@ -1062,6 +1062,9 @@ struct drm_gpuvm_map_req {
 struct drm_gpuva_ops *
 drm_gpuvm_sm_map_ops_create(struct drm_gpuvm *gpuvm,
 			    const struct drm_gpuvm_map_req *req);
+struct drm_gpuva_ops *
+drm_gpuvm_madvise_ops_create(struct drm_gpuvm *gpuvm,
+			     const struct drm_gpuvm_map_req *req);
 
 struct drm_gpuva_ops *
 drm_gpuvm_sm_unmap_ops_create(struct drm_gpuvm *gpuvm,

From dab74906423c5a0e41dfd4cefc3758d351ccc51e Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Tue, 19 Aug 2025 21:50:58 +0530
Subject: [PATCH 0547/1455] drm/gpusvm: Make drm_gpusvm_for_each_* macros
 public

The drm_gpusvm_for_each_notifier, drm_gpusvm_for_each_notifier_safe and
drm_gpusvm_for_each_range_safe macros are useful for locating notifiers
and ranges within a user-specified range. By making these macros public,
we enable broader access and utility for developers who need to leverage
them in their implementations.

v2 (Matthew Brost)
- drop inline __drm_gpusvm_range_find
- /s/notifier_iter_first/drm_gpusvm_notifier_find

Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250819162058.2777306-5-himal.prasad.ghimiray@intel.com
---
 drivers/gpu/drm/drm_gpusvm.c | 124 ++++++++---------------------------
 include/drm/drm_gpusvm.h     |  70 ++++++++++++++++++++
 2 files changed, 96 insertions(+), 98 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index 5bb4c77db2c3..647b49ff2da5 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -270,6 +270,29 @@ npages_in_range(unsigned long start, unsigned long end)
 	return (end - start) >> PAGE_SHIFT;
 }
 
+/**
+ * drm_gpusvm_notifier_find() - Find GPU SVM notifier from GPU SVM
+ * @gpusvm: Pointer to the GPU SVM structure.
+ * @start: Start address of the notifier
+ * @end: End address of the notifier
+ *
+ * Return: A pointer to the drm_gpusvm_notifier if found or NULL
+ */
+struct drm_gpusvm_notifier *
+drm_gpusvm_notifier_find(struct drm_gpusvm *gpusvm, unsigned long start,
+			 unsigned long end)
+{
+	struct interval_tree_node *itree;
+
+	itree = interval_tree_iter_first(&gpusvm->root, start, end - 1);
+
+	if (itree)
+		return container_of(itree, struct drm_gpusvm_notifier, itree);
+	else
+		return NULL;
+}
+EXPORT_SYMBOL_GPL(drm_gpusvm_notifier_find);
+
 /**
  * drm_gpusvm_range_find() - Find GPU SVM range from GPU SVM notifier
  * @notifier: Pointer to the GPU SVM notifier structure.
@@ -293,86 +316,6 @@ drm_gpusvm_range_find(struct drm_gpusvm_notifier *notifier, unsigned long start,
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_find);
 
-/**
- * drm_gpusvm_for_each_range_safe() - Safely iterate over GPU SVM ranges in a notifier
- * @range__: Iterator variable for the ranges
- * @next__: Iterator variable for the ranges temporay storage
- * @notifier__: Pointer to the GPU SVM notifier
- * @start__: Start address of the range
- * @end__: End address of the range
- *
- * This macro is used to iterate over GPU SVM ranges in a notifier while
- * removing ranges from it.
- */
-#define drm_gpusvm_for_each_range_safe(range__, next__, notifier__, start__, end__)	\
-	for ((range__) = drm_gpusvm_range_find((notifier__), (start__), (end__)),	\
-	     (next__) = __drm_gpusvm_range_next(range__);				\
-	     (range__) && (drm_gpusvm_range_start(range__) < (end__));			\
-	     (range__) = (next__), (next__) = __drm_gpusvm_range_next(range__))
-
-/**
- * __drm_gpusvm_notifier_next() - get the next drm_gpusvm_notifier in the list
- * @notifier: a pointer to the current drm_gpusvm_notifier
- *
- * Return: A pointer to the next drm_gpusvm_notifier if available, or NULL if
- *         the current notifier is the last one or if the input notifier is
- *         NULL.
- */
-static struct drm_gpusvm_notifier *
-__drm_gpusvm_notifier_next(struct drm_gpusvm_notifier *notifier)
-{
-	if (notifier && !list_is_last(&notifier->entry,
-				      &notifier->gpusvm->notifier_list))
-		return list_next_entry(notifier, entry);
-
-	return NULL;
-}
-
-static struct drm_gpusvm_notifier *
-notifier_iter_first(struct rb_root_cached *root, unsigned long start,
-		    unsigned long last)
-{
-	struct interval_tree_node *itree;
-
-	itree = interval_tree_iter_first(root, start, last);
-
-	if (itree)
-		return container_of(itree, struct drm_gpusvm_notifier, itree);
-	else
-		return NULL;
-}
-
-/**
- * drm_gpusvm_for_each_notifier() - Iterate over GPU SVM notifiers in a gpusvm
- * @notifier__: Iterator variable for the notifiers
- * @notifier__: Pointer to the GPU SVM notifier
- * @start__: Start address of the notifier
- * @end__: End address of the notifier
- *
- * This macro is used to iterate over GPU SVM notifiers in a gpusvm.
- */
-#define drm_gpusvm_for_each_notifier(notifier__, gpusvm__, start__, end__)		\
-	for ((notifier__) = notifier_iter_first(&(gpusvm__)->root, (start__), (end__) - 1);	\
-	     (notifier__) && (drm_gpusvm_notifier_start(notifier__) < (end__));		\
-	     (notifier__) = __drm_gpusvm_notifier_next(notifier__))
-
-/**
- * drm_gpusvm_for_each_notifier_safe() - Safely iterate over GPU SVM notifiers in a gpusvm
- * @notifier__: Iterator variable for the notifiers
- * @next__: Iterator variable for the notifiers temporay storage
- * @notifier__: Pointer to the GPU SVM notifier
- * @start__: Start address of the notifier
- * @end__: End address of the notifier
- *
- * This macro is used to iterate over GPU SVM notifiers in a gpusvm while
- * removing notifiers from it.
- */
-#define drm_gpusvm_for_each_notifier_safe(notifier__, next__, gpusvm__, start__, end__)	\
-	for ((notifier__) = notifier_iter_first(&(gpusvm__)->root, (start__), (end__) - 1),	\
-	     (next__) = __drm_gpusvm_notifier_next(notifier__);				\
-	     (notifier__) && (drm_gpusvm_notifier_start(notifier__) < (end__));		\
-	     (notifier__) = (next__), (next__) = __drm_gpusvm_notifier_next(notifier__))
-
 /**
  * drm_gpusvm_notifier_invalidate() - Invalidate a GPU SVM notifier.
  * @mni: Pointer to the mmu_interval_notifier structure.
@@ -472,22 +415,6 @@ int drm_gpusvm_init(struct drm_gpusvm *gpusvm,
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_init);
 
-/**
- * drm_gpusvm_notifier_find() - Find GPU SVM notifier
- * @gpusvm: Pointer to the GPU SVM structure
- * @fault_addr: Fault address
- *
- * This function finds the GPU SVM notifier associated with the fault address.
- *
- * Return: Pointer to the GPU SVM notifier on success, NULL otherwise.
- */
-static struct drm_gpusvm_notifier *
-drm_gpusvm_notifier_find(struct drm_gpusvm *gpusvm,
-			 unsigned long fault_addr)
-{
-	return notifier_iter_first(&gpusvm->root, fault_addr, fault_addr + 1);
-}
-
 /**
  * to_drm_gpusvm_notifier() - retrieve the container struct for a given rbtree node
  * @node: a pointer to the rbtree node embedded within a drm_gpusvm_notifier struct
@@ -943,7 +870,7 @@ drm_gpusvm_range_find_or_insert(struct drm_gpusvm *gpusvm,
 	if (!mmget_not_zero(mm))
 		return ERR_PTR(-EFAULT);
 
-	notifier = drm_gpusvm_notifier_find(gpusvm, fault_addr);
+	notifier = drm_gpusvm_notifier_find(gpusvm, fault_addr, fault_addr + 1);
 	if (!notifier) {
 		notifier = drm_gpusvm_notifier_alloc(gpusvm, fault_addr);
 		if (IS_ERR(notifier)) {
@@ -1107,7 +1034,8 @@ void drm_gpusvm_range_remove(struct drm_gpusvm *gpusvm,
 	drm_gpusvm_driver_lock_held(gpusvm);
 
 	notifier = drm_gpusvm_notifier_find(gpusvm,
-					    drm_gpusvm_range_start(range));
+					    drm_gpusvm_range_start(range),
+					    drm_gpusvm_range_start(range) + 1);
 	if (WARN_ON_ONCE(!notifier))
 		return;
 
diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h
index 4aedc5423aff..142fc2af1716 100644
--- a/include/drm/drm_gpusvm.h
+++ b/include/drm/drm_gpusvm.h
@@ -282,6 +282,10 @@ void drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 bool drm_gpusvm_has_mapping(struct drm_gpusvm *gpusvm, unsigned long start,
 			    unsigned long end);
 
+struct drm_gpusvm_notifier *
+drm_gpusvm_notifier_find(struct drm_gpusvm *gpusvm, unsigned long start,
+			 unsigned long end);
+
 struct drm_gpusvm_range *
 drm_gpusvm_range_find(struct drm_gpusvm_notifier *notifier, unsigned long start,
 		      unsigned long end);
@@ -434,4 +438,70 @@ __drm_gpusvm_range_next(struct drm_gpusvm_range *range)
 	     (range__) && (drm_gpusvm_range_start(range__) < (end__));	\
 	     (range__) = __drm_gpusvm_range_next(range__))
 
+/**
+ * drm_gpusvm_for_each_range_safe() - Safely iterate over GPU SVM ranges in a notifier
+ * @range__: Iterator variable for the ranges
+ * @next__: Iterator variable for the ranges temporay storage
+ * @notifier__: Pointer to the GPU SVM notifier
+ * @start__: Start address of the range
+ * @end__: End address of the range
+ *
+ * This macro is used to iterate over GPU SVM ranges in a notifier while
+ * removing ranges from it.
+ */
+#define drm_gpusvm_for_each_range_safe(range__, next__, notifier__, start__, end__)	\
+	for ((range__) = drm_gpusvm_range_find((notifier__), (start__), (end__)),	\
+	     (next__) = __drm_gpusvm_range_next(range__);				\
+	     (range__) && (drm_gpusvm_range_start(range__) < (end__));			\
+	     (range__) = (next__), (next__) = __drm_gpusvm_range_next(range__))
+
+/**
+ * __drm_gpusvm_notifier_next() - get the next drm_gpusvm_notifier in the list
+ * @notifier: a pointer to the current drm_gpusvm_notifier
+ *
+ * Return: A pointer to the next drm_gpusvm_notifier if available, or NULL if
+ *         the current notifier is the last one or if the input notifier is
+ *         NULL.
+ */
+static inline struct drm_gpusvm_notifier *
+__drm_gpusvm_notifier_next(struct drm_gpusvm_notifier *notifier)
+{
+	if (notifier && !list_is_last(&notifier->entry,
+				      &notifier->gpusvm->notifier_list))
+		return list_next_entry(notifier, entry);
+
+	return NULL;
+}
+
+/**
+ * drm_gpusvm_for_each_notifier() - Iterate over GPU SVM notifiers in a gpusvm
+ * @notifier__: Iterator variable for the notifiers
+ * @gpusvm__: Pointer to the GPU SVM notifier
+ * @start__: Start address of the notifier
+ * @end__: End address of the notifier
+ *
+ * This macro is used to iterate over GPU SVM notifiers in a gpusvm.
+ */
+#define drm_gpusvm_for_each_notifier(notifier__, gpusvm__, start__, end__)		\
+	for ((notifier__) = drm_gpusvm_notifier_find((gpusvm__), (start__), (end__));	\
+	     (notifier__) && (drm_gpusvm_notifier_start(notifier__) < (end__));		\
+	     (notifier__) = __drm_gpusvm_notifier_next(notifier__))
+
+/**
+ * drm_gpusvm_for_each_notifier_safe() - Safely iterate over GPU SVM notifiers in a gpusvm
+ * @notifier__: Iterator variable for the notifiers
+ * @next__: Iterator variable for the notifiers temporay storage
+ * @gpusvm__: Pointer to the GPU SVM notifier
+ * @start__: Start address of the notifier
+ * @end__: End address of the notifier
+ *
+ * This macro is used to iterate over GPU SVM notifiers in a gpusvm while
+ * removing notifiers from it.
+ */
+#define drm_gpusvm_for_each_notifier_safe(notifier__, next__, gpusvm__, start__, end__)	\
+	for ((notifier__) = drm_gpusvm_notifier_find((gpusvm__), (start__), (end__)),	\
+	     (next__) = __drm_gpusvm_notifier_next(notifier__);				\
+	     (notifier__) && (drm_gpusvm_notifier_start(notifier__) < (end__));		\
+	     (notifier__) = (next__), (next__) = __drm_gpusvm_notifier_next(notifier__))
+
 #endif /* __DRM_GPUSVM_H__ */

From 9b1795e9b0ae6d8b2c50987048170f77d0dc0300 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Mon, 19 May 2025 10:52:23 +0300
Subject: [PATCH 0548/1455] drm/i915/psr: Underrun on idle PSR wa only when
 pkgc latency > delayed vblank
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Underrun on idle PSR workaround (Wa_16025596647) is supposed to be
applied only when pkg c latency > delayed vblank. Currently we are
applying it always when other criterias are met.

Fix this by adding new boolean flag which is supposed to be set when
calculating watermark levels and pkgc latency > delayed vblank is
detected. currently this scenario is blocked but might be added
later. Due to this add also TODO comment into
skl_max_wm_level_for_vblank.

Bspec: 74151
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Link: https://lore.kernel.org/r/20250519075223.443266-1-jouni.hogander@intel.com
---
 .../gpu/drm/i915/display/intel_display_types.h   |  2 ++
 drivers/gpu/drm/i915/display/intel_psr.c         | 16 ++++++++++------
 drivers/gpu/drm/i915/display/skl_watermark.c     |  5 +++++
 3 files changed, 17 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 0d945d1fedd6..fd9d2527889b 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -1121,6 +1121,7 @@ struct intel_crtc_state {
 	bool req_psr2_sdp_prior_scanline;
 	bool has_panel_replay;
 	bool wm_level_disabled;
+	bool pkg_c_latency_used;
 	u32 dc3co_exitline;
 	u16 su_y_granularity;
 	u8 active_non_psr_pipes;
@@ -1683,6 +1684,7 @@ struct intel_psr {
 	u8 entry_setup_frames;
 
 	bool link_ok;
+	bool pkg_c_latency_used;
 
 	u8 active_non_psr_pipes;
 };
diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 8bea2b8188a7..609df53f1ef1 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -915,7 +915,7 @@ static void hsw_activate_psr1(struct intel_dp *intel_dp)
 	/* Wa_16025596647 */
 	if ((DISPLAY_VER(display) == 20 ||
 	     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
-	    is_dc5_dc6_blocked(intel_dp))
+	    is_dc5_dc6_blocked(intel_dp) && intel_dp->psr.pkg_c_latency_used)
 		intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display,
 								       intel_dp->psr.pipe,
 								       true);
@@ -1005,7 +1005,7 @@ static void hsw_activate_psr2(struct intel_dp *intel_dp)
 	/* Wa_16025596647 */
 	if ((DISPLAY_VER(display) == 20 ||
 	     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
-	    is_dc5_dc6_blocked(intel_dp))
+	    is_dc5_dc6_blocked(intel_dp) && intel_dp->psr.pkg_c_latency_used)
 		idle_frames = 0;
 	else
 		idle_frames = psr_compute_idle_frames(intel_dp);
@@ -2006,6 +2006,7 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp,
 	intel_dp->psr.req_psr2_sdp_prior_scanline =
 		crtc_state->req_psr2_sdp_prior_scanline;
 	intel_dp->psr.active_non_psr_pipes = crtc_state->active_non_psr_pipes;
+	intel_dp->psr.pkg_c_latency_used = crtc_state->pkg_c_latency_used;
 
 	if (!psr_interrupt_error_check(intel_dp))
 		return;
@@ -2186,6 +2187,7 @@ static void intel_psr_disable_locked(struct intel_dp *intel_dp)
 	intel_dp->psr.su_region_et_enabled = false;
 	intel_dp->psr.psr2_sel_fetch_cff_enabled = false;
 	intel_dp->psr.active_non_psr_pipes = 0;
+	intel_dp->psr.pkg_c_latency_used = 0;
 }
 
 /**
@@ -3702,7 +3704,7 @@ static void intel_psr_apply_underrun_on_idle_wa_locked(struct intel_dp *intel_dp
 	struct intel_display *display = to_intel_display(intel_dp);
 	bool dc5_dc6_blocked;
 
-	if (!intel_dp->psr.active)
+	if (!intel_dp->psr.active || !intel_dp->psr.pkg_c_latency_used)
 		return;
 
 	dc5_dc6_blocked = is_dc5_dc6_blocked(intel_dp);
@@ -3727,7 +3729,8 @@ static void psr_dc5_dc6_wa_work(struct work_struct *work)
 
 		mutex_lock(&intel_dp->psr.lock);
 
-		if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled)
+		if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled &&
+		    !intel_dp->psr.pkg_c_latency_used)
 			intel_psr_apply_underrun_on_idle_wa_locked(intel_dp);
 
 		mutex_unlock(&intel_dp->psr.lock);
@@ -3805,7 +3808,8 @@ void intel_psr_notify_pipe_change(struct intel_atomic_state *state,
 			goto unlock;
 
 		if ((enable && intel_dp->psr.active_non_psr_pipes) ||
-		    (!enable && !intel_dp->psr.active_non_psr_pipes)) {
+		    (!enable && !intel_dp->psr.active_non_psr_pipes) ||
+		    !intel_dp->psr.pkg_c_latency_used) {
 			intel_dp->psr.active_non_psr_pipes = active_non_psr_pipes;
 			goto unlock;
 		}
@@ -3840,7 +3844,7 @@ void intel_psr_notify_vblank_enable_disable(struct intel_display *display,
 			break;
 		}
 
-		if (intel_dp->psr.enabled)
+		if (intel_dp->psr.enabled && intel_dp->psr.pkg_c_latency_used)
 			intel_psr_apply_underrun_on_idle_wa_locked(intel_dp);
 
 		mutex_unlock(&intel_dp->psr.lock);
diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index def5150231a4..33885d619a97 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -2272,6 +2272,11 @@ static int skl_max_wm0_lines(const struct intel_crtc_state *crtc_state)
 	return wm0_lines;
 }
 
+/*
+ * TODO: In case we use PKG_C_LATENCY to allow C-states when the delayed vblank
+ * size is too small for the package C exit latency we need to notify PSR about
+ * the scenario to apply Wa_16025596647.
+ */
 static int skl_max_wm_level_for_vblank(struct intel_crtc_state *crtc_state,
 				       int wm0_lines)
 {

From 38580d1e7f18230b10e8ba6d9a44a9fcc532f47f Mon Sep 17 00:00:00 2001
From: Athul Raj Kollareth <krathul3152@gmail.com>
Date: Wed, 13 Aug 2025 11:51:05 +0530
Subject: [PATCH 0549/1455] drm/virtio: clean up minor codestyle issues

Fix codestyle warnings and errors generated by CHECKPATCH in virtio
source files.

Signed-off-by: Athul Raj Kollareth <krathul3152@gmail.com>
Signed-off-by: Dmitry Osipenko <dmitry.osipenko@collabora.com>
Link: https://lore.kernel.org/r/20250813062109.5326-1-krathul3152@gmail.com
---
 drivers/gpu/drm/virtio/virtgpu_display.c |  3 +--
 drivers/gpu/drm/virtio/virtgpu_kms.c     | 20 ++++++++++----------
 drivers/gpu/drm/virtio/virtgpu_object.c  |  4 ++--
 drivers/gpu/drm/virtio/virtgpu_plane.c   |  2 +-
 drivers/gpu/drm/virtio/virtgpu_vq.c      |  2 ++
 5 files changed, 16 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/virtio/virtgpu_display.c b/drivers/gpu/drm/virtio/virtgpu_display.c
index e5805ca646c7..c3315935d8bc 100644
--- a/drivers/gpu/drm/virtio/virtgpu_display.c
+++ b/drivers/gpu/drm/virtio/virtgpu_display.c
@@ -131,9 +131,8 @@ static void virtio_gpu_crtc_atomic_flush(struct drm_crtc *crtc,
 	 * in the plane update callback, and here we just check
 	 * whenever we must force the modeset.
 	 */
-	if (drm_atomic_crtc_needs_modeset(crtc_state)) {
+	if (drm_atomic_crtc_needs_modeset(crtc_state))
 		output->needs_modeset = true;
-	}
 }
 
 static const struct drm_crtc_helper_funcs virtio_gpu_crtc_helper_funcs = {
diff --git a/drivers/gpu/drm/virtio/virtgpu_kms.c b/drivers/gpu/drm/virtio/virtgpu_kms.c
index 7dfb2006c561..1c15cbf326b7 100644
--- a/drivers/gpu/drm/virtio/virtgpu_kms.c
+++ b/drivers/gpu/drm/virtio/virtgpu_kms.c
@@ -162,18 +162,18 @@ int virtio_gpu_init(struct virtio_device *vdev, struct drm_device *dev)
 	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_VIRGL))
 		vgdev->has_virgl_3d = true;
 #endif
-	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_EDID)) {
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_EDID))
 		vgdev->has_edid = true;
-	}
-	if (virtio_has_feature(vgdev->vdev, VIRTIO_RING_F_INDIRECT_DESC)) {
+
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_RING_F_INDIRECT_DESC))
 		vgdev->has_indirect = true;
-	}
-	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_RESOURCE_UUID)) {
+
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_RESOURCE_UUID))
 		vgdev->has_resource_assign_uuid = true;
-	}
-	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_RESOURCE_BLOB)) {
+
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_RESOURCE_BLOB))
 		vgdev->has_resource_blob = true;
-	}
+
 	if (virtio_get_shm_region(vgdev->vdev, &vgdev->host_visible_region,
 				  VIRTIO_GPU_SHM_ID_HOST_VISIBLE)) {
 		if (!devm_request_mem_region(&vgdev->vdev->dev,
@@ -193,9 +193,9 @@ int virtio_gpu_init(struct virtio_device *vdev, struct drm_device *dev)
 			    (unsigned long)vgdev->host_visible_region.addr,
 			    (unsigned long)vgdev->host_visible_region.len);
 	}
-	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_CONTEXT_INIT)) {
+
+	if (virtio_has_feature(vgdev->vdev, VIRTIO_GPU_F_CONTEXT_INIT))
 		vgdev->has_context_init = true;
-	}
 
 	DRM_INFO("features: %cvirgl %cedid %cresource_blob %chost_visible",
 		 vgdev->has_virgl_3d    ? '+' : '-',
diff --git a/drivers/gpu/drm/virtio/virtgpu_object.c b/drivers/gpu/drm/virtio/virtgpu_object.c
index 5517cff8715c..e6363c887500 100644
--- a/drivers/gpu/drm/virtio/virtgpu_object.c
+++ b/drivers/gpu/drm/virtio/virtgpu_object.c
@@ -47,6 +47,7 @@ int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev, uint32_t *resid)
 		*resid = handle + 1;
 	} else {
 		int handle = ida_alloc(&vgdev->resource_ida, GFP_KERNEL);
+
 		if (handle < 0)
 			return handle;
 		*resid = handle + 1;
@@ -56,9 +57,8 @@ int virtio_gpu_resource_id_get(struct virtio_gpu_device *vgdev, uint32_t *resid)
 
 static void virtio_gpu_resource_id_put(struct virtio_gpu_device *vgdev, uint32_t id)
 {
-	if (!virtio_gpu_virglrenderer_workaround) {
+	if (!virtio_gpu_virglrenderer_workaround)
 		ida_free(&vgdev->resource_ida, id - 1);
-	}
 }
 
 void virtio_gpu_cleanup_object(struct virtio_gpu_object *bo)
diff --git a/drivers/gpu/drm/virtio/virtgpu_plane.c b/drivers/gpu/drm/virtio/virtgpu_plane.c
index 698ea7adb951..29e4b458ae57 100644
--- a/drivers/gpu/drm/virtio/virtgpu_plane.c
+++ b/drivers/gpu/drm/virtio/virtgpu_plane.c
@@ -120,7 +120,7 @@ static int virtio_gpu_plane_atomic_check(struct drm_plane *plane,
 	crtc_state = drm_atomic_get_crtc_state(state,
 					       new_plane_state->crtc);
 	if (IS_ERR(crtc_state))
-                return PTR_ERR(crtc_state);
+		return PTR_ERR(crtc_state);
 
 	ret = drm_atomic_helper_check_plane_state(new_plane_state, crtc_state,
 						  DRM_PLANE_NO_SCALING,
diff --git a/drivers/gpu/drm/virtio/virtgpu_vq.c b/drivers/gpu/drm/virtio/virtgpu_vq.c
index 55a15e247dd1..8181b22b9b46 100644
--- a/drivers/gpu/drm/virtio/virtgpu_vq.c
+++ b/drivers/gpu/drm/virtio/virtgpu_vq.c
@@ -248,6 +248,7 @@ void virtio_gpu_dequeue_ctrl_func(struct work_struct *work)
 		if (resp->type != cpu_to_le32(VIRTIO_GPU_RESP_OK_NODATA)) {
 			if (le32_to_cpu(resp->type) >= VIRTIO_GPU_RESP_ERR_UNSPEC) {
 				struct virtio_gpu_ctrl_hdr *cmd;
+
 				cmd = virtio_gpu_vbuf_ctrl_hdr(entry);
 				DRM_ERROR_RATELIMITED("response 0x%x (command 0x%x)\n",
 						      le32_to_cpu(resp->type),
@@ -468,6 +469,7 @@ static int virtio_gpu_queue_fenced_ctrl_buffer(struct virtio_gpu_device *vgdev,
 	if (vbuf->data_size) {
 		if (is_vmalloc_addr(vbuf->data_buf)) {
 			int sg_ents;
+
 			sgt = vmalloc_to_sgt(vbuf->data_buf, vbuf->data_size,
 					     &sg_ents);
 			if (!sgt) {

From 2f44bb65f28f5f058fe4ab9bc81e5fde5575c473 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 12 Aug 2025 10:25:09 +0200
Subject: [PATCH 0550/1455] drm/panel: panel-samsung-s6e88a0-ams427ap24: Fix
 includes

Include <linux/property.h> to declare device_property_read_bool() and
<linux/mod_devicetable.h> to declare struct of_device_id. Avoids the
dependency on the backlight header to include both.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250812082509.227879-1-tzimmermann@suse.de
---
 drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams427ap24.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams427ap24.c b/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams427ap24.c
index e91f50662997..7e2f4e043d62 100644
--- a/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams427ap24.c
+++ b/drivers/gpu/drm/panel/panel-samsung-s6e88a0-ams427ap24.c
@@ -7,7 +7,9 @@
 #include <linux/backlight.h>
 #include <linux/delay.h>
 #include <linux/gpio/consumer.h>
+#include <linux/mod_devicetable.h>
 #include <linux/module.h>
+#include <linux/property.h>
 #include <linux/regulator/consumer.h>
 
 #include <video/mipi_display.h>

From 407a2fab3c99c40ad1acedaf028e8222da1f0433 Mon Sep 17 00:00:00 2001
From: Hsin-Yi Wang <hsinyi@chromium.org>
Date: Tue, 12 Aug 2025 16:17:58 +0800
Subject: [PATCH 0551/1455] drm_bridge: register content protect property

Some bridges can update HDCP status based on userspace requests if they
support HDCP.

The HDCP property is created after connector initialization and before
registration, just like other connector properties.

Add the content protection property to the connector if a bridge
supports HDCP.

Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Fei Shao <fshao@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250812082135.3351172-2-fshao@chromium.org
---
 drivers/gpu/drm/display/drm_bridge_connector.c | 9 +++++++++
 include/drm/drm_bridge.h                       | 4 ++++
 2 files changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c
index 8c915427d053..091c5335355a 100644
--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -20,6 +20,7 @@
 #include <drm/drm_modeset_helper_vtables.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
+#include <drm/display/drm_hdcp_helper.h>
 #include <drm/display/drm_hdmi_audio_helper.h>
 #include <drm/display/drm_hdmi_cec_helper.h>
 #include <drm/display/drm_hdmi_helper.h>
@@ -641,6 +642,7 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 	struct drm_bridge *bridge, *panel_bridge = NULL;
 	unsigned int supported_formats = BIT(HDMI_COLORSPACE_RGB);
 	unsigned int max_bpc = 8;
+	bool support_hdcp = false;
 	int connector_type;
 	int ret;
 
@@ -763,6 +765,9 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 
 		if (drm_bridge_is_panel(bridge))
 			panel_bridge = bridge;
+
+		if (bridge->support_hdcp)
+			support_hdcp = true;
 	}
 
 	if (connector_type == DRM_MODE_CONNECTOR_Unknown)
@@ -849,6 +854,10 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 	if (panel_bridge)
 		drm_panel_bridge_set_orientation(connector, panel_bridge);
 
+	if (support_hdcp && IS_REACHABLE(CONFIG_DRM_DISPLAY_HELPER) &&
+	    IS_ENABLED(CONFIG_DRM_DISPLAY_HDCP_HELPER))
+		drm_connector_attach_content_protection_property(connector, true);
+
 	return connector;
 }
 EXPORT_SYMBOL_GPL(drm_bridge_connector_init);
diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index 2ec1b136d603..8d9d4fd078e7 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -1171,6 +1171,10 @@ struct drm_bridge {
 	 * before the peripheral.
 	 */
 	bool pre_enable_prev_first;
+	/**
+	 * @support_hdcp: Indicate that the bridge supports HDCP.
+	 */
+	bool support_hdcp;
 	/**
 	 * @ddc: Associated I2C adapter for DDC access, if any.
 	 */

From 043d9c6928b010be7902a01b5cdfa7d754535b1a Mon Sep 17 00:00:00 2001
From: Hsin-Yi Wang <hsinyi@chromium.org>
Date: Tue, 12 Aug 2025 16:17:59 +0800
Subject: [PATCH 0552/1455] drm/bridge: anx7625: register content protect
 property

Set the `support_hdcp` bit to enable the connector to register content
protection during initialization.

Signed-off-by: Hsin-Yi Wang <hsinyi@chromium.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Fei Shao <fshao@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250812082135.3351172-3-fshao@chromium.org
---
 drivers/gpu/drm/bridge/analogix/anx7625.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c
index c0ad8f59e483..609cdb9d371e 100644
--- a/drivers/gpu/drm/bridge/analogix/anx7625.c
+++ b/drivers/gpu/drm/bridge/analogix/anx7625.c
@@ -2604,6 +2604,7 @@ static int anx7625_link_bridge(struct drm_dp_aux *aux)
 	platform->bridge.type = platform->pdata.panel_bridge ?
 				    DRM_MODE_CONNECTOR_eDP :
 				    DRM_MODE_CONNECTOR_DisplayPort;
+	platform->bridge.support_hdcp = true;
 
 	drm_bridge_add(&platform->bridge);
 

From 6627be03d3e1195b22172554191365378c3460e7 Mon Sep 17 00:00:00 2001
From: Nitin Gote <nitin.r.gote@intel.com>
Date: Tue, 19 Aug 2025 11:41:51 +0530
Subject: [PATCH 0553/1455] drm/xe/tuning: Apply "Disable NULL query for Anyhit
 Shader" to Xe2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Extend the "Disable NULL query for Anyhit Shader" tuning to Xe2
(graphics version 2000+) platforms, in addition to Xe3.

This sets the DIS_NULL_QUERY bit in RT_CTRL to disable null query
for Anyhit shaders on both Xe2 and Xe3.

This is a change in behavior that can regress a userspace not
prepared for it. However it's not feasible to change dynamically
the option per client or per exec queue via an opt-in flag. Mesa
is already prepared for that and it got propagated to their
stable versions. Even if it was possible, at this point adding a
flag would mean mesa would also need to propagate such a flag to
their stable versions, otherwise the previous fix would not be
used.

Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
Acked-by: Tapani Pälli <tapani.palli@intel.com>
Link: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35044
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/35044
Link: https://lore.kernel.org/r/20250819061151.1272622-1-nitin.r.gote@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_tuning.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_tuning.c b/drivers/gpu/drm/xe/xe_tuning.c
index 828b45b24c23..a524170a04d0 100644
--- a/drivers/gpu/drm/xe/xe_tuning.c
+++ b/drivers/gpu/drm/xe/xe_tuning.c
@@ -99,7 +99,7 @@ static const struct xe_rtp_entry_sr engine_tunings[] = {
 	  XE_RTP_ACTIONS(SET(SAMPLER_MODE, INDIRECT_STATE_BASE_ADDR_OVERRIDE))
 	},
 	{ XE_RTP_NAME("Tuning: Disable NULL query for Anyhit Shader"),
-	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, XE_RTP_END_VERSION_UNDEFINED),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(2000, XE_RTP_END_VERSION_UNDEFINED),
 		       FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(RT_CTRL, DIS_NULL_QUERY))
 	},

From 97a18f8349423371650996fc506dfd6cc4d3e1ad Mon Sep 17 00:00:00 2001
From: Gustavo Sousa <gustavo.sousa@intel.com>
Date: Mon, 18 Aug 2025 15:15:46 -0300
Subject: [PATCH 0554/1455] drm/xe: Probe for tile count during device info
 initialization

The function mmio_multi_tile_setup() does not look like the proper
location for probing for the number of existing tiles in the hardware.
It should not be that function's responsibility and such information
should instead be already available when it gets called.

The fact that we need to adjust gt_count is a symptom of that.

Move probing of available tile count to a dedicated function named
xe_info_probe_tile_count() and call it from xe_info_init(), which seems
like a more appropriate place for that.

With that move, we no longer need to (and shouldn't) adjust gt_count as
a part of xe_info_probe_tile_count(), as that field will be initialized
later in xe_info_init().

v4:
  - Only probe for tile count if the default tile_count != 1 (just like
    was done in mmio_multi_tile_setup()). (CI)
v3:
  - Unchanged.
v2:
  - Use KUnit static stub so that we do not try to query hardware when
    running KUnit tests. (CI)
  - Tweak last paragraph of commit message to make it clearer.
    (Jonathan)

Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250818-gt_count-improvements-v4-1-ee12870c6f57@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 drivers/gpu/drm/xe/tests/xe_pci.c |  7 ++++++
 drivers/gpu/drm/xe/xe_mmio.c      | 33 -------------------------
 drivers/gpu/drm/xe/xe_pci.c       | 41 +++++++++++++++++++++++++++++++
 3 files changed, 48 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c
index 9c715e59f030..db30c5156d0c 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci.c
+++ b/drivers/gpu/drm/xe/tests/xe_pci.c
@@ -101,6 +101,11 @@ static void fake_read_gmdid(struct xe_device *xe, enum xe_gmdid_type type,
 	}
 }
 
+static void fake_xe_info_probe_tile_count(struct xe_device *xe)
+{
+	/* Nothing to do, just use the statically defined value. */
+}
+
 int xe_pci_fake_device_init(struct xe_device *xe)
 {
 	struct kunit *test = kunit_get_current_test();
@@ -138,6 +143,8 @@ done:
 			   data->sriov_mode : XE_SRIOV_MODE_NONE;
 
 	kunit_activate_static_stub(test, read_gmdid, fake_read_gmdid);
+	kunit_activate_static_stub(test, xe_info_probe_tile_count,
+				   fake_xe_info_probe_tile_count);
 
 	xe_info_init_early(xe, desc, subplatform_desc);
 	xe_info_init(xe, desc);
diff --git a/drivers/gpu/drm/xe/xe_mmio.c b/drivers/gpu/drm/xe/xe_mmio.c
index e4db8d58ea2d..ef6f3ea573a2 100644
--- a/drivers/gpu/drm/xe/xe_mmio.c
+++ b/drivers/gpu/drm/xe/xe_mmio.c
@@ -58,7 +58,6 @@ static void tiles_fini(void *arg)
 static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size)
 {
 	struct xe_tile *tile;
-	struct xe_gt *gt;
 	u8 id;
 
 	/*
@@ -68,38 +67,6 @@ static void mmio_multi_tile_setup(struct xe_device *xe, size_t tile_mmio_size)
 	if (xe->info.tile_count == 1)
 		return;
 
-	/* Possibly override number of tile based on configuration register */
-	if (!xe->info.skip_mtcfg) {
-		struct xe_mmio *mmio = xe_root_tile_mmio(xe);
-		u8 tile_count, gt_count;
-		u32 mtcfg;
-
-		/*
-		 * Although the per-tile mmio regs are not yet initialized, this
-		 * is fine as it's going to the root tile's mmio, that's
-		 * guaranteed to be initialized earlier in xe_mmio_probe_early()
-		 */
-		mtcfg = xe_mmio_read32(mmio, XEHP_MTCFG_ADDR);
-		tile_count = REG_FIELD_GET(TILE_COUNT, mtcfg) + 1;
-
-		if (tile_count < xe->info.tile_count) {
-			drm_info(&xe->drm, "tile_count: %d, reduced_tile_count %d\n",
-				 xe->info.tile_count, tile_count);
-			xe->info.tile_count = tile_count;
-
-			/*
-			 * We've already setup gt_count according to the full
-			 * tile count.  Re-calculate it to only include the GTs
-			 * that belong to the remaining tile(s).
-			 */
-			gt_count = 0;
-			for_each_gt(gt, xe, id)
-				if (gt->info.id < tile_count * xe->info.max_gt_per_tile)
-					gt_count++;
-			xe->info.gt_count = gt_count;
-		}
-	}
-
 	for_each_remote_tile(tile, xe, id)
 		xe_mmio_init(&tile->mmio, tile, xe->mmio.regs + id * tile_mmio_size, SZ_4M);
 }
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 053faa7f9b01..bc73fe176165 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -17,6 +17,7 @@
 
 #include "display/xe_display.h"
 #include "regs/xe_gt_regs.h"
+#include "regs/xe_regs.h"
 #include "xe_device.h"
 #include "xe_drv.h"
 #include "xe_gt.h"
@@ -603,6 +604,44 @@ static int xe_info_init_early(struct xe_device *xe,
 	return 0;
 }
 
+/*
+ * Possibly override number of tile based on configuration register.
+ */
+static void xe_info_probe_tile_count(struct xe_device *xe)
+{
+	struct xe_mmio *mmio;
+	u8 tile_count;
+	u32 mtcfg;
+
+	KUNIT_STATIC_STUB_REDIRECT(xe_info_probe_tile_count, xe);
+
+	/*
+	 * Probe for tile count only for platforms that support multiple
+	 * tiles.
+	 */
+	if (xe->info.tile_count == 1)
+		return;
+
+	if (xe->info.skip_mtcfg)
+		return;
+
+	mmio = xe_root_tile_mmio(xe);
+
+	/*
+	 * Although the per-tile mmio regs are not yet initialized, this
+	 * is fine as it's going to the root tile's mmio, that's
+	 * guaranteed to be initialized earlier in xe_mmio_probe_early()
+	 */
+	mtcfg = xe_mmio_read32(mmio, XEHP_MTCFG_ADDR);
+	tile_count = REG_FIELD_GET(TILE_COUNT, mtcfg) + 1;
+
+	if (tile_count < xe->info.tile_count) {
+		drm_info(&xe->drm, "tile_count: %d, reduced_tile_count %d\n",
+			 xe->info.tile_count, tile_count);
+		xe->info.tile_count = tile_count;
+	}
+}
+
 /*
  * Initialize device info content that does require knowledge about
  * graphics / media IP version.
@@ -677,6 +716,8 @@ static int xe_info_init(struct xe_device *xe,
 	xe->info.has_usm = graphics_desc->has_usm;
 	xe->info.has_64bit_timestamp = graphics_desc->has_64bit_timestamp;
 
+	xe_info_probe_tile_count(xe);
+
 	for_each_remote_tile(tile, xe, id) {
 		int err;
 

From 1046be1db620d00eb390a94ac7b67109e0beeff5 Mon Sep 17 00:00:00 2001
From: Gustavo Sousa <gustavo.sousa@intel.com>
Date: Mon, 18 Aug 2025 15:15:47 -0300
Subject: [PATCH 0555/1455] drm/xe: Use for_each_gt to define gt_count

We are currently bumping gt_count as we define GTs for each tile. While
that works with our current code, there are reasons to improve that:

* That section of the code is dedicated to define each tile's set of GTs
  and their respective info. The gt_count can be seen as a device level
  property. While it is fair to bump it as we define each GT, we can
  also focus on the GT themselves and count after we are done with the
  definitions.

* More *importantly*, gt_count should reflect the number of GTs that we
  would get when looping over them with for_each_gt(). Bumping gt_count
  the way we are currently doing makes that value not really connected
  to for_each_gt(). This could bite us in the future if in the loop gets
  extra checks that are not accounted for in each existing "gt_count++".

As such, let's use for_each_gt() and extract the calculation of gt_count
into a separate block, just after we define the set of GTs for each
tile.

Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250818-gt_count-improvements-v4-2-ee12870c6f57@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 drivers/gpu/drm/xe/xe_pci.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index bc73fe176165..419e29cebbba 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -740,7 +740,6 @@ static int xe_info_init(struct xe_device *xe,
 		gt->info.id = tile->id * xe->info.max_gt_per_tile;
 		gt->info.has_indirect_ring_state = graphics_desc->has_indirect_ring_state;
 		gt->info.engine_mask = graphics_desc->hw_engine_mask;
-		xe->info.gt_count++;
 
 		err = xe_tile_alloc_vram(tile);
 		if (err)
@@ -765,9 +764,15 @@ static int xe_info_init(struct xe_device *xe,
 		gt->info.id = tile->id * xe->info.max_gt_per_tile + 1;
 		gt->info.has_indirect_ring_state = media_desc->has_indirect_ring_state;
 		gt->info.engine_mask = media_desc->hw_engine_mask;
-		xe->info.gt_count++;
 	}
 
+	/*
+	 * Now that we have tiles and GTs defined, let's loop over valid GTs
+	 * in order to define gt_count.
+	 */
+	for_each_gt(gt, xe, id)
+		xe->info.gt_count++;
+
 	return 0;
 }
 

From 11c09b55b3d36f5cfa15873e2e0f1e04166bcfd4 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Tue, 19 Aug 2025 21:34:39 +0530
Subject: [PATCH 0556/1455] drm/i915/backlight: Fix divide by 0 error in
 i9xx_set_backlight

pwm_level_max maybe 0 we do throw a warning but move ahead with
execution which may later cause a /0 error.

--v2
-return if the warn_on gets hit [Jani]

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/20250819160438.145734-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_backlight.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_backlight.c b/drivers/gpu/drm/i915/display/intel_backlight.c
index e007380e9a63..3b14f929825a 100644
--- a/drivers/gpu/drm/i915/display/intel_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_backlight.c
@@ -236,7 +236,8 @@ static void i9xx_set_backlight(const struct drm_connector_state *conn_state, u32
 	struct intel_panel *panel = &connector->panel;
 	u32 tmp, mask;
 
-	drm_WARN_ON(display->drm, panel->backlight.pwm_level_max == 0);
+	if (drm_WARN_ON(display->drm, panel->backlight.pwm_level_max == 0))
+		return;
 
 	if (panel->backlight.combination_mode) {
 		struct pci_dev *pdev = to_pci_dev(display->drm->dev);

From f9ad930c52a8d7e945b65776e379ed327cf3b904 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 15 Aug 2025 11:45:32 +0300
Subject: [PATCH 0557/1455] drm/i915/psr: drm_WARN_ON when activating disabled
 PSR
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add drm_WARN_ON for scenario where PSR is activated while it is disabled.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250815084534.1637030-2-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 609df53f1ef1..4a80f12563f8 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -1787,6 +1787,8 @@ static void intel_psr_activate(struct intel_dp *intel_dp)
 
 	drm_WARN_ON(display->drm, intel_dp->psr.active);
 
+	drm_WARN_ON(display->drm, !intel_dp->psr.enabled);
+
 	lockdep_assert_held(&intel_dp->psr.lock);
 
 	/* psr1, psr2 and panel-replay are mutually exclusive.*/

From 73b60be7f62f1edfef7aac4783878932b2b59ad9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 15 Aug 2025 11:45:33 +0300
Subject: [PATCH 0558/1455] drm/i915/psr: Do not activate disabled PSR on
 irq_aux_error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently intel_psr_work is continuing to activation of PSR which was just
disabled when irq_aux_error == true.

Fix this by skipping everything else than intel_psr_handle_irq in
intel_psr_work when irq_aux_error == true.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250815084534.1637030-3-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 4a80f12563f8..86836aa61413 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -3211,8 +3211,10 @@ static void intel_psr_work(struct work_struct *work)
 	if (!intel_dp->psr.enabled)
 		goto unlock;
 
-	if (READ_ONCE(intel_dp->psr.irq_aux_error))
+	if (READ_ONCE(intel_dp->psr.irq_aux_error)) {
 		intel_psr_handle_irq(intel_dp);
+		goto unlock;
+	}
 
 	/*
 	 * We have to make sure PSR is ready for re-enable

From 7c8c76272e1732c63aab760b0b656f00bdc67050 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 15 Aug 2025 11:45:34 +0300
Subject: [PATCH 0559/1455] drm/i915/psr: Check pause counter before continuing
 to PSR activation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently intel_psr_work is re-activating PSR even when pause_counter > 0
which is incorrect. Fix this by checking pause_counter before re-activating
PSR.

Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14822
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250815084534.1637030-4-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 86836aa61413..465f12c39330 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -3216,6 +3216,9 @@ static void intel_psr_work(struct work_struct *work)
 		goto unlock;
 	}
 
+	if (intel_dp->psr.pause_counter)
+		goto unlock;
+
 	/*
 	 * We have to make sure PSR is ready for re-enable
 	 * otherwise it keeps disabled until next full enable/disable cycle.

From 4aa8961b1b9c7498550b41168a91cf1558133dd3 Mon Sep 17 00:00:00 2001
From: Val Packett <val@packett.cool>
Date: Sun, 6 Jul 2025 17:42:24 -0300
Subject: [PATCH 0560/1455] drm/dp: drm_edp_backlight_set_level: do not always
 send 3-byte commands

At least some panels using the LSB register are not happy with the
unconditional increase of the command buffer to 3 bytes.

With the BOE NE14QDM in my Dell Latitude 7455, the recent patches for
luminance based brightness have introduced a regression: the brightness
range stopped being contiguous and became nonsensical (it probably was
interpreting the last 2 bytes of the buffer and not the first 2).

Change from using a fixed sizeof() to a length variable that's only
set to 3 when luminance is used. Let's leave the default as 2 even for
the single-byte version, since that's how it worked before.

Fixes: f2db78e37fe7 ("drm/dp: Modify drm_edp_backlight_set_level")
Signed-off-by: Val Packett <val@packett.cool>
Tested-by: Abel Vesa <abel.vesa@linaro.org>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250706204446.8918-1-val@packett.cool
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/display/drm_dp_helper.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/display/drm_dp_helper.c b/drivers/gpu/drm/display/drm_dp_helper.c
index 1ecc3df7e316..4aaeae4fa03c 100644
--- a/drivers/gpu/drm/display/drm_dp_helper.c
+++ b/drivers/gpu/drm/display/drm_dp_helper.c
@@ -3962,6 +3962,7 @@ int drm_edp_backlight_set_level(struct drm_dp_aux *aux, const struct drm_edp_bac
 	int ret;
 	unsigned int offset = DP_EDP_BACKLIGHT_BRIGHTNESS_MSB;
 	u8 buf[3] = { 0 };
+	size_t len = 2;
 
 	/* The panel uses the PWM for controlling brightness levels */
 	if (!(bl->aux_set || bl->luminance_set))
@@ -3974,6 +3975,7 @@ int drm_edp_backlight_set_level(struct drm_dp_aux *aux, const struct drm_edp_bac
 		buf[1] = (level & 0x00ff00) >> 8;
 		buf[2] = (level & 0xff0000) >> 16;
 		offset = DP_EDP_PANEL_TARGET_LUMINANCE_VALUE;
+		len = 3;
 	} else if (bl->lsb_reg_used) {
 		buf[0] = (level & 0xff00) >> 8;
 		buf[1] = (level & 0x00ff);
@@ -3981,7 +3983,7 @@ int drm_edp_backlight_set_level(struct drm_dp_aux *aux, const struct drm_edp_bac
 		buf[0] = level;
 	}
 
-	ret = drm_dp_dpcd_write_data(aux, offset, buf, sizeof(buf));
+	ret = drm_dp_dpcd_write_data(aux, offset, buf, len);
 	if (ret < 0) {
 		drm_err(aux->drm_dev,
 			"%s: Failed to write aux backlight level: %d\n",

From d234f3aef5928c2f0d692cf43768ecced0abd672 Mon Sep 17 00:00:00 2001
From: Shankari Anand <shankari.ak0208@gmail.com>
Date: Wed, 20 Aug 2025 16:58:46 +0530
Subject: [PATCH 0561/1455] gpu: nova-core: Update ARef imports from sync::aref

Update call sites in nova-core to import `ARef`
from `sync::aref` instead of `types`.

This aligns with the ongoing effort to move `ARef` and
`AlwaysRefCounted` to sync.

[acourbot@nvidia.com: use standard prefix for nova-core.]

Suggested-by: Benno Lossin <lossin@kernel.org>
Link: https://github.com/Rust-for-Linux/linux/issues/1173
Signed-off-by: Shankari Anand <shankari.ak0208@gmail.com>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Link: https://lore.kernel.org/r/20250820112846.9665-1-shankari.ak0208@gmail.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/falcon.rs | 2 +-
 drivers/gpu/nova-core/fb.rs     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index d235a6f9efca..645e65175cb6 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -7,8 +7,8 @@ use hal::FalconHal;
 use kernel::bindings;
 use kernel::device;
 use kernel::prelude::*;
+use kernel::sync::aref::ARef;
 use kernel::time::Delta;
-use kernel::types::ARef;
 
 use crate::dma::DmaObject;
 use crate::driver::Bar0;
diff --git a/drivers/gpu/nova-core/fb.rs b/drivers/gpu/nova-core/fb.rs
index 4a702525fff4..68559902ae78 100644
--- a/drivers/gpu/nova-core/fb.rs
+++ b/drivers/gpu/nova-core/fb.rs
@@ -4,7 +4,7 @@ use core::ops::Range;
 
 use kernel::prelude::*;
 use kernel::sizes::*;
-use kernel::types::ARef;
+use kernel::sync::aref::ARef;
 use kernel::{dev_warn, device};
 
 use crate::dma::DmaObject;

From 96481012628c2d9a03fb758763ad7bff786203eb Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <rgallaispou@gmail.com>
Date: Fri, 28 Mar 2025 12:41:48 +0100
Subject: [PATCH 0562/1455] drm: of: fix documentation reference

Documentation/devicetree/bindings/graph.txt content has move directly to
the dt-schema repo in commit 4b52be0ce6ad ("dt-bindings: Remove plain
text OF graph binding").

Point to the YAML of the official repo instead of the old file.

Signed-off-by: Raphael Gallais-Pou <rgallaispou@gmail.com>
Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250328114148.260322-1-rgallaispou@gmail.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 drivers/gpu/drm/drm_of.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
index d0183dea7703..4f65ce729a47 100644
--- a/drivers/gpu/drm/drm_of.c
+++ b/drivers/gpu/drm/drm_of.c
@@ -55,7 +55,8 @@ EXPORT_SYMBOL(drm_of_crtc_port_mask);
  * and generate the DRM mask of CRTCs which may be attached to this
  * encoder.
  *
- * See Documentation/devicetree/bindings/graph.txt for the bindings.
+ * See https://github.com/devicetree-org/dt-schema/blob/main/dtschema/schemas/graph.yaml
+ * for the bindings.
  */
 uint32_t drm_of_find_possible_crtcs(struct drm_device *dev,
 				    struct device_node *port)
@@ -106,7 +107,9 @@ EXPORT_SYMBOL_GPL(drm_of_component_match_add);
  * Parse the platform device OF node and bind all the components associated
  * with the master. Interface ports are added before the encoders in order to
  * satisfy their .bind requirements
- * See Documentation/devicetree/bindings/graph.txt for the bindings.
+ *
+ * See https://github.com/devicetree-org/dt-schema/blob/main/dtschema/schemas/graph.yaml
+ * for the bindings.
  *
  * Returns zero if successful, or one of the standard error codes if it fails.
  */

From 61ce50fd8196c8782b9620bb60d33649ec429f64 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 21 Aug 2025 15:30:54 +0300
Subject: [PATCH 0563/1455] drm/panel: jdi-lpm102a188a: Fix error code in
 jdi_panel_prepare()

If the mipi_dsi_dual() macro fails, the error code is stored in
dsi_ctx.accum_err.  Propagate that error back to the caller instead
of returning success as the current code does.

Fixes: a6adf47d30cc ("drm/panel: jdi-lpm102a188a: Fix bug and clean up driver")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/aKcRfq8xBrFmhqmO@stanley.mountain
---
 drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c b/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c
index 83656bb4b0b2..23462065d726 100644
--- a/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c
+++ b/drivers/gpu/drm/panel/panel-jdi-lpm102a188a.c
@@ -233,8 +233,10 @@ static int jdi_panel_prepare(struct drm_panel *panel)
 	mipi_dsi_dual(mipi_dsi_dcs_set_display_on_multi,
 		      &dsi_ctx, jdi->link1, jdi->link2);
 
-	if (dsi_ctx.accum_err < 0)
+	if (dsi_ctx.accum_err < 0) {
+		err = dsi_ctx.accum_err;
 		goto poweroff;
+	}
 
 	jdi->link1->mode_flags &= ~MIPI_DSI_MODE_LPM;
 	jdi->link2->mode_flags &= ~MIPI_DSI_MODE_LPM;

From 3f13bcc886fc034113cb75cb32b8d9db1216b846 Mon Sep 17 00:00:00 2001
From: Lizhi Hou <lizhi.hou@amd.com>
Date: Wed, 20 Aug 2025 16:38:17 -0700
Subject: [PATCH 0564/1455] Documentation: accel: amdxdna: Update compiler
 information

The compiler information is outdated. Update it to the latest.

Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250820233817.4050006-1-lizhi.hou@amd.com
---
 Documentation/accel/amdxdna/amdnpu.rst | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/Documentation/accel/amdxdna/amdnpu.rst b/Documentation/accel/amdxdna/amdnpu.rst
index fbe0a7585345..42e54904f9a8 100644
--- a/Documentation/accel/amdxdna/amdnpu.rst
+++ b/Documentation/accel/amdxdna/amdnpu.rst
@@ -223,13 +223,13 @@ Userspace components
 Compiler
 --------
 
-Peano is an LLVM based open-source compiler for AMD XDNA Array compute tile
-available at:
+Peano is an LLVM based open-source single core compiler for AMD XDNA Array
+compute tile. Peano is available at:
 https://github.com/Xilinx/llvm-aie
 
-The open-source IREE compiler supports graph compilation of ML models for AMD
-NPU and uses Peano underneath. It is available at:
-https://github.com/nod-ai/iree-amd-aie
+IRON is an open-source array compiler for AMD XDNA Array based NPU which uses
+Peano underneath. IRON is available at:
+https://github.com/Xilinx/mlir-aie
 
 Usermode Driver (UMD)
 ---------------------

From bc2b2062682a7f8d828b0020e37a84a9234f03a6 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 18 Aug 2025 21:20:32 +0200
Subject: [PATCH 0565/1455] drm/xe/kunit: Extend platform generator with PTL

Our list of typical platforms used to generate test device objects
does not contain any PANTHERLAKE. Add one.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250818192032.633-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_wa_test.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/tests/xe_wa_test.c b/drivers/gpu/drm/xe/tests/xe_wa_test.c
index c96d1fe34151..416258c193f6 100644
--- a/drivers/gpu/drm/xe/tests/xe_wa_test.c
+++ b/drivers/gpu/drm/xe/tests/xe_wa_test.c
@@ -75,6 +75,7 @@ static const struct platform_test_case cases[] = {
 	GMDID_CASE(LUNARLAKE, 2004, A0, 2000, A0),
 	GMDID_CASE(LUNARLAKE, 2004, B0, 2000, A0),
 	GMDID_CASE(BATTLEMAGE, 2001, A0, 1301, A1),
+	GMDID_CASE(PANTHERLAKE, 3000, A0, 3000, A0),
 };
 
 static void platform_desc(const struct platform_test_case *t, char *desc)

From 99e7f8e83a9c5a56ca20c4948b6126248c82c67d Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Thu, 21 Aug 2025 13:49:32 +0900
Subject: [PATCH 0566/1455] gpu: nova-core: falcon: align DMA transfers to 256
 bytes

Falcon DMA transfers are done in 256 bytes increments, and the method
responsible for initiating the transfer checked that the required length
was indeed a multiple of 256. While correct, this also requires callers
to specifically account for this limitation of DMA transfers, and we had
for instance the fwsec code performing a seemingly arbitrary (and
potentially overflowing) upwards alignment of the DMEM load size to
match this requirement.

Let's move that alignment into the loading code itself instead: since it
is working in terms of number of transfers, we can turn this upwards
alignment into a non-overflowing operation, and check that the requested
transfer remains into the limits of the DMA object. This also allows us
to remove a DMA-specific constant in the fwsec code.

Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
Link: https://lore.kernel.org/r/20250821-falcondma_256b-v2-1-83e8647a24b5@nvidia.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/nova-core/falcon.rs         | 31 ++++++++++++++++++-------
 drivers/gpu/nova-core/firmware/fwsec.rs |  9 +------
 2 files changed, 23 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index 645e65175cb6..be91aac6976a 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -463,14 +463,27 @@ impl<E: FalconEngine + 'static> Falcon<E> {
             );
             return Err(EINVAL);
         }
-        if load_offsets.len % DMA_LEN > 0 {
-            dev_err!(
-                self.dev,
-                "DMA transfer length must be a multiple of {}",
-                DMA_LEN
-            );
-            return Err(EINVAL);
-        }
+
+        // DMA transfers can only be done in units of 256 bytes. Compute how many such transfers we
+        // need to perform.
+        let num_transfers = load_offsets.len.div_ceil(DMA_LEN);
+
+        // Check that the area we are about to transfer is within the bounds of the DMA object.
+        // Upper limit of transfer is `(num_transfers * DMA_LEN) + load_offsets.src_start`.
+        match num_transfers
+            .checked_mul(DMA_LEN)
+            .and_then(|size| size.checked_add(load_offsets.src_start))
+        {
+            None => {
+                dev_err!(self.dev, "DMA transfer length overflow");
+                return Err(EOVERFLOW);
+            }
+            Some(upper_bound) if upper_bound as usize > fw.size() => {
+                dev_err!(self.dev, "DMA transfer goes beyond range of DMA object");
+                return Err(EINVAL);
+            }
+            Some(_) => (),
+        };
 
         // Set up the base source DMA address.
 
@@ -486,7 +499,7 @@ impl<E: FalconEngine + 'static> Falcon<E> {
             .set_imem(target_mem == FalconMem::Imem)
             .set_sec(if sec { 1 } else { 0 });
 
-        for pos in (0..load_offsets.len).step_by(DMA_LEN as usize) {
+        for pos in (0..num_transfers).map(|i| i * DMA_LEN) {
             // Perform a transfer of size `DMA_LEN`.
             regs::NV_PFALCON_FALCON_DMATRFMOFFS::default()
                 .set_offs(load_offsets.dst_start + pos)
diff --git a/drivers/gpu/nova-core/firmware/fwsec.rs b/drivers/gpu/nova-core/firmware/fwsec.rs
index 0dff3cfa90af..47f5e4524072 100644
--- a/drivers/gpu/nova-core/firmware/fwsec.rs
+++ b/drivers/gpu/nova-core/firmware/fwsec.rs
@@ -202,9 +202,6 @@ pub(crate) struct FwsecFirmware {
     ucode: FirmwareDmaObject<Self, Signed>,
 }
 
-// We need to load full DMEM pages.
-const DMEM_LOAD_SIZE_ALIGN: u32 = 256;
-
 impl FalconLoadParams for FwsecFirmware {
     fn imem_load_params(&self) -> FalconLoadTarget {
         FalconLoadTarget {
@@ -218,11 +215,7 @@ impl FalconLoadParams for FwsecFirmware {
         FalconLoadTarget {
             src_start: self.desc.imem_load_size,
             dst_start: self.desc.dmem_phys_base,
-            // TODO[NUMM]: replace with `align_up` once it lands.
-            len: self
-                .desc
-                .dmem_load_size
-                .next_multiple_of(DMEM_LOAD_SIZE_ALIGN),
+            len: self.desc.dmem_load_size,
         }
     }
 

From 1db476d294c031c38d3e5cbc4d41faf6dfd77ffb Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 1 Aug 2025 22:24:09 +0900
Subject: [PATCH 0567/1455] rust: transmute: add `as_bytes` method for
 `AsBytes` trait

Every type that implements `AsBytes` should be able to provide its byte
representation. Introduce the `as_bytes` method that returns the
implementer as a stream of bytes, and provide a default implementation
that should be suitable for any type that satisfies `AsBytes`'s safety
requirements.

[acourbot@nvidia.com: use fully qualified `core::mem::size_of_val` to
build with Rust 1.78.]

Reviewed-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Gary Guo <gary@garyguo.net>
Acked-by: Miguel Ojeda <ojeda@kernel.org>
Link: https://lore.kernel.org/r/20250801-as_bytes-v5-1-975f87d5dc85@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 rust/kernel/transmute.rs | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/rust/kernel/transmute.rs b/rust/kernel/transmute.rs
index 1c7d43771a37..a2d8f39182b2 100644
--- a/rust/kernel/transmute.rs
+++ b/rust/kernel/transmute.rs
@@ -47,7 +47,17 @@ impl_frombytes! {
 ///
 /// Values of this type may not contain any uninitialized bytes. This type must not have interior
 /// mutability.
-pub unsafe trait AsBytes {}
+pub unsafe trait AsBytes {
+    /// Returns `self` as a slice of bytes.
+    fn as_bytes(&self) -> &[u8] {
+        // CAST: `Self` implements `AsBytes` thus all bytes of `self` are initialized.
+        let data = core::ptr::from_ref(self).cast::<u8>();
+        let len = core::mem::size_of_val(self);
+
+        // SAFETY: `data` is non-null and valid for reads of `len * sizeof::<u8>()` bytes.
+        unsafe { core::slice::from_raw_parts(data, len) }
+    }
+}
 
 macro_rules! impl_asbytes {
     ($($({$($generics:tt)*})? $t:ty, )*) => {

From 331c24e6ce814af2af74bac648d1ac1708873e9c Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 1 Aug 2025 22:24:10 +0900
Subject: [PATCH 0568/1455] rust: transmute: add `as_bytes_mut` method to
 `AsBytes` trait

Types that implement both `AsBytes` and `FromBytes` can be safely
modified as a slice of bytes. Add a `as_bytes_mut` method for that
purpose.

[acourbot@nvidia.com: use fully qualified `core::mem::size_of_val` to
build with Rust 1.78.]

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Gary Guo <gary@garyguo.net>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Acked-by: Miguel Ojeda <ojeda@kernel.org>
Link: https://lore.kernel.org/r/20250801-as_bytes-v5-2-975f87d5dc85@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 rust/kernel/transmute.rs | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/rust/kernel/transmute.rs b/rust/kernel/transmute.rs
index a2d8f39182b2..517a432ac693 100644
--- a/rust/kernel/transmute.rs
+++ b/rust/kernel/transmute.rs
@@ -57,6 +57,21 @@ pub unsafe trait AsBytes {
         // SAFETY: `data` is non-null and valid for reads of `len * sizeof::<u8>()` bytes.
         unsafe { core::slice::from_raw_parts(data, len) }
     }
+
+    /// Returns `self` as a mutable slice of bytes.
+    fn as_bytes_mut(&mut self) -> &mut [u8]
+    where
+        Self: FromBytes,
+    {
+        // CAST: `Self` implements both `AsBytes` and `FromBytes` thus making `Self`
+        // bi-directionally transmutable to `[u8; size_of_val(self)]`.
+        let data = core::ptr::from_mut(self).cast::<u8>();
+        let len = core::mem::size_of_val(self);
+
+        // SAFETY: `data` is non-null and valid for read and writes of `len * sizeof::<u8>()`
+        // bytes.
+        unsafe { core::slice::from_raw_parts_mut(data, len) }
+    }
 }
 
 macro_rules! impl_asbytes {

From 2515d2b9ab4108c11a0b23935e68de27abb8b2a7 Mon Sep 17 00:00:00 2001
From: Stuart Summers <stuart.summers@intel.com>
Date: Tue, 19 Aug 2025 20:10:54 +0000
Subject: [PATCH 0569/1455] drm/xe/pcode: Initialize data0 for pcode read
 routine

There are two registers filled in when reading data from
pcode besides the mailbox itself. Currently, we allow a NULL
value for the second of these two (data1) and assume the first
is defined. However, many of the routines that are calling
this function assume that pcode will ignore the value being
passed in and so leave that first value (data0) defined but
uninitialized. To be safe, make sure this value is always
initialized to something (0 generally) in the event pcode
behavior changes and starts using this value.

v2: Fix sob/author

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250819201054.393220-1-stuart.summers@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_device_sysfs.c | 8 ++++----
 drivers/gpu/drm/xe/xe_hwmon.c        | 8 ++++----
 drivers/gpu/drm/xe/xe_vram_freq.c    | 4 ++--
 3 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index bd9015761aa0..6ee422594b56 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -76,7 +76,7 @@ lb_fan_control_version_show(struct device *dev, struct device_attribute *attr, c
 {
 	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
 	struct xe_tile *root = xe_device_get_root_tile(xe);
-	u32 cap, ver_low = FAN_TABLE, ver_high = FAN_TABLE;
+	u32 cap = 0, ver_low = FAN_TABLE, ver_high = FAN_TABLE;
 	u16 major = 0, minor = 0, hotfix = 0, build = 0;
 	int ret;
 
@@ -115,7 +115,7 @@ lb_voltage_regulator_version_show(struct device *dev, struct device_attribute *a
 {
 	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
 	struct xe_tile *root = xe_device_get_root_tile(xe);
-	u32 cap, ver_low = VR_CONFIG, ver_high = VR_CONFIG;
+	u32 cap = 0, ver_low = VR_CONFIG, ver_high = VR_CONFIG;
 	u16 major = 0, minor = 0, hotfix = 0, build = 0;
 	int ret;
 
@@ -153,7 +153,7 @@ static int late_bind_create_files(struct device *dev)
 {
 	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
 	struct xe_tile *root = xe_device_get_root_tile(xe);
-	u32 cap;
+	u32 cap = 0;
 	int ret;
 
 	xe_pm_runtime_get(xe);
@@ -186,7 +186,7 @@ static void late_bind_remove_files(struct device *dev)
 {
 	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
 	struct xe_tile *root = xe_device_get_root_tile(xe);
-	u32 cap;
+	u32 cap = 0;
 	int ret;
 
 	xe_pm_runtime_get(xe);
diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c
index c17ed1ae8649..32a76ae6e9dc 100644
--- a/drivers/gpu/drm/xe/xe_hwmon.c
+++ b/drivers/gpu/drm/xe/xe_hwmon.c
@@ -179,7 +179,7 @@ static int xe_hwmon_pcode_rmw_power_limit(const struct xe_hwmon *hwmon, u32 attr
 					  u32 clr, u32 set)
 {
 	struct xe_tile *root_tile = xe_device_get_root_tile(hwmon->xe);
-	u32 val0, val1;
+	u32 val0 = 0, val1 = 0;
 	int ret = 0;
 
 	ret = xe_pcode_read(root_tile, PCODE_MBOX(PCODE_POWER_SETUP,
@@ -734,7 +734,7 @@ static int xe_hwmon_power_curr_crit_read(struct xe_hwmon *hwmon, int channel,
 					 long *value, u32 scale_factor)
 {
 	int ret;
-	u32 uval;
+	u32 uval = 0;
 
 	mutex_lock(&hwmon->hwmon_lock);
 
@@ -918,7 +918,7 @@ xe_hwmon_power_write(struct xe_hwmon *hwmon, u32 attr, int channel, long val)
 static umode_t
 xe_hwmon_curr_is_visible(const struct xe_hwmon *hwmon, u32 attr, int channel)
 {
-	u32 uval;
+	u32 uval = 0;
 
 	/* hwmon sysfs attribute of current available only for package */
 	if (channel != CHANNEL_PKG)
@@ -1020,7 +1020,7 @@ xe_hwmon_energy_read(struct xe_hwmon *hwmon, u32 attr, int channel, long *val)
 static umode_t
 xe_hwmon_fan_is_visible(struct xe_hwmon *hwmon, u32 attr, int channel)
 {
-	u32 uval;
+	u32 uval = 0;
 
 	if (!hwmon->xe->info.has_fan_control)
 		return 0;
diff --git a/drivers/gpu/drm/xe/xe_vram_freq.c b/drivers/gpu/drm/xe/xe_vram_freq.c
index b26e26d73dae..17bc84da4cdc 100644
--- a/drivers/gpu/drm/xe/xe_vram_freq.c
+++ b/drivers/gpu/drm/xe/xe_vram_freq.c
@@ -34,7 +34,7 @@ static ssize_t max_freq_show(struct device *dev, struct device_attribute *attr,
 			     char *buf)
 {
 	struct xe_tile *tile = dev_to_tile(dev);
-	u32 val, mbox;
+	u32 val = 0, mbox;
 	int err;
 
 	mbox = REG_FIELD_PREP(PCODE_MB_COMMAND, PCODE_FREQUENCY_CONFIG)
@@ -56,7 +56,7 @@ static ssize_t min_freq_show(struct device *dev, struct device_attribute *attr,
 			     char *buf)
 {
 	struct xe_tile *tile = dev_to_tile(dev);
-	u32 val, mbox;
+	u32 val = 0, mbox;
 	int err;
 
 	mbox = REG_FIELD_PREP(PCODE_MB_COMMAND, PCODE_FREQUENCY_CONFIG)

From b065bd213caf6d35b57c5089d6507d7e8598a586 Mon Sep 17 00:00:00 2001
From: Xaver Hugl <xaver.hugl@kde.org>
Date: Fri, 22 Aug 2025 17:28:49 +0200
Subject: [PATCH 0570/1455] drm: re-allow no-op changes on non-primary planes
 in async flips
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit fd40a63c63a1 ("drm/atomic: Let drivers decide which planes to
async flip") unintentionally disallowed no-op changes on non-primary
planes that the driver doesn't allow async flips on. This broke async
flips for compositors that disable the cursor plane in every async
atomic commit. To fix that, change drm_atomic_set_property to again
only run atomic_async_check if the plane would actually be changed by
the atomic commit.

Fixes: fd40a63c63a1 ("drm/atomic: Let drivers decide which planes to async flip")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4263
Signed-off-by: Xaver Hugl <xaver.hugl@kde.org>
Reviewed-by: André Almeida <andrealmeid@igalia.com>
Link: https://lore.kernel.org/r/20250822152849.87843-1-xaver.hugl@kde.org
[andrealmeid: fix checkpatch warning]
Signed-off-by: André Almeida <andrealmeid@igalia.com>
---
 drivers/gpu/drm/drm_atomic_uapi.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/drm_atomic_uapi.c b/drivers/gpu/drm/drm_atomic_uapi.c
index ecc73d52bfae..85dbdaa4a2e2 100644
--- a/drivers/gpu/drm/drm_atomic_uapi.c
+++ b/drivers/gpu/drm/drm_atomic_uapi.c
@@ -1078,19 +1078,20 @@ int drm_atomic_set_property(struct drm_atomic_state *state,
 		}
 
 		if (async_flip) {
-			/* check if the prop does a nop change */
-			if ((prop != config->prop_fb_id &&
-			     prop != config->prop_in_fence_fd &&
-			     prop != config->prop_fb_damage_clips)) {
-				ret = drm_atomic_plane_get_property(plane, plane_state,
-								    prop, &old_val);
-				ret = drm_atomic_check_prop_changes(ret, old_val, prop_value, prop);
+			/* no-op changes are always allowed */
+			ret = drm_atomic_plane_get_property(plane, plane_state,
+							    prop, &old_val);
+			ret = drm_atomic_check_prop_changes(ret, old_val, prop_value, prop);
+
+			/* fail everything that isn't no-op or a pure flip */
+			if (ret && prop != config->prop_fb_id &&
+			    prop != config->prop_in_fence_fd &&
+			    prop != config->prop_fb_damage_clips) {
+				break;
 			}
 
-			/* ask the driver if this non-primary plane is supported */
-			if (plane->type != DRM_PLANE_TYPE_PRIMARY) {
-				ret = -EINVAL;
-
+			if (ret && plane->type != DRM_PLANE_TYPE_PRIMARY) {
+				/* ask the driver if this non-primary plane is supported */
 				if (plane_funcs && plane_funcs->atomic_async_check)
 					ret = plane_funcs->atomic_async_check(plane, state, true);
 

From 480a76c64f4f6a7f8d4aa2ac86cd2178687716b9 Mon Sep 17 00:00:00 2001
From: Andy Yan <andy.yan@rock-chips.com>
Date: Fri, 22 Aug 2025 14:39:49 +0800
Subject: [PATCH 0571/1455] dt-bindings: display: simple-bridge: Add ra620
 compatible

RA620 is a DP to HDMI bridge converter from RADXA, which first
found be used on ROCK 5 ITX.

This chip can be used without involving software.

Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20250822063959.692098-6-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../devicetree/bindings/display/bridge/simple-bridge.yaml        | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml b/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml
index 43cf4df9811a..421f99ca42d9 100644
--- a/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml
@@ -28,6 +28,7 @@ properties:
       - enum:
           - adi,adv7123
           - dumb-vga-dac
+          - radxa,ra620
           - ti,opa362
           - ti,ths8134
           - ti,ths8135

From 73b81fcb12df1051bd169e488245f8bf4c37bc93 Mon Sep 17 00:00:00 2001
From: Andy Yan <andy.yan@rock-chips.com>
Date: Fri, 22 Aug 2025 14:39:50 +0800
Subject: [PATCH 0572/1455] drm/bridge: simple-bridge: Add support for radxa
 ra620

The RA620 is an active DP to HDMI converter chip, basically
no software is involved to drive it.

Add it to simple bridge to make it can be find by the drm bridge chain.

Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250822063959.692098-7-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/simple-bridge.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/bridge/simple-bridge.c b/drivers/gpu/drm/bridge/simple-bridge.c
index 3d15ddd39470..1f16d568bcc4 100644
--- a/drivers/gpu/drm/bridge/simple-bridge.c
+++ b/drivers/gpu/drm/bridge/simple-bridge.c
@@ -261,6 +261,11 @@ static const struct of_device_id simple_bridge_match[] = {
 			.timings = &default_bridge_timings,
 			.connector_type = DRM_MODE_CONNECTOR_VGA,
 		},
+	}, {
+		.compatible = "radxa,ra620",
+		.data = &(const struct simple_bridge_info) {
+			.connector_type = DRM_MODE_CONNECTOR_HDMIA,
+		},
 	}, {
 		.compatible = "ti,opa362",
 		.data = &(const struct simple_bridge_info) {

From efeb036ffd2a0d1c1afba3d74bd5965825d1a646 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:17:56 -0700
Subject: [PATCH 0573/1455] drm/xe/psmi: Add GuC flag to enable PSMI

PSMI allows to capture data from the GPU useful for early
validation. From the kernel side there isn't much to be done, just a few
things:

	1) Toggle the feature support in GuC
	2) Enable some additional WAs
	3) Allocate buffers

Here is the first step, with the next ones to follow. For now everything
is disabled through a check in configfs that is currently hardcoded to
disabled.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-1-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.h | 2 ++
 drivers/gpu/drm/xe/xe_guc.c      | 7 ++++++-
 drivers/gpu/drm/xe/xe_guc_fwif.h | 1 +
 3 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
index fb8764008089..c14588b86e83 100644
--- a/drivers/gpu/drm/xe/xe_configfs.h
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -16,12 +16,14 @@ void xe_configfs_exit(void);
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
 void xe_configfs_clear_survivability_mode(struct pci_dev *pdev);
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
+static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
 #else
 static inline int xe_configfs_init(void) { return 0; }
 static inline void xe_configfs_exit(void) { }
 static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; }
 static inline void xe_configfs_clear_survivability_mode(struct pci_dev *pdev) { }
 static inline u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { return U64_MAX; }
+static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
 #endif
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 433abc787f7b..883f700e532a 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -16,6 +16,7 @@
 #include "regs/xe_guc_regs.h"
 #include "regs/xe_irq_regs.h"
 #include "xe_bo.h"
+#include "xe_configfs.h"
 #include "xe_device.h"
 #include "xe_force_wake.h"
 #include "xe_gt.h"
@@ -81,11 +82,15 @@ static u32 guc_ctl_debug_flags(struct xe_guc *guc)
 
 static u32 guc_ctl_feature_flags(struct xe_guc *guc)
 {
+	struct xe_device *xe = guc_to_xe(guc);
 	u32 flags = GUC_CTL_ENABLE_LITE_RESTORE;
 
-	if (!guc_to_xe(guc)->info.skip_guc_pc)
+	if (!xe->info.skip_guc_pc)
 		flags |= GUC_CTL_ENABLE_SLPC;
 
+	if (xe_configfs_get_psmi_enabled(to_pci_dev(xe->drm.dev)))
+		flags |= GUC_CTL_ENABLE_PSMI_LOGGING;
+
 	return flags;
 }
 
diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index ca9f999d38d1..400004f1838a 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -112,6 +112,7 @@ struct guc_update_exec_queue_policy {
 #define GUC_CTL_FEATURE			2
 #define   GUC_CTL_ENABLE_SLPC		BIT(2)
 #define   GUC_CTL_ENABLE_LITE_RESTORE	BIT(4)
+#define   GUC_CTL_ENABLE_PSMI_LOGGING	BIT(7)
 #define   GUC_CTL_DISABLE_SCHEDULER	BIT(14)
 
 #define GUC_CTL_DEBUG			3

From aaa0c1f50a3d3fc1624360eaeb7fe69af62f7107 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:17:57 -0700
Subject: [PATCH 0574/1455] drm/xe/psmi: Add debugfs interface for PSMI

Requirement for PSMI capture is to have a physically contiguous buffer.
All the needed configuration is done by the userspace tool directly to
the GPU via mmio access.

This interface only support allocating from VRAM regions. For integrated
devices, the PSMI buffer is in SYSTEM memory and should be allocated by
userspace using hugetlbfs.

Here we add the ability to allocate a region of physically contiguous
memory by writing to debugfs file (listed below). For multi-tile devices,
the capture tool requires ability to allocate a capture buffer per tile
(VRAM region) and so user can specify a region_mask. The tool then
can mmap the buffers via direct mmap of the PCIBAR via sysfs.

To support the capture tool, 3 new debugfs entries are added:

   psmi_capture_addr - physical address per VRAM region's capture buffer
   psmi_capture_region_mask - select which region(s) to allocate a buffer
   psmi_capture_size - size of current capture buffer

Writing psmi_capture_size will allocate new buffer of requested size per
region after freeing any current buffers.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Original-author: Brian Welty <brian.welty@intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com> # v2
Link: https://lore.kernel.org/r/20250821-psmi-v5-2-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/Makefile          |   1 +
 drivers/gpu/drm/xe/xe_debugfs.c      |   3 +
 drivers/gpu/drm/xe/xe_device.c       |   5 +
 drivers/gpu/drm/xe/xe_device_types.h |   8 +
 drivers/gpu/drm/xe/xe_psmi.c         | 306 +++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_psmi.h         |  14 ++
 6 files changed, 337 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_psmi.c
 create mode 100644 drivers/gpu/drm/xe/xe_psmi.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 8e0c3412a757..85b8d3a59ef0 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -98,6 +98,7 @@ xe-y += xe_bb.o \
 	xe_pcode.o \
 	xe_pm.o \
 	xe_preempt_fence.o \
+	xe_psmi.o \
 	xe_pt.o \
 	xe_pt_walk.o \
 	xe_pxp.o \
diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 0b4a532f7c45..bc717519502d 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -20,6 +20,7 @@
 #include "xe_guc_ads.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
+#include "xe_psmi.h"
 #include "xe_pxp_debugfs.h"
 #include "xe_sriov.h"
 #include "xe_sriov_pf.h"
@@ -400,6 +401,8 @@ void xe_debugfs_register(struct xe_device *xe)
 
 	xe_pxp_debugfs_register(xe->pxp);
 
+	xe_psmi_debugfs_register(xe);
+
 	fault_create_debugfs_attr("fail_gt_reset", root, &gt_reset_failure);
 
 	if (IS_SRIOV_PF(xe))
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 3e0402dff423..d4e186c6545a 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -54,6 +54,7 @@
 #include "xe_pcode.h"
 #include "xe_pm.h"
 #include "xe_pmu.h"
+#include "xe_psmi.h"
 #include "xe_pxp.h"
 #include "xe_query.h"
 #include "xe_shrinker.h"
@@ -908,6 +909,10 @@ int xe_device_probe(struct xe_device *xe)
 	if (err)
 		return err;
 
+	err = xe_psmi_init(xe);
+	if (err)
+		return err;
+
 	err = drm_dev_register(&xe->drm, 0);
 	if (err)
 		return err;
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 38c8329b4d2c..86506cdfc801 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -576,6 +576,14 @@ struct xe_device {
 	atomic64_t global_total_pages;
 #endif
 
+	/** @psmi: GPU debugging via additional validation HW */
+	struct {
+		/** @psmi.capture_obj: PSMI buffer for VRAM */
+		struct xe_bo *capture_obj[XE_MAX_TILES_PER_DEVICE + 1];
+		/** @psmi.region_mask: Mask of valid memory regions */
+		u8 region_mask;
+	} psmi;
+
 	/* private: */
 
 #if IS_ENABLED(CONFIG_DRM_XE_DISPLAY)
diff --git a/drivers/gpu/drm/xe/xe_psmi.c b/drivers/gpu/drm/xe/xe_psmi.c
new file mode 100644
index 000000000000..a2c9ff5bfd59
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_psmi.c
@@ -0,0 +1,306 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/debugfs.h>
+
+#include "xe_bo.h"
+#include "xe_device.h"
+#include "xe_configfs.h"
+#include "xe_psmi.h"
+
+/*
+ * PSMI capture support
+ *
+ * Requirement for PSMI capture is to have a physically contiguous buffer.  The
+ * PSMI tool owns doing all necessary configuration (MMIO register writes are
+ * done from user-space). However, KMD needs to provide the PSMI tool with the
+ * required physical address of the base of PSMI buffer in case of VRAM.
+ *
+ * VRAM backed PSMI buffer:
+ * Buffer is allocated as GEM object and with XE_BO_CREATE_PINNED_BIT flag which
+ * creates a contiguous allocation. The physical address is returned from
+ * psmi_debugfs_capture_addr_show(). PSMI tool can mmap the buffer via the
+ * PCIBAR through sysfs.
+ *
+ * SYSTEM memory backed PSMI buffer:
+ * Interface here does not support allocating from SYSTEM memory region.  The
+ * PSMI tool needs to allocate memory themselves using hugetlbfs. In order to
+ * get the physical address, user-space can query /proc/[pid]/pagemap. As an
+ * alternative, CMA debugfs could also be used to allocate reserved CMA memory.
+ */
+
+static bool psmi_enabled(struct xe_device *xe)
+{
+	return xe_configfs_get_psmi_enabled(to_pci_dev(xe->drm.dev));
+}
+
+static void psmi_free_object(struct xe_bo *bo)
+{
+	xe_bo_lock(bo, NULL);
+	xe_bo_unpin(bo);
+	xe_bo_unlock(bo);
+	xe_bo_put(bo);
+}
+
+/*
+ * Free PSMI capture buffer objects.
+ */
+static void psmi_cleanup(struct xe_device *xe)
+{
+	unsigned long id, region_mask = xe->psmi.region_mask;
+	struct xe_bo *bo;
+
+	for_each_set_bit(id, &region_mask,
+			 ARRAY_SIZE(xe->psmi.capture_obj)) {
+		/* smem should never be set */
+		xe_assert(xe, id);
+
+		bo = xe->psmi.capture_obj[id];
+		if (bo) {
+			psmi_free_object(bo);
+			xe->psmi.capture_obj[id] = NULL;
+		}
+	}
+}
+
+static struct xe_bo *psmi_alloc_object(struct xe_device *xe,
+				       unsigned int id, size_t bo_size)
+{
+	struct xe_bo *bo = NULL;
+	struct xe_tile *tile;
+	int err;
+
+	if (!id || !bo_size)
+		return NULL;
+
+	tile = &xe->tiles[id - 1];
+
+	/* VRAM: Allocate GEM object for the capture buffer */
+	bo = xe_bo_create_locked(xe, tile, NULL, bo_size,
+				 ttm_bo_type_kernel,
+				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+				 XE_BO_FLAG_PINNED |
+				 XE_BO_FLAG_PINNED_LATE_RESTORE |
+				 XE_BO_FLAG_NEEDS_CPU_ACCESS);
+
+	if (!IS_ERR(bo)) {
+		/* Buffer written by HW, ensure stays resident */
+		err = xe_bo_pin(bo);
+		if (err)
+			bo = ERR_PTR(err);
+		xe_bo_unlock(bo);
+	}
+
+	return bo;
+}
+
+/*
+ * Allocate PSMI capture buffer objects (via debugfs set function), based on
+ * which regions the user has selected in region_mask.  @size: size in bytes
+ * (should be power of 2)
+ *
+ * Always release/free the current buffer objects before attempting to allocate
+ * new ones.  Size == 0 will free all current buffers.
+ *
+ * Note, we don't write any registers as the capture tool is already configuring
+ * all PSMI registers itself via mmio space.
+ */
+static int psmi_resize_object(struct xe_device *xe, size_t size)
+{
+	unsigned long id, region_mask = xe->psmi.region_mask;
+	struct xe_bo *bo = NULL;
+	int err = 0;
+
+	/* if resizing, free currently allocated buffers first */
+	psmi_cleanup(xe);
+
+	/* can set size to 0, in which case, now done */
+	if (!size)
+		return 0;
+
+	for_each_set_bit(id, &region_mask,
+			 ARRAY_SIZE(xe->psmi.capture_obj)) {
+		/* smem should never be set */
+		xe_assert(xe, id);
+
+		bo = psmi_alloc_object(xe, id, size);
+		if (IS_ERR(bo)) {
+			err = PTR_ERR(bo);
+			break;
+		}
+		xe->psmi.capture_obj[id] = bo;
+
+		drm_info(&xe->drm,
+			 "PSMI capture size requested: %zu bytes, allocated: %lu:%zu\n",
+			 size, id, bo ? xe_bo_size(bo) : 0);
+	}
+
+	/* on error, reverse what was allocated */
+	if (err)
+		psmi_cleanup(xe);
+
+	return err;
+}
+
+/*
+ * Returns an address for the capture tool to use to find start of capture
+ * buffer. Capture tool requires the capability to have a buffer allocated per
+ * each tile (VRAM region), thus we return an address for each region.
+ */
+static int psmi_debugfs_capture_addr_show(struct seq_file *m, void *data)
+{
+	struct xe_device *xe = m->private;
+	unsigned long id, region_mask;
+	struct xe_bo *bo;
+	u64 val;
+
+	region_mask = xe->psmi.region_mask;
+	for_each_set_bit(id, &region_mask,
+			 ARRAY_SIZE(xe->psmi.capture_obj)) {
+		/* smem should never be set */
+		xe_assert(xe, id);
+
+		/* VRAM region */
+		bo = xe->psmi.capture_obj[id];
+		if (!bo)
+			continue;
+
+		/* pinned, so don't need bo_lock */
+		val = __xe_bo_addr(bo, 0, PAGE_SIZE);
+		seq_printf(m, "%ld: 0x%llx\n", id, val);
+	}
+
+	return 0;
+}
+
+/*
+ * Return capture buffer size, using the size from first allocated object that
+ * is found. This works because all objects must be of the same size.
+ */
+static int psmi_debugfs_capture_size_get(void *data, u64 *val)
+{
+	unsigned long id, region_mask;
+	struct xe_device *xe = data;
+	struct xe_bo *bo;
+
+	region_mask = xe->psmi.region_mask;
+	for_each_set_bit(id, &region_mask,
+			 ARRAY_SIZE(xe->psmi.capture_obj)) {
+		/* smem should never be set */
+		xe_assert(xe, id);
+
+		bo = xe->psmi.capture_obj[id];
+		if (bo) {
+			*val = xe_bo_size(bo);
+			return 0;
+		}
+	}
+
+	/* no capture objects are allocated */
+	*val = 0;
+
+	return 0;
+}
+
+/*
+ * Set size of PSMI capture buffer. This triggers the allocation of capture
+ * buffer in each memory region as specified with prior write to
+ * psmi_capture_region_mask.
+ */
+static int psmi_debugfs_capture_size_set(void *data, u64 val)
+{
+	struct xe_device *xe = data;
+
+	/* user must have specified at least one region */
+	if (!xe->psmi.region_mask)
+		return -EINVAL;
+
+	return psmi_resize_object(xe, val);
+}
+
+static int psmi_debugfs_capture_region_mask_get(void *data, u64 *val)
+{
+	struct xe_device *xe = data;
+
+	*val = xe->psmi.region_mask;
+
+	return 0;
+}
+
+/*
+ * Select VRAM regions for multi-tile devices, only allowed when buffer is not
+ * currently allocated.
+ */
+static int psmi_debugfs_capture_region_mask_set(void *data, u64 region_mask)
+{
+	struct xe_device *xe = data;
+	u64 size = 0;
+
+	/* SMEM is not supported (see comments at top of file) */
+	if (region_mask & 0x1)
+		return -EOPNOTSUPP;
+
+	/* input bitmask should contain only valid TTM regions */
+	if (!region_mask || region_mask & ~xe->info.mem_region_mask)
+		return -EINVAL;
+
+	/* only allow setting mask if buffer is not yet allocated */
+	psmi_debugfs_capture_size_get(xe, &size);
+	if (size)
+		return -EBUSY;
+
+	xe->psmi.region_mask = region_mask;
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(psmi_debugfs_capture_addr);
+
+DEFINE_DEBUGFS_ATTRIBUTE(psmi_debugfs_capture_region_mask_fops,
+			 psmi_debugfs_capture_region_mask_get,
+			 psmi_debugfs_capture_region_mask_set,
+			 "0x%llx\n");
+
+DEFINE_DEBUGFS_ATTRIBUTE(psmi_debugfs_capture_size_fops,
+			 psmi_debugfs_capture_size_get,
+			 psmi_debugfs_capture_size_set,
+			 "%lld\n");
+
+void xe_psmi_debugfs_register(struct xe_device *xe)
+{
+	struct drm_minor *minor;
+
+	if (!psmi_enabled(xe))
+		return;
+
+	minor = xe->drm.primary;
+	if (!minor->debugfs_root)
+		return;
+
+	debugfs_create_file("psmi_capture_addr",
+			    0400, minor->debugfs_root, xe,
+			    &psmi_debugfs_capture_addr_fops);
+
+	debugfs_create_file("psmi_capture_region_mask",
+			    0600, minor->debugfs_root, xe,
+			    &psmi_debugfs_capture_region_mask_fops);
+
+	debugfs_create_file("psmi_capture_size",
+			    0600, minor->debugfs_root, xe,
+			    &psmi_debugfs_capture_size_fops);
+}
+
+static void psmi_fini(void *arg)
+{
+	psmi_cleanup(arg);
+}
+
+int xe_psmi_init(struct xe_device *xe)
+{
+	if (!psmi_enabled(xe))
+		return 0;
+
+	return devm_add_action(xe->drm.dev, psmi_fini, xe);
+}
diff --git a/drivers/gpu/drm/xe/xe_psmi.h b/drivers/gpu/drm/xe/xe_psmi.h
new file mode 100644
index 000000000000..b1dfba80d893
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_psmi.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_PSMI_H_
+#define _XE_PSMI_H_
+
+struct xe_device;
+
+int xe_psmi_init(struct xe_device *xe);
+void xe_psmi_debugfs_register(struct xe_device *xe);
+
+#endif

From d67b1dfad00d7c28155a434b2a8a0012adc1b458 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:17:58 -0700
Subject: [PATCH 0575/1455] drm/xe/rtp: Add match for psmi

Add match to be used on WAs for only enabling workarounds if psmi is
intended to be used.

Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-3-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_rtp.c | 7 +++++++
 drivers/gpu/drm/xe/xe_rtp.h | 3 +++
 2 files changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_rtp.c b/drivers/gpu/drm/xe/xe_rtp.c
index 95571b87aa73..47ea1521dc80 100644
--- a/drivers/gpu/drm/xe/xe_rtp.c
+++ b/drivers/gpu/drm/xe/xe_rtp.c
@@ -9,6 +9,7 @@
 
 #include <uapi/drm/xe_drm.h>
 
+#include "xe_configfs.h"
 #include "xe_gt.h"
 #include "xe_gt_topology.h"
 #include "xe_macros.h"
@@ -363,3 +364,9 @@ bool xe_rtp_match_not_sriov_vf(const struct xe_gt *gt,
 {
 	return !IS_SRIOV_VF(gt_to_xe(gt));
 }
+
+bool xe_rtp_match_psmi_enabled(const struct xe_gt *gt,
+			       const struct xe_hw_engine *hwe)
+{
+	return xe_configfs_get_psmi_enabled(to_pci_dev(gt_to_xe(gt)->drm.dev));
+}
diff --git a/drivers/gpu/drm/xe/xe_rtp.h b/drivers/gpu/drm/xe/xe_rtp.h
index 5ed6c14b9ae3..7951fefdbe04 100644
--- a/drivers/gpu/drm/xe/xe_rtp.h
+++ b/drivers/gpu/drm/xe/xe_rtp.h
@@ -477,4 +477,7 @@ bool xe_rtp_match_first_render_or_compute(const struct xe_gt *gt,
 bool xe_rtp_match_not_sriov_vf(const struct xe_gt *gt,
 			       const struct xe_hw_engine *hwe);
 
+bool xe_rtp_match_psmi_enabled(const struct xe_gt *gt,
+			       const struct xe_hw_engine *hwe);
+
 #endif

From 29042df3acdc7364af1c251b2a05f7c1c8fe0401 Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Thu, 21 Aug 2025 09:17:59 -0700
Subject: [PATCH 0576/1455] drm/xe/psmi: Add Wa_14020001231

Enable Wa 14020001231 to block psmi interrupts during C6 entry exit
flow. It's only enabled if PSMI is enabled in runtime.

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-4-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/abi/guc_klvs_abi.h | 1 +
 drivers/gpu/drm/xe/xe_guc_ads.c       | 4 ++++
 drivers/gpu/drm/xe/xe_wa_oob.rules    | 4 ++++
 3 files changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
index 31dbfeee289e..0e78351c6ef5 100644
--- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
@@ -390,6 +390,7 @@ enum  {
  */
 enum xe_guc_klv_ids {
 	GUC_WORKAROUND_KLV_BLOCK_INTERRUPTS_WHEN_MGSR_BLOCKED				= 0x9002,
+	GUC_WORKAROUND_KLV_DISABLE_PSMI_INTERRUPTS_AT_C6_ENTRY_RESTORE_AT_EXIT		= 0x9004,
 	GUC_WORKAROUND_KLV_ID_GAM_PFQ_SHADOW_TAIL_POLLING				= 0x9005,
 	GUC_WORKAROUND_KLV_ID_DISABLE_MTP_DURING_ASYNC_COMPUTE				= 0x9007,
 	GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE			= 0x9008,
diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
index d7da67637079..5631722f34f5 100644
--- a/drivers/gpu/drm/xe/xe_guc_ads.c
+++ b/drivers/gpu/drm/xe/xe_guc_ads.c
@@ -359,6 +359,10 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 				 GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG);
 	}
 
+	if (XE_GT_WA(gt, 14020001231))
+		guc_waklv_enable(ads, NULL, 0, &offset, &remain,
+				 GUC_WORKAROUND_KLV_DISABLE_PSMI_INTERRUPTS_AT_C6_ENTRY_RESTORE_AT_EXIT);
+
 	size = guc_ads_waklv_size(ads) - remain;
 	if (!size)
 		return;
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 8d0aabab6777..303a5e05d993 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -68,6 +68,10 @@ no_media_l3	MEDIA_VERSION(3000)
 		MEDIA_VERSION_RANGE(1300, 3000)
 		MEDIA_VERSION(3002)
 		GRAPHICS_VERSION(3003)
+14020001231	GRAPHICS_VERSION_RANGE(2001,2004), FUNC(xe_rtp_match_psmi_enabled)
+		MEDIA_VERSION(2000), FUNC(xe_rtp_match_psmi_enabled)
+		MEDIA_VERSION(3000), FUNC(xe_rtp_match_psmi_enabled)
+		MEDIA_VERSION(3002), FUNC(xe_rtp_match_psmi_enabled)
 
 # SoC workaround - currently applies to all platforms with the following
 # primary GT GMDID

From 95b3899b4d25a49abf830ee1bc97940a7c1c7e97 Mon Sep 17 00:00:00 2001
From: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Date: Thu, 21 Aug 2025 09:18:00 -0700
Subject: [PATCH 0577/1455] drm/xe/psmi: Add Wa_16023683509

This WA ensures GuC will restore the media MCFG registers at C6
exit.

Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-5-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_guc.c        | 3 +++
 drivers/gpu/drm/xe/xe_guc_fwif.h   | 1 +
 drivers/gpu/drm/xe/xe_wa_oob.rules | 2 ++
 3 files changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 883f700e532a..37d06c51180c 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -219,6 +219,9 @@ static u32 guc_ctl_wa_flags(struct xe_guc *guc)
 	if (XE_GT_WA(gt, 14018913170))
 		flags |= GUC_WA_ENABLE_TSC_CHECK_ON_RC6;
 
+	if (XE_GT_WA(gt, 16023683509))
+		flags |= GUC_WA_SAVE_RESTORE_MCFG_REG_AT_MC6;
+
 	return flags;
 }
 
diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index 400004f1838a..0508f1064178 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -108,6 +108,7 @@ struct guc_update_exec_queue_policy {
 #define   GUC_WA_RENDER_RST_RC6_EXIT	BIT(19)
 #define   GUC_WA_RCS_REGS_IN_CCS_REGS_LIST	BIT(21)
 #define   GUC_WA_ENABLE_TSC_CHECK_ON_RC6	BIT(22)
+#define   GUC_WA_SAVE_RESTORE_MCFG_REG_AT_MC6	BIT(25)
 
 #define GUC_CTL_FEATURE			2
 #define   GUC_CTL_ENABLE_SLPC		BIT(2)
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 303a5e05d993..27cd19b98f8e 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -72,6 +72,8 @@ no_media_l3	MEDIA_VERSION(3000)
 		MEDIA_VERSION(2000), FUNC(xe_rtp_match_psmi_enabled)
 		MEDIA_VERSION(3000), FUNC(xe_rtp_match_psmi_enabled)
 		MEDIA_VERSION(3002), FUNC(xe_rtp_match_psmi_enabled)
+16023683509	MEDIA_VERSION(2000), FUNC(xe_rtp_match_psmi_enabled)
+		MEDIA_VERSION(3000), MEDIA_STEP(A0, B0), FUNC(xe_rtp_match_psmi_enabled)
 
 # SoC workaround - currently applies to all platforms with the following
 # primary GT GMDID

From 49245b49614870a0a135468be1bd0a6e0006a128 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:01 -0700
Subject: [PATCH 0578/1455] drm/xe/configfs: Simplify kernel doc

From the caller perspective reading the documentation, there's no need
to be so specific about everything the function is doing/checking. Just
document the functionality a caller cares about.

Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-6-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 853da2ee837a..17b1d6ae1ff6 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -389,10 +389,7 @@ static struct xe_config_group_device *find_xe_config_group_device(struct pci_dev
  * xe_configfs_get_survivability_mode - get configfs survivability mode attribute
  * @pdev: pci device
  *
- * find the configfs group that belongs to the pci device and return
- * the survivability mode attribute
- *
- * Return: survivability mode if config group is found, false otherwise
+ * Return: survivability_mode attribute in configfs
  */
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 {
@@ -409,11 +406,8 @@ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 }
 
 /**
- * xe_configfs_clear_survivability_mode - clear configfs survivability mode attribute
+ * xe_configfs_clear_survivability_mode - clear configfs survivability mode
  * @pdev: pci device
- *
- * find the configfs group that belongs to the pci device and clear survivability
- * mode attribute
  */
 void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
 {
@@ -433,10 +427,7 @@ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
  * xe_configfs_get_engines_allowed - get engine allowed mask from configfs
  * @pdev: pci device
  *
- * Find the configfs group that belongs to the pci device and return
- * the mask of engines allowed to be used.
- *
- * Return: engine mask with allowed engines
+ * Return: engine mask with allowed engines set in configfs
  */
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 {

From afe902848b417da81f60b3d15cbae40a67286dbf Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:02 -0700
Subject: [PATCH 0579/1455] drm/xe/configfs: Allow to enable PSMI

Now that additional WAs are in place and it's possible to allocate
buffers through debugfs, add the configfs attribute to turn PSMI on.

Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-7-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 66 ++++++++++++++++++++++++++++++--
 drivers/gpu/drm/xe/xe_configfs.h |  2 +-
 2 files changed, 63 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 17b1d6ae1ff6..61376dfc74bb 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -77,6 +77,16 @@
  * available for migrations, but it's disabled. This is intended for debugging
  * purposes only.
  *
+ * PSMI
+ * ----
+ *
+ * Enable extra debugging capabilities to trace engine execution. Only useful
+ * during early platform enabling and requires additional hardware connected.
+ * Once it's enabled, additionals WAs are added and runtime configuration is
+ * done via debugfs. Example to enable it::
+ *
+ *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/enable_psmi
+ *
  * Remove devices
  * ==============
  *
@@ -89,8 +99,9 @@ struct xe_config_group_device {
 	struct config_group group;
 
 	struct xe_config_device {
-		bool survivability_mode;
 		u64 engines_allowed;
+		bool survivability_mode;
+		bool enable_psmi;
 	} config;
 
 	/* protects attributes */
@@ -98,8 +109,9 @@ struct xe_config_group_device {
 };
 
 static const struct xe_config_device device_defaults = {
-	.survivability_mode = false,
 	.engines_allowed = U64_MAX,
+	.survivability_mode = false,
+	.enable_psmi = false,
 };
 
 static void set_device_defaults(struct xe_config_device *config)
@@ -243,12 +255,38 @@ static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 	return len;
 }
 
-CONFIGFS_ATTR(, survivability_mode);
+static ssize_t enable_psmi_show(struct config_item *item, char *page)
+{
+	struct xe_config_device *dev = to_xe_config_device(item);
+
+	return sprintf(page, "%d\n", dev->enable_psmi);
+}
+
+static ssize_t enable_psmi_store(struct config_item *item, const char *page, size_t len)
+{
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+	bool val;
+	int ret;
+
+	ret = kstrtobool(page, &val);
+	if (ret)
+		return ret;
+
+	mutex_lock(&dev->lock);
+	dev->config.enable_psmi = val;
+	mutex_unlock(&dev->lock);
+
+	return len;
+}
+
+CONFIGFS_ATTR(, enable_psmi);
 CONFIGFS_ATTR(, engines_allowed);
+CONFIGFS_ATTR(, survivability_mode);
 
 static struct configfs_attribute *xe_config_device_attrs[] = {
-	&attr_survivability_mode,
+	&attr_enable_psmi,
 	&attr_engines_allowed,
+	&attr_survivability_mode,
 	NULL,
 };
 
@@ -443,6 +481,26 @@ u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev)
 	return engines_allowed;
 }
 
+/**
+ * xe_configfs_get_psmi_enabled - get configfs enable_psmi setting
+ * @pdev: pci device
+ *
+ * Return: enable_psmi setting in configfs
+ */
+bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev)
+{
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
+	bool ret;
+
+	if (!dev)
+		return false;
+
+	ret = dev->config.enable_psmi;
+	config_item_put(&dev->group.cg_item);
+
+	return ret;
+}
+
 int __init xe_configfs_init(void)
 {
 	int ret;
diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
index c14588b86e83..603dd7796c8b 100644
--- a/drivers/gpu/drm/xe/xe_configfs.h
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -16,7 +16,7 @@ void xe_configfs_exit(void);
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
 void xe_configfs_clear_survivability_mode(struct pci_dev *pdev);
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
-static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
+bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev);
 #else
 static inline int xe_configfs_init(void) { return 0; }
 static inline void xe_configfs_exit(void) { }

From 734197a933cd2fb4adac7213fd1fc4df284cae14 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:03 -0700
Subject: [PATCH 0580/1455] drm/xe/configfs: Use guard() for dev->lock

Instead of the manual lock()/unlock() pattern, use guard() which will
make things easier for handling errors or early returns.

Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-8-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 61376dfc74bb..de73c4ff8f4f 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -5,6 +5,7 @@
 
 #include <linux/bitops.h>
 #include <linux/configfs.h>
+#include <linux/cleanup.h>
 #include <linux/find.h>
 #include <linux/init.h>
 #include <linux/module.h>
@@ -164,9 +165,8 @@ static ssize_t survivability_mode_store(struct config_item *item, const char *pa
 	if (ret)
 		return ret;
 
-	mutex_lock(&dev->lock);
+	guard(mutex)(&dev->lock);
 	dev->config.survivability_mode = survivability_mode;
-	mutex_unlock(&dev->lock);
 
 	return len;
 }
@@ -248,9 +248,8 @@ static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 		val |= mask;
 	}
 
-	mutex_lock(&dev->lock);
+	guard(mutex)(&dev->lock);
 	dev->config.engines_allowed = val;
-	mutex_unlock(&dev->lock);
 
 	return len;
 }
@@ -272,9 +271,8 @@ static ssize_t enable_psmi_store(struct config_item *item, const char *page, siz
 	if (ret)
 		return ret;
 
-	mutex_lock(&dev->lock);
+	guard(mutex)(&dev->lock);
 	dev->config.enable_psmi = val;
-	mutex_unlock(&dev->lock);
 
 	return len;
 }
@@ -454,9 +452,8 @@ void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
 	if (!dev)
 		return;
 
-	mutex_lock(&dev->lock);
+	guard(mutex)(&dev->lock);
 	dev->config.survivability_mode = 0;
-	mutex_unlock(&dev->lock);
 
 	config_group_put(&dev->group);
 }

From 3eb2280f6aa424bc99d59ff91ca741d6e240caee Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:04 -0700
Subject: [PATCH 0581/1455] drm/xe/configfs: Use tree-like output in
 documentation

When documenting the directories, use an output similar to the `tree`
command and add VFs and missing attributes.

Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-9-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index de73c4ff8f4f..25b47106413b 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -39,8 +39,16 @@
  * used to configure it::
  *
  *	/sys/kernel/config/xe/
- *		.. 0000:03:00.0/
- *			... survivability_mode
+ *	├── 0000:00:02.0
+ *	│   └── ...
+ *	├── 0000:00:02.1
+ *	│   └── ...
+ *	:
+ *	└── 0000:03:00.0
+ *	    ├── survivability_mode
+ *	    ├── engines_allowed
+ *	    └── enable_psmi
+ *
  *
  * Configure Attributes
  * ====================

From e2b33fce5eb08b37e38188b219a32c105cdc606e Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:05 -0700
Subject: [PATCH 0582/1455] drm/xe/configfs: Improve documentation steps

The steps are roughly:

1. Load the module without binding to the device
2. Configure the desired device
3. Bind the device

Move the binding part to the "Create devices" since it's not exclusive
to the survivability_mode attribute and better document the steps.

Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-10-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 25b47106413b..089e21fefcfd 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -29,11 +29,18 @@
  * See Documentation/filesystems/configfs.rst for more information about how configfs works.
  *
  * Create devices
- * ===============
+ * ==============
  *
- * In order to create a device, the user has to create a directory inside ``'xe'``::
+ * To create a device, the ``xe`` module should already be loaded, but some
+ * attributes can only be set before binding the device. It can be accomplished
+ * by blocking the driver autoprobe:
  *
- *	mkdir /sys/kernel/config/xe/0000:03:00.0/
+ *	# echo 0 > /sys/bus/pci/drivers_autoprobe
+ *	# modprobe xe
+ *
+ * In order to create a device, the user has to create a directory inside ``xe``::
+ *
+ *	# mkdir /sys/kernel/config/xe/0000:03:00.0/
  *
  * Every device created is populated by the driver with entries that can be
  * used to configure it::
@@ -49,6 +56,12 @@
  *	    ├── engines_allowed
  *	    └── enable_psmi
  *
+ * After configuring the attributes as per next section, the device can be
+ * probed with::
+ *
+ *	# echo 0000:03:00.0 > /sys/bus/pci/drivers/xe/bind
+ *	# # or
+ *	# echo 0000:03:00.0 > /sys/bus/pci/drivers_probe
  *
  * Configure Attributes
  * ====================
@@ -60,7 +73,6 @@
  * effect when probing the device. Example to enable it::
  *
  *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/survivability_mode
- *	# echo 0000:03:00.0 > /sys/bus/pci/drivers/xe/bind  (Enters survivability mode if supported)
  *
  * Allowed engines:
  * ----------------

From 66b21c338e6f05a980fae76c221d7aea4f1538c2 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:06 -0700
Subject: [PATCH 0583/1455] drm/xe/configfs: Minor fixes to documentation

Add a few missing punctuation and line breaks and make the syntax for
code snippets common to all of them.

Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-11-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 089e21fefcfd..26c5e1b29263 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -24,9 +24,10 @@
  * =========
  *
  * Configfs is a filesystem-based manager of kernel objects. XE KMD registers a
- * configfs subsystem called ``'xe'`` that creates a directory in the mounted configfs directory
- * The user can create devices under this directory and configure them as necessary
- * See Documentation/filesystems/configfs.rst for more information about how configfs works.
+ * configfs subsystem called ``xe`` that creates a directory in the mounted
+ * configfs directory. The user can create devices under this directory and
+ * configure them as necessary. See Documentation/filesystems/configfs.rst for
+ * more information about how configfs works.
  *
  * Create devices
  * ==============
@@ -113,7 +114,7 @@
  *
  * The created device directories can be removed using ``rmdir``::
  *
- *	rmdir /sys/kernel/config/xe/0000:03:00.0/
+ *	# rmdir /sys/kernel/config/xe/0000:03:00.0/
  */
 
 struct xe_config_group_device {

From 13dda74a164b570d93d412e812dc80549506410f Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 21 Aug 2025 09:18:07 -0700
Subject: [PATCH 0584/1455] drm/xe/configfs: Dump custom settings when binding

Device configuration using configfs could be prepared long time prior
the driver load. Currently all the xe configfs entries are for things
that are important to have in the log if a non-default value is being
used. Add a info-level message about that with the individual entries
that are different than the default.

Based on previous patch by Michal Wajdeczko.

Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250821-psmi-v5-12-34ab7550d3d8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 39 ++++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_configfs.h |  2 ++
 drivers/gpu/drm/xe/xe_pci.c      |  3 +++
 3 files changed, 44 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 26c5e1b29263..29a12d02443d 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -442,6 +442,45 @@ static struct xe_config_group_device *find_xe_config_group_device(struct pci_dev
 	return to_xe_config_group_device(item);
 }
 
+static void dump_custom_dev_config(struct pci_dev *pdev,
+				   struct xe_config_group_device *dev)
+{
+#define PRI_CUSTOM_ATTR(fmt_, attr_) do { \
+		if (dev->config.attr_ != device_defaults.attr_) \
+			pci_info(pdev, "configfs: " __stringify(attr_) " = " fmt_ "\n", \
+				 dev->config.attr_); \
+	} while (0)
+
+	PRI_CUSTOM_ATTR("%llx", engines_allowed);
+	PRI_CUSTOM_ATTR("%d", enable_psmi);
+	PRI_CUSTOM_ATTR("%d", survivability_mode);
+
+#undef PRI_CUSTOM_ATTR
+}
+
+/**
+ * xe_configfs_check_device() - Test if device was configured by configfs
+ * @pdev: the &pci_dev device to test
+ *
+ * Try to find the configfs group that belongs to the specified pci device
+ * and print a diagnostic message if different than the default value.
+ */
+void xe_configfs_check_device(struct pci_dev *pdev)
+{
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
+
+	if (!dev)
+		return;
+
+	/* memcmp here is safe as both are zero-initialized */
+	if (memcmp(&dev->config, &device_defaults, sizeof(dev->config))) {
+		pci_info(pdev, "Found custom settings in configfs\n");
+		dump_custom_dev_config(pdev, dev);
+	}
+
+	config_group_put(&dev->group);
+}
+
 /**
  * xe_configfs_get_survivability_mode - get configfs survivability mode attribute
  * @pdev: pci device
diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
index 603dd7796c8b..58c8c3164000 100644
--- a/drivers/gpu/drm/xe/xe_configfs.h
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -13,6 +13,7 @@ struct pci_dev;
 #if IS_ENABLED(CONFIG_CONFIGFS_FS)
 int xe_configfs_init(void);
 void xe_configfs_exit(void);
+void xe_configfs_check_device(struct pci_dev *pdev);
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
 void xe_configfs_clear_survivability_mode(struct pci_dev *pdev);
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
@@ -20,6 +21,7 @@ bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev);
 #else
 static inline int xe_configfs_init(void) { return 0; }
 static inline void xe_configfs_exit(void) { }
+static inline void xe_configfs_check_device(struct pci_dev *pdev) { }
 static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; }
 static inline void xe_configfs_clear_survivability_mode(struct pci_dev *pdev) { }
 static inline u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { return U64_MAX; }
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 419e29cebbba..a8bab19111c3 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -18,6 +18,7 @@
 #include "display/xe_display.h"
 #include "regs/xe_gt_regs.h"
 #include "regs/xe_regs.h"
+#include "xe_configfs.h"
 #include "xe_device.h"
 #include "xe_drv.h"
 #include "xe_gt.h"
@@ -816,6 +817,8 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct xe_device *xe;
 	int err;
 
+	xe_configfs_check_device(pdev);
+
 	if (desc->require_force_probe && !id_forced(pdev->device)) {
 		dev_info(&pdev->dev,
 			 "Your graphics device %04x is not officially supported\n"

From 6c66eba502709a78281333187c1add7b71f7201f Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:48:54 +0200
Subject: [PATCH 0585/1455] drm/panel: ilitek-ili9881c: turn off power-supply
 when init fails

The prepare function turns on the power-supply regulator first, when
preparing the display. But in an error case, just returns the error
code, but does not power off the regulator again, fix that.

Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-2-heiko@sntech.de
---
 drivers/gpu/drm/panel/panel-ilitek-ili9881c.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
index ac433345a179..3af22a5f5700 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
@@ -1486,7 +1486,7 @@ static int ili9881c_prepare(struct drm_panel *panel)
 						      instr->arg.cmd.data);
 
 		if (ret)
-			return ret;
+			goto disable_power;
 	}
 
 	ret = ili9881c_switch_page(ctx, 0);
@@ -1498,18 +1498,22 @@ static int ili9881c_prepare(struct drm_panel *panel)
 					 &ctx->address_mode,
 					 sizeof(ctx->address_mode));
 		if (ret < 0)
-			return ret;
+			goto disable_power;
 	}
 
 	ret = mipi_dsi_dcs_set_tear_on(ctx->dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
 	if (ret)
-		return ret;
+		goto disable_power;
 
 	ret = mipi_dsi_dcs_exit_sleep_mode(ctx->dsi);
 	if (ret)
-		return ret;
+		goto disable_power;
 
 	return 0;
+
+disable_power:
+	regulator_disable(ctx->power);
+	return ret;
 }
 
 static int ili9881c_enable(struct drm_panel *panel)

From 5efa82492066fcb32308210fb3f0b752af74334f Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:48:55 +0200
Subject: [PATCH 0586/1455] drm/panel: ilitek-ili9881c: move display_on/_off
 dcs calls to (un-)prepare

At least for panel-bridges, the atomic_enable call is defined as being
called right after the preceding element in the display pipe is enabled.

It is also stated that "The bridge can assume that the display pipe (i.e.
clocks and timing signals) feeding it is running when this callback is
called"

This means the DSI controller driving this display would have already
switched over to video-mode from command mode and thus dcs functions
should not be called anymore at this point.

This caused a non-working display for me, when trying to enable
the rk3576 dsi controller using a display using this controller.

Therefore move the display_on/off calls the more appropriate
prepare/unprepare callbacks.

Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-3-heiko@sntech.de
---
 drivers/gpu/drm/panel/panel-ilitek-ili9881c.c | 27 +++++--------------
 1 file changed, 7 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
index 3af22a5f5700..7ed65d6762d8 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
@@ -1509,6 +1509,12 @@ static int ili9881c_prepare(struct drm_panel *panel)
 	if (ret)
 		goto disable_power;
 
+	msleep(120);
+
+	ret = mipi_dsi_dcs_set_display_on(ctx->dsi);
+	if (ret)
+		goto disable_power;
+
 	return 0;
 
 disable_power:
@@ -1516,28 +1522,11 @@ disable_power:
 	return ret;
 }
 
-static int ili9881c_enable(struct drm_panel *panel)
-{
-	struct ili9881c *ctx = panel_to_ili9881c(panel);
-
-	msleep(120);
-
-	mipi_dsi_dcs_set_display_on(ctx->dsi);
-
-	return 0;
-}
-
-static int ili9881c_disable(struct drm_panel *panel)
-{
-	struct ili9881c *ctx = panel_to_ili9881c(panel);
-
-	return mipi_dsi_dcs_set_display_off(ctx->dsi);
-}
-
 static int ili9881c_unprepare(struct drm_panel *panel)
 {
 	struct ili9881c *ctx = panel_to_ili9881c(panel);
 
+	mipi_dsi_dcs_set_display_off(ctx->dsi);
 	mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
 	regulator_disable(ctx->power);
 	gpiod_set_value_cansleep(ctx->reset, 1);
@@ -1710,8 +1699,6 @@ static enum drm_panel_orientation ili9881c_get_orientation(struct drm_panel *pan
 static const struct drm_panel_funcs ili9881c_funcs = {
 	.prepare	= ili9881c_prepare,
 	.unprepare	= ili9881c_unprepare,
-	.enable		= ili9881c_enable,
-	.disable	= ili9881c_disable,
 	.get_modes	= ili9881c_get_modes,
 	.get_orientation = ili9881c_get_orientation,
 };

From 9002f55ee4480f23bd2ce91fb5fca536ce31717e Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:48:56 +0200
Subject: [PATCH 0587/1455] drm/panel: ilitek-ili9881c: convert (un-)prepare to
 mipi_dsi_multi_context

This saves some lines for error handling.

For the address mode change, that value is always an u8, so switching
to dcs_write_buffer function should be appropriate.

Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-4-heiko@sntech.de
---
 drivers/gpu/drm/panel/panel-ilitek-ili9881c.c | 66 ++++++-------------
 1 file changed, 21 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
index 7ed65d6762d8..8b208f5701f6 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
@@ -1433,33 +1433,24 @@ static inline struct ili9881c *panel_to_ili9881c(struct drm_panel *panel)
  * So before any attempt at sending a command or data, we have to be
  * sure if we're in the right page or not.
  */
-static int ili9881c_switch_page(struct ili9881c *ctx, u8 page)
+static void ili9881c_switch_page(struct mipi_dsi_multi_context *mctx, u8 page)
 {
 	u8 buf[4] = { 0xff, 0x98, 0x81, page };
-	int ret;
 
-	ret = mipi_dsi_dcs_write_buffer(ctx->dsi, buf, sizeof(buf));
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	mipi_dsi_dcs_write_buffer_multi(mctx, buf, sizeof(buf));
 }
 
-static int ili9881c_send_cmd_data(struct ili9881c *ctx, u8 cmd, u8 data)
+static void ili9881c_send_cmd_data(struct mipi_dsi_multi_context *mctx, u8 cmd, u8 data)
 {
 	u8 buf[2] = { cmd, data };
-	int ret;
 
-	ret = mipi_dsi_dcs_write_buffer(ctx->dsi, buf, sizeof(buf));
-	if (ret < 0)
-		return ret;
-
-	return 0;
+	mipi_dsi_dcs_write_buffer_multi(mctx, buf, sizeof(buf));
 }
 
 static int ili9881c_prepare(struct drm_panel *panel)
 {
 	struct ili9881c *ctx = panel_to_ili9881c(panel);
+	struct mipi_dsi_multi_context mctx = { .dsi = ctx->dsi };
 	unsigned int i;
 	int ret;
 
@@ -1480,54 +1471,39 @@ static int ili9881c_prepare(struct drm_panel *panel)
 		const struct ili9881c_instr *instr = &ctx->desc->init[i];
 
 		if (instr->op == ILI9881C_SWITCH_PAGE)
-			ret = ili9881c_switch_page(ctx, instr->arg.page);
+			ili9881c_switch_page(&mctx, instr->arg.page);
 		else if (instr->op == ILI9881C_COMMAND)
-			ret = ili9881c_send_cmd_data(ctx, instr->arg.cmd.cmd,
-						      instr->arg.cmd.data);
-
-		if (ret)
-			goto disable_power;
+			ili9881c_send_cmd_data(&mctx, instr->arg.cmd.cmd,
+					       instr->arg.cmd.data);
 	}
 
-	ret = ili9881c_switch_page(ctx, 0);
-	if (ret)
-		return ret;
+	ili9881c_switch_page(&mctx, 0);
 
-	if (ctx->address_mode) {
-		ret = mipi_dsi_dcs_write(ctx->dsi, MIPI_DCS_SET_ADDRESS_MODE,
-					 &ctx->address_mode,
-					 sizeof(ctx->address_mode));
-		if (ret < 0)
-			goto disable_power;
-	}
+	if (ctx->address_mode)
+		ili9881c_send_cmd_data(&mctx, MIPI_DCS_SET_ADDRESS_MODE,
+				       ctx->address_mode);
 
-	ret = mipi_dsi_dcs_set_tear_on(ctx->dsi, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
-	if (ret)
-		goto disable_power;
-
-	ret = mipi_dsi_dcs_exit_sleep_mode(ctx->dsi);
-	if (ret)
-		goto disable_power;
-
-	msleep(120);
-
-	ret = mipi_dsi_dcs_set_display_on(ctx->dsi);
-	if (ret)
+	mipi_dsi_dcs_set_tear_on_multi(&mctx, MIPI_DSI_DCS_TEAR_MODE_VBLANK);
+	mipi_dsi_dcs_exit_sleep_mode_multi(&mctx);
+	mipi_dsi_msleep(&mctx, 120);
+	mipi_dsi_dcs_set_display_on_multi(&mctx);
+	if (mctx.accum_err)
 		goto disable_power;
 
 	return 0;
 
 disable_power:
 	regulator_disable(ctx->power);
-	return ret;
+	return mctx.accum_err;
 }
 
 static int ili9881c_unprepare(struct drm_panel *panel)
 {
 	struct ili9881c *ctx = panel_to_ili9881c(panel);
+	struct mipi_dsi_multi_context mctx = { .dsi = ctx->dsi };
 
-	mipi_dsi_dcs_set_display_off(ctx->dsi);
-	mipi_dsi_dcs_enter_sleep_mode(ctx->dsi);
+	mipi_dsi_dcs_set_display_off_multi(&mctx);
+	mipi_dsi_dcs_enter_sleep_mode_multi(&mctx);
 	regulator_disable(ctx->power);
 	gpiod_set_value_cansleep(ctx->reset, 1);
 

From 8e484ff181b177ea2e86e537fd7a5c8f9d7532ad Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:48:57 +0200
Subject: [PATCH 0588/1455] dt-bindings: vendor-prefixes: Add prefix for
 Shenzhen Bestar Electronic

Add the prefix for Bestar, named in full both on Panelook.com and their
display datasheets as
  Shenzhen Bestar Electronic Technology Co., Ltd.
which produces at least DSI displays and maybe more.

Acked-by: "Rob Herring (Arm)" <robh@kernel.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-5-heiko@sntech.de
---
 Documentation/devicetree/bindings/vendor-prefixes.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml
index 77160cd47f54..abc43c43813a 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.yaml
+++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml
@@ -221,6 +221,8 @@ patternProperties:
     description: BeagleBoard.org Foundation
   "^belling,.*":
     description: Shanghai Belling Co., Ltd.
+  "^bestar,.*":
+    description: Shenzhen Bestar Electronic Technology Co., Ltd.
   "^bhf,.*":
     description: Beckhoff Automation GmbH & Co. KG
   "^bigtreetech,.*":

From 157ba8c6742931a617a34555c27dd10b36385cf3 Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:48:58 +0200
Subject: [PATCH 0589/1455] dt-bindings: display: ili9881c: Add Bestar
 BSD1218-A101KL68 LCD panel

Document the compatible value for Bestar BSD1218-A101KL68 LCD panels
that are based around the ili9881c controller.

Acked-by: "Rob Herring (Arm)" <robh@kernel.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-6-heiko@sntech.de
---
 .../devicetree/bindings/display/panel/ilitek,ili9881c.yaml       | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml b/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml
index a51af61d4846..434cc6af9c95 100644
--- a/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml
+++ b/Documentation/devicetree/bindings/display/panel/ilitek,ili9881c.yaml
@@ -18,6 +18,7 @@ properties:
       - enum:
           - ampire,am8001280g
           - bananapi,lhr050h41
+          - bestar,bsd1218-a101kl68
           - feixin,k101-im2byl02
           - raspberrypi,dsi-7inch
           - startek,kd050hdfia020

From 38dbbbb41c515d72b0a82fde3cec450d068f5e94 Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:48:59 +0200
Subject: [PATCH 0590/1455] drm/panel: ilitek-ili9881c: Add Bestar
 BSD1218-A101KL68 support

Add support for the Bestar BSD1218-A101KL68 panel.

The init table comes from the Firefly BSP ([1]).

[1] https://gitlab.com/firefly-linux/kernel/-/blob/rk3588/firefly/arch/arm64/boot/dts/rockchip/aio-3588l-mipi101-BSD1218-A101KL68.dtsi

Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-7-heiko@sntech.de
---
 drivers/gpu/drm/panel/panel-ilitek-ili9881c.c | 221 ++++++++++++++++++
 1 file changed, 221 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
index 8b208f5701f6..ad4993b2f92a 100644
--- a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
+++ b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c
@@ -1417,6 +1417,200 @@ static const struct ili9881c_instr rpi_7inch_init[] = {
 	ILI9881C_COMMAND_INSTR(0xD3, 0x39),
 };
 
+static const struct ili9881c_instr bsd1218_a101kl68_init[] = {
+	ILI9881C_SWITCH_PAGE_INSTR(3),
+	ILI9881C_COMMAND_INSTR(0x01, 0x00),
+	ILI9881C_COMMAND_INSTR(0x02, 0x00),
+	ILI9881C_COMMAND_INSTR(0x03, 0x55),
+	ILI9881C_COMMAND_INSTR(0x04, 0x55),
+	ILI9881C_COMMAND_INSTR(0x05, 0x03),
+	ILI9881C_COMMAND_INSTR(0x06, 0x06),
+	ILI9881C_COMMAND_INSTR(0x07, 0x00),
+	ILI9881C_COMMAND_INSTR(0x08, 0x07),
+	ILI9881C_COMMAND_INSTR(0x09, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0a, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0b, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0c, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0d, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0e, 0x00),
+	ILI9881C_COMMAND_INSTR(0x0f, 0x00),
+	ILI9881C_COMMAND_INSTR(0x10, 0x00),
+	ILI9881C_COMMAND_INSTR(0x11, 0x00),
+	ILI9881C_COMMAND_INSTR(0x12, 0x00),
+	ILI9881C_COMMAND_INSTR(0x13, 0x00),
+	ILI9881C_COMMAND_INSTR(0x14, 0x00),
+	ILI9881C_COMMAND_INSTR(0x15, 0x00),
+	ILI9881C_COMMAND_INSTR(0x16, 0x00),
+	ILI9881C_COMMAND_INSTR(0x17, 0x00),
+	ILI9881C_COMMAND_INSTR(0x18, 0x00),
+	ILI9881C_COMMAND_INSTR(0x19, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1a, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1b, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1c, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1d, 0x00),
+	ILI9881C_COMMAND_INSTR(0x1e, 0xc0),
+	ILI9881C_COMMAND_INSTR(0x1f, 0x80),
+	ILI9881C_COMMAND_INSTR(0x20, 0x04),
+	ILI9881C_COMMAND_INSTR(0x21, 0x03),
+	ILI9881C_COMMAND_INSTR(0x22, 0x00),
+	ILI9881C_COMMAND_INSTR(0x23, 0x00),
+	ILI9881C_COMMAND_INSTR(0x24, 0x00),
+	ILI9881C_COMMAND_INSTR(0x25, 0x00),
+	ILI9881C_COMMAND_INSTR(0x26, 0x00),
+	ILI9881C_COMMAND_INSTR(0x27, 0x00),
+	ILI9881C_COMMAND_INSTR(0x28, 0x33),
+	ILI9881C_COMMAND_INSTR(0x29, 0x33),
+	ILI9881C_COMMAND_INSTR(0x2a, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2b, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2c, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2d, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2e, 0x00),
+	ILI9881C_COMMAND_INSTR(0x2f, 0x00),
+	ILI9881C_COMMAND_INSTR(0x30, 0x00),
+	ILI9881C_COMMAND_INSTR(0x31, 0x00),
+	ILI9881C_COMMAND_INSTR(0x32, 0x00),
+	ILI9881C_COMMAND_INSTR(0x33, 0x00),
+	ILI9881C_COMMAND_INSTR(0x34, 0x04),
+	ILI9881C_COMMAND_INSTR(0x35, 0x00),
+	ILI9881C_COMMAND_INSTR(0x36, 0x00),
+	ILI9881C_COMMAND_INSTR(0x37, 0x00),
+	ILI9881C_COMMAND_INSTR(0x38, 0x3c),
+	ILI9881C_COMMAND_INSTR(0x39, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3a, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3b, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3c, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3d, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3e, 0x00),
+	ILI9881C_COMMAND_INSTR(0x3f, 0x00),
+	ILI9881C_COMMAND_INSTR(0x40, 0x00),
+	ILI9881C_COMMAND_INSTR(0x41, 0x00),
+	ILI9881C_COMMAND_INSTR(0x42, 0x00),
+	ILI9881C_COMMAND_INSTR(0x43, 0x00),
+	ILI9881C_COMMAND_INSTR(0x44, 0x00),
+	ILI9881C_COMMAND_INSTR(0x50, 0x00),
+	ILI9881C_COMMAND_INSTR(0x51, 0x11),
+	ILI9881C_COMMAND_INSTR(0x52, 0x44),
+	ILI9881C_COMMAND_INSTR(0x53, 0x55),
+	ILI9881C_COMMAND_INSTR(0x54, 0x88),
+	ILI9881C_COMMAND_INSTR(0x55, 0xab),
+	ILI9881C_COMMAND_INSTR(0x56, 0x00),
+	ILI9881C_COMMAND_INSTR(0x57, 0x11),
+	ILI9881C_COMMAND_INSTR(0x58, 0x22),
+	ILI9881C_COMMAND_INSTR(0x59, 0x33),
+	ILI9881C_COMMAND_INSTR(0x5a, 0x44),
+	ILI9881C_COMMAND_INSTR(0x5b, 0x55),
+	ILI9881C_COMMAND_INSTR(0x5c, 0x66),
+	ILI9881C_COMMAND_INSTR(0x5d, 0x77),
+	ILI9881C_COMMAND_INSTR(0x5e, 0x00),
+	ILI9881C_COMMAND_INSTR(0x5f, 0x02),
+	ILI9881C_COMMAND_INSTR(0x60, 0x02),
+	ILI9881C_COMMAND_INSTR(0x61, 0x0a),
+	ILI9881C_COMMAND_INSTR(0x62, 0x09),
+	ILI9881C_COMMAND_INSTR(0x63, 0x08),
+	ILI9881C_COMMAND_INSTR(0x64, 0x13),
+	ILI9881C_COMMAND_INSTR(0x65, 0x12),
+	ILI9881C_COMMAND_INSTR(0x66, 0x11),
+	ILI9881C_COMMAND_INSTR(0x67, 0x10),
+	ILI9881C_COMMAND_INSTR(0x68, 0x0f),
+	ILI9881C_COMMAND_INSTR(0x69, 0x0e),
+	ILI9881C_COMMAND_INSTR(0x6a, 0x0d),
+	ILI9881C_COMMAND_INSTR(0x6b, 0x0c),
+	ILI9881C_COMMAND_INSTR(0x6c, 0x06),
+	ILI9881C_COMMAND_INSTR(0x6d, 0x07),
+	ILI9881C_COMMAND_INSTR(0x6e, 0x02),
+	ILI9881C_COMMAND_INSTR(0x6f, 0x02),
+	ILI9881C_COMMAND_INSTR(0x70, 0x02),
+	ILI9881C_COMMAND_INSTR(0x71, 0x02),
+	ILI9881C_COMMAND_INSTR(0x72, 0x02),
+	ILI9881C_COMMAND_INSTR(0x73, 0x02),
+	ILI9881C_COMMAND_INSTR(0x74, 0x02),
+	ILI9881C_COMMAND_INSTR(0x75, 0x02),
+	ILI9881C_COMMAND_INSTR(0x76, 0x02),
+	ILI9881C_COMMAND_INSTR(0x77, 0x0a),
+	ILI9881C_COMMAND_INSTR(0x78, 0x06),
+	ILI9881C_COMMAND_INSTR(0x79, 0x07),
+	ILI9881C_COMMAND_INSTR(0x7a, 0x10),
+	ILI9881C_COMMAND_INSTR(0x7b, 0x11),
+	ILI9881C_COMMAND_INSTR(0x7c, 0x12),
+	ILI9881C_COMMAND_INSTR(0x7d, 0x13),
+	ILI9881C_COMMAND_INSTR(0x7e, 0x0c),
+	ILI9881C_COMMAND_INSTR(0x7f, 0x0d),
+	ILI9881C_COMMAND_INSTR(0x80, 0x0e),
+	ILI9881C_COMMAND_INSTR(0x81, 0x0f),
+	ILI9881C_COMMAND_INSTR(0x82, 0x09),
+	ILI9881C_COMMAND_INSTR(0x83, 0x08),
+	ILI9881C_COMMAND_INSTR(0x84, 0x02),
+	ILI9881C_COMMAND_INSTR(0x85, 0x02),
+	ILI9881C_COMMAND_INSTR(0x86, 0x02),
+	ILI9881C_COMMAND_INSTR(0x87, 0x02),
+	ILI9881C_COMMAND_INSTR(0x88, 0x02),
+	ILI9881C_COMMAND_INSTR(0x89, 0x02),
+	ILI9881C_COMMAND_INSTR(0x8a, 0x02),
+
+	ILI9881C_SWITCH_PAGE_INSTR(4),
+	ILI9881C_COMMAND_INSTR(0x6e, 0x2a),
+	ILI9881C_COMMAND_INSTR(0x6f, 0x37),
+	ILI9881C_COMMAND_INSTR(0x3a, 0x24),
+	ILI9881C_COMMAND_INSTR(0x8d, 0x19),
+	ILI9881C_COMMAND_INSTR(0x87, 0xba),
+	ILI9881C_COMMAND_INSTR(0xb2, 0xd1),
+	ILI9881C_COMMAND_INSTR(0x88, 0x0b),
+	ILI9881C_COMMAND_INSTR(0x38, 0x01),
+	ILI9881C_COMMAND_INSTR(0x39, 0x00),
+	ILI9881C_COMMAND_INSTR(0xb5, 0x02),
+	ILI9881C_COMMAND_INSTR(0x31, 0x25),
+	ILI9881C_COMMAND_INSTR(0x3b, 0x98),
+
+	ILI9881C_SWITCH_PAGE_INSTR(1),
+	ILI9881C_COMMAND_INSTR(0x22, 0x0a),
+	ILI9881C_COMMAND_INSTR(0x31, 0x0c),
+	ILI9881C_COMMAND_INSTR(0x53, 0x40),
+	ILI9881C_COMMAND_INSTR(0x55, 0x45),
+	ILI9881C_COMMAND_INSTR(0x50, 0xb7),
+	ILI9881C_COMMAND_INSTR(0x51, 0xb2),
+	ILI9881C_COMMAND_INSTR(0x60, 0x07),
+	ILI9881C_COMMAND_INSTR(0xa0, 0x22),
+	ILI9881C_COMMAND_INSTR(0xa1, 0x3f),
+	ILI9881C_COMMAND_INSTR(0xa2, 0x4e),
+	ILI9881C_COMMAND_INSTR(0xa3, 0x17),
+	ILI9881C_COMMAND_INSTR(0xa4, 0x1a),
+	ILI9881C_COMMAND_INSTR(0xa5, 0x2d),
+	ILI9881C_COMMAND_INSTR(0xa6, 0x21),
+	ILI9881C_COMMAND_INSTR(0xa7, 0x22),
+	ILI9881C_COMMAND_INSTR(0xa8, 0xc4),
+	ILI9881C_COMMAND_INSTR(0xa9, 0x1b),
+	ILI9881C_COMMAND_INSTR(0xaa, 0x25),
+	ILI9881C_COMMAND_INSTR(0xab, 0xa7),
+	ILI9881C_COMMAND_INSTR(0xac, 0x1a),
+	ILI9881C_COMMAND_INSTR(0xad, 0x19),
+	ILI9881C_COMMAND_INSTR(0xae, 0x4b),
+	ILI9881C_COMMAND_INSTR(0xaf, 0x1f),
+	ILI9881C_COMMAND_INSTR(0xb0, 0x2a),
+	ILI9881C_COMMAND_INSTR(0xb1, 0x59),
+	ILI9881C_COMMAND_INSTR(0xb2, 0x64),
+	ILI9881C_COMMAND_INSTR(0xb3, 0x3f),
+	ILI9881C_COMMAND_INSTR(0xc0, 0x22),
+	ILI9881C_COMMAND_INSTR(0xc1, 0x48),
+	ILI9881C_COMMAND_INSTR(0xc2, 0x59),
+	ILI9881C_COMMAND_INSTR(0xc3, 0x15),
+	ILI9881C_COMMAND_INSTR(0xc4, 0x15),
+	ILI9881C_COMMAND_INSTR(0xc5, 0x28),
+	ILI9881C_COMMAND_INSTR(0xc6, 0x1c),
+	ILI9881C_COMMAND_INSTR(0xc7, 0x1e),
+	ILI9881C_COMMAND_INSTR(0xc8, 0xc4),
+	ILI9881C_COMMAND_INSTR(0xc9, 0x1c),
+	ILI9881C_COMMAND_INSTR(0xca, 0x2b),
+	ILI9881C_COMMAND_INSTR(0xcb, 0xa3),
+	ILI9881C_COMMAND_INSTR(0xcc, 0x1f),
+	ILI9881C_COMMAND_INSTR(0xcd, 0x1e),
+	ILI9881C_COMMAND_INSTR(0xce, 0x52),
+	ILI9881C_COMMAND_INSTR(0xcf, 0x24),
+	ILI9881C_COMMAND_INSTR(0xd0, 0x2a),
+	ILI9881C_COMMAND_INSTR(0xd1, 0x58),
+	ILI9881C_COMMAND_INSTR(0xd2, 0x68),
+	ILI9881C_COMMAND_INSTR(0xd3, 0x3f),
+};
+
 static inline struct ili9881c *panel_to_ili9881c(struct drm_panel *panel)
 {
 	return container_of(panel, struct ili9881c, panel);
@@ -1629,6 +1823,23 @@ static const struct drm_display_mode rpi_7inch_default_mode = {
 	.height_mm	= 151,
 };
 
+static const struct drm_display_mode bsd1218_a101kl68_default_mode = {
+	.clock		= 70000,
+
+	.hdisplay	= 800,
+	.hsync_start	= 800 + 40,
+	.hsync_end	= 800 + 40 + 20,
+	.htotal		= 800 + 40 + 20 + 20,
+
+	.vdisplay	= 1280,
+	.vsync_start	= 1280 + 20,
+	.vsync_end	= 1280 + 20 + 4,
+	.vtotal		= 1280 + 20 + 4 + 20,
+
+	.width_mm	= 120,
+	.height_mm	= 170,
+};
+
 static int ili9881c_get_modes(struct drm_panel *panel,
 			      struct drm_connector *connector)
 {
@@ -1797,8 +2008,18 @@ static const struct ili9881c_desc rpi_7inch_desc = {
 	.lanes = 2,
 };
 
+static const struct ili9881c_desc bsd1218_a101kl68_desc = {
+	.init = bsd1218_a101kl68_init,
+	.init_length = ARRAY_SIZE(bsd1218_a101kl68_init),
+	.mode = &bsd1218_a101kl68_default_mode,
+	.mode_flags = MIPI_DSI_MODE_VIDEO | MIPI_DSI_MODE_VIDEO_BURST |
+		      MIPI_DSI_MODE_LPM | MIPI_DSI_MODE_NO_EOT_PACKET,
+	.lanes = 4,
+};
+
 static const struct of_device_id ili9881c_of_match[] = {
 	{ .compatible = "bananapi,lhr050h41", .data = &lhr050h41_desc },
+	{ .compatible = "bestar,bsd1218-a101kl68", .data = &bsd1218_a101kl68_desc },
 	{ .compatible = "feixin,k101-im2byl02", .data = &k101_im2byl02_desc },
 	{ .compatible = "startek,kd050hdfia020", .data = &kd050hdfia020_desc },
 	{ .compatible = "tdo,tl050hdv35", .data = &tl050hdv35_desc },

From f05530bdaf42aa0e6bb4cde76ba6a081cf473d44 Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 7 Jul 2025 18:49:01 +0200
Subject: [PATCH 0591/1455] dt-bindings: display: rockchip: Add rk3576 to
 RK3588 DW DSI2 controller schema

The rk3576 controller is based on the same newer Synopsis IP as the one
found in the rk3588.

Its external setting bits in the GRF are different though, so it needs
its own distinct compatible.

Acked-by: "Rob Herring (Arm)" <robh@kernel.org>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-9-heiko@sntech.de
---
 .../bindings/display/rockchip/rockchip,rk3588-mipi-dsi2.yaml     | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-mipi-dsi2.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-mipi-dsi2.yaml
index 53384e47b507..75cd1c13fa52 100644
--- a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-mipi-dsi2.yaml
+++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-mipi-dsi2.yaml
@@ -12,6 +12,7 @@ maintainers:
 properties:
   compatible:
     enum:
+      - rockchip,rk3576-mipi-dsi2
       - rockchip,rk3588-mipi-dsi2
 
   reg:

From b6f11f114759a088acf44e86b5cd72f24de85d44 Mon Sep 17 00:00:00 2001
From: Guochun Huang <hero.huang@rock-chips.com>
Date: Mon, 7 Jul 2025 18:49:02 +0200
Subject: [PATCH 0592/1455] drm/rockchip: dsi2: add support rk3576

Add the necessary specific bits to support the rk3576-variant of the
DW-DSI2 controller.

Signed-off-by: Guochun Huang <hero.huang@rock-chips.com>
[adapted from the vendor-kernel for mainline]
Reviewed-by: Andy Yan <andy.yan@rock-chips.com>
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250707164906.1445288-10-heiko@sntech.de
---
 .../gpu/drm/rockchip/dw-mipi-dsi2-rockchip.c  | 21 +++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi2-rockchip.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi2-rockchip.c
index cdd490778756..0aea764e29b2 100644
--- a/drivers/gpu/drm/rockchip/dw-mipi-dsi2-rockchip.c
+++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi2-rockchip.c
@@ -437,6 +437,15 @@ static void dw_mipi_dsi2_rockchip_remove(struct platform_device *pdev)
 	dw_mipi_dsi2_remove(dsi2->dmd);
 }
 
+static const struct dsigrf_reg rk3576_dsi_grf_reg_fields[MAX_FIELDS] = {
+	[TXREQCLKHS_EN]		= { 0x0028, 1,  1 },
+	[GATING_EN]		= { 0x0028, 0,  0 },
+	[IPI_SHUTDN]		= { 0x0028, 3,  3 },
+	[IPI_COLORM]		= { 0x0028, 2,  2 },
+	[IPI_COLOR_DEPTH]	= { 0x0028, 8, 11 },
+	[IPI_FORMAT]		= { 0x0028, 4,  7 },
+};
+
 static const struct dsigrf_reg rk3588_dsi0_grf_reg_fields[MAX_FIELDS] = {
 	[TXREQCLKHS_EN]		= { 0x0000, 11, 11 },
 	[GATING_EN]		= { 0x0000, 10, 10 },
@@ -455,6 +464,15 @@ static const struct dsigrf_reg rk3588_dsi1_grf_reg_fields[MAX_FIELDS] = {
 	[IPI_FORMAT]		= { 0x0004,  0,  3 },
 };
 
+static const struct rockchip_dw_dsi2_chip_data rk3576_chip_data[] = {
+	{
+		.reg = 0x27d80000,
+		.grf_regs = rk3576_dsi_grf_reg_fields,
+		.max_bit_rate_per_lane = 2500000ULL,
+	},
+	{ /* sentinel */ }
+};
+
 static const struct rockchip_dw_dsi2_chip_data rk3588_chip_data[] = {
 	{
 		.reg = 0xfde20000,
@@ -470,6 +488,9 @@ static const struct rockchip_dw_dsi2_chip_data rk3588_chip_data[] = {
 
 static const struct of_device_id dw_mipi_dsi2_rockchip_dt_ids[] = {
 	{
+		.compatible = "rockchip,rk3576-mipi-dsi2",
+		.data = &rk3576_chip_data,
+	}, {
 		.compatible = "rockchip,rk3588-mipi-dsi2",
 		.data = &rk3588_chip_data,
 	},

From efe927b9702643a1d80472664c2642f0304cb608 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Sat, 23 Aug 2025 03:12:00 +0300
Subject: [PATCH 0593/1455] drm/msm: fix msm_gem_vma_new() allocations for
 managed GPUVMs

Since commit 3309323241fb ("drm/gpuvm: Kill drm_gpuva_init()") MSM
driver fails to init, failing with "[drm:msm_gpu_init] *ERROR* could not
allocate memptrs: -22" errors. The mentioned commit reworked the
function, but didn't take into account that op_map is initialized at the
top of the function, while ranges might change if GPUVM is managed by
the kernel.

Move op_mode initialization after finalizing all addresses and right
before the drm_gpuva_init_from_op() call.

Reported-by: Danct12 <danct12@disroot.org>
Fixes: 3309323241fb ("drm/gpuvm: Kill drm_gpuva_init()")
Suggested-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Rob Clark <robin.clark@oss.qualcomm.com>
Acked-by: Rob Clark <robin.clark@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250823-msm-fix-gpuvm-init-v1-1-e199cd5b1983@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/msm_gem_vma.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 210604181c05..c632db9005e2 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -371,12 +371,6 @@ struct drm_gpuva *
 msm_gem_vma_new(struct drm_gpuvm *gpuvm, struct drm_gem_object *obj,
 		u64 offset, u64 range_start, u64 range_end)
 {
-	struct drm_gpuva_op_map op_map = {
-		.va.addr = range_start,
-		.va.range = range_end - range_start,
-		.gem.obj = obj,
-		.gem.offset = offset,
-	};
 	struct msm_gem_vm *vm = to_msm_vm(gpuvm);
 	struct drm_gpuvm_bo *vm_bo;
 	struct msm_gem_vma *vma;
@@ -405,6 +399,13 @@ msm_gem_vma_new(struct drm_gpuvm *gpuvm, struct drm_gem_object *obj,
 	if (obj)
 		GEM_WARN_ON((range_end - range_start) > obj->size);
 
+	struct drm_gpuva_op_map op_map = {
+		.va.addr = range_start,
+		.va.range = range_end - range_start,
+		.gem.obj = obj,
+		.gem.offset = offset,
+	};
+
 	drm_gpuva_init_from_op(&vma->base, &op_map);
 	vma->mapped = false;
 

From 4d33c77cf2c3e8842c2f722e1efb6010b021dbc0 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Thu, 21 Aug 2025 09:56:51 +0530
Subject: [PATCH 0594/1455] drm/i915/vbt: Add eDP Data rate overrride field in
 VBT

Add edp_data_rate_override field VBT which gives us a mask
of rates which needs to be skipped in favour of
subsequent higher rate.

--v2
-Rename vbt field [Jani]
-Fix comment to 263+ [Jani]
-Use BIT_U32 [Jani]
-Fix the bits assignment in vbt [Jani]

--v3
-Add a mask which represents all link rates [Ankit]

Bspec: 20124
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250821042653.269227-2-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_bios.c     |  4 +++-
 drivers/gpu/drm/i915/display/intel_vbt_defs.h | 18 ++++++++++++++++++
 2 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 8de332c627f8..cf464c68441a 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -2748,8 +2748,10 @@ static int child_device_expected_size(u16 version)
 {
 	BUILD_BUG_ON(sizeof(struct child_device_config) < 40);
 
-	if (version > 256)
+	if (version > 263)
 		return -ENOENT;
+	else if (version >= 263)
+		return 44;
 	else if (version >= 256)
 		return 40;
 	else if (version >= 216)
diff --git a/drivers/gpu/drm/i915/display/intel_vbt_defs.h b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
index 6612d3a4ec49..70e31520c560 100644
--- a/drivers/gpu/drm/i915/display/intel_vbt_defs.h
+++ b/drivers/gpu/drm/i915/display/intel_vbt_defs.h
@@ -437,6 +437,22 @@ enum vbt_gmbus_ddi {
 #define BDB_230_VBT_DP_MAX_LINK_RATE_UHBR13P5	6
 #define BDB_230_VBT_DP_MAX_LINK_RATE_UHBR20	7
 
+/* EDP link rate 263+ */
+#define BDB_263_VBT_EDP_LINK_RATE_1_62		BIT_U32(0)
+#define BDB_263_VBT_EDP_LINK_RATE_2_16		BIT_U32(1)
+#define BDB_263_VBT_EDP_LINK_RATE_2_43		BIT_U32(2)
+#define BDB_263_VBT_EDP_LINK_RATE_2_7		BIT_U32(3)
+#define BDB_263_VBT_EDP_LINK_RATE_3_24		BIT_U32(4)
+#define BDB_263_VBT_EDP_LINK_RATE_4_32		BIT_U32(5)
+#define BDB_263_VBT_EDP_LINK_RATE_5_4		BIT_U32(6)
+#define BDB_263_VBT_EDP_LINK_RATE_6_75		BIT_U32(7)
+#define BDB_263_VBT_EDP_LINK_RATE_8_1		BIT_U32(8)
+#define BDB_263_VBT_EDP_LINK_RATE_10		BIT_U32(9)
+#define BDB_263_VBT_EDP_LINK_RATE_13_5		BIT_U32(10)
+#define BDB_263_VBT_EDP_LINK_RATE_20		BIT_U32(11)
+#define BDB_263_VBT_EDP_NUM_RATES		12
+#define BDB_263_VBT_EDP_RATES_MASK		GENMASK(BDB_263_VBT_EDP_NUM_RATES - 1, 0)
+
 /*
  * The child device config, aka the display device data structure, provides a
  * description of a port and its configuration on the platform.
@@ -547,6 +563,8 @@ struct child_device_config {
 	u8 dp_max_link_rate:3;					/* 216+ */
 	u8 dp_max_link_rate_reserved:5;				/* 216+ */
 	u8 efp_index;						/* 256+ */
+	u32 edp_data_rate_override:12;				/* 263+ */
+	u32 edp_data_rate_override_reserved:20;			/* 263+ */
 } __packed;
 
 struct bdb_general_definitions {

From 835a0d544defa05d2425496d16d006b147df5bde Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Thu, 21 Aug 2025 09:56:52 +0530
Subject: [PATCH 0595/1455] drm/i915/bios: Add function to check if edp data
 override is needed

Add a function that helps identify if the rate provided needs to
be overridden. For this we need a function that compares the rate
provided and bitmask of rates provided in VBT.

--v2
-Rename functions [Jani]
-Return the mask instead of parsing it in function [Jani]
-Move the declaration in header [Jani]

--v3
-Change function name to depict what the function does [Ankit]

--v4
-Lets not use hweight [Ankit]

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250821042653.269227-3-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_bios.c | 32 +++++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_bios.h |  2 ++
 2 files changed, 34 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index cf464c68441a..7d7193a05d88 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -2481,6 +2481,25 @@ static int parse_bdb_216_dp_max_link_rate(const int vbt_max_link_rate)
 	}
 }
 
+static u32 edp_rate_override_mask(int rate)
+{
+	switch (rate) {
+	case 2000000: return BDB_263_VBT_EDP_LINK_RATE_20;
+	case 1350000: return BDB_263_VBT_EDP_LINK_RATE_13_5;
+	case 1000000: return BDB_263_VBT_EDP_LINK_RATE_10;
+	case 810000: return BDB_263_VBT_EDP_LINK_RATE_8_1;
+	case 675000: return BDB_263_VBT_EDP_LINK_RATE_6_75;
+	case 540000: return BDB_263_VBT_EDP_LINK_RATE_5_4;
+	case 432000: return BDB_263_VBT_EDP_LINK_RATE_4_32;
+	case 324000: return BDB_263_VBT_EDP_LINK_RATE_3_24;
+	case 270000: return BDB_263_VBT_EDP_LINK_RATE_2_7;
+	case 243000: return BDB_263_VBT_EDP_LINK_RATE_2_43;
+	case 216000: return BDB_263_VBT_EDP_LINK_RATE_2_16;
+	case 162000: return BDB_263_VBT_EDP_LINK_RATE_1_62;
+	default: return 0;
+	}
+}
+
 int intel_bios_dp_max_link_rate(const struct intel_bios_encoder_data *devdata)
 {
 	if (!devdata || devdata->display->vbt.version < 216)
@@ -2500,6 +2519,19 @@ int intel_bios_dp_max_lane_count(const struct intel_bios_encoder_data *devdata)
 	return devdata->child.dp_max_lane_count + 1;
 }
 
+bool
+intel_bios_encoder_reject_edp_rate(const struct intel_bios_encoder_data *devdata,
+				   int rate)
+{
+	if (!devdata || devdata->display->vbt.version < 263)
+		return false;
+
+	if (devdata->child.edp_data_rate_override == BDB_263_VBT_EDP_RATES_MASK)
+		return false;
+
+	return devdata->child.edp_data_rate_override & edp_rate_override_mask(rate);
+}
+
 static void sanitize_device_type(struct intel_bios_encoder_data *devdata,
 				 enum port port)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_bios.h b/drivers/gpu/drm/i915/display/intel_bios.h
index 8fdde85f7939..f9e438b2787b 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.h
+++ b/drivers/gpu/drm/i915/display/intel_bios.h
@@ -85,6 +85,8 @@ bool intel_bios_encoder_is_lspcon(const struct intel_bios_encoder_data *devdata)
 bool intel_bios_encoder_lane_reversal(const struct intel_bios_encoder_data *devdata);
 bool intel_bios_encoder_hpd_invert(const struct intel_bios_encoder_data *devdata);
 enum port intel_bios_encoder_port(const struct intel_bios_encoder_data *devdata);
+bool intel_bios_encoder_reject_edp_rate(const struct intel_bios_encoder_data *devdata,
+					int rate);
 enum aux_ch intel_bios_dp_aux_ch(const struct intel_bios_encoder_data *devdata);
 int intel_bios_dp_boost_level(const struct intel_bios_encoder_data *devdata);
 int intel_bios_dp_max_lane_count(const struct intel_bios_encoder_data *devdata);

From 2dbb3d70af6003dbf3c248cc01e36d7df142543a Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Thu, 21 Aug 2025 09:56:53 +0530
Subject: [PATCH 0596/1455] drm/i915/edp: eDP Data Overrride

We need override certain link rates in favour of the next available
higher link rate. The Link rates that need to be overridden are
indicated by a mask in VBT. To make sure these modes are skipped we
don't add them in them in the sink rates array.

--v2
-Update the link rates after we have a final set of link rates [Ankit]
-Break this patch up [Ankit]
-Optimize the assingment during loop [Ankit]

--v3
-Add protection against broken VBTs [Jani]

--v4
-Fix build errors
-Create a seprate function to check if edp data override is selected
and using the correct vbt

--v5
-Use correct number to check the num of edp rates [Ankit]

--v6
-No seprate function check if vbt is broken in the reject rate function
[Jani]

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250821042653.269227-4-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp.c | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 996edb8deded..bed7cba1ca68 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -4312,6 +4312,23 @@ static void intel_edp_mso_init(struct intel_dp *intel_dp)
 	intel_dp->mso_pixel_overlap = mso ? info->mso_pixel_overlap : 0;
 }
 
+static void
+intel_edp_set_data_override_rates(struct intel_dp *intel_dp)
+{
+	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
+	int *sink_rates = intel_dp->sink_rates;
+	int i, count = 0;
+
+	for (i = 0; i < intel_dp->num_sink_rates; i++) {
+		if (intel_bios_encoder_reject_edp_rate(encoder->devdata,
+						       intel_dp->sink_rates[i]))
+			continue;
+
+		sink_rates[count++] = intel_dp->sink_rates[i];
+	}
+	intel_dp->num_sink_rates = count;
+}
+
 static void
 intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 {
@@ -4362,6 +4379,8 @@ intel_edp_set_sink_rates(struct intel_dp *intel_dp)
 		intel_dp->use_rate_select = true;
 	else
 		intel_dp_set_sink_rates(intel_dp);
+
+	intel_edp_set_data_override_rates(intel_dp);
 }
 
 static bool

From 8ae04fe9ffc93d6bc3bc63ac08375427d69cee06 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Zbigniew=20Kempczy=C5=84ski?=
 <zbigniew.kempczynski@intel.com>
Date: Wed, 20 Aug 2025 10:39:04 +0200
Subject: [PATCH 0597/1455] drm/xe/xe_sync: avoid race during ufence signaling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Marking ufence as signalled after copy_to_user() is too late.
Worker thread which signals ufence by memory write might be raced
with another userspace vm-bind call. In map/unmap scenario unmap
may still see ufence is not signalled causing -EBUSY. Change the
order of marking / write to user-fence fixes this issue.

Fixes: 977e5b82e090 ("drm/xe: Expose user fence from xe_sync_entry")
Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5536
Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250820083903.2109891-2-zbigniew.kempczynski@intel.com
---
 drivers/gpu/drm/xe/xe_sync.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_sync.c b/drivers/gpu/drm/xe/xe_sync.c
index f87276df18f2..82872a51f098 100644
--- a/drivers/gpu/drm/xe/xe_sync.c
+++ b/drivers/gpu/drm/xe/xe_sync.c
@@ -77,6 +77,7 @@ static void user_fence_worker(struct work_struct *w)
 {
 	struct xe_user_fence *ufence = container_of(w, struct xe_user_fence, worker);
 
+	WRITE_ONCE(ufence->signalled, 1);
 	if (mmget_not_zero(ufence->mm)) {
 		kthread_use_mm(ufence->mm);
 		if (copy_to_user(ufence->addr, &ufence->value, sizeof(ufence->value)))
@@ -91,7 +92,6 @@ static void user_fence_worker(struct work_struct *w)
 	 * Wake up waiters only after updating the ufence state, allowing the UMD
 	 * to safely reuse the same ufence without encountering -EBUSY errors.
 	 */
-	WRITE_ONCE(ufence->signalled, 1);
 	wake_up_all(&ufence->xe->ufence_wq);
 	user_fence_put(ufence);
 }

From 9cc10041e9fe7f32c4817e3cdd806ff1986d266c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Thu, 21 Aug 2025 07:59:17 +0300
Subject: [PATCH 0598/1455] drm/i915/psr: Check drm_dp_dpcd_read return value
 on PSR dpcd init
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we are ignoriong drm_dp_dpcd_read return values when reading PSR
and Panel Replay capability DPCD register. Rework intel_psr_dpcd a bit to
take care of checking the return value.

v2: use drm_dp_dpcd_read_data

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/20250821045918.17757-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 32 ++++++++++++++++--------
 1 file changed, 21 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 465f12c39330..eec4f7dc2d66 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -580,6 +580,16 @@ exit:
 static void _panel_replay_init_dpcd(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
+	int ret;
+
+	ret = drm_dp_dpcd_read_data(&intel_dp->aux, DP_PANEL_REPLAY_CAP_SUPPORT,
+				    &intel_dp->pr_dpcd, sizeof(intel_dp->pr_dpcd));
+	if (ret < 0)
+		return;
+
+	if (!(intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_SUPPORT)] &
+	      DP_PANEL_REPLAY_SUPPORT))
+		return;
 
 	if (intel_dp_is_edp(intel_dp)) {
 		if (!intel_alpm_aux_less_wake_supported(intel_dp)) {
@@ -611,6 +621,15 @@ static void _panel_replay_init_dpcd(struct intel_dp *intel_dp)
 static void _psr_init_dpcd(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
+	int ret;
+
+	ret = drm_dp_dpcd_read_data(&intel_dp->aux, DP_PSR_SUPPORT, intel_dp->psr_dpcd,
+				    sizeof(intel_dp->psr_dpcd));
+	if (ret < 0)
+		return;
+
+	if (!intel_dp->psr_dpcd[0])
+		return;
 
 	drm_dbg_kms(display->drm, "eDP panel supports PSR version %x\n",
 		    intel_dp->psr_dpcd[0]);
@@ -656,18 +675,9 @@ static void _psr_init_dpcd(struct intel_dp *intel_dp)
 
 void intel_psr_init_dpcd(struct intel_dp *intel_dp)
 {
-	drm_dp_dpcd_read(&intel_dp->aux, DP_PSR_SUPPORT, intel_dp->psr_dpcd,
-			 sizeof(intel_dp->psr_dpcd));
+	_psr_init_dpcd(intel_dp);
 
-	drm_dp_dpcd_read(&intel_dp->aux, DP_PANEL_REPLAY_CAP_SUPPORT,
-			 &intel_dp->pr_dpcd, sizeof(intel_dp->pr_dpcd));
-
-	if (intel_dp->pr_dpcd[INTEL_PR_DPCD_INDEX(DP_PANEL_REPLAY_CAP_SUPPORT)] &
-	    DP_PANEL_REPLAY_SUPPORT)
-		_panel_replay_init_dpcd(intel_dp);
-
-	if (intel_dp->psr_dpcd[0])
-		_psr_init_dpcd(intel_dp);
+	_panel_replay_init_dpcd(intel_dp);
 
 	if (intel_dp->psr.sink_psr2_support ||
 	    intel_dp->psr.sink_panel_replay_su_support)

From 0a51bf3e54dd8b77e6f1febbbb66def0660862d2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 21 Aug 2025 16:30:43 +0200
Subject: [PATCH 0599/1455] drm/xe/vm: Don't pin the vm_resv during validation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The pinning has the odd side-effect that unlocking *any* resv
during validation triggers an "unlocking pinned lock" warning.

Cc: Matthew Brost <matthew.brost@intel.com>
Fixes: 9d5558649f68 ("drm/xe: Rework eviction rejection of bound external bos")
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821143045.106005-2-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_bo.c |  5 ++---
 drivers/gpu/drm/xe/xe_vm.h | 15 ++-------------
 2 files changed, 4 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 6fea39842e1e..11eaf3b06766 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -2468,7 +2468,6 @@ int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict)
 		.no_wait_gpu = false,
 		.gfp_retry_mayfail = true,
 	};
-	struct pin_cookie cookie;
 	int ret;
 
 	if (vm) {
@@ -2479,10 +2478,10 @@ int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict)
 		ctx.resv = xe_vm_resv(vm);
 	}
 
-	cookie = xe_vm_set_validating(vm, allow_res_evict);
+	xe_vm_set_validating(vm, allow_res_evict);
 	trace_xe_bo_validate(bo);
 	ret = ttm_bo_validate(&bo->ttm, &bo->placement, &ctx);
-	xe_vm_clear_validating(vm, allow_res_evict, cookie);
+	xe_vm_clear_validating(vm, allow_res_evict);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 2f213737c7e5..2ecb417c19a2 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -315,22 +315,14 @@ void xe_vm_snapshot_free(struct xe_vm_snapshot *snap);
  * Register this task as currently making bos resident for the vm. Intended
  * to avoid eviction by the same task of shared bos bound to the vm.
  * Call with the vm's resv lock held.
- *
- * Return: A pin cookie that should be used for xe_vm_clear_validating().
  */
-static inline struct pin_cookie xe_vm_set_validating(struct xe_vm *vm,
-						     bool allow_res_evict)
+static inline void xe_vm_set_validating(struct xe_vm *vm, bool allow_res_evict)
 {
-	struct pin_cookie cookie = {};
-
 	if (vm && !allow_res_evict) {
 		xe_vm_assert_held(vm);
-		cookie = lockdep_pin_lock(&xe_vm_resv(vm)->lock.base);
 		/* Pairs with READ_ONCE in xe_vm_is_validating() */
 		WRITE_ONCE(vm->validating, current);
 	}
-
-	return cookie;
 }
 
 /**
@@ -338,17 +330,14 @@ static inline struct pin_cookie xe_vm_set_validating(struct xe_vm *vm,
  * @vm: Pointer to the vm or NULL
  * @allow_res_evict: Eviction from @vm was allowed. Must be set to the same
  * value as for xe_vm_set_validation().
- * @cookie: Cookie obtained from xe_vm_set_validating().
  *
  * Register this task as currently making bos resident for the vm. Intended
  * to avoid eviction by the same task of shared bos bound to the vm.
  * Call with the vm's resv lock held.
  */
-static inline void xe_vm_clear_validating(struct xe_vm *vm, bool allow_res_evict,
-					  struct pin_cookie cookie)
+static inline void xe_vm_clear_validating(struct xe_vm *vm, bool allow_res_evict)
 {
 	if (vm && !allow_res_evict) {
-		lockdep_unpin_lock(&xe_vm_resv(vm)->lock.base, cookie);
 		/* Pairs with READ_ONCE in xe_vm_is_validating() */
 		WRITE_ONCE(vm->validating, NULL);
 	}

From b5dd1505a31c5cbe62ec50a70e9c9a7bd2a8f703 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 21 Aug 2025 16:30:44 +0200
Subject: [PATCH 0600/1455] drm/xe/tests/xe_dma_buf: Set the
 drm_object::dma_buf member
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This member is set when exporting using prime. However
the xe_gem_prime_export() alone doesn't set it, since it's done
later in the prime export flow.
For the test, set it manually and remove the hack that set it
temporarily when it was really needed.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821143045.106005-3-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/tests/xe_dma_buf.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
index c53f67ce4b0a..cde9530bef8c 100644
--- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
@@ -57,16 +57,12 @@ static void check_residency(struct kunit *test, struct xe_bo *exported,
 		return;
 
 	/*
-	 * Evict exporter. Note that the gem object dma_buf member isn't
-	 * set from xe_gem_prime_export(), and it's needed for the move_notify()
-	 * functionality, so hack that up here. Evicting the exported bo will
+	 * Evict exporter. Evicting the exported bo will
 	 * evict also the imported bo through the move_notify() functionality if
 	 * importer is on a different device. If they're on the same device,
 	 * the exporter and the importer should be the same bo.
 	 */
-	swap(exported->ttm.base.dma_buf, dmabuf);
 	ret = xe_bo_evict(exported);
-	swap(exported->ttm.base.dma_buf, dmabuf);
 	if (ret) {
 		if (ret != -EINTR && ret != -ERESTARTSYS)
 			KUNIT_FAIL(test, "Evicting exporter failed with err=%d.\n",
@@ -139,6 +135,7 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe)
 			   PTR_ERR(dmabuf));
 		goto out;
 	}
+	bo->ttm.base.dma_buf = dmabuf;
 
 	import = xe_gem_prime_import(&xe->drm, dmabuf);
 	if (!IS_ERR(import)) {
@@ -186,6 +183,7 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe)
 		KUNIT_FAIL(test, "dynamic p2p attachment failed with err=%ld\n",
 			   PTR_ERR(import));
 	}
+	bo->ttm.base.dma_buf = NULL;
 	dma_buf_put(dmabuf);
 out:
 	drm_gem_object_put(&bo->ttm.base);

From 358ee50ab565f3c8ea32480e9d03127a81ba32f8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 21 Aug 2025 16:30:45 +0200
Subject: [PATCH 0601/1455] drm/xe/vm: Clear the scratch_pt pointer on error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Avoid triggering a dereference of an error pointer on cleanup in
xe_vm_free_scratch() by clearing any scratch_pt error pointer.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Fixes: 06951c2ee72d ("drm/xe: Use NULL PTEs as scratch PTEs")
Cc: Brian Welty <brian.welty@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: <stable@vger.kernel.org> # v6.8+
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821143045.106005-4-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_vm.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index c86337e08a55..d3f6dc6b1779 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1635,8 +1635,12 @@ static int xe_vm_create_scratch(struct xe_device *xe, struct xe_tile *tile,
 
 	for (i = MAX_HUGEPTE_LEVEL; i < vm->pt_root[id]->level; i++) {
 		vm->scratch_pt[id][i] = xe_pt_create(vm, tile, i);
-		if (IS_ERR(vm->scratch_pt[id][i]))
-			return PTR_ERR(vm->scratch_pt[id][i]);
+		if (IS_ERR(vm->scratch_pt[id][i])) {
+			int err = PTR_ERR(vm->scratch_pt[id][i]);
+
+			vm->scratch_pt[id][i] = NULL;
+			return err;
+		}
 
 		xe_pt_populate_empty(tile, vm, vm->scratch_pt[id][i]);
 	}

From 2eb22214c132374e11e681c44d7879c91f67f614 Mon Sep 17 00:00:00 2001
From: Pin-yen Lin <treapking@chromium.org>
Date: Mon, 18 Aug 2025 19:49:33 +0800
Subject: [PATCH 0602/1455] drm/panel: Allow powering on panel follower after
 panel is enabled

Some touch controllers have to be powered on after the panel's backlight
is enabled. To support these controllers, introduce .panel_enabled() and
.panel_disabling() to panel_follower_funcs and use them to power on the
device after the panel and its backlight are enabled.

Signed-off-by: Pin-yen Lin <treapking@chromium.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250818115015.2909525-1-treapking@chromium.org
---
 drivers/gpu/drm/drm_panel.c | 73 +++++++++++++++++++++++++++++++------
 include/drm/drm_panel.h     | 14 +++++++
 2 files changed, 76 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/drm_panel.c b/drivers/gpu/drm/drm_panel.c
index c8bb28dccdc1..d1e6598ea3bc 100644
--- a/drivers/gpu/drm/drm_panel.c
+++ b/drivers/gpu/drm/drm_panel.c
@@ -134,6 +134,9 @@ void drm_panel_prepare(struct drm_panel *panel)
 	panel->prepared = true;
 
 	list_for_each_entry(follower, &panel->followers, list) {
+		if (!follower->funcs->panel_prepared)
+			continue;
+
 		ret = follower->funcs->panel_prepared(follower);
 		if (ret < 0)
 			dev_info(panel->dev, "%ps failed: %d\n",
@@ -179,6 +182,9 @@ void drm_panel_unprepare(struct drm_panel *panel)
 	mutex_lock(&panel->follower_lock);
 
 	list_for_each_entry(follower, &panel->followers, list) {
+		if (!follower->funcs->panel_unpreparing)
+			continue;
+
 		ret = follower->funcs->panel_unpreparing(follower);
 		if (ret < 0)
 			dev_info(panel->dev, "%ps failed: %d\n",
@@ -209,6 +215,7 @@ EXPORT_SYMBOL(drm_panel_unprepare);
  */
 void drm_panel_enable(struct drm_panel *panel)
 {
+	struct drm_panel_follower *follower;
 	int ret;
 
 	if (!panel)
@@ -219,10 +226,12 @@ void drm_panel_enable(struct drm_panel *panel)
 		return;
 	}
 
+	mutex_lock(&panel->follower_lock);
+
 	if (panel->funcs && panel->funcs->enable) {
 		ret = panel->funcs->enable(panel);
 		if (ret < 0)
-			return;
+			goto exit;
 	}
 	panel->enabled = true;
 
@@ -230,6 +239,19 @@ void drm_panel_enable(struct drm_panel *panel)
 	if (ret < 0)
 		DRM_DEV_INFO(panel->dev, "failed to enable backlight: %d\n",
 			     ret);
+
+	list_for_each_entry(follower, &panel->followers, list) {
+		if (!follower->funcs->panel_enabled)
+			continue;
+
+		ret = follower->funcs->panel_enabled(follower);
+		if (ret < 0)
+			dev_info(panel->dev, "%ps failed: %d\n",
+				 follower->funcs->panel_enabled, ret);
+	}
+
+exit:
+	mutex_unlock(&panel->follower_lock);
 }
 EXPORT_SYMBOL(drm_panel_enable);
 
@@ -243,6 +265,7 @@ EXPORT_SYMBOL(drm_panel_enable);
  */
 void drm_panel_disable(struct drm_panel *panel)
 {
+	struct drm_panel_follower *follower;
 	int ret;
 
 	if (!panel)
@@ -262,6 +285,18 @@ void drm_panel_disable(struct drm_panel *panel)
 		return;
 	}
 
+	mutex_lock(&panel->follower_lock);
+
+	list_for_each_entry(follower, &panel->followers, list) {
+		if (!follower->funcs->panel_disabling)
+			continue;
+
+		ret = follower->funcs->panel_disabling(follower);
+		if (ret < 0)
+			dev_info(panel->dev, "%ps failed: %d\n",
+				 follower->funcs->panel_disabling, ret);
+	}
+
 	ret = backlight_disable(panel->backlight);
 	if (ret < 0)
 		DRM_DEV_INFO(panel->dev, "failed to disable backlight: %d\n",
@@ -270,9 +305,12 @@ void drm_panel_disable(struct drm_panel *panel)
 	if (panel->funcs && panel->funcs->disable) {
 		ret = panel->funcs->disable(panel);
 		if (ret < 0)
-			return;
+			goto exit;
 	}
 	panel->enabled = false;
+
+exit:
+	mutex_unlock(&panel->follower_lock);
 }
 EXPORT_SYMBOL(drm_panel_disable);
 
@@ -539,13 +577,13 @@ EXPORT_SYMBOL(drm_is_panel_follower);
  * @follower_dev: The 'struct device' for the follower.
  * @follower:     The panel follower descriptor for the follower.
  *
- * A panel follower is called right after preparing the panel and right before
- * unpreparing the panel. It's primary intention is to power on an associated
- * touchscreen, though it could be used for any similar devices. Multiple
- * devices are allowed the follow the same panel.
+ * A panel follower is called right after preparing/enabling the panel and right
+ * before unpreparing/disabling the panel. It's primary intention is to power on
+ * an associated touchscreen, though it could be used for any similar devices.
+ * Multiple devices are allowed the follow the same panel.
  *
- * If a follower is added to a panel that's already been turned on, the
- * follower's prepare callback is called right away.
+ * If a follower is added to a panel that's already been prepared/enabled, the
+ * follower's prepared/enabled callback is called right away.
  *
  * The "panel" property of the follower points to the panel to be followed.
  *
@@ -569,12 +607,18 @@ int drm_panel_add_follower(struct device *follower_dev,
 	mutex_lock(&panel->follower_lock);
 
 	list_add_tail(&follower->list, &panel->followers);
-	if (panel->prepared) {
+	if (panel->prepared && follower->funcs->panel_prepared) {
 		ret = follower->funcs->panel_prepared(follower);
 		if (ret < 0)
 			dev_info(panel->dev, "%ps failed: %d\n",
 				 follower->funcs->panel_prepared, ret);
 	}
+	if (panel->enabled && follower->funcs->panel_enabled) {
+		ret = follower->funcs->panel_enabled(follower);
+		if (ret < 0)
+			dev_info(panel->dev, "%ps failed: %d\n",
+				 follower->funcs->panel_enabled, ret);
+	}
 
 	mutex_unlock(&panel->follower_lock);
 
@@ -587,7 +631,8 @@ EXPORT_SYMBOL(drm_panel_add_follower);
  * @follower:     The panel follower descriptor for the follower.
  *
  * Undo drm_panel_add_follower(). This includes calling the follower's
- * unprepare function if we're removed from a panel that's currently prepared.
+ * unpreparing/disabling function if we're removed from a panel that's currently
+ * prepared/enabled.
  *
  * Return: 0 or an error code.
  */
@@ -598,7 +643,13 @@ void drm_panel_remove_follower(struct drm_panel_follower *follower)
 
 	mutex_lock(&panel->follower_lock);
 
-	if (panel->prepared) {
+	if (panel->enabled && follower->funcs->panel_disabling) {
+		ret = follower->funcs->panel_disabling(follower);
+		if (ret < 0)
+			dev_info(panel->dev, "%ps failed: %d\n",
+				 follower->funcs->panel_disabling, ret);
+	}
+	if (panel->prepared && follower->funcs->panel_unpreparing) {
 		ret = follower->funcs->panel_unpreparing(follower);
 		if (ret < 0)
 			dev_info(panel->dev, "%ps failed: %d\n",
diff --git a/include/drm/drm_panel.h b/include/drm/drm_panel.h
index 843fb756a295..2407bfa60236 100644
--- a/include/drm/drm_panel.h
+++ b/include/drm/drm_panel.h
@@ -160,6 +160,20 @@ struct drm_panel_follower_funcs {
 	 * Called before the panel is powered off.
 	 */
 	int (*panel_unpreparing)(struct drm_panel_follower *follower);
+
+	/**
+	 * @panel_enabled:
+	 *
+	 * Called after the panel and the backlight have been enabled.
+	 */
+	int (*panel_enabled)(struct drm_panel_follower *follower);
+
+	/**
+	 * @panel_disabling:
+	 *
+	 * Called before the panel and the backlight are disabled.
+	 */
+	int (*panel_disabling)(struct drm_panel_follower *follower);
 };
 
 struct drm_panel_follower {

From cbdd16b818eef876dd2de9d503fe7397a0666cbe Mon Sep 17 00:00:00 2001
From: Pin-yen Lin <treapking@chromium.org>
Date: Mon, 18 Aug 2025 19:49:34 +0800
Subject: [PATCH 0603/1455] HID: i2c-hid: Make elan touch controllers power on
 after panel is enabled

Introduce a new HID quirk to indicate that this device has to be enabled
after the panel's backlight is enabled, and update the driver data for
the elan devices to enable this quirk. This cannot be a I2C HID quirk
because the kernel needs to acknowledge this before powering up the
device and read the VID/PID. When this quirk is enabled, register
.panel_enabled()/.panel_disabling() instead for the panel follower.

Also rename the *panel_prepare* functions into *panel_follower* because
they could be called in other situations now.

Fixes: bd3cba00dcc63 ("HID: i2c-hid: elan: Add support for Elan eKTH6915 i2c-hid touchscreens")
Fixes: d06651bebf99e ("HID: i2c-hid: elan: Add elan-ekth6a12nay timing")
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Pin-yen Lin <treapking@chromium.org>
Acked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250818115015.2909525-2-treapking@chromium.org
---
 drivers/hid/i2c-hid/i2c-hid-core.c    | 46 ++++++++++++++++-----------
 drivers/hid/i2c-hid/i2c-hid-of-elan.c | 11 ++++++-
 include/linux/hid.h                   |  2 ++
 3 files changed, 40 insertions(+), 19 deletions(-)

diff --git a/drivers/hid/i2c-hid/i2c-hid-core.c b/drivers/hid/i2c-hid/i2c-hid-core.c
index d3912e3f2f13..99ce6386176c 100644
--- a/drivers/hid/i2c-hid/i2c-hid-core.c
+++ b/drivers/hid/i2c-hid/i2c-hid-core.c
@@ -112,9 +112,9 @@ struct i2c_hid {
 
 	struct i2chid_ops	*ops;
 	struct drm_panel_follower panel_follower;
-	struct work_struct	panel_follower_prepare_work;
+	struct work_struct	panel_follower_work;
 	bool			is_panel_follower;
-	bool			prepare_work_finished;
+	bool			panel_follower_work_finished;
 };
 
 static const struct i2c_hid_quirks {
@@ -1110,10 +1110,10 @@ err_power_down:
 	return ret;
 }
 
-static void ihid_core_panel_prepare_work(struct work_struct *work)
+static void ihid_core_panel_follower_work(struct work_struct *work)
 {
 	struct i2c_hid *ihid = container_of(work, struct i2c_hid,
-					    panel_follower_prepare_work);
+					    panel_follower_work);
 	struct hid_device *hid = ihid->hid;
 	int ret;
 
@@ -1130,7 +1130,7 @@ static void ihid_core_panel_prepare_work(struct work_struct *work)
 	if (ret)
 		dev_warn(&ihid->client->dev, "Power on failed: %d\n", ret);
 	else
-		WRITE_ONCE(ihid->prepare_work_finished, true);
+		WRITE_ONCE(ihid->panel_follower_work_finished, true);
 
 	/*
 	 * The work APIs provide a number of memory ordering guarantees
@@ -1139,12 +1139,12 @@ static void ihid_core_panel_prepare_work(struct work_struct *work)
 	 * guarantee that a write that happened in the work is visible after
 	 * cancel_work_sync(). We'll add a write memory barrier here to match
 	 * with i2c_hid_core_panel_unpreparing() to ensure that our write to
-	 * prepare_work_finished is visible there.
+	 * panel_follower_work_finished is visible there.
 	 */
 	smp_wmb();
 }
 
-static int i2c_hid_core_panel_prepared(struct drm_panel_follower *follower)
+static int i2c_hid_core_panel_follower_resume(struct drm_panel_follower *follower)
 {
 	struct i2c_hid *ihid = container_of(follower, struct i2c_hid, panel_follower);
 
@@ -1152,29 +1152,36 @@ static int i2c_hid_core_panel_prepared(struct drm_panel_follower *follower)
 	 * Powering on a touchscreen can be a slow process. Queue the work to
 	 * the system workqueue so we don't block the panel's power up.
 	 */
-	WRITE_ONCE(ihid->prepare_work_finished, false);
-	schedule_work(&ihid->panel_follower_prepare_work);
+	WRITE_ONCE(ihid->panel_follower_work_finished, false);
+	schedule_work(&ihid->panel_follower_work);
 
 	return 0;
 }
 
-static int i2c_hid_core_panel_unpreparing(struct drm_panel_follower *follower)
+static int i2c_hid_core_panel_follower_suspend(struct drm_panel_follower *follower)
 {
 	struct i2c_hid *ihid = container_of(follower, struct i2c_hid, panel_follower);
 
-	cancel_work_sync(&ihid->panel_follower_prepare_work);
+	cancel_work_sync(&ihid->panel_follower_work);
 
-	/* Match with ihid_core_panel_prepare_work() */
+	/* Match with ihid_core_panel_follower_work() */
 	smp_rmb();
-	if (!READ_ONCE(ihid->prepare_work_finished))
+	if (!READ_ONCE(ihid->panel_follower_work_finished))
 		return 0;
 
 	return i2c_hid_core_suspend(ihid, true);
 }
 
-static const struct drm_panel_follower_funcs i2c_hid_core_panel_follower_funcs = {
-	.panel_prepared = i2c_hid_core_panel_prepared,
-	.panel_unpreparing = i2c_hid_core_panel_unpreparing,
+static const struct drm_panel_follower_funcs
+				i2c_hid_core_panel_follower_prepare_funcs = {
+	.panel_prepared = i2c_hid_core_panel_follower_resume,
+	.panel_unpreparing = i2c_hid_core_panel_follower_suspend,
+};
+
+static const struct drm_panel_follower_funcs
+				i2c_hid_core_panel_follower_enable_funcs = {
+	.panel_enabled = i2c_hid_core_panel_follower_resume,
+	.panel_disabling = i2c_hid_core_panel_follower_suspend,
 };
 
 static int i2c_hid_core_register_panel_follower(struct i2c_hid *ihid)
@@ -1182,7 +1189,10 @@ static int i2c_hid_core_register_panel_follower(struct i2c_hid *ihid)
 	struct device *dev = &ihid->client->dev;
 	int ret;
 
-	ihid->panel_follower.funcs = &i2c_hid_core_panel_follower_funcs;
+	if (ihid->hid->initial_quirks | HID_QUIRK_POWER_ON_AFTER_BACKLIGHT)
+		ihid->panel_follower.funcs = &i2c_hid_core_panel_follower_enable_funcs;
+	else
+		ihid->panel_follower.funcs = &i2c_hid_core_panel_follower_prepare_funcs;
 
 	/*
 	 * If we're not in control of our own power up/power down then we can't
@@ -1237,7 +1247,7 @@ int i2c_hid_core_probe(struct i2c_client *client, struct i2chid_ops *ops,
 	init_waitqueue_head(&ihid->wait);
 	mutex_init(&ihid->cmd_lock);
 	mutex_init(&ihid->reset_lock);
-	INIT_WORK(&ihid->panel_follower_prepare_work, ihid_core_panel_prepare_work);
+	INIT_WORK(&ihid->panel_follower_work, ihid_core_panel_follower_work);
 
 	/* we need to allocate the command buffer without knowing the maximum
 	 * size of the reports. Let's use HID_MIN_BUFFER_SIZE, then we do the
diff --git a/drivers/hid/i2c-hid/i2c-hid-of-elan.c b/drivers/hid/i2c-hid/i2c-hid-of-elan.c
index 3fcff6daa0d3..0215f217f6d8 100644
--- a/drivers/hid/i2c-hid/i2c-hid-of-elan.c
+++ b/drivers/hid/i2c-hid/i2c-hid-of-elan.c
@@ -8,6 +8,7 @@
 #include <linux/delay.h>
 #include <linux/device.h>
 #include <linux/gpio/consumer.h>
+#include <linux/hid.h>
 #include <linux/i2c.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -23,6 +24,7 @@ struct elan_i2c_hid_chip_data {
 	unsigned int post_power_delay_ms;
 	u16 hid_descriptor_address;
 	const char *main_supply_name;
+	bool power_after_backlight;
 };
 
 struct i2c_hid_of_elan {
@@ -97,6 +99,7 @@ static int i2c_hid_of_elan_probe(struct i2c_client *client)
 {
 	struct i2c_hid_of_elan *ihid_elan;
 	int ret;
+	u32 quirks = 0;
 
 	ihid_elan = devm_kzalloc(&client->dev, sizeof(*ihid_elan), GFP_KERNEL);
 	if (!ihid_elan)
@@ -131,8 +134,12 @@ static int i2c_hid_of_elan_probe(struct i2c_client *client)
 		}
 	}
 
+	if (ihid_elan->chip_data->power_after_backlight)
+		quirks = HID_QUIRK_POWER_ON_AFTER_BACKLIGHT;
+
 	ret = i2c_hid_core_probe(client, &ihid_elan->ops,
-				 ihid_elan->chip_data->hid_descriptor_address, 0);
+				 ihid_elan->chip_data->hid_descriptor_address,
+				 quirks);
 	if (ret)
 		goto err_deassert_reset;
 
@@ -150,6 +157,7 @@ static const struct elan_i2c_hid_chip_data elan_ekth6915_chip_data = {
 	.post_gpio_reset_on_delay_ms = 300,
 	.hid_descriptor_address = 0x0001,
 	.main_supply_name = "vcc33",
+	.power_after_backlight = true,
 };
 
 static const struct elan_i2c_hid_chip_data elan_ekth6a12nay_chip_data = {
@@ -157,6 +165,7 @@ static const struct elan_i2c_hid_chip_data elan_ekth6a12nay_chip_data = {
 	.post_gpio_reset_on_delay_ms = 300,
 	.hid_descriptor_address = 0x0001,
 	.main_supply_name = "vcc33",
+	.power_after_backlight = true,
 };
 
 static const struct elan_i2c_hid_chip_data ilitek_ili9882t_chip_data = {
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 2cc4f1e4ea96..c32425b5d011 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -364,6 +364,7 @@ struct hid_item {
  * | @HID_QUIRK_HAVE_SPECIAL_DRIVER:
  * | @HID_QUIRK_INCREMENT_USAGE_ON_DUPLICATE:
  * | @HID_QUIRK_IGNORE_SPECIAL_DRIVER
+ * | @HID_QUIRK_POWER_ON_AFTER_BACKLIGHT
  * | @HID_QUIRK_FULLSPEED_INTERVAL:
  * | @HID_QUIRK_NO_INIT_REPORTS:
  * | @HID_QUIRK_NO_IGNORE:
@@ -391,6 +392,7 @@ struct hid_item {
 #define HID_QUIRK_INCREMENT_USAGE_ON_DUPLICATE	BIT(20)
 #define HID_QUIRK_NOINVERT			BIT(21)
 #define HID_QUIRK_IGNORE_SPECIAL_DRIVER		BIT(22)
+#define HID_QUIRK_POWER_ON_AFTER_BACKLIGHT	BIT(23)
 #define HID_QUIRK_FULLSPEED_INTERVAL		BIT(28)
 #define HID_QUIRK_NO_INIT_REPORTS		BIT(29)
 #define HID_QUIRK_NO_IGNORE			BIT(30)

From ffdf968762e4fb3cdae54e811ec3525e67440a60 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Mon, 25 Aug 2025 08:28:41 -0700
Subject: [PATCH 0604/1455] drm/xe: Don't trigger rebind on initial dma-buf
 validation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On the first validate of an imported dma-buf (initial bind), the device
has no GPU mappings, so a rebind is unnecessary. Rebinding here is
harmful in multi-GPU setups and for VMs using preempt-fence mode, as it
would evict in-flight GPU work.

v2:
 - Drop dma_buf_validated, check for XE_PL_SYSTEM (Thomas)

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250825152841.3837378-1-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_bo.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 11eaf3b06766..0a6a32cfb771 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -816,7 +816,8 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,
 	}
 
 	if (ttm_bo->type == ttm_bo_type_sg) {
-		ret = xe_bo_move_notify(bo, ctx);
+		if (new_mem->mem_type == XE_PL_SYSTEM)
+			ret = xe_bo_move_notify(bo, ctx);
 		if (!ret)
 			ret = xe_bo_move_dmabuf(ttm_bo, new_mem);
 		return ret;

From 41be792f5baaf90d744a9a9e82994ce560ca9582 Mon Sep 17 00:00:00 2001
From: Carlos Llamas <cmllamas@google.com>
Date: Mon, 25 Aug 2025 15:57:42 +0000
Subject: [PATCH 0605/1455] drm/xe: switch to local xbasename() helper
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit b0a2ee5567ab ("drm/xe: prepare xe_gen_wa_oob to be multi-use")
introduced a call to basename(). The GNU version of this function is not
portable and fails to build with alternative libc implementations like
musl or bionic. This causes the following build error:

  drivers/gpu/drm/xe/xe_gen_wa_oob.c:130:12: error: assignment to ‘const char *’ from ‘int’ makes pointer from integer without a cast [-Wint-conversion]
    130 |         fn = basename(fn);
        |            ^

While a POSIX version of basename() could be used, it would require a
separate header plus the behavior differs from GNU version in that it
might modify its argument. Not great.

Instead, implement a local xbasename() helper based on strrchr() that
provides the same functionality and avoids portability issues.

Fixes: b0a2ee5567ab ("drm/xe: prepare xe_gen_wa_oob to be multi-use")
Suggested-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Tiffany Yang <ynaffit@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250825155743.1132433-1-cmllamas@google.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_gen_wa_oob.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_gen_wa_oob.c b/drivers/gpu/drm/xe/xe_gen_wa_oob.c
index 6581cb0f0e59..247e41c1c48d 100644
--- a/drivers/gpu/drm/xe/xe_gen_wa_oob.c
+++ b/drivers/gpu/drm/xe/xe_gen_wa_oob.c
@@ -123,11 +123,19 @@ static int parse(FILE *input, FILE *csource, FILE *cheader, char *prefix)
 	return 0;
 }
 
+/* Avoid GNU vs POSIX basename() discrepancy, just use our own */
+static const char *xbasename(const char *s)
+{
+	const char *p = strrchr(s, '/');
+
+	return p ? p + 1 : s;
+}
+
 static int fn_to_prefix(const char *fn, char *prefix, size_t size)
 {
 	size_t len;
 
-	fn = basename(fn);
+	fn = xbasename(fn);
 	len = strlen(fn);
 
 	if (len > size - 1)

From 231bb0ee7aa5d1f0d077d3a30663f8ffd6860fa3 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:45 +0530
Subject: [PATCH 0606/1455] drm/xe/uapi: Add madvise interface

This commit introduces a new madvise interface to support
driver-specific ioctl operations. The madvise interface allows for more
efficient memory management by providing hints to the driver about the
expected memory usage and pte update policy for gpuvma.

v2 (Matthew/Thomas)
- Drop num_ops support
- Drop purgeable support
- Add kernel-docs
- IOWR/IOW

v3 (Matthew/Thomas)
- Reorder attributes
- use __u16 for migration_policy
- use __u64 for reserved in unions
- Avoid usage of vma

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-2-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 include/uapi/drm/xe_drm.h | 130 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 130 insertions(+)

diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index c721e130c1d2..4e6e9a9164ee 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -81,6 +81,7 @@ extern "C" {
  *  - &DRM_IOCTL_XE_EXEC
  *  - &DRM_IOCTL_XE_WAIT_USER_FENCE
  *  - &DRM_IOCTL_XE_OBSERVATION
+ *  - &DRM_IOCTL_XE_MADVISE
  */
 
 /*
@@ -102,6 +103,7 @@ extern "C" {
 #define DRM_XE_EXEC			0x09
 #define DRM_XE_WAIT_USER_FENCE		0x0a
 #define DRM_XE_OBSERVATION		0x0b
+#define DRM_XE_MADVISE			0x0c
 
 /* Must be kept compact -- no holes */
 
@@ -117,6 +119,7 @@ extern "C" {
 #define DRM_IOCTL_XE_EXEC			DRM_IOW(DRM_COMMAND_BASE + DRM_XE_EXEC, struct drm_xe_exec)
 #define DRM_IOCTL_XE_WAIT_USER_FENCE		DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_WAIT_USER_FENCE, struct drm_xe_wait_user_fence)
 #define DRM_IOCTL_XE_OBSERVATION		DRM_IOW(DRM_COMMAND_BASE + DRM_XE_OBSERVATION, struct drm_xe_observation_param)
+#define DRM_IOCTL_XE_MADVISE			DRM_IOW(DRM_COMMAND_BASE + DRM_XE_MADVISE, struct drm_xe_madvise)
 
 /**
  * DOC: Xe IOCTL Extensions
@@ -1978,6 +1981,133 @@ struct drm_xe_query_eu_stall {
 	__u64 sampling_rates[];
 };
 
+/**
+ * struct drm_xe_madvise - Input of &DRM_IOCTL_XE_MADVISE
+ *
+ * This structure is used to set memory attributes for a virtual address range
+ * in a VM. The type of attribute is specified by @type, and the corresponding
+ * union member is used to provide additional parameters for @type.
+ *
+ * Supported attribute types:
+ * - DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC: Set preferred memory location.
+ * - DRM_XE_MEM_RANGE_ATTR_ATOMIC: Set atomic access policy.
+ * - DRM_XE_MEM_RANGE_ATTR_PAT: Set page attribute table index.
+ *
+ * Example:
+ *
+ * .. code-block:: C
+ *
+ * struct drm_xe_madvise madvise = {
+ *          .vm_id = vm_id,
+ *          .start = 0x100000,
+ *          .range = 0x2000,
+ *          .type = DRM_XE_MEM_RANGE_ATTR_ATOMIC,
+ *          .atomic_val = DRM_XE_ATOMIC_DEVICE,
+ *         };
+ *
+ * ioctl(fd, DRM_IOCTL_XE_MADVISE, &madvise);
+ *
+ */
+struct drm_xe_madvise {
+	/** @extensions: Pointer to the first extension struct, if any */
+	__u64 extensions;
+
+	/** @start: start of the virtual address range */
+	__u64 start;
+
+	/** @range: size of the virtual address range */
+	__u64 range;
+
+	/** @vm_id: vm_id of the virtual range */
+	__u32 vm_id;
+
+#define DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC	0
+#define DRM_XE_MEM_RANGE_ATTR_ATOMIC		1
+#define DRM_XE_MEM_RANGE_ATTR_PAT		2
+	/** @type: type of attribute */
+	__u32 type;
+
+	union {
+		/**
+		 * @preferred_mem_loc: preferred memory location
+		 *
+		 * Used when @type == DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC
+		 *
+		 * Supported values for @preferred_mem_loc.devmem_fd:
+		 * - DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE: set vram of faulting tile as preferred loc
+		 * - DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM: set smem as preferred loc
+		 *
+		 * Supported values for @preferred_mem_loc.migration_policy:
+		 * - DRM_XE_MIGRATE_ALL_PAGES
+		 * - DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES
+		 */
+		struct {
+#define DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE	0
+#define DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM	-1
+			/** @preferred_mem_loc.devmem_fd: fd for preferred loc */
+			__u32 devmem_fd;
+
+#define DRM_XE_MIGRATE_ALL_PAGES		0
+#define DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES	1
+			/** @preferred_mem_loc.migration_policy: Page migration policy */
+			__u16 migration_policy;
+
+			/** @preferred_mem_loc.pad : MBZ */
+			__u16 pad;
+
+			/** @preferred_mem_loc.reserved : Reserved */
+			__u64 reserved;
+		} preferred_mem_loc;
+
+		/**
+		 * @atomic: Atomic access policy
+		 *
+		 * Used when @type == DRM_XE_MEM_RANGE_ATTR_ATOMIC.
+		 *
+		 * Supported values for @atomic.val:
+		 * - DRM_XE_ATOMIC_UNDEFINED: Undefined or default behaviour
+		 *   Support both GPU and CPU atomic operations for system allocator
+		 *   Support GPU atomic operations for normal(bo) allocator
+		 * - DRM_XE_ATOMIC_DEVICE: Support GPU atomic operations
+		 * - DRM_XE_ATOMIC_GLOBAL: Support both GPU and CPU atomic operations
+		 * - DRM_XE_ATOMIC_CPU: Support CPU atomic
+		 */
+		struct {
+#define DRM_XE_ATOMIC_UNDEFINED	0
+#define DRM_XE_ATOMIC_DEVICE	1
+#define DRM_XE_ATOMIC_GLOBAL	2
+#define DRM_XE_ATOMIC_CPU	3
+			/** @atomic.val: value of atomic operation */
+			__u32 val;
+
+			/** @atomic.pad: MBZ */
+			__u32 pad;
+
+			/** @atomic.reserved: Reserved */
+			__u64 reserved;
+		} atomic;
+
+		/**
+		 * @pat_index: Page attribute table index
+		 *
+		 * Used when @type == DRM_XE_MEM_RANGE_ATTR_PAT.
+		 */
+		struct {
+			/** @pat_index.val: PAT index value */
+			__u32 val;
+
+			/** @pat_index.pad: MBZ */
+			__u32 pad;
+
+			/** @pat_index.reserved: Reserved */
+			__u64 reserved;
+		} pat_index;
+	};
+
+	/** @reserved: Reserved */
+	__u64 reserved[2];
+};
+
 #if defined(__cplusplus)
 }
 #endif

From 99a89e4e2d0c4e806af3dea4538c1fc9b208df3b Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:46 +0530
Subject: [PATCH 0607/1455] drm/xe/vm: Add attributes struct as member of vma
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The attribute of xe_vma will determine the migration policy and the
encoding of the page table entries (PTEs) for that vma.
This attribute helps manage how memory pages are moved and how their
addresses are translated. It will be used by madvise to set the
behavior of the vma.

v2 (Matthew Brost)
- Add docs

v3 (Matthew Brost)
- Add uapi references
- 80 characters line wrap

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-3-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm_types.h | 33 ++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 8a07feef503b..8e8138c2b80a 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -77,6 +77,33 @@ struct xe_userptr {
 #endif
 };
 
+/**
+ * struct xe_vma_mem_attr - memory attributes associated with vma
+ */
+struct xe_vma_mem_attr {
+	/** @preferred_loc: perferred memory_location */
+	struct {
+		/** @preferred_loc.migration_policy: Pages migration policy */
+		u32 migration_policy;
+
+		/**
+		 * @preferred_loc.devmem_fd: used for determining pagemap_fd
+		 * requested by user DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM and
+		 * DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE mean system memory or
+		 * closest device memory respectively.
+		 */
+		u32 devmem_fd;
+	} preferred_loc;
+
+	/**
+	 * @atomic_access: The atomic access type for the vma
+	 * See %DRM_XE_VMA_ATOMIC_UNDEFINED, %DRM_XE_VMA_ATOMIC_DEVICE,
+	 * %DRM_XE_VMA_ATOMIC_GLOBAL, and %DRM_XE_VMA_ATOMIC_CPU for possible
+	 * values. These are defined in uapi/drm/xe_drm.h.
+	 */
+	u32 atomic_access;
+};
+
 struct xe_vma {
 	/** @gpuva: Base GPUVA object */
 	struct drm_gpuva gpuva;
@@ -135,6 +162,12 @@ struct xe_vma {
 	 * Needs to be signalled before UNMAP can be processed.
 	 */
 	struct xe_user_fence *ufence;
+
+	/**
+	 * @attr: The attributes of vma which determines the migration policy
+	 * and encoding of the PTEs for this vma.
+	 */
+	struct xe_vma_mem_attr attr;
 };
 
 /**

From 11974fe8c7eac3ee3f35dfe6696f097284bd67eb Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:47 +0530
Subject: [PATCH 0608/1455] drm/xe/vma: Move pat_index to vma attributes

The PAT index determines how PTEs are encoded and can be modified by
madvise. Therefore, it is now part of the vma attributes.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-4-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_pt.c       |  2 +-
 drivers/gpu/drm/xe/xe_vm.c       |  6 +++---
 drivers/gpu/drm/xe/xe_vm_types.h | 10 +++++-----
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index f3a39e734a90..ba7a50bf3a2d 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -518,7 +518,7 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset,
 {
 	struct xe_pt_stage_bind_walk *xe_walk =
 		container_of(walk, typeof(*xe_walk), base);
-	u16 pat_index = xe_walk->vma->pat_index;
+	u16 pat_index = xe_walk->vma->attr.pat_index;
 	struct xe_pt *xe_parent = container_of(parent, typeof(*xe_parent), base);
 	struct xe_vm *vm = xe_walk->vm;
 	struct xe_pt *xe_child;
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 529b6767caac..bb831629d149 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1223,7 +1223,7 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm,
 	if (vm->xe->info.has_atomic_enable_pte_bit)
 		vma->gpuva.flags |= XE_VMA_ATOMIC_PTE_BIT;
 
-	vma->pat_index = pat_index;
+	vma->attr.pat_index = pat_index;
 
 	if (bo) {
 		struct drm_gpuvm_bo *vm_bo;
@@ -2704,7 +2704,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 
 			if (op->base.remap.prev) {
 				vma = new_vma(vm, op->base.remap.prev,
-					      old->pat_index, flags);
+					      old->attr.pat_index, flags);
 				if (IS_ERR(vma))
 					return PTR_ERR(vma);
 
@@ -2734,7 +2734,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 
 			if (op->base.remap.next) {
 				vma = new_vma(vm, op->base.remap.next,
-					      old->pat_index, flags);
+					      old->attr.pat_index, flags);
 				if (IS_ERR(vma))
 					return PTR_ERR(vma);
 
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 8e8138c2b80a..c7b2bfa0a0d1 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -102,6 +102,11 @@ struct xe_vma_mem_attr {
 	 * values. These are defined in uapi/drm/xe_drm.h.
 	 */
 	u32 atomic_access;
+
+	/**
+	 * @pat_index: The pat index to use when encoding the PTEs for this vma.
+	 */
+	u16 pat_index;
 };
 
 struct xe_vma {
@@ -152,11 +157,6 @@ struct xe_vma {
 	/** @tile_staged: bind is staged for this VMA */
 	u8 tile_staged;
 
-	/**
-	 * @pat_index: The pat index to use when encoding the PTEs for this vma.
-	 */
-	u16 pat_index;
-
 	/**
 	 * @ufence: The user fence that was provided with MAP.
 	 * Needs to be signalled before UNMAP can be processed.

From 29c39c56a08b238d1a069cd0e16adfbef30c4968 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:48 +0530
Subject: [PATCH 0609/1455] drm/xe/vma: Modify new_vma to accept struct
 xe_vma_mem_attr as parameter

This change simplifies the logic by ensuring that remapped previous or
next VMAs are created with the same memory attributes as the original VMA.
By passing struct xe_vma_mem_attr as a parameter, we maintain consistency
in memory attributes.

-v2
 *dst = *src (Matthew Brost)

-v3 (Matthew Brost)
 Drop unnecessary helper
 pass attr ptr as input to new_vma and vma_create

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-5-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c | 24 +++++++++++++++++-------
 1 file changed, 17 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index bb831629d149..12b478e7ef01 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1168,7 +1168,8 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm,
 				    struct xe_bo *bo,
 				    u64 bo_offset_or_userptr,
 				    u64 start, u64 end,
-				    u16 pat_index, unsigned int flags)
+				    struct xe_vma_mem_attr *attr,
+				    unsigned int flags)
 {
 	struct xe_vma *vma;
 	struct xe_tile *tile;
@@ -1223,7 +1224,7 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm,
 	if (vm->xe->info.has_atomic_enable_pte_bit)
 		vma->gpuva.flags |= XE_VMA_ATOMIC_PTE_BIT;
 
-	vma->attr.pat_index = pat_index;
+	vma->attr = *attr;
 
 	if (bo) {
 		struct drm_gpuvm_bo *vm_bo;
@@ -2475,7 +2476,7 @@ unwind_prefetch_ops:
 ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_create, ERRNO);
 
 static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op,
-			      u16 pat_index, unsigned int flags)
+			      struct xe_vma_mem_attr *attr, unsigned int flags)
 {
 	struct xe_bo *bo = op->gem.obj ? gem_to_xe_bo(op->gem.obj) : NULL;
 	struct drm_exec exec;
@@ -2504,7 +2505,7 @@ static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op,
 	}
 	vma = xe_vma_create(vm, bo, op->gem.offset,
 			    op->va.addr, op->va.addr +
-			    op->va.range - 1, pat_index, flags);
+			    op->va.range - 1, attr, flags);
 	if (IS_ERR(vma))
 		goto err_unlock;
 
@@ -2647,6 +2648,15 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 		switch (op->base.op) {
 		case DRM_GPUVA_OP_MAP:
 		{
+			struct xe_vma_mem_attr default_attr = {
+				.preferred_loc = {
+					.devmem_fd = DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE,
+					.migration_policy = DRM_XE_MIGRATE_ALL_PAGES,
+				},
+				.atomic_access = DRM_XE_ATOMIC_UNDEFINED,
+				.pat_index = op->map.pat_index,
+			};
+
 			flags |= op->map.read_only ?
 				VMA_CREATE_FLAG_READ_ONLY : 0;
 			flags |= op->map.is_null ?
@@ -2656,7 +2666,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 			flags |= op->map.is_cpu_addr_mirror ?
 				VMA_CREATE_FLAG_IS_SYSTEM_ALLOCATOR : 0;
 
-			vma = new_vma(vm, &op->base.map, op->map.pat_index,
+			vma = new_vma(vm, &op->base.map, &default_attr,
 				      flags);
 			if (IS_ERR(vma))
 				return PTR_ERR(vma);
@@ -2704,7 +2714,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 
 			if (op->base.remap.prev) {
 				vma = new_vma(vm, op->base.remap.prev,
-					      old->attr.pat_index, flags);
+					      &old->attr, flags);
 				if (IS_ERR(vma))
 					return PTR_ERR(vma);
 
@@ -2734,7 +2744,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 
 			if (op->base.remap.next) {
 				vma = new_vma(vm, op->base.remap.next,
-					      old->attr.pat_index, flags);
+					      &old->attr, flags);
 				if (IS_ERR(vma))
 					return PTR_ERR(vma);
 

From 186b526abd6c20713d0dc9992ad736027e866a8d Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:49 +0530
Subject: [PATCH 0610/1455] drm/xe/svm: Split system allocator vma incase of
 madvise call

If the start or end of input address range lies within system allocator
vma split the vma to create new vma's as per input range.

v2 (Matthew Brost)
- Add lockdep_assert_write for vm->lock
- Remove unnecessary page aligned checks
- Add kerrnel-doc and comments
- Remove unnecessary unwind_ops and return

v3
- Fix copying of attributes

v4
- Nit fixes

v5
- Squash identifier for madvise in xe_vma_ops to this patch

v6/v7/v8
- Rebase on drm_gpuvm changes

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-6-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c       | 108 +++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vm.h       |   2 +
 drivers/gpu/drm/xe/xe_vm_types.h |   1 +
 3 files changed, 111 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 12b478e7ef01..9c76cb31025d 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -4207,3 +4207,111 @@ void xe_vm_snapshot_free(struct xe_vm_snapshot *snap)
 	}
 	kvfree(snap);
 }
+
+/**
+ * xe_vm_alloc_madvise_vma - Allocate VMA's with madvise ops
+ * @vm: Pointer to the xe_vm structure
+ * @start: Starting input address
+ * @range: Size of the input range
+ *
+ * This function splits existing vma to create new vma for user provided input range
+ *
+ *  Return: 0 if success
+ */
+int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
+{
+	struct drm_gpuvm_map_req map_req = {
+		.map.va.addr = start,
+		.map.va.range = range,
+	};
+
+	struct xe_vma_ops vops;
+	struct drm_gpuva_ops *ops = NULL;
+	struct drm_gpuva_op *__op;
+	bool is_cpu_addr_mirror = false;
+	bool remap_op = false;
+	struct xe_vma_mem_attr tmp_attr;
+	int err;
+
+	lockdep_assert_held_write(&vm->lock);
+
+	vm_dbg(&vm->xe->drm, "MADVISE_OPS_CREATE: addr=0x%016llx, size=0x%016llx", start, range);
+	ops = drm_gpuvm_madvise_ops_create(&vm->gpuvm, &map_req);
+	if (IS_ERR(ops))
+		return PTR_ERR(ops);
+
+	if (list_empty(&ops->list)) {
+		err = 0;
+		goto free_ops;
+	}
+
+	drm_gpuva_for_each_op(__op, ops) {
+		struct xe_vma_op *op = gpuva_op_to_vma_op(__op);
+
+		if (__op->op == DRM_GPUVA_OP_REMAP) {
+			xe_assert(vm->xe, !remap_op);
+			remap_op = true;
+
+			if (xe_vma_is_cpu_addr_mirror(gpuva_to_vma(op->base.remap.unmap->va)))
+				is_cpu_addr_mirror = true;
+			else
+				is_cpu_addr_mirror = false;
+		}
+
+		if (__op->op == DRM_GPUVA_OP_MAP) {
+			xe_assert(vm->xe, remap_op);
+			remap_op = false;
+
+			/* In case of madvise ops DRM_GPUVA_OP_MAP is always after
+			 * DRM_GPUVA_OP_REMAP, so ensure we assign op->map.is_cpu_addr_mirror true
+			 * if REMAP is for xe_vma_is_cpu_addr_mirror vma
+			 */
+			op->map.is_cpu_addr_mirror = is_cpu_addr_mirror;
+		}
+
+		print_op(vm->xe, __op);
+	}
+
+	xe_vma_ops_init(&vops, vm, NULL, NULL, 0);
+	vops.flags |= XE_VMA_OPS_FLAG_MADVISE;
+	err = vm_bind_ioctl_ops_parse(vm, ops, &vops);
+	if (err)
+		goto unwind_ops;
+
+	xe_vm_lock(vm, false);
+
+	drm_gpuva_for_each_op(__op, ops) {
+		struct xe_vma_op *op = gpuva_op_to_vma_op(__op);
+		struct xe_vma *vma;
+
+		if (__op->op == DRM_GPUVA_OP_UNMAP) {
+			/* There should be no unmap */
+			XE_WARN_ON("UNEXPECTED UNMAP");
+			xe_vma_destroy(gpuva_to_vma(op->base.unmap.va), NULL);
+		} else if (__op->op == DRM_GPUVA_OP_REMAP) {
+			vma = gpuva_to_vma(op->base.remap.unmap->va);
+			/* Store attributes for REMAP UNMAPPED VMA, so they can be assigned
+			 * to newly MAP created vma.
+			 */
+			tmp_attr = vma->attr;
+			xe_vma_destroy(gpuva_to_vma(op->base.remap.unmap->va), NULL);
+		} else if (__op->op == DRM_GPUVA_OP_MAP) {
+			vma = op->map.vma;
+			/* In case of madvise call, MAP will always be follwed by REMAP.
+			 * Therefore temp_attr will always have sane values, making it safe to
+			 * copy them to new vma.
+			 */
+			vma->attr = tmp_attr;
+		}
+	}
+
+	xe_vm_unlock(vm);
+	drm_gpuva_ops_free(&vm->gpuvm, ops);
+	return 0;
+
+unwind_ops:
+	vm_bind_ioctl_ops_unwind(vm, &ops, 1);
+free_ops:
+	drm_gpuva_ops_free(&vm->gpuvm, ops);
+	return err;
+}
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 2ecb417c19a2..ccc549e39cfc 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -171,6 +171,8 @@ static inline bool xe_vma_is_userptr(struct xe_vma *vma)
 
 struct xe_vma *xe_vm_find_vma_by_addr(struct xe_vm *vm, u64 page_addr);
 
+int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t addr, uint64_t size);
+
 /**
  * to_userptr_vma() - Return a pointer to an embedding userptr vma
  * @vma: Pointer to the embedded struct xe_vma
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index c7b2bfa0a0d1..dde7218ceba6 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -495,6 +495,7 @@ struct xe_vma_ops {
 	struct xe_vm_pgtable_update_ops pt_update_ops[XE_MAX_TILES_PER_DEVICE];
 	/** @flag: signify the properties within xe_vma_ops*/
 #define XE_VMA_OPS_FLAG_HAS_SVM_PREFETCH BIT(0)
+#define XE_VMA_OPS_FLAG_MADVISE          BIT(1)
 	u32 flags;
 #ifdef TEST_VM_OPS_ERROR
 	/** @inject_error: inject error to test error handling */

From 6ad887f378c858ef3f721d5c1d8dad17842cc58b Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:50 +0530
Subject: [PATCH 0611/1455] drm/xe: Allow CPU address mirror VMA unbind with
 gpu bindings for madvise

In the case of the MADVISE ioctl, if the start or end addresses fall
within a VMA and existing SVM ranges are present, remove the existing
SVM mappings. Then, continue with ops_parse to create new VMAs by REMAP
unmapping of old one.

v2 (Matthew Brost)
- Use vops flag to call unmapping of ranges in vm_bind_ioctl_ops_parse
- Rename the function

v3
- Fix doc

v4
- check if range is already in garbage collector (Matthew Brost)

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-7-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_svm.c | 35 +++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_svm.h |  7 +++++++
 drivers/gpu/drm/xe/xe_vm.c  |  8 ++++++--
 3 files changed, 48 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index e35c6d4def20..ce42100cb753 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -932,6 +932,41 @@ bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end)
 	return drm_gpusvm_has_mapping(&vm->svm.gpusvm, start, end);
 }
 
+/**
+ * xe_svm_unmap_address_range - UNMAP SVM mappings and ranges
+ * @vm: The VM
+ * @start: start addr
+ * @end: end addr
+ *
+ * This function UNMAPS svm ranges if start or end address are inside them.
+ */
+void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end)
+{
+	struct drm_gpusvm_notifier *notifier, *next;
+
+	lockdep_assert_held_write(&vm->lock);
+
+	drm_gpusvm_for_each_notifier_safe(notifier, next, &vm->svm.gpusvm, start, end) {
+		struct drm_gpusvm_range *range, *__next;
+
+		drm_gpusvm_for_each_range_safe(range, __next, notifier, start, end) {
+			if (start > drm_gpusvm_range_start(range) ||
+			    end < drm_gpusvm_range_end(range)) {
+				if (IS_DGFX(vm->xe) && xe_svm_range_in_vram(to_xe_range(range)))
+					drm_gpusvm_range_evict(&vm->svm.gpusvm, range);
+				drm_gpusvm_range_get(range);
+				__xe_svm_garbage_collector(vm, to_xe_range(range));
+				if (!list_empty(&to_xe_range(range)->garbage_collector_link)) {
+					spin_lock(&vm->svm.garbage_collector.lock);
+					list_del(&to_xe_range(range)->garbage_collector_link);
+					spin_unlock(&vm->svm.garbage_collector.lock);
+				}
+				drm_gpusvm_range_put(range);
+			}
+		}
+	}
+}
+
 /**
  * xe_svm_bo_evict() - SVM evict BO to system memory
  * @bo: BO to evict
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index 4bdccb56d25f..184b3f4f0b2a 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -90,6 +90,8 @@ bool xe_svm_range_validate(struct xe_vm *vm,
 
 u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 addr, u64 end,  struct xe_vma *vma);
 
+void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end);
+
 /**
  * xe_svm_range_has_dma_mapping() - SVM range has DMA mapping
  * @range: SVM range
@@ -303,6 +305,11 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 addr, u64 end, struct xe_vma *vm
 	return ULONG_MAX;
 }
 
+static inline
+void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end)
+{
+}
+
 #define xe_svm_assert_in_notifier(...) do {} while (0)
 #define xe_svm_range_has_dma_mapping(...) false
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 9c76cb31025d..9fef01400c74 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2694,8 +2694,12 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 				end = op->base.remap.next->va.addr;
 
 			if (xe_vma_is_cpu_addr_mirror(old) &&
-			    xe_svm_has_mapping(vm, start, end))
-				return -EBUSY;
+			    xe_svm_has_mapping(vm, start, end)) {
+				if (vops->flags & XE_VMA_OPS_FLAG_MADVISE)
+					xe_svm_unmap_address_range(vm, start, end);
+				else
+					return -EBUSY;
+			}
 
 			op->remap.start = xe_vma_start(old);
 			op->remap.range = xe_vma_size(old);

From 6ca463ef0d97a20eba559e699f0f6c4dd71e5011 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:51 +0530
Subject: [PATCH 0612/1455] drm/xe/svm: Add xe_svm_ranges_zap_ptes_in_range()
 for PTE zapping

Introduce xe_svm_ranges_zap_ptes_in_range(), a function to zap page table
entries (PTEs) for all SVM ranges within a user-specified address range.

-v2 (Matthew Brost)
Lock should be called even for tlb_invalidation

v3(Matthew Brost)
- Update comment
- s/notifier->itree.start/drm_gpusvm_notifier_start
- s/notifier->itree.last + 1/drm_gpusvm_notifier_end
- use WRITE_ONCE

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-8-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_pt.c  | 14 ++++++++++-
 drivers/gpu/drm/xe/xe_svm.c | 50 +++++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_svm.h |  8 ++++++
 3 files changed, 71 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index ba7a50bf3a2d..bf50a821853e 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -950,7 +950,19 @@ bool xe_pt_zap_ptes_range(struct xe_tile *tile, struct xe_vm *vm,
 	struct xe_pt *pt = vm->pt_root[tile->id];
 	u8 pt_mask = (range->tile_present & ~range->tile_invalidated);
 
-	xe_svm_assert_in_notifier(vm);
+	/*
+	 * Locking rules:
+	 *
+	 * - notifier_lock (write): full protection against page table changes
+	 *   and MMU notifier invalidations.
+	 *
+	 * - notifier_lock (read) + vm_lock (write): combined protection against
+	 *   invalidations and concurrent page table modifications. (e.g., madvise)
+	 *
+	 */
+	lockdep_assert(lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 0) ||
+		       (lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 1) &&
+		       lockdep_is_held_type(&vm->lock, 0)));
 
 	if (!(pt_mask & BIT(tile->id)))
 		return false;
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index ce42100cb753..c2306000f15e 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -1031,6 +1031,56 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range,
 	return err;
 }
 
+/**
+ * xe_svm_ranges_zap_ptes_in_range - clear ptes of svm ranges in input range
+ * @vm: Pointer to the xe_vm structure
+ * @start: Start of the input range
+ * @end: End of the input range
+ *
+ * This function removes the page table entries (PTEs) associated
+ * with the svm ranges within the given input start and end
+ *
+ * Return: tile_mask for which gt's need to be tlb invalidated.
+ */
+u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end)
+{
+	struct drm_gpusvm_notifier *notifier;
+	struct xe_svm_range *range;
+	u64 adj_start, adj_end;
+	struct xe_tile *tile;
+	u8 tile_mask = 0;
+	u8 id;
+
+	lockdep_assert(lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 1) &&
+		       lockdep_is_held_type(&vm->lock, 0));
+
+	drm_gpusvm_for_each_notifier(notifier, &vm->svm.gpusvm, start, end) {
+		struct drm_gpusvm_range *r = NULL;
+
+		adj_start = max(start, drm_gpusvm_notifier_start(notifier));
+		adj_end = min(end, drm_gpusvm_notifier_end(notifier));
+		drm_gpusvm_for_each_range(r, notifier, adj_start, adj_end) {
+			range = to_xe_range(r);
+			for_each_tile(tile, vm->xe, id) {
+				if (xe_pt_zap_ptes_range(tile, vm, range)) {
+					tile_mask |= BIT(id);
+					/*
+					 * WRITE_ONCE pairs with READ_ONCE in
+					 * xe_vm_has_valid_gpu_mapping().
+					 * Must not fail after setting
+					 * tile_invalidated and before
+					 * TLB invalidation.
+					 */
+					WRITE_ONCE(range->tile_invalidated,
+						   range->tile_invalidated | BIT(id));
+				}
+			}
+		}
+	}
+
+	return tile_mask;
+}
+
 #if IS_ENABLED(CONFIG_DRM_XE_PAGEMAP)
 
 static struct drm_pagemap *tile_local_pagemap(struct xe_tile *tile)
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index 184b3f4f0b2a..046a9c4e95c2 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -92,6 +92,8 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 addr, u64 end,  struct xe_vma *v
 
 void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end);
 
+u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end);
+
 /**
  * xe_svm_range_has_dma_mapping() - SVM range has DMA mapping
  * @range: SVM range
@@ -310,6 +312,12 @@ void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end)
 {
 }
 
+static inline
+u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end)
+{
+	return 0;
+}
+
 #define xe_svm_assert_in_notifier(...) do {} while (0)
 #define xe_svm_range_has_dma_mapping(...) false
 

From ada7486c5668db542a7d361268df931aca5b726a Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:52 +0530
Subject: [PATCH 0613/1455] drm/xe: Implement madvise ioctl for xe

This driver-specific ioctl enables UMDs to control the memory attributes
for GPU VMAs within a specified input range. If the start or end
addresses fall within an existing VMA, the VMA is split accordingly. The
attributes of the VMA are modified as provided by the users. The old
mappings of the VMAs are invalidated, and TLB invalidation is performed
if necessary.

v2(Matthew brost)
- xe_vm_in_fault_mode can't be enabled by Mesa, hence allow ioctl in non
fault mode too
- fix tlb invalidation skip for same ranges in multiple op
- use helper for tlb invalidation
- use xe_svm_notifier_lock/unlock helper
- s/lockdep_assert_held/lockdep_assert_held_write
- Add kernel-doc

v3(Matthew Brost)
- make vfunc fail safe
- Add sanitizing input args before vfunc

v4(Matthew Brost/Shuicheng)
- Make locks interruptable
- Error handling fixes
- vm_put fixes

v5(Matthew Brost)
- Flush garbage collector before any locking.
- Add check for null vma

Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-9-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/Makefile        |   1 +
 drivers/gpu/drm/xe/xe_vm_madvise.c | 308 +++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vm_madvise.h |  15 ++
 3 files changed, 324 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_vm_madvise.c
 create mode 100644 drivers/gpu/drm/xe/xe_vm_madvise.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 85b8d3a59ef0..c52cac979e05 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -129,6 +129,7 @@ xe-y += xe_bb.o \
 	xe_uc.o \
 	xe_uc_fw.o \
 	xe_vm.o \
+	xe_vm_madvise.o \
 	xe_vram.o \
 	xe_vram_freq.o \
 	xe_vsec.o \
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
new file mode 100644
index 000000000000..b861c3349b0a
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -0,0 +1,308 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "xe_vm_madvise.h"
+
+#include <linux/nospec.h>
+#include <drm/xe_drm.h>
+
+#include "xe_bo.h"
+#include "xe_pt.h"
+#include "xe_svm.h"
+
+struct xe_vmas_in_madvise_range {
+	u64 addr;
+	u64 range;
+	struct xe_vma **vmas;
+	int num_vmas;
+	bool has_svm_vmas;
+	bool has_bo_vmas;
+	bool has_userptr_vmas;
+};
+
+static int get_vmas(struct xe_vm *vm, struct xe_vmas_in_madvise_range *madvise_range)
+{
+	u64 addr = madvise_range->addr;
+	u64 range = madvise_range->range;
+
+	struct xe_vma  **__vmas;
+	struct drm_gpuva *gpuva;
+	int max_vmas = 8;
+
+	lockdep_assert_held(&vm->lock);
+
+	madvise_range->num_vmas = 0;
+	madvise_range->vmas = kmalloc_array(max_vmas, sizeof(*madvise_range->vmas), GFP_KERNEL);
+	if (!madvise_range->vmas)
+		return -ENOMEM;
+
+	vm_dbg(&vm->xe->drm, "VMA's in range: start=0x%016llx, end=0x%016llx", addr, addr + range);
+
+	drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, addr, addr + range) {
+		struct xe_vma *vma = gpuva_to_vma(gpuva);
+
+		if (xe_vma_bo(vma))
+			madvise_range->has_bo_vmas = true;
+		else if (xe_vma_is_cpu_addr_mirror(vma))
+			madvise_range->has_svm_vmas = true;
+		else if (xe_vma_is_userptr(vma))
+			madvise_range->has_userptr_vmas = true;
+
+		if (madvise_range->num_vmas == max_vmas) {
+			max_vmas <<= 1;
+			__vmas = krealloc(madvise_range->vmas,
+					  max_vmas * sizeof(*madvise_range->vmas),
+					  GFP_KERNEL);
+			if (!__vmas) {
+				kfree(madvise_range->vmas);
+				return -ENOMEM;
+			}
+			madvise_range->vmas = __vmas;
+		}
+
+		madvise_range->vmas[madvise_range->num_vmas] = vma;
+		(madvise_range->num_vmas)++;
+	}
+
+	if (!madvise_range->num_vmas)
+		kfree(madvise_range->vmas);
+
+	vm_dbg(&vm->xe->drm, "madvise_range-num_vmas = %d\n", madvise_range->num_vmas);
+
+	return 0;
+}
+
+static void madvise_preferred_mem_loc(struct xe_device *xe, struct xe_vm *vm,
+				      struct xe_vma **vmas, int num_vmas,
+				      struct drm_xe_madvise *op)
+{
+	/* Implementation pending */
+}
+
+static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
+			   struct xe_vma **vmas, int num_vmas,
+			   struct drm_xe_madvise *op)
+{
+	/* Implementation pending */
+}
+
+static void madvise_pat_index(struct xe_device *xe, struct xe_vm *vm,
+			      struct xe_vma **vmas, int num_vmas,
+			      struct drm_xe_madvise *op)
+{
+	/* Implementation pending */
+}
+
+typedef void (*madvise_func)(struct xe_device *xe, struct xe_vm *vm,
+			     struct xe_vma **vmas, int num_vmas,
+			     struct drm_xe_madvise *op);
+
+static const madvise_func madvise_funcs[] = {
+	[DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC] = madvise_preferred_mem_loc,
+	[DRM_XE_MEM_RANGE_ATTR_ATOMIC] = madvise_atomic,
+	[DRM_XE_MEM_RANGE_ATTR_PAT] = madvise_pat_index,
+};
+
+static u8 xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end)
+{
+	struct drm_gpuva *gpuva;
+	struct xe_tile *tile;
+	u8 id, tile_mask;
+
+	lockdep_assert_held_write(&vm->lock);
+
+	/* Wait for pending binds */
+	if (dma_resv_wait_timeout(xe_vm_resv(vm), DMA_RESV_USAGE_BOOKKEEP,
+				  false, MAX_SCHEDULE_TIMEOUT) <= 0)
+		XE_WARN_ON(1);
+
+	tile_mask = xe_svm_ranges_zap_ptes_in_range(vm, start, end);
+
+	drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) {
+		struct xe_vma *vma = gpuva_to_vma(gpuva);
+
+		if (xe_vma_is_cpu_addr_mirror(vma) || xe_vma_is_null(vma))
+			continue;
+
+		for_each_tile(tile, vm->xe, id) {
+			if (xe_pt_zap_ptes(tile, vma)) {
+				tile_mask |= BIT(id);
+
+				/*
+				 * WRITE_ONCE pairs with READ_ONCE
+				 * in xe_vm_has_valid_gpu_mapping()
+				 */
+				WRITE_ONCE(vma->tile_invalidated,
+					   vma->tile_invalidated | BIT(id));
+			}
+		}
+	}
+
+	return tile_mask;
+}
+
+static int xe_vm_invalidate_madvise_range(struct xe_vm *vm, u64 start, u64 end)
+{
+	u8 tile_mask = xe_zap_ptes_in_madvise_range(vm, start, end);
+
+	if (!tile_mask)
+		return 0;
+
+	xe_device_wmb(vm->xe);
+
+	return xe_vm_range_tilemask_tlb_invalidation(vm, start, end, tile_mask);
+}
+
+static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madvise *args)
+{
+	if (XE_IOCTL_DBG(xe, !args))
+		return false;
+
+	if (XE_IOCTL_DBG(xe, !IS_ALIGNED(args->start, SZ_4K)))
+		return false;
+
+	if (XE_IOCTL_DBG(xe, !IS_ALIGNED(args->range, SZ_4K)))
+		return false;
+
+	if (XE_IOCTL_DBG(xe, args->range < SZ_4K))
+		return false;
+
+	switch (args->type) {
+	case DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC:
+		if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.migration_policy >
+				     DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.pad))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, args->atomic.reserved))
+			return false;
+		break;
+	case DRM_XE_MEM_RANGE_ATTR_ATOMIC:
+		if (XE_IOCTL_DBG(xe, args->atomic.val > DRM_XE_ATOMIC_CPU))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, args->atomic.pad))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, args->atomic.reserved))
+			return false;
+
+		break;
+	case DRM_XE_MEM_RANGE_ATTR_PAT:
+		/*TODO: Add valid pat check */
+		break;
+	default:
+		if (XE_IOCTL_DBG(xe, 1))
+			return false;
+	}
+
+	if (XE_IOCTL_DBG(xe, args->reserved[0] || args->reserved[1]))
+		return false;
+
+	return true;
+}
+
+/**
+ * xe_vm_madvise_ioctl - Handle MADVise ioctl for a VM
+ * @dev: DRM device pointer
+ * @data: Pointer to ioctl data (drm_xe_madvise*)
+ * @file: DRM file pointer
+ *
+ * Handles the MADVISE ioctl to provide memory advice for vma's within
+ * input range.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct xe_device *xe = to_xe_device(dev);
+	struct xe_file *xef = to_xe_file(file);
+	struct drm_xe_madvise *args = data;
+	struct xe_vmas_in_madvise_range madvise_range = {.addr = args->start,
+							 .range =  args->range, };
+	struct xe_vm *vm;
+	struct drm_exec exec;
+	int err, attr_type;
+
+	vm = xe_vm_lookup(xef, args->vm_id);
+	if (XE_IOCTL_DBG(xe, !vm))
+		return -EINVAL;
+
+	if (!madvise_args_are_sane(vm->xe, args)) {
+		err = -EINVAL;
+		goto put_vm;
+	}
+
+	xe_svm_flush(vm);
+
+	err = down_write_killable(&vm->lock);
+	if (err)
+		goto put_vm;
+
+	if (XE_IOCTL_DBG(xe, xe_vm_is_closed_or_banned(vm))) {
+		err = -ENOENT;
+		goto unlock_vm;
+	}
+
+	err = xe_vm_alloc_madvise_vma(vm, args->start, args->range);
+	if (err)
+		goto unlock_vm;
+
+	err = get_vmas(vm, &madvise_range);
+	if (err || !madvise_range.num_vmas)
+		goto unlock_vm;
+
+	if (madvise_range.has_bo_vmas) {
+		drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES | DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
+		drm_exec_until_all_locked(&exec) {
+			for (int i = 0; i < madvise_range.num_vmas; i++) {
+				struct xe_bo *bo = xe_vma_bo(madvise_range.vmas[i]);
+
+				if (!bo)
+					continue;
+				err = drm_exec_lock_obj(&exec, &bo->ttm.base);
+				drm_exec_retry_on_contention(&exec);
+				if (err)
+					goto err_fini;
+			}
+		}
+	}
+
+	if (madvise_range.has_userptr_vmas) {
+		err = down_read_interruptible(&vm->userptr.notifier_lock);
+		if (err)
+			goto err_fini;
+	}
+
+	if (madvise_range.has_svm_vmas) {
+		err = down_read_interruptible(&vm->svm.gpusvm.notifier_lock);
+		if (err)
+			goto unlock_userptr;
+	}
+
+	attr_type = array_index_nospec(args->type, ARRAY_SIZE(madvise_funcs));
+	madvise_funcs[attr_type](xe, vm, madvise_range.vmas, madvise_range.num_vmas, args);
+
+	err = xe_vm_invalidate_madvise_range(vm, args->start, args->start + args->range);
+
+	if (madvise_range.has_svm_vmas)
+		xe_svm_notifier_unlock(vm);
+
+unlock_userptr:
+	if (madvise_range.has_userptr_vmas)
+		up_read(&vm->userptr.notifier_lock);
+err_fini:
+	if (madvise_range.has_bo_vmas)
+		drm_exec_fini(&exec);
+	kfree(madvise_range.vmas);
+	madvise_range.vmas = NULL;
+unlock_vm:
+	up_write(&vm->lock);
+put_vm:
+	xe_vm_put(vm);
+	return err;
+}
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.h b/drivers/gpu/drm/xe/xe_vm_madvise.h
new file mode 100644
index 000000000000..b0e1fc445f23
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_VM_MADVISE_H_
+#define _XE_VM_MADVISE_H_
+
+struct drm_device;
+struct drm_file;
+
+int xe_vm_madvise_ioctl(struct drm_device *dev, void *data,
+			struct drm_file *file);
+
+#endif

From d6db171167f886d5f173c67044e413e14b02c880 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:53 +0530
Subject: [PATCH 0614/1455] drm/xe/svm: Add svm ranges migration policy on
 atomic access

If the platform does not support atomic access on system memory, and the
ranges are in system memory, but the user requires atomic accesses on
the VMA, then migrate the ranges to VRAM. Apply this policy for prefetch
operations as well.

v2
- Drop unnecessary vm_dbg

v3 (Matthew Brost)
- fix atomic policy
- prefetch shouldn't have any impact of atomic
- bo can be accessed from vma, avoid duplicate parameter

v4 (Matthew Brost)
- Remove TODO comment
- Fix comment
- Dont allow gpu atomic ops when user is setting atomic attr as CPU

v5 (Matthew Brost)
- Fix atomic checks
- Add userptr checks

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-10-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_pt.c         | 23 ++++++++------
 drivers/gpu/drm/xe/xe_svm.c        | 50 ++++++++++++++++++------------
 drivers/gpu/drm/xe/xe_vm.c         | 39 +++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vm.h         |  2 ++
 drivers/gpu/drm/xe/xe_vm_madvise.c | 15 ++++++++-
 5 files changed, 99 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index bf50a821853e..c0a70c80dff9 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -640,28 +640,31 @@ static const struct xe_pt_walk_ops xe_pt_stage_bind_ops = {
  *    - In all other cases device atomics will be disabled with AE=0 until an application
  *      request differently using a ioctl like madvise.
  */
-static bool xe_atomic_for_vram(struct xe_vm *vm)
+static bool xe_atomic_for_vram(struct xe_vm *vm, struct xe_vma *vma)
 {
+	if (vma->attr.atomic_access == DRM_XE_ATOMIC_CPU)
+		return false;
+
 	return true;
 }
 
-static bool xe_atomic_for_system(struct xe_vm *vm, struct xe_bo *bo)
+static bool xe_atomic_for_system(struct xe_vm *vm, struct xe_vma *vma)
 {
 	struct xe_device *xe = vm->xe;
+	struct xe_bo *bo = xe_vma_bo(vma);
 
-	if (!xe->info.has_device_atomics_on_smem)
+	if (!xe->info.has_device_atomics_on_smem ||
+	    vma->attr.atomic_access == DRM_XE_ATOMIC_CPU)
 		return false;
 
+	if (vma->attr.atomic_access == DRM_XE_ATOMIC_DEVICE)
+		return true;
+
 	/*
 	 * If a SMEM+LMEM allocation is backed by SMEM, a device
 	 * atomics will cause a gpu page fault and which then
 	 * gets migrated to LMEM, bind such allocations with
 	 * device atomics enabled.
-	 *
-	 * TODO: Revisit this. Perhaps add something like a
-	 * fault_on_atomics_in_system UAPI flag.
-	 * Note that this also prohibits GPU atomics in LR mode for
-	 * userptr and system memory on DGFX.
 	 */
 	return (!IS_DGFX(xe) || (!xe_vm_in_lr_mode(vm) ||
 				 (bo && xe_bo_has_single_placement(bo))));
@@ -744,8 +747,8 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 		goto walk_pt;
 
 	if (vma->gpuva.flags & XE_VMA_ATOMIC_PTE_BIT) {
-		xe_walk.default_vram_pte = xe_atomic_for_vram(vm) ? XE_USM_PPGTT_PTE_AE : 0;
-		xe_walk.default_system_pte = xe_atomic_for_system(vm, bo) ?
+		xe_walk.default_vram_pte = xe_atomic_for_vram(vm, vma) ? XE_USM_PPGTT_PTE_AE : 0;
+		xe_walk.default_system_pte = xe_atomic_for_system(vm, vma) ?
 			XE_USM_PPGTT_PTE_AE : 0;
 	}
 
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index c2306000f15e..c660ccb21945 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -789,22 +789,9 @@ bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range, struct xe_vm
 	return true;
 }
 
-/**
- * xe_svm_handle_pagefault() - SVM handle page fault
- * @vm: The VM.
- * @vma: The CPU address mirror VMA.
- * @gt: The gt upon the fault occurred.
- * @fault_addr: The GPU fault address.
- * @atomic: The fault atomic access bit.
- *
- * Create GPU bindings for a SVM page fault. Optionally migrate to device
- * memory.
- *
- * Return: 0 on success, negative error code on error.
- */
-int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
-			    struct xe_gt *gt, u64 fault_addr,
-			    bool atomic)
+static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
+				     struct xe_gt *gt, u64 fault_addr,
+				     bool need_vram)
 {
 	struct drm_gpusvm_ctx ctx = {
 		.read_only = xe_vma_read_only(vma),
@@ -812,9 +799,8 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP),
 		.check_pages_threshold = IS_DGFX(vm->xe) &&
 			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ? SZ_64K : 0,
-		.devmem_only = atomic && IS_DGFX(vm->xe) &&
-			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP),
-		.timeslice_ms = atomic && IS_DGFX(vm->xe) &&
+		.devmem_only = need_vram && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP),
+		.timeslice_ms = need_vram && IS_DGFX(vm->xe) &&
 			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ?
 			vm->xe->atomic_svm_timeslice_ms : 0,
 	};
@@ -917,6 +903,32 @@ err_out:
 	return err;
 }
 
+/**
+ * xe_svm_handle_pagefault() - SVM handle page fault
+ * @vm: The VM.
+ * @vma: The CPU address mirror VMA.
+ * @gt: The gt upon the fault occurred.
+ * @fault_addr: The GPU fault address.
+ * @atomic: The fault atomic access bit.
+ *
+ * Create GPU bindings for a SVM page fault. Optionally migrate to device
+ * memory.
+ *
+ * Return: 0 on success, negative error code on error.
+ */
+int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
+			    struct xe_gt *gt, u64 fault_addr,
+			    bool atomic)
+{
+	int need_vram;
+
+	need_vram = xe_vma_need_vram_for_atomic(vm->xe, vma, atomic);
+	if (need_vram < 0)
+		return need_vram;
+
+	return __xe_svm_handle_pagefault(vm, vma, gt, fault_addr, need_vram ? true : false);
+}
+
 /**
  * xe_svm_has_mapping() - SVM has mappings
  * @vm: The VM.
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 9fef01400c74..7bbf12742606 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -4212,6 +4212,45 @@ void xe_vm_snapshot_free(struct xe_vm_snapshot *snap)
 	kvfree(snap);
 }
 
+/**
+ * xe_vma_need_vram_for_atomic - Check if VMA needs VRAM migration for atomic operations
+ * @xe: Pointer to the XE device structure
+ * @vma: Pointer to the virtual memory area (VMA) structure
+ * @is_atomic: In pagefault path and atomic operation
+ *
+ * This function determines whether the given VMA needs to be migrated to
+ * VRAM in order to do atomic GPU operation.
+ *
+ * Return:
+ *   1        - Migration to VRAM is required
+ *   0        - Migration is not required
+ *   -EACCES  - Invalid access for atomic memory attr
+ *
+ */
+int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool is_atomic)
+{
+	if (!IS_DGFX(xe) || !is_atomic)
+		return 0;
+
+	/*
+	 * NOTE: The checks implemented here are platform-specific. For
+	 * instance, on a device supporting CXL atomics, these would ideally
+	 * work universally without additional handling.
+	 */
+	switch (vma->attr.atomic_access) {
+	case DRM_XE_ATOMIC_DEVICE:
+		return !xe->info.has_device_atomics_on_smem;
+
+	case DRM_XE_ATOMIC_CPU:
+		return -EACCES;
+
+	case DRM_XE_ATOMIC_UNDEFINED:
+	case DRM_XE_ATOMIC_GLOBAL:
+	default:
+		return 1;
+	}
+}
+
 /**
  * xe_vm_alloc_madvise_vma - Allocate VMA's with madvise ops
  * @vm: Pointer to the xe_vm structure
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index ccc549e39cfc..dbb1bf3caf2d 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -171,6 +171,8 @@ static inline bool xe_vma_is_userptr(struct xe_vma *vma)
 
 struct xe_vma *xe_vm_find_vma_by_addr(struct xe_vm *vm, u64 page_addr);
 
+int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool is_atomic);
+
 int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t addr, uint64_t size);
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index b861c3349b0a..95258bb6a8ee 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -85,7 +85,20 @@ static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
 			   struct xe_vma **vmas, int num_vmas,
 			   struct drm_xe_madvise *op)
 {
-	/* Implementation pending */
+	int i;
+
+	xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_ATOMIC);
+	xe_assert(vm->xe, op->atomic.val <= DRM_XE_ATOMIC_CPU);
+
+	for (i = 0; i < num_vmas; i++) {
+		if ((xe_vma_is_userptr(vmas[i]) &&
+		     !(op->atomic.val == DRM_XE_ATOMIC_DEVICE &&
+		       xe->info.has_device_atomics_on_smem)))
+			continue;
+
+		vmas[i]->attr.atomic_access = op->atomic.val;
+	/*TODO: handle bo backed vmas */
+	}
 }
 
 static void madvise_pat_index(struct xe_device *xe, struct xe_vm *vm,

From a894c27407402b6f980876603e1fa66470555ff6 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:54 +0530
Subject: [PATCH 0615/1455] drm/xe/madvise: Update migration policy based on
 preferred location

When the user sets the valid devmem_fd as a preferred location, GPU fault
will trigger migration to tile of device associated with devmem_fd.

If the user sets an invalid devmem_fd the preferred location is current
placement(smem) only.

v2(Matthew Brost)
- Default should be faulting tile
- remove devmem_fd used as region

v3 (Matthew Brost)
- Add migration_policy
- Fix return condition
- fix migrate condition

v4
-Rebase

v5
- Add check for userptr and bo based vmas

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-11-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_svm.c        | 45 +++++++++++++++++++++++++++++-
 drivers/gpu/drm/xe/xe_svm.h        |  8 ++++++
 drivers/gpu/drm/xe/xe_vm_madvise.c | 25 ++++++++++++++++-
 3 files changed, 76 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index c660ccb21945..19585a3d9f69 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -806,6 +806,7 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 	};
 	struct xe_svm_range *range;
 	struct dma_fence *fence;
+	struct drm_pagemap *dpagemap;
 	struct xe_tile *tile = gt_to_tile(gt);
 	int migrate_try_count = ctx.devmem_only ? 3 : 1;
 	ktime_t end = 0;
@@ -835,8 +836,14 @@ retry:
 
 	range_debug(range, "PAGE FAULT");
 
+	dpagemap = xe_vma_resolve_pagemap(vma, tile);
 	if (--migrate_try_count >= 0 &&
-	    xe_svm_range_needs_migrate_to_vram(range, vma, IS_DGFX(vm->xe))) {
+	    xe_svm_range_needs_migrate_to_vram(range, vma, !!dpagemap || ctx.devmem_only)) {
+		/* TODO : For multi-device dpagemap will be used to find the
+		 * remote tile and remote device. Will need to modify
+		 * xe_svm_alloc_vram to use dpagemap for future multi-device
+		 * support.
+		 */
 		err = xe_svm_alloc_vram(tile, range, &ctx);
 		ctx.timeslice_ms <<= 1;	/* Double timeslice if we have to retry */
 		if (err) {
@@ -1100,6 +1107,37 @@ static struct drm_pagemap *tile_local_pagemap(struct xe_tile *tile)
 	return &tile->mem.vram->dpagemap;
 }
 
+/**
+ * xe_vma_resolve_pagemap - Resolve the appropriate DRM pagemap for a VMA
+ * @vma: Pointer to the xe_vma structure containing memory attributes
+ * @tile: Pointer to the xe_tile structure used as fallback for VRAM mapping
+ *
+ * This function determines the correct DRM pagemap to use for a given VMA.
+ * It first checks if a valid devmem_fd is provided in the VMA's preferred
+ * location. If the devmem_fd is negative, it returns NULL, indicating no
+ * pagemap is available and smem to be used as preferred location.
+ * If the devmem_fd is equal to the default faulting
+ * GT identifier, it returns the VRAM pagemap associated with the tile.
+ *
+ * Future support for multi-device configurations may use drm_pagemap_from_fd()
+ * to resolve pagemaps from arbitrary file descriptors.
+ *
+ * Return: A pointer to the resolved drm_pagemap, or NULL if none is applicable.
+ */
+struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile)
+{
+	s32 fd = (s32)vma->attr.preferred_loc.devmem_fd;
+
+	if (fd == DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM)
+		return NULL;
+
+	if (fd == DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE)
+		return IS_DGFX(tile_to_xe(tile)) ? tile_local_pagemap(tile) : NULL;
+
+	/* TODO: Support multi-device with drm_pagemap_from_fd(fd) */
+	return NULL;
+}
+
 /**
  * xe_svm_alloc_vram()- Allocate device memory pages for range,
  * migrating existing data.
@@ -1212,6 +1250,11 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr)
 {
 	return 0;
 }
+
+struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile)
+{
+	return NULL;
+}
 #endif
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index 046a9c4e95c2..9d6a8840a8b7 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -94,6 +94,8 @@ void xe_svm_unmap_address_range(struct xe_vm *vm, u64 start, u64 end);
 
 u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end);
 
+struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile);
+
 /**
  * xe_svm_range_has_dma_mapping() - SVM range has DMA mapping
  * @range: SVM range
@@ -318,6 +320,12 @@ u8 xe_svm_ranges_zap_ptes_in_range(struct xe_vm *vm, u64 start, u64 end)
 	return 0;
 }
 
+static inline
+struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *tile)
+{
+	return NULL;
+}
+
 #define xe_svm_assert_in_notifier(...) do {} while (0)
 #define xe_svm_range_has_dma_mapping(...) false
 
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 95258bb6a8ee..b5fc1eedf095 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -78,7 +78,23 @@ static void madvise_preferred_mem_loc(struct xe_device *xe, struct xe_vm *vm,
 				      struct xe_vma **vmas, int num_vmas,
 				      struct drm_xe_madvise *op)
 {
-	/* Implementation pending */
+	int i;
+
+	xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC);
+
+	for (i = 0; i < num_vmas; i++) {
+		/*TODO: Extend attributes to bo based vmas */
+		if (!xe_vma_is_cpu_addr_mirror(vmas[i]))
+			continue;
+
+		vmas[i]->attr.preferred_loc.devmem_fd = op->preferred_mem_loc.devmem_fd;
+
+		/* Till multi-device support is not added migration_policy
+		 * is of no use and can be ignored.
+		 */
+		vmas[i]->attr.preferred_loc.migration_policy =
+						op->preferred_mem_loc.migration_policy;
+	}
 }
 
 static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
@@ -184,6 +200,12 @@ static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madv
 
 	switch (args->type) {
 	case DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC:
+	{
+		s32 fd = (s32)args->preferred_mem_loc.devmem_fd;
+
+		if (XE_IOCTL_DBG(xe, fd < DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM))
+			return false;
+
 		if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.migration_policy >
 				     DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES))
 			return false;
@@ -194,6 +216,7 @@ static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madv
 		if (XE_IOCTL_DBG(xe, args->atomic.reserved))
 			return false;
 		break;
+	}
 	case DRM_XE_MEM_RANGE_ATTR_ATOMIC:
 		if (XE_IOCTL_DBG(xe, args->atomic.val > DRM_XE_ATOMIC_CPU))
 			return false;

From 18d36fd6d13b295ebf111de2651bb398fcd1a040 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:55 +0530
Subject: [PATCH 0616/1455] drm/xe/svm: Support DRM_XE_SVM_MEM_RANGE_ATTR_PAT
 memory attribute

This attributes sets the pat_index for the svm used vma range, which is
utilized to ascertain the coherence.

v2 (Matthew Brost)
- Pat index sanity check

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-12-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm_madvise.c | 24 ++++++++++++++++++++++--
 1 file changed, 22 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index b5fc1eedf095..3bd52063f9c2 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -9,6 +9,7 @@
 #include <drm/xe_drm.h>
 
 #include "xe_bo.h"
+#include "xe_pat.h"
 #include "xe_pt.h"
 #include "xe_svm.h"
 
@@ -121,7 +122,12 @@ static void madvise_pat_index(struct xe_device *xe, struct xe_vm *vm,
 			      struct xe_vma **vmas, int num_vmas,
 			      struct drm_xe_madvise *op)
 {
-	/* Implementation pending */
+	int i;
+
+	xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_PAT);
+
+	for (i = 0; i < num_vmas; i++)
+		vmas[i]->attr.pat_index = op->pat_index.val;
 }
 
 typedef void (*madvise_func)(struct xe_device *xe, struct xe_vm *vm,
@@ -229,8 +235,22 @@ static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madv
 
 		break;
 	case DRM_XE_MEM_RANGE_ATTR_PAT:
-		/*TODO: Add valid pat check */
+	{
+		u16 coh_mode = xe_pat_index_get_coh_mode(xe, args->pat_index.val);
+
+		if (XE_IOCTL_DBG(xe, !coh_mode))
+			return false;
+
+		if (XE_WARN_ON(coh_mode > XE_COH_AT_LEAST_1WAY))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, args->pat_index.pad))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, args->pat_index.reserved))
+			return false;
 		break;
+	}
 	default:
 		if (XE_IOCTL_DBG(xe, 1))
 			return false;

From fa1a82c985dba642de66f0a1918fc531007bf90f Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:56 +0530
Subject: [PATCH 0617/1455] drm/xe/uapi: Add flag for consulting madvise hints
 on svm prefetch
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce flag DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC to ensure prefetching
in madvise-advised memory regions

v2 (Matthew Brost)
- Add kernel-doc

v3 (Matthew Brost)
- Fix kernel-doc

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-13-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 include/uapi/drm/xe_drm.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index 4e6e9a9164ee..115b9bca2a25 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -1010,6 +1010,10 @@ struct drm_xe_vm_destroy {
  *    valid on VMs with DRM_XE_VM_CREATE_FLAG_FAULT_MODE set. The CPU address
  *    mirror flag are only valid for DRM_XE_VM_BIND_OP_MAP operations, the BO
  *    handle MBZ, and the BO offset MBZ.
+ *
+ * The @prefetch_mem_region_instance for %DRM_XE_VM_BIND_OP_PREFETCH can also be:
+ *  - %DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC, which ensures prefetching occurs in
+ *    the memory region advised by madvise.
  */
 struct drm_xe_vm_bind_op {
 	/** @extensions: Pointer to the first extension struct, if any */
@@ -1115,6 +1119,7 @@ struct drm_xe_vm_bind_op {
 	/** @flags: Bind flags */
 	__u32 flags;
 
+#define DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC	-1
 	/**
 	 * @prefetch_mem_region_instance: Memory region to prefetch VMA to.
 	 * It is a region instance, not a mask.

From c1bb69a2e8e2d55c3725e304d8b8fd189ee43fbe Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:57 +0530
Subject: [PATCH 0618/1455] drm/xe/svm: Consult madvise preferred location in
 prefetch

When prefetch region is DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC, prefetch svm
ranges to preferred location provided by madvise.

v2 (Matthew Brost)
- Fix region, devmem_fd usages
- consult madvise is applicable for other vma's too.

v3
- Fix atomic handling

v4
- Fix xe_svm_range_validate to check for
  DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC too.

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-14-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c       | 47 ++++++++++++++++++++------------
 drivers/gpu/drm/xe/xe_vm_types.h |  7 +++--
 2 files changed, 35 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 7bbf12742606..c765c96b9bde 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -38,6 +38,7 @@
 #include "xe_res_cursor.h"
 #include "xe_svm.h"
 #include "xe_sync.h"
+#include "xe_tile.h"
 #include "xe_trace_bo.h"
 #include "xe_wa.h"
 #include "xe_hmm.h"
@@ -2400,9 +2401,10 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops,
 				__xe_vm_needs_clear_scratch_pages(vm, flags);
 		} else if (__op->op == DRM_GPUVA_OP_PREFETCH) {
 			struct xe_vma *vma = gpuva_to_vma(op->base.prefetch.va);
+			struct xe_tile *tile;
 			struct xe_svm_range *svm_range;
 			struct drm_gpusvm_ctx ctx = {};
-			struct xe_tile *tile;
+			struct drm_pagemap *dpagemap;
 			u8 id, tile_mask = 0;
 			u32 i;
 
@@ -2419,8 +2421,24 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops,
 				tile_mask |= 0x1 << id;
 
 			xa_init_flags(&op->prefetch_range.range, XA_FLAGS_ALLOC);
-			op->prefetch_range.region = prefetch_region;
 			op->prefetch_range.ranges_count = 0;
+			tile = NULL;
+
+			if (prefetch_region == DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC) {
+				dpagemap = xe_vma_resolve_pagemap(vma,
+								  xe_device_get_root_tile(vm->xe));
+				/*
+				 * TODO: Once multigpu support is enabled will need
+				 * something to dereference tile from dpagemap.
+				 */
+				if (dpagemap)
+					tile = xe_device_get_root_tile(vm->xe);
+			} else if (prefetch_region) {
+				tile = &vm->xe->tiles[region_to_mem_type[prefetch_region] -
+						      XE_PL_VRAM0];
+			}
+
+			op->prefetch_range.tile = tile;
 alloc_next_range:
 			svm_range = xe_svm_range_find_or_insert(vm, addr, vma, &ctx);
 
@@ -2439,7 +2457,7 @@ alloc_next_range:
 				goto unwind_prefetch_ops;
 			}
 
-			if (xe_svm_range_validate(vm, svm_range, tile_mask, !!prefetch_region)) {
+			if (xe_svm_range_validate(vm, svm_range, tile_mask, !!tile)) {
 				xe_svm_range_debug(svm_range, "PREFETCH - RANGE IS VALID");
 				goto check_next_range;
 			}
@@ -2935,30 +2953,26 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op)
 {
 	bool devmem_possible = IS_DGFX(vm->xe) && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP);
 	struct xe_vma *vma = gpuva_to_vma(op->base.prefetch.va);
+	struct xe_tile *tile = op->prefetch_range.tile;
 	int err = 0;
 
 	struct xe_svm_range *svm_range;
 	struct drm_gpusvm_ctx ctx = {};
-	struct xe_tile *tile;
 	unsigned long i;
-	u32 region;
 
 	if (!xe_vma_is_cpu_addr_mirror(vma))
 		return 0;
 
-	region = op->prefetch_range.region;
-
 	ctx.read_only = xe_vma_read_only(vma);
 	ctx.devmem_possible = devmem_possible;
 	ctx.check_pages_threshold = devmem_possible ? SZ_64K : 0;
 
 	/* TODO: Threading the migration */
 	xa_for_each(&op->prefetch_range.range, i, svm_range) {
-		if (!region)
+		if (!tile)
 			xe_svm_range_migrate_to_smem(vm, svm_range);
 
-		if (xe_svm_range_needs_migrate_to_vram(svm_range, vma, region)) {
-			tile = &vm->xe->tiles[region_to_mem_type[region] - XE_PL_VRAM0];
+		if (xe_svm_range_needs_migrate_to_vram(svm_range, vma, !!tile)) {
 			err = xe_svm_alloc_vram(tile, svm_range, &ctx);
 			if (err) {
 				drm_dbg(&vm->xe->drm, "VRAM allocation failed, retry from userspace, asid=%u, gpusvm=%p, errno=%pe\n",
@@ -3021,12 +3035,11 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm,
 		struct xe_vma *vma = gpuva_to_vma(op->base.prefetch.va);
 		u32 region;
 
-		if (xe_vma_is_cpu_addr_mirror(vma))
-			region = op->prefetch_range.region;
-		else
+		if (!xe_vma_is_cpu_addr_mirror(vma)) {
 			region = op->prefetch.region;
-
-		xe_assert(vm->xe, region <= ARRAY_SIZE(region_to_mem_type));
+			xe_assert(vm->xe, region == DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC ||
+				  region <= ARRAY_SIZE(region_to_mem_type));
+		}
 
 		err = vma_lock_and_validate(exec,
 					    gpuva_to_vma(op->base.prefetch.va),
@@ -3444,8 +3457,8 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
 				 op == DRM_XE_VM_BIND_OP_PREFETCH) ||
 		    XE_IOCTL_DBG(xe, prefetch_region &&
 				 op != DRM_XE_VM_BIND_OP_PREFETCH) ||
-		    XE_IOCTL_DBG(xe, !(BIT(prefetch_region) &
-				       xe->info.mem_region_mask)) ||
+		    XE_IOCTL_DBG(xe,  (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
+				       !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
 		    XE_IOCTL_DBG(xe, obj &&
 				 op == DRM_XE_VM_BIND_OP_UNMAP)) {
 			err = -EINVAL;
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index dde7218ceba6..5704f379c225 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -428,8 +428,11 @@ struct xe_vma_op_prefetch_range {
 	struct xarray range;
 	/** @ranges_count: number of svm ranges to map */
 	u32 ranges_count;
-	/** @region: memory region to prefetch to */
-	u32 region;
+	/**
+	 * @tile: Pointer to the tile structure containing memory to prefetch.
+	 *        NULL if prefetch requested region is smem
+	 */
+	struct xe_tile *tile;
 };
 
 /** enum xe_vma_op_flags - flags for VMA operation */

From 072e299982635aa8d96683071930e3bf0a1c1a2f Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:58 +0530
Subject: [PATCH 0619/1455] drm/xe/bo: Add attributes field to xe_bo
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A single BO can be linked to multiple VMAs, making VMA attributes
insufficient for determining the placement and PTE update attributes
of the BO. To address this, an attributes field has been added to the
BO.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-15-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_bo_types.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index cf604adc13a3..314652afdca7 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -61,6 +61,14 @@ struct xe_bo {
 	 */
 	struct list_head client_link;
 #endif
+	/** @attr: User controlled attributes for bo */
+	struct {
+		/**
+		 * @atomic_access: type of atomic access bo needs
+		 * protected by bo dma-resv lock
+		 */
+		u32 atomic_access;
+	} attr;
 	/**
 	 * @pxp_key_instance: PXP key instance this BO was created against. A
 	 * 0 in this variable indicates that the BO does not use PXP encryption.

From 293032eec4baa04374d62dd44de61e355296ad32 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:00:59 +0530
Subject: [PATCH 0620/1455] drm/xe/bo: Update atomic_access attribute on
 madvise

Update the bo_atomic_access based on user-provided input and determine
the migration to smem during a CPU fault

v2 (Matthew Brost)
- Avoid cpu unmapping if bo is already in smem
- check atomics on smem too for ioctl
- Add comments

v3
- Avoid migration in prefetch

v4 (Matthew Brost)
- make sanity check function bool
- add assert for smem placement
- fix doc

v5 (Matthew Brost)
- NACK atomic fault with  DRM_XE_ATOMIC_CPU

Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-16-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_bo.c           | 29 ++++++++++++--
 drivers/gpu/drm/xe/xe_gt_pagefault.c | 35 ++++++-----------
 drivers/gpu/drm/xe/xe_vm.c           |  7 +++-
 drivers/gpu/drm/xe/xe_vm_madvise.c   | 59 +++++++++++++++++++++++++++-
 4 files changed, 102 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 0a6a32cfb771..7d1ff642b02a 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1712,6 +1712,18 @@ static void xe_gem_object_close(struct drm_gem_object *obj,
 	}
 }
 
+static bool should_migrate_to_smem(struct xe_bo *bo)
+{
+	/*
+	 * NOTE: The following atomic checks are platform-specific. For example,
+	 * if a device supports CXL atomics, these may not be necessary or
+	 * may behave differently.
+	 */
+
+	return bo->attr.atomic_access == DRM_XE_ATOMIC_GLOBAL ||
+	       bo->attr.atomic_access == DRM_XE_ATOMIC_CPU;
+}
+
 static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *tbo = vmf->vma->vm_private_data;
@@ -1720,7 +1732,7 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
 	struct xe_bo *bo = ttm_to_xe_bo(tbo);
 	bool needs_rpm = bo->flags & XE_BO_FLAG_VRAM_MASK;
 	vm_fault_t ret;
-	int idx;
+	int idx, r = 0;
 
 	if (needs_rpm)
 		xe_pm_runtime_get(xe);
@@ -1732,8 +1744,19 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
 	if (drm_dev_enter(ddev, &idx)) {
 		trace_xe_bo_cpu_fault(bo);
 
-		ret = ttm_bo_vm_fault_reserved(vmf, vmf->vma->vm_page_prot,
-					       TTM_BO_VM_NUM_PREFAULT);
+		if (should_migrate_to_smem(bo)) {
+			xe_assert(xe, bo->flags & XE_BO_FLAG_SYSTEM);
+
+			r = xe_bo_migrate(bo, XE_PL_TT);
+			if (r == -EBUSY || r == -ERESTARTSYS || r == -EINTR)
+				ret = VM_FAULT_NOPAGE;
+			else if (r)
+				ret = VM_FAULT_SIGBUS;
+		}
+		if (!ret)
+			ret = ttm_bo_vm_fault_reserved(vmf,
+						       vmf->vma->vm_page_prot,
+						       TTM_BO_VM_NUM_PREFAULT);
 		drm_dev_exit(idx);
 
 		if (ret == VM_FAULT_RETRY &&
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index ab43dec52776..4ea30fbce9bd 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -75,7 +75,7 @@ static bool vma_is_valid(struct xe_tile *tile, struct xe_vma *vma)
 }
 
 static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma,
-		       bool atomic, struct xe_vram_region *vram)
+		       bool need_vram_move, struct xe_vram_region *vram)
 {
 	struct xe_bo *bo = xe_vma_bo(vma);
 	struct xe_vm *vm = xe_vma_vm(vma);
@@ -85,26 +85,13 @@ static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma,
 	if (err)
 		return err;
 
-	if (atomic && vram) {
-		xe_assert(vm->xe, IS_DGFX(vm->xe));
+	if (!bo)
+		return 0;
 
-		if (xe_vma_is_userptr(vma)) {
-			err = -EACCES;
-			return err;
-		}
+	err = need_vram_move ? xe_bo_migrate(bo, vram->placement) :
+			       xe_bo_validate(bo, vm, true);
 
-		/* Migrate to VRAM, move should invalidate the VMA first */
-		err = xe_bo_migrate(bo, vram->placement);
-		if (err)
-			return err;
-	} else if (bo) {
-		/* Create backing store if needed */
-		err = xe_bo_validate(bo, vm, true);
-		if (err)
-			return err;
-	}
-
-	return 0;
+	return err;
 }
 
 static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
@@ -115,10 +102,14 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
 	struct drm_exec exec;
 	struct dma_fence *fence;
 	ktime_t end = 0;
-	int err;
+	int err, needs_vram;
 
 	lockdep_assert_held_write(&vm->lock);
 
+	needs_vram = xe_vma_need_vram_for_atomic(vm->xe, vma, atomic);
+	if (needs_vram < 0 || (needs_vram && xe_vma_is_userptr(vma)))
+		return needs_vram < 0 ? needs_vram : -EACCES;
+
 	xe_gt_stats_incr(gt, XE_GT_STATS_ID_VMA_PAGEFAULT_COUNT, 1);
 	xe_gt_stats_incr(gt, XE_GT_STATS_ID_VMA_PAGEFAULT_KB, xe_vma_size(vma) / 1024);
 
@@ -141,7 +132,7 @@ retry_userptr:
 	/* Lock VM and BOs dma-resv */
 	drm_exec_init(&exec, 0, 0);
 	drm_exec_until_all_locked(&exec) {
-		err = xe_pf_begin(&exec, vma, atomic, tile->mem.vram);
+		err = xe_pf_begin(&exec, vma, needs_vram == 1, tile->mem.vram);
 		drm_exec_retry_on_contention(&exec);
 		if (xe_vm_validate_should_retry(&exec, err, &end))
 			err = -EAGAIN;
@@ -576,7 +567,7 @@ static int handle_acc(struct xe_gt *gt, struct acc *acc)
 	/* Lock VM and BOs dma-resv */
 	drm_exec_init(&exec, 0, 0);
 	drm_exec_until_all_locked(&exec) {
-		ret = xe_pf_begin(&exec, vma, true, tile->mem.vram);
+		ret = xe_pf_begin(&exec, vma, IS_DGFX(vm->xe), tile->mem.vram);
 		drm_exec_retry_on_contention(&exec);
 		if (ret)
 			break;
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index c765c96b9bde..6574ae56a730 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -4242,15 +4242,18 @@ void xe_vm_snapshot_free(struct xe_vm_snapshot *snap)
  */
 int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool is_atomic)
 {
+	u32 atomic_access = xe_vma_bo(vma) ? xe_vma_bo(vma)->attr.atomic_access :
+					     vma->attr.atomic_access;
+
 	if (!IS_DGFX(xe) || !is_atomic)
-		return 0;
+		return false;
 
 	/*
 	 * NOTE: The checks implemented here are platform-specific. For
 	 * instance, on a device supporting CXL atomics, these would ideally
 	 * work universally without additional handling.
 	 */
-	switch (vma->attr.atomic_access) {
+	switch (atomic_access) {
 	case DRM_XE_ATOMIC_DEVICE:
 		return !xe->info.has_device_atomics_on_smem;
 
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 3bd52063f9c2..212a03178d65 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -102,6 +102,7 @@ static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
 			   struct xe_vma **vmas, int num_vmas,
 			   struct drm_xe_madvise *op)
 {
+	struct xe_bo *bo;
 	int i;
 
 	xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_ATOMIC);
@@ -114,7 +115,19 @@ static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
 			continue;
 
 		vmas[i]->attr.atomic_access = op->atomic.val;
-	/*TODO: handle bo backed vmas */
+
+		bo = xe_vma_bo(vmas[i]);
+		if (!bo)
+			continue;
+
+		xe_bo_assert_held(bo);
+		bo->attr.atomic_access = op->atomic.val;
+
+		/* Invalidate cpu page table, so bo can migrate to smem in next access */
+		if (xe_bo_is_vram(bo) &&
+		    (bo->attr.atomic_access == DRM_XE_ATOMIC_CPU ||
+		     bo->attr.atomic_access == DRM_XE_ATOMIC_GLOBAL))
+			ttm_bo_unmap_virtual(&bo->ttm);
 	}
 }
 
@@ -262,6 +275,41 @@ static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madv
 	return true;
 }
 
+static bool check_bo_args_are_sane(struct xe_vm *vm, struct xe_vma **vmas,
+				   int num_vmas, u32 atomic_val)
+{
+	struct xe_device *xe = vm->xe;
+	struct xe_bo *bo;
+	int i;
+
+	for (i = 0; i < num_vmas; i++) {
+		bo = xe_vma_bo(vmas[i]);
+		if (!bo)
+			continue;
+		/*
+		 * NOTE: The following atomic checks are platform-specific. For example,
+		 * if a device supports CXL atomics, these may not be necessary or
+		 * may behave differently.
+		 */
+		if (XE_IOCTL_DBG(xe, atomic_val == DRM_XE_ATOMIC_CPU &&
+				 !(bo->flags & XE_BO_FLAG_SYSTEM)))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, atomic_val == DRM_XE_ATOMIC_DEVICE &&
+				 !(bo->flags & XE_BO_FLAG_VRAM0) &&
+				 !(bo->flags & XE_BO_FLAG_VRAM1) &&
+				 !(bo->flags & XE_BO_FLAG_SYSTEM &&
+				   xe->info.has_device_atomics_on_smem)))
+			return false;
+
+		if (XE_IOCTL_DBG(xe, atomic_val == DRM_XE_ATOMIC_GLOBAL &&
+				 (!(bo->flags & XE_BO_FLAG_SYSTEM) ||
+				  (!(bo->flags & XE_BO_FLAG_VRAM0) &&
+				   !(bo->flags & XE_BO_FLAG_VRAM1)))))
+			return false;
+	}
+	return true;
+}
 /**
  * xe_vm_madvise_ioctl - Handle MADVise ioctl for a VM
  * @dev: DRM device pointer
@@ -313,6 +361,15 @@ int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *fil
 		goto unlock_vm;
 
 	if (madvise_range.has_bo_vmas) {
+		if (args->type == DRM_XE_MEM_RANGE_ATTR_ATOMIC) {
+			if (!check_bo_args_are_sane(vm, madvise_range.vmas,
+						    madvise_range.num_vmas,
+						    args->atomic.val)) {
+				err = -EINVAL;
+				goto unlock_vm;
+			}
+		}
+
 		drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES | DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
 		drm_exec_until_all_locked(&exec) {
 			for (int i = 0; i < madvise_range.num_vmas; i++) {

From 002f817d61ce2cfb0be850c3a6036adfbb4fd5e9 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:01:00 +0530
Subject: [PATCH 0621/1455] drm/xe/madvise: Skip vma invalidation if mem attr
 are unchanged

If a VMA within the madvise input range already has the same memory
attribute as the one requested by the user, skip PTE zapping for that
VMA to avoid unnecessary invalidation.

v2 (Matthew Brost)
- fix skip_invalidation for new attributes
- s/u32/bool
- Remove unnecessary assignment  for kzalloc'ed

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-17-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm_madvise.c | 68 ++++++++++++++++++++----------
 drivers/gpu/drm/xe/xe_vm_types.h   |  6 +++
 2 files changed, 52 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 212a03178d65..7813bdedacaa 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -85,16 +85,20 @@ static void madvise_preferred_mem_loc(struct xe_device *xe, struct xe_vm *vm,
 
 	for (i = 0; i < num_vmas; i++) {
 		/*TODO: Extend attributes to bo based vmas */
-		if (!xe_vma_is_cpu_addr_mirror(vmas[i]))
-			continue;
-
-		vmas[i]->attr.preferred_loc.devmem_fd = op->preferred_mem_loc.devmem_fd;
-
-		/* Till multi-device support is not added migration_policy
-		 * is of no use and can be ignored.
-		 */
-		vmas[i]->attr.preferred_loc.migration_policy =
+		if ((vmas[i]->attr.preferred_loc.devmem_fd == op->preferred_mem_loc.devmem_fd &&
+		     vmas[i]->attr.preferred_loc.migration_policy ==
+		     op->preferred_mem_loc.migration_policy) ||
+		    !xe_vma_is_cpu_addr_mirror(vmas[i])) {
+			vmas[i]->skip_invalidation = true;
+		} else {
+			vmas[i]->skip_invalidation = false;
+			vmas[i]->attr.preferred_loc.devmem_fd = op->preferred_mem_loc.devmem_fd;
+			/* Till multi-device support is not added migration_policy
+			 * is of no use and can be ignored.
+			 */
+			vmas[i]->attr.preferred_loc.migration_policy =
 						op->preferred_mem_loc.migration_policy;
+		}
 	}
 }
 
@@ -109,17 +113,27 @@ static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
 	xe_assert(vm->xe, op->atomic.val <= DRM_XE_ATOMIC_CPU);
 
 	for (i = 0; i < num_vmas; i++) {
-		if ((xe_vma_is_userptr(vmas[i]) &&
-		     !(op->atomic.val == DRM_XE_ATOMIC_DEVICE &&
-		       xe->info.has_device_atomics_on_smem)))
+		if (xe_vma_is_userptr(vmas[i]) &&
+		    !(op->atomic.val == DRM_XE_ATOMIC_DEVICE &&
+		      xe->info.has_device_atomics_on_smem)) {
+			vmas[i]->skip_invalidation = true;
 			continue;
+		}
+
+		if (vmas[i]->attr.atomic_access == op->atomic.val) {
+			vmas[i]->skip_invalidation = true;
+		} else {
+			vmas[i]->skip_invalidation = false;
+			vmas[i]->attr.atomic_access = op->atomic.val;
+		}
 
 		vmas[i]->attr.atomic_access = op->atomic.val;
 
 		bo = xe_vma_bo(vmas[i]);
-		if (!bo)
+		if (!bo || bo->attr.atomic_access == op->atomic.val)
 			continue;
 
+		vmas[i]->skip_invalidation = false;
 		xe_bo_assert_held(bo);
 		bo->attr.atomic_access = op->atomic.val;
 
@@ -139,8 +153,14 @@ static void madvise_pat_index(struct xe_device *xe, struct xe_vm *vm,
 
 	xe_assert(vm->xe, op->type == DRM_XE_MEM_RANGE_ATTR_PAT);
 
-	for (i = 0; i < num_vmas; i++)
-		vmas[i]->attr.pat_index = op->pat_index.val;
+	for (i = 0; i < num_vmas; i++) {
+		if (vmas[i]->attr.pat_index == op->pat_index.val) {
+			vmas[i]->skip_invalidation = true;
+		} else {
+			vmas[i]->skip_invalidation = false;
+			vmas[i]->attr.pat_index = op->pat_index.val;
+		}
+	}
 }
 
 typedef void (*madvise_func)(struct xe_device *xe, struct xe_vm *vm,
@@ -157,7 +177,7 @@ static u8 xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end)
 {
 	struct drm_gpuva *gpuva;
 	struct xe_tile *tile;
-	u8 id, tile_mask;
+	u8 id, tile_mask = 0;
 
 	lockdep_assert_held_write(&vm->lock);
 
@@ -166,17 +186,20 @@ static u8 xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end)
 				  false, MAX_SCHEDULE_TIMEOUT) <= 0)
 		XE_WARN_ON(1);
 
-	tile_mask = xe_svm_ranges_zap_ptes_in_range(vm, start, end);
-
 	drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) {
 		struct xe_vma *vma = gpuva_to_vma(gpuva);
 
-		if (xe_vma_is_cpu_addr_mirror(vma) || xe_vma_is_null(vma))
+		if (vma->skip_invalidation || xe_vma_is_null(vma))
 			continue;
 
-		for_each_tile(tile, vm->xe, id) {
-			if (xe_pt_zap_ptes(tile, vma)) {
-				tile_mask |= BIT(id);
+		if (xe_vma_is_cpu_addr_mirror(vma)) {
+			tile_mask |= xe_svm_ranges_zap_ptes_in_range(vm,
+								      xe_vma_start(vma),
+								      xe_vma_end(vma));
+		} else {
+			for_each_tile(tile, vm->xe, id) {
+				if (xe_pt_zap_ptes(tile, vma)) {
+					tile_mask |= BIT(id);
 
 				/*
 				 * WRITE_ONCE pairs with READ_ONCE
@@ -184,6 +207,7 @@ static u8 xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end)
 				 */
 				WRITE_ONCE(vma->tile_invalidated,
 					   vma->tile_invalidated | BIT(id));
+				}
 			}
 		}
 	}
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 5704f379c225..35b6344158f5 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -157,6 +157,12 @@ struct xe_vma {
 	/** @tile_staged: bind is staged for this VMA */
 	u8 tile_staged;
 
+	/**
+	 * @skip_invalidation: Used in madvise to avoid invalidation
+	 * if mem attributes doesn't change
+	 */
+	bool skip_invalidation;
+
 	/**
 	 * @ufence: The user fence that was provided with MAP.
 	 * Needs to be signalled before UNMAP can be processed.

From 58dc430d89ae2cf41a8919ac99c36c4a17a3fdc3 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:01:01 +0530
Subject: [PATCH 0622/1455] drm/xe/vm: Add helper to check for default VMA
 memory attributes

Introduce a new helper function `xe_vma_has_default_mem_attrs()` to
determine whether a VMA's memory attributes are set to their default
values. This includes checks for atomic access, PAT index, and preferred
location.

Also, add a new field `default_pat_index` to `struct xe_vma_mem_attr`
to track the initial PAT index set during the first bind. This helps
distinguish between default and user-modified pat index, such as those
changed via madvise.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-18-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c       | 24 ++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vm.h       |  2 ++
 drivers/gpu/drm/xe/xe_vm_types.h |  6 ++++++
 3 files changed, 32 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 6574ae56a730..e9615d5ef944 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2640,6 +2640,29 @@ static int xe_vma_op_commit(struct xe_vm *vm, struct xe_vma_op *op)
 	return err;
 }
 
+/**
+ * xe_vma_has_default_mem_attrs - Check if a VMA has default memory attributes
+ * @vma: Pointer to the xe_vma structure to check
+ *
+ * This function determines whether the given VMA (Virtual Memory Area)
+ * has its memory attributes set to their default values. Specifically,
+ * it checks the following conditions:
+ *
+ * - `atomic_access` is `DRM_XE_VMA_ATOMIC_UNDEFINED`
+ * - `pat_index` is equal to `default_pat_index`
+ * - `preferred_loc.devmem_fd` is `DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE`
+ * - `preferred_loc.migration_policy` is `DRM_XE_MIGRATE_ALL_PAGES`
+ *
+ * Return: true if all attributes are at their default values, false otherwise.
+ */
+bool xe_vma_has_default_mem_attrs(struct xe_vma *vma)
+{
+	return (vma->attr.atomic_access == DRM_XE_ATOMIC_UNDEFINED &&
+		vma->attr.pat_index ==  vma->attr.default_pat_index &&
+		vma->attr.preferred_loc.devmem_fd == DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE &&
+		vma->attr.preferred_loc.migration_policy == DRM_XE_MIGRATE_ALL_PAGES);
+}
+
 static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 				   struct xe_vma_ops *vops)
 {
@@ -2672,6 +2695,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct drm_gpuva_ops *ops,
 					.migration_policy = DRM_XE_MIGRATE_ALL_PAGES,
 				},
 				.atomic_access = DRM_XE_ATOMIC_UNDEFINED,
+				.default_pat_index = op->map.pat_index,
 				.pat_index = op->map.pat_index,
 			};
 
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index dbb1bf3caf2d..1a4d67f29c6b 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -66,6 +66,8 @@ static inline bool xe_vm_is_closed_or_banned(struct xe_vm *vm)
 struct xe_vma *
 xe_vm_find_overlapping_vma(struct xe_vm *vm, u64 start, u64 range);
 
+bool xe_vma_has_default_mem_attrs(struct xe_vma *vma);
+
 /**
  * xe_vm_has_scratch() - Whether the vm is configured for scratch PTEs
  * @vm: The vm
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 35b6344158f5..b5108d010786 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -103,8 +103,14 @@ struct xe_vma_mem_attr {
 	 */
 	u32 atomic_access;
 
+	/**
+	 * @default_pat_index: The pat index for VMA set during first bind by user.
+	 */
+	u16 default_pat_index;
+
 	/**
 	 * @pat_index: The pat index to use when encoding the PTEs for this vma.
+	 * same as default_pat_index unless overwritten by madvise.
 	 */
 	u16 pat_index;
 };

From a2eb8aec3ebe474ce0fe0b6ebb18aade8c1a5c00 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:01:02 +0530
Subject: [PATCH 0623/1455] drm/xe: Reset VMA attributes to default in SVM
 garbage collector

Restore default memory attributes for VMAs during garbage collection
if they were modified by madvise. Reuse existing VMA if fully overlapping;
otherwise, allocate a new mirror VMA.

v2 (Matthew Brost)
- Add helper for vma split
- Add retry to get updated vma

v3
- Rebase on gpuvm layer

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-19-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_svm.c |  80 ++++++++++++++++--
 drivers/gpu/drm/xe/xe_vm.c  | 160 ++++++++++++++++++++++++++----------
 drivers/gpu/drm/xe/xe_vm.h  |   2 +
 3 files changed, 192 insertions(+), 50 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 19585a3d9f69..0596039ef0a1 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -253,10 +253,56 @@ static int __xe_svm_garbage_collector(struct xe_vm *vm,
 	return 0;
 }
 
+static int xe_svm_range_set_default_attr(struct xe_vm *vm, u64 range_start, u64 range_end)
+{
+	struct xe_vma *vma;
+	struct xe_vma_mem_attr default_attr = {
+		.preferred_loc = {
+			.devmem_fd = DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE,
+			.migration_policy = DRM_XE_MIGRATE_ALL_PAGES,
+		},
+		.atomic_access = DRM_XE_ATOMIC_UNDEFINED,
+	};
+	int err = 0;
+
+	vma = xe_vm_find_vma_by_addr(vm, range_start);
+	if (!vma)
+		return -EINVAL;
+
+	if (xe_vma_has_default_mem_attrs(vma))
+		return 0;
+
+	vm_dbg(&vm->xe->drm, "Existing VMA start=0x%016llx, vma_end=0x%016llx",
+	       xe_vma_start(vma), xe_vma_end(vma));
+
+	if (xe_vma_start(vma) == range_start && xe_vma_end(vma) == range_end) {
+		default_attr.pat_index = vma->attr.default_pat_index;
+		default_attr.default_pat_index  = vma->attr.default_pat_index;
+		vma->attr = default_attr;
+	} else {
+		vm_dbg(&vm->xe->drm, "Split VMA start=0x%016llx, vma_end=0x%016llx",
+		       range_start, range_end);
+		err = xe_vm_alloc_cpu_addr_mirror_vma(vm, range_start, range_end - range_start);
+		if (err) {
+			drm_warn(&vm->xe->drm, "VMA SPLIT failed: %pe\n", ERR_PTR(err));
+			xe_vm_kill(vm, true);
+			return err;
+		}
+	}
+
+	/*
+	 * On call from xe_svm_handle_pagefault original VMA might be changed
+	 * signal this to lookup for VMA again.
+	 */
+	return -EAGAIN;
+}
+
 static int xe_svm_garbage_collector(struct xe_vm *vm)
 {
 	struct xe_svm_range *range;
-	int err;
+	u64 range_start;
+	u64 range_end;
+	int err, ret = 0;
 
 	lockdep_assert_held_write(&vm->lock);
 
@@ -271,6 +317,9 @@ static int xe_svm_garbage_collector(struct xe_vm *vm)
 		if (!range)
 			break;
 
+		range_start = xe_svm_range_start(range);
+		range_end = xe_svm_range_end(range);
+
 		list_del(&range->garbage_collector_link);
 		spin_unlock(&vm->svm.garbage_collector.lock);
 
@@ -283,11 +332,19 @@ static int xe_svm_garbage_collector(struct xe_vm *vm)
 			return err;
 		}
 
+		err = xe_svm_range_set_default_attr(vm, range_start, range_end);
+		if (err) {
+			if (err == -EAGAIN)
+				ret = -EAGAIN;
+			else
+				return err;
+		}
+
 		spin_lock(&vm->svm.garbage_collector.lock);
 	}
 	spin_unlock(&vm->svm.garbage_collector.lock);
 
-	return 0;
+	return ret;
 }
 
 static void xe_svm_garbage_collector_work_func(struct work_struct *w)
@@ -927,13 +984,26 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 			    struct xe_gt *gt, u64 fault_addr,
 			    bool atomic)
 {
-	int need_vram;
-
+	int need_vram, ret;
+retry:
 	need_vram = xe_vma_need_vram_for_atomic(vm->xe, vma, atomic);
 	if (need_vram < 0)
 		return need_vram;
 
-	return __xe_svm_handle_pagefault(vm, vma, gt, fault_addr, need_vram ? true : false);
+	ret =  __xe_svm_handle_pagefault(vm, vma, gt, fault_addr,
+					 need_vram ? true : false);
+	if (ret == -EAGAIN) {
+		/*
+		 * Retry once on -EAGAIN to re-lookup the VMA, as the original VMA
+		 * may have been split by xe_svm_range_set_default_attr.
+		 */
+		vma = xe_vm_find_vma_by_addr(vm, fault_addr);
+		if (!vma)
+			return -EINVAL;
+
+		goto retry;
+	}
+	return ret;
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index e9615d5ef944..ae95407691d8 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -4291,35 +4291,26 @@ int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool i
 	}
 }
 
-/**
- * xe_vm_alloc_madvise_vma - Allocate VMA's with madvise ops
- * @vm: Pointer to the xe_vm structure
- * @start: Starting input address
- * @range: Size of the input range
- *
- * This function splits existing vma to create new vma for user provided input range
- *
- *  Return: 0 if success
- */
-int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
+static int xe_vm_alloc_vma(struct xe_vm *vm,
+			   struct drm_gpuvm_map_req *map_req,
+			   bool is_madvise)
 {
-	struct drm_gpuvm_map_req map_req = {
-		.map.va.addr = start,
-		.map.va.range = range,
-	};
-
 	struct xe_vma_ops vops;
 	struct drm_gpuva_ops *ops = NULL;
 	struct drm_gpuva_op *__op;
 	bool is_cpu_addr_mirror = false;
 	bool remap_op = false;
 	struct xe_vma_mem_attr tmp_attr;
+	u16 default_pat;
 	int err;
 
 	lockdep_assert_held_write(&vm->lock);
 
-	vm_dbg(&vm->xe->drm, "MADVISE_OPS_CREATE: addr=0x%016llx, size=0x%016llx", start, range);
-	ops = drm_gpuvm_madvise_ops_create(&vm->gpuvm, &map_req);
+	if (is_madvise)
+		ops = drm_gpuvm_madvise_ops_create(&vm->gpuvm, map_req);
+	else
+		ops = drm_gpuvm_sm_map_ops_create(&vm->gpuvm, map_req);
+
 	if (IS_ERR(ops))
 		return PTR_ERR(ops);
 
@@ -4330,33 +4321,57 @@ int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
 
 	drm_gpuva_for_each_op(__op, ops) {
 		struct xe_vma_op *op = gpuva_op_to_vma_op(__op);
+		struct xe_vma *vma = NULL;
 
-		if (__op->op == DRM_GPUVA_OP_REMAP) {
-			xe_assert(vm->xe, !remap_op);
-			remap_op = true;
+		if (!is_madvise) {
+			if (__op->op == DRM_GPUVA_OP_UNMAP) {
+				vma = gpuva_to_vma(op->base.unmap.va);
+				XE_WARN_ON(!xe_vma_has_default_mem_attrs(vma));
+				default_pat = vma->attr.default_pat_index;
+			}
 
-			if (xe_vma_is_cpu_addr_mirror(gpuva_to_vma(op->base.remap.unmap->va)))
-				is_cpu_addr_mirror = true;
-			else
-				is_cpu_addr_mirror = false;
+			if (__op->op == DRM_GPUVA_OP_REMAP) {
+				vma = gpuva_to_vma(op->base.remap.unmap->va);
+				default_pat = vma->attr.default_pat_index;
+			}
+
+			if (__op->op == DRM_GPUVA_OP_MAP) {
+				op->map.is_cpu_addr_mirror = true;
+				op->map.pat_index = default_pat;
+			}
+		} else {
+			if (__op->op == DRM_GPUVA_OP_REMAP) {
+				vma = gpuva_to_vma(op->base.remap.unmap->va);
+				xe_assert(vm->xe, !remap_op);
+				xe_assert(vm->xe, xe_vma_has_no_bo(vma));
+				remap_op = true;
+
+				if (xe_vma_is_cpu_addr_mirror(vma))
+					is_cpu_addr_mirror = true;
+				else
+					is_cpu_addr_mirror = false;
+			}
+
+			if (__op->op == DRM_GPUVA_OP_MAP) {
+				xe_assert(vm->xe, remap_op);
+				remap_op = false;
+				/*
+				 * In case of madvise ops DRM_GPUVA_OP_MAP is
+				 * always after DRM_GPUVA_OP_REMAP, so ensure
+				 * we assign op->map.is_cpu_addr_mirror true
+				 * if REMAP is for xe_vma_is_cpu_addr_mirror vma
+				 */
+				op->map.is_cpu_addr_mirror = is_cpu_addr_mirror;
+			}
 		}
-
-		if (__op->op == DRM_GPUVA_OP_MAP) {
-			xe_assert(vm->xe, remap_op);
-			remap_op = false;
-
-			/* In case of madvise ops DRM_GPUVA_OP_MAP is always after
-			 * DRM_GPUVA_OP_REMAP, so ensure we assign op->map.is_cpu_addr_mirror true
-			 * if REMAP is for xe_vma_is_cpu_addr_mirror vma
-			 */
-			op->map.is_cpu_addr_mirror = is_cpu_addr_mirror;
-		}
-
 		print_op(vm->xe, __op);
 	}
 
 	xe_vma_ops_init(&vops, vm, NULL, NULL, 0);
-	vops.flags |= XE_VMA_OPS_FLAG_MADVISE;
+
+	if (is_madvise)
+		vops.flags |= XE_VMA_OPS_FLAG_MADVISE;
+
 	err = vm_bind_ioctl_ops_parse(vm, ops, &vops);
 	if (err)
 		goto unwind_ops;
@@ -4368,15 +4383,20 @@ int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
 		struct xe_vma *vma;
 
 		if (__op->op == DRM_GPUVA_OP_UNMAP) {
-			/* There should be no unmap */
-			XE_WARN_ON("UNEXPECTED UNMAP");
-			xe_vma_destroy(gpuva_to_vma(op->base.unmap.va), NULL);
+			vma = gpuva_to_vma(op->base.unmap.va);
+			/* There should be no unmap for madvise */
+			if (is_madvise)
+				XE_WARN_ON("UNEXPECTED UNMAP");
+
+			xe_vma_destroy(vma, NULL);
 		} else if (__op->op == DRM_GPUVA_OP_REMAP) {
 			vma = gpuva_to_vma(op->base.remap.unmap->va);
-			/* Store attributes for REMAP UNMAPPED VMA, so they can be assigned
-			 * to newly MAP created vma.
+			/* In case of madvise ops Store attributes for REMAP UNMAPPED
+			 * VMA, so they can be assigned to newly MAP created vma.
 			 */
-			tmp_attr = vma->attr;
+			if (is_madvise)
+				tmp_attr = vma->attr;
+
 			xe_vma_destroy(gpuva_to_vma(op->base.remap.unmap->va), NULL);
 		} else if (__op->op == DRM_GPUVA_OP_MAP) {
 			vma = op->map.vma;
@@ -4384,7 +4404,8 @@ int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
 			 * Therefore temp_attr will always have sane values, making it safe to
 			 * copy them to new vma.
 			 */
-			vma->attr = tmp_attr;
+			if (is_madvise)
+				vma->attr = tmp_attr;
 		}
 	}
 
@@ -4398,3 +4419,52 @@ free_ops:
 	drm_gpuva_ops_free(&vm->gpuvm, ops);
 	return err;
 }
+
+/**
+ * xe_vm_alloc_madvise_vma - Allocate VMA's with madvise ops
+ * @vm: Pointer to the xe_vm structure
+ * @start: Starting input address
+ * @range: Size of the input range
+ *
+ * This function splits existing vma to create new vma for user provided input range
+ *
+ * Return: 0 if success
+ */
+int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
+{
+	struct drm_gpuvm_map_req map_req = {
+		.map.va.addr = start,
+		.map.va.range = range,
+	};
+
+	lockdep_assert_held_write(&vm->lock);
+
+	vm_dbg(&vm->xe->drm, "MADVISE_OPS_CREATE: addr=0x%016llx, size=0x%016llx", start, range);
+
+	return xe_vm_alloc_vma(vm, &map_req, true);
+}
+
+/**
+ * xe_vm_alloc_cpu_addr_mirror_vma - Allocate CPU addr mirror vma
+ * @vm: Pointer to the xe_vm structure
+ * @start: Starting input address
+ * @range: Size of the input range
+ *
+ * This function splits/merges existing vma to create new vma for user provided input range
+ *
+ * Return: 0 if success
+ */
+int xe_vm_alloc_cpu_addr_mirror_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
+{
+	struct drm_gpuvm_map_req map_req = {
+		.map.va.addr = start,
+		.map.va.range = range,
+	};
+
+	lockdep_assert_held_write(&vm->lock);
+
+	vm_dbg(&vm->xe->drm, "CPU_ADDR_MIRROR_VMA_OPS_CREATE: addr=0x%016llx, size=0x%016llx",
+	       start, range);
+
+	return xe_vm_alloc_vma(vm, &map_req, false);
+}
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 1a4d67f29c6b..015ddb8b8f99 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -177,6 +177,8 @@ int xe_vma_need_vram_for_atomic(struct xe_device *xe, struct xe_vma *vma, bool i
 
 int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t addr, uint64_t size);
 
+int xe_vm_alloc_cpu_addr_mirror_vma(struct xe_vm *vm, uint64_t addr, uint64_t size);
+
 /**
  * to_userptr_vma() - Return a pointer to an embedding userptr vma
  * @vma: Pointer to the embedded struct xe_vma

From e80b05b09f0e38aabcd5185dc292db0eff7e6e22 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:01:03 +0530
Subject: [PATCH 0624/1455] drm/xe: Enable madvise ioctl for xe

Ioctl enables setting up of memory attributes in user provided range.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-20-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index d4e186c6545a..42eb55fd80ce 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -64,6 +64,7 @@
 #include "xe_ttm_stolen_mgr.h"
 #include "xe_ttm_sys_mgr.h"
 #include "xe_vm.h"
+#include "xe_vm_madvise.h"
 #include "xe_vram.h"
 #include "xe_vram_types.h"
 #include "xe_vsec.h"
@@ -202,6 +203,7 @@ static const struct drm_ioctl_desc xe_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(XE_WAIT_USER_FENCE, xe_wait_user_fence_ioctl,
 			  DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(XE_MADVISE, xe_vm_madvise_ioctl, DRM_RENDER_ALLOW),
 };
 
 static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)

From 418807860e94eb9c2fe07a6f5bf67de4c59a97e4 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 21 Aug 2025 23:01:04 +0530
Subject: [PATCH 0625/1455] drm/xe/uapi: Add UAPI for querying VMA count and
 memory attributes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce the DRM_IOCTL_XE_VM_QUERY_MEMORY_RANGE_ATTRS ioctl to allow
userspace to query memory attributes of VMAs within a user specified
virtual address range.

Userspace first calls the ioctl with num_mem_ranges = 0,
sizeof_mem_ranges_attr = 0 and vector_of_vma_mem_attr = NULL to retrieve
the number of memory ranges (vmas) and size of each memory range attribute.
Then, it allocates a buffer of that size and calls the ioctl again to fill
the buffer with memory range attributes.

This two-step interface allows userspace to first query the required
buffer size, then retrieve detailed attributes efficiently.

v2 (Matthew Brost)
- Use same ioctl to overload functionality

v3
- Add kernel-doc

v4
- Make uapi future proof by passing struct size (Matthew Brost)
- make lock interruptible (Matthew Brost)
- set reserved bits to zero (Matthew Brost)
- s/__copy_to_user/copy_to_user (Matthew Brost)
- Avod using VMA term in uapi (Thomas)
- xe_vm_put(vm) is missing (Shuicheng)

v5
- Nits
- Fix kernel-doc

Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250821173104.3030148-21-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c |   2 +
 drivers/gpu/drm/xe/xe_vm.c     | 102 ++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_vm.h     |   2 +-
 include/uapi/drm/xe_drm.h      | 140 +++++++++++++++++++++++++++++++++
 4 files changed, 245 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 42eb55fd80ce..90617e627bc4 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -204,6 +204,8 @@ static const struct drm_ioctl_desc xe_ioctls[] = {
 			  DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(XE_OBSERVATION, xe_observation_ioctl, DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(XE_MADVISE, xe_vm_madvise_ioctl, DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(XE_VM_QUERY_MEM_RANGE_ATTRS, xe_vm_query_vmas_attrs_ioctl,
+			  DRM_RENDER_ALLOW),
 };
 
 static long xe_drm_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index ae95407691d8..e913efa2057b 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2196,6 +2196,108 @@ int xe_vm_destroy_ioctl(struct drm_device *dev, void *data,
 	return err;
 }
 
+static int xe_vm_query_vmas(struct xe_vm *vm, u64 start, u64 end)
+{
+	struct drm_gpuva *gpuva;
+	u32 num_vmas = 0;
+
+	lockdep_assert_held(&vm->lock);
+	drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end)
+		num_vmas++;
+
+	return num_vmas;
+}
+
+static int get_mem_attrs(struct xe_vm *vm, u32 *num_vmas, u64 start,
+			 u64 end, struct drm_xe_mem_range_attr *attrs)
+{
+	struct drm_gpuva *gpuva;
+	int i = 0;
+
+	lockdep_assert_held(&vm->lock);
+
+	drm_gpuvm_for_each_va_range(gpuva, &vm->gpuvm, start, end) {
+		struct xe_vma *vma = gpuva_to_vma(gpuva);
+
+		if (i == *num_vmas)
+			return -ENOSPC;
+
+		attrs[i].start = xe_vma_start(vma);
+		attrs[i].end = xe_vma_end(vma);
+		attrs[i].atomic.val = vma->attr.atomic_access;
+		attrs[i].pat_index.val = vma->attr.pat_index;
+		attrs[i].preferred_mem_loc.devmem_fd = vma->attr.preferred_loc.devmem_fd;
+		attrs[i].preferred_mem_loc.migration_policy =
+		vma->attr.preferred_loc.migration_policy;
+
+		i++;
+	}
+
+	*num_vmas = i;
+	return 0;
+}
+
+int xe_vm_query_vmas_attrs_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
+{
+	struct xe_device *xe = to_xe_device(dev);
+	struct xe_file *xef = to_xe_file(file);
+	struct drm_xe_mem_range_attr *mem_attrs;
+	struct drm_xe_vm_query_mem_range_attr *args = data;
+	u64 __user *attrs_user = u64_to_user_ptr(args->vector_of_mem_attr);
+	struct xe_vm *vm;
+	int err = 0;
+
+	if (XE_IOCTL_DBG(xe,
+			 ((args->num_mem_ranges == 0 &&
+			  (attrs_user || args->sizeof_mem_range_attr != 0)) ||
+			 (args->num_mem_ranges > 0 &&
+			  (!attrs_user ||
+			   args->sizeof_mem_range_attr !=
+			   sizeof(struct drm_xe_mem_range_attr))))))
+		return -EINVAL;
+
+	vm = xe_vm_lookup(xef, args->vm_id);
+	if (XE_IOCTL_DBG(xe, !vm))
+		return -EINVAL;
+
+	err = down_read_interruptible(&vm->lock);
+	if (err)
+		goto put_vm;
+
+	attrs_user = u64_to_user_ptr(args->vector_of_mem_attr);
+
+	if (args->num_mem_ranges == 0 && !attrs_user) {
+		args->num_mem_ranges = xe_vm_query_vmas(vm, args->start, args->start + args->range);
+		args->sizeof_mem_range_attr = sizeof(struct drm_xe_mem_range_attr);
+		goto unlock_vm;
+	}
+
+	mem_attrs = kvmalloc_array(args->num_mem_ranges, args->sizeof_mem_range_attr,
+				   GFP_KERNEL | __GFP_ACCOUNT |
+				   __GFP_RETRY_MAYFAIL | __GFP_NOWARN);
+	if (!mem_attrs) {
+		err = args->num_mem_ranges > 1 ? -ENOBUFS : -ENOMEM;
+		goto unlock_vm;
+	}
+
+	memset(mem_attrs, 0, args->num_mem_ranges * args->sizeof_mem_range_attr);
+	err = get_mem_attrs(vm, &args->num_mem_ranges, args->start,
+			    args->start + args->range, mem_attrs);
+	if (err)
+		goto free_mem_attrs;
+
+	err = copy_to_user(attrs_user, mem_attrs,
+			   args->sizeof_mem_range_attr * args->num_mem_ranges);
+
+free_mem_attrs:
+	kvfree(mem_attrs);
+unlock_vm:
+	up_read(&vm->lock);
+put_vm:
+	xe_vm_put(vm);
+	return err;
+}
+
 static bool vma_matches(struct xe_vma *vma, u64 page_addr)
 {
 	if (page_addr > xe_vma_end(vma) - 1 ||
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 015ddb8b8f99..8afd2cd4c9dc 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -199,7 +199,7 @@ int xe_vm_destroy_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file);
 int xe_vm_bind_ioctl(struct drm_device *dev, void *data,
 		     struct drm_file *file);
-
+int xe_vm_query_vmas_attrs_ioctl(struct drm_device *dev, void *data, struct drm_file *file);
 void xe_vm_close_and_put(struct xe_vm *vm);
 
 static inline bool xe_vm_in_fault_mode(struct xe_vm *vm)
diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index 115b9bca2a25..7dedd45ab995 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -82,6 +82,7 @@ extern "C" {
  *  - &DRM_IOCTL_XE_WAIT_USER_FENCE
  *  - &DRM_IOCTL_XE_OBSERVATION
  *  - &DRM_IOCTL_XE_MADVISE
+ *  - &DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS
  */
 
 /*
@@ -104,6 +105,7 @@ extern "C" {
 #define DRM_XE_WAIT_USER_FENCE		0x0a
 #define DRM_XE_OBSERVATION		0x0b
 #define DRM_XE_MADVISE			0x0c
+#define DRM_XE_VM_QUERY_MEM_RANGE_ATTRS	0x0d
 
 /* Must be kept compact -- no holes */
 
@@ -120,6 +122,7 @@ extern "C" {
 #define DRM_IOCTL_XE_WAIT_USER_FENCE		DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_WAIT_USER_FENCE, struct drm_xe_wait_user_fence)
 #define DRM_IOCTL_XE_OBSERVATION		DRM_IOW(DRM_COMMAND_BASE + DRM_XE_OBSERVATION, struct drm_xe_observation_param)
 #define DRM_IOCTL_XE_MADVISE			DRM_IOW(DRM_COMMAND_BASE + DRM_XE_MADVISE, struct drm_xe_madvise)
+#define DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS	DRM_IOWR(DRM_COMMAND_BASE + DRM_XE_VM_QUERY_MEM_RANGE_ATTRS, struct drm_xe_vm_query_mem_range_attr)
 
 /**
  * DOC: Xe IOCTL Extensions
@@ -2113,6 +2116,143 @@ struct drm_xe_madvise {
 	__u64 reserved[2];
 };
 
+/**
+ * struct drm_xe_mem_range_attr - Output of &DRM_IOCTL_XE_VM_QUERY_MEM_RANGES_ATTRS
+ *
+ * This structure is provided by userspace and filled by KMD in response to the
+ * DRM_IOCTL_XE_VM_QUERY_MEM_RANGES_ATTRS ioctl. It describes memory attributes of
+ * a memory ranges within a user specified address range in a VM.
+ *
+ * The structure includes information such as atomic access policy,
+ * page attribute table (PAT) index, and preferred memory location.
+ * Userspace allocates an array of these structures and passes a pointer to the
+ * ioctl to retrieve attributes for each memory ranges
+ *
+ * @extensions: Pointer to the first extension struct, if any
+ * @start: Start address of the memory range
+ * @end: End address of the virtual memory range
+ *
+ */
+struct drm_xe_mem_range_attr {
+	 /** @extensions: Pointer to the first extension struct, if any */
+	__u64 extensions;
+
+	/** @start: start of the memory range */
+	__u64 start;
+
+	/** @end: end of the memory range */
+	__u64 end;
+
+	/** @preferred_mem_loc: preferred memory location */
+	struct {
+		/** @preferred_mem_loc.devmem_fd: fd for preferred loc */
+		__u32 devmem_fd;
+
+		/** @preferred_mem_loc.migration_policy: Page migration policy */
+		__u32 migration_policy;
+	} preferred_mem_loc;
+
+	/** @atomic: Atomic access policy */
+	struct {
+		/** @atomic.val: atomic attribute */
+		__u32 val;
+
+		/** @atomic.reserved: Reserved */
+		__u32 reserved;
+	} atomic;
+
+	 /** @pat_index: Page attribute table index */
+	struct {
+		/** @pat_index.val: PAT index */
+		__u32 val;
+
+		/** @pat_index.reserved: Reserved */
+		__u32 reserved;
+	} pat_index;
+
+	/** @reserved: Reserved */
+	__u64 reserved[2];
+};
+
+/**
+ * struct drm_xe_vm_query_mem_range_attr - Input of &DRM_IOCTL_XE_VM_QUERY_MEM_ATTRIBUTES
+ *
+ * This structure is used to query memory attributes of memory regions
+ * within a user specified address range in a VM. It provides detailed
+ * information about each memory range, including atomic access policy,
+ * page attribute table (PAT) index, and preferred memory location.
+ *
+ * Userspace first calls the ioctl with @num_mem_ranges = 0,
+ * @sizeof_mem_ranges_attr = 0 and @vector_of_vma_mem_attr = NULL to retrieve
+ * the number of memory regions and size of each memory range attribute.
+ * Then, it allocates a buffer of that size and calls the ioctl again to fill
+ * the buffer with memory range attributes.
+ *
+ * If second call fails with -ENOSPC, it means memory ranges changed between
+ * first call and now, retry IOCTL again with @num_mem_ranges = 0,
+ * @sizeof_mem_ranges_attr = 0 and @vector_of_vma_mem_attr = NULL followed by
+ * Second ioctl call.
+ *
+ * Example:
+ *
+ * .. code-block:: C
+ *    struct drm_xe_vm_query_mem_range_attr query = {
+ *         .vm_id = vm_id,
+ *         .start = 0x100000,
+ *         .range = 0x2000,
+ *     };
+ *
+ *    // First ioctl call to get num of mem regions and sizeof each attribute
+ *    ioctl(fd, DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS, &query);
+ *
+ *    // Allocate buffer for the memory region attributes
+ *    void *ptr = malloc(query.num_mem_ranges * query.sizeof_mem_range_attr);
+ *    void *ptr_start = ptr;
+ *
+ *    query.vector_of_mem_attr = (uintptr_t)ptr;
+ *
+ *    // Second ioctl call to actually fill the memory attributes
+ *    ioctl(fd, DRM_IOCTL_XE_VM_QUERY_MEM_RANGE_ATTRS, &query);
+ *
+ *    // Iterate over the returned memory region attributes
+ *    for (unsigned int i = 0; i < query.num_mem_ranges; ++i) {
+ *       struct drm_xe_mem_range_attr *attr = (struct drm_xe_mem_range_attr *)ptr;
+ *
+ *       // Do something with attr
+ *
+ *       // Move pointer by one entry
+ *       ptr += query.sizeof_mem_range_attr;
+ *     }
+ *
+ *    free(ptr_start);
+ */
+struct drm_xe_vm_query_mem_range_attr {
+	/** @extensions: Pointer to the first extension struct, if any */
+	__u64 extensions;
+
+	/** @vm_id: vm_id of the virtual range */
+	__u32 vm_id;
+
+	/** @num_mem_ranges: number of mem_ranges in range */
+	__u32 num_mem_ranges;
+
+	/** @start: start of the virtual address range */
+	__u64 start;
+
+	/** @range: size of the virtual address range */
+	__u64 range;
+
+	/** @sizeof_mem_range_attr: size of struct drm_xe_mem_range_attr */
+	__u64 sizeof_mem_range_attr;
+
+	/** @vector_of_mem_attr: userptr to array of struct drm_xe_mem_range_attr */
+	__u64 vector_of_mem_attr;
+
+	/** @reserved: Reserved */
+	__u64 reserved[2];
+
+};
+
 #if defined(__cplusplus)
 }
 #endif

From c65ee1b1ffc68e274d6c1cebdf9df09407d180fd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 22 Aug 2025 13:58:46 +0300
Subject: [PATCH 0626/1455] drm/i915/psr: Do not unnecessarily remove underrun
 on idle PSR WA
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are currently removing underrun on idle PSR WA even if it's not
applied. Fix this by checking pkg_c_latency_used on PSR exit as well.

Fixes: 9b1795e9b0ae ("drm/i915/psr: Underrun on idle PSR wa only when pkgc latency > delayed vblank")
Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Uma Shankar <uma.shankar@intel.com>
Link: https://lore.kernel.org/r/20250822105846.1023631-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index eec4f7dc2d66..681e20d92376 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -2095,8 +2095,9 @@ static void intel_psr_exit(struct intel_dp *intel_dp)
 
 		drm_WARN_ON(display->drm, !(val & EDP_PSR2_ENABLE));
 	} else {
-		if (DISPLAY_VER(display) == 20 ||
-		    IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0))
+		if ((DISPLAY_VER(display) == 20 ||
+		     IS_DISPLAY_VERx100_STEP(display, 3000, STEP_A0, STEP_B0)) &&
+			intel_dp->psr.pkg_c_latency_used)
 			intel_dmc_start_pkgc_exit_at_start_of_undelayed_vblank(display,
 								       intel_dp->psr.pipe,
 								       false);

From cff5fb82733c4f1acda458ffd2bb5c948fb59bd6 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:01 +0200
Subject: [PATCH 0627/1455] video: pixel_format: Add compare helpers

Add helpers that compare two pixel-format descriptions against
each other.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-2-tzimmermann@suse.de
---
 include/video/pixel_format.h | 58 ++++++++++++++++++++++++++++++++++++
 1 file changed, 58 insertions(+)

diff --git a/include/video/pixel_format.h b/include/video/pixel_format.h
index b5104b2a3a13..c57019cd6ea8 100644
--- a/include/video/pixel_format.h
+++ b/include/video/pixel_format.h
@@ -38,4 +38,62 @@ struct pixel_format {
 #define PIXEL_FORMAT_XRGB2101010 \
 	{ 32, false, { .alpha = {0, 0}, .red = {20, 10}, .green = {10, 10}, .blue = {0, 10} } }
 
+#define __pixel_format_cmp_field(lhs, rhs, name) \
+	{ \
+		int ret = ((lhs)->name) - ((rhs)->name); \
+		if (ret) \
+			return ret; \
+	}
+
+#define __pixel_format_cmp_bitfield(lhs, rhs, name) \
+	{ \
+		__pixel_format_cmp_field(lhs, rhs, name.offset); \
+		__pixel_format_cmp_field(lhs, rhs, name.length); \
+	}
+
+/**
+ * pixel_format_cmp - Compares two pixel-format descriptions
+ *
+ * @lhs: a pixel-format description
+ * @rhs: a pixel-format description
+ *
+ * Compares two pixel-format descriptions for their order. The semantics
+ * are equivalent to memcmp().
+ *
+ * Returns:
+ * 0 if both arguments describe the same pixel format, less-than-zero if lhs < rhs,
+ * or greater-than-zero if lhs > rhs.
+ */
+static inline int pixel_format_cmp(const struct pixel_format *lhs, const struct pixel_format *rhs)
+{
+	__pixel_format_cmp_field(lhs, rhs, bits_per_pixel);
+	__pixel_format_cmp_field(lhs, rhs, indexed);
+
+	if (lhs->indexed) {
+		__pixel_format_cmp_bitfield(lhs, rhs, index);
+	} else {
+		__pixel_format_cmp_bitfield(lhs, rhs, alpha);
+		__pixel_format_cmp_bitfield(lhs, rhs, red);
+		__pixel_format_cmp_bitfield(lhs, rhs, green);
+		__pixel_format_cmp_bitfield(lhs, rhs, blue);
+	}
+
+	return 0;
+}
+
+/**
+ * pixel_format_equal - Compares two pixel-format descriptions for equality
+ *
+ * @lhs: a pixel-format description
+ * @rhs: a pixel-format description
+ *
+ * Returns:
+ * True if both arguments describe the same pixel format, or false otherwise.
+ */
+static inline bool pixel_format_equal(const struct pixel_format *lhs,
+				      const struct pixel_format *rhs)
+{
+	return !pixel_format_cmp(lhs, rhs);
+}
+
 #endif

From d6d05e2af796ca25094f80a73d8841505d54368b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:02 +0200
Subject: [PATCH 0628/1455] video: screen_info: Add pixel-format helper for
 linear framebuffers

Add screen_info_pixel_format(), which converts a screen_info's
information about the color format to struct pixel_format. The encoding
within the screen_info structure is complex and therefore prone to
errors. Later patches will convert callers to use the pixel format.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-3-tzimmermann@suse.de
---
 drivers/video/screen_info_generic.c | 55 +++++++++++++++++++++++++++++
 include/linux/screen_info.h         |  2 ++
 2 files changed, 57 insertions(+)

diff --git a/drivers/video/screen_info_generic.c b/drivers/video/screen_info_generic.c
index 900e9386eceb..763adbba71cb 100644
--- a/drivers/video/screen_info_generic.c
+++ b/drivers/video/screen_info_generic.c
@@ -5,6 +5,8 @@
 #include <linux/screen_info.h>
 #include <linux/string.h>
 
+#include <video/pixel_format.h>
+
 static void resource_init_named(struct resource *r,
 				resource_size_t start, resource_size_t size,
 				const char *name, unsigned int flags)
@@ -180,3 +182,56 @@ u32 __screen_info_lfb_bits_per_pixel(const struct screen_info *si)
 	return bits_per_pixel;
 }
 EXPORT_SYMBOL(__screen_info_lfb_bits_per_pixel);
+
+static int __screen_info_lfb_pixel_format(const struct screen_info *si, struct pixel_format *f)
+{
+	u32 bits_per_pixel = __screen_info_lfb_bits_per_pixel(si);
+
+	if (bits_per_pixel > U8_MAX)
+		return -EINVAL;
+
+	f->bits_per_pixel = bits_per_pixel;
+
+	if (si->lfb_depth > 8) {
+		f->indexed = false;
+		f->alpha.offset = 0;
+		f->alpha.length = 0;
+		f->red.offset = si->red_pos;
+		f->red.length = si->red_size;
+		f->green.offset = si->green_pos;
+		f->green.length = si->green_size;
+		f->blue.offset = si->blue_pos;
+		f->blue.length = si->blue_size;
+	} else {
+		f->indexed = true;
+		f->index.offset = 0;
+		f->index.length = si->lfb_depth;
+	}
+
+	return 0;
+}
+
+/**
+ * screen_info_pixel_format - Returns the screen-info format as pixel-format description
+ *
+ * @si: the screen_info
+ * @f: pointer to return pixel-format description
+ *
+ * Returns:
+ * 0 on success, or a negative errno code otherwise.
+ */
+int screen_info_pixel_format(const struct screen_info *si, struct pixel_format *f)
+{
+	unsigned int type = screen_info_video_type(si);
+
+	/* TODO: Add support for additional types as needed. */
+	switch (type) {
+	case VIDEO_TYPE_VLFB:
+	case VIDEO_TYPE_EFI:
+		return __screen_info_lfb_pixel_format(si, f);
+	}
+
+	/* not supported */
+	return -EINVAL;
+}
+EXPORT_SYMBOL(screen_info_pixel_format);
diff --git a/include/linux/screen_info.h b/include/linux/screen_info.h
index 923d68e07679..1690706206e8 100644
--- a/include/linux/screen_info.h
+++ b/include/linux/screen_info.h
@@ -12,6 +12,7 @@
 #define SCREEN_INFO_MAX_RESOURCES	3
 
 struct pci_dev;
+struct pixel_format;
 struct resource;
 
 static inline bool __screen_info_has_lfb(unsigned int type)
@@ -136,6 +137,7 @@ static inline u32 __screen_info_vesapm_info_base(const struct screen_info *si)
 ssize_t screen_info_resources(const struct screen_info *si, struct resource *r, size_t num);
 
 u32 __screen_info_lfb_bits_per_pixel(const struct screen_info *si);
+int screen_info_pixel_format(const struct screen_info *si, struct pixel_format *f);
 
 #if defined(CONFIG_PCI)
 void screen_info_apply_fixups(void);

From 31eea29d727ce35b747e68c6be350ca07b7ecd9b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:03 +0200
Subject: [PATCH 0629/1455] drm/sysfb: Find screen_info format with helpers

Convert drm_sysfb_get_format_si() to lookup the screen_info color
format as struct pixel_format with screen_info_pixel_format(). Then
search the list of given formats for the screen_info format with
pixel_format_equal().

Replaces custom code with helpers. The pixel-compare helper
pixel_format_equal() also handles indexed color formats. Prepares
for sysfb drivers to support color palettes.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-4-tzimmermann@suse.de
---
 drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c | 19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c b/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c
index 0b3fb874a51f..885864168c54 100644
--- a/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_screen_info.c
@@ -79,22 +79,19 @@ const struct drm_format_info *drm_sysfb_get_format_si(struct drm_device *dev,
 						      const struct screen_info *si)
 {
 	const struct drm_format_info *format = NULL;
-	u32 bits_per_pixel;
+	struct pixel_format pixel;
 	size_t i;
+	int ret;
 
-	bits_per_pixel = __screen_info_lfb_bits_per_pixel(si);
+	ret = screen_info_pixel_format(si, &pixel);
+	if (ret)
+		return NULL;
 
 	for (i = 0; i < nformats; ++i) {
-		const struct pixel_format *f = &formats[i].pixel;
+		const struct drm_sysfb_format *f = &formats[i];
 
-		if (bits_per_pixel == f->bits_per_pixel &&
-		    si->red_size == f->red.length &&
-		    si->red_pos == f->red.offset &&
-		    si->green_size == f->green.length &&
-		    si->green_pos == f->green.offset &&
-		    si->blue_size == f->blue.length &&
-		    si->blue_pos == f->blue.offset) {
-			format = drm_format_info(formats[i].fourcc);
+		if (pixel_format_equal(&pixel, &f->pixel)) {
+			format = drm_format_info(f->fourcc);
 			break;
 		}
 	}

From 061963cd9e5b60672695e9a592be440469a6bf72 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:04 +0200
Subject: [PATCH 0630/1455] drm/sysfb: Blit to CRTC destination format

Use the color format stored in struct drm_sysfb_crtc_state for
color-format conversion instead of the scanout-buffer format
announced by firmware. Currently, both values are identical.

This will allow drivers to modify the CRTC's input format to a
certain extend. Specifically, vesadrm will be able to display RGB
framebuffers when the scanout buffer is of C8 format. With color-
format conversion to RGB332 and correct setup of the C8 palette,
displaying XRGB8888-based buffers under C8 can be achieved.

v2:
- refer to RGB332 as CRTC input format

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-5-tzimmermann@suse.de
---
 drivers/gpu/drm/sysfb/drm_sysfb_helper.h  |  2 +-
 drivers/gpu/drm/sysfb/drm_sysfb_modeset.c | 29 +++++++++++++++--------
 2 files changed, 20 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_helper.h b/drivers/gpu/drm/sysfb/drm_sysfb_helper.h
index 1424b63dde99..89633e30ca62 100644
--- a/drivers/gpu/drm/sysfb/drm_sysfb_helper.h
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_helper.h
@@ -132,7 +132,7 @@ int drm_sysfb_plane_helper_get_scanout_buffer(struct drm_plane *plane,
 struct drm_sysfb_crtc_state {
 	struct drm_crtc_state base;
 
-	/* Primary-plane format; required for color mgmt. */
+	/* CRTC input color format; required for color mgmt. */
 	const struct drm_format_info *format;
 };
 
diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
index 1bcdb5ee8f09..0e05345cdeee 100644
--- a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
@@ -210,7 +210,12 @@ int drm_sysfb_plane_helper_atomic_check(struct drm_plane *plane,
 	else if (!new_plane_state->visible)
 		return 0;
 
-	if (new_fb->format != sysfb->fb_format) {
+	new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
+
+	new_sysfb_crtc_state = to_drm_sysfb_crtc_state(new_crtc_state);
+	new_sysfb_crtc_state->format = sysfb->fb_format;
+
+	if (new_fb->format != new_sysfb_crtc_state->format) {
 		void *buf;
 
 		/* format conversion necessary; reserve buffer */
@@ -220,11 +225,6 @@ int drm_sysfb_plane_helper_atomic_check(struct drm_plane *plane,
 			return -ENOMEM;
 	}
 
-	new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
-
-	new_sysfb_crtc_state = to_drm_sysfb_crtc_state(new_crtc_state);
-	new_sysfb_crtc_state->format = new_fb->format;
-
 	return 0;
 }
 EXPORT_SYMBOL(drm_sysfb_plane_helper_atomic_check);
@@ -238,7 +238,10 @@ void drm_sysfb_plane_helper_atomic_update(struct drm_plane *plane, struct drm_at
 	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
 	struct drm_framebuffer *fb = plane_state->fb;
 	unsigned int dst_pitch = sysfb->fb_pitch;
-	const struct drm_format_info *dst_format = sysfb->fb_format;
+	struct drm_crtc_state *crtc_state = crtc_state =
+		drm_atomic_get_new_crtc_state(state, plane_state->crtc);
+	struct drm_sysfb_crtc_state *sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
+	const struct drm_format_info *dst_format = sysfb_crtc_state->format;
 	struct drm_atomic_helper_damage_iter iter;
 	struct drm_rect damage;
 	int ret, idx;
@@ -278,7 +281,10 @@ void drm_sysfb_plane_helper_atomic_disable(struct drm_plane *plane,
 	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
 	void __iomem *dst_vmap = dst.vaddr_iomem; /* TODO: Use mapping abstraction */
 	unsigned int dst_pitch = sysfb->fb_pitch;
-	const struct drm_format_info *dst_format = sysfb->fb_format;
+	struct drm_crtc_state *crtc_state = crtc_state =
+		drm_atomic_get_new_crtc_state(state, plane_state->crtc);
+	struct drm_sysfb_crtc_state *sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
+	const struct drm_format_info *dst_format = sysfb_crtc_state->format;
 	struct drm_rect dst_clip;
 	unsigned long lines, linepixels, i;
 	int idx;
@@ -370,16 +376,19 @@ EXPORT_SYMBOL(drm_sysfb_crtc_helper_atomic_check);
 
 void drm_sysfb_crtc_reset(struct drm_crtc *crtc)
 {
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(crtc->dev);
 	struct drm_sysfb_crtc_state *sysfb_crtc_state;
 
 	if (crtc->state)
 		drm_sysfb_crtc_state_destroy(to_drm_sysfb_crtc_state(crtc->state));
 
 	sysfb_crtc_state = kzalloc(sizeof(*sysfb_crtc_state), GFP_KERNEL);
-	if (sysfb_crtc_state)
+	if (sysfb_crtc_state) {
+		sysfb_crtc_state->format = sysfb->fb_format;
 		__drm_atomic_helper_crtc_reset(crtc, &sysfb_crtc_state->base);
-	else
+	} else {
 		__drm_atomic_helper_crtc_reset(crtc, NULL);
+	}
 }
 EXPORT_SYMBOL(drm_sysfb_crtc_reset);
 

From 7ff61177b7116825085587f007dcdfd042c7b33b Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:05 +0200
Subject: [PATCH 0631/1455] drm/color-mgmt: Prepare for RGB332 palettes

Add helper drm_crtc_fill_palette_332(), which fills palettes with
RGB332 color data. Each color in RGB332 format serves as an index
into an 8-bit palette that stores the corresponding component-based
colors.

Vesadrm will use the new helper to emulate RGB formats on top of
framebuffers in C8 format.

v2:
- add comments on bit operations (Javier)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-6-tzimmermann@suse.de
---
 drivers/gpu/drm/drm_color_mgmt.c | 34 ++++++++++++++++++++++++++++++++
 include/drm/drm_color_mgmt.h     |  1 +
 2 files changed, 35 insertions(+)

diff --git a/drivers/gpu/drm/drm_color_mgmt.c b/drivers/gpu/drm/drm_color_mgmt.c
index 37a3270bc3c2..131c1c9ae92f 100644
--- a/drivers/gpu/drm/drm_color_mgmt.c
+++ b/drivers/gpu/drm/drm_color_mgmt.c
@@ -817,6 +817,40 @@ void drm_crtc_load_palette_8(struct drm_crtc *crtc, const struct drm_color_lut *
 }
 EXPORT_SYMBOL(drm_crtc_load_palette_8);
 
+static void fill_palette_332(struct drm_crtc *crtc, u16 r, u16 g, u16 b,
+			     drm_crtc_set_lut_func set_palette)
+{
+	unsigned int i = (r << 5) | (g << 2) | b; /* 8-bit palette index */
+
+	/* Expand R (3-bit) G (3-bit) and B (2-bit) values to 16-bit values */
+	r = (r << 13) | (r << 10) | (r << 7) | (r << 4) | (r << 1) | (r >> 2);
+	g = (g << 13) | (g << 10) | (g << 7) | (g << 4) | (g << 1) | (g >> 2);
+	b = (b << 14) | (b << 12) | (b << 10) | (b << 8) | (b << 6) | (b << 4) | (b << 2) | b;
+
+	set_palette(crtc, i, r, g, b);
+}
+
+/**
+ * drm_crtc_fill_palette_332 - Programs a default palette for R332-like formats
+ * @crtc: The displaying CRTC
+ * @set_palette: Callback for programming the hardware gamma LUT
+ *
+ * Programs an RGB332 palette to hardware.
+ */
+void drm_crtc_fill_palette_332(struct drm_crtc *crtc, drm_crtc_set_lut_func set_palette)
+{
+	unsigned int r, g, b;
+
+	/* Limits of 8-8-4 are the maximum number of values for each channel. */
+	for (r = 0; r < 8; ++r) {
+		for (g = 0; g < 8; ++g) {
+			for (b = 0; b < 4; ++b)
+				fill_palette_332(crtc, r, g, b, set_palette);
+		}
+	}
+}
+EXPORT_SYMBOL(drm_crtc_fill_palette_332);
+
 static void fill_palette_8(struct drm_crtc *crtc, unsigned int i,
 			   drm_crtc_set_lut_func set_palette)
 {
diff --git a/include/drm/drm_color_mgmt.h b/include/drm/drm_color_mgmt.h
index 6cb577f6dba6..eccb71ab335a 100644
--- a/include/drm/drm_color_mgmt.h
+++ b/include/drm/drm_color_mgmt.h
@@ -143,6 +143,7 @@ void drm_crtc_fill_gamma_555(struct drm_crtc *crtc, drm_crtc_set_lut_func set_ga
 void drm_crtc_load_palette_8(struct drm_crtc *crtc, const struct drm_color_lut *lut,
 			     drm_crtc_set_lut_func set_palette);
 
+void drm_crtc_fill_palette_332(struct drm_crtc *crtc, drm_crtc_set_lut_func set_palette);
 void drm_crtc_fill_palette_8(struct drm_crtc *crtc, drm_crtc_set_lut_func set_palette);
 
 #endif

From b3c7edd5e50e0e92ead8794b084e92d89002dd6f Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:06 +0200
Subject: [PATCH 0632/1455] drm/format-helper: Add XRGB8888-to-RGB332 to
 drm_fb_blit()

List the conversion from XRGB8888-to-RGB332 in drm_fb_blit(), so that
drivers based on sysfb-helpers can use it.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-7-tzimmermann@suse.de
---
 drivers/gpu/drm/drm_format_helper.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c
index 8f3daf38ca63..37d5e0a7eb46 100644
--- a/drivers/gpu/drm/drm_format_helper.c
+++ b/drivers/gpu/drm/drm_format_helper.c
@@ -1243,6 +1243,9 @@ int drm_fb_blit(struct iosys_map *dst, const unsigned int *dst_pitch, uint32_t d
 		} else if (dst_format == DRM_FORMAT_BGRX8888) {
 			drm_fb_swab(dst, dst_pitch, src, fb, clip, false, state);
 			return 0;
+		} else if (dst_format == DRM_FORMAT_RGB332) {
+			drm_fb_xrgb8888_to_rgb332(dst, dst_pitch, src, fb, clip, state);
+			return 0;
 		}
 	}
 

From cbc383cea0942cd827c6f1a3408793ac3a97ea00 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:07 +0200
Subject: [PATCH 0633/1455] drm/vesadrm: Rename vesadrm_set_gamma_lut() to
 vesadrm_set_color_lut()

Rename vesadrm's gamma helpers in preparation of the upcoming support
for color palettes. Gamma correction and color palettes share the same
hardware features, but the driver's old naming only indicated support
for gamma LUTs.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-8-tzimmermann@suse.de
---
 drivers/gpu/drm/sysfb/vesadrm.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
index 90615e9ac86b..9596322c232b 100644
--- a/drivers/gpu/drm/sysfb/vesadrm.c
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -82,7 +82,7 @@ static struct vesadrm_device *to_vesadrm_device(struct drm_device *dev)
 }
 
 /*
- * Palette
+ * Color LUT
  */
 
 static void vesadrm_vga_cmap_write(struct vesadrm_device *vesa, unsigned int index,
@@ -128,7 +128,7 @@ static void vesadrm_pmi_cmap_write(struct vesadrm_device *vesa, unsigned int ind
 }
 #endif
 
-static void vesadrm_set_gamma_lut(struct drm_crtc *crtc, unsigned int index,
+static void vesadrm_set_color_lut(struct drm_crtc *crtc, unsigned int index,
 				  u16 red, u16 green, u16 blue)
 {
 	struct drm_device *dev = crtc->dev;
@@ -149,15 +149,15 @@ static void vesadrm_fill_gamma_lut(struct vesadrm_device *vesa,
 
 	switch (format->format) {
 	case DRM_FORMAT_XRGB1555:
-		drm_crtc_fill_gamma_555(crtc, vesadrm_set_gamma_lut);
+		drm_crtc_fill_gamma_555(crtc, vesadrm_set_color_lut);
 		break;
 	case DRM_FORMAT_RGB565:
-		drm_crtc_fill_gamma_565(crtc, vesadrm_set_gamma_lut);
+		drm_crtc_fill_gamma_565(crtc, vesadrm_set_color_lut);
 		break;
 	case DRM_FORMAT_RGB888:
 	case DRM_FORMAT_XRGB8888:
 	case DRM_FORMAT_BGRX8888:
-		drm_crtc_fill_gamma_888(crtc, vesadrm_set_gamma_lut);
+		drm_crtc_fill_gamma_888(crtc, vesadrm_set_color_lut);
 		break;
 	default:
 		drm_warn_once(dev, "Unsupported format %p4cc for gamma correction\n",
@@ -175,15 +175,15 @@ static void vesadrm_load_gamma_lut(struct vesadrm_device *vesa,
 
 	switch (format->format) {
 	case DRM_FORMAT_XRGB1555:
-		drm_crtc_load_gamma_555_from_888(crtc, lut, vesadrm_set_gamma_lut);
+		drm_crtc_load_gamma_555_from_888(crtc, lut, vesadrm_set_color_lut);
 		break;
 	case DRM_FORMAT_RGB565:
-		drm_crtc_load_gamma_565_from_888(crtc, lut, vesadrm_set_gamma_lut);
+		drm_crtc_load_gamma_565_from_888(crtc, lut, vesadrm_set_color_lut);
 		break;
 	case DRM_FORMAT_RGB888:
 	case DRM_FORMAT_XRGB8888:
 	case DRM_FORMAT_BGRX8888:
-		drm_crtc_load_gamma_888(crtc, lut, vesadrm_set_gamma_lut);
+		drm_crtc_load_gamma_888(crtc, lut, vesadrm_set_color_lut);
 		break;
 	default:
 		drm_warn_once(dev, "Unsupported format %p4cc for gamma correction\n",

From 1adb35c2523fcd06cf1c24d788d6a958a24bc143 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:08 +0200
Subject: [PATCH 0634/1455] drm/vesadrm: Prepare color management for
 palette-based framebuffers

Distiguish between component-based formats and 'the rest' in vesadrm's
color management. Scanout buffers with component-based format allow
for gamma correction. Palette-based formats (i.e., 'the rest') require
palette setup.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-9-tzimmermann@suse.de
---
 drivers/gpu/drm/sysfb/vesadrm.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
index 9596322c232b..b6e5ed1863f9 100644
--- a/drivers/gpu/drm/sysfb/vesadrm.c
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -223,15 +223,22 @@ static void vesadrm_crtc_helper_atomic_flush(struct drm_crtc *crtc,
 	 * plane's color format.
 	 */
 	if (crtc_state->enable && crtc_state->color_mgmt_changed) {
-		if (sysfb_crtc_state->format == sysfb->fb_format) {
-			if (crtc_state->gamma_lut)
-				vesadrm_load_gamma_lut(vesa,
-						       sysfb_crtc_state->format,
-						       crtc_state->gamma_lut->data);
-			else
+		switch (sysfb->fb_format->format) {
+		/*
+		 * Component formats
+		 */
+		default:
+			if (sysfb_crtc_state->format == sysfb->fb_format) {
+				if (crtc_state->gamma_lut)
+					vesadrm_load_gamma_lut(vesa,
+							       sysfb_crtc_state->format,
+							       crtc_state->gamma_lut->data);
+				else
+					vesadrm_fill_gamma_lut(vesa, sysfb_crtc_state->format);
+			} else {
 				vesadrm_fill_gamma_lut(vesa, sysfb_crtc_state->format);
-		} else {
-			vesadrm_fill_gamma_lut(vesa, sysfb_crtc_state->format);
+			}
+			break;
 		}
 	}
 }

From 7399c13f619f33dc8bdce838f3c83e88a18765ee Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Mon, 14 Jul 2025 17:13:09 +0200
Subject: [PATCH 0635/1455] drm/vesadrm: Support DRM_FORMAT_C8

Add support for DRM_FORMAT_C8 to vesadrm. The new pixel-format
description PIXEL_FORMAT_C8 describes the layout. Vesadrm's helpers
vesadrm_fill_palette_lut() and vesadrm_load_palette_lut() set the
hardware palette according to the CRTC's output format.

The driver emulates XRGB8888 by converting the source buffer to
RGB332 and using the resulting 256 colors as index into the hardware
palette. The hardware palette converts back to RGB during scanout.
This has no overhead compared to other format conversion, but allows
common userspace, such as Wayland compositors, to operate on the
display.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250714151513.309475-10-tzimmermann@suse.de
---
 drivers/gpu/drm/sysfb/vesadrm.c | 114 +++++++++++++++++++++++++++++++-
 include/video/pixel_format.h    |   3 +
 2 files changed, 116 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
index b6e5ed1863f9..1170e71e4208 100644
--- a/drivers/gpu/drm/sysfb/vesadrm.c
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -46,6 +46,7 @@ static const struct drm_format_info *vesadrm_get_format_si(struct drm_device *de
 		{ PIXEL_FORMAT_RGB888, DRM_FORMAT_RGB888, },
 		{ PIXEL_FORMAT_XRGB8888, DRM_FORMAT_XRGB8888, },
 		{ PIXEL_FORMAT_XBGR8888, DRM_FORMAT_XBGR8888, },
+		{ PIXEL_FORMAT_C8, DRM_FORMAT_C8, },
 	};
 
 	return drm_sysfb_get_format_si(dev, formats, ARRAY_SIZE(formats), si);
@@ -192,6 +193,44 @@ static void vesadrm_load_gamma_lut(struct vesadrm_device *vesa,
 	}
 }
 
+static void vesadrm_fill_palette_lut(struct vesadrm_device *vesa,
+				     const struct drm_format_info *format)
+{
+	struct drm_device *dev = &vesa->sysfb.dev;
+	struct drm_crtc *crtc = &vesa->crtc;
+
+	switch (format->format) {
+	case DRM_FORMAT_C8:
+		drm_crtc_fill_palette_8(crtc, vesadrm_set_color_lut);
+		break;
+	case DRM_FORMAT_RGB332:
+		drm_crtc_fill_palette_332(crtc, vesadrm_set_color_lut);
+		break;
+	default:
+		drm_warn_once(dev, "Unsupported format %p4cc for palette\n",
+			      &format->format);
+		break;
+	}
+}
+
+static void vesadrm_load_palette_lut(struct vesadrm_device *vesa,
+				     const struct drm_format_info *format,
+				     struct drm_color_lut *lut)
+{
+	struct drm_device *dev = &vesa->sysfb.dev;
+	struct drm_crtc *crtc = &vesa->crtc;
+
+	switch (format->format) {
+	case DRM_FORMAT_C8:
+		drm_crtc_load_palette_8(crtc, lut, vesadrm_set_color_lut);
+		break;
+	default:
+		drm_warn_once(dev, "Unsupported format %p4cc for gamma correction\n",
+			      &format->format);
+		break;
+	}
+}
+
 /*
  * Modesetting
  */
@@ -200,8 +239,67 @@ static const u64 vesadrm_primary_plane_format_modifiers[] = {
 	DRM_SYSFB_PLANE_FORMAT_MODIFIERS,
 };
 
+static int vesadrm_primary_plane_helper_atomic_check(struct drm_plane *plane,
+						     struct drm_atomic_state *new_state)
+{
+	struct drm_sysfb_device *sysfb = to_drm_sysfb_device(plane->dev);
+	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(new_state, plane);
+	struct drm_framebuffer *new_fb = new_plane_state->fb;
+	struct drm_crtc_state *new_crtc_state;
+	struct drm_sysfb_crtc_state *new_sysfb_crtc_state;
+	int ret;
+
+	ret = drm_sysfb_plane_helper_atomic_check(plane, new_state);
+	if (ret)
+		return ret;
+	else if (!new_plane_state->visible)
+		return 0;
+
+	/*
+	 * Fix up format conversion for specific cases
+	 */
+
+	switch (sysfb->fb_format->format) {
+	case DRM_FORMAT_C8:
+		new_crtc_state = drm_atomic_get_new_crtc_state(new_state, new_plane_state->crtc);
+		new_sysfb_crtc_state = to_drm_sysfb_crtc_state(new_crtc_state);
+
+		switch (new_fb->format->format) {
+		case DRM_FORMAT_XRGB8888:
+			/*
+			 * Reduce XRGB8888 to RGB332. Each resulting pixel is an index
+			 * into the C8 hardware palette, which stores RGB332 colors.
+			 */
+			if (new_sysfb_crtc_state->format->format != DRM_FORMAT_RGB332) {
+				new_sysfb_crtc_state->format =
+					drm_format_info(DRM_FORMAT_RGB332);
+				new_crtc_state->color_mgmt_changed = true;
+			}
+			break;
+		case DRM_FORMAT_C8:
+			/*
+			 * Restore original output. Emulation of XRGB8888 set RBG332
+			 * output format and hardware palette. This needs to be undone
+			 * when we switch back to DRM_FORMAT_C8.
+			 */
+			if (new_sysfb_crtc_state->format->format == DRM_FORMAT_RGB332) {
+				new_sysfb_crtc_state->format = sysfb->fb_format;
+				new_crtc_state->color_mgmt_changed = true;
+			}
+			break;
+		}
+		break;
+	};
+
+	return 0;
+}
+
 static const struct drm_plane_helper_funcs vesadrm_primary_plane_helper_funcs = {
-	DRM_SYSFB_PLANE_HELPER_FUNCS,
+	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
+	.atomic_check = vesadrm_primary_plane_helper_atomic_check,
+	.atomic_update = drm_sysfb_plane_helper_atomic_update,
+	.atomic_disable = drm_sysfb_plane_helper_atomic_disable,
+	.get_scanout_buffer = drm_sysfb_plane_helper_get_scanout_buffer,
 };
 
 static const struct drm_plane_funcs vesadrm_primary_plane_funcs = {
@@ -224,6 +322,20 @@ static void vesadrm_crtc_helper_atomic_flush(struct drm_crtc *crtc,
 	 */
 	if (crtc_state->enable && crtc_state->color_mgmt_changed) {
 		switch (sysfb->fb_format->format) {
+		/*
+		 * Index formats
+		 */
+		case DRM_FORMAT_C8:
+			if (sysfb_crtc_state->format->format == DRM_FORMAT_RGB332) {
+				vesadrm_fill_palette_lut(vesa, sysfb_crtc_state->format);
+			} else if (crtc->state->gamma_lut) {
+				vesadrm_load_palette_lut(vesa,
+							 sysfb_crtc_state->format,
+							 crtc_state->gamma_lut->data);
+			} else {
+				vesadrm_fill_palette_lut(vesa, sysfb_crtc_state->format);
+			}
+			break;
 		/*
 		 * Component formats
 		 */
diff --git a/include/video/pixel_format.h b/include/video/pixel_format.h
index c57019cd6ea8..6874754b0474 100644
--- a/include/video/pixel_format.h
+++ b/include/video/pixel_format.h
@@ -20,6 +20,9 @@ struct pixel_format {
 	};
 };
 
+#define PIXEL_FORMAT_C8 \
+	{ 8, true, { .index = {0, 8}, } }
+
 #define PIXEL_FORMAT_XRGB1555 \
 	{ 16, false, { .alpha = {0, 0}, .red = {10, 5}, .green = {5, 5}, .blue = {0, 5} } }
 

From 38fc73b8c7d692a099ddda37b700eeb330a03ff1 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:08 +0530
Subject: [PATCH 0636/1455] drm/xe: Add documentation for Xe Device Wedging

Add documentation for Xe Device Wedging so that
file can be referenced in following patches.

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-2-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 Documentation/gpu/xe/index.rst     |  1 +
 Documentation/gpu/xe/xe_device.rst | 10 ++++++++++
 drivers/gpu/drm/xe/xe_device.c     | 10 ++++++++++
 3 files changed, 21 insertions(+)
 create mode 100644 Documentation/gpu/xe/xe_device.rst

diff --git a/Documentation/gpu/xe/index.rst b/Documentation/gpu/xe/index.rst
index 42ba6c263cd0..88b22fad880e 100644
--- a/Documentation/gpu/xe/index.rst
+++ b/Documentation/gpu/xe/index.rst
@@ -25,5 +25,6 @@ DG2, etc is provided to prototype the driver.
    xe_tile
    xe_debugging
    xe_devcoredump
+   xe_device
    xe-drm-usage-stats.rst
    xe_configfs
diff --git a/Documentation/gpu/xe/xe_device.rst b/Documentation/gpu/xe/xe_device.rst
new file mode 100644
index 000000000000..39a937b97cd3
--- /dev/null
+++ b/Documentation/gpu/xe/xe_device.rst
@@ -0,0 +1,10 @@
+.. SPDX-License-Identifier: (GPL-2.0+ OR MIT)
+
+.. _xe-device-wedging:
+
+==================
+Xe Device Wedging
+==================
+
+.. kernel-doc:: drivers/gpu/drm/xe/xe_device.c
+   :doc: Xe Device Wedging
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 90617e627bc4..6f7f551bd263 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -1166,6 +1166,16 @@ static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
 	xe_pm_runtime_put(xe);
 }
 
+/**
+ * DOC: Xe Device Wedging
+ *
+ * Xe driver uses drm device wedged uevent as documented in Documentation/gpu/drm-uapi.rst.
+ * When device is in wedged state, every IOCTL will be blocked and GT cannot be
+ * used. Certain critical errors like gt reset failure, firmware failures can cause
+ * the device to be wedged. The default recovery method for a wedged state
+ * is rebind/bus-reset.
+ */
+
 /**
  * xe_device_declare_wedged - Declare device wedged
  * @xe: xe device instance

From 9c857a9d84e01332d031b55c4e38a66daecbae73 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:09 +0530
Subject: [PATCH 0637/1455] drm: Add a vendor-specific recovery method to drm
 device wedged uevent
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Address the need for a recovery method (firmware flash on Firmware errors)
introduced in the later patches of Xe KMD.
Whenever XE KMD detects a firmware error, a firmware flash is required to
recover the device to normal operation.

The initial proposal to use 'firmware-flash' as a recovery method was
not applicable to other drivers and could cause multiple recovery
methods specific to vendors to be added.
To address this a more generic 'vendor-specific' method is introduced,
guiding users to refer to vendor specific documentation and system logs
for detailed vendor specific recovery procedure.

Add a recovery method 'WEDGED=vendor-specific' for such errors.
Vendors must provide additional recovery documentation if this method
is used.

It is the responsibility of the consumer to refer to the correct vendor
specific documentation and usecase before attempting a recovery.

For example: If driver is XE KMD, the consumer must refer
to the documentation of 'Device Wedging' under 'Documentation/gpu/xe/'.

v2: fix documentation (Raag)
v3: add more details to commit message (Sima, Rodrigo, Raag)
    add an example script to the documentation (Raag)
v4: use consistent naming (Raag)
v5: fix commit message
v6: add more documentation

Cc: André Almeida <andrealmeid@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona.vetter@ffwll.ch>
Cc: Maxime Ripard <mripard@kernel.org>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250826063419.3022216-3-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 Documentation/gpu/drm-uapi.rst | 47 +++++++++++++++++++++++++++++-----
 drivers/gpu/drm/drm_drv.c      |  2 ++
 include/drm/drm_device.h       |  4 +++
 3 files changed, 46 insertions(+), 7 deletions(-)

diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst
index 843facf01b2d..d98428a592f1 100644
--- a/Documentation/gpu/drm-uapi.rst
+++ b/Documentation/gpu/drm-uapi.rst
@@ -418,13 +418,12 @@ needed.
 Recovery
 --------
 
-Current implementation defines three recovery methods, out of which, drivers
+Current implementation defines four recovery methods, out of which, drivers
 can use any one, multiple or none. Method(s) of choice will be sent in the
 uevent environment as ``WEDGED=<method1>[,..,<methodN>]`` in order of less to
-more side-effects. If driver is unsure about recovery or method is unknown
-(like soft/hard system reboot, firmware flashing, physical device replacement
-or any other procedure which can't be attempted on the fly), ``WEDGED=unknown``
-will be sent instead.
+more side-effects. See the section `Vendor Specific Recovery`_
+for ``WEDGED=vendor-specific``. If driver is unsure about recovery or
+method is unknown, ``WEDGED=unknown`` will be sent instead.
 
 Userspace consumers can parse this event and attempt recovery as per the
 following expectations.
@@ -435,6 +434,7 @@ following expectations.
     none            optional telemetry collection
     rebind          unbind + bind driver
     bus-reset       unbind + bus reset/re-enumeration + bind
+    vendor-specific vendor specific recovery method
     unknown         consumer policy
     =============== ========================================
 
@@ -446,6 +446,35 @@ telemetry information (devcoredump, syslog). This is useful because the first
 hang is usually the most critical one which can result in consequential hangs or
 complete wedging.
 
+
+Vendor Specific Recovery
+------------------------
+
+When ``WEDGED=vendor-specific`` is sent, it indicates that the device requires
+a recovery procedure specific to the hardware vendor and is not one of the
+standardized approaches.
+
+``WEDGED=vendor-specific`` may be used to indicate different cases within a
+single vendor driver, each requiring a distinct recovery procedure.
+In such scenarios, the vendor driver must provide comprehensive documentation
+that describes each case, include additional hints to identify specific case and
+outline the corresponding recovery procedure. The documentation includes:
+
+Case - A list of all cases that sends the ``WEDGED=vendor-specific`` recovery method.
+
+Hints - Additional Information to assist the userspace consumer in identifying and
+differentiating between different cases. This can be exposed through sysfs, debugfs,
+traces, dmesg etc.
+
+Recovery Procedure - Clear instructions and guidance for recovering each case.
+This may include userspace scripts, tools needed for the recovery procedure.
+
+It is the responsibility of the admin/userspace consumer to identify the case and
+verify additional identification hints before attempting a recovery procedure.
+
+Example: If the device uses the Xe driver, then userspace consumer should refer to
+:ref:`Xe Device Wedging <xe-device-wedging>` for the detailed documentation.
+
 Task information
 ----------------
 
@@ -472,8 +501,12 @@ erroring out, all device memory should be unmapped and file descriptors should
 be closed to prevent leaks or undefined behaviour. The idea here is to clear the
 device of all user context beforehand and set the stage for a clean recovery.
 
-Example
--------
+For ``WEDGED=vendor-specific`` recovery method, it is the responsibility of the
+consumer to check the driver documentation and the usecase before attempting
+a recovery.
+
+Example - rebind
+----------------
 
 Udev rule::
 
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index cdd591b11488..0ac723a46a91 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -532,6 +532,8 @@ static const char *drm_get_wedge_recovery(unsigned int opt)
 		return "rebind";
 	case DRM_WEDGE_RECOVERY_BUS_RESET:
 		return "bus-reset";
+	case DRM_WEDGE_RECOVERY_VENDOR:
+		return "vendor-specific";
 	default:
 		return NULL;
 	}
diff --git a/include/drm/drm_device.h b/include/drm/drm_device.h
index a33aedd5e9ec..59fd3f4d5995 100644
--- a/include/drm/drm_device.h
+++ b/include/drm/drm_device.h
@@ -26,10 +26,14 @@ struct pci_controller;
  * Recovery methods for wedged device in order of less to more side-effects.
  * To be used with drm_dev_wedged_event() as recovery @method. Callers can
  * use any one, multiple (or'd) or none depending on their needs.
+ *
+ * Refer to "Device Wedging" chapter in Documentation/gpu/drm-uapi.rst for more
+ * details.
  */
 #define DRM_WEDGE_RECOVERY_NONE		BIT(0)	/* optional telemetry collection */
 #define DRM_WEDGE_RECOVERY_REBIND	BIT(1)	/* unbind + bind driver */
 #define DRM_WEDGE_RECOVERY_BUS_RESET	BIT(2)	/* unbind + reset bus device + bind */
+#define DRM_WEDGE_RECOVERY_VENDOR	BIT(3)	/* vendor specific recovery method */
 
 /**
  * struct drm_wedge_task_info - information about the guilty task of a wedge dev

From 90fdcf5f89e9288c153923f16a60e6f7da18ba76 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:10 +0530
Subject: [PATCH 0638/1455] drm/xe: Set GT as wedged before sending wedged
 uevent

Userspace should be notified after setting the device as wedged.
Re-order function calls to set gt wedged before sending uevent.

Cc: Matthew Brost <matthew.brost@intel.com>
Suggested-by: Raag Jadav <raag.jadav@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-4-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 6f7f551bd263..bc8be6fbd2e8 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -1180,8 +1180,10 @@ static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
  * xe_device_declare_wedged - Declare device wedged
  * @xe: xe device instance
  *
- * This is a final state that can only be cleared with a module
+ * This is a final state that can only be cleared with the recovery method
+ * specified in the drm wedged uevent. The default recovery method is
  * re-probe (unbind + bind).
+ *
  * In this state every IOCTL will be blocked so the GT cannot be used.
  * In general it will be called upon any critical error such as gt reset
  * failure or guc loading failure. Userspace will be notified of this state
@@ -1215,13 +1217,15 @@ void xe_device_declare_wedged(struct xe_device *xe)
 			"IOCTLs and executions are blocked. Only a rebind may clear the failure\n"
 			"Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/xe/kernel/issues/new\n",
 			dev_name(xe->drm.dev));
+	}
 
+	for_each_gt(gt, xe, id)
+		xe_gt_declare_wedged(gt);
+
+	if (xe_device_wedged(xe)) {
 		/* Notify userspace of wedged device */
 		drm_dev_wedged_event(&xe->drm,
 				     DRM_WEDGE_RECOVERY_REBIND | DRM_WEDGE_RECOVERY_BUS_RESET,
 				     NULL);
 	}
-
-	for_each_gt(gt, xe, id)
-		xe_gt_declare_wedged(gt);
 }

From 60439ac3f23549ff272e626bd78502123086eaa5 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:11 +0530
Subject: [PATCH 0639/1455] drm/xe: Add a helper function to set recovery
 method

Add a helper function to set recovery method. The recovery
method can be set before declaring the device wedged and sending the
drm wedged uevent. If no method is set, default unbind/re-bind method
will be set.

v2: fix documentation (Raag)

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-5-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c       | 26 +++++++++++++++++++++-----
 drivers/gpu/drm/xe/xe_device.h       |  1 +
 drivers/gpu/drm/xe/xe_device_types.h |  2 ++
 3 files changed, 24 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index bc8be6fbd2e8..ab9f3a57dcd9 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -1176,13 +1176,26 @@ static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
  * is rebind/bus-reset.
  */
 
+/**
+ * xe_device_set_wedged_method - Set wedged recovery method
+ * @xe: xe device instance
+ * @method: recovery method to set
+ *
+ * Set wedged recovery method to be sent in drm wedged uevent.
+ */
+void xe_device_set_wedged_method(struct xe_device *xe, unsigned long method)
+{
+	xe->wedged.method = method;
+}
+
 /**
  * xe_device_declare_wedged - Declare device wedged
  * @xe: xe device instance
  *
  * This is a final state that can only be cleared with the recovery method
- * specified in the drm wedged uevent. The default recovery method is
- * re-probe (unbind + bind).
+ * specified in the drm wedged uevent. The method can be set using
+ * xe_device_set_wedged_method before declaring the device as wedged. If no method
+ * is set, reprobe (unbind/re-bind) will be sent by default.
  *
  * In this state every IOCTL will be blocked so the GT cannot be used.
  * In general it will be called upon any critical error such as gt reset
@@ -1223,9 +1236,12 @@ void xe_device_declare_wedged(struct xe_device *xe)
 		xe_gt_declare_wedged(gt);
 
 	if (xe_device_wedged(xe)) {
+		/* If no wedge recovery method is set, use default */
+		if (!xe->wedged.method)
+			xe_device_set_wedged_method(xe, DRM_WEDGE_RECOVERY_REBIND |
+						    DRM_WEDGE_RECOVERY_BUS_RESET);
+
 		/* Notify userspace of wedged device */
-		drm_dev_wedged_event(&xe->drm,
-				     DRM_WEDGE_RECOVERY_REBIND | DRM_WEDGE_RECOVERY_BUS_RESET,
-				     NULL);
+		drm_dev_wedged_event(&xe->drm, xe->wedged.method, NULL);
 	}
 }
diff --git a/drivers/gpu/drm/xe/xe_device.h b/drivers/gpu/drm/xe/xe_device.h
index bc802e066a7d..32cc6323b7f6 100644
--- a/drivers/gpu/drm/xe/xe_device.h
+++ b/drivers/gpu/drm/xe/xe_device.h
@@ -187,6 +187,7 @@ static inline bool xe_device_wedged(struct xe_device *xe)
 	return atomic_read(&xe->wedged.flag);
 }
 
+void xe_device_set_wedged_method(struct xe_device *xe, unsigned long method);
 void xe_device_declare_wedged(struct xe_device *xe);
 
 struct xe_file *xe_file_get(struct xe_file *xef);
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 86506cdfc801..333c7aa6bbb3 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -541,6 +541,8 @@ struct xe_device {
 		atomic_t flag;
 		/** @wedged.mode: Mode controlled by kernel parameter and debugfs */
 		int mode;
+		/** @wedged.method: Recovery method to be sent in the drm device wedged uevent */
+		unsigned long method;
 	} wedged;
 
 	/** @bo_device: Struct to control async free of BOs */

From 41ff795aff53613d4184a35bbfae11a15caf07c5 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:12 +0530
Subject: [PATCH 0640/1455] drm/xe/xe_survivability: Refactor survivability
 mode

Refactor survivability mode code to support both boot
and runtime survivability.

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-6-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c                |  2 +-
 drivers/gpu/drm/xe/xe_heci_gsc.c              |  2 +-
 drivers/gpu/drm/xe/xe_pci.c                   |  6 +-
 drivers/gpu/drm/xe/xe_survivability_mode.c    | 93 +++++++++++++------
 drivers/gpu/drm/xe/xe_survivability_mode.h    |  4 +-
 .../gpu/drm/xe/xe_survivability_mode_types.h  |  7 ++
 6 files changed, 81 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index ab9f3a57dcd9..d6f10ee6730d 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -743,7 +743,7 @@ int xe_device_probe_early(struct xe_device *xe)
 		 * possible, but still return the previous error for error
 		 * propagation
 		 */
-		err = xe_survivability_mode_enable(xe);
+		err = xe_survivability_mode_boot_enable(xe);
 		if (err)
 			return err;
 
diff --git a/drivers/gpu/drm/xe/xe_heci_gsc.c b/drivers/gpu/drm/xe/xe_heci_gsc.c
index 6d7b62724126..a415ca488791 100644
--- a/drivers/gpu/drm/xe/xe_heci_gsc.c
+++ b/drivers/gpu/drm/xe/xe_heci_gsc.c
@@ -197,7 +197,7 @@ int xe_heci_gsc_init(struct xe_device *xe)
 	if (ret)
 		return ret;
 
-	if (!def->use_polling && !xe_survivability_mode_is_enabled(xe)) {
+	if (!def->use_polling && !xe_survivability_mode_is_boot_enabled(xe)) {
 		ret = heci_gsc_irq_setup(xe);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index a8bab19111c3..15a863491cdd 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -784,7 +784,7 @@ static void xe_pci_remove(struct pci_dev *pdev)
 	if (IS_SRIOV_PF(xe))
 		xe_pci_sriov_configure(pdev, 0);
 
-	if (xe_survivability_mode_is_enabled(xe))
+	if (xe_survivability_mode_is_boot_enabled(xe))
 		return;
 
 	xe_device_remove(xe);
@@ -866,7 +866,7 @@ static int xe_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	 * flashed through mei. Return success, if survivability mode
 	 * is enabled due to pcode failure or configfs being set
 	 */
-	if (xe_survivability_mode_is_enabled(xe))
+	if (xe_survivability_mode_is_boot_enabled(xe))
 		return 0;
 
 	if (err)
@@ -960,7 +960,7 @@ static int xe_pci_suspend(struct device *dev)
 	struct xe_device *xe = pdev_to_xe_device(pdev);
 	int err;
 
-	if (xe_survivability_mode_is_enabled(xe))
+	if (xe_survivability_mode_is_boot_enabled(xe))
 		return -EBUSY;
 
 	err = xe_pm_suspend(xe);
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index 41705f5d52e3..4a2d1cff65d2 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -121,6 +121,14 @@ static void log_survivability_info(struct pci_dev *pdev)
 	}
 }
 
+static int check_boot_failure(struct xe_device *xe)
+{
+	struct xe_survivability *survivability = &xe->survivability;
+
+	return survivability->boot_status == NON_CRITICAL_FAILURE ||
+		survivability->boot_status == CRITICAL_FAILURE;
+}
+
 static ssize_t survivability_mode_show(struct device *dev,
 				       struct device_attribute *attr, char *buff)
 {
@@ -130,6 +138,11 @@ static ssize_t survivability_mode_show(struct device *dev,
 	struct xe_survivability_info *info = survivability->info;
 	int index = 0, count = 0;
 
+	count += sysfs_emit_at(buff, count, "Survivability mode type: Boot\n");
+
+	if (!check_boot_failure(xe))
+		return count;
+
 	for (index = 0; index < MAX_SCRATCH_MMIO; index++) {
 		if (info[index].reg)
 			count += sysfs_emit_at(buff, count, "%s: 0x%x - 0x%x\n", info[index].name,
@@ -151,12 +164,11 @@ static void xe_survivability_mode_fini(void *arg)
 	sysfs_remove_file(&dev->kobj, &dev_attr_survivability_mode.attr);
 }
 
-static int enable_survivability_mode(struct pci_dev *pdev)
+static int create_survivability_sysfs(struct pci_dev *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct xe_device *xe = pdev_to_xe_device(pdev);
-	struct xe_survivability *survivability = &xe->survivability;
-	int ret = 0;
+	int ret;
 
 	/* create survivability mode sysfs */
 	ret = sysfs_create_file(&dev->kobj, &dev_attr_survivability_mode.attr);
@@ -170,6 +182,20 @@ static int enable_survivability_mode(struct pci_dev *pdev)
 	if (ret)
 		return ret;
 
+	return 0;
+}
+
+static int enable_boot_survivability_mode(struct pci_dev *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct xe_device *xe = pdev_to_xe_device(pdev);
+	struct xe_survivability *survivability = &xe->survivability;
+	int ret = 0;
+
+	ret = create_survivability_sysfs(pdev);
+	if (ret)
+		return ret;
+
 	/* Make sure xe_heci_gsc_init() knows about survivability mode */
 	survivability->mode = true;
 
@@ -192,15 +218,36 @@ err:
 	return ret;
 }
 
+static int init_survivability_mode(struct xe_device *xe)
+{
+	struct xe_survivability *survivability = &xe->survivability;
+	struct xe_survivability_info *info;
+
+	survivability->size = MAX_SCRATCH_MMIO;
+
+	info = devm_kcalloc(xe->drm.dev, survivability->size, sizeof(*info),
+			    GFP_KERNEL);
+	if (!info)
+		return -ENOMEM;
+
+	survivability->info = info;
+
+	populate_survivability_info(xe);
+
+	return 0;
+}
+
 /**
- * xe_survivability_mode_is_enabled - check if survivability mode is enabled
+ * xe_survivability_mode_is_boot_enabled- check if boot survivability mode is enabled
  * @xe: xe device instance
  *
- * Returns true if in survivability mode, false otherwise
+ * Returns true if in boot survivability mode of type, else false
  */
-bool xe_survivability_mode_is_enabled(struct xe_device *xe)
+bool xe_survivability_mode_is_boot_enabled(struct xe_device *xe)
 {
-	return xe->survivability.mode;
+	struct xe_survivability *survivability = &xe->survivability;
+
+	return survivability->mode && survivability->type == XE_SURVIVABILITY_TYPE_BOOT;
 }
 
 /**
@@ -241,44 +288,38 @@ bool xe_survivability_mode_is_requested(struct xe_device *xe)
 	data = xe_mmio_read32(mmio, PCODE_SCRATCH(0));
 	survivability->boot_status = REG_FIELD_GET(BOOT_STATUS, data);
 
-	return survivability->boot_status == NON_CRITICAL_FAILURE ||
-		survivability->boot_status == CRITICAL_FAILURE;
+	return check_boot_failure(xe);
 }
 
 /**
- * xe_survivability_mode_enable - Initialize and enable the survivability mode
+ * xe_survivability_mode_boot_enable - Initialize and enable boot survivability mode
  * @xe: xe device instance
  *
- * Initialize survivability information and enable survivability mode
+ * Initialize survivability information and enable boot survivability mode
  *
- * Return: 0 if survivability mode is enabled or not requested; negative error
+ * Return: 0 if boot survivability mode is enabled or not requested, negative error
  * code otherwise.
  */
-int xe_survivability_mode_enable(struct xe_device *xe)
+int xe_survivability_mode_boot_enable(struct xe_device *xe)
 {
 	struct xe_survivability *survivability = &xe->survivability;
-	struct xe_survivability_info *info;
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
+	int ret;
 
 	if (!xe_survivability_mode_is_requested(xe))
 		return 0;
 
-	survivability->size = MAX_SCRATCH_MMIO;
+	ret = init_survivability_mode(xe);
+	if (ret)
+		return ret;
 
-	info = devm_kcalloc(xe->drm.dev, survivability->size, sizeof(*info),
-			    GFP_KERNEL);
-	if (!info)
-		return -ENOMEM;
-
-	survivability->info = info;
-
-	populate_survivability_info(xe);
-
-	/* Only log debug information and exit if it is a critical failure */
+	/* Log breadcrumbs but do not enter survivability mode for Critical boot errors */
 	if (survivability->boot_status == CRITICAL_FAILURE) {
 		log_survivability_info(pdev);
 		return -ENXIO;
 	}
 
-	return enable_survivability_mode(pdev);
+	survivability->type = XE_SURVIVABILITY_TYPE_BOOT;
+
+	return enable_boot_survivability_mode(pdev);
 }
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.h b/drivers/gpu/drm/xe/xe_survivability_mode.h
index 02231c2bf008..f6ee283ea5e8 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.h
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.h
@@ -10,8 +10,8 @@
 
 struct xe_device;
 
-int xe_survivability_mode_enable(struct xe_device *xe);
-bool xe_survivability_mode_is_enabled(struct xe_device *xe);
+int xe_survivability_mode_boot_enable(struct xe_device *xe);
+bool xe_survivability_mode_is_boot_enabled(struct xe_device *xe);
 bool xe_survivability_mode_is_requested(struct xe_device *xe);
 
 #endif /* _XE_SURVIVABILITY_MODE_H_ */
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode_types.h b/drivers/gpu/drm/xe/xe_survivability_mode_types.h
index 19d433e253df..5dce393498da 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode_types.h
+++ b/drivers/gpu/drm/xe/xe_survivability_mode_types.h
@@ -9,6 +9,10 @@
 #include <linux/limits.h>
 #include <linux/types.h>
 
+enum xe_survivability_type {
+	XE_SURVIVABILITY_TYPE_BOOT,
+};
+
 struct xe_survivability_info {
 	char name[NAME_MAX];
 	u32 reg;
@@ -30,6 +34,9 @@ struct xe_survivability {
 
 	/** @mode: boolean to indicate survivability mode */
 	bool mode;
+
+	/** @type: survivability type */
+	enum xe_survivability_type type;
 };
 
 #endif /* _XE_SURVIVABILITY_MODE_TYPES_H_ */

From a2ca0633a0fef925a0d8125d8f3e4495a5ecb310 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:13 +0530
Subject: [PATCH 0641/1455] drm/xe/xe_survivability: Add support for Runtime
 survivability mode

Certain runtime firmware errors can cause the device to be in a unusable
state requiring a firmware flash to restore normal operation.
Runtime Survivability Mode indicates firmware flash is necessary by
wedging the device and exposing survivability mode sysfs.

The below sysfs is an indication that device is in survivability mode

/sys/bus/pci/devices/<device>/survivability_mode

v2: Fix kernel-doc (Umesh)
v3: Add user friendly dmesg (Frank)

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-7-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_survivability_mode.c    | 43 ++++++++++++++++++-
 drivers/gpu/drm/xe/xe_survivability_mode.h    |  1 +
 .../gpu/drm/xe/xe_survivability_mode_types.h  |  1 +
 3 files changed, 44 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index 4a2d1cff65d2..267d0e3fd85a 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -138,7 +138,8 @@ static ssize_t survivability_mode_show(struct device *dev,
 	struct xe_survivability_info *info = survivability->info;
 	int index = 0, count = 0;
 
-	count += sysfs_emit_at(buff, count, "Survivability mode type: Boot\n");
+	count += sysfs_emit_at(buff, count, "Survivability mode type: %s\n",
+			       survivability->type ? "Runtime" : "Boot");
 
 	if (!check_boot_failure(xe))
 		return count;
@@ -291,6 +292,46 @@ bool xe_survivability_mode_is_requested(struct xe_device *xe)
 	return check_boot_failure(xe);
 }
 
+/**
+ * xe_survivability_mode_runtime_enable - Initialize and enable runtime survivability mode
+ * @xe: xe device instance
+ *
+ * Initialize survivability information and enable runtime survivability mode.
+ * Runtime survivability mode is enabled when certain errors cause the device to be
+ * in non-recoverable state. The device is declared wedged with the appropriate
+ * recovery method and survivability mode sysfs exposed to userspace
+ *
+ * Return: 0 if runtime survivability mode is enabled, negative error code otherwise.
+ */
+int xe_survivability_mode_runtime_enable(struct xe_device *xe)
+{
+	struct xe_survivability *survivability = &xe->survivability;
+	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
+	int ret;
+
+	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe) || xe->info.platform < XE_BATTLEMAGE) {
+		dev_err(&pdev->dev, "Runtime Survivability Mode not supported\n");
+		return -EINVAL;
+	}
+
+	ret = init_survivability_mode(xe);
+	if (ret)
+		return ret;
+
+	ret = create_survivability_sysfs(pdev);
+	if (ret)
+		dev_err(&pdev->dev, "Failed to create survivability mode sysfs\n");
+
+	survivability->type = XE_SURVIVABILITY_TYPE_RUNTIME;
+	dev_err(&pdev->dev, "Runtime Survivability mode enabled\n");
+
+	xe_device_set_wedged_method(xe, DRM_WEDGE_RECOVERY_VENDOR);
+	xe_device_declare_wedged(xe);
+	dev_err(&pdev->dev, "Firmware update required, Refer the userspace documentation for more details!\n");
+
+	return 0;
+}
+
 /**
  * xe_survivability_mode_boot_enable - Initialize and enable boot survivability mode
  * @xe: xe device instance
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.h b/drivers/gpu/drm/xe/xe_survivability_mode.h
index f6ee283ea5e8..1cc94226aa82 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.h
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.h
@@ -11,6 +11,7 @@
 struct xe_device;
 
 int xe_survivability_mode_boot_enable(struct xe_device *xe);
+int xe_survivability_mode_runtime_enable(struct xe_device *xe);
 bool xe_survivability_mode_is_boot_enabled(struct xe_device *xe);
 bool xe_survivability_mode_is_requested(struct xe_device *xe);
 
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode_types.h b/drivers/gpu/drm/xe/xe_survivability_mode_types.h
index 5dce393498da..cd65a5d167c9 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode_types.h
+++ b/drivers/gpu/drm/xe/xe_survivability_mode_types.h
@@ -11,6 +11,7 @@
 
 enum xe_survivability_type {
 	XE_SURVIVABILITY_TYPE_BOOT,
+	XE_SURVIVABILITY_TYPE_RUNTIME,
 };
 
 struct xe_survivability_info {

From f646c9f9371b28b8f93e619fe003415f6aaeb416 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:14 +0530
Subject: [PATCH 0642/1455] drm/xe/doc: Document device wedged and runtime
 survivability

Add documentation for vendor specific device wedged recovery method
and runtime survivability.

v2: fix documentation (Raag)
v3: add userspace tool for firmware update (Raag)
v4: use consistent documentation (Raag)
v5: add more documentation

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-8-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 Documentation/gpu/xe/xe_pcode.rst          |  6 ++--
 drivers/gpu/drm/xe/xe_device.c             | 27 +++++++++++++++++
 drivers/gpu/drm/xe/xe_survivability_mode.c | 35 +++++++++++++++++-----
 3 files changed, 58 insertions(+), 10 deletions(-)

diff --git a/Documentation/gpu/xe/xe_pcode.rst b/Documentation/gpu/xe/xe_pcode.rst
index 5937ef3599b0..2a43601123cb 100644
--- a/Documentation/gpu/xe/xe_pcode.rst
+++ b/Documentation/gpu/xe/xe_pcode.rst
@@ -13,9 +13,11 @@ Internal API
 .. kernel-doc:: drivers/gpu/drm/xe/xe_pcode.c
    :internal:
 
+.. _xe-survivability-mode:
+
 ==================
-Boot Survivability
+Survivability Mode
 ==================
 
 .. kernel-doc:: drivers/gpu/drm/xe/xe_survivability_mode.c
-   :doc: Xe Boot Survivability
+   :doc: Survivability Mode
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index d6f10ee6730d..9e4773a17ef8 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -1174,6 +1174,33 @@ static void xe_device_wedged_fini(struct drm_device *drm, void *arg)
  * used. Certain critical errors like gt reset failure, firmware failures can cause
  * the device to be wedged. The default recovery method for a wedged state
  * is rebind/bus-reset.
+ *
+ * Another recovery method is vendor-specific. Below are the cases that send
+ * ``WEDGED=vendor-specific`` recovery method in drm device wedged uevent.
+ *
+ * Case: Firmware Flash
+ * --------------------
+ *
+ * Identification Hint
+ * +++++++++++++++++++
+ *
+ * ``WEDGED=vendor-specific`` drm device wedged uevent with
+ * :ref:`Runtime Survivability mode <xe-survivability-mode>` is used to notify
+ * admin/userspace consumer about the need for a firmware flash.
+ *
+ * Recovery Procedure
+ * ++++++++++++++++++
+ *
+ * Once ``WEDGED=vendor-specific`` drm device wedged uevent is received, follow
+ * the below steps
+ *
+ * - Check Runtime Survivability mode sysfs.
+ *   If enabled, firmware flash is required to recover the device.
+ *
+ *   /sys/bus/pci/devices/<device>/survivability_mode
+ *
+ * - Admin/userpsace consumer can use firmware flashing tools like fwupd to flash
+ *   firmware and restore device to normal operation.
  */
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index 267d0e3fd85a..86ba767c4e44 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -22,15 +22,18 @@
 #define MAX_SCRATCH_MMIO 8
 
 /**
- * DOC: Xe Boot Survivability
+ * DOC: Survivability Mode
  *
- * Boot Survivability is a software based workflow for recovering a system in a failed boot state
+ * Survivability Mode is a software based workflow for recovering a system in a failed boot state
  * Here system recoverability is concerned with recovering the firmware responsible for boot.
  *
- * This is implemented by loading the driver with bare minimum (no drm card) to allow the firmware
- * to be flashed through mei and collect telemetry. The driver's probe flow is modified
- * such that it enters survivability mode when pcode initialization is incomplete and boot status
- * denotes a failure.
+ * Boot Survivability
+ * ===================
+ *
+ * Boot Survivability is implemented by loading the driver with bare minimum (no drm card) to allow
+ * the firmware to be flashed through mei driver and collect telemetry. The driver's probe flow is
+ * modified such that it enters survivability mode when pcode initialization is incomplete and boot
+ * status denotes a failure.
  *
  * Survivability mode can also be entered manually using the survivability mode attribute available
  * through configfs which is beneficial in several usecases. It can be used to address scenarios
@@ -46,7 +49,7 @@
  * Survivability mode is indicated by the below admin-only readable sysfs which provides additional
  * debug information::
  *
- *	/sys/bus/pci/devices/<device>/surivability_mode
+ *	/sys/bus/pci/devices/<device>/survivability_mode
  *
  * Capability Information:
  *	Provides boot status
@@ -56,6 +59,22 @@
  *	Provides history of previous failures
  * Auxiliary Information
  *	Certain failures may have information in addition to postcode information
+ *
+ * Runtime Survivability
+ * =====================
+ *
+ * Certain runtime firmware errors can cause the device to enter a wedged state
+ * (:ref:`xe-device-wedging`) requiring a firmware flash to restore normal operation.
+ * Runtime Survivability Mode indicates that a firmware flash is necessary to recover the device and
+ * is indicated by the presence of survivability mode sysfs::
+ *
+ *	/sys/bus/pci/devices/<device>/survivability_mode
+ *
+ * Survivability mode sysfs provides information about the type of survivability mode.
+ *
+ * When such errors occur, userspace is notified with the drm device wedged uevent and runtime
+ * survivability mode. User can then initiate a firmware flash using userspace tools like fwupd
+ * to restore device to normal operation.
  */
 
 static u32 aux_history_offset(u32 reg_value)
@@ -327,7 +346,7 @@ int xe_survivability_mode_runtime_enable(struct xe_device *xe)
 
 	xe_device_set_wedged_method(xe, DRM_WEDGE_RECOVERY_VENDOR);
 	xe_device_declare_wedged(xe);
-	dev_err(&pdev->dev, "Firmware update required, Refer the userspace documentation for more details!\n");
+	dev_err(&pdev->dev, "Firmware flash required, Refer the userspace documentation for more details!\n");
 
 	return 0;
 }

From 0a2a873d615a39e8a87d3f15285ed888341ddce8 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:15 +0530
Subject: [PATCH 0643/1455] drm/xe: Add support to handle hardware errors

Gfx device reports two classes of errors: uncorrectable and
correctable. Depending on the severity uncorrectable errors are further
classified Non-Fatal and Fatal.

Correctable and Non-Fatal errors: These errors are reported as MSI. Bits in
the Master Interrupt Register indicate the class of the error.
The source of the error is then read from the Device Error Source
Register.

Fatal errors: These are reported as PCIe errors
When a PCIe error is asserted, the OS will perform a SBR (Secondary
Bus reset) which causes the driver to reload. The error registers are
sticky and the values are maintained through SBR.

Add basic support to handle these errors.

Bspec: 50875, 53073, 53074, 53075, 53076

v2: Format commit message (Umesh)
v3: fix documentation (Stuart)

Cc: Stuart Summers <stuart.summers@intel.com>
Co-developed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-9-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/Makefile                |   1 +
 drivers/gpu/drm/xe/regs/xe_hw_error_regs.h |  15 +++
 drivers/gpu/drm/xe/regs/xe_irq_regs.h      |   1 +
 drivers/gpu/drm/xe/xe_hw_error.c           | 109 +++++++++++++++++++++
 drivers/gpu/drm/xe/xe_hw_error.h           |  15 +++
 drivers/gpu/drm/xe/xe_irq.c                |   4 +
 6 files changed, 145 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
 create mode 100644 drivers/gpu/drm/xe/xe_hw_error.c
 create mode 100644 drivers/gpu/drm/xe/xe_hw_error.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index c52cac979e05..0abd8078cfb0 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -82,6 +82,7 @@ xe-y += xe_bb.o \
 	xe_hw_engine.o \
 	xe_hw_engine_class_sysfs.o \
 	xe_hw_engine_group.o \
+	xe_hw_error.o \
 	xe_hw_fence.o \
 	xe_irq.o \
 	xe_lrc.o \
diff --git a/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
new file mode 100644
index 000000000000..ed9b81fb28a0
--- /dev/null
+++ b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_HW_ERROR_REGS_H_
+#define _XE_HW_ERROR_REGS_H_
+
+#define DEV_ERR_STAT_NONFATAL			0x100178
+#define DEV_ERR_STAT_CORRECTABLE		0x10017c
+#define DEV_ERR_STAT_REG(x)			XE_REG(_PICK_EVEN((x), \
+								  DEV_ERR_STAT_CORRECTABLE, \
+								  DEV_ERR_STAT_NONFATAL))
+
+#endif
diff --git a/drivers/gpu/drm/xe/regs/xe_irq_regs.h b/drivers/gpu/drm/xe/regs/xe_irq_regs.h
index 13635e4331d4..7c2a3a140142 100644
--- a/drivers/gpu/drm/xe/regs/xe_irq_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_irq_regs.h
@@ -18,6 +18,7 @@
 #define GFX_MSTR_IRQ				XE_REG(0x190010, XE_REG_OPTION_VF)
 #define   MASTER_IRQ				REG_BIT(31)
 #define   GU_MISC_IRQ				REG_BIT(29)
+#define   ERROR_IRQ(x)				REG_BIT(26 + (x))
 #define   DISPLAY_IRQ				REG_BIT(16)
 #define   I2C_IRQ				REG_BIT(12)
 #define   GT_DW_IRQ(x)				REG_BIT(x)
diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c
new file mode 100644
index 000000000000..0fa45792f1c0
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_hw_error.c
@@ -0,0 +1,109 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "regs/xe_hw_error_regs.h"
+#include "regs/xe_irq_regs.h"
+
+#include "xe_device.h"
+#include "xe_hw_error.h"
+#include "xe_mmio.h"
+
+/* Error categories reported by hardware */
+enum hardware_error {
+	HARDWARE_ERROR_CORRECTABLE = 0,
+	HARDWARE_ERROR_NONFATAL = 1,
+	HARDWARE_ERROR_FATAL = 2,
+	HARDWARE_ERROR_MAX,
+};
+
+static const char *hw_error_to_str(const enum hardware_error hw_err)
+{
+	switch (hw_err) {
+	case HARDWARE_ERROR_CORRECTABLE:
+		return "CORRECTABLE";
+	case HARDWARE_ERROR_NONFATAL:
+		return "NONFATAL";
+	case HARDWARE_ERROR_FATAL:
+		return "FATAL";
+	default:
+		return "UNKNOWN";
+	}
+}
+
+static void hw_error_source_handler(struct xe_tile *tile, const enum hardware_error hw_err)
+{
+	const char *hw_err_str = hw_error_to_str(hw_err);
+	struct xe_device *xe = tile_to_xe(tile);
+	unsigned long flags;
+	u32 err_src;
+
+	if (xe->info.platform != XE_BATTLEMAGE)
+		return;
+
+	spin_lock_irqsave(&xe->irq.lock, flags);
+	err_src = xe_mmio_read32(&tile->mmio, DEV_ERR_STAT_REG(hw_err));
+	if (!err_src) {
+		drm_err_ratelimited(&xe->drm, HW_ERR "Tile%d reported DEV_ERR_STAT_%s blank!\n",
+				    tile->id, hw_err_str);
+		goto unlock;
+	}
+
+	/* TODO: Process errrors per source */
+
+	xe_mmio_write32(&tile->mmio, DEV_ERR_STAT_REG(hw_err), err_src);
+
+unlock:
+	spin_unlock_irqrestore(&xe->irq.lock, flags);
+}
+
+/**
+ * xe_hw_error_irq_handler - irq handling for hw errors
+ * @tile: tile instance
+ * @master_ctl: value read from master interrupt register
+ *
+ * Xe platforms add three error bits to the master interrupt register to support error handling.
+ * These three bits are used to convey the class of error FATAL, NONFATAL, or CORRECTABLE.
+ * To process the interrupt, determine the source of error by reading the Device Error Source
+ * Register that corresponds to the class of error being serviced.
+ */
+void xe_hw_error_irq_handler(struct xe_tile *tile, const u32 master_ctl)
+{
+	enum hardware_error hw_err;
+
+	for (hw_err = 0; hw_err < HARDWARE_ERROR_MAX; hw_err++)
+		if (master_ctl & ERROR_IRQ(hw_err))
+			hw_error_source_handler(tile, hw_err);
+}
+
+/*
+ * Process hardware errors during boot
+ */
+static void process_hw_errors(struct xe_device *xe)
+{
+	struct xe_tile *tile;
+	u32 master_ctl;
+	u8 id;
+
+	for_each_tile(tile, xe, id) {
+		master_ctl = xe_mmio_read32(&tile->mmio, GFX_MSTR_IRQ);
+		xe_hw_error_irq_handler(tile, master_ctl);
+		xe_mmio_write32(&tile->mmio, GFX_MSTR_IRQ, master_ctl);
+	}
+}
+
+/**
+ * xe_hw_error_init - Initialize hw errors
+ * @xe: xe device instance
+ *
+ * Initialize and check for errors that occurred during boot
+ * prior to driver load
+ */
+void xe_hw_error_init(struct xe_device *xe)
+{
+	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
+		return;
+
+	process_hw_errors(xe);
+}
diff --git a/drivers/gpu/drm/xe/xe_hw_error.h b/drivers/gpu/drm/xe/xe_hw_error.h
new file mode 100644
index 000000000000..d86e28c5180c
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_hw_error.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+#ifndef XE_HW_ERROR_H_
+#define XE_HW_ERROR_H_
+
+#include <linux/types.h>
+
+struct xe_tile;
+struct xe_device;
+
+void xe_hw_error_irq_handler(struct xe_tile *tile, const u32 master_ctl);
+void xe_hw_error_init(struct xe_device *xe);
+#endif
diff --git a/drivers/gpu/drm/xe/xe_irq.c b/drivers/gpu/drm/xe/xe_irq.c
index 5df5b8c2a3e4..870edaf69388 100644
--- a/drivers/gpu/drm/xe/xe_irq.c
+++ b/drivers/gpu/drm/xe/xe_irq.c
@@ -18,6 +18,7 @@
 #include "xe_gt.h"
 #include "xe_guc.h"
 #include "xe_hw_engine.h"
+#include "xe_hw_error.h"
 #include "xe_i2c.h"
 #include "xe_memirq.h"
 #include "xe_mmio.h"
@@ -468,6 +469,7 @@ static irqreturn_t dg1_irq_handler(int irq, void *arg)
 		xe_mmio_write32(mmio, GFX_MSTR_IRQ, master_ctl);
 
 		gt_irq_handler(tile, master_ctl, intr_dw, identity);
+		xe_hw_error_irq_handler(tile, master_ctl);
 
 		/*
 		 * Display interrupts (including display backlight operations
@@ -756,6 +758,8 @@ int xe_irq_install(struct xe_device *xe)
 	int nvec = 1;
 	int err;
 
+	xe_hw_error_init(xe);
+
 	xe_irq_reset(xe);
 
 	if (xe_device_has_msix(xe)) {

From a7df563b45b03ac8aa480051cce8d49d7b489ec6 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:16 +0530
Subject: [PATCH 0644/1455] drm/xe/xe_hw_error: Handle CSC Firmware reported
 Hardware errors

Add support to handle CSC firmware reported errors. When CSC firmware
errors are encoutered, a error interrupt is received by the GFX device as
a MSI interrupt.

Device Source control registers indicates the source of the error as CSC
The HEC error status register indicates that the error is firmware reported
Depending on the type of error, the error cause is written to the HEC
Firmware error register.

On encountering such CSC firmware errors, the graphics device is
non-recoverable from driver context. The only way to recover from these
errors is firmware flash.

System admin/userspace is notified of the necessity of firmware flash
with a combination of vendor-specific drm device edged uevent, dmesg logs
and runtime survivability sysfs. It is the responsiblity of the consumer
to verify all the actions and then trigger a firmware flash using tools
like fwupd.

$ udevadm monitor --property --kernel
monitor will print the received events for:
KERNEL - the kernel uevent

KERNEL[754.709341] change   /devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:01.0/0000:03:00.0/drm/card0 (drm)
ACTION=change
DEVPATH=/devices/pci0000:00/0000:00:01.0/0000:01:00.0/0000:02:01.0/0000:03:00.0/drm/card0
SUBSYSTEM=drm
WEDGED=vendor-specific
DEVNAME=/dev/dri/card0
DEVTYPE=drm_minor
SEQNUM=5973
MAJOR=226
MINOR=0

Logs

xe 0000:03:00.0: [drm] *ERROR* [Hardware Error]: Tile0 reported NONFATAL error 0x20000
xe 0000:03:00.0: [drm] *ERROR* [Hardware Error]: NONFATAL: HEC Uncorrected FW FD Corruption error reported, bit[2] is set
xe 0000:03:00.0: Runtime Survivability mode enabled
xe 0000:03:00.0: [drm] *ERROR* CRITICAL: Xe has declared device 0000:03:00.0 as wedged.
               IOCTLs and executions are blocked. Only a rebind may clear the failure
               Please file a _new_ bug report at https://gitlab.freedesktop.org/drm/xe/kernel/issues/new
xe 0000:03:00.0: [drm] device wedged, needs recovery
xe 0000:03:00.0: Firmware flash required, Please refer to the userspace documentation for more details!

Runtime survivability Sysfs:

/sys/bus/pci/devices/<device>/survivability_mode

v2: use vendor recovery method with
    runtime survivability (Christian, Rodrigo, Raag)
v3: move declare wedged to runtime survivability mode (Rodrigo)
v4: update commit message

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-10-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_gsc_regs.h      |  2 +
 drivers/gpu/drm/xe/regs/xe_hw_error_regs.h |  7 ++-
 drivers/gpu/drm/xe/xe_device_types.h       |  3 +
 drivers/gpu/drm/xe/xe_hw_error.c           | 64 +++++++++++++++++++++-
 drivers/gpu/drm/xe/xe_survivability_mode.c |  2 +-
 5 files changed, 75 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/regs/xe_gsc_regs.h b/drivers/gpu/drm/xe/regs/xe_gsc_regs.h
index 9b66cc972a63..180be82672ab 100644
--- a/drivers/gpu/drm/xe/regs/xe_gsc_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_gsc_regs.h
@@ -13,6 +13,8 @@
 
 /* Definitions of GSC H/W registers, bits, etc */
 
+#define BMG_GSC_HECI1_BASE	0x373000
+
 #define MTL_GSC_HECI1_BASE	0x00116000
 #define MTL_GSC_HECI2_BASE	0x00117000
 
diff --git a/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
index ed9b81fb28a0..c146b9ef44eb 100644
--- a/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_hw_error_regs.h
@@ -6,10 +6,15 @@
 #ifndef _XE_HW_ERROR_REGS_H_
 #define _XE_HW_ERROR_REGS_H_
 
+#define HEC_UNCORR_ERR_STATUS(base)                    XE_REG((base) + 0x118)
+#define    UNCORR_FW_REPORTED_ERR                      BIT(6)
+
+#define HEC_UNCORR_FW_ERR_DW0(base)                    XE_REG((base) + 0x124)
+
 #define DEV_ERR_STAT_NONFATAL			0x100178
 #define DEV_ERR_STAT_CORRECTABLE		0x10017c
 #define DEV_ERR_STAT_REG(x)			XE_REG(_PICK_EVEN((x), \
 								  DEV_ERR_STAT_CORRECTABLE, \
 								  DEV_ERR_STAT_NONFATAL))
-
+#define   XE_CSC_ERROR				BIT(17)
 #endif
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 333c7aa6bbb3..389996970564 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -192,6 +192,9 @@ struct xe_tile {
 	/** @memirq: Memory Based Interrupts. */
 	struct xe_memirq memirq;
 
+	/** @csc_hw_error_work: worker to report CSC HW errors */
+	struct work_struct csc_hw_error_work;
+
 	/** @pcode: tile's PCODE */
 	struct {
 		/** @pcode.lock: protecting tile's PCODE mailbox data */
diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c
index 0fa45792f1c0..4099ed1262e1 100644
--- a/drivers/gpu/drm/xe/xe_hw_error.c
+++ b/drivers/gpu/drm/xe/xe_hw_error.c
@@ -3,12 +3,16 @@
  * Copyright © 2025 Intel Corporation
  */
 
+#include "regs/xe_gsc_regs.h"
 #include "regs/xe_hw_error_regs.h"
 #include "regs/xe_irq_regs.h"
 
 #include "xe_device.h"
 #include "xe_hw_error.h"
 #include "xe_mmio.h"
+#include "xe_survivability_mode.h"
+
+#define  HEC_UNCORR_FW_ERR_BITS 4
 
 /* Error categories reported by hardware */
 enum hardware_error {
@@ -18,6 +22,13 @@ enum hardware_error {
 	HARDWARE_ERROR_MAX,
 };
 
+static const char * const hec_uncorrected_fw_errors[] = {
+	"Fatal",
+	"CSE Disabled",
+	"FD Corruption",
+	"Data Corruption"
+};
+
 static const char *hw_error_to_str(const enum hardware_error hw_err)
 {
 	switch (hw_err) {
@@ -32,6 +43,52 @@ static const char *hw_error_to_str(const enum hardware_error hw_err)
 	}
 }
 
+static void csc_hw_error_work(struct work_struct *work)
+{
+	struct xe_tile *tile = container_of(work, typeof(*tile), csc_hw_error_work);
+	struct xe_device *xe = tile_to_xe(tile);
+	int ret;
+
+	ret = xe_survivability_mode_runtime_enable(xe);
+	if (ret)
+		drm_err(&xe->drm, "Failed to enable runtime survivability mode\n");
+}
+
+static void csc_hw_error_handler(struct xe_tile *tile, const enum hardware_error hw_err)
+{
+	const char *hw_err_str = hw_error_to_str(hw_err);
+	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_mmio *mmio = &tile->mmio;
+	u32 base, err_bit, err_src;
+	unsigned long fw_err;
+
+	if (xe->info.platform != XE_BATTLEMAGE)
+		return;
+
+	base = BMG_GSC_HECI1_BASE;
+	lockdep_assert_held(&xe->irq.lock);
+	err_src = xe_mmio_read32(mmio, HEC_UNCORR_ERR_STATUS(base));
+	if (!err_src) {
+		drm_err_ratelimited(&xe->drm, HW_ERR "Tile%d reported HEC_ERR_STATUS_%s blank\n",
+				    tile->id, hw_err_str);
+		return;
+	}
+
+	if (err_src & UNCORR_FW_REPORTED_ERR) {
+		fw_err = xe_mmio_read32(mmio, HEC_UNCORR_FW_ERR_DW0(base));
+		for_each_set_bit(err_bit, &fw_err, HEC_UNCORR_FW_ERR_BITS) {
+			drm_err_ratelimited(&xe->drm, HW_ERR
+					    "%s: HEC Uncorrected FW %s error reported, bit[%d] is set\n",
+					     hw_err_str, hec_uncorrected_fw_errors[err_bit],
+					     err_bit);
+
+			schedule_work(&tile->csc_hw_error_work);
+		}
+	}
+
+	xe_mmio_write32(mmio, HEC_UNCORR_ERR_STATUS(base), err_src);
+}
+
 static void hw_error_source_handler(struct xe_tile *tile, const enum hardware_error hw_err)
 {
 	const char *hw_err_str = hw_error_to_str(hw_err);
@@ -50,7 +107,8 @@ static void hw_error_source_handler(struct xe_tile *tile, const enum hardware_er
 		goto unlock;
 	}
 
-	/* TODO: Process errrors per source */
+	if (err_src & XE_CSC_ERROR)
+		csc_hw_error_handler(tile, hw_err);
 
 	xe_mmio_write32(&tile->mmio, DEV_ERR_STAT_REG(hw_err), err_src);
 
@@ -102,8 +160,12 @@ static void process_hw_errors(struct xe_device *xe)
  */
 void xe_hw_error_init(struct xe_device *xe)
 {
+	struct xe_tile *tile = xe_device_get_root_tile(xe);
+
 	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
 		return;
 
+	INIT_WORK(&tile->csc_hw_error_work, csc_hw_error_work);
+
 	process_hw_errors(xe);
 }
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index 86ba767c4e44..53c5af4b810c 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -346,7 +346,7 @@ int xe_survivability_mode_runtime_enable(struct xe_device *xe)
 
 	xe_device_set_wedged_method(xe, DRM_WEDGE_RECOVERY_VENDOR);
 	xe_device_declare_wedged(xe);
-	dev_err(&pdev->dev, "Firmware flash required, Refer the userspace documentation for more details!\n");
+	dev_err(&pdev->dev, "Firmware flash required, Please refer to the userspace documentation for more details!\n");
 
 	return 0;
 }

From d1f51a4f953da4ada15fa585c800ca98d627ce47 Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Tue, 26 Aug 2025 12:04:17 +0530
Subject: [PATCH 0645/1455] drm/xe/xe_hw_error: Add fault injection to trigger
 csc error handler

Add a debugfs fault handler to trigger csc error handler that
wedges the device and enables runtime survivability mode.

v2: add debugfs only for bmg (Umesh)
v3: do not use csc_fault attribute if debugfs is not enabled
v4: rebase

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250826063419.3022216-11-riana.tauro@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_debugfs.c  |  6 +++++-
 drivers/gpu/drm/xe/xe_hw_error.c | 11 +++++++++++
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index bc717519502d..8d6df6bd885e 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -35,6 +35,7 @@
 #endif
 
 DECLARE_FAULT_ATTR(gt_reset_failure);
+DECLARE_FAULT_ATTR(inject_csc_hw_error);
 
 static void read_residency_counter(struct xe_device *xe, struct xe_mmio *mmio,
 				   u32 offset, char *name, struct drm_printer *p)
@@ -361,10 +362,13 @@ void xe_debugfs_register(struct xe_device *xe)
 				 ARRAY_SIZE(debugfs_list),
 				 root, minor);
 
-	if (xe->info.platform == XE_BATTLEMAGE)
+	if (xe->info.platform == XE_BATTLEMAGE) {
 		drm_debugfs_create_files(debugfs_residencies,
 					 ARRAY_SIZE(debugfs_residencies),
 					 root, minor);
+		fault_create_debugfs_attr("inject_csc_hw_error", root,
+					  &inject_csc_hw_error);
+	}
 
 	debugfs_create_file("forcewake_all", 0400, root, xe,
 			    &forcewake_all_fops);
diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c
index 4099ed1262e1..8c65291f36fc 100644
--- a/drivers/gpu/drm/xe/xe_hw_error.c
+++ b/drivers/gpu/drm/xe/xe_hw_error.c
@@ -3,6 +3,8 @@
  * Copyright © 2025 Intel Corporation
  */
 
+#include <linux/fault-inject.h>
+
 #include "regs/xe_gsc_regs.h"
 #include "regs/xe_hw_error_regs.h"
 #include "regs/xe_irq_regs.h"
@@ -13,6 +15,7 @@
 #include "xe_survivability_mode.h"
 
 #define  HEC_UNCORR_FW_ERR_BITS 4
+extern struct fault_attr inject_csc_hw_error;
 
 /* Error categories reported by hardware */
 enum hardware_error {
@@ -43,6 +46,11 @@ static const char *hw_error_to_str(const enum hardware_error hw_err)
 	}
 }
 
+static bool fault_inject_csc_hw_error(void)
+{
+	return IS_ENABLED(CONFIG_DEBUG_FS) && should_fail(&inject_csc_hw_error, 1);
+}
+
 static void csc_hw_error_work(struct work_struct *work)
 {
 	struct xe_tile *tile = container_of(work, typeof(*tile), csc_hw_error_work);
@@ -130,6 +138,9 @@ void xe_hw_error_irq_handler(struct xe_tile *tile, const u32 master_ctl)
 {
 	enum hardware_error hw_err;
 
+	if (fault_inject_csc_hw_error())
+		schedule_work(&tile->csc_hw_error_work);
+
 	for (hw_err = 0; hw_err < HARDWARE_ERROR_MAX; hw_err++)
 		if (master_ctl & ERROR_IRQ(hw_err))
 			hw_error_source_handler(tile, hw_err);

From 6380b1ceba7783ccd2c7fb2e003412c1b8bb4235 Mon Sep 17 00:00:00 2001
From: Lizhi Hou <lizhi.hou@amd.com>
Date: Tue, 26 Aug 2025 10:19:51 -0700
Subject: [PATCH 0646/1455] accel/amdxdna: Fix incorrect type used for a local
 variable

drivers/accel/amdxdna/aie2_pci.c:794:13: sparse: sparse: incorrect type in assignment (different address spaces)

Fixes: c8cea4371e5e ("accel/amdxdna: Add a function to walk hardware contexts")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202508230855.0b9efFl6-lkp@intel.com/
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250826171951.801585-1-lizhi.hou@amd.com
---
 drivers/accel/amdxdna/aie2_pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/accel/amdxdna/aie2_pci.c b/drivers/accel/amdxdna/aie2_pci.c
index 16ac0cab4f44..7a3449541107 100644
--- a/drivers/accel/amdxdna/aie2_pci.c
+++ b/drivers/accel/amdxdna/aie2_pci.c
@@ -785,8 +785,9 @@ static int aie2_get_clock_metadata(struct amdxdna_client *client,
 
 static int aie2_hwctx_status_cb(struct amdxdna_hwctx *hwctx, void *arg)
 {
-	struct amdxdna_drm_query_hwctx __user *buf, *tmp __free(kfree) = NULL;
+	struct amdxdna_drm_query_hwctx *tmp __free(kfree) = NULL;
 	struct amdxdna_drm_get_info *get_info_args = arg;
+	struct amdxdna_drm_query_hwctx __user *buf;
 
 	if (get_info_args->buffer_size < sizeof(*tmp))
 		return -EINVAL;

From d738e1be2b2b4364403babc43ae7343d45e99d41 Mon Sep 17 00:00:00 2001
From: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Date: Fri, 22 Aug 2025 05:55:12 +0530
Subject: [PATCH 0647/1455] drm/xe/wcl: Extend L3bank mask workaround

The commit 9ab440a9d042 ("drm/xe/ptl: L3bank mask is not
available on the media GT") added a workaround to ignore
the fuse register that L3 bank availability as it did not
contain valid values. Same is true for WCL therefore extend
the workaround to cover it.

Signed-off-by: Chaitanya Kumar Borah <chaitanya.kumar.borah@intel.com>
Reviewed-by: Dnyaneshwar Bhadane <dnyaneshwar.bhadane@intel.com>
Link: https://lore.kernel.org/r/20250822002512.1129144-1-chaitanya.kumar.borah@intel.com
Signed-off-by: Gustavo Sousa <gustavo.sousa@intel.com>
---
 drivers/gpu/drm/xe/xe_wa_oob.rules | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 27cd19b98f8e..0cf2b3c03532 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -48,7 +48,7 @@
 16023588340	GRAPHICS_VERSION(2001), FUNC(xe_rtp_match_not_sriov_vf)
 14019789679	GRAPHICS_VERSION(1255)
 		GRAPHICS_VERSION_RANGE(1270, 2004)
-no_media_l3	MEDIA_VERSION(3000)
+no_media_l3	MEDIA_VERSION_RANGE(3000, 3002)
 14022866841	GRAPHICS_VERSION(3000), GRAPHICS_STEP(A0, B0)
 		MEDIA_VERSION(3000), MEDIA_STEP(A0, B0)
 16021333562	GRAPHICS_VERSION_RANGE(1200, 1274)

From 61a0ef5921d603e64beef12f4a3bf7bfde409f18 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Tue, 26 Aug 2025 10:44:57 +0300
Subject: [PATCH 0648/1455] drm/i915/psr: Check PSR pause counter in
 __psr_wait_for_idle_locked
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

PSR work is using __psr_wait_for_idle_locked to ensure PSR exit is
completed before continuing to PSR activation. __psr_wait_for_idle_locked
is unlocking PSR mutex which allows PSR disable/enable and PSR
pause/resume while PSR idle is being wait. PSR enable status is already
checked after locking again PSR mutex but PSR pause counter check is
missing. Due to this PSR work may continue to PSR activation even PSR is
paused.

Fix this by checking PSR pause counter in __psr_wait_for_idle_locked after
PSR mutex is locked again.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Jeevan B <jeevan.b@intel.com>
Link: https://lore.kernel.org/r/20250826074457.1992524-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 681e20d92376..22433fe2ee14 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -3093,7 +3093,7 @@ static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp)
 
 	/* After the unlocked wait, verify that PSR is still wanted! */
 	mutex_lock(&intel_dp->psr.lock);
-	return err == 0 && intel_dp->psr.enabled;
+	return err == 0 && intel_dp->psr.enabled && !intel_dp->psr.pause_counter;
 }
 
 static int intel_psr_fastset_force(struct intel_display *display)

From 27a7e8b6c58144c8c8bd49765364c93f756abc33 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 16:50:25 +0200
Subject: [PATCH 0649/1455] drm/sysfb: Do not deref unexisting CRTC state in
 atomic_disable

Do not access CRTC state in drm_sysfb_plane_helper_atomic_disable().
Use format from sysfb device for clearing scanout buffer. This is
the behavior from before commit 061963cd9e5b ("drm/sysfb: Blit to
CRTC destination format").

When being disabled, the plane has no associated CRTC. Trying to deref
the format pointer results in a segmentation fault. An example stack
track is shown below.

[   58.948915] Oops: general protection fault, probably for non-canonical address 0xdffffc0000000023: 0000 [#1] SMP KASAN PTI
[   58.959971] KASAN: null-ptr-deref in range [0x0000000000000118-0x000000000000011f]
[...]
[   58.979308] RIP: 0010:drm_sysfb_plane_helper_atomic_disable+0x1af/0x520
[...]
[   59.084227] Call Trace:
[   59.086682]  <TASK>
[   59.088793]  ? __pfx_drm_sysfb_plane_helper_atomic_disable+0x10/0x10
[   59.095155]  ? crtc_disable+0xf2/0x5a0
[   59.098920]  drm_atomic_helper_commit_planes+0x848/0x1030
[   59.104336]  drm_atomic_helper_commit_tail+0x41/0xb0
[   59.109316]  commit_tail+0x204/0x330
[   59.112903]  drm_atomic_helper_commit+0x242/0x2e0
[   59.117618]  ? __pfx_drm_atomic_helper_commit+0x10/0x10
[   59.122851]  drm_atomic_commit+0x1e1/0x290
[   59.126957]  ? drm_atomic_add_affected_connectors+0x266/0x330
[   59.132714]  ? __pfx_drm_atomic_commit+0x10/0x10
[   59.137343]  ? __pfx___drm_printfn_info+0x10/0x10
[   59.142058]  ? drm_atomic_set_crtc_for_connector+0x436/0x630
[   59.147729]  atomic_remove_fb+0x631/0x920
[   59.151751]  ? save_trace+0xcf/0x180
[   59.155343]  ? __pfx_atomic_remove_fb+0x10/0x10
[   59.159890]  ? __pfx___drm_dev_dbg+0x10/0x10
[   59.164173]  drm_framebuffer_remove+0x19a/0x710

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Fixes: 061963cd9e5b ("drm/sysfb: Blit to CRTC destination format")
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14874
Cc: Thomas Zimmermann <tzimmermann@suse.de>
Cc: Javier Martinez Canillas <javierm@redhat.com>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/20250826145044.954396-1-tzimmermann@suse.de
---
 drivers/gpu/drm/sysfb/drm_sysfb_modeset.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
index 0e05345cdeee..963c380fea64 100644
--- a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
@@ -281,10 +281,7 @@ void drm_sysfb_plane_helper_atomic_disable(struct drm_plane *plane,
 	struct drm_plane_state *plane_state = drm_atomic_get_new_plane_state(state, plane);
 	void __iomem *dst_vmap = dst.vaddr_iomem; /* TODO: Use mapping abstraction */
 	unsigned int dst_pitch = sysfb->fb_pitch;
-	struct drm_crtc_state *crtc_state = crtc_state =
-		drm_atomic_get_new_crtc_state(state, plane_state->crtc);
-	struct drm_sysfb_crtc_state *sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
-	const struct drm_format_info *dst_format = sysfb_crtc_state->format;
+	const struct drm_format_info *dst_format = sysfb->fb_format;
 	struct drm_rect dst_clip;
 	unsigned long lines, linepixels, i;
 	int idx;

From f315e030a202012a551eb648fa10cd90bd925b20 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Tue, 26 Aug 2025 13:54:27 +0200
Subject: [PATCH 0650/1455] dt-bindings: panel: lvds: Append
 ampire,amp19201200b5tzqw-t03 in panel-lvds

List Ampire AMP19201200B5TZQW-T03 in the LVDS panel enumeration.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20250826-drm-misc-next-v1-1-980d0a0592b9@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 Documentation/devicetree/bindings/display/panel/panel-lvds.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml b/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml
index fcb5834f799a..f8f95e772778 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml
@@ -41,6 +41,8 @@ properties:
       - enum:
           # Admatec 9904379 10.1" 1024x600 LVDS panel
           - admatec,9904379
+          # Ampire AMP19201200B5TZQW-T03 10.1" WUXGA (1920x1200) color TFT LCD panel
+          - ampire,amp19201200b5tzqw-t03
           - auo,b101ew05
           # AUO G084SN05 V9 8.4" 800x600 LVDS panel
           - auo,g084sn05

From 1e00a12dbb09a7cf173ed60fdbdc33f79a6fff7f Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Mon, 25 Aug 2025 15:29:51 +0200
Subject: [PATCH 0651/1455] drm/stm: ltdc: unify log system

DRM_ERROR and similar are deprecated.  Use drm_dev based logging.

Link: https://lore.kernel.org/r/20250821130356.883553-1-raphael.gallais-pou@foss.st.com
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Link: https://lore.kernel.org/r/20250825132951.547899-1-raphael.gallais-pou@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 drivers/gpu/drm/stm/ltdc.c | 139 +++++++++++++++++++------------------
 1 file changed, 70 insertions(+), 69 deletions(-)

diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c
index ba315c66a04d..b9477fbec1c1 100644
--- a/drivers/gpu/drm/stm/ltdc.c
+++ b/drivers/gpu/drm/stm/ltdc.c
@@ -641,7 +641,7 @@ static inline void ltdc_set_ycbcr_config(struct drm_plane *plane, u32 drm_pix_fm
 		break;
 	default:
 		/* RGB or not a YCbCr supported format */
-		DRM_ERROR("Unsupported pixel format: %u\n", drm_pix_fmt);
+		drm_err(plane->dev, "Unsupported pixel format: %u\n", drm_pix_fmt);
 		return;
 	}
 
@@ -664,18 +664,19 @@ static inline void ltdc_set_ycbcr_coeffs(struct drm_plane *plane)
 	u32 lofs = plane->index * LAY_OFS;
 
 	if (enc != DRM_COLOR_YCBCR_BT601 && enc != DRM_COLOR_YCBCR_BT709) {
-		DRM_ERROR("color encoding %d not supported, use bt601 by default\n", enc);
+		drm_err(plane->dev, "color encoding %d not supported, use bt601 by default\n", enc);
 		/* set by default color encoding to DRM_COLOR_YCBCR_BT601 */
 		enc = DRM_COLOR_YCBCR_BT601;
 	}
 
 	if (ran != DRM_COLOR_YCBCR_LIMITED_RANGE && ran != DRM_COLOR_YCBCR_FULL_RANGE) {
-		DRM_ERROR("color range %d not supported, use limited range by default\n", ran);
+		drm_err(plane->dev,
+			"color range %d not supported, use limited range by default\n", ran);
 		/* set by default color range to DRM_COLOR_YCBCR_LIMITED_RANGE */
 		ran = DRM_COLOR_YCBCR_LIMITED_RANGE;
 	}
 
-	DRM_DEBUG_DRIVER("Color encoding=%d, range=%d\n", enc, ran);
+	drm_err(plane->dev, "Color encoding=%d, range=%d\n", enc, ran);
 	regmap_write(ldev->regmap, LTDC_L1CYR0R + lofs,
 		     ltdc_ycbcr2rgb_coeffs[enc][ran][0]);
 	regmap_write(ldev->regmap, LTDC_L1CYR1R + lofs,
@@ -774,7 +775,7 @@ static void ltdc_crtc_atomic_enable(struct drm_crtc *crtc,
 	struct ltdc_device *ldev = crtc_to_ltdc(crtc);
 	struct drm_device *ddev = crtc->dev;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(crtc->dev, "\n");
 
 	pm_runtime_get_sync(ddev->dev);
 
@@ -798,7 +799,7 @@ static void ltdc_crtc_atomic_disable(struct drm_crtc *crtc,
 	struct drm_device *ddev = crtc->dev;
 	int layer_index = 0;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(crtc->dev, "\n");
 
 	drm_crtc_vblank_off(crtc);
 
@@ -837,7 +838,7 @@ ltdc_crtc_mode_valid(struct drm_crtc *crtc,
 
 	result = clk_round_rate(ldev->pixel_clk, target);
 
-	DRM_DEBUG_DRIVER("clk rate target %d, available %d\n", target, result);
+	drm_dbg_driver(crtc->dev, "clk rate target %d, available %d\n", target, result);
 
 	/* Filter modes according to the max frequency supported by the pads */
 	if (result > ldev->caps.pad_max_freq_hz)
@@ -872,14 +873,14 @@ static bool ltdc_crtc_mode_fixup(struct drm_crtc *crtc,
 	int rate = mode->clock * 1000;
 
 	if (clk_set_rate(ldev->pixel_clk, rate) < 0) {
-		DRM_ERROR("Cannot set rate (%dHz) for pixel clk\n", rate);
+		drm_err(crtc->dev, "Cannot set rate (%dHz) for pixel clk\n", rate);
 		return false;
 	}
 
 	adjusted_mode->clock = clk_get_rate(ldev->pixel_clk) / 1000;
 
-	DRM_DEBUG_DRIVER("requested clock %dkHz, adjusted clock %dkHz\n",
-			 mode->clock, adjusted_mode->clock);
+	drm_dbg_driver(crtc->dev, "requested clock %dkHz, adjusted clock %dkHz\n",
+		       mode->clock, adjusted_mode->clock);
 
 	return true;
 }
@@ -934,20 +935,20 @@ static void ltdc_crtc_mode_set_nofb(struct drm_crtc *crtc)
 	if (!pm_runtime_active(ddev->dev)) {
 		ret = pm_runtime_get_sync(ddev->dev);
 		if (ret) {
-			DRM_ERROR("Failed to set mode, cannot get sync\n");
+			drm_err(crtc->dev, "Failed to set mode, cannot get sync\n");
 			return;
 		}
 	}
 
-	DRM_DEBUG_DRIVER("CRTC:%d mode:%s\n", crtc->base.id, mode->name);
-	DRM_DEBUG_DRIVER("Video mode: %dx%d", mode->hdisplay, mode->vdisplay);
-	DRM_DEBUG_DRIVER(" hfp %d hbp %d hsl %d vfp %d vbp %d vsl %d\n",
-			 mode->hsync_start - mode->hdisplay,
-			 mode->htotal - mode->hsync_end,
-			 mode->hsync_end - mode->hsync_start,
-			 mode->vsync_start - mode->vdisplay,
-			 mode->vtotal - mode->vsync_end,
-			 mode->vsync_end - mode->vsync_start);
+	drm_dbg_driver(crtc->dev, "CRTC:%d mode:%s\n", crtc->base.id, mode->name);
+	drm_dbg_driver(crtc->dev, "Video mode: %dx%d", mode->hdisplay, mode->vdisplay);
+	drm_dbg_driver(crtc->dev, " hfp %d hbp %d hsl %d vfp %d vbp %d vsl %d\n",
+		       mode->hsync_start - mode->hdisplay,
+		       mode->htotal - mode->hsync_end,
+		       mode->hsync_end - mode->hsync_start,
+		       mode->vsync_start - mode->vdisplay,
+		       mode->vtotal - mode->vsync_end,
+		       mode->vsync_end - mode->vsync_start);
 
 	/* Convert video timings to ltdc timings */
 	hsync = mode->hsync_end - mode->hsync_start - 1;
@@ -1033,7 +1034,7 @@ static void ltdc_crtc_atomic_flush(struct drm_crtc *crtc,
 	struct drm_device *ddev = crtc->dev;
 	struct drm_pending_vblank_event *event = crtc->state->event;
 
-	DRM_DEBUG_ATOMIC("\n");
+	drm_dbg_atomic(crtc->dev, "\n");
 
 	ltdc_crtc_update_clut(crtc);
 
@@ -1121,7 +1122,7 @@ static int ltdc_crtc_enable_vblank(struct drm_crtc *crtc)
 	struct ltdc_device *ldev = crtc_to_ltdc(crtc);
 	struct drm_crtc_state *state = crtc->state;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(crtc->dev, "\n");
 
 	if (state->enable)
 		regmap_set_bits(ldev->regmap, LTDC_IER, IER_LIE);
@@ -1135,7 +1136,7 @@ static void ltdc_crtc_disable_vblank(struct drm_crtc *crtc)
 {
 	struct ltdc_device *ldev = crtc_to_ltdc(crtc);
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(crtc->dev, "\n");
 	regmap_clear_bits(ldev->regmap, LTDC_IER, IER_LIE);
 }
 
@@ -1144,11 +1145,11 @@ static int ltdc_crtc_set_crc_source(struct drm_crtc *crtc, const char *source)
 	struct ltdc_device *ldev;
 	int ret;
 
-	DRM_DEBUG_DRIVER("\n");
-
 	if (!crtc)
 		return -ENODEV;
 
+	drm_dbg_driver(crtc->dev, "\n");
+
 	ldev = crtc_to_ltdc(crtc);
 
 	if (source && strcmp(source, "auto") == 0) {
@@ -1168,14 +1169,14 @@ static int ltdc_crtc_set_crc_source(struct drm_crtc *crtc, const char *source)
 static int ltdc_crtc_verify_crc_source(struct drm_crtc *crtc,
 				       const char *source, size_t *values_cnt)
 {
-	DRM_DEBUG_DRIVER("\n");
-
 	if (!crtc)
 		return -ENODEV;
 
+	drm_dbg_driver(crtc->dev, "\n");
+
 	if (source && strcmp(source, "auto") != 0) {
-		DRM_DEBUG_DRIVER("Unknown CRC source %s for %s\n",
-				 source, crtc->name);
+		drm_dbg_driver(crtc->dev, "Unknown CRC source %s for %s\n",
+			       source, crtc->name);
 		return -EINVAL;
 	}
 
@@ -1233,7 +1234,7 @@ static int ltdc_plane_atomic_check(struct drm_plane *plane,
 	struct drm_framebuffer *fb = new_plane_state->fb;
 	u32 src_w, src_h;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(plane->dev, "\n");
 
 	if (!fb)
 		return 0;
@@ -1244,7 +1245,7 @@ static int ltdc_plane_atomic_check(struct drm_plane *plane,
 
 	/* Reject scaling */
 	if (src_w != new_plane_state->crtc_w || src_h != new_plane_state->crtc_h) {
-		DRM_DEBUG_DRIVER("Scaling is not supported");
+		drm_dbg_driver(plane->dev, "Scaling is not supported");
 
 		return -EINVAL;
 	}
@@ -1270,7 +1271,7 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane,
 	enum ltdc_pix_fmt pf;
 
 	if (!newstate->crtc || !fb) {
-		DRM_DEBUG_DRIVER("fb or crtc NULL");
+		drm_dbg_driver(plane->dev, "fb or crtc NULL");
 		return;
 	}
 
@@ -1280,11 +1281,11 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane,
 	src_w = newstate->src_w >> 16;
 	src_h = newstate->src_h >> 16;
 
-	DRM_DEBUG_DRIVER("plane:%d fb:%d (%dx%d)@(%d,%d) -> (%dx%d)@(%d,%d)\n",
-			 plane->base.id, fb->base.id,
-			 src_w, src_h, src_x, src_y,
-			 newstate->crtc_w, newstate->crtc_h,
-			 newstate->crtc_x, newstate->crtc_y);
+	drm_dbg_driver(plane->dev, "plane:%d fb:%d (%dx%d)@(%d,%d) -> (%dx%d)@(%d,%d)\n",
+		       plane->base.id, fb->base.id,
+		       src_w, src_h, src_x, src_y,
+		       newstate->crtc_w, newstate->crtc_h,
+		       newstate->crtc_x, newstate->crtc_y);
 
 	regmap_read(ldev->regmap, LTDC_BPCR, &bpcr);
 
@@ -1312,8 +1313,8 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane,
 		val = ltdc_set_flexible_pixel_format(plane, pf);
 
 	if (val == NB_PF) {
-		DRM_ERROR("Pixel format %.4s not supported\n",
-			  (char *)&fb->format->format);
+		drm_err(fb->dev, "Pixel format %.4s not supported\n",
+			(char *)&fb->format->format);
 		val = 0;	/* set by default ARGB 32 bits */
 	}
 	regmap_write_bits(ldev->regmap, LTDC_L1PFCR + lofs, LXPFCR_PF, val);
@@ -1350,7 +1351,7 @@ static void ltdc_plane_atomic_update(struct drm_plane *plane,
 	if (newstate->rotation & DRM_MODE_REFLECT_Y)
 		paddr += (fb->pitches[0] * (y1 - y0));
 
-	DRM_DEBUG_DRIVER("fb: phys 0x%08x", paddr);
+	drm_dbg_driver(fb->dev, "fb: phys 0x%08x", paddr);
 	regmap_write(ldev->regmap, LTDC_L1CFBAR + lofs, paddr);
 
 	/* Configures the color frame buffer pitch in bytes & line length */
@@ -1517,8 +1518,8 @@ static void ltdc_plane_atomic_disable(struct drm_plane *plane,
 		regmap_write_bits(ldev->regmap, LTDC_L1RCR + lofs,
 				  LXRCR_IMR | LXRCR_VBR | LXRCR_GRMSK, LXRCR_VBR);
 
-	DRM_DEBUG_DRIVER("CRTC:%d plane:%d\n",
-			 oldstate->crtc->base.id, plane->base.id);
+	drm_dbg_driver(plane->dev, "CRTC:%d plane:%d\n",
+		       oldstate->crtc->base.id, plane->base.id);
 }
 
 static void ltdc_plane_atomic_print_state(struct drm_printer *p,
@@ -1632,7 +1633,7 @@ static struct drm_plane *ltdc_plane_create(struct drm_device *ddev,
 
 	drm_plane_create_alpha_property(plane);
 
-	DRM_DEBUG_DRIVER("plane:%d created\n", plane->base.id);
+	drm_dbg_driver(plane->dev, "plane:%d created\n", plane->base.id);
 
 	return plane;
 }
@@ -1647,7 +1648,7 @@ static int ltdc_crtc_init(struct drm_device *ddev, struct drm_crtc *crtc)
 
 	primary = ltdc_plane_create(ddev, DRM_PLANE_TYPE_PRIMARY, 0);
 	if (!primary) {
-		DRM_ERROR("Can not create primary plane\n");
+		drm_err(ddev, "Can not create primary plane\n");
 		return -EINVAL;
 	}
 
@@ -1668,7 +1669,7 @@ static int ltdc_crtc_init(struct drm_device *ddev, struct drm_crtc *crtc)
 		ret = drmm_crtc_init_with_planes(ddev, crtc, primary, NULL,
 						 &ltdc_crtc_funcs, NULL);
 	if (ret) {
-		DRM_ERROR("Can not initialize CRTC\n");
+		drm_err(ddev, "Can not initialize CRTC\n");
 		return ret;
 	}
 
@@ -1677,13 +1678,13 @@ static int ltdc_crtc_init(struct drm_device *ddev, struct drm_crtc *crtc)
 	drm_mode_crtc_set_gamma_size(crtc, CLUT_SIZE);
 	drm_crtc_enable_color_mgmt(crtc, 0, false, CLUT_SIZE);
 
-	DRM_DEBUG_DRIVER("CRTC:%d created\n", crtc->base.id);
+	drm_dbg_driver(ddev, "CRTC:%d created\n", crtc->base.id);
 
 	/* Add planes. Note : the first layer is used by primary plane */
 	for (i = 1; i < ldev->caps.nb_layers; i++) {
 		overlay = ltdc_plane_create(ddev, DRM_PLANE_TYPE_OVERLAY, i);
 		if (!overlay) {
-			DRM_ERROR("Can not create overlay plane %d\n", i);
+			drm_err(ddev, "Can not create overlay plane %d\n", i);
 			return -ENOMEM;
 		}
 		if (ldev->caps.dynamic_zorder)
@@ -1704,7 +1705,7 @@ static void ltdc_encoder_disable(struct drm_encoder *encoder)
 	struct drm_device *ddev = encoder->dev;
 	struct ltdc_device *ldev = ddev->dev_private;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(encoder->dev, "\n");
 
 	/* Disable LTDC */
 	regmap_clear_bits(ldev->regmap, LTDC_GCR, GCR_LTDCEN);
@@ -1718,7 +1719,7 @@ static void ltdc_encoder_enable(struct drm_encoder *encoder)
 	struct drm_device *ddev = encoder->dev;
 	struct ltdc_device *ldev = ddev->dev_private;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(encoder->dev, "\n");
 
 	/* set fifo underrun threshold register */
 	if (ldev->caps.fifo_threshold)
@@ -1734,7 +1735,7 @@ static void ltdc_encoder_mode_set(struct drm_encoder *encoder,
 {
 	struct drm_device *ddev = encoder->dev;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(encoder->dev, "\n");
 
 	/*
 	 * Set to default state the pinctrl only with DPI type.
@@ -1770,7 +1771,7 @@ static int ltdc_encoder_init(struct drm_device *ddev, struct drm_bridge *bridge)
 	if (ret)
 		return ret;
 
-	DRM_DEBUG_DRIVER("Bridge encoder:%d created\n", encoder->base.id);
+	drm_dbg_driver(encoder->dev, "Bridge encoder:%d created\n", encoder->base.id);
 
 	return 0;
 }
@@ -1870,7 +1871,7 @@ void ltdc_suspend(struct drm_device *ddev)
 {
 	struct ltdc_device *ldev = ddev->dev_private;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(ddev, "\n");
 	clk_disable_unprepare(ldev->pixel_clk);
 }
 
@@ -1879,11 +1880,11 @@ int ltdc_resume(struct drm_device *ddev)
 	struct ltdc_device *ldev = ddev->dev_private;
 	int ret;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(ddev, "\n");
 
 	ret = clk_prepare_enable(ldev->pixel_clk);
 	if (ret) {
-		DRM_ERROR("failed to enable pixel clock (%d)\n", ret);
+		drm_err(ddev, "failed to enable pixel clock (%d)\n", ret);
 		return ret;
 	}
 
@@ -1903,7 +1904,7 @@ int ltdc_load(struct drm_device *ddev)
 	int irq, i, nb_endpoints;
 	int ret = -ENODEV;
 
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(ddev, "\n");
 
 	/* Get number of endpoints */
 	nb_endpoints = of_graph_get_endpoint_count(np);
@@ -1913,12 +1914,12 @@ int ltdc_load(struct drm_device *ddev)
 	ldev->pixel_clk = devm_clk_get(dev, "lcd");
 	if (IS_ERR(ldev->pixel_clk)) {
 		if (PTR_ERR(ldev->pixel_clk) != -EPROBE_DEFER)
-			DRM_ERROR("Unable to get lcd clock\n");
+			drm_err(ddev, "Unable to get lcd clock\n");
 		return PTR_ERR(ldev->pixel_clk);
 	}
 
 	if (clk_prepare_enable(ldev->pixel_clk)) {
-		DRM_ERROR("Unable to prepare pixel clock\n");
+		drm_err(ddev, "Unable to prepare pixel clock\n");
 		return -ENODEV;
 	}
 
@@ -1939,7 +1940,7 @@ int ltdc_load(struct drm_device *ddev)
 		if (panel) {
 			bridge = drmm_panel_bridge_add(ddev, panel);
 			if (IS_ERR(bridge)) {
-				DRM_ERROR("panel-bridge endpoint %d\n", i);
+				drm_err(ddev, "panel-bridge endpoint %d\n", i);
 				ret = PTR_ERR(bridge);
 				goto err;
 			}
@@ -1949,7 +1950,7 @@ int ltdc_load(struct drm_device *ddev)
 			ret = ltdc_encoder_init(ddev, bridge);
 			if (ret) {
 				if (ret != -EPROBE_DEFER)
-					DRM_ERROR("init encoder endpoint %d\n", i);
+					drm_err(ddev, "init encoder endpoint %d\n", i);
 				goto err;
 			}
 		}
@@ -1967,29 +1968,29 @@ int ltdc_load(struct drm_device *ddev)
 
 	ldev->regs = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(ldev->regs)) {
-		DRM_ERROR("Unable to get ltdc registers\n");
+		drm_err(ddev, "Unable to get ltdc registers\n");
 		ret = PTR_ERR(ldev->regs);
 		goto err;
 	}
 
 	ldev->regmap = devm_regmap_init_mmio(&pdev->dev, ldev->regs, &stm32_ltdc_regmap_cfg);
 	if (IS_ERR(ldev->regmap)) {
-		DRM_ERROR("Unable to regmap ltdc registers\n");
+		drm_err(ddev, "Unable to regmap ltdc registers\n");
 		ret = PTR_ERR(ldev->regmap);
 		goto err;
 	}
 
 	ret = ltdc_get_caps(ddev);
 	if (ret) {
-		DRM_ERROR("hardware identifier (0x%08x) not supported!\n",
-			  ldev->caps.hw_version);
+		drm_err(ddev, "hardware identifier (0x%08x) not supported!\n",
+			ldev->caps.hw_version);
 		goto err;
 	}
 
 	/* Disable all interrupts */
 	regmap_clear_bits(ldev->regmap, LTDC_IER, IER_MASK);
 
-	DRM_DEBUG_DRIVER("ltdc hw version 0x%08x\n", ldev->caps.hw_version);
+	drm_dbg_driver(ddev, "ltdc hw version 0x%08x\n", ldev->caps.hw_version);
 
 	/* initialize default value for fifo underrun threshold & clear interrupt error counters */
 	ldev->transfer_err = 0;
@@ -2008,27 +2009,27 @@ int ltdc_load(struct drm_device *ddev)
 						ltdc_irq_thread, IRQF_ONESHOT,
 						dev_name(dev), ddev);
 		if (ret) {
-			DRM_ERROR("Failed to register LTDC interrupt\n");
+			drm_err(ddev, "Failed to register LTDC interrupt\n");
 			goto err;
 		}
 	}
 
 	crtc = drmm_kzalloc(ddev, sizeof(*crtc), GFP_KERNEL);
 	if (!crtc) {
-		DRM_ERROR("Failed to allocate crtc\n");
+		drm_err(ddev, "Failed to allocate crtc\n");
 		ret = -ENOMEM;
 		goto err;
 	}
 
 	ret = ltdc_crtc_init(ddev, crtc);
 	if (ret) {
-		DRM_ERROR("Failed to init crtc\n");
+		drm_err(ddev, "Failed to init crtc\n");
 		goto err;
 	}
 
 	ret = drm_vblank_init(ddev, NB_CRTC);
 	if (ret) {
-		DRM_ERROR("Failed calling drm_vblank_init()\n");
+		drm_err(ddev, "Failed calling drm_vblank_init()\n");
 		goto err;
 	}
 
@@ -2047,7 +2048,7 @@ err:
 
 void ltdc_unload(struct drm_device *ddev)
 {
-	DRM_DEBUG_DRIVER("\n");
+	drm_dbg_driver(ddev, "\n");
 
 	pm_runtime_disable(ddev->dev);
 }

From 1313351e71181a4818afeb8dfe202e4162091ef6 Mon Sep 17 00:00:00 2001
From: Xin Wang <x.wang@intel.com>
Date: Tue, 26 Aug 2025 17:06:32 -0700
Subject: [PATCH 0652/1455] drm/xe: make xe_gt_idle_disable_c6() handle the
 forcewake internally

Move forcewake_get() into xe_gt_idle_enable_c6() to streamline the
code and make it easier to use.

Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Xin Wang <x.wang@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250827000633.1369890-2-x.wang@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_idle.c | 21 +++++++++++++--------
 drivers/gpu/drm/xe/xe_gt_idle.h |  2 +-
 drivers/gpu/drm/xe/xe_guc_pc.c  | 13 +------------
 3 files changed, 15 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_idle.c b/drivers/gpu/drm/xe/xe_gt_idle.c
index ffb210216aa9..f8950a52d0a4 100644
--- a/drivers/gpu/drm/xe/xe_gt_idle.c
+++ b/drivers/gpu/drm/xe/xe_gt_idle.c
@@ -322,15 +322,11 @@ static void gt_idle_fini(void *arg)
 {
 	struct kobject *kobj = arg;
 	struct xe_gt *gt = kobj_to_gt(kobj->parent);
-	unsigned int fw_ref;
 
 	xe_gt_idle_disable_pg(gt);
 
-	if (gt_to_xe(gt)->info.skip_guc_pc) {
-		fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+	if (gt_to_xe(gt)->info.skip_guc_pc)
 		xe_gt_idle_disable_c6(gt);
-		xe_force_wake_put(gt_to_fw(gt), fw_ref);
-	}
 
 	sysfs_remove_files(kobj, gt_idle_attrs);
 	kobject_put(kobj);
@@ -390,14 +386,23 @@ void xe_gt_idle_enable_c6(struct xe_gt *gt)
 			RC_CTL_HW_ENABLE | RC_CTL_TO_MODE | RC_CTL_RC6_ENABLE);
 }
 
-void xe_gt_idle_disable_c6(struct xe_gt *gt)
+int xe_gt_idle_disable_c6(struct xe_gt *gt)
 {
+	unsigned int fw_ref;
+
 	xe_device_assert_mem_access(gt_to_xe(gt));
-	xe_force_wake_assert_held(gt_to_fw(gt), XE_FW_GT);
 
 	if (IS_SRIOV_VF(gt_to_xe(gt)))
-		return;
+		return 0;
+
+	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+	if (!fw_ref)
+		return -ETIMEDOUT;
 
 	xe_mmio_write32(&gt->mmio, RC_CONTROL, 0);
 	xe_mmio_write32(&gt->mmio, RC_STATE, 0);
+
+	xe_force_wake_put(gt_to_fw(gt), fw_ref);
+
+	return 0;
 }
diff --git a/drivers/gpu/drm/xe/xe_gt_idle.h b/drivers/gpu/drm/xe/xe_gt_idle.h
index 591a01e181bc..9c34a155e102 100644
--- a/drivers/gpu/drm/xe/xe_gt_idle.h
+++ b/drivers/gpu/drm/xe/xe_gt_idle.h
@@ -13,7 +13,7 @@ struct xe_gt;
 
 int xe_gt_idle_init(struct xe_gt_idle *gtidle);
 void xe_gt_idle_enable_c6(struct xe_gt *gt);
-void xe_gt_idle_disable_c6(struct xe_gt *gt);
+int xe_gt_idle_disable_c6(struct xe_gt *gt);
 void xe_gt_idle_enable_pg(struct xe_gt *gt);
 void xe_gt_idle_disable_pg(struct xe_gt *gt);
 int xe_gt_idle_pg_print(struct xe_gt *gt, struct drm_printer *p);
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c
index 0f8430acd16d..88557e86d637 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.c
+++ b/drivers/gpu/drm/xe/xe_guc_pc.c
@@ -1076,7 +1076,6 @@ int xe_guc_pc_gucrc_disable(struct xe_guc_pc *pc)
 {
 	struct xe_device *xe = pc_to_xe(pc);
 	struct xe_gt *gt = pc_to_gt(pc);
-	unsigned int fw_ref;
 	int ret = 0;
 
 	if (xe->info.skip_guc_pc)
@@ -1086,17 +1085,7 @@ int xe_guc_pc_gucrc_disable(struct xe_guc_pc *pc)
 	if (ret)
 		return ret;
 
-	fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FORCEWAKE_ALL);
-	if (!xe_force_wake_ref_has_domain(fw_ref, XE_FORCEWAKE_ALL)) {
-		xe_force_wake_put(gt_to_fw(gt), fw_ref);
-		return -ETIMEDOUT;
-	}
-
-	xe_gt_idle_disable_c6(gt);
-
-	xe_force_wake_put(gt_to_fw(gt), fw_ref);
-
-	return 0;
+	return xe_gt_idle_disable_c6(gt);
 }
 
 /**

From 95d0883ac8105717f59c2dcdc0d8b9150f13aa12 Mon Sep 17 00:00:00 2001
From: Xin Wang <x.wang@intel.com>
Date: Tue, 26 Aug 2025 17:06:33 -0700
Subject: [PATCH 0653/1455] drm/xe: Ensure GT is in C0 during resumes

This patch ensures the gt will be awake for the entire duration
of the resume sequences until GuCRC takes over and GT-C6 gets
re-enabled.

Before suspending GT-C6 is kept enabled, but upon resume, GuCRC
is not yet alive to properly control the exits and some cases of
instability and corruption related to GT-C6 can be observed.

Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4037

Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Xin Wang <x.wang@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4037
Link: https://lore.kernel.org/r/20250827000633.1369890-3-x.wang@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_pm.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index 51ed7270994a..d4b54cd34cff 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -18,7 +18,7 @@
 #include "xe_device.h"
 #include "xe_ggtt.h"
 #include "xe_gt.h"
-#include "xe_guc.h"
+#include "xe_gt_idle.h"
 #include "xe_i2c.h"
 #include "xe_irq.h"
 #include "xe_pcode.h"
@@ -177,6 +177,9 @@ int xe_pm_resume(struct xe_device *xe)
 	drm_dbg(&xe->drm, "Resuming device\n");
 	trace_xe_pm_resume(xe, __builtin_return_address(0));
 
+	for_each_gt(gt, xe, id)
+		xe_gt_idle_disable_c6(gt);
+
 	for_each_tile(tile, xe, id)
 		xe_wa_apply_tile_workarounds(tile);
 
@@ -537,6 +540,9 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 
 	xe_rpm_lockmap_acquire(xe);
 
+	for_each_gt(gt, xe, id)
+		xe_gt_idle_disable_c6(gt);
+
 	if (xe->d3cold.allowed) {
 		err = xe_pcode_ready(xe, true);
 		if (err)

From 063d6683208722b1875f888a45084e3d112701ac Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 24 Jun 2025 11:37:16 -0400
Subject: [PATCH 0654/1455] drm/amdgpu/gfx11: set MQD as appriopriate for queue
 types
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set the MQD as appropriate for the kernel vs user queues.

Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index c01c241a1b06..456ba758fa94 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -4129,6 +4129,8 @@ static int gfx_v11_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 #endif
 	if (prop->tmz_queue)
 		tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, TMZ_MATCH, 1);
+	if (!prop->kernel_queue)
+		tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, RB_NON_PRIV, 1);
 	mqd->cp_gfx_hqd_cntl = tmp;
 
 	/* set up cp_doorbell_control */
@@ -4281,8 +4283,10 @@ static int gfx_v11_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 1);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH,
 			    prop->allow_tunneling);
-	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
-	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	if (prop->kernel_queue) {
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	}
 	if (prop->tmz_queue)
 		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TMZ, 1);
 	mqd->cp_hqd_pq_control = tmp;

From 7b9110f2897957efd9715b52fc01986509729db3 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 24 Jun 2025 11:38:14 -0400
Subject: [PATCH 0655/1455] drm/amdgpu/gfx12: set MQD as appriopriate for queue
 types
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set the MQD as appropriate for the kernel vs user queues.

Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index 3e138527d534..fd44d5503e28 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -3026,6 +3026,8 @@ static int gfx_v12_0_gfx_mqd_init(struct amdgpu_device *adev, void *m,
 #endif
 	if (prop->tmz_queue)
 		tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, TMZ_MATCH, 1);
+	if (!prop->kernel_queue)
+		tmp = REG_SET_FIELD(tmp, CP_GFX_HQD_CNTL, RB_NON_PRIV, 1);
 	mqd->cp_gfx_hqd_cntl = tmp;
 
 	/* set up cp_doorbell_control */
@@ -3175,8 +3177,10 @@ static int gfx_v12_0_compute_mqd_init(struct amdgpu_device *adev, void *m,
 			    (order_base_2(AMDGPU_GPU_PAGE_SIZE / 4) - 1));
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, UNORD_DISPATCH, 1);
 	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TUNNEL_DISPATCH, 0);
-	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
-	tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	if (prop->kernel_queue) {
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, PRIV_STATE, 1);
+		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, KMD_QUEUE, 1);
+	}
 	if (prop->tmz_queue)
 		tmp = REG_SET_FIELD(tmp, CP_HQD_PQ_CONTROL, TMZ, 1);
 	mqd->cp_hqd_pq_control = tmp;

From 22dcb283d63d5677a5875d0002d04d2c61720f78 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Tue, 19 Aug 2025 16:21:02 +0800
Subject: [PATCH 0656/1455] drm/amdgpu: Allocate psp fw private buffer in vram

It's not necessarily to allocate psp firmware private
buffer in different memory domain in sriov and bare
metal environment

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 771a43681d72..772682974ac6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -506,8 +506,7 @@ static int psp_sw_init(struct amdgpu_ip_block *ip_block)
 	}
 
 	ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG,
-				      (amdgpu_sriov_vf(adev) || adev->debug_use_vram_fw_buf) ?
-				      AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT,
+				      AMDGPU_GEM_DOMAIN_VRAM,
 				      &psp->fw_pri_bo,
 				      &psp->fw_pri_mc_addr,
 				      &psp->fw_pri_buf);

From ee6ba1e69dd7d969ab36f16d81c930b1a7fbf61c Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Tue, 19 Aug 2025 16:25:21 +0800
Subject: [PATCH 0657/1455] drm/amdgpu/fence: Remove redundant 0 value
 initialization
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The amdgpu_fence struct is already zeroed by kzalloc(). It's redundant to
initialize am_fence->context to 0.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index bcb74286a78a..fd8cca241da6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -120,7 +120,6 @@ int amdgpu_fence_emit(struct amdgpu_ring *ring, struct dma_fence **f,
 		am_fence = kzalloc(sizeof(*am_fence), GFP_KERNEL);
 		if (!am_fence)
 			return -ENOMEM;
-		am_fence->context = 0;
 	} else {
 		am_fence = af;
 	}

From 8ed5f8d9fe59cc1b1ac267182c3ce7d2210855f4 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Tue, 19 Aug 2025 22:24:50 +0800
Subject: [PATCH 0658/1455] drm/amd/display: Remove redundant header files

The header file "dc_stream.h" is already included on line 1507. Remove the
redundant include.

This is because the header file was initially included towards the latter
part of the code. Subsequent commits had to include the header file again
earlier in the code. In my opinion, this doesn't count as a fix; it just
requires removing the redundant header inclusion.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index eb7019ed92b2..3431dee6389d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1863,8 +1863,6 @@ void dc_3dlut_func_retain(struct dc_3dlut *lut);
 void dc_post_update_surfaces_to_stream(
 		struct dc *dc);
 
-#include "dc_stream.h"
-
 /**
  * struct dc_validation_set - Struct to store surface/stream associations for validation
  */

From 907813e5d7cadfeafab12467d748705a5309efb0 Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Thu, 7 Aug 2025 12:36:05 +0800
Subject: [PATCH 0659/1455] drm/amdgpu: Correct the counts of nr_banks and
 nr_errors

Correct the counts of nr_banks and nr_errors

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
index d1e431818212..9b3180449150 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_aca.c
@@ -76,6 +76,7 @@ static void aca_banks_release(struct aca_banks *banks)
 	list_for_each_entry_safe(node, tmp, &banks->list, node) {
 		list_del(&node->node);
 		kvfree(node);
+		banks->nr_banks--;
 	}
 }
 
@@ -238,6 +239,7 @@ static struct aca_bank_error *new_bank_error(struct aca_error *aerr, struct aca_
 
 	mutex_lock(&aerr->lock);
 	list_add_tail(&bank_error->node, &aerr->list);
+	aerr->nr_errors++;
 	mutex_unlock(&aerr->lock);
 
 	return bank_error;

From 0989b764f43d4de2c6665c15165c251d9cfde9c0 Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Tue, 19 Aug 2025 14:47:05 +0800
Subject: [PATCH 0660/1455] drm/amdgpu: Add a mutex lock to protect poison
 injection

When poison is triggered multiple times, competition will occur.
Add a mutex lock to protect poison injection

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h | 2 ++
 2 files changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index e6b8c2cf80ea..ff2b522d86c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -3512,6 +3512,7 @@ static int amdgpu_ras_page_retirement_thread(void *param)
 		if (kthread_should_stop())
 			break;
 
+		mutex_lock(&con->poison_lock);
 		gpu_reset = 0;
 
 		do {
@@ -3568,6 +3569,7 @@ static int amdgpu_ras_page_retirement_thread(void *param)
 			/* Wake up work to save bad pages to eeprom */
 			schedule_delayed_work(&con->page_retirement_dwork, 0);
 		}
+		mutex_unlock(&con->poison_lock);
 	}
 
 	return 0;
@@ -3648,6 +3650,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info)
 	}
 
 	mutex_init(&con->recovery_lock);
+	mutex_init(&con->poison_lock);
 	INIT_WORK(&con->recovery_work, amdgpu_ras_do_recovery);
 	atomic_set(&con->in_recovery, 0);
 	atomic_set(&con->rma_in_recovery, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index ff63020f9c6c..2a70782b07bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -587,6 +587,8 @@ struct amdgpu_ras {
 	struct list_head critical_region_head;
 	struct mutex critical_region_lock;
 
+	/* Protect poison injection */
+	struct mutex poison_lock;
 };
 
 struct ras_fs_data {

From d8442bcad0764c5613e9f8b2356f3e0a48327e20 Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Wed, 20 Aug 2025 17:18:57 +0800
Subject: [PATCH 0661/1455] drm/amdgpu: Correct the loss of aca bank reg info

By polling, poll ACA bank count to ensure that valid
ACA bank reg info can be obtained

v2: add corresponding delay before send msg to SMU to query mca bank info
(Stanley)

v3: the loop cannot exit. (Thomas)

v4: remove amdgpu_aca_clear_bank_count. (Kevin)

v5: continuously inject ce. If a creation interruption
occurs at this time, bank reg info will be lost. (Thomas)
v5: each cycle is delayed by 100ms. (Tao)

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 50 +++++++++++--------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h |  5 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c |  1 +
 drivers/gpu/drm/amd/amdgpu/umc_v12_0.c  |  5 ++-
 4 files changed, 29 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index ff2b522d86c6..21678b3d620c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -122,7 +122,7 @@ const char *get_ras_block_str(struct ras_common_if *ras_block)
 /* typical ECC bad page rate is 1 bad page per 100MB VRAM */
 #define RAS_BAD_PAGE_COVER              (100 * 1024 * 1024ULL)
 
-#define MAX_UMC_POISON_POLLING_TIME_ASYNC  300  //ms
+#define MAX_UMC_POISON_POLLING_TIME_ASYNC  10
 
 #define AMDGPU_RAS_RETIRE_PAGE_INTERVAL 100  //ms
 
@@ -3317,7 +3317,7 @@ static void amdgpu_ras_ecc_log_init(struct ras_ecc_log_info *ecc_log)
 
 	INIT_RADIX_TREE(&ecc_log->de_page_tree, GFP_KERNEL);
 	ecc_log->de_queried_count = 0;
-	ecc_log->prev_de_queried_count = 0;
+	ecc_log->consumption_q_count = 0;
 }
 
 static void amdgpu_ras_ecc_log_fini(struct ras_ecc_log_info *ecc_log)
@@ -3337,7 +3337,7 @@ static void amdgpu_ras_ecc_log_fini(struct ras_ecc_log_info *ecc_log)
 
 	mutex_destroy(&ecc_log->lock);
 	ecc_log->de_queried_count = 0;
-	ecc_log->prev_de_queried_count = 0;
+	ecc_log->consumption_q_count = 0;
 }
 
 static bool amdgpu_ras_schedule_retirement_dwork(struct amdgpu_ras *con,
@@ -3387,47 +3387,34 @@ static int amdgpu_ras_poison_creation_handler(struct amdgpu_device *adev,
 	int ret = 0;
 	struct ras_ecc_log_info *ecc_log;
 	struct ras_query_if info;
-	uint32_t timeout = 0;
+	u32 timeout = MAX_UMC_POISON_POLLING_TIME_ASYNC;
 	struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
-	uint64_t de_queried_count;
-	uint32_t new_detect_count, total_detect_count;
-	uint32_t need_query_count = poison_creation_count;
+	u64 de_queried_count;
+	u64 consumption_q_count;
 	enum ras_event_type type = RAS_EVENT_TYPE_POISON_CREATION;
 
 	memset(&info, 0, sizeof(info));
 	info.head.block = AMDGPU_RAS_BLOCK__UMC;
 
 	ecc_log = &ras->umc_ecc_log;
-	total_detect_count = 0;
+	ecc_log->de_queried_count = 0;
+	ecc_log->consumption_q_count = 0;
+
 	do {
 		ret = amdgpu_ras_query_error_status_with_event(adev, &info, type);
 		if (ret)
 			return ret;
 
 		de_queried_count = ecc_log->de_queried_count;
-		if (de_queried_count > ecc_log->prev_de_queried_count) {
-			new_detect_count = de_queried_count - ecc_log->prev_de_queried_count;
-			ecc_log->prev_de_queried_count = de_queried_count;
-			timeout = 0;
-		} else {
-			new_detect_count = 0;
-		}
+		consumption_q_count = ecc_log->consumption_q_count;
 
-		if (new_detect_count) {
-			total_detect_count += new_detect_count;
-		} else {
-			if (!timeout && need_query_count)
-				timeout = MAX_UMC_POISON_POLLING_TIME_ASYNC;
+		if (de_queried_count && consumption_q_count)
+			break;
 
-			if (timeout) {
-				if (!--timeout)
-					break;
-				msleep(1);
-			}
-		}
-	} while (total_detect_count < need_query_count);
+		msleep(100);
+	} while (--timeout);
 
-	if (total_detect_count)
+	if (de_queried_count)
 		schedule_delayed_work(&ras->page_retirement_dwork, 0);
 
 	if (amdgpu_ras_is_rma(adev) && atomic_cmpxchg(&ras->rma_in_recovery, 0, 1) == 0)
@@ -3525,7 +3512,8 @@ static int amdgpu_ras_page_retirement_thread(void *param)
 				atomic_sub(poison_creation_count, &con->poison_creation_count);
 				atomic_sub(poison_creation_count, &con->page_retirement_req_cnt);
 			}
-		} while (atomic_read(&con->poison_creation_count));
+		} while (atomic_read(&con->poison_creation_count) &&
+			!atomic_read(&con->poison_consumption_count));
 
 		if (ret != -EIO) {
 			msg_count = kfifo_len(&con->poison_fifo);
@@ -3542,6 +3530,7 @@ static int amdgpu_ras_page_retirement_thread(void *param)
 			/* gpu mode-1 reset is ongoing or just completed ras mode-1 reset */
 			/* Clear poison creation request */
 			atomic_set(&con->poison_creation_count, 0);
+			atomic_set(&con->poison_consumption_count, 0);
 
 			/* Clear poison fifo */
 			amdgpu_ras_clear_poison_fifo(adev);
@@ -3566,6 +3555,8 @@ static int amdgpu_ras_page_retirement_thread(void *param)
 				atomic_sub(msg_count, &con->page_retirement_req_cnt);
 			}
 
+			atomic_set(&con->poison_consumption_count, 0);
+
 			/* Wake up work to save bad pages to eeprom */
 			schedule_delayed_work(&con->page_retirement_dwork, 0);
 		}
@@ -3671,6 +3662,7 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info)
 	init_waitqueue_head(&con->page_retirement_wq);
 	atomic_set(&con->page_retirement_req_cnt, 0);
 	atomic_set(&con->poison_creation_count, 0);
+	atomic_set(&con->poison_consumption_count, 0);
 	con->page_retirement_thread =
 		kthread_run(amdgpu_ras_page_retirement_thread, adev, "umc_page_retirement");
 	if (IS_ERR(con->page_retirement_thread)) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 2a70782b07bf..6cf0dfd38be8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -492,8 +492,8 @@ struct ras_ecc_err {
 struct ras_ecc_log_info {
 	struct mutex lock;
 	struct radix_tree_root de_page_tree;
-	uint64_t	de_queried_count;
-	uint64_t	prev_de_queried_count;
+	uint64_t de_queried_count;
+	uint64_t consumption_q_count;
 };
 
 struct ras_critical_region {
@@ -565,6 +565,7 @@ struct amdgpu_ras {
 	struct mutex page_retirement_lock;
 	atomic_t page_retirement_req_cnt;
 	atomic_t poison_creation_count;
+	atomic_t poison_consumption_count;
 	struct mutex page_rsv_lock;
 	DECLARE_KFIFO(poison_fifo, struct ras_poison_msg, 128);
 	struct ras_ecc_log_info  umc_ecc_log;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
index c92b8794aa73..2e039fb778ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umc.c
@@ -252,6 +252,7 @@ int amdgpu_umc_pasid_poison_handler(struct amdgpu_device *adev,
 				block, pasid, pasid_fn, data, reset);
 			if (!ret) {
 				atomic_inc(&con->page_retirement_req_cnt);
+				atomic_inc(&con->poison_consumption_count);
 				wake_up(&con->page_retirement_wq);
 			}
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
index e590cbdd8de9..8dc32787d625 100644
--- a/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/umc_v12_0.c
@@ -536,8 +536,11 @@ static int umc_v12_0_update_ecc_status(struct amdgpu_device *adev,
 	hwid = REG_GET_FIELD(ipid, MCMP1_IPIDT0, HardwareID);
 	mcatype = REG_GET_FIELD(ipid, MCMP1_IPIDT0, McaType);
 
-	if ((hwid != MCA_UMC_HWID_V12_0) || (mcatype != MCA_UMC_MCATYPE_V12_0))
+	/* The IP block decode of consumption is SMU */
+	if (hwid != MCA_UMC_HWID_V12_0 || mcatype != MCA_UMC_MCATYPE_V12_0) {
+		con->umc_ecc_log.consumption_q_count++;
 		return 0;
+	}
 
 	if (!status)
 		return 0;

From 01479d140686430ce51d01dc4ad4548323bd1232 Mon Sep 17 00:00:00 2001
From: Kathara Sasikumar <katharasasikumar007@gmail.com>
Date: Tue, 19 Aug 2025 21:19:58 +0000
Subject: [PATCH 0662/1455] Documentation/gpu/amdgpu: Fix duplicate word in
 driver-core.rst

Remove duplicate word 'and' in driver-core.rst.

Signed-off-by: Kathara Sasikumar <katharasasikumar007@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/driver-core.rst | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/gpu/amdgpu/driver-core.rst b/Documentation/gpu/amdgpu/driver-core.rst
index 81256318e93c..bd4be32f2725 100644
--- a/Documentation/gpu/amdgpu/driver-core.rst
+++ b/Documentation/gpu/amdgpu/driver-core.rst
@@ -65,7 +65,7 @@ SDMA (System DMA)
 
 GC (Graphics and Compute)
     This is the graphics and compute engine, i.e., the block that
-    encompasses the 3D pipeline and and shader blocks.  This is by far the
+    encompasses the 3D pipeline and shader blocks.  This is by far the
     largest block on the GPU.  The 3D pipeline has tons of sub-blocks.  In
     addition to that, it also contains the CP microcontrollers (ME, PFP, CE,
     MEC) and the RLC microcontroller.  It's exposed to userspace for user mode

From b9dba6a644fc31a7342962bf0a5c0142efd4a2f2 Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Thu, 21 Aug 2025 10:59:23 +0800
Subject: [PATCH 0663/1455] drm/amdgpu: Add support for dpc to a series of
 products

Add support for dpc to a series of products

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 627a8188d868..a4be035fff30 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3135,7 +3135,7 @@ static inline bool smu_v13_0_6_is_link_reset_supported(struct smu_context *smu)
 	struct amdgpu_device *adev = smu->adev;
 	int var = (adev->pdev->device & 0xF);
 
-	if (var == 0x1)
+	if (var == 0x1 || var == 0x0)
 		return true;
 
 	return false;

From e68197aa2bbf658a95a927db01c2933c8125bf17 Mon Sep 17 00:00:00 2001
From: Yifan Zhang <yifan1.zhang@amd.com>
Date: Wed, 20 Aug 2025 14:55:41 +0800
Subject: [PATCH 0664/1455] drm/amdgpu: remove redundant AMDGPU_HAS_VRAM
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

AMDGPU_HAS_VRAM is redundant with is_app_apu, as both refer to
APUs with no carve-out. Since AMDGPU_HAS_VRAM only occurs once,
remove AMDGPU_HAS_VRAM definition. The tmr allocation can be covered
with AMDGPU_GEM_DOMAIN_GTT | AMDGPU_GEM_DOMAIN_VRAM in both vram and
non vram ASICs.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     | 6 ------
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 4 +---
 2 files changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index ddd472e56f69..01f53700694b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -946,12 +946,6 @@ enum amdgpu_enforce_isolation_mode {
 	AMDGPU_ENFORCE_ISOLATION_NO_CLEANER_SHADER = 3,
 };
 
-
-/*
- * Non-zero (true) if the GPU has VRAM. Zero (false) otherwise.
- */
-#define AMDGPU_HAS_VRAM(_adev) ((_adev)->gmc.real_vram_size)
-
 struct amdgpu_device {
 	struct device			*dev;
 	struct pci_dev			*pdev;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 772682974ac6..2c4db7e9d7e8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -880,9 +880,7 @@ static int psp_tmr_init(struct psp_context *psp)
 		pptr = amdgpu_sriov_vf(psp->adev) ? &tmr_buf : NULL;
 		ret = amdgpu_bo_create_kernel(psp->adev, tmr_size,
 					      PSP_TMR_ALIGNMENT,
-					      AMDGPU_HAS_VRAM(psp->adev) ?
-					      AMDGPU_GEM_DOMAIN_VRAM :
-					      AMDGPU_GEM_DOMAIN_GTT,
+					      AMDGPU_GEM_DOMAIN_GTT | AMDGPU_GEM_DOMAIN_VRAM,
 					      &psp->tmr_bo, &psp->tmr_mc_addr,
 					      pptr);
 	}

From 863a17904a8250aeb2692f703721b70cda8a1c82 Mon Sep 17 00:00:00 2001
From: Ce Sun <cesun102@amd.com>
Date: Thu, 21 Aug 2025 16:27:57 +0800
Subject: [PATCH 0665/1455] drm/amdgpu: Add support for dpc to the product

Add support for dpc to the product

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index a4be035fff30..d65c912d2408 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3135,7 +3135,7 @@ static inline bool smu_v13_0_6_is_link_reset_supported(struct smu_context *smu)
 	struct amdgpu_device *adev = smu->adev;
 	int var = (adev->pdev->device & 0xF);
 
-	if (var == 0x1 || var == 0x0)
+	if (var == 0x0 || var == 0x1 || var == 0x3)
 		return true;
 
 	return false;

From f117eca965c797c5efb55ee40543cb37b9cb7d23 Mon Sep 17 00:00:00 2001
From: Brahmajit Das <listout@listout.xyz>
Date: Thu, 21 Aug 2025 17:31:33 +0530
Subject: [PATCH 0666/1455] drm/amd/display: clean-up dead code in
 dml2_mall_phantom
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

pipe_idx in funtion dml2_svp_validate_static_schedulabilit, although set
is never actually used. While building with GCC 16 this gives a warning:

drivers/gpu/drm/amd/amdgpu/../display/dc/dml2/dml2_mall_phantom.c: In function ‘set_phantom_stream_timing’:
drivers/gpu/drm/amd/amdgpu/../display/dc/dml2/dml2_mall_phantom.c:657:25: warning: variable ‘pipe_idx’ set but not used [-Wunused-but-set-variable=]
  657 |         unsigned int i, pipe_idx;
      |                         ^~~~~~~~

Signed-off-by: Brahmajit Das <listout@listout.xyz>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c
index a56e75cdf712..c59f825cfae9 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_mall_phantom.c
@@ -654,14 +654,14 @@ static void set_phantom_stream_timing(struct dml2_context *ctx, struct dc_state
 				     unsigned int svp_height,
 				     unsigned int svp_vstartup)
 {
-	unsigned int i, pipe_idx;
+	unsigned int i;
 	double line_time, fp_and_sync_width_time;
 	struct pipe_ctx *pipe;
 	uint32_t phantom_vactive, phantom_bp, pstate_width_fw_delay_lines;
 	static const double cvt_rb_vblank_max = ((double) 460 / (1000 * 1000));
 
 	// Find DML pipe index (pipe_idx) using dc_pipe_idx
-	for (i = 0, pipe_idx = 0; i < ctx->config.dcn_pipe_count; i++) {
+	for (i = 0; i < ctx->config.dcn_pipe_count; i++) {
 		pipe = &state->res_ctx.pipe_ctx[i];
 
 		if (!pipe->stream)
@@ -669,8 +669,6 @@ static void set_phantom_stream_timing(struct dml2_context *ctx, struct dc_state
 
 		if (i == dc_pipe_idx)
 			break;
-
-		pipe_idx++;
 	}
 
 	// Calculate lines required for pstate allow width and FW processing delays

From cbf3a5d67a5bcae620b5315e4891f8c3e939bc79 Mon Sep 17 00:00:00 2001
From: Rakuram Eswaran <rakuram.e96@gmail.com>
Date: Thu, 21 Aug 2025 08:29:55 +0530
Subject: [PATCH 0667/1455] docs: gpu: amdgpu: Fix spelling in amdgpu
 documentation

Fixed following typos reported by Codespell

1. propogated ==> propagated
   aperatures ==> apertures
In Documentation/gpu/amdgpu/debugfs.rst

2. parition ==> partition
In Documentation/gpu/amdgpu/process-isolation.rst

3. conections ==> connections
In Documentation/gpu/amdgpu/display/programming-model-dcn.rst

In addition to above,
Fixed wrong bit-partition naming in gpu/amdgpu/process-isolation.rst
from "fourth" partition to "third" partition.

Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Suggested-by: Randy Dunlap <rdunlap@infradead.org>
Suggested-by: Alexander Deucher <Alexander.Deucher@amd.com>
Signed-off-by: Rakuram Eswaran <rakuram.e96@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/debugfs.rst                       | 4 ++--
 Documentation/gpu/amdgpu/display/programming-model-dcn.rst | 2 +-
 Documentation/gpu/amdgpu/process-isolation.rst             | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/Documentation/gpu/amdgpu/debugfs.rst b/Documentation/gpu/amdgpu/debugfs.rst
index 5150d0a95658..151d8bfc79e2 100644
--- a/Documentation/gpu/amdgpu/debugfs.rst
+++ b/Documentation/gpu/amdgpu/debugfs.rst
@@ -94,7 +94,7 @@ amdgpu_error_<name>
 -------------------
 
 Provides an interface to set an error code on the dma fences associated with
-ring <name>.  The error code specified is propogated to all fences associated
+ring <name>.  The error code specified is propagated to all fences associated
 with the ring.  Use this to inject a fence error into a ring.
 
 amdgpu_pm_info
@@ -165,7 +165,7 @@ GTT memory.
 amdgpu_regs_*
 -------------
 
-Provides direct access to various register aperatures on the GPU.  Used
+Provides direct access to various register apertures on the GPU.  Used
 by tools like UMR to access GPU registers.
 
 amdgpu_regs2
diff --git a/Documentation/gpu/amdgpu/display/programming-model-dcn.rst b/Documentation/gpu/amdgpu/display/programming-model-dcn.rst
index c1b48d49fb0b..bc7de97a746f 100644
--- a/Documentation/gpu/amdgpu/display/programming-model-dcn.rst
+++ b/Documentation/gpu/amdgpu/display/programming-model-dcn.rst
@@ -100,7 +100,7 @@ represents the connected display.
    For historical reasons, we used the name `dc_link`, which gives the
    wrong impression that this abstraction only deals with physical connections
    that the developer can easily manipulate. However, this also covers
-   conections like eDP or cases where the output is connected to other devices.
+   connections like eDP or cases where the output is connected to other devices.
 
 There are two structs that are not represented in the diagram since they were
 elaborated in the DCN overview page  (check the DCN block diagram :ref:`Display
diff --git a/Documentation/gpu/amdgpu/process-isolation.rst b/Documentation/gpu/amdgpu/process-isolation.rst
index 6b6d70e357a7..25b06ffefc33 100644
--- a/Documentation/gpu/amdgpu/process-isolation.rst
+++ b/Documentation/gpu/amdgpu/process-isolation.rst
@@ -26,7 +26,7 @@ Example of enabling enforce isolation on a GPU with multiple partitions:
     $ cat /sys/class/drm/card0/device/enforce_isolation
     1 0 1 0
 
-The output indicates that enforce isolation is enabled on zeroth and second parition and disabled on first and fourth parition.
+The output indicates that enforce isolation is enabled on zeroth and second partition and disabled on first and third partition.
 
 For devices with a single partition or those that do not support partitions, there will be only one element:
 

From 7bbfa1b1fa7b91cb805563ebffd1d55df192df05 Mon Sep 17 00:00:00 2001
From: Sunday Clement <Sunday.Clement@amd.com>
Date: Fri, 30 May 2025 10:58:08 -0400
Subject: [PATCH 0668/1455] drm/amdkfd: Allow device error to be logged

The addition of a WARN_ON() check in order to return early in the
kq_initialize function retroactively causes the default case in the
following switch statement to never be executed, preventing dev_err
from logging device errors in the kernel. Both logs are now checked
in the default case.

Signed-off-by: Sunday Clement <Sunday.Clement@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
index 2b0a830f5b29..fb3129883a4c 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c
@@ -46,11 +46,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev,
 	int retval;
 	union PM4_MES_TYPE_3_HEADER nop;
 
-	if (WARN_ON(type != KFD_QUEUE_TYPE_DIQ && type != KFD_QUEUE_TYPE_HIQ))
-		return false;
-
-	pr_debug("Initializing queue type %d size %d\n", KFD_QUEUE_TYPE_HIQ,
-			queue_size);
+	pr_debug("Initializing queue type %d size %d\n", type, queue_size);
 
 	memset(&prop, 0, sizeof(prop));
 	memset(&nop, 0, sizeof(nop));
@@ -69,6 +65,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_node *dev,
 		kq->mqd_mgr = dev->dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ];
 		break;
 	default:
+		WARN(1, "Invalid queue type %d\n", type);
 		dev_err(dev->adev->dev, "Invalid queue type %d\n", type);
 		return false;
 	}

From 93aa919ca05bec544b17ee9a1bfe394ce6c94bd8 Mon Sep 17 00:00:00 2001
From: Eric Huang <jinhuieric.huang@amd.com>
Date: Mon, 18 Aug 2025 14:22:53 -0400
Subject: [PATCH 0669/1455] drm/amdkfd: fix vram allocation failure for a
 special case

When it only allocates vram without va, which is 0, and a
SVM range allocated stays in this range, the vram allocation
returns failure. It should be skipped for this case from
SVM usage check.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 79ed3be63d0d..43115a374469 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1070,7 +1070,12 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
 	svm_range_list_lock_and_flush_work(&p->svms, current->mm);
 	mutex_lock(&p->svms.lock);
 	mmap_write_unlock(current->mm);
-	if (interval_tree_iter_first(&p->svms.objects,
+
+	/* Skip a special case that allocates VRAM without VA,
+	 * VA will be invalid of 0.
+	 */
+	if (!(!args->va_addr && (flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM)) &&
+	    interval_tree_iter_first(&p->svms.objects,
 				     args->va_addr >> PAGE_SHIFT,
 				     (args->va_addr + args->size - 1) >> PAGE_SHIFT)) {
 		pr_err("Address: 0x%llx already allocated by SVM\n",

From 181d7dcc584f8ac575ea2b6e51498547186df9f3 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Tue, 19 Aug 2025 09:46:04 +0800
Subject: [PATCH 0670/1455] drm/amd/pm: Update SMU v13.0.6 PPT caps
 initialization

Update the conditions for setting the SMU vcn reset caps in the SMU v13.0.6 PPT
initialization function. Specifically:

- Add support for VCN reset capability for firmware versions 0x00558200 and
  above when the program version is 0.
- Add support for VCN reset capability for firmware versions 0x05551800 and
  above when the program version is 5.

v2: correct the smu mp1 version for program 5 (Lijo)

Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index d65c912d2408..fe2d48dc47af 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -312,6 +312,8 @@ static void smu_v13_0_14_init_caps(struct smu_context *smu)
 		smu_v13_0_6_cap_set(smu, SMU_CAP(PER_INST_METRICS));
 	if (fw_ver >= 0x5551200)
 		smu_v13_0_6_cap_set(smu, SMU_CAP(SDMA_RESET));
+	if (fw_ver >= 0x5551800)
+		smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET));
 	if (fw_ver >= 0x5551600) {
 		smu_v13_0_6_cap_set(smu, SMU_CAP(STATIC_METRICS));
 		smu_v13_0_6_cap_set(smu, SMU_CAP(BOARD_VOLTAGE));
@@ -437,7 +439,8 @@ static void smu_v13_0_6_init_caps(struct smu_context *smu)
 	    ((pgm == 4) && (fw_ver >= 0x4557000)))
 		smu_v13_0_6_cap_set(smu, SMU_CAP(SDMA_RESET));
 
-	if ((pgm == 4) && (fw_ver >= 0x04557100))
+	if (((pgm == 0) && (fw_ver >= 0x00558200)) ||
+	    ((pgm == 4) && (fw_ver >= 0x04557100)))
 		smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET));
 }
 

From f3820e9d356132e18405cd7606e22dc87ccfa6d1 Mon Sep 17 00:00:00 2001
From: Amber Lin <Amber.Lin@amd.com>
Date: Fri, 15 Aug 2025 14:04:15 -0400
Subject: [PATCH 0671/1455] drm/amdkfd: Tie UNMAP_LATENCY to queue_preemption

When KFD asks CP to preempt queues, other than preempt CP queues, CP
also requests SDMA to preempt SDMA queues with UNMAP_LATENCY timeout.
Currently queue_preemption_timeout_ms is 9000 ms by default but can be
configured via module parameter. KFD_UNMAP_LATENCY_MS is hard coded as
4000 ms though. This patch ties KFD_UNMAP_LATENCY_MS to
queue_preemption_timeout_ms so in a slow system such as emulator, both
CP and SDMA slowness are taken into account.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 67694bcd9464..d01ef5ac0766 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -111,7 +111,14 @@
 
 #define KFD_KERNEL_QUEUE_SIZE 2048
 
-#define KFD_UNMAP_LATENCY_MS	(4000)
+/*  KFD_UNMAP_LATENCY_MS is the timeout CP waiting for SDMA preemption. One XCC
+ *  can be associated to 2 SDMA engines. queue_preemption_timeout_ms is the time
+ *  driver waiting for CP returning the UNMAP_QUEUE fence. Thus the math is
+ *  queue_preemption_timeout_ms = sdma_preemption_time * 2 + cp workload
+ *  The format here makes CP workload 10% of total timeout
+ */
+#define KFD_UNMAP_LATENCY_MS	\
+	((queue_preemption_timeout_ms - queue_preemption_timeout_ms / 10) >> 1)
 
 #define KFD_MAX_SDMA_QUEUES	128
 

From fa064d50b7eee6f40749ff03de7ef99ff4acc8c9 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 13 Aug 2025 14:53:50 -0400
Subject: [PATCH 0672/1455] drm/amdgpu/vcn: drop extra
 cancel_delayed_work_sync()

We already call this in the hw_fini() methods for all
VCN instances, so no need to call it again in
amdgpu_vcn_suspend().

Tested-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 9a76e11d1c18..fd8ebf4b5a82 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -357,8 +357,6 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev, int i)
 	if (adev->vcn.harvest_config & (1 << i))
 		return 0;
 
-	cancel_delayed_work_sync(&adev->vcn.inst[i].idle_work);
-
 	/* err_event_athub and dpc recovery will corrupt VCPU buffer, so we need to
 	 * restore fw data and clear buffer in amdgpu_vcn_resume() */
 	if (in_ras_intr || adev->pcie_reset_ctx.in_link_reset)

From 1c65502f813318d48084b33a2c5713d317019dfe Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 18 Aug 2025 15:19:47 -0400
Subject: [PATCH 0673/1455] drm/amdgpu/vpe: add ring reset support

Implement ring reset for VPE.  Similar to VCN and JPEG,
just powergate the the IP to reset it.

v2: Properly set per queue reset flag

Reviewed-by: Jesse Zhang <jesse.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c
index 121ee17b522b..86573c8d9b91 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c
@@ -379,9 +379,10 @@ static int vpe_sw_init(struct amdgpu_ip_block *ip_block)
 	if (ret)
 		goto out;
 
-	/* TODO: Add queue reset mask when FW fully supports it */
 	adev->vpe.supported_reset =
 		 amdgpu_get_soft_full_reset_mask(&adev->vpe.ring);
+	if (!amdgpu_sriov_vf(adev))
+		adev->vpe.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
 	ret = amdgpu_vpe_sysfs_reset_mask_init(adev);
 	if (ret)
 		goto out;
@@ -874,6 +875,27 @@ static void vpe_ring_end_use(struct amdgpu_ring *ring)
 	schedule_delayed_work(&adev->vpe.idle_work, VPE_IDLE_TIMEOUT);
 }
 
+static int vpe_ring_reset(struct amdgpu_ring *ring,
+			  unsigned int vmid,
+			  struct amdgpu_fence *timedout_fence)
+{
+	struct amdgpu_device *adev = ring->adev;
+	int r;
+
+	amdgpu_ring_reset_helper_begin(ring, timedout_fence);
+
+	r = amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE,
+						   AMD_PG_STATE_GATE);
+	if (r)
+		return r;
+	r = amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VPE,
+						   AMD_PG_STATE_UNGATE);
+	if (r)
+		return r;
+
+	return amdgpu_ring_reset_helper_end(ring, timedout_fence);
+}
+
 static ssize_t amdgpu_get_vpe_reset_mask(struct device *dev,
 						struct device_attribute *attr,
 						char *buf)
@@ -942,6 +964,7 @@ static const struct amdgpu_ring_funcs vpe_ring_funcs = {
 	.preempt_ib = vpe_ring_preempt_ib,
 	.begin_use = vpe_ring_begin_use,
 	.end_use = vpe_ring_end_use,
+	.reset = vpe_ring_reset,
 };
 
 static void vpe_set_ring_funcs(struct amdgpu_device *adev)

From 88fdd89d1ca2d5e6385ec1f562bdb34c449cab47 Mon Sep 17 00:00:00 2001
From: Karthi Kandasamy <karthi.kandasamy@amd.com>
Date: Mon, 7 Jul 2025 13:28:12 +0200
Subject: [PATCH 0674/1455] drm/amd/display: Add control flags to force PSR /
 replay

To change PSR/Replay behavior based on OS preferences, add some
config options.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Karthi Kandasamy <karthi.kandasamy@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_types.h                     | 4 ++++
 drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c                 | 2 +-
 drivers/gpu/drm/amd/display/dc/inc/core_types.h               | 1 +
 .../amd/display/dc/link/protocols/link_edp_panel_control.c    | 4 ++++
 4 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 2e2dea21b332..619834a328a3 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -628,6 +628,7 @@ struct psr_config {
 	unsigned int line_time_in_us;
 	uint8_t rate_control_caps;
 	uint16_t dsc_slice_height;
+	bool os_request_force_ffu;
 };
 
 union dmcu_psr_level {
@@ -740,6 +741,7 @@ struct psr_context {
 	unsigned int line_time_in_us;
 	uint8_t rate_control_caps;
 	uint16_t dsc_slice_height;
+	bool os_request_force_ffu;
 };
 
 struct colorspace_transform {
@@ -1151,6 +1153,8 @@ struct replay_config {
 	bool replay_video_conferencing_optimization_enabled;
 	/* Replay alpm mode */
 	enum dc_alpm_mode alpm_mode;
+	/* Replay full screen only */
+	bool os_request_force_ffu;
 };
 
 /* Replay feature flags*/
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
index ff3b8244ba3d..87af4fdc04a6 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_psr.c
@@ -391,7 +391,7 @@ static bool dmub_psr_copy_settings(struct dmub_psr *dmub,
 			sizeof(DP_SINK_DEVICE_STR_ID_1)))
 		link->psr_settings.force_ffu_mode = 1;
 
-	copy_settings_data->force_ffu_mode = link->psr_settings.force_ffu_mode;
+	copy_settings_data->force_ffu_mode = link->psr_settings.force_ffu_mode || psr_context->os_request_force_ffu;
 
 	if (((link->dpcd_caps.fec_cap.bits.FEC_CAPABLE &&
 		!link->dc->debug.disable_fec) &&
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index 4387de044469..d30f94c35f11 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -684,6 +684,7 @@ struct replay_context {
 	/* Controller Id used for Dig Fe source select */
 	enum controller_id controllerId;
 	unsigned int line_time_in_ns;
+	bool os_request_force_ffu;
 };
 
 enum dc_replay_enable {
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index 2c3e2945124a..8b7b87b21c2e 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -870,6 +870,8 @@ bool edp_setup_psr(struct dc_link *link,
 
 	psr_context->dsc_slice_height = psr_config->dsc_slice_height;
 
+	psr_context->os_request_force_ffu = psr_config->os_request_force_ffu;
+
 	if (psr) {
 		link->psr_settings.psr_feature_enabled = psr->funcs->psr_copy_settings(psr,
 			link, psr_context, panel_inst);
@@ -1029,6 +1031,8 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream
 
 	replay_context.line_time_in_ns = lineTimeInNs;
 
+	replay_context.os_request_force_ffu = link->replay_settings.config.os_request_force_ffu;
+
 	link->replay_settings.replay_feature_enabled =
 			replay->funcs->replay_copy_settings(replay, link, &replay_context, panel_inst);
 	if (link->replay_settings.replay_feature_enabled) {

From 8b715afa8f2d74e7f3be1d8b2ff23adb53986c1b Mon Sep 17 00:00:00 2001
From: Cruise Hung <Cruise.Hung@amd.com>
Date: Mon, 11 Aug 2025 21:03:13 +0800
Subject: [PATCH 0675/1455] drm/amd/display: Reserve instance index notified by
 DMUB

[Why]
Reserve instance index notified by DMUB.

[How]
Add new variable for instance index.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc_stat.c        | 2 +-
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h          | 2 ++
 drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c | 8 ++++----
 3 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stat.c b/drivers/gpu/drm/amd/display/dc/core/dc_stat.c
index fe9f99f1bdf9..f976ffd6d466 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stat.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stat.c
@@ -65,7 +65,7 @@ void dc_stat_get_dmub_notification(const struct dc *dc, struct dmub_notification
 	    notify->type == DMUB_NOTIFICATION_DPIA_NOTIFICATION ||
 	    notify->type == DMUB_NOTIFICATION_SET_CONFIG_REPLY) {
 		notify->link_index =
-			get_link_index_from_dpia_port_index(dc, notify->link_index);
+			get_link_index_from_dpia_port_index(dc, notify->instance);
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 87b761ac3135..7abf7d0fd02b 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -598,6 +598,8 @@ struct dmub_notification {
 	enum dmub_notification_type type;
 	uint8_t link_index;
 	uint8_t result;
+	/* notify instance from DMUB */
+	uint8_t instance;
 	bool pending_notification;
 	union {
 		struct aux_reply_data aux_reply;
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c
index 567c5b1aeb7a..e7a58b140388 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_srv_stat.c
@@ -71,7 +71,7 @@ enum dmub_status dmub_srv_stat_get_notification(struct dmub_srv *dmub,
 	switch (cmd.cmd_common.header.type) {
 	case DMUB_OUT_CMD__DP_AUX_REPLY:
 		notify->type = DMUB_NOTIFICATION_AUX_REPLY;
-		notify->link_index = cmd.dp_aux_reply.control.instance;
+		notify->instance = cmd.dp_aux_reply.control.instance;
 		notify->result = cmd.dp_aux_reply.control.result;
 		dmub_memcpy((void *)&notify->aux_reply,
 			(void *)&cmd.dp_aux_reply.reply_data, sizeof(struct aux_reply_data));
@@ -84,17 +84,17 @@ enum dmub_status dmub_srv_stat_get_notification(struct dmub_srv *dmub,
 			notify->type = DMUB_NOTIFICATION_HPD_IRQ;
 		}
 
-		notify->link_index = cmd.dp_hpd_notify.hpd_data.instance;
+		notify->instance = cmd.dp_hpd_notify.hpd_data.instance;
 		notify->result = AUX_RET_SUCCESS;
 		break;
 	case DMUB_OUT_CMD__SET_CONFIG_REPLY:
 		notify->type = DMUB_NOTIFICATION_SET_CONFIG_REPLY;
-		notify->link_index = cmd.set_config_reply.set_config_reply_control.instance;
+		notify->instance = cmd.set_config_reply.set_config_reply_control.instance;
 		notify->sc_status = cmd.set_config_reply.set_config_reply_control.status;
 		break;
 	case DMUB_OUT_CMD__DPIA_NOTIFICATION:
 		notify->type = DMUB_NOTIFICATION_DPIA_NOTIFICATION;
-		notify->link_index = cmd.dpia_notification.payload.header.instance;
+		notify->instance = cmd.dpia_notification.payload.header.instance;
 		break;
 	case DMUB_OUT_CMD__HPD_SENSE_NOTIFY:
 		notify->type = DMUB_NOTIFICATION_HPD_SENSE_NOTIFY;

From 90868205b6f21896a62bf99e92ad1fb55eda0e2d Mon Sep 17 00:00:00 2001
From: Ausef Yousof <Ausef.Yousof@amd.com>
Date: Wed, 30 Jul 2025 16:08:10 -0400
Subject: [PATCH 0676/1455] drm/amd/display: track dpia support

[why&how]
initialize a flag to track if we previously
supported dpia and write that to boot options

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Signed-off-by: Ausef Yousof <Ausef.Yousof@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/dmub_srv.h       | 1 +
 drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c | 7 ++++---
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
index 7abf7d0fd02b..338fdc651f2c 100644
--- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
+++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h
@@ -568,6 +568,7 @@ struct dmub_srv {
 
 	bool sw_init;
 	bool hw_init;
+	bool dpia_supported;
 
 	uint64_t fb_base;
 	uint64_t fb_offset;
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c
index 2228d62adc7e..834e5434ccb8 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn35.c
@@ -400,13 +400,14 @@ union dmub_fw_boot_options dmub_dcn35_get_fw_boot_option(struct dmub_srv *dmub)
 void dmub_dcn35_enable_dmub_boot_options(struct dmub_srv *dmub, const struct dmub_srv_hw_params *params)
 {
 	union dmub_fw_boot_options boot_options = {0};
-	union dmub_fw_boot_options cur_boot_options = {0};
 
-	cur_boot_options = dmub_dcn35_get_fw_boot_option(dmub);
+	if (!dmub->dpia_supported) {
+		dmub->dpia_supported = dmub_dcn35_get_fw_boot_option(dmub).bits.enable_dpia;
+	}
 
 	boot_options.bits.z10_disable = params->disable_z10;
 	boot_options.bits.dpia_supported = params->dpia_supported;
-	boot_options.bits.enable_dpia = cur_boot_options.bits.enable_dpia && !params->disable_dpia;
+	boot_options.bits.enable_dpia = dmub->dpia_supported && !params->disable_dpia;
 	boot_options.bits.usb4_cm_version = params->usb4_cm_version;
 	boot_options.bits.dpia_hpd_int_enable_supported = params->dpia_hpd_int_enable_supported;
 	boot_options.bits.power_optimization = params->power_optimization;

From dbe88861742ebdb277aaf8963c9bfb876436106c Mon Sep 17 00:00:00 2001
From: Reza Amini <reza.amini@amd.com>
Date: Tue, 12 Aug 2025 10:02:45 -0400
Subject: [PATCH 0677/1455] drm/amd/display: Decrease stack size in logging
 path

[why]
Reducing stack size can avoid stack over flow

[how]
Make local variables const and static so they are not
on the stack.

Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Reza Amini <reza.amini@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index 79c9bea78c47..d633033c98f2 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -328,7 +328,7 @@ static void dcn10_log_hubp_states(struct dc *dc, void *log_ctx)
 	}
 
 	DTN_INFO("\n=======HUBP FL======\n");
-	char pLabels[18][50] = {
+	static const char * const pLabels[] = {
 		"inst", "Enabled ", "Done ", "adr_mode ", "width ", "mpc_width ",
 		"tmz", "xbar_sel_R", "xbar_sel_G", "xbar_sel_B", "adr_hi ",
 		"adr_low", "REFCYC", "Bias", "Scale", "Mode",
@@ -552,7 +552,7 @@ static void dcn10_log_color_state(struct dc *dc,
 		 dc->caps.color.mpc.ogam_ram,
 		 dc->caps.color.mpc.ocsc);
 	DTN_INFO("===== MPC RMCM 3DLUT =====\n");
-	char pLabels[19][50] = {
+	static const char * const pLabels[] = {
 		"MPCC", "SIZE", "MODE", "MODE_CUR", "RD_SEL",
 		"30BIT_EN", "WR_EN_MASK", "RAM_SEL", "OUT_NORM_FACTOR", "FL_SEL",
 		"OUT_OFFSET", "OUT_SCALE", "FL_DONE", "SOFT_UNDERFLOW", "HARD_UNDERFLOW",

From 8a359f0f138d5ac7ceffd21b73279be50e516c0a Mon Sep 17 00:00:00 2001
From: Ivan Lipski <ivan.lipski@amd.com>
Date: Thu, 7 Aug 2025 09:45:26 -0400
Subject: [PATCH 0678/1455] drm/amd/display: Support HW cursor 180 rot for any
 number of pipe splits

[Why]
For the HW cursor, its current position in the pipe_ctx->stream struct is
not affected by the 180 rotation, i. e. the top left corner is still at
0,0. However, the DPP & HUBP set_cursor_position functions require rotated
position.

The current approach is hard-coded for ODM 2:1, thus it's failing for
ODM 4:1, resulting in a double cursor.

[How]
Instead of calculating the new cursor position relatively to the
viewports, we calculate it using a viewavable clip_rect of each plane.

The clip_rects are first offset and scaled to the same space as the
src_rect, i. e. Stream space -> Plane space.

In case of a pipe split, which divides the plane into 2 or more viewports,
the clip_rect is the union of all the viewports of the given plane.

With the assumption that the viewports in HUBP's set_cursor_position are
in the Plane space as well, it should produce a correct cursor position
for any number of pipe splits.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/hwss/dcn10/dcn10_hwseq.c   | 73 +++++++------------
 1 file changed, 27 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index d633033c98f2..506c3bbbf221 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -3663,6 +3663,8 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 	int y_plane = pipe_ctx->plane_state->dst_rect.y;
 	int x_pos = pos_cpy.x;
 	int y_pos = pos_cpy.y;
+	int clip_x = pipe_ctx->plane_state->clip_rect.x;
+	int clip_width = pipe_ctx->plane_state->clip_rect.width;
 
 	if ((pipe_ctx->top_pipe != NULL) || (pipe_ctx->bottom_pipe != NULL)) {
 		if ((pipe_ctx->plane_state->src_rect.width != pipe_ctx->plane_res.scl_data.viewport.width) ||
@@ -3681,7 +3683,7 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 	 */
 
 	/**
-	 * Translate cursor from stream space to plane space.
+	 * Translate cursor and clip offset from stream space to plane space.
 	 *
 	 * If the cursor is scaled then we need to scale the position
 	 * to be in the approximately correct place. We can't do anything
@@ -3698,6 +3700,10 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 				pipe_ctx->plane_state->dst_rect.width;
 		y_pos = (y_pos - y_plane) * pipe_ctx->plane_state->src_rect.height /
 				pipe_ctx->plane_state->dst_rect.height;
+		clip_x = (clip_x - x_plane) * pipe_ctx->plane_state->src_rect.width /
+				pipe_ctx->plane_state->dst_rect.width;
+		clip_width = clip_width * pipe_ctx->plane_state->src_rect.width /
+				pipe_ctx->plane_state->dst_rect.width;
 	}
 
 	/**
@@ -3744,30 +3750,18 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 
 
 	if (param.rotation == ROTATION_ANGLE_0) {
-		int viewport_width =
-			pipe_ctx->plane_res.scl_data.viewport.width;
-		int viewport_x =
-			pipe_ctx->plane_res.scl_data.viewport.x;
 
 		if (param.mirror) {
-			if (pipe_split_on || odm_combine_on) {
-				if (pos_cpy.x >= viewport_width + viewport_x) {
-					pos_cpy.x = 2 * viewport_width
-							- pos_cpy.x + 2 * viewport_x;
-				} else {
-					uint32_t temp_x = pos_cpy.x;
-
-					pos_cpy.x = 2 * viewport_x - pos_cpy.x;
-					if (temp_x >= viewport_x +
-						(int)hubp->curs_attr.width || pos_cpy.x
-						<= (int)hubp->curs_attr.width +
-						pipe_ctx->plane_state->src_rect.x) {
-						pos_cpy.x = 2 * viewport_width - temp_x;
-					}
-				}
-			} else {
-				pos_cpy.x = viewport_width - pos_cpy.x + 2 * viewport_x;
-			}
+			/*
+			 * The plane is split into multiple viewports.
+			 * The combination of all viewports span the
+			 * entirety of the clip rect.
+			 *
+			 * For no pipe_split, viewport_width is represents
+			 * the full width of the clip_rect, so we can just
+			 * mirror it.
+			 */
+			pos_cpy.x = clip_width - pos_cpy.x + 2 * clip_x;
 		}
 	}
 	// Swap axis and mirror horizontally
@@ -3837,30 +3831,17 @@ void dcn10_set_cursor_position(struct pipe_ctx *pipe_ctx)
 	}
 	// Mirror horizontally and vertically
 	else if (param.rotation == ROTATION_ANGLE_180) {
-		int viewport_width =
-			pipe_ctx->plane_res.scl_data.viewport.width;
-		int viewport_x =
-			pipe_ctx->plane_res.scl_data.viewport.x;
-
 		if (!param.mirror) {
-			if (pipe_split_on || odm_combine_on) {
-				if (pos_cpy.x >= viewport_width + viewport_x) {
-					pos_cpy.x = 2 * viewport_width
-							- pos_cpy.x + 2 * viewport_x;
-				} else {
-					uint32_t temp_x = pos_cpy.x;
-
-					pos_cpy.x = 2 * viewport_x - pos_cpy.x;
-					if (temp_x >= viewport_x +
-						(int)hubp->curs_attr.width || pos_cpy.x
-						<= (int)hubp->curs_attr.width +
-						pipe_ctx->plane_state->src_rect.x) {
-						pos_cpy.x = temp_x + viewport_width;
-					}
-				}
-			} else {
-				pos_cpy.x = viewport_width - pos_cpy.x + 2 * viewport_x;
-			}
+			/*
+			 * The plane is split into multiple viewports.
+			 * The combination of all viewports span the
+			 * entirety of the clip rect.
+			 *
+			 * For no pipe_split, viewport_width is represents
+			 * the full width of the clip_rect, so we can just
+			 * mirror it.
+			 */
+			pos_cpy.x = clip_width - pos_cpy.x + 2 * clip_x;
 		}
 
 		/**

From 9ec77e3a4b8913508bfd10891ce339af05e671ef Mon Sep 17 00:00:00 2001
From: Clay King <clayking@amd.com>
Date: Thu, 14 Aug 2025 16:58:15 -0400
Subject: [PATCH 0679/1455] drm/amd/display: Multiplication result converted to
 larger type

Consolidating multiple CodeQL Fixes for alerts with rule id: cpp/integer-multiplication-cast-to-long

Reviewed-by: Joshua Aberback <joshua.aberback@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c    |  2 +-
 drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c   |  2 +-
 drivers/gpu/drm/amd/display/dc/basics/vector.c       |  6 +++---
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c |  2 +-
 .../amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c   |  2 +-
 .../amd/display/dc/resource/dce112/dce112_resource.c | 12 ++++++------
 .../amd/display/dc/resource/dce120/dce120_resource.c |  6 +++---
 7 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c b/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c
index d897f8a30ede..4da5adab799c 100644
--- a/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/basics/dce_calcs.c
@@ -1136,7 +1136,7 @@ static void calculate_bandwidth(
 			}
 		}
 	}
-	data->total_dmifmc_urgent_trips = bw_ceil2(bw_div(data->total_requests_for_adjusted_dmif_size, (bw_add(dceip->dmif_request_buffer_size, bw_int_to_fixed(vbios->number_of_request_slots_gmc_reserves_for_dmif_per_channel * data->number_of_dram_channels)))), bw_int_to_fixed(1));
+	data->total_dmifmc_urgent_trips = bw_ceil2(bw_div(data->total_requests_for_adjusted_dmif_size, (bw_add(dceip->dmif_request_buffer_size, bw_int_to_fixed((uint64_t)vbios->number_of_request_slots_gmc_reserves_for_dmif_per_channel * data->number_of_dram_channels)))), bw_int_to_fixed(1));
 	data->total_dmifmc_urgent_latency = bw_mul(vbios->dmifmc_urgent_latency, data->total_dmifmc_urgent_trips);
 	data->total_display_reads_required_data = bw_int_to_fixed(0);
 	data->total_display_reads_required_dram_access_data = bw_int_to_fixed(0);
diff --git a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
index 452206b5095e..6073cadde76c 100644
--- a/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
+++ b/drivers/gpu/drm/amd/display/dc/basics/fixpt31_32.c
@@ -284,7 +284,7 @@ struct fixed31_32 dc_fixpt_cos(struct fixed31_32 arg)
 				dc_fixpt_mul(
 					square,
 					res),
-				n * (n - 1)));
+				(long long)n * (n - 1)));
 
 		n -= 2;
 	} while (n != 0);
diff --git a/drivers/gpu/drm/amd/display/dc/basics/vector.c b/drivers/gpu/drm/amd/display/dc/basics/vector.c
index 6d2924114a3e..b413a672c2c0 100644
--- a/drivers/gpu/drm/amd/display/dc/basics/vector.c
+++ b/drivers/gpu/drm/amd/display/dc/basics/vector.c
@@ -170,7 +170,7 @@ bool dal_vector_remove_at_index(
 		memmove(
 			vector->container + (index * vector->struct_size),
 			vector->container + ((index + 1) * vector->struct_size),
-			(vector->count - index - 1) * vector->struct_size);
+			(size_t)(vector->count - index - 1) * vector->struct_size);
 	vector->count -= 1;
 
 	return true;
@@ -219,7 +219,7 @@ bool dal_vector_insert_at(
 		memmove(
 			insert_address + vector->struct_size,
 			insert_address,
-			vector->struct_size * (vector->count - position));
+			(size_t)vector->struct_size * (vector->count - position));
 
 	memmove(
 		insert_address,
@@ -271,7 +271,7 @@ struct vector *dal_vector_clone(
 
 	/* copy vector's data */
 	memmove(vec_cloned->container, vector->container,
-			vec_cloned->struct_size * vec_cloned->capacity);
+			(size_t)vec_cloned->struct_size * vec_cloned->capacity);
 
 	return vec_cloned;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 6160952245b4..7cd7bddea423 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -3401,7 +3401,7 @@ bool dcn32_allow_subvp_with_active_margin(struct pipe_ctx *pipe)
 		uint32_t height = subvp_active_margin_list.res[i].height;
 
 		refresh_rate = (pipe->stream->timing.pix_clk_100hz * (uint64_t)100 +
-			pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1);
+			(uint64_t)pipe->stream->timing.v_total * pipe->stream->timing.h_total - (uint64_t)1);
 		refresh_rate = div_u64(refresh_rate, pipe->stream->timing.v_total);
 		refresh_rate = div_u64(refresh_rate, pipe->stream->timing.h_total);
 
diff --git a/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c b/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c
index 259a98e4ee2c..2a422e223bf2 100644
--- a/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/mmhubbub/dcn20/dcn20_mmhubbub.c
@@ -284,7 +284,7 @@ void mcifwb2_dump_frame(struct mcif_wb *mcif_wb,
 
 	REG_UPDATE(MCIF_WB_BUFMGR_SW_CONTROL, MCIF_WB_BUFMGR_SW_LOCK, 0xf);
 
-	memcpy(dest_luma_buffer,   luma_buffer,   mcif_params->luma_pitch * dest_height);
+	memcpy(dest_luma_buffer,   luma_buffer,   (size_t)mcif_params->luma_pitch * dest_height);
 	memcpy(dest_chroma_buffer, chroma_buffer, mcif_params->chroma_pitch * dest_height / 2);
 
 	REG_UPDATE(MCIF_WB_BUFMGR_SW_CONTROL, MCIF_WB_BUFMGR_SW_LOCK, 0x0);
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
index 164ba796f64c..869a8e515fc0 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c
@@ -1111,12 +1111,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
 				&clks);
 
 		dc->bw_vbios->low_yclk = bw_frc_to_fixed(
-			clks.clocks_in_khz[0] * memory_type_multiplier, 1000);
+			(int64_t)clks.clocks_in_khz[0] * memory_type_multiplier, 1000);
 		dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
-			clks.clocks_in_khz[clks.num_levels>>1] * memory_type_multiplier,
+			(int64_t)clks.clocks_in_khz[clks.num_levels>>1] * memory_type_multiplier,
 			1000);
 		dc->bw_vbios->high_yclk = bw_frc_to_fixed(
-			clks.clocks_in_khz[clks.num_levels-1] * memory_type_multiplier,
+			(int64_t)clks.clocks_in_khz[clks.num_levels-1] * memory_type_multiplier,
 			1000);
 
 		return;
@@ -1152,12 +1152,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
 	 * YCLK = UMACLK*m_memoryTypeMultiplier
 	 */
 	dc->bw_vbios->low_yclk = bw_frc_to_fixed(
-		mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000);
+		(int64_t)mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000);
 	dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
-		mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier,
+		(int64_t)mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier,
 		1000);
 	dc->bw_vbios->high_yclk = bw_frc_to_fixed(
-		mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier,
+		(int64_t)mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier,
 		1000);
 
 	/* Now notify PPLib/SMU about which Watermarks sets they should select
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
index eb1e158d3436..2f23cc6df571 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
@@ -990,12 +990,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
 		memory_type_multiplier = MEMORY_TYPE_HBM;
 
 	dc->bw_vbios->low_yclk = bw_frc_to_fixed(
-		mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000);
+		(int64_t)mem_clks.data[0].clocks_in_khz * memory_type_multiplier, 1000);
 	dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
-		mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier,
+		(int64_t)mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * memory_type_multiplier,
 		1000);
 	dc->bw_vbios->high_yclk = bw_frc_to_fixed(
-		mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier,
+		(int64_t)mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * memory_type_multiplier,
 		1000);
 
 	/* Now notify PPLib/SMU about which Watermarks sets they should select

From 048ce48c3dc9ddb0026b750001283ecbb2828bde Mon Sep 17 00:00:00 2001
From: Clay King <clayking@amd.com>
Date: Thu, 14 Aug 2025 17:02:45 -0400
Subject: [PATCH 0680/1455] drm/amd/display: Incorrect 'not' operator usage

Consolidating multiple CodeQL Fixes for alerts with rule id: cpp/incorrect-not-operator-usage

Reviewed-by: Joshua Aberback <joshua.aberback@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c                 | 2 +-
 drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c                 | 2 +-
 .../gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c   | 2 +-
 .../gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c | 2 +-
 .../gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c   | 2 +-
 .../gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c   | 2 +-
 .../gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c   | 2 +-
 .../gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c | 2 +-
 8 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c
index 0421b267a0b5..365dd2e37aea 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_hw.c
@@ -591,7 +591,7 @@ static bool dce_i2c_hw_engine_submit_payload(struct dce_i2c_hw *dce_i2c_hw,
 			DCE_I2C_TRANSACTION_ACTION_I2C_WRITE;
 
 
-	request.address = (uint8_t) ((payload->address << 1) | !payload->write);
+	request.address = (uint8_t) ((payload->address << 1) | (payload->write ? 0 : 1));
 	request.length = payload->length;
 	request.data = payload->data;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c
index e188447c8156..2d73b94c515c 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_i2c_sw.c
@@ -451,7 +451,7 @@ static bool dce_i2c_sw_engine_submit_payload(struct dce_i2c_sw *engine,
 			DCE_I2C_TRANSACTION_ACTION_I2C_WRITE_MOT :
 			DCE_I2C_TRANSACTION_ACTION_I2C_WRITE;
 
-	request.address = (uint8_t) ((payload->address << 1) | !payload->write);
+	request.address = (uint8_t) ((payload->address << 1) | (payload->write ? 0 : 1));
 	request.length = payload->length;
 	request.data = payload->data;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c
index 390c1a77fda6..9c58ff1069d6 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c
@@ -646,7 +646,7 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib,
 
 	// the dpte_group_bytes is reduced for the specific case of vertical
 	// access of a tile surface that has dpte request of 8x1 ptes.
-	if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group
+	if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group
 		rq_sizing_param->dpte_group_bytes = 512;
 	else
 		//full size
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c
index 843d6004258c..570e6e39eb45 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c
@@ -646,7 +646,7 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib,
 
 	// the dpte_group_bytes is reduced for the specific case of vertical
 	// access of a tile surface that has dpte request of 8x1 ptes.
-	if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group
+	if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group
 		rq_sizing_param->dpte_group_bytes = 512;
 	else
 		//full size
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c
index 5718000627b0..f549da082c01 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c
@@ -652,7 +652,7 @@ static void get_meta_and_pte_attr(
 	if (hostvm_enable)
 		rq_sizing_param->dpte_group_bytes = 512;
 	else {
-		if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group
+		if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group
 			rq_sizing_param->dpte_group_bytes = 512;
 		else
 			//full size
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
index 8d4873f80df0..4fb37df54d59 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c
@@ -620,7 +620,7 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib,
 	if (hostvm_enable)
 		rq_sizing_param->dpte_group_bytes = 512;
 	else {
-		if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group
+		if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group
 			rq_sizing_param->dpte_group_bytes = 512;
 		else
 			rq_sizing_param->dpte_group_bytes = 2048;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
index c46bda2141ac..bfeb01477f0c 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c
@@ -615,7 +615,7 @@ static void get_meta_and_pte_attr(
 	if (hostvm_enable)
 		rq_sizing_param->dpte_group_bytes = 512;
 	else {
-		if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group
+		if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group
 			rq_sizing_param->dpte_group_bytes = 512;
 		else
 			rq_sizing_param->dpte_group_bytes = 2048;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
index b7d2a0caec11..04df263ff65e 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c
@@ -703,7 +703,7 @@ static void get_meta_and_pte_attr(
 	if (hostvm_enable)
 		rq_sizing_param->dpte_group_bytes = 512;
 	else {
-		if (!surf_linear & (log2_dpte_req_height_ptes == 0) & surf_vert) //reduced, in this case, will have page fault within a group
+		if (!surf_linear && (log2_dpte_req_height_ptes == 0) && surf_vert) //reduced, in this case, will have page fault within a group
 			rq_sizing_param->dpte_group_bytes = 512;
 		else
 			rq_sizing_param->dpte_group_bytes = 2048;

From f382e2d0faad0e0d73f626dbd71f2a4fce03975b Mon Sep 17 00:00:00 2001
From: Yihan Zhu <Yihan.Zhu@amd.com>
Date: Wed, 13 Aug 2025 16:05:10 -0400
Subject: [PATCH 0681/1455] drm/amd/display: wait for otg update pending latch
 before clock optimization

[WHY & HOW]
OTG pending update unlatched will cause system fail, wait OTG fully disabled to
avoid this error.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Yihan Zhu <Yihan.Zhu@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/core/dc_hw_sequencer.c  |  2 ++
 .../amd/display/dc/inc/hw/timing_generator.h   |  1 +
 .../drm/amd/display/dc/optc/dcn32/dcn32_optc.h |  1 +
 .../drm/amd/display/dc/optc/dcn35/dcn35_optc.c | 18 ++++++++++++++++++
 4 files changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
index ec4e80e5b6eb..d82b1cb467f4 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c
@@ -1177,6 +1177,8 @@ void hwss_wait_for_odm_update_pending_complete(struct dc *dc, struct dc_state *c
 		tg = otg_master->stream_res.tg;
 		if (tg->funcs->wait_odm_doublebuffer_pending_clear)
 			tg->funcs->wait_odm_doublebuffer_pending_clear(tg);
+		if (tg->funcs->wait_otg_disable)
+			tg->funcs->wait_otg_disable(tg);
 	}
 
 	/* ODM update may require to reprogram blank pattern for each OPP */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
index 267ace4eef8a..f2de2cf23859 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h
@@ -374,6 +374,7 @@ struct timing_generator_funcs {
 	void (*wait_drr_doublebuffer_pending_clear)(struct timing_generator *tg);
 	void (*set_long_vtotal)(struct timing_generator *optc, const struct long_vtotal_params *params);
 	void (*wait_odm_doublebuffer_pending_clear)(struct timing_generator *tg);
+	void (*wait_otg_disable)(struct timing_generator *optc);
 	bool (*get_optc_double_buffer_pending)(struct timing_generator *tg);
 	bool (*get_otg_double_buffer_pending)(struct timing_generator *tg);
 	bool (*get_pipe_update_pending)(struct timing_generator *tg);
diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h
index d159e3ed3bb3..ead92ad78a23 100644
--- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h
+++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h
@@ -62,6 +62,7 @@
 	SF(OTG0_OTG_CONTROL, OTG_DISABLE_POINT_CNTL, mask_sh),\
 	SF(OTG0_OTG_CONTROL, OTG_FIELD_NUMBER_CNTL, mask_sh),\
 	SF(OTG0_OTG_CONTROL, OTG_OUT_MUX, mask_sh),\
+	SF(OTG0_OTG_CONTROL, OTG_CURRENT_MASTER_EN_STATE, mask_sh),\
 	SF(OTG0_OTG_STEREO_CONTROL, OTG_STEREO_EN, mask_sh),\
 	SF(OTG0_OTG_STEREO_CONTROL, OTG_STEREO_SYNC_OUTPUT_LINE_NUM, mask_sh),\
 	SF(OTG0_OTG_STEREO_CONTROL, OTG_STEREO_SYNC_OUTPUT_POLARITY, mask_sh),\
diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c
index 72bff94cb57d..52d5ea98c86b 100644
--- a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.c
@@ -162,6 +162,8 @@ static bool optc35_disable_crtc(struct timing_generator *optc)
 	REG_WAIT(OTG_CLOCK_CONTROL,
 			OTG_BUSY, 0,
 			1, 100000);
+	REG_WAIT(OTG_CONTROL, OTG_CURRENT_MASTER_EN_STATE, 0, 1, 100000);
+
 	optc1_clear_optc_underflow(optc);
 
 	return true;
@@ -428,6 +430,21 @@ static void optc35_set_long_vtotal(
 	}
 }
 
+static void optc35_wait_otg_disable(struct timing_generator *optc)
+{
+	struct optc *optc1;
+	uint32_t is_master_en;
+
+	if (!optc || !optc->ctx)
+		return;
+
+	optc1 = DCN10TG_FROM_TG(optc);
+
+	REG_GET(OTG_CONTROL, OTG_MASTER_EN, &is_master_en);
+	if (!is_master_en)
+		REG_WAIT(OTG_CLOCK_CONTROL, OTG_CURRENT_MASTER_EN_STATE, 0, 1, 100000);
+}
+
 static const struct timing_generator_funcs dcn35_tg_funcs = {
 		.validate_timing = optc1_validate_timing,
 		.program_timing = optc1_program_timing,
@@ -479,6 +496,7 @@ static const struct timing_generator_funcs dcn35_tg_funcs = {
 		.set_odm_bypass = optc32_set_odm_bypass,
 		.set_odm_combine = optc35_set_odm_combine,
 		.get_optc_source = optc2_get_optc_source,
+		.wait_otg_disable = optc35_wait_otg_disable,
 		.set_h_timing_div_manual_mode = optc32_set_h_timing_div_manual_mode,
 		.set_out_mux = optc3_set_out_mux,
 		.set_drr_trigger_window = optc3_set_drr_trigger_window,

From bdbb6a14db574677254cb7846a3774f5fc57e549 Mon Sep 17 00:00:00 2001
From: Clay King <clayking@amd.com>
Date: Thu, 14 Aug 2025 17:01:04 -0400
Subject: [PATCH 0682/1455] drm/amd/display: Array offset used before range
 check

Consolidating multiple CodeQL Fixes for alerts with rule id: cpp/offset-use-before-range-check

Reviewed-by: Joshua Aberback <joshua.aberback@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c         | 2 +-
 drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c         | 2 +-
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c           | 2 +-
 drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c         | 2 +-
 drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c | 2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c
index 8da97a96b1ce..8d7c59ec701d 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn302/dcn302_fpu.c
@@ -280,7 +280,7 @@ void dcn302_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p
 		j = 0;
 		/* create the final dcfclk and uclk table */
 		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
-			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
 				dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
 				dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
 			} else {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c
index e968870a4b81..b5d3fd4c3694 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn303/dcn303_fpu.c
@@ -285,7 +285,7 @@ void dcn303_fpu_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_p
 		j = 0;
 		/* create the final dcfclk and uclk table */
 		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
-			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
 				dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
 				dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
 			} else {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 7cd7bddea423..18388fb00be8 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -3229,7 +3229,7 @@ void dcn32_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_pa
 			j = 0;
 			// create the final dcfclk and uclk table
 			while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
-				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
+				if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
 					dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
 					dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
 				} else {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
index 8839faf42207..e0a1dc89ce43 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn321/dcn321_fpu.c
@@ -779,7 +779,7 @@ void dcn321_update_bw_bounding_box_fpu(struct dc *dc, struct clk_bw_params *bw_p
 		j = 0;
 		// create the final dcfclk and uclk table
 		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
-			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
 				dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
 				dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
 			} else {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
index 895349d9ca07..201ed863b69e 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
@@ -2192,7 +2192,7 @@ void dcn30_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params
 		j = 0;
 		// create the final dcfclk and uclk table
 		while (i < num_dcfclk_sta_targets && j < num_uclk_states && num_states < DC__VOLTAGE_STATES) {
-			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j] && i < num_dcfclk_sta_targets) {
+			if (dcfclk_sta_targets[i] < optimal_dcfclk_for_uclk[j]) {
 				dcfclk_mhz[num_states] = dcfclk_sta_targets[i];
 				dram_speed_mts[num_states++] = optimal_uclk_for_dcfclk_sta_targets[i++];
 			} else {

From 6b34e7ed4ba583ee77032a4c850ff97ba16ad870 Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Thu, 14 Aug 2025 12:01:15 -0400
Subject: [PATCH 0683/1455] drm/amd/display: Consider sink max slice width
 limitation for dsc

[WHY&HOW]
The sink max slice width limitation should be considered for DSC, but
was removed in "refactor DSC cap calculations".
This patch adds it back and takes the valid minimum between the sink and
source.

Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
index 1f53a9f0c0ac..e4144b244332 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dc_dsc.c
@@ -1157,6 +1157,11 @@ static bool setup_dsc_config(
 	if (!is_dsc_possible)
 		goto done;
 
+	/* increase miniumum slice count to meet sink slice width limitations */
+	min_slices_h = dc_fixpt_ceil(dc_fixpt_max(
+			dc_fixpt_div_int(dc_fixpt_from_int(pic_width), dsc_common_caps.max_slice_width), // sink min
+			dc_fixpt_from_int(min_slices_h))); // source min
+
 	min_slices_h = fit_num_slices_up(dsc_common_caps.slice_caps, min_slices_h);
 
 	/* increase minimum slice count to meet sink throughput limitations */

From be33e8a239aac204d7e9e673c4220ef244eb1ba3 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 25 Aug 2025 13:40:22 -0400
Subject: [PATCH 0684/1455] Revert "drm/amdgpu: fix incorrect vm flags to map
 bo"

This reverts commit b08425fa77ad2f305fe57a33dceb456be03b653f.

Revert this to align with 6.17 because the fixes tag
was wrong on this commit.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
index dfb6cfd83760..02138aa55793 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -88,8 +88,8 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	}
 
 	r = amdgpu_vm_bo_map(adev, *bo_va, csa_addr, 0, size,
-			     AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE |
-			     AMDGPU_VM_PAGE_EXECUTABLE);
+			     AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
+			     AMDGPU_PTE_EXECUTABLE);
 
 	if (r) {
 		DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r);

From c350d9e2684f5ee9e753b825cf4bc3fbac8fe865 Mon Sep 17 00:00:00 2001
From: Jack Xiao <Jack.Xiao@amd.com>
Date: Mon, 25 Aug 2025 13:41:23 -0400
Subject: [PATCH 0685/1455] Reapply "drm/amdgpu: fix incorrect vm flags to map
 bo"

It should use vm flags instead of pte flags
to specify bo vm attributes.

This reverts commit 1263ceea2a1327014d9de2858a122f3c27dfa4dd.

Reapply this patch with the proper fixes tag.

Fixes: 6716a823d18d ("drm/amdgpu: rework how PTE flags are generated v3")
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Likun Gao <Likun.Gao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
index 02138aa55793..dfb6cfd83760 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -88,8 +88,8 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	}
 
 	r = amdgpu_vm_bo_map(adev, *bo_va, csa_addr, 0, size,
-			     AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
-			     AMDGPU_PTE_EXECUTABLE);
+			     AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE |
+			     AMDGPU_VM_PAGE_EXECUTABLE);
 
 	if (r) {
 		DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r);

From 3dc9c55d9c075b1b1337a2b6234d4c7a99a482ab Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 15 Aug 2025 19:23:50 -0400
Subject: [PATCH 0686/1455] drm/amd/display: [FW Promotion] Release 0.1.24.0

Add two new IPS residency data modes.

Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index d7008d84c1ec..b7d49a117fa7 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -6021,6 +6021,8 @@ enum ips_residency_mode {
 	IPS_RESIDENCY__IPS1_RCG,
 	IPS_RESIDENCY__IPS1_ONO2_ON,
 	IPS_RESIDENCY__IPS1_Z8_RETENTION,
+	IPS_RESIDENCY__PG_ONO_LAST_SEEN_IN_IPS,
+	IPS_RESIDENCY__PG_ONO_CURRENT_STATE
 };
 
 #define NUM_IPS_HISTOGRAM_BUCKETS 16
@@ -6034,6 +6036,8 @@ struct dmub_ips_residency_info {
 	uint32_t histogram[NUM_IPS_HISTOGRAM_BUCKETS];
 	uint64_t total_time_us;
 	uint64_t total_inactive_time_us;
+	uint32_t ono_pg_state_at_collection;
+	uint32_t ono_pg_state_last_seen_in_ips;
 };
 
 /**

From 268da6fc8d89f322358750922b626c8fda33043b Mon Sep 17 00:00:00 2001
From: Nicholas Carbones <ncarbone@amd.com>
Date: Fri, 15 Aug 2025 20:45:14 -0400
Subject: [PATCH 0687/1455] drm/amd/display: DC v3.2.347

DC Release v3.2.347

 * Firmware releases for multiple asics
 * CodeQL fixes
 * Fix for double cursor with 180 degree rotation on large resolutions
 * Misc bug fixes for DSC, PSR/Replay, DPIA etc.

Signed-off-by: Nicholas Carbones <ncarbone@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Leo Li <sunpeng.li@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 3431dee6389d..274ab744a331 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.346"
+#define DC_VER "3.2.347"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From a7a411e2462675327e07dd35f226c48d236bdbb3 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Sat, 23 Aug 2025 15:00:06 +0800
Subject: [PATCH 0688/1455] drm/amdgpu: fix shift-out-of-bounds in
 amdgpu_debugfs_jpeg_sched_mask_set

Fix a UBSAN shift-out-of-bounds warning in amdgpu_debugfs_jpeg_sched_mask_set
when the shift exponent reaches or exceeds 32 bits. The issue occurred because
a 32-bit integer '1' was being shifted by up to 32 bits, which is undefined
behavior.

Replace '1' with '1ULL' to ensure 64-bit arithmetic, matching the u64 type of
'val' and preventing the shift overflow. This is consistent with the existing
mask calculation that already uses 1ULL.

The error manifested as:
UBSAN: shift-out-of-bounds in drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c:373:17
shift exponent 32 is too large for 32-bit type 'int'
v2: remove debug log

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index 5d5e9ee83a5d..f0d7e2487237 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -370,7 +370,7 @@ static int amdgpu_debugfs_jpeg_sched_mask_set(void *data, u64 val)
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
 		for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) {
 			ring = &adev->jpeg.inst[i].ring_dec[j];
-			if (val & (1 << ((i * adev->jpeg.num_jpeg_rings) + j)))
+			if (val & (BIT_ULL(1) << ((i * adev->jpeg.num_jpeg_rings) + j)))
 				ring->sched.ready = true;
 			else
 				ring->sched.ready = false;

From e86a411b4293a1db58954ce7c99db9903d530b9f Mon Sep 17 00:00:00 2001
From: Shaoyun Liu <shaoyun.liu@amd.com>
Date: Fri, 4 Jul 2025 12:30:10 -0400
Subject: [PATCH 0689/1455] drm/amd/include : Update MES v12 API
 header(INV_TLBS)

The requirement from driver side is to have an API that can do the
tlb invalidation on dedicate pasid since driver don't know the vmid
and process mapping.
Make the API generic to support different tlb invalidation related
request. Driver can specify pasid, vmid, hub_id and vm address range
need to be invalidated.
With this API the old INV_GART in MISC Op can be deprecated.

Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/include/mes_v12_api_def.h | 30 +++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/drivers/gpu/drm/amd/include/mes_v12_api_def.h b/drivers/gpu/drm/amd/include/mes_v12_api_def.h
index d85ffab2aff9..c04bd351b250 100644
--- a/drivers/gpu/drm/amd/include/mes_v12_api_def.h
+++ b/drivers/gpu/drm/amd/include/mes_v12_api_def.h
@@ -66,6 +66,7 @@ enum MES_SCH_API_OPCODE {
 	MES_SCH_API_SET_SE_MODE			= 17,
 	MES_SCH_API_SET_GANG_SUBMIT		= 18,
 	MES_SCH_API_SET_HW_RSRC_1               = 19,
+	MES_SCH_API_INV_TLBS                    = 20,
 
 	MES_SCH_API_MAX = 0xFF
 };
@@ -870,6 +871,35 @@ union MESAPI__SET_GANG_SUBMIT {
 	uint32_t max_dwords_in_api[API_FRAME_SIZE_IN_DWORDS];
 };
 
+/*
+ * @inv_sel        0-select pasid as input to do the invalidation , 1-select vmid
+ * @flush_type     0-old style, 1-light weight, 2-heavyweight, 3-heavyweight2
+ * @inv_sel_id     specific pasid when inv_sel is 0 and specific vmid if inv_sel is 1
+ * @hub_id         0-gc_hub, 1-mm_hub
+ */
+struct INV_TLBS {
+	uint8_t     inv_sel;
+	uint8_t     flush_type;
+	uint16_t    inv_sel_id;
+	uint32_t    hub_id;
+	/* If following two inv_range setting are all 0 , whole VM will be invalidated,
+	 * otherwise only required range be invalidated
+	 */
+	uint64_t    inv_range_va_start;
+	uint64_t    inv_range_size;
+	uint64_t    reserved;
+};
+
+union MESAPI__INV_TLBS {
+	struct {
+		union MES_API_HEADER    header;
+		struct MES_API_STATUS   api_status;
+		struct INV_TLBS         invalidate_tlbs;
+	};
+
+	uint32_t max_dwords_in_api[API_FRAME_SIZE_IN_DWORDS];
+};
+
 #pragma pack(pop)
 
 #endif

From 87e65052616c53676a4f9ae3def290fc489215e2 Mon Sep 17 00:00:00 2001
From: Shaoyun Liu <shaoyun.liu@amd.com>
Date: Thu, 10 Jul 2025 21:42:16 -0400
Subject: [PATCH 0690/1455] drm/amd/amdgpu : Use the MES INV_TLBS API for tlb
 invalidation on gfx12

From MES version 0x81, it provide the new API INV_TLBS that support
invalidate tlbs with PASID.

Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h | 10 ++++++
 drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c  | 16 ++++++++++
 drivers/gpu/drm/amd/amdgpu/mes_v12_0.c  | 42 +++++++++++++++++++++++++
 3 files changed, 68 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index c0d2c195fe2e..489a4a0f0610 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -280,6 +280,13 @@ struct mes_reset_queue_input {
 	bool                               is_kq;
 };
 
+struct mes_inv_tlbs_pasid_input {
+	uint32_t        xcc_id;
+	uint16_t        pasid;
+	uint8_t         hub_id;
+	uint8_t         flush_type;
+};
+
 enum mes_misc_opcode {
 	MES_MISC_OP_WRITE_REG,
 	MES_MISC_OP_READ_REG,
@@ -367,6 +374,9 @@ struct amdgpu_mes_funcs {
 
 	int (*reset_hw_queue)(struct amdgpu_mes *mes,
 			      struct mes_reset_queue_input *input);
+
+	int (*invalidate_tlbs_pasid)(struct amdgpu_mes *mes,
+			      struct mes_inv_tlbs_pasid_input *input);
 };
 
 #define amdgpu_mes_kiq_hw_init(adev) (adev)->mes.kiq_hw_init((adev))
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
index a0a5367f9dc4..9ba055ddc00f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
@@ -336,6 +336,22 @@ static void gmc_v12_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	uint16_t queried;
 	int vmid, i;
 
+	if (adev->enable_uni_mes && adev->mes.ring[AMDGPU_MES_SCHED_PIPE].sched.ready &&
+	    (adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x81) {
+		struct mes_inv_tlbs_pasid_input input = {0};
+		input.pasid = pasid;
+		input.flush_type = flush_type;
+		input.hub_id = AMDGPU_GFXHUB(0);
+		/* MES will invalidate all gc_hub for the device from master */
+		adev->mes.funcs->invalidate_tlbs_pasid(&adev->mes, &input);
+		if (all_hub) {
+			/* Only need to invalidate mm_hub now, gfx12 only support one mmhub */
+			input.hub_id = AMDGPU_MMHUB0(0);
+			adev->mes.funcs->invalidate_tlbs_pasid(&adev->mes, &input);
+		}
+		return;
+	}
+
 	for (vmid = 1; vmid < 16; vmid++) {
 		bool valid;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index 6b222630f3fa..cd5c966cee95 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -108,6 +108,7 @@ static const char *mes_v12_0_opcodes[] = {
 	"SET_SE_MODE",
 	"SET_GANG_SUBMIT",
 	"SET_HW_RSRC_1",
+	"INVALIDATE_TLBS",
 };
 
 static const char *mes_v12_0_misc_opcodes[] = {
@@ -879,6 +880,46 @@ static int mes_v12_0_reset_hw_queue(struct amdgpu_mes *mes,
 			offsetof(union MESAPI__RESET, api_status));
 }
 
+static int mes_v12_inv_tlb_convert_hub_id(uint8_t id)
+{
+	/*
+	 * MES doesn't support invalidate gc_hub on slave xcc individually
+	 * master xcc will invalidate all gc_hub for the partition
+	 */
+	if (AMDGPU_IS_GFXHUB(id))
+		return 0;
+	else if (AMDGPU_IS_MMHUB0(id))
+		return 1;
+	else
+		return -EINVAL;
+
+}
+
+static int mes_v12_0_inv_tlbs_pasid(struct amdgpu_mes *mes,
+				    struct mes_inv_tlbs_pasid_input *input)
+{
+	union MESAPI__INV_TLBS mes_inv_tlbs;
+
+	memset(&mes_inv_tlbs, 0, sizeof(mes_inv_tlbs));
+
+	mes_inv_tlbs.header.type = MES_API_TYPE_SCHEDULER;
+	mes_inv_tlbs.header.opcode = MES_SCH_API_INV_TLBS;
+	mes_inv_tlbs.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
+
+	mes_inv_tlbs.invalidate_tlbs.inv_sel = 0;
+	mes_inv_tlbs.invalidate_tlbs.flush_type = input->flush_type;
+	mes_inv_tlbs.invalidate_tlbs.inv_sel_id = input->pasid;
+
+	/*convert amdgpu_mes_hub_id to mes expected hub_id */
+	mes_inv_tlbs.invalidate_tlbs.hub_id = mes_v12_inv_tlb_convert_hub_id(input->hub_id);
+	if (mes_inv_tlbs.invalidate_tlbs.hub_id < 0)
+		return -EINVAL;
+	return mes_v12_0_submit_pkt_and_poll_completion(mes, AMDGPU_MES_KIQ_PIPE,
+			&mes_inv_tlbs, sizeof(mes_inv_tlbs),
+			offsetof(union MESAPI__INV_TLBS, api_status));
+
+}
+
 static const struct amdgpu_mes_funcs mes_v12_0_funcs = {
 	.add_hw_queue = mes_v12_0_add_hw_queue,
 	.remove_hw_queue = mes_v12_0_remove_hw_queue,
@@ -888,6 +929,7 @@ static const struct amdgpu_mes_funcs mes_v12_0_funcs = {
 	.resume_gang = mes_v12_0_resume_gang,
 	.misc_op = mes_v12_0_misc_op,
 	.reset_hw_queue = mes_v12_0_reset_hw_queue,
+	.invalidate_tlbs_pasid = mes_v12_0_inv_tlbs_pasid,
 };
 
 static int mes_v12_0_allocate_ucode_buffer(struct amdgpu_device *adev,

From 061a09b4dcc4ebbc8ab8b648f7da697962aae68a Mon Sep 17 00:00:00 2001
From: "David (Ming Qiang) Wu" <David.Wu3@amd.com>
Date: Mon, 25 Aug 2025 15:12:18 -0400
Subject: [PATCH 0691/1455] drm/amdgpu/vcn: remove unused code in vcn_v1_0.c

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index c74947705d77..1e89ba153d9d 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -1338,7 +1338,6 @@ static int vcn_v1_0_pause_dpg_mode(struct amdgpu_vcn_inst *vinst,
 				WREG32_SOC15(UVD, 0, mmUVD_RB_RPTR2, lower_32_bits(ring->wptr));
 				WREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2, lower_32_bits(ring->wptr));
 
-				ring = &adev->vcn.inst->ring_dec;
 				WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR,
 						   RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF);
 				SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS,
@@ -1399,7 +1398,6 @@ static int vcn_v1_0_pause_dpg_mode(struct amdgpu_vcn_inst *vinst,
 				WREG32_SOC15(UVD, 0, mmUVD_JRBC_RB_CNTL,
 							UVD_JRBC_RB_CNTL__RB_RPTR_WR_EN_MASK);
 
-				ring = &adev->vcn.inst->ring_dec;
 				WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR,
 						   RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF);
 				SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS,

From ec813f384b1a9df332e86ff46c422e5d2d00217f Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Mon, 25 Aug 2025 09:35:07 -0400
Subject: [PATCH 0692/1455] drm/amdgpu/vpe: cancel delayed work in hw_fini

We need to cancel any outstanding work at both suspend
and driver teardown. Move the cancel to hw_fini which
gets called in both cases.

Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c
index 86573c8d9b91..474bfe36c0c2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c
@@ -436,6 +436,8 @@ static int vpe_hw_fini(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	struct amdgpu_vpe *vpe = &adev->vpe;
 
+	cancel_delayed_work_sync(&adev->vpe.idle_work);
+
 	vpe_ring_stop(vpe);
 
 	/* Power off VPE */
@@ -446,10 +448,6 @@ static int vpe_hw_fini(struct amdgpu_ip_block *ip_block)
 
 static int vpe_suspend(struct amdgpu_ip_block *ip_block)
 {
-	struct amdgpu_device *adev = ip_block->adev;
-
-	cancel_delayed_work_sync(&adev->vpe.idle_work);
-
 	return vpe_hw_fini(ip_block);
 }
 

From e82a8d441038d8cb10b63047a9e705c42479d156 Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Mon, 25 Aug 2025 12:54:01 +0800
Subject: [PATCH 0693/1455] drm/amd/amdgpu: disable hwmon power1_cap* for gfx
 11.0.3 on vf mode

the PPSMC_MSG_GetPptLimit msg is not valid for gfx 11.0.3 on vf mode,
so skiped to create power1_cap* hwmon sysfs node.

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 30a504122a5a..5230276628a3 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -3586,14 +3586,16 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 		effective_mode &= ~S_IWUSR;
 
 	/* not implemented yet for APUs other than GC 10.3.1 (vangogh) and 9.4.3 */
-	if (((adev->family == AMDGPU_FAMILY_SI) ||
-	     ((adev->flags & AMD_IS_APU) && (gc_ver != IP_VERSION(10, 3, 1)) &&
-	      (gc_ver != IP_VERSION(9, 4, 3) && gc_ver != IP_VERSION(9, 4, 4)))) &&
-	    (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr ||
-	     attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr ||
-	     attr == &sensor_dev_attr_power1_cap.dev_attr.attr ||
-	     attr == &sensor_dev_attr_power1_cap_default.dev_attr.attr))
-		return 0;
+	if (attr == &sensor_dev_attr_power1_cap_max.dev_attr.attr ||
+	    attr == &sensor_dev_attr_power1_cap_min.dev_attr.attr ||
+	    attr == &sensor_dev_attr_power1_cap.dev_attr.attr ||
+	    attr == &sensor_dev_attr_power1_cap_default.dev_attr.attr) {
+		if (adev->family == AMDGPU_FAMILY_SI ||
+		    ((adev->flags & AMD_IS_APU) && gc_ver != IP_VERSION(10, 3, 1) &&
+		     (gc_ver != IP_VERSION(9, 4, 3) && gc_ver != IP_VERSION(9, 4, 4))) ||
+		    (amdgpu_sriov_vf(adev) && gc_ver == IP_VERSION(11, 0, 3)))
+			return 0;
+	}
 
 	/* not implemented yet for APUs having < GC 9.3.0 (Renoir) */
 	if (((adev->family == AMDGPU_FAMILY_SI) ||

From 5f976c9939f0d5916d2b8ef3156a6d1799781df1 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Tue, 26 Aug 2025 17:30:58 +0800
Subject: [PATCH 0694/1455] drm/amdgpu: update firmware version checks for user
 queue support

The minimum firmware versions required for user queue functionality
have been increased to address an issue where the queue privilege
state was lost during queue connect operations.

The problem occurred because the privilege state was being restored
to its initial value at the beginning of the function, overwriting
the state that was properly set during the queue connect case.

This commit updates the minimum version requirements:
- ME firmware from 2390 to 2420
- PFP firmware from 2530 to 2580
- MEC firmware from 2600 to 2650
- MES firmware remains at 120

These updated firmware versions contain the necessary fixes to
properly maintain queue privilege state throughout connect operations.

Fixes: 61ca97e9590c ("drm/amdgpu: Add fw minimum version check for usermode queue")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 456ba758fa94..c85de8c8f6f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1612,9 +1612,9 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 	case IP_VERSION(11, 0, 2):
 	case IP_VERSION(11, 0, 3):
 		if (!adev->gfx.disable_uq &&
-		    adev->gfx.me_fw_version  >= 2390 &&
-		    adev->gfx.pfp_fw_version >= 2530 &&
-		    adev->gfx.mec_fw_version >= 2600 &&
+		    adev->gfx.me_fw_version  >= 2420 &&
+		    adev->gfx.pfp_fw_version >= 2580 &&
+		    adev->gfx.mec_fw_version >= 2650 &&
 		    adev->mes.fw_version[0] >= 120) {
 			adev->userq_funcs[AMDGPU_HW_IP_GFX] = &userq_mes_funcs;
 			adev->userq_funcs[AMDGPU_HW_IP_COMPUTE] = &userq_mes_funcs;

From 7e2a5b0a9a165a7c51274aa01b18be29491b4345 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 22 Aug 2025 12:12:37 -0400
Subject: [PATCH 0695/1455] drm/amdgpu/userq: fix error handling of invalid
 doorbell

If the doorbell is invalid, be sure to set the r to an error
state so the function returns an error.

Reviewed-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 86cabb20bff1..f72de06a4ac8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -471,6 +471,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args)
 	if (index == (uint64_t)-EINVAL) {
 		drm_file_err(uq_mgr->file, "Failed to get doorbell for queue\n");
 		kfree(queue);
+		r = -EINVAL;
 		goto unlock;
 	}
 

From 2674f1ef29f46ae81464d0ac44b7824deb9de85c Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 26 Aug 2025 08:32:11 -0700
Subject: [PATCH 0696/1455] drm/xe/configfs: Block runtime attribute changes

Although it's possible to change the attributes in runtime, they have no
effect after the driver is already bound to the device. Check for that
and return -EBUSY in that case.

This should help users understand what's going on when the behavior is
not changing even if the value from the configfs is "right", but it got
to that state too late.

Reviewed-by: Riana Tauro <riana.tauro@intel.com>
Link: https://lore.kernel.org/r/20250826153210.3068808-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 41 ++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 29a12d02443d..1025d3979b06 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -75,6 +75,8 @@
  *
  *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/survivability_mode
  *
+ * This attribute can only be set before binding to the device.
+ *
  * Allowed engines:
  * ----------------
  *
@@ -99,6 +101,8 @@
  * available for migrations, but it's disabled. This is intended for debugging
  * purposes only.
  *
+ * This attribute can only be set before binding to the device.
+ *
  * PSMI
  * ----
  *
@@ -109,6 +113,8 @@
  *
  *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/enable_psmi
  *
+ * This attribute can only be set before binding to the device.
+ *
  * Remove devices
  * ==============
  *
@@ -169,6 +175,32 @@ static struct xe_config_device *to_xe_config_device(struct config_item *item)
 	return &to_xe_config_group_device(item)->config;
 }
 
+static bool is_bound(struct xe_config_group_device *dev)
+{
+	unsigned int domain, bus, slot, function;
+	struct pci_dev *pdev;
+	const char *name;
+	bool ret;
+
+	lockdep_assert_held(&dev->lock);
+
+	name = dev->group.cg_item.ci_name;
+	if (sscanf(name, "%x:%x:%x.%x", &domain, &bus, &slot, &function) != 4)
+		return false;
+
+	pdev = pci_get_domain_bus_and_slot(domain, bus, PCI_DEVFN(slot, function));
+	if (!pdev)
+		return false;
+
+	ret = pci_get_drvdata(pdev);
+	pci_dev_put(pdev);
+
+	if (ret)
+		pci_dbg(pdev, "Already bound to driver\n");
+
+	return ret;
+}
+
 static ssize_t survivability_mode_show(struct config_item *item, char *page)
 {
 	struct xe_config_device *dev = to_xe_config_device(item);
@@ -187,6 +219,9 @@ static ssize_t survivability_mode_store(struct config_item *item, const char *pa
 		return ret;
 
 	guard(mutex)(&dev->lock);
+	if (is_bound(dev))
+		return -EBUSY;
+
 	dev->config.survivability_mode = survivability_mode;
 
 	return len;
@@ -270,6 +305,9 @@ static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 	}
 
 	guard(mutex)(&dev->lock);
+	if (is_bound(dev))
+		return -EBUSY;
+
 	dev->config.engines_allowed = val;
 
 	return len;
@@ -293,6 +331,9 @@ static ssize_t enable_psmi_store(struct config_item *item, const char *page, siz
 		return ret;
 
 	guard(mutex)(&dev->lock);
+	if (is_bound(dev))
+		return -EBUSY;
+
 	dev->config.enable_psmi = val;
 
 	return len;

From ce5059bf851b7286ec27797d786aab28c8a62e5e Mon Sep 17 00:00:00 2001
From: Stuart Summers <stuart.summers@intel.com>
Date: Tue, 26 Aug 2025 18:29:03 +0000
Subject: [PATCH 0697/1455] drm/xe: Move explicit CT lock in TLB invalidation
 sequence

Currently the CT lock is used to cover TLB invalidation
sequence number updates. In an effort to separate the GuC
back end tracking of communication with the firmware from
the front end TLB sequence number tracking, add a new lock
here to specifically track those sequence number updates
coming in from the user.

Apart from the CT lock, we also have a pending lock to
cover both pending fences and sequence numbers received
from the back end. Those cover interrupt cases and so
it makes not to overload those with sequence numbers
coming in from new transactions. In that way, we'll employ
a mutex here.

v2: Actually add the correct lock rather than just dropping
    it... (Matt)

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-2-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 19 +++++++++++++------
 drivers/gpu/drm/xe/xe_gt_types.h            |  2 ++
 2 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
index 02f0bb92d6e0..75854b963d66 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
@@ -118,6 +118,9 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
  */
 int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt)
 {
+	struct xe_device *xe = gt_to_xe(gt);
+	int err;
+
 	gt->tlb_invalidation.seqno = 1;
 	INIT_LIST_HEAD(&gt->tlb_invalidation.pending_fences);
 	spin_lock_init(&gt->tlb_invalidation.pending_lock);
@@ -125,6 +128,10 @@ int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt)
 	INIT_DELAYED_WORK(&gt->tlb_invalidation.fence_tdr,
 			  xe_gt_tlb_fence_timeout);
 
+	err = drmm_mutex_init(&xe->drm, &gt->tlb_invalidation.seqno_lock);
+	if (err)
+		return err;
+
 	gt->tlb_invalidation.job_wq =
 		drmm_alloc_ordered_workqueue(&gt_to_xe(gt)->drm, "gt-tbl-inval-job-wq",
 					     WQ_MEM_RECLAIM);
@@ -158,7 +165,7 @@ void xe_gt_tlb_invalidation_reset(struct xe_gt *gt)
 	 * appear.
 	 */
 
-	mutex_lock(&gt->uc.guc.ct.lock);
+	mutex_lock(&gt->tlb_invalidation.seqno_lock);
 	spin_lock_irq(&gt->tlb_invalidation.pending_lock);
 	cancel_delayed_work(&gt->tlb_invalidation.fence_tdr);
 	/*
@@ -178,7 +185,7 @@ void xe_gt_tlb_invalidation_reset(struct xe_gt *gt)
 				 &gt->tlb_invalidation.pending_fences, link)
 		invalidation_fence_signal(gt_to_xe(gt), fence);
 	spin_unlock_irq(&gt->tlb_invalidation.pending_lock);
-	mutex_unlock(&gt->uc.guc.ct.lock);
+	mutex_unlock(&gt->tlb_invalidation.seqno_lock);
 }
 
 static bool tlb_invalidation_seqno_past(struct xe_gt *gt, int seqno)
@@ -211,13 +218,13 @@ static int send_tlb_invalidation(struct xe_guc *guc,
 	 * need to be updated.
 	 */
 
-	mutex_lock(&guc->ct.lock);
+	mutex_lock(&gt->tlb_invalidation.seqno_lock);
 	seqno = gt->tlb_invalidation.seqno;
 	fence->seqno = seqno;
 	trace_xe_gt_tlb_invalidation_fence_send(xe, fence);
 	action[1] = seqno;
-	ret = xe_guc_ct_send_locked(&guc->ct, action, len,
-				    G2H_LEN_DW_TLB_INVALIDATE, 1);
+	ret = xe_guc_ct_send(&guc->ct, action, len,
+			     G2H_LEN_DW_TLB_INVALIDATE, 1);
 	if (!ret) {
 		spin_lock_irq(&gt->tlb_invalidation.pending_lock);
 		/*
@@ -248,7 +255,7 @@ static int send_tlb_invalidation(struct xe_guc *guc,
 		if (!gt->tlb_invalidation.seqno)
 			gt->tlb_invalidation.seqno = 1;
 	}
-	mutex_unlock(&guc->ct.lock);
+	mutex_unlock(&gt->tlb_invalidation.seqno_lock);
 	xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1);
 
 	return ret;
diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
index ef0f2eecfa29..4dbc40fa6639 100644
--- a/drivers/gpu/drm/xe/xe_gt_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_types.h
@@ -190,6 +190,8 @@ struct xe_gt {
 		/** @tlb_invalidation.seqno: TLB invalidation seqno, protected by CT lock */
 #define TLB_INVALIDATION_SEQNO_MAX	0x100000
 		int seqno;
+		/** @tlb_invalidation.seqno_lock: protects @tlb_invalidation.seqno */
+		struct mutex seqno_lock;
 		/**
 		 * @tlb_invalidation.seqno_recv: last received TLB invalidation seqno,
 		 * protected by CT lock

From 76186a253a4b9eb41c5a83224c14efdf30960a71 Mon Sep 17 00:00:00 2001
From: Stuart Summers <stuart.summers@intel.com>
Date: Tue, 26 Aug 2025 18:29:04 +0000
Subject: [PATCH 0698/1455] drm/xe: Cancel pending TLB inval workers on
 teardown

Add a new _fini() routine on the GT TLB invalidation
side to handle this worker cleanup on driver teardown.

v2: Move the TLB teardown to the gt fini() routine called during
    gt_init rather than in gt_alloc. This way the GT structure stays
    alive for while we reset the TLB state.

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-3-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/xe_gt.c                  |  2 ++
 drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c | 12 ++++++++++++
 drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h |  1 +
 3 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index a3397f04abcc..178c4783bbda 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -603,6 +603,8 @@ static void xe_gt_fini(void *arg)
 	struct xe_gt *gt = arg;
 	int i;
 
+	xe_gt_tlb_invalidation_fini(gt);
+
 	for (i = 0; i < XE_ENGINE_CLASS_MAX; ++i)
 		xe_hw_fence_irq_finish(&gt->fence_irq[i]);
 
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
index 75854b963d66..db00c5adead9 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
@@ -188,6 +188,18 @@ void xe_gt_tlb_invalidation_reset(struct xe_gt *gt)
 	mutex_unlock(&gt->tlb_invalidation.seqno_lock);
 }
 
+/**
+ *
+ * xe_gt_tlb_invalidation_fini - Clean up GT TLB invalidation state
+ *
+ * Cancel pending fence workers and clean up any additional
+ * GT TLB invalidation state.
+ */
+void xe_gt_tlb_invalidation_fini(struct xe_gt *gt)
+{
+	xe_gt_tlb_invalidation_reset(gt);
+}
+
 static bool tlb_invalidation_seqno_past(struct xe_gt *gt, int seqno)
 {
 	int seqno_recv = READ_ONCE(gt->tlb_invalidation.seqno_recv);
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h
index f7f0f2eaf4b5..3e4cff3922d6 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h
@@ -16,6 +16,7 @@ struct xe_vm;
 struct xe_vma;
 
 int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt);
+void xe_gt_tlb_invalidation_fini(struct xe_gt *gt);
 
 void xe_gt_tlb_invalidation_reset(struct xe_gt *gt);
 int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt);

From c697ddcf27bd5c71b89539a9b79b7d078777b394 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:05 +0000
Subject: [PATCH 0699/1455] drm/xe: s/tlb_invalidation/tlb_inval

tlb_invalidation is a bit verbose leading to ugly wraps in the code,
shorten to tlb_inval.

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-4-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/Makefile                   |   2 +-
 drivers/gpu/drm/xe/xe_device_types.h          |   4 +-
 drivers/gpu/drm/xe/xe_exec_queue.c            |   2 +-
 drivers/gpu/drm/xe/xe_ggtt.c                  |   4 +-
 drivers/gpu/drm/xe/xe_gt.c                    |  10 +-
 drivers/gpu/drm/xe/xe_gt_pagefault.c          |   1 -
 ...t_tlb_invalidation.c => xe_gt_tlb_inval.c} | 262 +++++++++---------
 drivers/gpu/drm/xe/xe_gt_tlb_inval.h          |  41 +++
 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c      |  18 +-
 ...dation_types.h => xe_gt_tlb_inval_types.h} |  14 +-
 drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h   |  41 ---
 drivers/gpu/drm/xe/xe_gt_types.h              |  18 +-
 drivers/gpu/drm/xe/xe_guc_ct.c                |   8 +-
 drivers/gpu/drm/xe/xe_lmtt.c                  |  12 +-
 drivers/gpu/drm/xe/xe_pci.c                   |   6 +-
 drivers/gpu/drm/xe/xe_pci_types.h             |   2 +-
 drivers/gpu/drm/xe/xe_svm.c                   |   4 +-
 drivers/gpu/drm/xe/xe_trace.h                 |  24 +-
 drivers/gpu/drm/xe/xe_vm.c                    |  62 ++---
 drivers/gpu/drm/xe/xe_vm.h                    |   4 +-
 drivers/gpu/drm/xe/xe_vm_madvise.c            |   2 +-
 21 files changed, 265 insertions(+), 276 deletions(-)
 rename drivers/gpu/drm/xe/{xe_gt_tlb_invalidation.c => xe_gt_tlb_inval.c} (61%)
 create mode 100644 drivers/gpu/drm/xe/xe_gt_tlb_inval.h
 rename drivers/gpu/drm/xe/{xe_gt_tlb_invalidation_types.h => xe_gt_tlb_inval_types.h} (55%)
 delete mode 100644 drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 0abd8078cfb0..a1e4b3b1a1b6 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -61,7 +61,7 @@ xe-y += xe_bb.o \
 	xe_gt_pagefault.o \
 	xe_gt_sysfs.o \
 	xe_gt_throttle.o \
-	xe_gt_tlb_invalidation.o \
+	xe_gt_tlb_inval.o \
 	xe_gt_tlb_inval_job.o \
 	xe_gt_topology.o \
 	xe_guc.o \
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 389996970564..646d04aec68c 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -290,8 +290,8 @@ struct xe_device {
 		u8 has_mbx_power_limits:1;
 		/** @info.has_pxp: Device has PXP support */
 		u8 has_pxp:1;
-		/** @info.has_range_tlb_invalidation: Has range based TLB invalidations */
-		u8 has_range_tlb_invalidation:1;
+		/** @info.has_range_tlb_inval: Has range based TLB invalidations */
+		u8 has_range_tlb_inval:1;
 		/** @info.has_sriov: Supports SR-IOV */
 		u8 has_sriov:1;
 		/** @info.has_usm: Device has unified shared memory support */
diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 2d10a53f701d..063c89d981e5 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -75,7 +75,7 @@ static int alloc_dep_schedulers(struct xe_device *xe, struct xe_exec_queue *q)
 		if (!gt)
 			continue;
 
-		wq = gt->tlb_invalidation.job_wq;
+		wq = gt->tlb_inval.job_wq;
 
 #define MAX_TLB_INVAL_JOBS	16	/* Picking a reasonable value */
 		dep_scheduler = xe_dep_scheduler_create(xe, wq, q->name,
diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index e03222f5ac5a..c3e46c270117 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -23,7 +23,7 @@
 #include "xe_device.h"
 #include "xe_gt.h"
 #include "xe_gt_printk.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_map.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
@@ -438,7 +438,7 @@ static void ggtt_invalidate_gt_tlb(struct xe_gt *gt)
 	if (!gt)
 		return;
 
-	err = xe_gt_tlb_invalidation_ggtt(gt);
+	err = xe_gt_tlb_inval_ggtt(gt);
 	xe_gt_WARN(gt, err, "Failed to invalidate GGTT (%pe)", ERR_PTR(err));
 }
 
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 178c4783bbda..9a4639732bd7 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -37,7 +37,7 @@
 #include "xe_gt_sriov_pf.h"
 #include "xe_gt_sriov_vf.h"
 #include "xe_gt_sysfs.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_gt_topology.h"
 #include "xe_guc_exec_queue_types.h"
 #include "xe_guc_pc.h"
@@ -413,7 +413,7 @@ int xe_gt_init_early(struct xe_gt *gt)
 	xe_force_wake_init_gt(gt, gt_to_fw(gt));
 	spin_lock_init(&gt->global_invl_lock);
 
-	err = xe_gt_tlb_invalidation_init_early(gt);
+	err = xe_gt_tlb_inval_init_early(gt);
 	if (err)
 		return err;
 
@@ -603,7 +603,7 @@ static void xe_gt_fini(void *arg)
 	struct xe_gt *gt = arg;
 	int i;
 
-	xe_gt_tlb_invalidation_fini(gt);
+	xe_gt_tlb_inval_fini(gt);
 
 	for (i = 0; i < XE_ENGINE_CLASS_MAX; ++i)
 		xe_hw_fence_irq_finish(&gt->fence_irq[i]);
@@ -852,7 +852,7 @@ static int gt_reset(struct xe_gt *gt)
 
 	xe_uc_stop(&gt->uc);
 
-	xe_gt_tlb_invalidation_reset(gt);
+	xe_gt_tlb_inval_reset(gt);
 
 	err = do_gt_reset(gt);
 	if (err)
@@ -1066,5 +1066,5 @@ void xe_gt_declare_wedged(struct xe_gt *gt)
 	xe_gt_assert(gt, gt_to_xe(gt)->wedged.mode);
 
 	xe_uc_declare_wedged(&gt->uc);
-	xe_gt_tlb_invalidation_reset(gt);
+	xe_gt_tlb_inval_reset(gt);
 }
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index 4ea30fbce9bd..d02d22fb3659 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -16,7 +16,6 @@
 #include "xe_gt.h"
 #include "xe_gt_printk.h"
 #include "xe_gt_stats.h"
-#include "xe_gt_tlb_invalidation.h"
 #include "xe_guc.h"
 #include "xe_guc_ct.h"
 #include "xe_migrate.h"
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c b/drivers/gpu/drm/xe/xe_gt_tlb_inval.c
similarity index 61%
rename from drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
rename to drivers/gpu/drm/xe/xe_gt_tlb_inval.c
index db00c5adead9..1571fd917830 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval.c
@@ -5,8 +5,6 @@
 
 #include <drm/drm_managed.h>
 
-#include "xe_gt_tlb_invalidation.h"
-
 #include "abi/guc_actions_abi.h"
 #include "xe_device.h"
 #include "xe_force_wake.h"
@@ -15,6 +13,7 @@
 #include "xe_guc.h"
 #include "xe_guc_ct.h"
 #include "xe_gt_stats.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
 #include "xe_sriov.h"
@@ -39,7 +38,7 @@ static long tlb_timeout_jiffies(struct xe_gt *gt)
 	return hw_tlb_timeout + 2 * delay;
 }
 
-static void xe_gt_tlb_invalidation_fence_fini(struct xe_gt_tlb_invalidation_fence *fence)
+static void xe_gt_tlb_inval_fence_fini(struct xe_gt_tlb_inval_fence *fence)
 {
 	if (WARN_ON_ONCE(!fence->gt))
 		return;
@@ -49,66 +48,66 @@ static void xe_gt_tlb_invalidation_fence_fini(struct xe_gt_tlb_invalidation_fenc
 }
 
 static void
-__invalidation_fence_signal(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence)
+__inval_fence_signal(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence)
 {
 	bool stack = test_bit(FENCE_STACK_BIT, &fence->base.flags);
 
-	trace_xe_gt_tlb_invalidation_fence_signal(xe, fence);
-	xe_gt_tlb_invalidation_fence_fini(fence);
+	trace_xe_gt_tlb_inval_fence_signal(xe, fence);
+	xe_gt_tlb_inval_fence_fini(fence);
 	dma_fence_signal(&fence->base);
 	if (!stack)
 		dma_fence_put(&fence->base);
 }
 
 static void
-invalidation_fence_signal(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence)
+inval_fence_signal(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence)
 {
 	list_del(&fence->link);
-	__invalidation_fence_signal(xe, fence);
+	__inval_fence_signal(xe, fence);
 }
 
-void xe_gt_tlb_invalidation_fence_signal(struct xe_gt_tlb_invalidation_fence *fence)
+void xe_gt_tlb_inval_fence_signal(struct xe_gt_tlb_inval_fence *fence)
 {
 	if (WARN_ON_ONCE(!fence->gt))
 		return;
 
-	__invalidation_fence_signal(gt_to_xe(fence->gt), fence);
+	__inval_fence_signal(gt_to_xe(fence->gt), fence);
 }
 
 static void xe_gt_tlb_fence_timeout(struct work_struct *work)
 {
 	struct xe_gt *gt = container_of(work, struct xe_gt,
-					tlb_invalidation.fence_tdr.work);
+					tlb_inval.fence_tdr.work);
 	struct xe_device *xe = gt_to_xe(gt);
-	struct xe_gt_tlb_invalidation_fence *fence, *next;
+	struct xe_gt_tlb_inval_fence *fence, *next;
 
 	LNL_FLUSH_WORK(&gt->uc.guc.ct.g2h_worker);
 
-	spin_lock_irq(&gt->tlb_invalidation.pending_lock);
+	spin_lock_irq(&gt->tlb_inval.pending_lock);
 	list_for_each_entry_safe(fence, next,
-				 &gt->tlb_invalidation.pending_fences, link) {
+				 &gt->tlb_inval.pending_fences, link) {
 		s64 since_inval_ms = ktime_ms_delta(ktime_get(),
-						    fence->invalidation_time);
+						    fence->inval_time);
 
 		if (msecs_to_jiffies(since_inval_ms) < tlb_timeout_jiffies(gt))
 			break;
 
-		trace_xe_gt_tlb_invalidation_fence_timeout(xe, fence);
+		trace_xe_gt_tlb_inval_fence_timeout(xe, fence);
 		xe_gt_err(gt, "TLB invalidation fence timeout, seqno=%d recv=%d",
-			  fence->seqno, gt->tlb_invalidation.seqno_recv);
+			  fence->seqno, gt->tlb_inval.seqno_recv);
 
 		fence->base.error = -ETIME;
-		invalidation_fence_signal(xe, fence);
+		inval_fence_signal(xe, fence);
 	}
-	if (!list_empty(&gt->tlb_invalidation.pending_fences))
+	if (!list_empty(&gt->tlb_inval.pending_fences))
 		queue_delayed_work(system_wq,
-				   &gt->tlb_invalidation.fence_tdr,
+				   &gt->tlb_inval.fence_tdr,
 				   tlb_timeout_jiffies(gt));
-	spin_unlock_irq(&gt->tlb_invalidation.pending_lock);
+	spin_unlock_irq(&gt->tlb_inval.pending_lock);
 }
 
 /**
- * xe_gt_tlb_invalidation_init_early - Initialize GT TLB invalidation state
+ * xe_gt_tlb_inval_init_early - Initialize GT TLB invalidation state
  * @gt: GT structure
  *
  * Initialize GT TLB invalidation state, purely software initialization, should
@@ -116,40 +115,40 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
  *
  * Return: 0 on success, negative error code on error.
  */
-int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt)
+int xe_gt_tlb_inval_init_early(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 	int err;
 
-	gt->tlb_invalidation.seqno = 1;
-	INIT_LIST_HEAD(&gt->tlb_invalidation.pending_fences);
-	spin_lock_init(&gt->tlb_invalidation.pending_lock);
-	spin_lock_init(&gt->tlb_invalidation.lock);
-	INIT_DELAYED_WORK(&gt->tlb_invalidation.fence_tdr,
+	gt->tlb_inval.seqno = 1;
+	INIT_LIST_HEAD(&gt->tlb_inval.pending_fences);
+	spin_lock_init(&gt->tlb_inval.pending_lock);
+	spin_lock_init(&gt->tlb_inval.lock);
+	INIT_DELAYED_WORK(&gt->tlb_inval.fence_tdr,
 			  xe_gt_tlb_fence_timeout);
 
-	err = drmm_mutex_init(&xe->drm, &gt->tlb_invalidation.seqno_lock);
+	err = drmm_mutex_init(&xe->drm, &gt->tlb_inval.seqno_lock);
 	if (err)
 		return err;
 
-	gt->tlb_invalidation.job_wq =
+	gt->tlb_inval.job_wq =
 		drmm_alloc_ordered_workqueue(&gt_to_xe(gt)->drm, "gt-tbl-inval-job-wq",
 					     WQ_MEM_RECLAIM);
-	if (IS_ERR(gt->tlb_invalidation.job_wq))
-		return PTR_ERR(gt->tlb_invalidation.job_wq);
+	if (IS_ERR(gt->tlb_inval.job_wq))
+		return PTR_ERR(gt->tlb_inval.job_wq);
 
 	return 0;
 }
 
 /**
- * xe_gt_tlb_invalidation_reset - Initialize GT TLB invalidation reset
+ * xe_gt_tlb_inval_reset - Initialize GT TLB invalidation reset
  * @gt: GT structure
  *
  * Signal any pending invalidation fences, should be called during a GT reset
  */
-void xe_gt_tlb_invalidation_reset(struct xe_gt *gt)
+void xe_gt_tlb_inval_reset(struct xe_gt *gt)
 {
-	struct xe_gt_tlb_invalidation_fence *fence, *next;
+	struct xe_gt_tlb_inval_fence *fence, *next;
 	int pending_seqno;
 
 	/*
@@ -165,9 +164,9 @@ void xe_gt_tlb_invalidation_reset(struct xe_gt *gt)
 	 * appear.
 	 */
 
-	mutex_lock(&gt->tlb_invalidation.seqno_lock);
-	spin_lock_irq(&gt->tlb_invalidation.pending_lock);
-	cancel_delayed_work(&gt->tlb_invalidation.fence_tdr);
+	mutex_lock(&gt->tlb_inval.seqno_lock);
+	spin_lock_irq(&gt->tlb_inval.pending_lock);
+	cancel_delayed_work(&gt->tlb_inval.fence_tdr);
 	/*
 	 * We might have various kworkers waiting for TLB flushes to complete
 	 * which are not tracked with an explicit TLB fence, however at this
@@ -175,34 +174,34 @@ void xe_gt_tlb_invalidation_reset(struct xe_gt *gt)
 	 * make sure we signal them here under the assumption that we have
 	 * completed a full GT reset.
 	 */
-	if (gt->tlb_invalidation.seqno == 1)
+	if (gt->tlb_inval.seqno == 1)
 		pending_seqno = TLB_INVALIDATION_SEQNO_MAX - 1;
 	else
-		pending_seqno = gt->tlb_invalidation.seqno - 1;
-	WRITE_ONCE(gt->tlb_invalidation.seqno_recv, pending_seqno);
+		pending_seqno = gt->tlb_inval.seqno - 1;
+	WRITE_ONCE(gt->tlb_inval.seqno_recv, pending_seqno);
 
 	list_for_each_entry_safe(fence, next,
-				 &gt->tlb_invalidation.pending_fences, link)
-		invalidation_fence_signal(gt_to_xe(gt), fence);
-	spin_unlock_irq(&gt->tlb_invalidation.pending_lock);
-	mutex_unlock(&gt->tlb_invalidation.seqno_lock);
+				 &gt->tlb_inval.pending_fences, link)
+		inval_fence_signal(gt_to_xe(gt), fence);
+	spin_unlock_irq(&gt->tlb_inval.pending_lock);
+	mutex_unlock(&gt->tlb_inval.seqno_lock);
 }
 
 /**
  *
- * xe_gt_tlb_invalidation_fini - Clean up GT TLB invalidation state
+ * xe_gt_tlb_inval_fini - Clean up GT TLB invalidation state
  *
  * Cancel pending fence workers and clean up any additional
  * GT TLB invalidation state.
  */
-void xe_gt_tlb_invalidation_fini(struct xe_gt *gt)
+void xe_gt_tlb_inval_fini(struct xe_gt *gt)
 {
-	xe_gt_tlb_invalidation_reset(gt);
+	xe_gt_tlb_inval_reset(gt);
 }
 
-static bool tlb_invalidation_seqno_past(struct xe_gt *gt, int seqno)
+static bool tlb_inval_seqno_past(struct xe_gt *gt, int seqno)
 {
-	int seqno_recv = READ_ONCE(gt->tlb_invalidation.seqno_recv);
+	int seqno_recv = READ_ONCE(gt->tlb_inval.seqno_recv);
 
 	if (seqno - seqno_recv < -(TLB_INVALIDATION_SEQNO_MAX / 2))
 		return false;
@@ -213,9 +212,9 @@ static bool tlb_invalidation_seqno_past(struct xe_gt *gt, int seqno)
 	return seqno_recv >= seqno;
 }
 
-static int send_tlb_invalidation(struct xe_guc *guc,
-				 struct xe_gt_tlb_invalidation_fence *fence,
-				 u32 *action, int len)
+static int send_tlb_inval(struct xe_guc *guc,
+			  struct xe_gt_tlb_inval_fence *fence,
+			  u32 *action, int len)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
 	struct xe_device *xe = gt_to_xe(gt);
@@ -230,44 +229,44 @@ static int send_tlb_invalidation(struct xe_guc *guc,
 	 * need to be updated.
 	 */
 
-	mutex_lock(&gt->tlb_invalidation.seqno_lock);
-	seqno = gt->tlb_invalidation.seqno;
+	mutex_lock(&gt->tlb_inval.seqno_lock);
+	seqno = gt->tlb_inval.seqno;
 	fence->seqno = seqno;
-	trace_xe_gt_tlb_invalidation_fence_send(xe, fence);
+	trace_xe_gt_tlb_inval_fence_send(xe, fence);
 	action[1] = seqno;
 	ret = xe_guc_ct_send(&guc->ct, action, len,
 			     G2H_LEN_DW_TLB_INVALIDATE, 1);
 	if (!ret) {
-		spin_lock_irq(&gt->tlb_invalidation.pending_lock);
+		spin_lock_irq(&gt->tlb_inval.pending_lock);
 		/*
 		 * We haven't actually published the TLB fence as per
 		 * pending_fences, but in theory our seqno could have already
 		 * been written as we acquired the pending_lock. In such a case
 		 * we can just go ahead and signal the fence here.
 		 */
-		if (tlb_invalidation_seqno_past(gt, seqno)) {
-			__invalidation_fence_signal(xe, fence);
+		if (tlb_inval_seqno_past(gt, seqno)) {
+			__inval_fence_signal(xe, fence);
 		} else {
-			fence->invalidation_time = ktime_get();
+			fence->inval_time = ktime_get();
 			list_add_tail(&fence->link,
-				      &gt->tlb_invalidation.pending_fences);
+				      &gt->tlb_inval.pending_fences);
 
-			if (list_is_singular(&gt->tlb_invalidation.pending_fences))
+			if (list_is_singular(&gt->tlb_inval.pending_fences))
 				queue_delayed_work(system_wq,
-						   &gt->tlb_invalidation.fence_tdr,
+						   &gt->tlb_inval.fence_tdr,
 						   tlb_timeout_jiffies(gt));
 		}
-		spin_unlock_irq(&gt->tlb_invalidation.pending_lock);
+		spin_unlock_irq(&gt->tlb_inval.pending_lock);
 	} else {
-		__invalidation_fence_signal(xe, fence);
+		__inval_fence_signal(xe, fence);
 	}
 	if (!ret) {
-		gt->tlb_invalidation.seqno = (gt->tlb_invalidation.seqno + 1) %
+		gt->tlb_inval.seqno = (gt->tlb_inval.seqno + 1) %
 			TLB_INVALIDATION_SEQNO_MAX;
-		if (!gt->tlb_invalidation.seqno)
-			gt->tlb_invalidation.seqno = 1;
+		if (!gt->tlb_inval.seqno)
+			gt->tlb_inval.seqno = 1;
 	}
-	mutex_unlock(&gt->tlb_invalidation.seqno_lock);
+	mutex_unlock(&gt->tlb_inval.seqno_lock);
 	xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1);
 
 	return ret;
@@ -278,7 +277,7 @@ static int send_tlb_invalidation(struct xe_guc *guc,
 		XE_GUC_TLB_INVAL_FLUSH_CACHE)
 
 /**
- * xe_gt_tlb_invalidation_guc - Issue a TLB invalidation on this GT for the GuC
+ * xe_gt_tlb_inval_guc - Issue a TLB invalidation on this GT for the GuC
  * @gt: GT structure
  * @fence: invalidation fence which will be signal on TLB invalidation
  * completion
@@ -288,18 +287,17 @@ static int send_tlb_invalidation(struct xe_guc *guc,
  *
  * Return: 0 on success, negative error code on error
  */
-static int xe_gt_tlb_invalidation_guc(struct xe_gt *gt,
-				      struct xe_gt_tlb_invalidation_fence *fence)
+static int xe_gt_tlb_inval_guc(struct xe_gt *gt,
+			       struct xe_gt_tlb_inval_fence *fence)
 {
 	u32 action[] = {
 		XE_GUC_ACTION_TLB_INVALIDATION,
-		0,  /* seqno, replaced in send_tlb_invalidation */
+		0,  /* seqno, replaced in send_tlb_inval */
 		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_GUC),
 	};
 	int ret;
 
-	ret = send_tlb_invalidation(&gt->uc.guc, fence, action,
-				    ARRAY_SIZE(action));
+	ret = send_tlb_inval(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
 	/*
 	 * -ECANCELED indicates the CT is stopped for a GT reset. TLB caches
 	 *  should be nuked on a GT reset so this error can be ignored.
@@ -311,7 +309,7 @@ static int xe_gt_tlb_invalidation_guc(struct xe_gt *gt,
 }
 
 /**
- * xe_gt_tlb_invalidation_ggtt - Issue a TLB invalidation on this GT for the GGTT
+ * xe_gt_tlb_inval_ggtt - Issue a TLB invalidation on this GT for the GGTT
  * @gt: GT structure
  *
  * Issue a TLB invalidation for the GGTT. Completion of TLB invalidation is
@@ -319,22 +317,22 @@ static int xe_gt_tlb_invalidation_guc(struct xe_gt *gt,
  *
  * Return: 0 on success, negative error code on error
  */
-int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt)
+int xe_gt_tlb_inval_ggtt(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 	unsigned int fw_ref;
 
 	if (xe_guc_ct_enabled(&gt->uc.guc.ct) &&
 	    gt->uc.guc.submission_state.enabled) {
-		struct xe_gt_tlb_invalidation_fence fence;
+		struct xe_gt_tlb_inval_fence fence;
 		int ret;
 
-		xe_gt_tlb_invalidation_fence_init(gt, &fence, true);
-		ret = xe_gt_tlb_invalidation_guc(gt, &fence);
+		xe_gt_tlb_inval_fence_init(gt, &fence, true);
+		ret = xe_gt_tlb_inval_guc(gt, &fence);
 		if (ret)
 			return ret;
 
-		xe_gt_tlb_invalidation_fence_wait(&fence);
+		xe_gt_tlb_inval_fence_wait(&fence);
 	} else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) {
 		struct xe_mmio *mmio = &gt->mmio;
 
@@ -357,34 +355,34 @@ int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt)
 	return 0;
 }
 
-static int send_tlb_invalidation_all(struct xe_gt *gt,
-				     struct xe_gt_tlb_invalidation_fence *fence)
+static int send_tlb_inval_all(struct xe_gt *gt,
+			      struct xe_gt_tlb_inval_fence *fence)
 {
 	u32 action[] = {
 		XE_GUC_ACTION_TLB_INVALIDATION_ALL,
-		0,  /* seqno, replaced in send_tlb_invalidation */
+		0,  /* seqno, replaced in send_tlb_inval */
 		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL),
 	};
 
-	return send_tlb_invalidation(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
+	return send_tlb_inval(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
 }
 
 /**
  * xe_gt_tlb_invalidation_all - Invalidate all TLBs across PF and all VFs.
  * @gt: the &xe_gt structure
- * @fence: the &xe_gt_tlb_invalidation_fence to be signaled on completion
+ * @fence: the &xe_gt_tlb_inval_fence to be signaled on completion
  *
  * Send a request to invalidate all TLBs across PF and all VFs.
  *
  * Return: 0 on success, negative error code on error
  */
-int xe_gt_tlb_invalidation_all(struct xe_gt *gt, struct xe_gt_tlb_invalidation_fence *fence)
+int xe_gt_tlb_inval_all(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence)
 {
 	int err;
 
 	xe_gt_assert(gt, gt == fence->gt);
 
-	err = send_tlb_invalidation_all(gt, fence);
+	err = send_tlb_inval_all(gt, fence);
 	if (err)
 		xe_gt_err(gt, "TLB invalidation request failed (%pe)", ERR_PTR(err));
 
@@ -399,8 +397,7 @@ int xe_gt_tlb_invalidation_all(struct xe_gt *gt, struct xe_gt_tlb_invalidation_f
 #define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX))
 
 /**
- * xe_gt_tlb_invalidation_range - Issue a TLB invalidation on this GT for an
- * address range
+ * xe_gt_tlb_inval_range - Issue a TLB invalidation on this GT for an address range
  *
  * @gt: GT structure
  * @fence: invalidation fence which will be signal on TLB invalidation
@@ -415,9 +412,8 @@ int xe_gt_tlb_invalidation_all(struct xe_gt *gt, struct xe_gt_tlb_invalidation_f
  *
  * Return: Negative error code on error, 0 on success
  */
-int xe_gt_tlb_invalidation_range(struct xe_gt *gt,
-				 struct xe_gt_tlb_invalidation_fence *fence,
-				 u64 start, u64 end, u32 asid)
+int xe_gt_tlb_inval_range(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence,
+			  u64 start, u64 end, u32 asid)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 #define MAX_TLB_INVALIDATION_LEN	7
@@ -429,13 +425,13 @@ int xe_gt_tlb_invalidation_range(struct xe_gt *gt,
 
 	/* Execlists not supported */
 	if (gt_to_xe(gt)->info.force_execlist) {
-		__invalidation_fence_signal(xe, fence);
+		__inval_fence_signal(xe, fence);
 		return 0;
 	}
 
 	action[len++] = XE_GUC_ACTION_TLB_INVALIDATION;
-	action[len++] = 0; /* seqno, replaced in send_tlb_invalidation */
-	if (!xe->info.has_range_tlb_invalidation ||
+	action[len++] = 0; /* seqno, replaced in send_tlb_inval */
+	if (!xe->info.has_range_tlb_inval ||
 	    length > MAX_RANGE_TLB_INVALIDATION_LENGTH) {
 		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL);
 	} else {
@@ -484,33 +480,33 @@ int xe_gt_tlb_invalidation_range(struct xe_gt *gt,
 
 	xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN);
 
-	return send_tlb_invalidation(&gt->uc.guc, fence, action, len);
+	return send_tlb_inval(&gt->uc.guc, fence, action, len);
 }
 
 /**
- * xe_gt_tlb_invalidation_vm - Issue a TLB invalidation on this GT for a VM
+ * xe_gt_tlb_inval_vm - Issue a TLB invalidation on this GT for a VM
  * @gt: graphics tile
  * @vm: VM to invalidate
  *
  * Invalidate entire VM's address space
  */
-void xe_gt_tlb_invalidation_vm(struct xe_gt *gt, struct xe_vm *vm)
+void xe_gt_tlb_inval_vm(struct xe_gt *gt, struct xe_vm *vm)
 {
-	struct xe_gt_tlb_invalidation_fence fence;
+	struct xe_gt_tlb_inval_fence fence;
 	u64 range = 1ull << vm->xe->info.va_bits;
 	int ret;
 
-	xe_gt_tlb_invalidation_fence_init(gt, &fence, true);
+	xe_gt_tlb_inval_fence_init(gt, &fence, true);
 
-	ret = xe_gt_tlb_invalidation_range(gt, &fence, 0, range, vm->usm.asid);
+	ret = xe_gt_tlb_inval_range(gt, &fence, 0, range, vm->usm.asid);
 	if (ret < 0)
 		return;
 
-	xe_gt_tlb_invalidation_fence_wait(&fence);
+	xe_gt_tlb_inval_fence_wait(&fence);
 }
 
 /**
- * xe_guc_tlb_invalidation_done_handler - TLB invalidation done handler
+ * xe_guc_tlb_inval_done_handler - TLB invalidation done handler
  * @guc: guc
  * @msg: message indicating TLB invalidation done
  * @len: length of message
@@ -521,11 +517,11 @@ void xe_gt_tlb_invalidation_vm(struct xe_gt *gt, struct xe_vm *vm)
  *
  * Return: 0 on success, -EPROTO for malformed messages.
  */
-int xe_guc_tlb_invalidation_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
+int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
 	struct xe_device *xe = gt_to_xe(gt);
-	struct xe_gt_tlb_invalidation_fence *fence, *next;
+	struct xe_gt_tlb_inval_fence *fence, *next;
 	unsigned long flags;
 
 	if (unlikely(len != 1))
@@ -546,74 +542,74 @@ int xe_guc_tlb_invalidation_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
 	 * officially process the CT message like if racing against
 	 * process_g2h_msg().
 	 */
-	spin_lock_irqsave(&gt->tlb_invalidation.pending_lock, flags);
-	if (tlb_invalidation_seqno_past(gt, msg[0])) {
-		spin_unlock_irqrestore(&gt->tlb_invalidation.pending_lock, flags);
+	spin_lock_irqsave(&gt->tlb_inval.pending_lock, flags);
+	if (tlb_inval_seqno_past(gt, msg[0])) {
+		spin_unlock_irqrestore(&gt->tlb_inval.pending_lock, flags);
 		return 0;
 	}
 
-	WRITE_ONCE(gt->tlb_invalidation.seqno_recv, msg[0]);
+	WRITE_ONCE(gt->tlb_inval.seqno_recv, msg[0]);
 
 	list_for_each_entry_safe(fence, next,
-				 &gt->tlb_invalidation.pending_fences, link) {
-		trace_xe_gt_tlb_invalidation_fence_recv(xe, fence);
+				 &gt->tlb_inval.pending_fences, link) {
+		trace_xe_gt_tlb_inval_fence_recv(xe, fence);
 
-		if (!tlb_invalidation_seqno_past(gt, fence->seqno))
+		if (!tlb_inval_seqno_past(gt, fence->seqno))
 			break;
 
-		invalidation_fence_signal(xe, fence);
+		inval_fence_signal(xe, fence);
 	}
 
-	if (!list_empty(&gt->tlb_invalidation.pending_fences))
+	if (!list_empty(&gt->tlb_inval.pending_fences))
 		mod_delayed_work(system_wq,
-				 &gt->tlb_invalidation.fence_tdr,
+				 &gt->tlb_inval.fence_tdr,
 				 tlb_timeout_jiffies(gt));
 	else
-		cancel_delayed_work(&gt->tlb_invalidation.fence_tdr);
+		cancel_delayed_work(&gt->tlb_inval.fence_tdr);
 
-	spin_unlock_irqrestore(&gt->tlb_invalidation.pending_lock, flags);
+	spin_unlock_irqrestore(&gt->tlb_inval.pending_lock, flags);
 
 	return 0;
 }
 
 static const char *
-invalidation_fence_get_driver_name(struct dma_fence *dma_fence)
+inval_fence_get_driver_name(struct dma_fence *dma_fence)
 {
 	return "xe";
 }
 
 static const char *
-invalidation_fence_get_timeline_name(struct dma_fence *dma_fence)
+inval_fence_get_timeline_name(struct dma_fence *dma_fence)
 {
-	return "invalidation_fence";
+	return "inval_fence";
 }
 
-static const struct dma_fence_ops invalidation_fence_ops = {
-	.get_driver_name = invalidation_fence_get_driver_name,
-	.get_timeline_name = invalidation_fence_get_timeline_name,
+static const struct dma_fence_ops inval_fence_ops = {
+	.get_driver_name = inval_fence_get_driver_name,
+	.get_timeline_name = inval_fence_get_timeline_name,
 };
 
 /**
- * xe_gt_tlb_invalidation_fence_init - Initialize TLB invalidation fence
+ * xe_gt_tlb_inval_fence_init - Initialize TLB invalidation fence
  * @gt: GT
  * @fence: TLB invalidation fence to initialize
  * @stack: fence is stack variable
  *
- * Initialize TLB invalidation fence for use. xe_gt_tlb_invalidation_fence_fini
+ * Initialize TLB invalidation fence for use. xe_gt_tlb_inval_fence_fini
  * will be automatically called when fence is signalled (all fences must signal),
  * even on error.
  */
-void xe_gt_tlb_invalidation_fence_init(struct xe_gt *gt,
-				       struct xe_gt_tlb_invalidation_fence *fence,
-				       bool stack)
+void xe_gt_tlb_inval_fence_init(struct xe_gt *gt,
+				struct xe_gt_tlb_inval_fence *fence,
+				bool stack)
 {
 	xe_pm_runtime_get_noresume(gt_to_xe(gt));
 
-	spin_lock_irq(&gt->tlb_invalidation.lock);
-	dma_fence_init(&fence->base, &invalidation_fence_ops,
-		       &gt->tlb_invalidation.lock,
+	spin_lock_irq(&gt->tlb_inval.lock);
+	dma_fence_init(&fence->base, &inval_fence_ops,
+		       &gt->tlb_inval.lock,
 		       dma_fence_context_alloc(1), 1);
-	spin_unlock_irq(&gt->tlb_invalidation.lock);
+	spin_unlock_irq(&gt->tlb_inval.lock);
 	INIT_LIST_HEAD(&fence->link);
 	if (stack)
 		set_bit(FENCE_STACK_BIT, &fence->base.flags);
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval.h b/drivers/gpu/drm/xe/xe_gt_tlb_inval.h
new file mode 100644
index 000000000000..b1258ac4e8fb
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2023 Intel Corporation
+ */
+
+#ifndef _XE_GT_TLB_INVAL_H_
+#define _XE_GT_TLB_INVAL_H_
+
+#include <linux/types.h>
+
+#include "xe_gt_tlb_inval_types.h"
+
+struct xe_gt;
+struct xe_guc;
+struct xe_vm;
+struct xe_vma;
+
+int xe_gt_tlb_inval_init_early(struct xe_gt *gt);
+void xe_gt_tlb_inval_fini(struct xe_gt *gt);
+
+void xe_gt_tlb_inval_reset(struct xe_gt *gt);
+int xe_gt_tlb_inval_ggtt(struct xe_gt *gt);
+void xe_gt_tlb_inval_vm(struct xe_gt *gt, struct xe_vm *vm);
+int xe_gt_tlb_inval_all(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence);
+int xe_gt_tlb_inval_range(struct xe_gt *gt,
+			  struct xe_gt_tlb_inval_fence *fence,
+			  u64 start, u64 end, u32 asid);
+int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len);
+
+void xe_gt_tlb_inval_fence_init(struct xe_gt *gt,
+				struct xe_gt_tlb_inval_fence *fence,
+				bool stack);
+void xe_gt_tlb_inval_fence_signal(struct xe_gt_tlb_inval_fence *fence);
+
+static inline void
+xe_gt_tlb_inval_fence_wait(struct xe_gt_tlb_inval_fence *fence)
+{
+	dma_fence_wait(&fence->base, false);
+}
+
+#endif	/* _XE_GT_TLB_INVAL_ */
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
index e9255be26467..41e0ea92ea5a 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
@@ -7,7 +7,7 @@
 #include "xe_dep_scheduler.h"
 #include "xe_exec_queue.h"
 #include "xe_gt.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_gt_tlb_inval_job.h"
 #include "xe_migrate.h"
 #include "xe_pm.h"
@@ -41,11 +41,11 @@ static struct dma_fence *xe_gt_tlb_inval_job_run(struct xe_dep_job *dep_job)
 {
 	struct xe_gt_tlb_inval_job *job =
 		container_of(dep_job, typeof(*job), dep);
-	struct xe_gt_tlb_invalidation_fence *ifence =
+	struct xe_gt_tlb_inval_fence *ifence =
 		container_of(job->fence, typeof(*ifence), base);
 
-	xe_gt_tlb_invalidation_range(job->gt, ifence, job->start,
-				     job->end, job->asid);
+	xe_gt_tlb_inval_range(job->gt, ifence, job->start,
+			      job->end, job->asid);
 
 	return job->fence;
 }
@@ -93,7 +93,7 @@ struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
 		q->tlb_inval[xe_gt_tlb_inval_context(gt)].dep_scheduler;
 	struct drm_sched_entity *entity =
 		xe_dep_scheduler_entity(dep_scheduler);
-	struct xe_gt_tlb_invalidation_fence *ifence;
+	struct xe_gt_tlb_inval_fence *ifence;
 	int err;
 
 	job = kmalloc(sizeof(*job), GFP_KERNEL);
@@ -140,7 +140,7 @@ static void xe_gt_tlb_inval_job_destroy(struct kref *ref)
 {
 	struct xe_gt_tlb_inval_job *job = container_of(ref, typeof(*job),
 						       refcount);
-	struct xe_gt_tlb_invalidation_fence *ifence =
+	struct xe_gt_tlb_inval_fence *ifence =
 		container_of(job->fence, typeof(*ifence), base);
 	struct xe_device *xe = gt_to_xe(job->gt);
 	struct xe_exec_queue *q = job->q;
@@ -148,7 +148,7 @@ static void xe_gt_tlb_inval_job_destroy(struct kref *ref)
 	if (!job->fence_armed)
 		kfree(ifence);
 	else
-		/* Ref from xe_gt_tlb_invalidation_fence_init */
+		/* Ref from xe_gt_tlb_inval_fence_init */
 		dma_fence_put(job->fence);
 
 	drm_sched_job_cleanup(&job->dep.drm);
@@ -194,7 +194,7 @@ struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
 					   struct xe_migrate *m,
 					   struct dma_fence *fence)
 {
-	struct xe_gt_tlb_invalidation_fence *ifence =
+	struct xe_gt_tlb_inval_fence *ifence =
 		container_of(job->fence, typeof(*ifence), base);
 
 	if (!dma_fence_is_signaled(fence)) {
@@ -226,7 +226,7 @@ struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
 	xe_migrate_job_lock(m, job->q);
 
 	/* Creation ref pairs with put in xe_gt_tlb_inval_job_destroy */
-	xe_gt_tlb_invalidation_fence_init(job->gt, ifence, false);
+	xe_gt_tlb_inval_fence_init(job->gt, ifence, false);
 	dma_fence_get(job->fence);	/* Pairs with put in DRM scheduler */
 
 	drm_sched_job_arm(&job->dep.drm);
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation_types.h b/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
similarity index 55%
rename from drivers/gpu/drm/xe/xe_gt_tlb_invalidation_types.h
rename to drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
index de6e825e0851..919430359103 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
@@ -3,20 +3,20 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#ifndef _XE_GT_TLB_INVALIDATION_TYPES_H_
-#define _XE_GT_TLB_INVALIDATION_TYPES_H_
+#ifndef _XE_GT_TLB_INVAL_TYPES_H_
+#define _XE_GT_TLB_INVAL_TYPES_H_
 
 #include <linux/dma-fence.h>
 
 struct xe_gt;
 
 /**
- * struct xe_gt_tlb_invalidation_fence - XE GT TLB invalidation fence
+ * struct xe_gt_tlb_inval_fence - XE GT TLB invalidation fence
  *
- * Optionally passed to xe_gt_tlb_invalidation and will be signaled upon TLB
+ * Optionally passed to xe_gt_tlb_inval and will be signaled upon TLB
  * invalidation completion.
  */
-struct xe_gt_tlb_invalidation_fence {
+struct xe_gt_tlb_inval_fence {
 	/** @base: dma fence base */
 	struct dma_fence base;
 	/** @gt: GT which fence belong to */
@@ -25,8 +25,8 @@ struct xe_gt_tlb_invalidation_fence {
 	struct list_head link;
 	/** @seqno: seqno of TLB invalidation to signal fence one */
 	int seqno;
-	/** @invalidation_time: time of TLB invalidation */
-	ktime_t invalidation_time;
+	/** @inval_time: time of TLB invalidation */
+	ktime_t inval_time;
 };
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h b/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h
deleted file mode 100644
index 3e4cff3922d6..000000000000
--- a/drivers/gpu/drm/xe/xe_gt_tlb_invalidation.h
+++ /dev/null
@@ -1,41 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright © 2023 Intel Corporation
- */
-
-#ifndef _XE_GT_TLB_INVALIDATION_H_
-#define _XE_GT_TLB_INVALIDATION_H_
-
-#include <linux/types.h>
-
-#include "xe_gt_tlb_invalidation_types.h"
-
-struct xe_gt;
-struct xe_guc;
-struct xe_vm;
-struct xe_vma;
-
-int xe_gt_tlb_invalidation_init_early(struct xe_gt *gt);
-void xe_gt_tlb_invalidation_fini(struct xe_gt *gt);
-
-void xe_gt_tlb_invalidation_reset(struct xe_gt *gt);
-int xe_gt_tlb_invalidation_ggtt(struct xe_gt *gt);
-void xe_gt_tlb_invalidation_vm(struct xe_gt *gt, struct xe_vm *vm);
-int xe_gt_tlb_invalidation_all(struct xe_gt *gt, struct xe_gt_tlb_invalidation_fence *fence);
-int xe_gt_tlb_invalidation_range(struct xe_gt *gt,
-				 struct xe_gt_tlb_invalidation_fence *fence,
-				 u64 start, u64 end, u32 asid);
-int xe_guc_tlb_invalidation_done_handler(struct xe_guc *guc, u32 *msg, u32 len);
-
-void xe_gt_tlb_invalidation_fence_init(struct xe_gt *gt,
-				       struct xe_gt_tlb_invalidation_fence *fence,
-				       bool stack);
-void xe_gt_tlb_invalidation_fence_signal(struct xe_gt_tlb_invalidation_fence *fence);
-
-static inline void
-xe_gt_tlb_invalidation_fence_wait(struct xe_gt_tlb_invalidation_fence *fence)
-{
-	dma_fence_wait(&fence->base, false);
-}
-
-#endif	/* _XE_GT_TLB_INVALIDATION_ */
diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
index 4dbc40fa6639..85cfcc49472b 100644
--- a/drivers/gpu/drm/xe/xe_gt_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_types.h
@@ -185,38 +185,38 @@ struct xe_gt {
 		struct work_struct worker;
 	} reset;
 
-	/** @tlb_invalidation: TLB invalidation state */
+	/** @tlb_inval: TLB invalidation state */
 	struct {
-		/** @tlb_invalidation.seqno: TLB invalidation seqno, protected by CT lock */
+		/** @tlb_inval.seqno: TLB invalidation seqno, protected by CT lock */
 #define TLB_INVALIDATION_SEQNO_MAX	0x100000
 		int seqno;
 		/** @tlb_invalidation.seqno_lock: protects @tlb_invalidation.seqno */
 		struct mutex seqno_lock;
 		/**
-		 * @tlb_invalidation.seqno_recv: last received TLB invalidation seqno,
+		 * @tlb_inval.seqno_recv: last received TLB invalidation seqno,
 		 * protected by CT lock
 		 */
 		int seqno_recv;
 		/**
-		 * @tlb_invalidation.pending_fences: list of pending fences waiting TLB
+		 * @tlb_inval.pending_fences: list of pending fences waiting TLB
 		 * invaliations, protected by CT lock
 		 */
 		struct list_head pending_fences;
 		/**
-		 * @tlb_invalidation.pending_lock: protects @tlb_invalidation.pending_fences
-		 * and updating @tlb_invalidation.seqno_recv.
+		 * @tlb_inval.pending_lock: protects @tlb_inval.pending_fences
+		 * and updating @tlb_inval.seqno_recv.
 		 */
 		spinlock_t pending_lock;
 		/**
-		 * @tlb_invalidation.fence_tdr: schedules a delayed call to
+		 * @tlb_inval.fence_tdr: schedules a delayed call to
 		 * xe_gt_tlb_fence_timeout after the timeut interval is over.
 		 */
 		struct delayed_work fence_tdr;
 		/** @wtlb_invalidation.wq: schedules GT TLB invalidation jobs */
 		struct workqueue_struct *job_wq;
-		/** @tlb_invalidation.lock: protects TLB invalidation fences */
+		/** @tlb_inval.lock: protects TLB invalidation fences */
 		spinlock_t lock;
-	} tlb_invalidation;
+	} tlb_inval;
 
 	/**
 	 * @ccs_mode: Number of compute engines enabled.
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 3f4e6a46ff16..9131d121d941 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -26,7 +26,7 @@
 #include "xe_gt_sriov_pf_control.h"
 #include "xe_gt_sriov_pf_monitor.h"
 #include "xe_gt_sriov_printk.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_guc.h"
 #include "xe_guc_log.h"
 #include "xe_guc_relay.h"
@@ -1416,8 +1416,7 @@ static int process_g2h_msg(struct xe_guc_ct *ct, u32 *msg, u32 len)
 		ret = xe_guc_pagefault_handler(guc, payload, adj_len);
 		break;
 	case XE_GUC_ACTION_TLB_INVALIDATION_DONE:
-		ret = xe_guc_tlb_invalidation_done_handler(guc, payload,
-							   adj_len);
+		ret = xe_guc_tlb_inval_done_handler(guc, payload, adj_len);
 		break;
 	case XE_GUC_ACTION_ACCESS_COUNTER_NOTIFY:
 		ret = xe_guc_access_counter_notify_handler(guc, payload,
@@ -1618,8 +1617,7 @@ static void g2h_fast_path(struct xe_guc_ct *ct, u32 *msg, u32 len)
 		break;
 	case XE_GUC_ACTION_TLB_INVALIDATION_DONE:
 		__g2h_release_space(ct, len);
-		ret = xe_guc_tlb_invalidation_done_handler(guc, payload,
-							   adj_len);
+		ret = xe_guc_tlb_inval_done_handler(guc, payload, adj_len);
 		break;
 	default:
 		xe_gt_warn(gt, "NOT_POSSIBLE");
diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c
index a78c9d474a6e..e5aba03ff8ac 100644
--- a/drivers/gpu/drm/xe/xe_lmtt.c
+++ b/drivers/gpu/drm/xe/xe_lmtt.c
@@ -11,7 +11,7 @@
 
 #include "xe_assert.h"
 #include "xe_bo.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_lmtt.h"
 #include "xe_map.h"
 #include "xe_mmio.h"
@@ -228,8 +228,8 @@ void xe_lmtt_init_hw(struct xe_lmtt *lmtt)
 
 static int lmtt_invalidate_hw(struct xe_lmtt *lmtt)
 {
-	struct xe_gt_tlb_invalidation_fence fences[XE_MAX_GT_PER_TILE];
-	struct xe_gt_tlb_invalidation_fence *fence = fences;
+	struct xe_gt_tlb_inval_fence fences[XE_MAX_GT_PER_TILE];
+	struct xe_gt_tlb_inval_fence *fence = fences;
 	struct xe_tile *tile = lmtt_to_tile(lmtt);
 	struct xe_gt *gt;
 	int result = 0;
@@ -237,8 +237,8 @@ static int lmtt_invalidate_hw(struct xe_lmtt *lmtt)
 	u8 id;
 
 	for_each_gt_on_tile(gt, tile, id) {
-		xe_gt_tlb_invalidation_fence_init(gt, fence, true);
-		err = xe_gt_tlb_invalidation_all(gt, fence);
+		xe_gt_tlb_inval_fence_init(gt, fence, true);
+		err = xe_gt_tlb_inval_all(gt, fence);
 		result = result ?: err;
 		fence++;
 	}
@@ -252,7 +252,7 @@ static int lmtt_invalidate_hw(struct xe_lmtt *lmtt)
 	 */
 	fence = fences;
 	for_each_gt_on_tile(gt, tile, id)
-		xe_gt_tlb_invalidation_fence_wait(fence++);
+		xe_gt_tlb_inval_fence_wait(fence++);
 
 	return result;
 }
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 15a863491cdd..046d330bad34 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -57,7 +57,7 @@ static const struct xe_graphics_desc graphics_xelp = {
 };
 
 #define XE_HP_FEATURES \
-	.has_range_tlb_invalidation = true, \
+	.has_range_tlb_inval = true, \
 	.va_bits = 48, \
 	.vm_max_level = 3
 
@@ -105,7 +105,7 @@ static const struct xe_graphics_desc graphics_xelpg = {
 	.has_asid = 1, \
 	.has_atomic_enable_pte_bit = 1, \
 	.has_flat_ccs = 1, \
-	.has_range_tlb_invalidation = 1, \
+	.has_range_tlb_inval = 1, \
 	.has_usm = 1, \
 	.has_64bit_timestamp = 1, \
 	.va_bits = 48, \
@@ -713,7 +713,7 @@ static int xe_info_init(struct xe_device *xe,
 	/* Runtime detection may change this later */
 	xe->info.has_flat_ccs = graphics_desc->has_flat_ccs;
 
-	xe->info.has_range_tlb_invalidation = graphics_desc->has_range_tlb_invalidation;
+	xe->info.has_range_tlb_inval = graphics_desc->has_range_tlb_inval;
 	xe->info.has_usm = graphics_desc->has_usm;
 	xe->info.has_64bit_timestamp = graphics_desc->has_64bit_timestamp;
 
diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h
index 4de6f69ed975..b63002fc0f67 100644
--- a/drivers/gpu/drm/xe/xe_pci_types.h
+++ b/drivers/gpu/drm/xe/xe_pci_types.h
@@ -60,7 +60,7 @@ struct xe_graphics_desc {
 	u8 has_atomic_enable_pte_bit:1;
 	u8 has_flat_ccs:1;
 	u8 has_indirect_ring_state:1;
-	u8 has_range_tlb_invalidation:1;
+	u8 has_range_tlb_inval:1;
 	u8 has_usm:1;
 	u8 has_64bit_timestamp:1;
 };
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 0596039ef0a1..babc64485b52 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -7,7 +7,7 @@
 
 #include "xe_bo.h"
 #include "xe_gt_stats.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_migrate.h"
 #include "xe_module.h"
 #include "xe_pm.h"
@@ -225,7 +225,7 @@ static void xe_svm_invalidate(struct drm_gpusvm *gpusvm,
 
 	xe_device_wmb(xe);
 
-	err = xe_vm_range_tilemask_tlb_invalidation(vm, adj_start, adj_end, tile_mask);
+	err = xe_vm_range_tilemask_tlb_inval(vm, adj_start, adj_end, tile_mask);
 	WARN_ON_ONCE(err);
 
 range_notifier_event_end:
diff --git a/drivers/gpu/drm/xe/xe_trace.h b/drivers/gpu/drm/xe/xe_trace.h
index 21486a6f693a..36538f50d06f 100644
--- a/drivers/gpu/drm/xe/xe_trace.h
+++ b/drivers/gpu/drm/xe/xe_trace.h
@@ -14,7 +14,7 @@
 
 #include "xe_exec_queue_types.h"
 #include "xe_gpu_scheduler_types.h"
-#include "xe_gt_tlb_invalidation_types.h"
+#include "xe_gt_tlb_inval_types.h"
 #include "xe_gt_types.h"
 #include "xe_guc_exec_queue_types.h"
 #include "xe_sched_job.h"
@@ -25,13 +25,13 @@
 #define __dev_name_gt(gt)	__dev_name_xe(gt_to_xe((gt)))
 #define __dev_name_eq(q)	__dev_name_gt((q)->gt)
 
-DECLARE_EVENT_CLASS(xe_gt_tlb_invalidation_fence,
-		    TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
+DECLARE_EVENT_CLASS(xe_gt_tlb_inval_fence,
+		    TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
 		    TP_ARGS(xe, fence),
 
 		    TP_STRUCT__entry(
 			     __string(dev, __dev_name_xe(xe))
-			     __field(struct xe_gt_tlb_invalidation_fence *, fence)
+			     __field(struct xe_gt_tlb_inval_fence *, fence)
 			     __field(int, seqno)
 			     ),
 
@@ -45,23 +45,23 @@ DECLARE_EVENT_CLASS(xe_gt_tlb_invalidation_fence,
 			      __get_str(dev), __entry->fence, __entry->seqno)
 );
 
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_send,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
+DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_send,
+	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_recv,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
+DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_recv,
+	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_signal,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
+DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_signal,
+	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
-DEFINE_EVENT(xe_gt_tlb_invalidation_fence, xe_gt_tlb_invalidation_fence_timeout,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_invalidation_fence *fence),
+DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_timeout,
+	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index e913efa2057b..b758e9a6837a 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -28,7 +28,7 @@
 #include "xe_drm_client.h"
 #include "xe_exec_queue.h"
 #include "xe_gt_pagefault.h"
-#include "xe_gt_tlb_invalidation.h"
+#include "xe_gt_tlb_inval.h"
 #include "xe_migrate.h"
 #include "xe_pat.h"
 #include "xe_pm.h"
@@ -1898,7 +1898,7 @@ static void xe_vm_close(struct xe_vm *vm)
 					xe_pt_clear(xe, vm->pt_root[id]);
 
 			for_each_gt(gt, xe, id)
-				xe_gt_tlb_invalidation_vm(gt, vm);
+				xe_gt_tlb_inval_vm(gt, vm);
 		}
 	}
 
@@ -4032,7 +4032,7 @@ void xe_vm_unlock(struct xe_vm *vm)
 }
 
 /**
- * xe_vm_range_tilemask_tlb_invalidation - Issue a TLB invalidation on this tilemask for an
+ * xe_vm_range_tilemask_tlb_inval - Issue a TLB invalidation on this tilemask for an
  * address range
  * @vm: The VM
  * @start: start address
@@ -4043,10 +4043,11 @@ void xe_vm_unlock(struct xe_vm *vm)
  *
  * Returns 0 for success, negative error code otherwise.
  */
-int xe_vm_range_tilemask_tlb_invalidation(struct xe_vm *vm, u64 start,
-					  u64 end, u8 tile_mask)
+int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start,
+				   u64 end, u8 tile_mask)
 {
-	struct xe_gt_tlb_invalidation_fence fence[XE_MAX_TILES_PER_DEVICE * XE_MAX_GT_PER_TILE];
+	struct xe_gt_tlb_inval_fence
+		fence[XE_MAX_TILES_PER_DEVICE * XE_MAX_GT_PER_TILE];
 	struct xe_tile *tile;
 	u32 fence_id = 0;
 	u8 id;
@@ -4056,39 +4057,34 @@ int xe_vm_range_tilemask_tlb_invalidation(struct xe_vm *vm, u64 start,
 		return 0;
 
 	for_each_tile(tile, vm->xe, id) {
-		if (tile_mask & BIT(id)) {
-			xe_gt_tlb_invalidation_fence_init(tile->primary_gt,
-							  &fence[fence_id], true);
+		if (!(tile_mask & BIT(id)))
+			continue;
 
-			err = xe_gt_tlb_invalidation_range(tile->primary_gt,
-							   &fence[fence_id],
-							   start,
-							   end,
-							   vm->usm.asid);
-			if (err)
-				goto wait;
-			++fence_id;
+		xe_gt_tlb_inval_fence_init(tile->primary_gt,
+					   &fence[fence_id], true);
 
-			if (!tile->media_gt)
-				continue;
+		err = xe_gt_tlb_inval_range(tile->primary_gt, &fence[fence_id],
+					    start, end, vm->usm.asid);
+		if (err)
+			goto wait;
+		++fence_id;
 
-			xe_gt_tlb_invalidation_fence_init(tile->media_gt,
-							  &fence[fence_id], true);
+		if (!tile->media_gt)
+			continue;
 
-			err = xe_gt_tlb_invalidation_range(tile->media_gt,
-							   &fence[fence_id],
-							   start,
-							   end,
-							   vm->usm.asid);
-			if (err)
-				goto wait;
-			++fence_id;
-		}
+		xe_gt_tlb_inval_fence_init(tile->media_gt,
+					   &fence[fence_id], true);
+
+		err = xe_gt_tlb_inval_range(tile->media_gt, &fence[fence_id],
+					    start, end, vm->usm.asid);
+		if (err)
+			goto wait;
+		++fence_id;
 	}
 
 wait:
 	for (id = 0; id < fence_id; ++id)
-		xe_gt_tlb_invalidation_fence_wait(&fence[id]);
+		xe_gt_tlb_inval_fence_wait(&fence[id]);
 
 	return err;
 }
@@ -4147,8 +4143,8 @@ int xe_vm_invalidate_vma(struct xe_vma *vma)
 
 	xe_device_wmb(xe);
 
-	ret = xe_vm_range_tilemask_tlb_invalidation(xe_vma_vm(vma), xe_vma_start(vma),
-						    xe_vma_end(vma), tile_mask);
+	ret = xe_vm_range_tilemask_tlb_inval(xe_vma_vm(vma), xe_vma_start(vma),
+					     xe_vma_end(vma), tile_mask);
 
 	/* WRITE_ONCE pairs with READ_ONCE in xe_vm_has_valid_gpu_mapping() */
 	WRITE_ONCE(vma->tile_invalidated, vma->tile_mask);
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 8afd2cd4c9dc..b3e5bec0fa58 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -236,8 +236,8 @@ struct dma_fence *xe_vm_range_rebind(struct xe_vm *vm,
 struct dma_fence *xe_vm_range_unbind(struct xe_vm *vm,
 				     struct xe_svm_range *range);
 
-int xe_vm_range_tilemask_tlb_invalidation(struct xe_vm *vm, u64 start,
-					  u64 end, u8 tile_mask);
+int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start,
+				   u64 end, u8 tile_mask);
 
 int xe_vm_invalidate_vma(struct xe_vma *vma);
 
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 7813bdedacaa..09c5783ee523 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -224,7 +224,7 @@ static int xe_vm_invalidate_madvise_range(struct xe_vm *vm, u64 start, u64 end)
 
 	xe_device_wmb(vm->xe);
 
-	return xe_vm_range_tilemask_tlb_invalidation(vm, start, end, tile_mask);
+	return xe_vm_range_tilemask_tlb_inval(vm, start, end, tile_mask);
 }
 
 static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madvise *args)

From 594bb930fc7dd40cea72c643047d642fadfc8b68 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:06 +0000
Subject: [PATCH 0700/1455] drm/xe: Add xe_tlb_inval structure

Extract TLB invalidation state into a structure to decouple TLB
invalidations from the GT, allowing the structure to be embedded
anywhere in the driver.

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-5-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h | 34 ++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_gt_types.h           | 33 ++-------------------
 2 files changed, 36 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h b/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
index 919430359103..442f72b78ccf 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
@@ -6,10 +6,44 @@
 #ifndef _XE_GT_TLB_INVAL_TYPES_H_
 #define _XE_GT_TLB_INVAL_TYPES_H_
 
+#include <linux/workqueue.h>
 #include <linux/dma-fence.h>
 
 struct xe_gt;
 
+/** struct xe_tlb_inval - TLB invalidation client */
+struct xe_tlb_inval {
+	/** @tlb_inval.seqno: TLB invalidation seqno, protected by CT lock */
+#define TLB_INVALIDATION_SEQNO_MAX	0x100000
+	int seqno;
+	/** @tlb_invalidation.seqno_lock: protects @tlb_invalidation.seqno */
+	struct mutex seqno_lock;
+	/**
+	 * @tlb_inval.seqno_recv: last received TLB invalidation seqno,
+	 * protected by CT lock
+	 */
+	int seqno_recv;
+	/**
+	 * @tlb_inval.pending_fences: list of pending fences waiting TLB
+	 * invaliations, protected by CT lock
+	 */
+	struct list_head pending_fences;
+	/**
+	 * @tlb_inval.pending_lock: protects @tlb_inval.pending_fences
+	 * and updating @tlb_inval.seqno_recv.
+	 */
+	spinlock_t pending_lock;
+	/**
+	 * @tlb_inval.fence_tdr: schedules a delayed call to
+	 * xe_gt_tlb_fence_timeout after the timeut interval is over.
+	 */
+	struct delayed_work fence_tdr;
+	/** @wtlb_invalidation.wq: schedules GT TLB invalidation jobs */
+	struct workqueue_struct *job_wq;
+	/** @tlb_inval.lock: protects TLB invalidation fences */
+	spinlock_t lock;
+};
+
 /**
  * struct xe_gt_tlb_inval_fence - XE GT TLB invalidation fence
  *
diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
index 85cfcc49472b..7dc5a3f310f1 100644
--- a/drivers/gpu/drm/xe/xe_gt_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_types.h
@@ -12,6 +12,7 @@
 #include "xe_gt_sriov_pf_types.h"
 #include "xe_gt_sriov_vf_types.h"
 #include "xe_gt_stats_types.h"
+#include "xe_gt_tlb_inval_types.h"
 #include "xe_hw_engine_types.h"
 #include "xe_hw_fence_types.h"
 #include "xe_oa_types.h"
@@ -186,37 +187,7 @@ struct xe_gt {
 	} reset;
 
 	/** @tlb_inval: TLB invalidation state */
-	struct {
-		/** @tlb_inval.seqno: TLB invalidation seqno, protected by CT lock */
-#define TLB_INVALIDATION_SEQNO_MAX	0x100000
-		int seqno;
-		/** @tlb_invalidation.seqno_lock: protects @tlb_invalidation.seqno */
-		struct mutex seqno_lock;
-		/**
-		 * @tlb_inval.seqno_recv: last received TLB invalidation seqno,
-		 * protected by CT lock
-		 */
-		int seqno_recv;
-		/**
-		 * @tlb_inval.pending_fences: list of pending fences waiting TLB
-		 * invaliations, protected by CT lock
-		 */
-		struct list_head pending_fences;
-		/**
-		 * @tlb_inval.pending_lock: protects @tlb_inval.pending_fences
-		 * and updating @tlb_inval.seqno_recv.
-		 */
-		spinlock_t pending_lock;
-		/**
-		 * @tlb_inval.fence_tdr: schedules a delayed call to
-		 * xe_gt_tlb_fence_timeout after the timeut interval is over.
-		 */
-		struct delayed_work fence_tdr;
-		/** @wtlb_invalidation.wq: schedules GT TLB invalidation jobs */
-		struct workqueue_struct *job_wq;
-		/** @tlb_inval.lock: protects TLB invalidation fences */
-		spinlock_t lock;
-	} tlb_inval;
+	struct xe_tlb_inval tlb_inval;
 
 	/**
 	 * @ccs_mode: Number of compute engines enabled.

From 6d1e452e0948fd69ec3aeea0e52807152dc93c2d Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:07 +0000
Subject: [PATCH 0701/1455] drm/xe: Add xe_gt_tlb_invalidation_done_handler

Decouple GT TLB seqno handling from G2H handler.

v2:
 - Add kernel doc

Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-6-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/xe_gt_tlb_inval.c | 47 ++++++++++++++++++----------
 1 file changed, 30 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval.c b/drivers/gpu/drm/xe/xe_gt_tlb_inval.c
index 1571fd917830..37b3b45ec230 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval.c
+++ b/drivers/gpu/drm/xe/xe_gt_tlb_inval.c
@@ -506,27 +506,18 @@ void xe_gt_tlb_inval_vm(struct xe_gt *gt, struct xe_vm *vm)
 }
 
 /**
- * xe_guc_tlb_inval_done_handler - TLB invalidation done handler
- * @guc: guc
- * @msg: message indicating TLB invalidation done
- * @len: length of message
+ * xe_gt_tlb_inval_done_handler - GT TLB invalidation done handler
+ * @gt: gt
+ * @seqno: seqno of invalidation that is done
  *
- * Parse seqno of TLB invalidation, wake any waiters for seqno, and signal any
- * invalidation fences for seqno. Algorithm for this depends on seqno being
- * received in-order and asserts this assumption.
- *
- * Return: 0 on success, -EPROTO for malformed messages.
+ * Update recv seqno, signal any GT TLB invalidation fences, and restart TDR
  */
-int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
+static void xe_gt_tlb_inval_done_handler(struct xe_gt *gt, int seqno)
 {
-	struct xe_gt *gt = guc_to_gt(guc);
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_gt_tlb_inval_fence *fence, *next;
 	unsigned long flags;
 
-	if (unlikely(len != 1))
-		return -EPROTO;
-
 	/*
 	 * This can also be run both directly from the IRQ handler and also in
 	 * process_g2h_msg(). Only one may process any individual CT message,
@@ -543,12 +534,12 @@ int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
 	 * process_g2h_msg().
 	 */
 	spin_lock_irqsave(&gt->tlb_inval.pending_lock, flags);
-	if (tlb_inval_seqno_past(gt, msg[0])) {
+	if (tlb_inval_seqno_past(gt, seqno)) {
 		spin_unlock_irqrestore(&gt->tlb_inval.pending_lock, flags);
-		return 0;
+		return;
 	}
 
-	WRITE_ONCE(gt->tlb_inval.seqno_recv, msg[0]);
+	WRITE_ONCE(gt->tlb_inval.seqno_recv, seqno);
 
 	list_for_each_entry_safe(fence, next,
 				 &gt->tlb_inval.pending_fences, link) {
@@ -568,6 +559,28 @@ int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
 		cancel_delayed_work(&gt->tlb_inval.fence_tdr);
 
 	spin_unlock_irqrestore(&gt->tlb_inval.pending_lock, flags);
+}
+
+/**
+ * xe_guc_tlb_inval_done_handler - TLB invalidation done handler
+ * @guc: guc
+ * @msg: message indicating TLB invalidation done
+ * @len: length of message
+ *
+ * Parse seqno of TLB invalidation, wake any waiters for seqno, and signal any
+ * invalidation fences for seqno. Algorithm for this depends on seqno being
+ * received in-order and asserts this assumption.
+ *
+ * Return: 0 on success, -EPROTO for malformed messages.
+ */
+int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
+{
+	struct xe_gt *gt = guc_to_gt(guc);
+
+	if (unlikely(len != 1))
+		return -EPROTO;
+
+	xe_gt_tlb_inval_done_handler(gt, msg[0]);
 
 	return 0;
 }

From 15366239e2130ebe662351c0995590f2af2ba3b3 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:08 +0000
Subject: [PATCH 0702/1455] drm/xe: Decouple TLB invalidations from GT

Decouple TLB invalidations from the GT by updating the TLB invalidation
layer to accept a `struct xe_tlb_inval` instead of a `struct xe_gt`.
Also, rename *gt_tlb* to *tlb*. The internals of the TLB invalidation
code still operate on a GT, but this is now hidden from the rest of the
driver.

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-7-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/Makefile                   |   4 +-
 drivers/gpu/drm/xe/xe_ggtt.c                  |   4 +-
 drivers/gpu/drm/xe/xe_gt.c                    |   6 +-
 drivers/gpu/drm/xe/xe_gt_tlb_inval.h          |  41 -----
 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h      |  34 ----
 drivers/gpu/drm/xe/xe_gt_types.h              |   2 +-
 drivers/gpu/drm/xe/xe_guc_ct.c                |   2 +-
 drivers/gpu/drm/xe/xe_lmtt.c                  |  12 +-
 drivers/gpu/drm/xe/xe_migrate.h               |  10 +-
 drivers/gpu/drm/xe/xe_pt.c                    |  63 ++++---
 drivers/gpu/drm/xe/xe_svm.c                   |   1 -
 .../xe/{xe_gt_tlb_inval.c => xe_tlb_inval.c}  | 142 +++++++++-------
 drivers/gpu/drm/xe/xe_tlb_inval.h             |  41 +++++
 ..._gt_tlb_inval_job.c => xe_tlb_inval_job.c} | 154 +++++++++---------
 drivers/gpu/drm/xe/xe_tlb_inval_job.h         |  33 ++++
 ...tlb_inval_types.h => xe_tlb_inval_types.h} |  35 ++--
 drivers/gpu/drm/xe/xe_trace.h                 |  24 +--
 drivers/gpu/drm/xe/xe_vm.c                    |  26 +--
 18 files changed, 330 insertions(+), 304 deletions(-)
 delete mode 100644 drivers/gpu/drm/xe/xe_gt_tlb_inval.h
 delete mode 100644 drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h
 rename drivers/gpu/drm/xe/{xe_gt_tlb_inval.c => xe_tlb_inval.c} (80%)
 create mode 100644 drivers/gpu/drm/xe/xe_tlb_inval.h
 rename drivers/gpu/drm/xe/{xe_gt_tlb_inval_job.c => xe_tlb_inval_job.c} (51%)
 create mode 100644 drivers/gpu/drm/xe/xe_tlb_inval_job.h
 rename drivers/gpu/drm/xe/{xe_gt_tlb_inval_types.h => xe_tlb_inval_types.h} (56%)

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index a1e4b3b1a1b6..2b67debb8bc4 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -61,8 +61,6 @@ xe-y += xe_bb.o \
 	xe_gt_pagefault.o \
 	xe_gt_sysfs.o \
 	xe_gt_throttle.o \
-	xe_gt_tlb_inval.o \
-	xe_gt_tlb_inval_job.o \
 	xe_gt_topology.o \
 	xe_guc.o \
 	xe_guc_ads.o \
@@ -119,6 +117,8 @@ xe-y += xe_bb.o \
 	xe_sync.o \
 	xe_tile.o \
 	xe_tile_sysfs.o \
+	xe_tlb_inval.o \
+	xe_tlb_inval_job.o \
 	xe_trace.o \
 	xe_trace_bo.o \
 	xe_trace_guc.o \
diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index c3e46c270117..71c7690a92b3 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -23,13 +23,13 @@
 #include "xe_device.h"
 #include "xe_gt.h"
 #include "xe_gt_printk.h"
-#include "xe_gt_tlb_inval.h"
 #include "xe_map.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
 #include "xe_res_cursor.h"
 #include "xe_sriov.h"
 #include "xe_tile_sriov_vf.h"
+#include "xe_tlb_inval.h"
 #include "xe_wa.h"
 #include "xe_wopcm.h"
 
@@ -438,7 +438,7 @@ static void ggtt_invalidate_gt_tlb(struct xe_gt *gt)
 	if (!gt)
 		return;
 
-	err = xe_gt_tlb_inval_ggtt(gt);
+	err = xe_tlb_inval_ggtt(&gt->tlb_inval);
 	xe_gt_WARN(gt, err, "Failed to invalidate GGTT (%pe)", ERR_PTR(err));
 }
 
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 9a4639732bd7..67ee7cdbd6ec 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -37,7 +37,6 @@
 #include "xe_gt_sriov_pf.h"
 #include "xe_gt_sriov_vf.h"
 #include "xe_gt_sysfs.h"
-#include "xe_gt_tlb_inval.h"
 #include "xe_gt_topology.h"
 #include "xe_guc_exec_queue_types.h"
 #include "xe_guc_pc.h"
@@ -58,6 +57,7 @@
 #include "xe_sa.h"
 #include "xe_sched_job.h"
 #include "xe_sriov.h"
+#include "xe_tlb_inval.h"
 #include "xe_tuning.h"
 #include "xe_uc.h"
 #include "xe_uc_fw.h"
@@ -852,7 +852,7 @@ static int gt_reset(struct xe_gt *gt)
 
 	xe_uc_stop(&gt->uc);
 
-	xe_gt_tlb_inval_reset(gt);
+	xe_tlb_inval_reset(&gt->tlb_inval);
 
 	err = do_gt_reset(gt);
 	if (err)
@@ -1066,5 +1066,5 @@ void xe_gt_declare_wedged(struct xe_gt *gt)
 	xe_gt_assert(gt, gt_to_xe(gt)->wedged.mode);
 
 	xe_uc_declare_wedged(&gt->uc);
-	xe_gt_tlb_inval_reset(gt);
+	xe_tlb_inval_reset(&gt->tlb_inval);
 }
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval.h b/drivers/gpu/drm/xe/xe_gt_tlb_inval.h
deleted file mode 100644
index b1258ac4e8fb..000000000000
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval.h
+++ /dev/null
@@ -1,41 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright © 2023 Intel Corporation
- */
-
-#ifndef _XE_GT_TLB_INVAL_H_
-#define _XE_GT_TLB_INVAL_H_
-
-#include <linux/types.h>
-
-#include "xe_gt_tlb_inval_types.h"
-
-struct xe_gt;
-struct xe_guc;
-struct xe_vm;
-struct xe_vma;
-
-int xe_gt_tlb_inval_init_early(struct xe_gt *gt);
-void xe_gt_tlb_inval_fini(struct xe_gt *gt);
-
-void xe_gt_tlb_inval_reset(struct xe_gt *gt);
-int xe_gt_tlb_inval_ggtt(struct xe_gt *gt);
-void xe_gt_tlb_inval_vm(struct xe_gt *gt, struct xe_vm *vm);
-int xe_gt_tlb_inval_all(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence);
-int xe_gt_tlb_inval_range(struct xe_gt *gt,
-			  struct xe_gt_tlb_inval_fence *fence,
-			  u64 start, u64 end, u32 asid);
-int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len);
-
-void xe_gt_tlb_inval_fence_init(struct xe_gt *gt,
-				struct xe_gt_tlb_inval_fence *fence,
-				bool stack);
-void xe_gt_tlb_inval_fence_signal(struct xe_gt_tlb_inval_fence *fence);
-
-static inline void
-xe_gt_tlb_inval_fence_wait(struct xe_gt_tlb_inval_fence *fence)
-{
-	dma_fence_wait(&fence->base, false);
-}
-
-#endif	/* _XE_GT_TLB_INVAL_ */
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h b/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h
deleted file mode 100644
index 883896194a34..000000000000
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.h
+++ /dev/null
@@ -1,34 +0,0 @@
-/* SPDX-License-Identifier: MIT */
-/*
- * Copyright © 2025 Intel Corporation
- */
-
-#ifndef _XE_GT_TLB_INVAL_JOB_H_
-#define _XE_GT_TLB_INVAL_JOB_H_
-
-#include <linux/types.h>
-
-struct dma_fence;
-struct drm_sched_job;
-struct kref;
-struct xe_exec_queue;
-struct xe_gt;
-struct xe_gt_tlb_inval_job;
-struct xe_migrate;
-
-struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
-						       struct xe_gt *gt,
-						       u64 start, u64 end,
-						       u32 asid);
-
-int xe_gt_tlb_inval_job_alloc_dep(struct xe_gt_tlb_inval_job *job);
-
-struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
-					   struct xe_migrate *m,
-					   struct dma_fence *fence);
-
-void xe_gt_tlb_inval_job_get(struct xe_gt_tlb_inval_job *job);
-
-void xe_gt_tlb_inval_job_put(struct xe_gt_tlb_inval_job *job);
-
-#endif
diff --git a/drivers/gpu/drm/xe/xe_gt_types.h b/drivers/gpu/drm/xe/xe_gt_types.h
index 7dc5a3f310f1..66158105aca5 100644
--- a/drivers/gpu/drm/xe/xe_gt_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_types.h
@@ -12,12 +12,12 @@
 #include "xe_gt_sriov_pf_types.h"
 #include "xe_gt_sriov_vf_types.h"
 #include "xe_gt_stats_types.h"
-#include "xe_gt_tlb_inval_types.h"
 #include "xe_hw_engine_types.h"
 #include "xe_hw_fence_types.h"
 #include "xe_oa_types.h"
 #include "xe_reg_sr_types.h"
 #include "xe_sa_types.h"
+#include "xe_tlb_inval_types.h"
 #include "xe_uc_types.h"
 
 struct xe_exec_queue_ops;
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 9131d121d941..5f38041cff4c 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -26,13 +26,13 @@
 #include "xe_gt_sriov_pf_control.h"
 #include "xe_gt_sriov_pf_monitor.h"
 #include "xe_gt_sriov_printk.h"
-#include "xe_gt_tlb_inval.h"
 #include "xe_guc.h"
 #include "xe_guc_log.h"
 #include "xe_guc_relay.h"
 #include "xe_guc_submit.h"
 #include "xe_map.h"
 #include "xe_pm.h"
+#include "xe_tlb_inval.h"
 #include "xe_trace_guc.h"
 
 static void receive_g2h(struct xe_guc_ct *ct);
diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c
index e5aba03ff8ac..f2bfbfa3efa1 100644
--- a/drivers/gpu/drm/xe/xe_lmtt.c
+++ b/drivers/gpu/drm/xe/xe_lmtt.c
@@ -11,7 +11,7 @@
 
 #include "xe_assert.h"
 #include "xe_bo.h"
-#include "xe_gt_tlb_inval.h"
+#include "xe_tlb_inval.h"
 #include "xe_lmtt.h"
 #include "xe_map.h"
 #include "xe_mmio.h"
@@ -228,8 +228,8 @@ void xe_lmtt_init_hw(struct xe_lmtt *lmtt)
 
 static int lmtt_invalidate_hw(struct xe_lmtt *lmtt)
 {
-	struct xe_gt_tlb_inval_fence fences[XE_MAX_GT_PER_TILE];
-	struct xe_gt_tlb_inval_fence *fence = fences;
+	struct xe_tlb_inval_fence fences[XE_MAX_GT_PER_TILE];
+	struct xe_tlb_inval_fence *fence = fences;
 	struct xe_tile *tile = lmtt_to_tile(lmtt);
 	struct xe_gt *gt;
 	int result = 0;
@@ -237,8 +237,8 @@ static int lmtt_invalidate_hw(struct xe_lmtt *lmtt)
 	u8 id;
 
 	for_each_gt_on_tile(gt, tile, id) {
-		xe_gt_tlb_inval_fence_init(gt, fence, true);
-		err = xe_gt_tlb_inval_all(gt, fence);
+		xe_tlb_inval_fence_init(&gt->tlb_inval, fence, true);
+		err = xe_tlb_inval_all(&gt->tlb_inval, fence);
 		result = result ?: err;
 		fence++;
 	}
@@ -252,7 +252,7 @@ static int lmtt_invalidate_hw(struct xe_lmtt *lmtt)
 	 */
 	fence = fences;
 	for_each_gt_on_tile(gt, tile, id)
-		xe_gt_tlb_inval_fence_wait(fence++);
+		xe_tlb_inval_fence_wait(fence++);
 
 	return result;
 }
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 8978d2cc1a75..4fad324b6253 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -15,7 +15,7 @@ struct ttm_resource;
 
 struct xe_bo;
 struct xe_gt;
-struct xe_gt_tlb_inval_job;
+struct xe_tlb_inval_job;
 struct xe_exec_queue;
 struct xe_migrate;
 struct xe_migrate_pt_update;
@@ -94,13 +94,13 @@ struct xe_migrate_pt_update {
 	/** @job: The job if a GPU page-table update. NULL otherwise */
 	struct xe_sched_job *job;
 	/**
-	 * @ijob: The GT TLB invalidation job for primary tile. NULL otherwise
+	 * @ijob: The TLB invalidation job for primary GT. NULL otherwise
 	 */
-	struct xe_gt_tlb_inval_job *ijob;
+	struct xe_tlb_inval_job *ijob;
 	/**
-	 * @mjob: The GT TLB invalidation job for media tile. NULL otherwise
+	 * @mjob: The TLB invalidation job for media GT. NULL otherwise
 	 */
-	struct xe_gt_tlb_inval_job *mjob;
+	struct xe_tlb_inval_job *mjob;
 	/** @tile_id: Tile ID of the update */
 	u8 tile_id;
 };
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index c0a70c80dff9..c129048a9a09 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -13,7 +13,7 @@
 #include "xe_drm_client.h"
 #include "xe_exec_queue.h"
 #include "xe_gt.h"
-#include "xe_gt_tlb_inval_job.h"
+#include "xe_tlb_inval_job.h"
 #include "xe_migrate.h"
 #include "xe_pt_types.h"
 #include "xe_pt_walk.h"
@@ -21,6 +21,7 @@
 #include "xe_sched_job.h"
 #include "xe_sync.h"
 #include "xe_svm.h"
+#include "xe_tlb_inval_job.h"
 #include "xe_trace.h"
 #include "xe_ttm_stolen_mgr.h"
 #include "xe_vm.h"
@@ -1276,8 +1277,8 @@ static int op_add_deps(struct xe_vm *vm, struct xe_vma_op *op,
 }
 
 static int xe_pt_vm_dependencies(struct xe_sched_job *job,
-				 struct xe_gt_tlb_inval_job *ijob,
-				 struct xe_gt_tlb_inval_job *mjob,
+				 struct xe_tlb_inval_job *ijob,
+				 struct xe_tlb_inval_job *mjob,
 				 struct xe_vm *vm,
 				 struct xe_vma_ops *vops,
 				 struct xe_vm_pgtable_update_ops *pt_update_ops,
@@ -1347,13 +1348,13 @@ static int xe_pt_vm_dependencies(struct xe_sched_job *job,
 
 	if (job) {
 		if (ijob) {
-			err = xe_gt_tlb_inval_job_alloc_dep(ijob);
+			err = xe_tlb_inval_job_alloc_dep(ijob);
 			if (err)
 				return err;
 		}
 
 		if (mjob) {
-			err = xe_gt_tlb_inval_job_alloc_dep(mjob);
+			err = xe_tlb_inval_job_alloc_dep(mjob);
 			if (err)
 				return err;
 		}
@@ -2353,6 +2354,15 @@ static const struct xe_migrate_pt_update_ops svm_migrate_ops = {
 static const struct xe_migrate_pt_update_ops svm_migrate_ops;
 #endif
 
+static struct xe_dep_scheduler *to_dep_scheduler(struct xe_exec_queue *q,
+						 struct xe_gt *gt)
+{
+	if (xe_gt_is_media_type(gt))
+		return q->tlb_inval[XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT].dep_scheduler;
+
+	return q->tlb_inval[XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT].dep_scheduler;
+}
+
 /**
  * xe_pt_update_ops_run() - Run PT update operations
  * @tile: Tile of PT update operations
@@ -2371,7 +2381,7 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	struct xe_vm_pgtable_update_ops *pt_update_ops =
 		&vops->pt_update_ops[tile->id];
 	struct dma_fence *fence, *ifence, *mfence;
-	struct xe_gt_tlb_inval_job *ijob = NULL, *mjob = NULL;
+	struct xe_tlb_inval_job *ijob = NULL, *mjob = NULL;
 	struct dma_fence **fences = NULL;
 	struct dma_fence_array *cf = NULL;
 	struct xe_range_fence *rfence;
@@ -2403,11 +2413,15 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 #endif
 
 	if (pt_update_ops->needs_invalidation) {
-		ijob = xe_gt_tlb_inval_job_create(pt_update_ops->q,
-						  tile->primary_gt,
-						  pt_update_ops->start,
-						  pt_update_ops->last,
-						  vm->usm.asid);
+		struct xe_exec_queue *q = pt_update_ops->q;
+		struct xe_dep_scheduler *dep_scheduler =
+			to_dep_scheduler(q, tile->primary_gt);
+
+		ijob = xe_tlb_inval_job_create(q, &tile->primary_gt->tlb_inval,
+					       dep_scheduler,
+					       pt_update_ops->start,
+					       pt_update_ops->last,
+					       vm->usm.asid);
 		if (IS_ERR(ijob)) {
 			err = PTR_ERR(ijob);
 			goto kill_vm_tile1;
@@ -2415,11 +2429,14 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 		update.ijob = ijob;
 
 		if (tile->media_gt) {
-			mjob = xe_gt_tlb_inval_job_create(pt_update_ops->q,
-							  tile->media_gt,
-							  pt_update_ops->start,
-							  pt_update_ops->last,
-							  vm->usm.asid);
+			dep_scheduler = to_dep_scheduler(q, tile->media_gt);
+
+			mjob = xe_tlb_inval_job_create(q,
+						       &tile->media_gt->tlb_inval,
+						       dep_scheduler,
+						       pt_update_ops->start,
+						       pt_update_ops->last,
+						       vm->usm.asid);
 			if (IS_ERR(mjob)) {
 				err = PTR_ERR(mjob);
 				goto free_ijob;
@@ -2470,13 +2487,13 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	if (ijob) {
 		struct dma_fence *__fence;
 
-		ifence = xe_gt_tlb_inval_job_push(ijob, tile->migrate, fence);
+		ifence = xe_tlb_inval_job_push(ijob, tile->migrate, fence);
 		__fence = ifence;
 
 		if (mjob) {
 			fences[0] = ifence;
-			mfence = xe_gt_tlb_inval_job_push(mjob, tile->migrate,
-							  fence);
+			mfence = xe_tlb_inval_job_push(mjob, tile->migrate,
+						       fence);
 			fences[1] = mfence;
 
 			dma_fence_array_init(cf, 2, fences,
@@ -2519,8 +2536,8 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	if (pt_update_ops->needs_userptr_lock)
 		up_read(&vm->userptr.notifier_lock);
 
-	xe_gt_tlb_inval_job_put(mjob);
-	xe_gt_tlb_inval_job_put(ijob);
+	xe_tlb_inval_job_put(mjob);
+	xe_tlb_inval_job_put(ijob);
 
 	return fence;
 
@@ -2529,8 +2546,8 @@ free_rfence:
 free_ijob:
 	kfree(cf);
 	kfree(fences);
-	xe_gt_tlb_inval_job_put(mjob);
-	xe_gt_tlb_inval_job_put(ijob);
+	xe_tlb_inval_job_put(mjob);
+	xe_tlb_inval_job_put(ijob);
 kill_vm_tile1:
 	if (err != -EAGAIN && err != -ENODATA && tile->id)
 		xe_vm_kill(vops->vm, false);
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index babc64485b52..76c6d74c1208 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -7,7 +7,6 @@
 
 #include "xe_bo.h"
 #include "xe_gt_stats.h"
-#include "xe_gt_tlb_inval.h"
 #include "xe_migrate.h"
 #include "xe_module.h"
 #include "xe_pm.h"
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval.c b/drivers/gpu/drm/xe/xe_tlb_inval.c
similarity index 80%
rename from drivers/gpu/drm/xe/xe_gt_tlb_inval.c
rename to drivers/gpu/drm/xe/xe_tlb_inval.c
index 37b3b45ec230..f4b7c0c74894 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval.c
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.c
@@ -13,7 +13,7 @@
 #include "xe_guc.h"
 #include "xe_guc_ct.h"
 #include "xe_gt_stats.h"
-#include "xe_gt_tlb_inval.h"
+#include "xe_tlb_inval.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
 #include "xe_sriov.h"
@@ -38,40 +38,47 @@ static long tlb_timeout_jiffies(struct xe_gt *gt)
 	return hw_tlb_timeout + 2 * delay;
 }
 
-static void xe_gt_tlb_inval_fence_fini(struct xe_gt_tlb_inval_fence *fence)
+static void xe_tlb_inval_fence_fini(struct xe_tlb_inval_fence *fence)
 {
-	if (WARN_ON_ONCE(!fence->gt))
+	struct xe_gt *gt;
+
+	if (WARN_ON_ONCE(!fence->tlb_inval))
 		return;
 
-	xe_pm_runtime_put(gt_to_xe(fence->gt));
-	fence->gt = NULL; /* fini() should be called once */
+	gt = fence->tlb_inval->private;
+
+	xe_pm_runtime_put(gt_to_xe(gt));
+	fence->tlb_inval = NULL; /* fini() should be called once */
 }
 
 static void
-__inval_fence_signal(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence)
+__inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
 {
 	bool stack = test_bit(FENCE_STACK_BIT, &fence->base.flags);
 
-	trace_xe_gt_tlb_inval_fence_signal(xe, fence);
-	xe_gt_tlb_inval_fence_fini(fence);
+	trace_xe_tlb_inval_fence_signal(xe, fence);
+	xe_tlb_inval_fence_fini(fence);
 	dma_fence_signal(&fence->base);
 	if (!stack)
 		dma_fence_put(&fence->base);
 }
 
 static void
-inval_fence_signal(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence)
+inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
 {
 	list_del(&fence->link);
 	__inval_fence_signal(xe, fence);
 }
 
-void xe_gt_tlb_inval_fence_signal(struct xe_gt_tlb_inval_fence *fence)
+void xe_tlb_inval_fence_signal(struct xe_tlb_inval_fence *fence)
 {
-	if (WARN_ON_ONCE(!fence->gt))
+	struct xe_gt *gt;
+
+	if (WARN_ON_ONCE(!fence->tlb_inval))
 		return;
 
-	__inval_fence_signal(gt_to_xe(fence->gt), fence);
+	gt = fence->tlb_inval->private;
+	__inval_fence_signal(gt_to_xe(gt), fence);
 }
 
 static void xe_gt_tlb_fence_timeout(struct work_struct *work)
@@ -79,7 +86,7 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
 	struct xe_gt *gt = container_of(work, struct xe_gt,
 					tlb_inval.fence_tdr.work);
 	struct xe_device *xe = gt_to_xe(gt);
-	struct xe_gt_tlb_inval_fence *fence, *next;
+	struct xe_tlb_inval_fence *fence, *next;
 
 	LNL_FLUSH_WORK(&gt->uc.guc.ct.g2h_worker);
 
@@ -92,7 +99,7 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
 		if (msecs_to_jiffies(since_inval_ms) < tlb_timeout_jiffies(gt))
 			break;
 
-		trace_xe_gt_tlb_inval_fence_timeout(xe, fence);
+		trace_xe_tlb_inval_fence_timeout(xe, fence);
 		xe_gt_err(gt, "TLB invalidation fence timeout, seqno=%d recv=%d",
 			  fence->seqno, gt->tlb_inval.seqno_recv);
 
@@ -110,7 +117,7 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
  * xe_gt_tlb_inval_init_early - Initialize GT TLB invalidation state
  * @gt: GT structure
  *
- * Initialize GT TLB invalidation state, purely software initialization, should
+ * Initialize TLB invalidation state, purely software initialization, should
  * be called once during driver load.
  *
  * Return: 0 on success, negative error code on error.
@@ -120,6 +127,7 @@ int xe_gt_tlb_inval_init_early(struct xe_gt *gt)
 	struct xe_device *xe = gt_to_xe(gt);
 	int err;
 
+	gt->tlb_inval.private = gt;
 	gt->tlb_inval.seqno = 1;
 	INIT_LIST_HEAD(&gt->tlb_inval.pending_fences);
 	spin_lock_init(&gt->tlb_inval.pending_lock);
@@ -141,14 +149,15 @@ int xe_gt_tlb_inval_init_early(struct xe_gt *gt)
 }
 
 /**
- * xe_gt_tlb_inval_reset - Initialize GT TLB invalidation reset
- * @gt: GT structure
+ * xe_tlb_inval_reset - Initialize TLB invalidation reset
+ * @tlb_inval: TLB invalidation client
  *
  * Signal any pending invalidation fences, should be called during a GT reset
  */
-void xe_gt_tlb_inval_reset(struct xe_gt *gt)
+void xe_tlb_inval_reset(struct xe_tlb_inval *tlb_inval)
 {
-	struct xe_gt_tlb_inval_fence *fence, *next;
+	struct xe_gt *gt = tlb_inval->private;
+	struct xe_tlb_inval_fence *fence, *next;
 	int pending_seqno;
 
 	/*
@@ -213,7 +222,7 @@ static bool tlb_inval_seqno_past(struct xe_gt *gt, int seqno)
 }
 
 static int send_tlb_inval(struct xe_guc *guc,
-			  struct xe_gt_tlb_inval_fence *fence,
+			  struct xe_tlb_inval_fence *fence,
 			  u32 *action, int len)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
@@ -232,7 +241,7 @@ static int send_tlb_inval(struct xe_guc *guc,
 	mutex_lock(&gt->tlb_inval.seqno_lock);
 	seqno = gt->tlb_inval.seqno;
 	fence->seqno = seqno;
-	trace_xe_gt_tlb_inval_fence_send(xe, fence);
+	trace_xe_tlb_inval_fence_send(xe, fence);
 	action[1] = seqno;
 	ret = xe_guc_ct_send(&guc->ct, action, len,
 			     G2H_LEN_DW_TLB_INVALIDATE, 1);
@@ -277,7 +286,7 @@ static int send_tlb_inval(struct xe_guc *guc,
 		XE_GUC_TLB_INVAL_FLUSH_CACHE)
 
 /**
- * xe_gt_tlb_inval_guc - Issue a TLB invalidation on this GT for the GuC
+ * xe_tlb_inval_guc - Issue a TLB invalidation on this GT for the GuC
  * @gt: GT structure
  * @fence: invalidation fence which will be signal on TLB invalidation
  * completion
@@ -287,8 +296,8 @@ static int send_tlb_inval(struct xe_guc *guc,
  *
  * Return: 0 on success, negative error code on error
  */
-static int xe_gt_tlb_inval_guc(struct xe_gt *gt,
-			       struct xe_gt_tlb_inval_fence *fence)
+static int xe_tlb_inval_guc(struct xe_gt *gt,
+			    struct xe_tlb_inval_fence *fence)
 {
 	u32 action[] = {
 		XE_GUC_ACTION_TLB_INVALIDATION,
@@ -309,30 +318,31 @@ static int xe_gt_tlb_inval_guc(struct xe_gt *gt,
 }
 
 /**
- * xe_gt_tlb_inval_ggtt - Issue a TLB invalidation on this GT for the GGTT
- * @gt: GT structure
+ * xe_tlb_inval_ggtt - Issue a TLB invalidation on this GT for the GGTT
+ * @tlb_inval: TLB invalidation client
  *
  * Issue a TLB invalidation for the GGTT. Completion of TLB invalidation is
  * synchronous.
  *
  * Return: 0 on success, negative error code on error
  */
-int xe_gt_tlb_inval_ggtt(struct xe_gt *gt)
+int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval)
 {
+	struct xe_gt *gt = tlb_inval->private;
 	struct xe_device *xe = gt_to_xe(gt);
 	unsigned int fw_ref;
 
 	if (xe_guc_ct_enabled(&gt->uc.guc.ct) &&
 	    gt->uc.guc.submission_state.enabled) {
-		struct xe_gt_tlb_inval_fence fence;
+		struct xe_tlb_inval_fence fence;
 		int ret;
 
-		xe_gt_tlb_inval_fence_init(gt, &fence, true);
-		ret = xe_gt_tlb_inval_guc(gt, &fence);
+		xe_tlb_inval_fence_init(tlb_inval, &fence, true);
+		ret = xe_tlb_inval_guc(gt, &fence);
 		if (ret)
 			return ret;
 
-		xe_gt_tlb_inval_fence_wait(&fence);
+		xe_tlb_inval_fence_wait(&fence);
 	} else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) {
 		struct xe_mmio *mmio = &gt->mmio;
 
@@ -355,14 +365,17 @@ int xe_gt_tlb_inval_ggtt(struct xe_gt *gt)
 	return 0;
 }
 
-static int send_tlb_inval_all(struct xe_gt *gt,
-			      struct xe_gt_tlb_inval_fence *fence)
+static int send_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
+			      struct xe_tlb_inval_fence *fence)
 {
 	u32 action[] = {
 		XE_GUC_ACTION_TLB_INVALIDATION_ALL,
 		0,  /* seqno, replaced in send_tlb_inval */
 		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL),
 	};
+	struct xe_gt *gt = tlb_inval->private;
+
+	xe_gt_assert(gt, fence);
 
 	return send_tlb_inval(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
 }
@@ -370,19 +383,19 @@ static int send_tlb_inval_all(struct xe_gt *gt,
 /**
  * xe_gt_tlb_invalidation_all - Invalidate all TLBs across PF and all VFs.
  * @gt: the &xe_gt structure
- * @fence: the &xe_gt_tlb_inval_fence to be signaled on completion
+ * @fence: the &xe_tlb_inval_fence to be signaled on completion
  *
  * Send a request to invalidate all TLBs across PF and all VFs.
  *
  * Return: 0 on success, negative error code on error
  */
-int xe_gt_tlb_inval_all(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence)
+int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
+		     struct xe_tlb_inval_fence *fence)
 {
+	struct xe_gt *gt = tlb_inval->private;
 	int err;
 
-	xe_gt_assert(gt, gt == fence->gt);
-
-	err = send_tlb_inval_all(gt, fence);
+	err = send_tlb_inval_all(tlb_inval, fence);
 	if (err)
 		xe_gt_err(gt, "TLB invalidation request failed (%pe)", ERR_PTR(err));
 
@@ -397,9 +410,8 @@ int xe_gt_tlb_inval_all(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence)
 #define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX))
 
 /**
- * xe_gt_tlb_inval_range - Issue a TLB invalidation on this GT for an address range
- *
- * @gt: GT structure
+ * xe_tlb_inval_range - Issue a TLB invalidation on this GT for an address range
+ * @tlb_inval: TLB invalidation client
  * @fence: invalidation fence which will be signal on TLB invalidation
  * completion
  * @start: start address
@@ -412,9 +424,11 @@ int xe_gt_tlb_inval_all(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence)
  *
  * Return: Negative error code on error, 0 on success
  */
-int xe_gt_tlb_inval_range(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence,
-			  u64 start, u64 end, u32 asid)
+int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
+		       struct xe_tlb_inval_fence *fence, u64 start, u64 end,
+		       u32 asid)
 {
+	struct xe_gt *gt = tlb_inval->private;
 	struct xe_device *xe = gt_to_xe(gt);
 #define MAX_TLB_INVALIDATION_LEN	7
 	u32 action[MAX_TLB_INVALIDATION_LEN];
@@ -484,38 +498,38 @@ int xe_gt_tlb_inval_range(struct xe_gt *gt, struct xe_gt_tlb_inval_fence *fence,
 }
 
 /**
- * xe_gt_tlb_inval_vm - Issue a TLB invalidation on this GT for a VM
- * @gt: graphics tile
+ * xe_tlb_inval_vm - Issue a TLB invalidation on this GT for a VM
+ * @tlb_inval: TLB invalidation client
  * @vm: VM to invalidate
  *
  * Invalidate entire VM's address space
  */
-void xe_gt_tlb_inval_vm(struct xe_gt *gt, struct xe_vm *vm)
+void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm)
 {
-	struct xe_gt_tlb_inval_fence fence;
+	struct xe_tlb_inval_fence fence;
 	u64 range = 1ull << vm->xe->info.va_bits;
 	int ret;
 
-	xe_gt_tlb_inval_fence_init(gt, &fence, true);
+	xe_tlb_inval_fence_init(tlb_inval, &fence, true);
 
-	ret = xe_gt_tlb_inval_range(gt, &fence, 0, range, vm->usm.asid);
+	ret = xe_tlb_inval_range(tlb_inval, &fence, 0, range, vm->usm.asid);
 	if (ret < 0)
 		return;
 
-	xe_gt_tlb_inval_fence_wait(&fence);
+	xe_tlb_inval_fence_wait(&fence);
 }
 
 /**
- * xe_gt_tlb_inval_done_handler - GT TLB invalidation done handler
+ * xe_tlb_inval_done_handler - TLB invalidation done handler
  * @gt: gt
  * @seqno: seqno of invalidation that is done
  *
- * Update recv seqno, signal any GT TLB invalidation fences, and restart TDR
+ * Update recv seqno, signal any TLB invalidation fences, and restart TDR
  */
-static void xe_gt_tlb_inval_done_handler(struct xe_gt *gt, int seqno)
+static void xe_tlb_inval_done_handler(struct xe_gt *gt, int seqno)
 {
 	struct xe_device *xe = gt_to_xe(gt);
-	struct xe_gt_tlb_inval_fence *fence, *next;
+	struct xe_tlb_inval_fence *fence, *next;
 	unsigned long flags;
 
 	/*
@@ -543,7 +557,7 @@ static void xe_gt_tlb_inval_done_handler(struct xe_gt *gt, int seqno)
 
 	list_for_each_entry_safe(fence, next,
 				 &gt->tlb_inval.pending_fences, link) {
-		trace_xe_gt_tlb_inval_fence_recv(xe, fence);
+		trace_xe_tlb_inval_fence_recv(xe, fence);
 
 		if (!tlb_inval_seqno_past(gt, fence->seqno))
 			break;
@@ -580,7 +594,7 @@ int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
 	if (unlikely(len != 1))
 		return -EPROTO;
 
-	xe_gt_tlb_inval_done_handler(gt, msg[0]);
+	xe_tlb_inval_done_handler(gt, msg[0]);
 
 	return 0;
 }
@@ -603,19 +617,21 @@ static const struct dma_fence_ops inval_fence_ops = {
 };
 
 /**
- * xe_gt_tlb_inval_fence_init - Initialize TLB invalidation fence
- * @gt: GT
+ * xe_tlb_inval_fence_init - Initialize TLB invalidation fence
+ * @tlb_inval: TLB invalidation client
  * @fence: TLB invalidation fence to initialize
  * @stack: fence is stack variable
  *
- * Initialize TLB invalidation fence for use. xe_gt_tlb_inval_fence_fini
+ * Initialize TLB invalidation fence for use. xe_tlb_inval_fence_fini
  * will be automatically called when fence is signalled (all fences must signal),
  * even on error.
  */
-void xe_gt_tlb_inval_fence_init(struct xe_gt *gt,
-				struct xe_gt_tlb_inval_fence *fence,
-				bool stack)
+void xe_tlb_inval_fence_init(struct xe_tlb_inval *tlb_inval,
+			     struct xe_tlb_inval_fence *fence,
+			     bool stack)
 {
+	struct xe_gt *gt = tlb_inval->private;
+
 	xe_pm_runtime_get_noresume(gt_to_xe(gt));
 
 	spin_lock_irq(&gt->tlb_inval.lock);
@@ -628,5 +644,5 @@ void xe_gt_tlb_inval_fence_init(struct xe_gt *gt,
 		set_bit(FENCE_STACK_BIT, &fence->base.flags);
 	else
 		dma_fence_get(&fence->base);
-	fence->gt = gt;
+	fence->tlb_inval = tlb_inval;
 }
diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.h b/drivers/gpu/drm/xe/xe_tlb_inval.h
new file mode 100644
index 000000000000..ab6f769c50be
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.h
@@ -0,0 +1,41 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_TLB_INVAL_H_
+#define _XE_TLB_INVAL_H_
+
+#include <linux/types.h>
+
+#include "xe_tlb_inval_types.h"
+
+struct xe_gt;
+struct xe_guc;
+struct xe_vm;
+
+int xe_gt_tlb_inval_init_early(struct xe_gt *gt);
+void xe_gt_tlb_inval_fini(struct xe_gt *gt);
+
+void xe_tlb_inval_reset(struct xe_tlb_inval *tlb_inval);
+int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval);
+void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm);
+int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
+		     struct xe_tlb_inval_fence *fence);
+int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
+		       struct xe_tlb_inval_fence *fence,
+		       u64 start, u64 end, u32 asid);
+int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len);
+
+void xe_tlb_inval_fence_init(struct xe_tlb_inval *tlb_inval,
+			     struct xe_tlb_inval_fence *fence,
+			     bool stack);
+void xe_tlb_inval_fence_signal(struct xe_tlb_inval_fence *fence);
+
+static inline void
+xe_tlb_inval_fence_wait(struct xe_tlb_inval_fence *fence)
+{
+	dma_fence_wait(&fence->base, false);
+}
+
+#endif	/* _XE_TLB_INVAL_ */
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c b/drivers/gpu/drm/xe/xe_tlb_inval_job.c
similarity index 51%
rename from drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
rename to drivers/gpu/drm/xe/xe_tlb_inval_job.c
index 41e0ea92ea5a..492def04a559 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval_job.c
+++ b/drivers/gpu/drm/xe/xe_tlb_inval_job.c
@@ -3,21 +3,22 @@
  * Copyright © 2025 Intel Corporation
  */
 
+#include "xe_assert.h"
 #include "xe_dep_job_types.h"
 #include "xe_dep_scheduler.h"
 #include "xe_exec_queue.h"
-#include "xe_gt.h"
-#include "xe_gt_tlb_inval.h"
-#include "xe_gt_tlb_inval_job.h"
+#include "xe_gt_types.h"
+#include "xe_tlb_inval.h"
+#include "xe_tlb_inval_job.h"
 #include "xe_migrate.h"
 #include "xe_pm.h"
 
-/** struct xe_gt_tlb_inval_job - GT TLB invalidation job */
-struct xe_gt_tlb_inval_job {
+/** struct xe_tlb_inval_job - TLB invalidation job */
+struct xe_tlb_inval_job {
 	/** @dep: base generic dependency Xe job */
 	struct xe_dep_job dep;
-	/** @gt: GT to invalidate */
-	struct xe_gt *gt;
+	/** @tlb_inval: TLB invalidation client */
+	struct xe_tlb_inval *tlb_inval;
 	/** @q: exec queue issuing the invalidate */
 	struct xe_exec_queue *q;
 	/** @refcount: ref count of this job */
@@ -37,63 +38,56 @@ struct xe_gt_tlb_inval_job {
 	bool fence_armed;
 };
 
-static struct dma_fence *xe_gt_tlb_inval_job_run(struct xe_dep_job *dep_job)
+static struct dma_fence *xe_tlb_inval_job_run(struct xe_dep_job *dep_job)
 {
-	struct xe_gt_tlb_inval_job *job =
+	struct xe_tlb_inval_job *job =
 		container_of(dep_job, typeof(*job), dep);
-	struct xe_gt_tlb_inval_fence *ifence =
+	struct xe_tlb_inval_fence *ifence =
 		container_of(job->fence, typeof(*ifence), base);
 
-	xe_gt_tlb_inval_range(job->gt, ifence, job->start,
-			      job->end, job->asid);
+	xe_tlb_inval_range(job->tlb_inval, ifence, job->start,
+			   job->end, job->asid);
 
 	return job->fence;
 }
 
-static void xe_gt_tlb_inval_job_free(struct xe_dep_job *dep_job)
+static void xe_tlb_inval_job_free(struct xe_dep_job *dep_job)
 {
-	struct xe_gt_tlb_inval_job *job =
+	struct xe_tlb_inval_job *job =
 		container_of(dep_job, typeof(*job), dep);
 
-	/* Pairs with get in xe_gt_tlb_inval_job_push */
-	xe_gt_tlb_inval_job_put(job);
+	/* Pairs with get in xe_tlb_inval_job_push */
+	xe_tlb_inval_job_put(job);
 }
 
 static const struct xe_dep_job_ops dep_job_ops = {
-	.run_job = xe_gt_tlb_inval_job_run,
-	.free_job = xe_gt_tlb_inval_job_free,
+	.run_job = xe_tlb_inval_job_run,
+	.free_job = xe_tlb_inval_job_free,
 };
 
-static int xe_gt_tlb_inval_context(struct xe_gt *gt)
-{
-	return xe_gt_is_media_type(gt) ? XE_EXEC_QUEUE_TLB_INVAL_MEDIA_GT :
-		XE_EXEC_QUEUE_TLB_INVAL_PRIMARY_GT;
-}
-
 /**
- * xe_gt_tlb_inval_job_create() - GT TLB invalidation job create
- * @gt: GT to invalidate
+ * xe_tlb_inval_job_create() - TLB invalidation job create
  * @q: exec queue issuing the invalidate
+ * @tlb_inval: TLB invalidation client
+ * @dep_scheduler: Dependency scheduler for job
  * @start: Start address to invalidate
  * @end: End address to invalidate
  * @asid: Address space ID to invalidate
  *
- * Create a GT TLB invalidation job and initialize internal fields. The caller is
+ * Create a TLB invalidation job and initialize internal fields. The caller is
  * responsible for releasing the creation reference.
  *
- * Return: GT TLB invalidation job object on success, ERR_PTR failure
+ * Return: TLB invalidation job object on success, ERR_PTR failure
  */
-struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
-						       struct xe_gt *gt,
-						       u64 start, u64 end,
-						       u32 asid)
+struct xe_tlb_inval_job *
+xe_tlb_inval_job_create(struct xe_exec_queue *q, struct xe_tlb_inval *tlb_inval,
+			struct xe_dep_scheduler *dep_scheduler, u64 start,
+			u64 end, u32 asid)
 {
-	struct xe_gt_tlb_inval_job *job;
-	struct xe_dep_scheduler *dep_scheduler =
-		q->tlb_inval[xe_gt_tlb_inval_context(gt)].dep_scheduler;
+	struct xe_tlb_inval_job *job;
 	struct drm_sched_entity *entity =
 		xe_dep_scheduler_entity(dep_scheduler);
-	struct xe_gt_tlb_inval_fence *ifence;
+	struct xe_tlb_inval_fence *ifence;
 	int err;
 
 	job = kmalloc(sizeof(*job), GFP_KERNEL);
@@ -101,14 +95,14 @@ struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
 		return ERR_PTR(-ENOMEM);
 
 	job->q = q;
-	job->gt = gt;
+	job->tlb_inval = tlb_inval;
 	job->start = start;
 	job->end = end;
 	job->asid = asid;
 	job->fence_armed = false;
 	job->dep.ops = &dep_job_ops;
 	kref_init(&job->refcount);
-	xe_exec_queue_get(q);	/* Pairs with put in xe_gt_tlb_inval_job_destroy */
+	xe_exec_queue_get(q);	/* Pairs with put in xe_tlb_inval_job_destroy */
 
 	ifence = kmalloc(sizeof(*ifence), GFP_KERNEL);
 	if (!ifence) {
@@ -122,8 +116,8 @@ struct xe_gt_tlb_inval_job *xe_gt_tlb_inval_job_create(struct xe_exec_queue *q,
 	if (err)
 		goto err_fence;
 
-	/* Pairs with put in xe_gt_tlb_inval_job_destroy */
-	xe_pm_runtime_get_noresume(gt_to_xe(job->gt));
+	/* Pairs with put in xe_tlb_inval_job_destroy */
+	xe_pm_runtime_get_noresume(gt_to_xe(q->gt));
 
 	return job;
 
@@ -136,40 +130,40 @@ err_job:
 	return ERR_PTR(err);
 }
 
-static void xe_gt_tlb_inval_job_destroy(struct kref *ref)
+static void xe_tlb_inval_job_destroy(struct kref *ref)
 {
-	struct xe_gt_tlb_inval_job *job = container_of(ref, typeof(*job),
-						       refcount);
-	struct xe_gt_tlb_inval_fence *ifence =
+	struct xe_tlb_inval_job *job = container_of(ref, typeof(*job),
+						    refcount);
+	struct xe_tlb_inval_fence *ifence =
 		container_of(job->fence, typeof(*ifence), base);
-	struct xe_device *xe = gt_to_xe(job->gt);
 	struct xe_exec_queue *q = job->q;
+	struct xe_device *xe = gt_to_xe(q->gt);
 
 	if (!job->fence_armed)
 		kfree(ifence);
 	else
-		/* Ref from xe_gt_tlb_inval_fence_init */
+		/* Ref from xe_tlb_inval_fence_init */
 		dma_fence_put(job->fence);
 
 	drm_sched_job_cleanup(&job->dep.drm);
 	kfree(job);
-	xe_exec_queue_put(q);	/* Pairs with get from xe_gt_tlb_inval_job_create */
-	xe_pm_runtime_put(xe);	/* Pairs with get from xe_gt_tlb_inval_job_create */
+	xe_exec_queue_put(q);	/* Pairs with get from xe_tlb_inval_job_create */
+	xe_pm_runtime_put(xe);	/* Pairs with get from xe_tlb_inval_job_create */
 }
 
 /**
- * xe_gt_tlb_inval_alloc_dep() - GT TLB invalidation job alloc dependency
- * @job: GT TLB invalidation job to alloc dependency for
+ * xe_tlb_inval_alloc_dep() - TLB invalidation job alloc dependency
+ * @job: TLB invalidation job to alloc dependency for
  *
- * Allocate storage for a dependency in the GT TLB invalidation fence. This
+ * Allocate storage for a dependency in the TLB invalidation fence. This
  * function should be called at most once per job and must be paired with
- * xe_gt_tlb_inval_job_push being called with a real fence.
+ * xe_tlb_inval_job_push being called with a real fence.
  *
  * Return: 0 on success, -errno on failure
  */
-int xe_gt_tlb_inval_job_alloc_dep(struct xe_gt_tlb_inval_job *job)
+int xe_tlb_inval_job_alloc_dep(struct xe_tlb_inval_job *job)
 {
-	xe_assert(gt_to_xe(job->gt), !xa_load(&job->dep.drm.dependencies, 0));
+	xe_assert(gt_to_xe(job->q->gt), !xa_load(&job->dep.drm.dependencies, 0));
 	might_alloc(GFP_KERNEL);
 
 	return drm_sched_job_add_dependency(&job->dep.drm,
@@ -177,24 +171,24 @@ int xe_gt_tlb_inval_job_alloc_dep(struct xe_gt_tlb_inval_job *job)
 }
 
 /**
- * xe_gt_tlb_inval_job_push() - GT TLB invalidation job push
- * @job: GT TLB invalidation job to push
+ * xe_tlb_inval_job_push() - TLB invalidation job push
+ * @job: TLB invalidation job to push
  * @m: The migration object being used
- * @fence: Dependency for GT TLB invalidation job
+ * @fence: Dependency for TLB invalidation job
  *
- * Pushes a GT TLB invalidation job for execution, using @fence as a dependency.
- * Storage for @fence must be preallocated with xe_gt_tlb_inval_job_alloc_dep
+ * Pushes a TLB invalidation job for execution, using @fence as a dependency.
+ * Storage for @fence must be preallocated with xe_tlb_inval_job_alloc_dep
  * prior to this call if @fence is not signaled. Takes a reference to the job’s
  * finished fence, which the caller is responsible for releasing, and return it
  * to the caller. This function is safe to be called in the path of reclaim.
  *
  * Return: Job's finished fence on success, cannot fail
  */
-struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
-					   struct xe_migrate *m,
-					   struct dma_fence *fence)
+struct dma_fence *xe_tlb_inval_job_push(struct xe_tlb_inval_job *job,
+					struct xe_migrate *m,
+					struct dma_fence *fence)
 {
-	struct xe_gt_tlb_inval_fence *ifence =
+	struct xe_tlb_inval_fence *ifence =
 		container_of(job->fence, typeof(*ifence), base);
 
 	if (!dma_fence_is_signaled(fence)) {
@@ -202,20 +196,20 @@ struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
 
 		/*
 		 * Can be in path of reclaim, hence the preallocation of fence
-		 * storage in xe_gt_tlb_inval_job_alloc_dep. Verify caller did
+		 * storage in xe_tlb_inval_job_alloc_dep. Verify caller did
 		 * this correctly.
 		 */
-		xe_assert(gt_to_xe(job->gt),
+		xe_assert(gt_to_xe(job->q->gt),
 			  xa_load(&job->dep.drm.dependencies, 0) ==
 			  dma_fence_get_stub());
 
 		dma_fence_get(fence);	/* ref released once dependency processed by scheduler */
 		ptr = xa_store(&job->dep.drm.dependencies, 0, fence,
 			       GFP_ATOMIC);
-		xe_assert(gt_to_xe(job->gt), !xa_is_err(ptr));
+		xe_assert(gt_to_xe(job->q->gt), !xa_is_err(ptr));
 	}
 
-	xe_gt_tlb_inval_job_get(job);	/* Pairs with put in free_job */
+	xe_tlb_inval_job_get(job);	/* Pairs with put in free_job */
 	job->fence_armed = true;
 
 	/*
@@ -225,8 +219,8 @@ struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
 	 */
 	xe_migrate_job_lock(m, job->q);
 
-	/* Creation ref pairs with put in xe_gt_tlb_inval_job_destroy */
-	xe_gt_tlb_inval_fence_init(job->gt, ifence, false);
+	/* Creation ref pairs with put in xe_tlb_inval_job_destroy */
+	xe_tlb_inval_fence_init(job->tlb_inval, ifence, false);
 	dma_fence_get(job->fence);	/* Pairs with put in DRM scheduler */
 
 	drm_sched_job_arm(&job->dep.drm);
@@ -241,7 +235,7 @@ struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
 
 	/*
 	 * Not using job->fence, as it has its own dma-fence context, which does
-	 * not allow GT TLB invalidation fences on the same queue, GT tuple to
+	 * not allow TLB invalidation fences on the same queue, GT tuple to
 	 * be squashed in dma-resv/DRM scheduler. Instead, we use the DRM scheduler
 	 * context and job's finished fence, which enables squashing.
 	 */
@@ -249,26 +243,26 @@ struct dma_fence *xe_gt_tlb_inval_job_push(struct xe_gt_tlb_inval_job *job,
 }
 
 /**
- * xe_gt_tlb_inval_job_get() - Get a reference to GT TLB invalidation job
- * @job: GT TLB invalidation job object
+ * xe_tlb_inval_job_get() - Get a reference to TLB invalidation job
+ * @job: TLB invalidation job object
  *
- * Increment the GT TLB invalidation job's reference count
+ * Increment the TLB invalidation job's reference count
  */
-void xe_gt_tlb_inval_job_get(struct xe_gt_tlb_inval_job *job)
+void xe_tlb_inval_job_get(struct xe_tlb_inval_job *job)
 {
 	kref_get(&job->refcount);
 }
 
 /**
- * xe_gt_tlb_inval_job_put() - Put a reference to GT TLB invalidation job
- * @job: GT TLB invalidation job object
+ * xe_tlb_inval_job_put() - Put a reference to TLB invalidation job
+ * @job: TLB invalidation job object
  *
- * Decrement the GT TLB invalidation job's reference count, call
- * xe_gt_tlb_inval_job_destroy when reference count == 0. Skips decrement if
+ * Decrement the TLB invalidation job's reference count, call
+ * xe_tlb_inval_job_destroy when reference count == 0. Skips decrement if
  * input @job is NULL or IS_ERR.
  */
-void xe_gt_tlb_inval_job_put(struct xe_gt_tlb_inval_job *job)
+void xe_tlb_inval_job_put(struct xe_tlb_inval_job *job)
 {
 	if (!IS_ERR_OR_NULL(job))
-		kref_put(&job->refcount, xe_gt_tlb_inval_job_destroy);
+		kref_put(&job->refcount, xe_tlb_inval_job_destroy);
 }
diff --git a/drivers/gpu/drm/xe/xe_tlb_inval_job.h b/drivers/gpu/drm/xe/xe_tlb_inval_job.h
new file mode 100644
index 000000000000..e63edcb26b50
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_tlb_inval_job.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_TLB_INVAL_JOB_H_
+#define _XE_TLB_INVAL_JOB_H_
+
+#include <linux/types.h>
+
+struct dma_fence;
+struct xe_dep_scheduler;
+struct xe_exec_queue;
+struct xe_tlb_inval;
+struct xe_tlb_inval_job;
+struct xe_migrate;
+
+struct xe_tlb_inval_job *
+xe_tlb_inval_job_create(struct xe_exec_queue *q, struct xe_tlb_inval *tlb_inval,
+			struct xe_dep_scheduler *dep_scheduler,
+			u64 start, u64 end, u32 asid);
+
+int xe_tlb_inval_job_alloc_dep(struct xe_tlb_inval_job *job);
+
+struct dma_fence *xe_tlb_inval_job_push(struct xe_tlb_inval_job *job,
+					struct xe_migrate *m,
+					struct dma_fence *fence);
+
+void xe_tlb_inval_job_get(struct xe_tlb_inval_job *job);
+
+void xe_tlb_inval_job_put(struct xe_tlb_inval_job *job);
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h b/drivers/gpu/drm/xe/xe_tlb_inval_types.h
similarity index 56%
rename from drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
rename to drivers/gpu/drm/xe/xe_tlb_inval_types.h
index 442f72b78ccf..6d14b9f17b91 100644
--- a/drivers/gpu/drm/xe/xe_gt_tlb_inval_types.h
+++ b/drivers/gpu/drm/xe/xe_tlb_inval_types.h
@@ -3,58 +3,57 @@
  * Copyright © 2023 Intel Corporation
  */
 
-#ifndef _XE_GT_TLB_INVAL_TYPES_H_
-#define _XE_GT_TLB_INVAL_TYPES_H_
+#ifndef _XE_TLB_INVAL_TYPES_H_
+#define _XE_TLB_INVAL_TYPES_H_
 
 #include <linux/workqueue.h>
 #include <linux/dma-fence.h>
 
-struct xe_gt;
-
 /** struct xe_tlb_inval - TLB invalidation client */
 struct xe_tlb_inval {
+	/** @private: Backend private pointer */
+	void *private;
 	/** @tlb_inval.seqno: TLB invalidation seqno, protected by CT lock */
 #define TLB_INVALIDATION_SEQNO_MAX	0x100000
 	int seqno;
 	/** @tlb_invalidation.seqno_lock: protects @tlb_invalidation.seqno */
 	struct mutex seqno_lock;
 	/**
-	 * @tlb_inval.seqno_recv: last received TLB invalidation seqno,
-	 * protected by CT lock
+	 * @seqno_recv: last received TLB invalidation seqno, protected by
+	 * CT lock
 	 */
 	int seqno_recv;
 	/**
-	 * @tlb_inval.pending_fences: list of pending fences waiting TLB
-	 * invaliations, protected by CT lock
+	 * @pending_fences: list of pending fences waiting TLB invaliations,
+	 * protected CT lock
 	 */
 	struct list_head pending_fences;
 	/**
-	 * @tlb_inval.pending_lock: protects @tlb_inval.pending_fences
-	 * and updating @tlb_inval.seqno_recv.
+	 * @pending_lock: protects @pending_fences and updating @seqno_recv.
 	 */
 	spinlock_t pending_lock;
 	/**
-	 * @tlb_inval.fence_tdr: schedules a delayed call to
-	 * xe_gt_tlb_fence_timeout after the timeut interval is over.
+	 * @fence_tdr: schedules a delayed call to xe_tlb_fence_timeout after
+	 * the timeout interval is over.
 	 */
 	struct delayed_work fence_tdr;
-	/** @wtlb_invalidation.wq: schedules GT TLB invalidation jobs */
+	/** @job_wq: schedules TLB invalidation jobs */
 	struct workqueue_struct *job_wq;
 	/** @tlb_inval.lock: protects TLB invalidation fences */
 	spinlock_t lock;
 };
 
 /**
- * struct xe_gt_tlb_inval_fence - XE GT TLB invalidation fence
+ * struct xe_tlb_inval_fence - TLB invalidation fence
  *
- * Optionally passed to xe_gt_tlb_inval and will be signaled upon TLB
+ * Optionally passed to xe_tlb_inval* functions and will be signaled upon TLB
  * invalidation completion.
  */
-struct xe_gt_tlb_inval_fence {
+struct xe_tlb_inval_fence {
 	/** @base: dma fence base */
 	struct dma_fence base;
-	/** @gt: GT which fence belong to */
-	struct xe_gt *gt;
+	/** @tlb_inval: TLB invalidation client which fence belong to */
+	struct xe_tlb_inval *tlb_inval;
 	/** @link: link into list of pending tlb fences */
 	struct list_head link;
 	/** @seqno: seqno of TLB invalidation to signal fence one */
diff --git a/drivers/gpu/drm/xe/xe_trace.h b/drivers/gpu/drm/xe/xe_trace.h
index 36538f50d06f..314f42fcbcbd 100644
--- a/drivers/gpu/drm/xe/xe_trace.h
+++ b/drivers/gpu/drm/xe/xe_trace.h
@@ -14,10 +14,10 @@
 
 #include "xe_exec_queue_types.h"
 #include "xe_gpu_scheduler_types.h"
-#include "xe_gt_tlb_inval_types.h"
 #include "xe_gt_types.h"
 #include "xe_guc_exec_queue_types.h"
 #include "xe_sched_job.h"
+#include "xe_tlb_inval_types.h"
 #include "xe_vm.h"
 
 #define __dev_name_xe(xe)	dev_name((xe)->drm.dev)
@@ -25,13 +25,13 @@
 #define __dev_name_gt(gt)	__dev_name_xe(gt_to_xe((gt)))
 #define __dev_name_eq(q)	__dev_name_gt((q)->gt)
 
-DECLARE_EVENT_CLASS(xe_gt_tlb_inval_fence,
-		    TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
+DECLARE_EVENT_CLASS(xe_tlb_inval_fence,
+		    TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence),
 		    TP_ARGS(xe, fence),
 
 		    TP_STRUCT__entry(
 			     __string(dev, __dev_name_xe(xe))
-			     __field(struct xe_gt_tlb_inval_fence *, fence)
+			     __field(struct xe_tlb_inval_fence *, fence)
 			     __field(int, seqno)
 			     ),
 
@@ -45,23 +45,23 @@ DECLARE_EVENT_CLASS(xe_gt_tlb_inval_fence,
 			      __get_str(dev), __entry->fence, __entry->seqno)
 );
 
-DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_send,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
+DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_send,
+	     TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
-DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_recv,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
+DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_recv,
+	     TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
-DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_signal,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
+DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_signal,
+	     TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
-DEFINE_EVENT(xe_gt_tlb_inval_fence, xe_gt_tlb_inval_fence_timeout,
-	     TP_PROTO(struct xe_device *xe, struct xe_gt_tlb_inval_fence *fence),
+DEFINE_EVENT(xe_tlb_inval_fence, xe_tlb_inval_fence_timeout,
+	     TP_PROTO(struct xe_device *xe, struct xe_tlb_inval_fence *fence),
 	     TP_ARGS(xe, fence)
 );
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index b758e9a6837a..3ff3c67aa79d 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -28,7 +28,6 @@
 #include "xe_drm_client.h"
 #include "xe_exec_queue.h"
 #include "xe_gt_pagefault.h"
-#include "xe_gt_tlb_inval.h"
 #include "xe_migrate.h"
 #include "xe_pat.h"
 #include "xe_pm.h"
@@ -39,6 +38,7 @@
 #include "xe_svm.h"
 #include "xe_sync.h"
 #include "xe_tile.h"
+#include "xe_tlb_inval.h"
 #include "xe_trace_bo.h"
 #include "xe_wa.h"
 #include "xe_hmm.h"
@@ -1898,7 +1898,7 @@ static void xe_vm_close(struct xe_vm *vm)
 					xe_pt_clear(xe, vm->pt_root[id]);
 
 			for_each_gt(gt, xe, id)
-				xe_gt_tlb_inval_vm(gt, vm);
+				xe_tlb_inval_vm(&gt->tlb_inval, vm);
 		}
 	}
 
@@ -4046,7 +4046,7 @@ void xe_vm_unlock(struct xe_vm *vm)
 int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start,
 				   u64 end, u8 tile_mask)
 {
-	struct xe_gt_tlb_inval_fence
+	struct xe_tlb_inval_fence
 		fence[XE_MAX_TILES_PER_DEVICE * XE_MAX_GT_PER_TILE];
 	struct xe_tile *tile;
 	u32 fence_id = 0;
@@ -4060,11 +4060,12 @@ int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start,
 		if (!(tile_mask & BIT(id)))
 			continue;
 
-		xe_gt_tlb_inval_fence_init(tile->primary_gt,
-					   &fence[fence_id], true);
+		xe_tlb_inval_fence_init(&tile->primary_gt->tlb_inval,
+					&fence[fence_id], true);
 
-		err = xe_gt_tlb_inval_range(tile->primary_gt, &fence[fence_id],
-					    start, end, vm->usm.asid);
+		err = xe_tlb_inval_range(&tile->primary_gt->tlb_inval,
+					 &fence[fence_id], start, end,
+					 vm->usm.asid);
 		if (err)
 			goto wait;
 		++fence_id;
@@ -4072,11 +4073,12 @@ int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start,
 		if (!tile->media_gt)
 			continue;
 
-		xe_gt_tlb_inval_fence_init(tile->media_gt,
-					   &fence[fence_id], true);
+		xe_tlb_inval_fence_init(&tile->media_gt->tlb_inval,
+					&fence[fence_id], true);
 
-		err = xe_gt_tlb_inval_range(tile->media_gt, &fence[fence_id],
-					    start, end, vm->usm.asid);
+		err = xe_tlb_inval_range(&tile->media_gt->tlb_inval,
+					 &fence[fence_id], start, end,
+					 vm->usm.asid);
 		if (err)
 			goto wait;
 		++fence_id;
@@ -4084,7 +4086,7 @@ int xe_vm_range_tilemask_tlb_inval(struct xe_vm *vm, u64 start,
 
 wait:
 	for (id = 0; id < fence_id; ++id)
-		xe_gt_tlb_inval_fence_wait(&fence[id]);
+		xe_tlb_inval_fence_wait(&fence[id]);
 
 	return err;
 }

From 9aff63cf3791e464e4762ca79bd5d5b9643edce6 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:09 +0000
Subject: [PATCH 0703/1455] drm/xe: Prep TLB invalidation fence before sending

It is a bit backwards to add a TLB invalidation fence to the pending
list after issuing the invalidation. Perform this step before issuing
the TLB invalidation in a helper function.

v2: Make sure the seqno_lock mutex covers the send as well (Matt)

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-8-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/xe_tlb_inval.c | 109 +++++++++++++++---------------
 1 file changed, 55 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.c b/drivers/gpu/drm/xe/xe_tlb_inval.c
index f4b7c0c74894..dff26f7954ec 100644
--- a/drivers/gpu/drm/xe/xe_tlb_inval.c
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.c
@@ -66,19 +66,19 @@ __inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
 static void
 inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
 {
+	lockdep_assert_held(&fence->tlb_inval->pending_lock);
+
 	list_del(&fence->link);
 	__inval_fence_signal(xe, fence);
 }
 
-void xe_tlb_inval_fence_signal(struct xe_tlb_inval_fence *fence)
+static void
+inval_fence_signal_unlocked(struct xe_device *xe,
+			    struct xe_tlb_inval_fence *fence)
 {
-	struct xe_gt *gt;
-
-	if (WARN_ON_ONCE(!fence->tlb_inval))
-		return;
-
-	gt = fence->tlb_inval->private;
-	__inval_fence_signal(gt_to_xe(gt), fence);
+	spin_lock_irq(&fence->tlb_inval->pending_lock);
+	inval_fence_signal(xe, fence);
+	spin_unlock_irq(&fence->tlb_inval->pending_lock);
 }
 
 static void xe_gt_tlb_fence_timeout(struct work_struct *work)
@@ -221,14 +221,10 @@ static bool tlb_inval_seqno_past(struct xe_gt *gt, int seqno)
 	return seqno_recv >= seqno;
 }
 
-static int send_tlb_inval(struct xe_guc *guc,
-			  struct xe_tlb_inval_fence *fence,
+static int send_tlb_inval(struct xe_guc *guc, struct xe_tlb_inval_fence *fence,
 			  u32 *action, int len)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
-	struct xe_device *xe = gt_to_xe(gt);
-	int seqno;
-	int ret;
 
 	xe_gt_assert(gt, fence);
 
@@ -238,47 +234,36 @@ static int send_tlb_inval(struct xe_guc *guc,
 	 * need to be updated.
 	 */
 
-	mutex_lock(&gt->tlb_inval.seqno_lock);
-	seqno = gt->tlb_inval.seqno;
-	fence->seqno = seqno;
-	trace_xe_tlb_inval_fence_send(xe, fence);
-	action[1] = seqno;
-	ret = xe_guc_ct_send(&guc->ct, action, len,
-			     G2H_LEN_DW_TLB_INVALIDATE, 1);
-	if (!ret) {
-		spin_lock_irq(&gt->tlb_inval.pending_lock);
-		/*
-		 * We haven't actually published the TLB fence as per
-		 * pending_fences, but in theory our seqno could have already
-		 * been written as we acquired the pending_lock. In such a case
-		 * we can just go ahead and signal the fence here.
-		 */
-		if (tlb_inval_seqno_past(gt, seqno)) {
-			__inval_fence_signal(xe, fence);
-		} else {
-			fence->inval_time = ktime_get();
-			list_add_tail(&fence->link,
-				      &gt->tlb_inval.pending_fences);
-
-			if (list_is_singular(&gt->tlb_inval.pending_fences))
-				queue_delayed_work(system_wq,
-						   &gt->tlb_inval.fence_tdr,
-						   tlb_timeout_jiffies(gt));
-		}
-		spin_unlock_irq(&gt->tlb_inval.pending_lock);
-	} else {
-		__inval_fence_signal(xe, fence);
-	}
-	if (!ret) {
-		gt->tlb_inval.seqno = (gt->tlb_inval.seqno + 1) %
-			TLB_INVALIDATION_SEQNO_MAX;
-		if (!gt->tlb_inval.seqno)
-			gt->tlb_inval.seqno = 1;
-	}
-	mutex_unlock(&gt->tlb_inval.seqno_lock);
 	xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1);
+	action[1] = fence->seqno;
 
-	return ret;
+	return xe_guc_ct_send(&guc->ct, action, len,
+			      G2H_LEN_DW_TLB_INVALIDATE, 1);
+}
+
+static void xe_tlb_inval_fence_prep(struct xe_tlb_inval_fence *fence)
+{
+	struct xe_tlb_inval *tlb_inval = fence->tlb_inval;
+	struct xe_gt *gt = tlb_inval->private;
+	struct xe_device *xe = gt_to_xe(gt);
+
+	fence->seqno = tlb_inval->seqno;
+	trace_xe_tlb_inval_fence_send(xe, fence);
+
+	spin_lock_irq(&tlb_inval->pending_lock);
+	fence->inval_time = ktime_get();
+	list_add_tail(&fence->link, &tlb_inval->pending_fences);
+
+	if (list_is_singular(&tlb_inval->pending_fences))
+		queue_delayed_work(system_wq,
+				   &tlb_inval->fence_tdr,
+				   tlb_timeout_jiffies(gt));
+	spin_unlock_irq(&tlb_inval->pending_lock);
+
+	tlb_inval->seqno = (tlb_inval->seqno + 1) %
+		TLB_INVALIDATION_SEQNO_MAX;
+	if (!tlb_inval->seqno)
+		tlb_inval->seqno = 1;
 }
 
 #define MAKE_INVAL_OP(type)	((type << XE_GUC_TLB_INVAL_TYPE_SHIFT) | \
@@ -306,7 +291,14 @@ static int xe_tlb_inval_guc(struct xe_gt *gt,
 	};
 	int ret;
 
+	mutex_lock(&gt->tlb_inval.seqno_lock);
+	xe_tlb_inval_fence_prep(fence);
+
 	ret = send_tlb_inval(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
+	if (ret < 0)
+		inval_fence_signal_unlocked(gt_to_xe(gt), fence);
+	mutex_unlock(&gt->tlb_inval.seqno_lock);
+
 	/*
 	 * -ECANCELED indicates the CT is stopped for a GT reset. TLB caches
 	 *  should be nuked on a GT reset so this error can be ignored.
@@ -433,7 +425,7 @@ int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
 #define MAX_TLB_INVALIDATION_LEN	7
 	u32 action[MAX_TLB_INVALIDATION_LEN];
 	u64 length = end - start;
-	int len = 0;
+	int len = 0, ret;
 
 	xe_gt_assert(gt, fence);
 
@@ -494,7 +486,16 @@ int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
 
 	xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN);
 
-	return send_tlb_inval(&gt->uc.guc, fence, action, len);
+	mutex_lock(&gt->tlb_inval.seqno_lock);
+	xe_tlb_inval_fence_prep(fence);
+
+	ret = send_tlb_inval(&gt->uc.guc, fence, action,
+			     ARRAY_SIZE(action));
+	if (ret < 0)
+		inval_fence_signal_unlocked(xe, fence);
+	mutex_unlock(&gt->tlb_inval.seqno_lock);
+
+	return ret;
 }
 
 /**

From 8443e8c448cfe1082f2bbad41fe6e56e8d3b5bba Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:10 +0000
Subject: [PATCH 0704/1455] drm/xe: Add helpers to send TLB invalidations

Break out the GuC specific code into helpers as part of the process to
decouple frontback TLB invalidation code from the backend.

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-9-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/xe_tlb_inval.c | 234 +++++++++++++++---------------
 1 file changed, 117 insertions(+), 117 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.c b/drivers/gpu/drm/xe/xe_tlb_inval.c
index dff26f7954ec..d81a45f1dd7c 100644
--- a/drivers/gpu/drm/xe/xe_tlb_inval.c
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.c
@@ -221,12 +221,11 @@ static bool tlb_inval_seqno_past(struct xe_gt *gt, int seqno)
 	return seqno_recv >= seqno;
 }
 
-static int send_tlb_inval(struct xe_guc *guc, struct xe_tlb_inval_fence *fence,
-			  u32 *action, int len)
+static int send_tlb_inval(struct xe_guc *guc, const u32 *action, int len)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
 
-	xe_gt_assert(gt, fence);
+	xe_gt_assert(gt, action[1]);	/* Seqno */
 
 	/*
 	 * XXX: The seqno algorithm relies on TLB invalidation being processed
@@ -235,7 +234,6 @@ static int send_tlb_inval(struct xe_guc *guc, struct xe_tlb_inval_fence *fence,
 	 */
 
 	xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1);
-	action[1] = fence->seqno;
 
 	return xe_guc_ct_send(&guc->ct, action, len,
 			      G2H_LEN_DW_TLB_INVALIDATE, 1);
@@ -270,91 +268,15 @@ static void xe_tlb_inval_fence_prep(struct xe_tlb_inval_fence *fence)
 		XE_GUC_TLB_INVAL_MODE_HEAVY << XE_GUC_TLB_INVAL_MODE_SHIFT | \
 		XE_GUC_TLB_INVAL_FLUSH_CACHE)
 
-/**
- * xe_tlb_inval_guc - Issue a TLB invalidation on this GT for the GuC
- * @gt: GT structure
- * @fence: invalidation fence which will be signal on TLB invalidation
- * completion
- *
- * Issue a TLB invalidation for the GuC. Completion of TLB is asynchronous and
- * caller can use the invalidation fence to wait for completion.
- *
- * Return: 0 on success, negative error code on error
- */
-static int xe_tlb_inval_guc(struct xe_gt *gt,
-			    struct xe_tlb_inval_fence *fence)
+static int send_tlb_inval_ggtt(struct xe_gt *gt, int seqno)
 {
 	u32 action[] = {
 		XE_GUC_ACTION_TLB_INVALIDATION,
-		0,  /* seqno, replaced in send_tlb_inval */
+		seqno,
 		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_GUC),
 	};
-	int ret;
 
-	mutex_lock(&gt->tlb_inval.seqno_lock);
-	xe_tlb_inval_fence_prep(fence);
-
-	ret = send_tlb_inval(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
-	if (ret < 0)
-		inval_fence_signal_unlocked(gt_to_xe(gt), fence);
-	mutex_unlock(&gt->tlb_inval.seqno_lock);
-
-	/*
-	 * -ECANCELED indicates the CT is stopped for a GT reset. TLB caches
-	 *  should be nuked on a GT reset so this error can be ignored.
-	 */
-	if (ret == -ECANCELED)
-		return 0;
-
-	return ret;
-}
-
-/**
- * xe_tlb_inval_ggtt - Issue a TLB invalidation on this GT for the GGTT
- * @tlb_inval: TLB invalidation client
- *
- * Issue a TLB invalidation for the GGTT. Completion of TLB invalidation is
- * synchronous.
- *
- * Return: 0 on success, negative error code on error
- */
-int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval)
-{
-	struct xe_gt *gt = tlb_inval->private;
-	struct xe_device *xe = gt_to_xe(gt);
-	unsigned int fw_ref;
-
-	if (xe_guc_ct_enabled(&gt->uc.guc.ct) &&
-	    gt->uc.guc.submission_state.enabled) {
-		struct xe_tlb_inval_fence fence;
-		int ret;
-
-		xe_tlb_inval_fence_init(tlb_inval, &fence, true);
-		ret = xe_tlb_inval_guc(gt, &fence);
-		if (ret)
-			return ret;
-
-		xe_tlb_inval_fence_wait(&fence);
-	} else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) {
-		struct xe_mmio *mmio = &gt->mmio;
-
-		if (IS_SRIOV_VF(xe))
-			return 0;
-
-		fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
-		if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) {
-			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC1,
-					PVC_GUC_TLB_INV_DESC1_INVALIDATE);
-			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC0,
-					PVC_GUC_TLB_INV_DESC0_VALID);
-		} else {
-			xe_mmio_write32(mmio, GUC_TLB_INV_CR,
-					GUC_TLB_INV_CR_INVALIDATE);
-		}
-		xe_force_wake_put(gt_to_fw(gt), fw_ref);
-	}
-
-	return 0;
+	return send_tlb_inval(&gt->uc.guc, action, ARRAY_SIZE(action));
 }
 
 static int send_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
@@ -369,7 +291,7 @@ static int send_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
 
 	xe_gt_assert(gt, fence);
 
-	return send_tlb_inval(&gt->uc.guc, fence, action, ARRAY_SIZE(action));
+	return send_tlb_inval(&gt->uc.guc, action, ARRAY_SIZE(action));
 }
 
 /**
@@ -401,43 +323,17 @@ int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
  */
 #define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX))
 
-/**
- * xe_tlb_inval_range - Issue a TLB invalidation on this GT for an address range
- * @tlb_inval: TLB invalidation client
- * @fence: invalidation fence which will be signal on TLB invalidation
- * completion
- * @start: start address
- * @end: end address
- * @asid: address space id
- *
- * Issue a range based TLB invalidation if supported, if not fallback to a full
- * TLB invalidation. Completion of TLB is asynchronous and caller can use
- * the invalidation fence to wait for completion.
- *
- * Return: Negative error code on error, 0 on success
- */
-int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
-		       struct xe_tlb_inval_fence *fence, u64 start, u64 end,
-		       u32 asid)
+static int send_tlb_inval_ppgtt(struct xe_gt *gt, u64 start, u64 end,
+				u32 asid, int seqno)
 {
-	struct xe_gt *gt = tlb_inval->private;
-	struct xe_device *xe = gt_to_xe(gt);
 #define MAX_TLB_INVALIDATION_LEN	7
 	u32 action[MAX_TLB_INVALIDATION_LEN];
 	u64 length = end - start;
-	int len = 0, ret;
-
-	xe_gt_assert(gt, fence);
-
-	/* Execlists not supported */
-	if (gt_to_xe(gt)->info.force_execlist) {
-		__inval_fence_signal(xe, fence);
-		return 0;
-	}
+	int len = 0;
 
 	action[len++] = XE_GUC_ACTION_TLB_INVALIDATION;
-	action[len++] = 0; /* seqno, replaced in send_tlb_inval */
-	if (!xe->info.has_range_tlb_inval ||
+	action[len++] = seqno;
+	if (!gt_to_xe(gt)->info.has_range_tlb_inval ||
 	    length > MAX_RANGE_TLB_INVALIDATION_LENGTH) {
 		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL);
 	} else {
@@ -486,11 +382,115 @@ int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
 
 	xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN);
 
+	return send_tlb_inval(&gt->uc.guc, action, len);
+}
+
+static int __xe_tlb_inval_ggtt(struct xe_gt *gt,
+			       struct xe_tlb_inval_fence *fence)
+{
+	int ret;
+
 	mutex_lock(&gt->tlb_inval.seqno_lock);
 	xe_tlb_inval_fence_prep(fence);
 
-	ret = send_tlb_inval(&gt->uc.guc, fence, action,
-			     ARRAY_SIZE(action));
+	ret = send_tlb_inval_ggtt(gt, fence->seqno);
+	if (ret < 0)
+		inval_fence_signal_unlocked(gt_to_xe(gt), fence);
+	mutex_unlock(&gt->tlb_inval.seqno_lock);
+
+	/*
+	 * -ECANCELED indicates the CT is stopped for a GT reset. TLB caches
+	 *  should be nuked on a GT reset so this error can be ignored.
+	 */
+	if (ret == -ECANCELED)
+		return 0;
+
+	return ret;
+}
+
+/**
+ * xe_tlb_inval_ggtt - Issue a TLB invalidation on this GT for the GGTT
+ * @tlb_inval: TLB invalidation client
+ *
+ * Issue a TLB invalidation for the GGTT. Completion of TLB invalidation is
+ * synchronous.
+ *
+ * Return: 0 on success, negative error code on error
+ */
+int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval)
+{
+	struct xe_gt *gt = tlb_inval->private;
+	struct xe_device *xe = gt_to_xe(gt);
+	unsigned int fw_ref;
+
+	if (xe_guc_ct_enabled(&gt->uc.guc.ct) &&
+	    gt->uc.guc.submission_state.enabled) {
+		struct xe_tlb_inval_fence fence;
+		int ret;
+
+		xe_tlb_inval_fence_init(tlb_inval, &fence, true);
+		ret = __xe_tlb_inval_ggtt(gt, &fence);
+		if (ret)
+			return ret;
+
+		xe_tlb_inval_fence_wait(&fence);
+	} else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) {
+		struct xe_mmio *mmio = &gt->mmio;
+
+		if (IS_SRIOV_VF(xe))
+			return 0;
+
+		fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+		if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) {
+			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC1,
+					PVC_GUC_TLB_INV_DESC1_INVALIDATE);
+			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC0,
+					PVC_GUC_TLB_INV_DESC0_VALID);
+		} else {
+			xe_mmio_write32(mmio, GUC_TLB_INV_CR,
+					GUC_TLB_INV_CR_INVALIDATE);
+		}
+		xe_force_wake_put(gt_to_fw(gt), fw_ref);
+	}
+
+	return 0;
+}
+
+/**
+ * xe_tlb_inval_range - Issue a TLB invalidation on this GT for an address range
+ * @tlb_inval: TLB invalidation client
+ * @fence: invalidation fence which will be signal on TLB invalidation
+ * completion
+ * @start: start address
+ * @end: end address
+ * @asid: address space id
+ *
+ * Issue a range based TLB invalidation if supported, if not fallback to a full
+ * TLB invalidation. Completion of TLB is asynchronous and caller can use
+ * the invalidation fence to wait for completion.
+ *
+ * Return: Negative error code on error, 0 on success
+ */
+int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
+		       struct xe_tlb_inval_fence *fence, u64 start, u64 end,
+		       u32 asid)
+{
+	struct xe_gt *gt = tlb_inval->private;
+	struct xe_device *xe = gt_to_xe(gt);
+	int  ret;
+
+	xe_gt_assert(gt, fence);
+
+	/* Execlists not supported */
+	if (xe->info.force_execlist) {
+		__inval_fence_signal(xe, fence);
+		return 0;
+	}
+
+	mutex_lock(&gt->tlb_inval.seqno_lock);
+	xe_tlb_inval_fence_prep(fence);
+
+	ret = send_tlb_inval_ppgtt(gt, start, end, asid, fence->seqno);
 	if (ret < 0)
 		inval_fence_signal_unlocked(xe, fence);
 	mutex_unlock(&gt->tlb_inval.seqno_lock);

From db16f9d90c1d97a2b5483b4b856625875bffe860 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Tue, 26 Aug 2025 18:29:11 +0000
Subject: [PATCH 0705/1455] drm/xe: Split TLB invalidation code in frontend and
 backend

The frontend exposes an API to the driver to send invalidations, handles
sequence number assignment, synchronization (fences), and provides a
timeout mechanism. The backend issues the actual invalidation to the
hardware (or firmware).

The new layering easily allows issuing TLB invalidations to different
hardware or firmware interfaces.

Normalize some naming while here too.

Signed-off-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250826182911.392550-10-stuart.summers@intel.com
---
 drivers/gpu/drm/xe/Makefile             |   1 +
 drivers/gpu/drm/xe/xe_gt.c              |   2 -
 drivers/gpu/drm/xe/xe_guc_ct.c          |   2 +-
 drivers/gpu/drm/xe/xe_guc_tlb_inval.c   | 242 +++++++++++
 drivers/gpu/drm/xe/xe_guc_tlb_inval.h   |  19 +
 drivers/gpu/drm/xe/xe_tlb_inval.c       | 537 +++++++-----------------
 drivers/gpu/drm/xe/xe_tlb_inval.h       |  15 +-
 drivers/gpu/drm/xe/xe_tlb_inval_types.h |  67 ++-
 8 files changed, 500 insertions(+), 385 deletions(-)
 create mode 100644 drivers/gpu/drm/xe/xe_guc_tlb_inval.c
 create mode 100644 drivers/gpu/drm/xe/xe_guc_tlb_inval.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 2b67debb8bc4..987e4fe10538 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -75,6 +75,7 @@ xe-y += xe_bb.o \
 	xe_guc_log.o \
 	xe_guc_pc.o \
 	xe_guc_submit.o \
+	xe_guc_tlb_inval.o \
 	xe_heci_gsc.o \
 	xe_huc.o \
 	xe_hw_engine.o \
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 67ee7cdbd6ec..34505a6d93ed 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -603,8 +603,6 @@ static void xe_gt_fini(void *arg)
 	struct xe_gt *gt = arg;
 	int i;
 
-	xe_gt_tlb_inval_fini(gt);
-
 	for (i = 0; i < XE_ENGINE_CLASS_MAX; ++i)
 		xe_hw_fence_irq_finish(&gt->fence_irq[i]);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 5f38041cff4c..848065a25c44 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -30,9 +30,9 @@
 #include "xe_guc_log.h"
 #include "xe_guc_relay.h"
 #include "xe_guc_submit.h"
+#include "xe_guc_tlb_inval.h"
 #include "xe_map.h"
 #include "xe_pm.h"
-#include "xe_tlb_inval.h"
 #include "xe_trace_guc.h"
 
 static void receive_g2h(struct xe_guc_ct *ct);
diff --git a/drivers/gpu/drm/xe/xe_guc_tlb_inval.c b/drivers/gpu/drm/xe/xe_guc_tlb_inval.c
new file mode 100644
index 000000000000..6bf2103602f8
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_guc_tlb_inval.c
@@ -0,0 +1,242 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "abi/guc_actions_abi.h"
+
+#include "xe_device.h"
+#include "xe_gt_stats.h"
+#include "xe_gt_types.h"
+#include "xe_guc.h"
+#include "xe_guc_ct.h"
+#include "xe_guc_tlb_inval.h"
+#include "xe_force_wake.h"
+#include "xe_mmio.h"
+#include "xe_tlb_inval.h"
+
+#include "regs/xe_guc_regs.h"
+
+/*
+ * XXX: The seqno algorithm relies on TLB invalidation being processed in order
+ * which they currently are by the GuC, if that changes the algorithm will need
+ * to be updated.
+ */
+
+static int send_tlb_inval(struct xe_guc *guc, const u32 *action, int len)
+{
+	struct xe_gt *gt = guc_to_gt(guc);
+
+	xe_gt_assert(gt, action[1]);	/* Seqno */
+
+	xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1);
+	return xe_guc_ct_send(&guc->ct, action, len,
+			      G2H_LEN_DW_TLB_INVALIDATE, 1);
+}
+
+#define MAKE_INVAL_OP(type)	((type << XE_GUC_TLB_INVAL_TYPE_SHIFT) | \
+		XE_GUC_TLB_INVAL_MODE_HEAVY << XE_GUC_TLB_INVAL_MODE_SHIFT | \
+		XE_GUC_TLB_INVAL_FLUSH_CACHE)
+
+static int send_tlb_inval_all(struct xe_tlb_inval *tlb_inval, u32 seqno)
+{
+	struct xe_guc *guc = tlb_inval->private;
+	u32 action[] = {
+		XE_GUC_ACTION_TLB_INVALIDATION_ALL,
+		seqno,
+		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL),
+	};
+
+	return send_tlb_inval(guc, action, ARRAY_SIZE(action));
+}
+
+static int send_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval, u32 seqno)
+{
+	struct xe_guc *guc = tlb_inval->private;
+	struct xe_gt *gt = guc_to_gt(guc);
+	struct xe_device *xe = guc_to_xe(guc);
+
+	/*
+	 * Returning -ECANCELED in this function is squashed at the caller and
+	 * signals waiters.
+	 */
+
+	if (xe_guc_ct_enabled(&guc->ct) && guc->submission_state.enabled) {
+		u32 action[] = {
+			XE_GUC_ACTION_TLB_INVALIDATION,
+			seqno,
+			MAKE_INVAL_OP(XE_GUC_TLB_INVAL_GUC),
+		};
+
+		return send_tlb_inval(guc, action, ARRAY_SIZE(action));
+	} else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) {
+		struct xe_mmio *mmio = &gt->mmio;
+		unsigned int fw_ref;
+
+		if (IS_SRIOV_VF(xe))
+			return -ECANCELED;
+
+		fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
+		if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) {
+			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC1,
+					PVC_GUC_TLB_INV_DESC1_INVALIDATE);
+			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC0,
+					PVC_GUC_TLB_INV_DESC0_VALID);
+		} else {
+			xe_mmio_write32(mmio, GUC_TLB_INV_CR,
+					GUC_TLB_INV_CR_INVALIDATE);
+		}
+		xe_force_wake_put(gt_to_fw(gt), fw_ref);
+	}
+
+	return -ECANCELED;
+}
+
+/*
+ * Ensure that roundup_pow_of_two(length) doesn't overflow.
+ * Note that roundup_pow_of_two() operates on unsigned long,
+ * not on u64.
+ */
+#define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX))
+
+static int send_tlb_inval_ppgtt(struct xe_tlb_inval *tlb_inval, u32 seqno,
+				u64 start, u64 end, u32 asid)
+{
+#define MAX_TLB_INVALIDATION_LEN	7
+	struct xe_guc *guc = tlb_inval->private;
+	struct xe_gt *gt = guc_to_gt(guc);
+	u32 action[MAX_TLB_INVALIDATION_LEN];
+	u64 length = end - start;
+	int len = 0;
+
+	if (guc_to_xe(guc)->info.force_execlist)
+		return -ECANCELED;
+
+	action[len++] = XE_GUC_ACTION_TLB_INVALIDATION;
+	action[len++] = seqno;
+	if (!gt_to_xe(gt)->info.has_range_tlb_inval ||
+	    length > MAX_RANGE_TLB_INVALIDATION_LENGTH) {
+		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL);
+	} else {
+		u64 orig_start = start;
+		u64 align;
+
+		if (length < SZ_4K)
+			length = SZ_4K;
+
+		/*
+		 * We need to invalidate a higher granularity if start address
+		 * is not aligned to length. When start is not aligned with
+		 * length we need to find the length large enough to create an
+		 * address mask covering the required range.
+		 */
+		align = roundup_pow_of_two(length);
+		start = ALIGN_DOWN(start, align);
+		end = ALIGN(end, align);
+		length = align;
+		while (start + length < end) {
+			length <<= 1;
+			start = ALIGN_DOWN(orig_start, length);
+		}
+
+		/*
+		 * Minimum invalidation size for a 2MB page that the hardware
+		 * expects is 16MB
+		 */
+		if (length >= SZ_2M) {
+			length = max_t(u64, SZ_16M, length);
+			start = ALIGN_DOWN(orig_start, length);
+		}
+
+		xe_gt_assert(gt, length >= SZ_4K);
+		xe_gt_assert(gt, is_power_of_2(length));
+		xe_gt_assert(gt, !(length & GENMASK(ilog2(SZ_16M) - 1,
+						    ilog2(SZ_2M) + 1)));
+		xe_gt_assert(gt, IS_ALIGNED(start, length));
+
+		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_PAGE_SELECTIVE);
+		action[len++] = asid;
+		action[len++] = lower_32_bits(start);
+		action[len++] = upper_32_bits(start);
+		action[len++] = ilog2(length) - ilog2(SZ_4K);
+	}
+
+	xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN);
+
+	return send_tlb_inval(guc, action, len);
+}
+
+static bool tlb_inval_initialized(struct xe_tlb_inval *tlb_inval)
+{
+	struct xe_guc *guc = tlb_inval->private;
+
+	return xe_guc_ct_initialized(&guc->ct);
+}
+
+static void tlb_inval_flush(struct xe_tlb_inval *tlb_inval)
+{
+	struct xe_guc *guc = tlb_inval->private;
+
+	LNL_FLUSH_WORK(&guc->ct.g2h_worker);
+}
+
+static long tlb_inval_timeout_delay(struct xe_tlb_inval *tlb_inval)
+{
+	struct xe_guc *guc = tlb_inval->private;
+
+	/* this reflects what HW/GuC needs to process TLB inv request */
+	const long hw_tlb_timeout = HZ / 4;
+
+	/* this estimates actual delay caused by the CTB transport */
+	long delay = xe_guc_ct_queue_proc_time_jiffies(&guc->ct);
+
+	return hw_tlb_timeout + 2 * delay;
+}
+
+static const struct xe_tlb_inval_ops guc_tlb_inval_ops = {
+	.all = send_tlb_inval_all,
+	.ggtt = send_tlb_inval_ggtt,
+	.ppgtt = send_tlb_inval_ppgtt,
+	.initialized = tlb_inval_initialized,
+	.flush = tlb_inval_flush,
+	.timeout_delay = tlb_inval_timeout_delay,
+};
+
+/**
+ * xe_guc_tlb_inval_init_early() - Init GuC TLB invalidation early
+ * @guc: GuC object
+ * @tlb_inval: TLB invalidation client
+ *
+ * Inititialize GuC TLB invalidation by setting back pointer in TLB invalidation
+ * client to the GuC and setting GuC backend ops.
+ */
+void xe_guc_tlb_inval_init_early(struct xe_guc *guc,
+				 struct xe_tlb_inval *tlb_inval)
+{
+	tlb_inval->private = guc;
+	tlb_inval->ops = &guc_tlb_inval_ops;
+}
+
+/**
+ * xe_guc_tlb_inval_done_handler() - TLB invalidation done handler
+ * @guc: guc
+ * @msg: message indicating TLB invalidation done
+ * @len: length of message
+ *
+ * Parse seqno of TLB invalidation, wake any waiters for seqno, and signal any
+ * invalidation fences for seqno. Algorithm for this depends on seqno being
+ * received in-order and asserts this assumption.
+ *
+ * Return: 0 on success, -EPROTO for malformed messages.
+ */
+int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
+{
+	struct xe_gt *gt = guc_to_gt(guc);
+
+	if (unlikely(len != 1))
+		return -EPROTO;
+
+	xe_tlb_inval_done_handler(&gt->tlb_inval, msg[0]);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/xe/xe_guc_tlb_inval.h b/drivers/gpu/drm/xe/xe_guc_tlb_inval.h
new file mode 100644
index 000000000000..07d668b02e3d
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_guc_tlb_inval.h
@@ -0,0 +1,19 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_GUC_TLB_INVAL_H_
+#define _XE_GUC_TLB_INVAL_H_
+
+#include <linux/types.h>
+
+struct xe_guc;
+struct xe_tlb_inval;
+
+void xe_guc_tlb_inval_init_early(struct xe_guc *guc,
+				 struct xe_tlb_inval *tlb_inval);
+
+int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len);
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.c b/drivers/gpu/drm/xe/xe_tlb_inval.c
index d81a45f1dd7c..e6e97b5a7b5c 100644
--- a/drivers/gpu/drm/xe/xe_tlb_inval.c
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.c
@@ -12,51 +12,45 @@
 #include "xe_gt_printk.h"
 #include "xe_guc.h"
 #include "xe_guc_ct.h"
+#include "xe_guc_tlb_inval.h"
 #include "xe_gt_stats.h"
 #include "xe_tlb_inval.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
-#include "xe_sriov.h"
+#include "xe_tlb_inval.h"
 #include "xe_trace.h"
-#include "regs/xe_guc_regs.h"
+
+/**
+ * DOC: Xe TLB invalidation
+ *
+ * Xe TLB invalidation is implemented in two layers. The first is the frontend
+ * API, which provides an interface for TLB invalidations to the driver code.
+ * The frontend handles seqno assignment, synchronization (fences), and the
+ * timeout mechanism. The frontend is implemented via an embedded structure
+ * xe_tlb_inval that includes a set of ops hooking into the backend. The backend
+ * interacts with the hardware (or firmware) to perform the actual invalidation.
+ */
 
 #define FENCE_STACK_BIT		DMA_FENCE_FLAG_USER_BITS
 
-/*
- * TLB inval depends on pending commands in the CT queue and then the real
- * invalidation time. Double up the time to process full CT queue
- * just to be on the safe side.
- */
-static long tlb_timeout_jiffies(struct xe_gt *gt)
-{
-	/* this reflects what HW/GuC needs to process TLB inv request */
-	const long hw_tlb_timeout = HZ / 4;
-
-	/* this estimates actual delay caused by the CTB transport */
-	long delay = xe_guc_ct_queue_proc_time_jiffies(&gt->uc.guc.ct);
-
-	return hw_tlb_timeout + 2 * delay;
-}
-
 static void xe_tlb_inval_fence_fini(struct xe_tlb_inval_fence *fence)
 {
-	struct xe_gt *gt;
-
 	if (WARN_ON_ONCE(!fence->tlb_inval))
 		return;
 
-	gt = fence->tlb_inval->private;
-
-	xe_pm_runtime_put(gt_to_xe(gt));
+	xe_pm_runtime_put(fence->tlb_inval->xe);
 	fence->tlb_inval = NULL; /* fini() should be called once */
 }
 
 static void
-__inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
+xe_tlb_inval_fence_signal(struct xe_tlb_inval_fence *fence)
 {
 	bool stack = test_bit(FENCE_STACK_BIT, &fence->base.flags);
 
-	trace_xe_tlb_inval_fence_signal(xe, fence);
+	lockdep_assert_held(&fence->tlb_inval->pending_lock);
+
+	list_del(&fence->link);
+	trace_xe_tlb_inval_fence_signal(fence->tlb_inval->xe, fence);
 	xe_tlb_inval_fence_fini(fence);
 	dma_fence_signal(&fence->base);
 	if (!stack)
@@ -64,57 +58,65 @@ __inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
 }
 
 static void
-inval_fence_signal(struct xe_device *xe, struct xe_tlb_inval_fence *fence)
+xe_tlb_inval_fence_signal_unlocked(struct xe_tlb_inval_fence *fence)
 {
-	lockdep_assert_held(&fence->tlb_inval->pending_lock);
+	struct xe_tlb_inval *tlb_inval = fence->tlb_inval;
 
-	list_del(&fence->link);
-	__inval_fence_signal(xe, fence);
+	spin_lock_irq(&tlb_inval->pending_lock);
+	xe_tlb_inval_fence_signal(fence);
+	spin_unlock_irq(&tlb_inval->pending_lock);
 }
 
-static void
-inval_fence_signal_unlocked(struct xe_device *xe,
-			    struct xe_tlb_inval_fence *fence)
+static void xe_tlb_inval_fence_timeout(struct work_struct *work)
 {
-	spin_lock_irq(&fence->tlb_inval->pending_lock);
-	inval_fence_signal(xe, fence);
-	spin_unlock_irq(&fence->tlb_inval->pending_lock);
-}
-
-static void xe_gt_tlb_fence_timeout(struct work_struct *work)
-{
-	struct xe_gt *gt = container_of(work, struct xe_gt,
-					tlb_inval.fence_tdr.work);
-	struct xe_device *xe = gt_to_xe(gt);
+	struct xe_tlb_inval *tlb_inval = container_of(work, struct xe_tlb_inval,
+						      fence_tdr.work);
+	struct xe_device *xe = tlb_inval->xe;
 	struct xe_tlb_inval_fence *fence, *next;
+	long timeout_delay = tlb_inval->ops->timeout_delay(tlb_inval);
 
-	LNL_FLUSH_WORK(&gt->uc.guc.ct.g2h_worker);
+	tlb_inval->ops->flush(tlb_inval);
 
-	spin_lock_irq(&gt->tlb_inval.pending_lock);
+	spin_lock_irq(&tlb_inval->pending_lock);
 	list_for_each_entry_safe(fence, next,
-				 &gt->tlb_inval.pending_fences, link) {
+				 &tlb_inval->pending_fences, link) {
 		s64 since_inval_ms = ktime_ms_delta(ktime_get(),
 						    fence->inval_time);
 
-		if (msecs_to_jiffies(since_inval_ms) < tlb_timeout_jiffies(gt))
+		if (msecs_to_jiffies(since_inval_ms) < timeout_delay)
 			break;
 
 		trace_xe_tlb_inval_fence_timeout(xe, fence);
-		xe_gt_err(gt, "TLB invalidation fence timeout, seqno=%d recv=%d",
-			  fence->seqno, gt->tlb_inval.seqno_recv);
+		drm_err(&xe->drm,
+			"TLB invalidation fence timeout, seqno=%d recv=%d",
+			fence->seqno, tlb_inval->seqno_recv);
 
 		fence->base.error = -ETIME;
-		inval_fence_signal(xe, fence);
+		xe_tlb_inval_fence_signal(fence);
 	}
-	if (!list_empty(&gt->tlb_inval.pending_fences))
-		queue_delayed_work(system_wq,
-				   &gt->tlb_inval.fence_tdr,
-				   tlb_timeout_jiffies(gt));
-	spin_unlock_irq(&gt->tlb_inval.pending_lock);
+	if (!list_empty(&tlb_inval->pending_fences))
+		queue_delayed_work(system_wq, &tlb_inval->fence_tdr,
+				   timeout_delay);
+	spin_unlock_irq(&tlb_inval->pending_lock);
 }
 
 /**
- * xe_gt_tlb_inval_init_early - Initialize GT TLB invalidation state
+ * tlb_inval_fini - Clean up TLB invalidation state
+ * @drm: @drm_device
+ * @arg: pointer to struct @xe_tlb_inval
+ *
+ * Cancel pending fence workers and clean up any additional
+ * TLB invalidation state.
+ */
+static void tlb_inval_fini(struct drm_device *drm, void *arg)
+{
+	struct xe_tlb_inval *tlb_inval = arg;
+
+	xe_tlb_inval_reset(tlb_inval);
+}
+
+/**
+ * xe_gt_tlb_inval_init - Initialize TLB invalidation state
  * @gt: GT structure
  *
  * Initialize TLB invalidation state, purely software initialization, should
@@ -125,92 +127,84 @@ static void xe_gt_tlb_fence_timeout(struct work_struct *work)
 int xe_gt_tlb_inval_init_early(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
+	struct xe_tlb_inval *tlb_inval = &gt->tlb_inval;
 	int err;
 
-	gt->tlb_inval.private = gt;
-	gt->tlb_inval.seqno = 1;
-	INIT_LIST_HEAD(&gt->tlb_inval.pending_fences);
-	spin_lock_init(&gt->tlb_inval.pending_lock);
-	spin_lock_init(&gt->tlb_inval.lock);
-	INIT_DELAYED_WORK(&gt->tlb_inval.fence_tdr,
-			  xe_gt_tlb_fence_timeout);
+	tlb_inval->xe = xe;
+	tlb_inval->seqno = 1;
+	INIT_LIST_HEAD(&tlb_inval->pending_fences);
+	spin_lock_init(&tlb_inval->pending_lock);
+	spin_lock_init(&tlb_inval->lock);
+	INIT_DELAYED_WORK(&tlb_inval->fence_tdr, xe_tlb_inval_fence_timeout);
 
-	err = drmm_mutex_init(&xe->drm, &gt->tlb_inval.seqno_lock);
+	err = drmm_mutex_init(&xe->drm, &tlb_inval->seqno_lock);
 	if (err)
 		return err;
 
-	gt->tlb_inval.job_wq =
-		drmm_alloc_ordered_workqueue(&gt_to_xe(gt)->drm, "gt-tbl-inval-job-wq",
-					     WQ_MEM_RECLAIM);
-	if (IS_ERR(gt->tlb_inval.job_wq))
-		return PTR_ERR(gt->tlb_inval.job_wq);
+	tlb_inval->job_wq = drmm_alloc_ordered_workqueue(&xe->drm,
+							 "gt-tbl-inval-job-wq",
+							 WQ_MEM_RECLAIM);
+	if (IS_ERR(tlb_inval->job_wq))
+		return PTR_ERR(tlb_inval->job_wq);
 
-	return 0;
+	/* XXX: Blindly setting up backend to GuC */
+	xe_guc_tlb_inval_init_early(&gt->uc.guc, tlb_inval);
+
+	return drmm_add_action_or_reset(&xe->drm, tlb_inval_fini, tlb_inval);
 }
 
 /**
- * xe_tlb_inval_reset - Initialize TLB invalidation reset
+ * xe_tlb_inval_reset() - TLB invalidation reset
  * @tlb_inval: TLB invalidation client
  *
  * Signal any pending invalidation fences, should be called during a GT reset
  */
 void xe_tlb_inval_reset(struct xe_tlb_inval *tlb_inval)
 {
-	struct xe_gt *gt = tlb_inval->private;
 	struct xe_tlb_inval_fence *fence, *next;
 	int pending_seqno;
 
 	/*
-	 * we can get here before the CTs are even initialized if we're wedging
-	 * very early, in which case there are not going to be any pending
-	 * fences so we can bail immediately.
+	 * we can get here before the backends are even initialized if we're
+	 * wedging very early, in which case there are not going to be any
+	 * pendind fences so we can bail immediately.
 	 */
-	if (!xe_guc_ct_initialized(&gt->uc.guc.ct))
+	if (!tlb_inval->ops->initialized(tlb_inval))
 		return;
 
 	/*
-	 * CT channel is already disabled at this point. No new TLB requests can
+	 * Backend is already disabled at this point. No new TLB requests can
 	 * appear.
 	 */
 
-	mutex_lock(&gt->tlb_inval.seqno_lock);
-	spin_lock_irq(&gt->tlb_inval.pending_lock);
-	cancel_delayed_work(&gt->tlb_inval.fence_tdr);
+	mutex_lock(&tlb_inval->seqno_lock);
+	spin_lock_irq(&tlb_inval->pending_lock);
+	cancel_delayed_work(&tlb_inval->fence_tdr);
 	/*
 	 * We might have various kworkers waiting for TLB flushes to complete
 	 * which are not tracked with an explicit TLB fence, however at this
-	 * stage that will never happen since the CT is already disabled, so
-	 * make sure we signal them here under the assumption that we have
+	 * stage that will never happen since the backend is already disabled,
+	 * so make sure we signal them here under the assumption that we have
 	 * completed a full GT reset.
 	 */
-	if (gt->tlb_inval.seqno == 1)
+	if (tlb_inval->seqno == 1)
 		pending_seqno = TLB_INVALIDATION_SEQNO_MAX - 1;
 	else
-		pending_seqno = gt->tlb_inval.seqno - 1;
-	WRITE_ONCE(gt->tlb_inval.seqno_recv, pending_seqno);
+		pending_seqno = tlb_inval->seqno - 1;
+	WRITE_ONCE(tlb_inval->seqno_recv, pending_seqno);
 
 	list_for_each_entry_safe(fence, next,
-				 &gt->tlb_inval.pending_fences, link)
-		inval_fence_signal(gt_to_xe(gt), fence);
-	spin_unlock_irq(&gt->tlb_inval.pending_lock);
-	mutex_unlock(&gt->tlb_inval.seqno_lock);
+				 &tlb_inval->pending_fences, link)
+		xe_tlb_inval_fence_signal(fence);
+	spin_unlock_irq(&tlb_inval->pending_lock);
+	mutex_unlock(&tlb_inval->seqno_lock);
 }
 
-/**
- *
- * xe_gt_tlb_inval_fini - Clean up GT TLB invalidation state
- *
- * Cancel pending fence workers and clean up any additional
- * GT TLB invalidation state.
- */
-void xe_gt_tlb_inval_fini(struct xe_gt *gt)
+static bool xe_tlb_inval_seqno_past(struct xe_tlb_inval *tlb_inval, int seqno)
 {
-	xe_gt_tlb_inval_reset(gt);
-}
+	int seqno_recv = READ_ONCE(tlb_inval->seqno_recv);
 
-static bool tlb_inval_seqno_past(struct xe_gt *gt, int seqno)
-{
-	int seqno_recv = READ_ONCE(gt->tlb_inval.seqno_recv);
+	lockdep_assert_held(&tlb_inval->pending_lock);
 
 	if (seqno - seqno_recv < -(TLB_INVALIDATION_SEQNO_MAX / 2))
 		return false;
@@ -221,41 +215,20 @@ static bool tlb_inval_seqno_past(struct xe_gt *gt, int seqno)
 	return seqno_recv >= seqno;
 }
 
-static int send_tlb_inval(struct xe_guc *guc, const u32 *action, int len)
-{
-	struct xe_gt *gt = guc_to_gt(guc);
-
-	xe_gt_assert(gt, action[1]);	/* Seqno */
-
-	/*
-	 * XXX: The seqno algorithm relies on TLB invalidation being processed
-	 * in order which they currently are, if that changes the algorithm will
-	 * need to be updated.
-	 */
-
-	xe_gt_stats_incr(gt, XE_GT_STATS_ID_TLB_INVAL, 1);
-
-	return xe_guc_ct_send(&guc->ct, action, len,
-			      G2H_LEN_DW_TLB_INVALIDATE, 1);
-}
-
 static void xe_tlb_inval_fence_prep(struct xe_tlb_inval_fence *fence)
 {
 	struct xe_tlb_inval *tlb_inval = fence->tlb_inval;
-	struct xe_gt *gt = tlb_inval->private;
-	struct xe_device *xe = gt_to_xe(gt);
 
 	fence->seqno = tlb_inval->seqno;
-	trace_xe_tlb_inval_fence_send(xe, fence);
+	trace_xe_tlb_inval_fence_send(tlb_inval->xe, fence);
 
 	spin_lock_irq(&tlb_inval->pending_lock);
 	fence->inval_time = ktime_get();
 	list_add_tail(&fence->link, &tlb_inval->pending_fences);
 
 	if (list_is_singular(&tlb_inval->pending_fences))
-		queue_delayed_work(system_wq,
-				   &tlb_inval->fence_tdr,
-				   tlb_timeout_jiffies(gt));
+		queue_delayed_work(system_wq, &tlb_inval->fence_tdr,
+				   tlb_inval->ops->timeout_delay(tlb_inval));
 	spin_unlock_irq(&tlb_inval->pending_lock);
 
 	tlb_inval->seqno = (tlb_inval->seqno + 1) %
@@ -264,200 +237,63 @@ static void xe_tlb_inval_fence_prep(struct xe_tlb_inval_fence *fence)
 		tlb_inval->seqno = 1;
 }
 
-#define MAKE_INVAL_OP(type)	((type << XE_GUC_TLB_INVAL_TYPE_SHIFT) | \
-		XE_GUC_TLB_INVAL_MODE_HEAVY << XE_GUC_TLB_INVAL_MODE_SHIFT | \
-		XE_GUC_TLB_INVAL_FLUSH_CACHE)
-
-static int send_tlb_inval_ggtt(struct xe_gt *gt, int seqno)
-{
-	u32 action[] = {
-		XE_GUC_ACTION_TLB_INVALIDATION,
-		seqno,
-		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_GUC),
-	};
-
-	return send_tlb_inval(&gt->uc.guc, action, ARRAY_SIZE(action));
-}
-
-static int send_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
-			      struct xe_tlb_inval_fence *fence)
-{
-	u32 action[] = {
-		XE_GUC_ACTION_TLB_INVALIDATION_ALL,
-		0,  /* seqno, replaced in send_tlb_inval */
-		MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL),
-	};
-	struct xe_gt *gt = tlb_inval->private;
-
-	xe_gt_assert(gt, fence);
-
-	return send_tlb_inval(&gt->uc.guc, action, ARRAY_SIZE(action));
-}
+#define xe_tlb_inval_issue(__tlb_inval, __fence, op, args...)	\
+({								\
+	int __ret;						\
+								\
+	xe_assert((__tlb_inval)->xe, (__tlb_inval)->ops);	\
+	xe_assert((__tlb_inval)->xe, (__fence));		\
+								\
+	mutex_lock(&(__tlb_inval)->seqno_lock); 		\
+	xe_tlb_inval_fence_prep((__fence));			\
+	__ret = op((__tlb_inval), (__fence)->seqno, ##args);	\
+	if (__ret < 0)						\
+		xe_tlb_inval_fence_signal_unlocked((__fence));	\
+	mutex_unlock(&(__tlb_inval)->seqno_lock);		\
+								\
+	__ret == -ECANCELED ? 0 : __ret;			\
+})
 
 /**
- * xe_gt_tlb_invalidation_all - Invalidate all TLBs across PF and all VFs.
- * @gt: the &xe_gt structure
- * @fence: the &xe_tlb_inval_fence to be signaled on completion
+ * xe_tlb_inval_all() - Issue a TLB invalidation for all TLBs
+ * @tlb_inval: TLB invalidation client
+ * @fence: invalidation fence which will be signal on TLB invalidation
+ * completion
  *
- * Send a request to invalidate all TLBs across PF and all VFs.
+ * Issue a TLB invalidation for all TLBs. Completion of TLB is asynchronous and
+ * caller can use the invalidation fence to wait for completion.
  *
  * Return: 0 on success, negative error code on error
  */
 int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
 		     struct xe_tlb_inval_fence *fence)
 {
-	struct xe_gt *gt = tlb_inval->private;
-	int err;
-
-	err = send_tlb_inval_all(tlb_inval, fence);
-	if (err)
-		xe_gt_err(gt, "TLB invalidation request failed (%pe)", ERR_PTR(err));
-
-	return err;
-}
-
-/*
- * Ensure that roundup_pow_of_two(length) doesn't overflow.
- * Note that roundup_pow_of_two() operates on unsigned long,
- * not on u64.
- */
-#define MAX_RANGE_TLB_INVALIDATION_LENGTH (rounddown_pow_of_two(ULONG_MAX))
-
-static int send_tlb_inval_ppgtt(struct xe_gt *gt, u64 start, u64 end,
-				u32 asid, int seqno)
-{
-#define MAX_TLB_INVALIDATION_LEN	7
-	u32 action[MAX_TLB_INVALIDATION_LEN];
-	u64 length = end - start;
-	int len = 0;
-
-	action[len++] = XE_GUC_ACTION_TLB_INVALIDATION;
-	action[len++] = seqno;
-	if (!gt_to_xe(gt)->info.has_range_tlb_inval ||
-	    length > MAX_RANGE_TLB_INVALIDATION_LENGTH) {
-		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_FULL);
-	} else {
-		u64 orig_start = start;
-		u64 align;
-
-		if (length < SZ_4K)
-			length = SZ_4K;
-
-		/*
-		 * We need to invalidate a higher granularity if start address
-		 * is not aligned to length. When start is not aligned with
-		 * length we need to find the length large enough to create an
-		 * address mask covering the required range.
-		 */
-		align = roundup_pow_of_two(length);
-		start = ALIGN_DOWN(start, align);
-		end = ALIGN(end, align);
-		length = align;
-		while (start + length < end) {
-			length <<= 1;
-			start = ALIGN_DOWN(orig_start, length);
-		}
-
-		/*
-		 * Minimum invalidation size for a 2MB page that the hardware
-		 * expects is 16MB
-		 */
-		if (length >= SZ_2M) {
-			length = max_t(u64, SZ_16M, length);
-			start = ALIGN_DOWN(orig_start, length);
-		}
-
-		xe_gt_assert(gt, length >= SZ_4K);
-		xe_gt_assert(gt, is_power_of_2(length));
-		xe_gt_assert(gt, !(length & GENMASK(ilog2(SZ_16M) - 1,
-						    ilog2(SZ_2M) + 1)));
-		xe_gt_assert(gt, IS_ALIGNED(start, length));
-
-		action[len++] = MAKE_INVAL_OP(XE_GUC_TLB_INVAL_PAGE_SELECTIVE);
-		action[len++] = asid;
-		action[len++] = lower_32_bits(start);
-		action[len++] = upper_32_bits(start);
-		action[len++] = ilog2(length) - ilog2(SZ_4K);
-	}
-
-	xe_gt_assert(gt, len <= MAX_TLB_INVALIDATION_LEN);
-
-	return send_tlb_inval(&gt->uc.guc, action, len);
-}
-
-static int __xe_tlb_inval_ggtt(struct xe_gt *gt,
-			       struct xe_tlb_inval_fence *fence)
-{
-	int ret;
-
-	mutex_lock(&gt->tlb_inval.seqno_lock);
-	xe_tlb_inval_fence_prep(fence);
-
-	ret = send_tlb_inval_ggtt(gt, fence->seqno);
-	if (ret < 0)
-		inval_fence_signal_unlocked(gt_to_xe(gt), fence);
-	mutex_unlock(&gt->tlb_inval.seqno_lock);
-
-	/*
-	 * -ECANCELED indicates the CT is stopped for a GT reset. TLB caches
-	 *  should be nuked on a GT reset so this error can be ignored.
-	 */
-	if (ret == -ECANCELED)
-		return 0;
-
-	return ret;
+	return xe_tlb_inval_issue(tlb_inval, fence, tlb_inval->ops->all);
 }
 
 /**
- * xe_tlb_inval_ggtt - Issue a TLB invalidation on this GT for the GGTT
+ * xe_tlb_inval_ggtt() - Issue a TLB invalidation for the GGTT
  * @tlb_inval: TLB invalidation client
  *
- * Issue a TLB invalidation for the GGTT. Completion of TLB invalidation is
- * synchronous.
+ * Issue a TLB invalidation for the GGTT. Completion of TLB is asynchronous and
+ * caller can use the invalidation fence to wait for completion.
  *
  * Return: 0 on success, negative error code on error
  */
 int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval)
 {
-	struct xe_gt *gt = tlb_inval->private;
-	struct xe_device *xe = gt_to_xe(gt);
-	unsigned int fw_ref;
+	struct xe_tlb_inval_fence fence, *fence_ptr = &fence;
+	int ret;
 
-	if (xe_guc_ct_enabled(&gt->uc.guc.ct) &&
-	    gt->uc.guc.submission_state.enabled) {
-		struct xe_tlb_inval_fence fence;
-		int ret;
+	xe_tlb_inval_fence_init(tlb_inval, fence_ptr, true);
+	ret = xe_tlb_inval_issue(tlb_inval, fence_ptr, tlb_inval->ops->ggtt);
+	xe_tlb_inval_fence_wait(fence_ptr);
 
-		xe_tlb_inval_fence_init(tlb_inval, &fence, true);
-		ret = __xe_tlb_inval_ggtt(gt, &fence);
-		if (ret)
-			return ret;
-
-		xe_tlb_inval_fence_wait(&fence);
-	} else if (xe_device_uc_enabled(xe) && !xe_device_wedged(xe)) {
-		struct xe_mmio *mmio = &gt->mmio;
-
-		if (IS_SRIOV_VF(xe))
-			return 0;
-
-		fw_ref = xe_force_wake_get(gt_to_fw(gt), XE_FW_GT);
-		if (xe->info.platform == XE_PVC || GRAPHICS_VER(xe) >= 20) {
-			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC1,
-					PVC_GUC_TLB_INV_DESC1_INVALIDATE);
-			xe_mmio_write32(mmio, PVC_GUC_TLB_INV_DESC0,
-					PVC_GUC_TLB_INV_DESC0_VALID);
-		} else {
-			xe_mmio_write32(mmio, GUC_TLB_INV_CR,
-					GUC_TLB_INV_CR_INVALIDATE);
-		}
-		xe_force_wake_put(gt_to_fw(gt), fw_ref);
-	}
-
-	return 0;
+	return ret;
 }
 
 /**
- * xe_tlb_inval_range - Issue a TLB invalidation on this GT for an address range
+ * xe_tlb_inval_range() - Issue a TLB invalidation for an address range
  * @tlb_inval: TLB invalidation client
  * @fence: invalidation fence which will be signal on TLB invalidation
  * completion
@@ -475,31 +311,12 @@ int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
 		       struct xe_tlb_inval_fence *fence, u64 start, u64 end,
 		       u32 asid)
 {
-	struct xe_gt *gt = tlb_inval->private;
-	struct xe_device *xe = gt_to_xe(gt);
-	int  ret;
-
-	xe_gt_assert(gt, fence);
-
-	/* Execlists not supported */
-	if (xe->info.force_execlist) {
-		__inval_fence_signal(xe, fence);
-		return 0;
-	}
-
-	mutex_lock(&gt->tlb_inval.seqno_lock);
-	xe_tlb_inval_fence_prep(fence);
-
-	ret = send_tlb_inval_ppgtt(gt, start, end, asid, fence->seqno);
-	if (ret < 0)
-		inval_fence_signal_unlocked(xe, fence);
-	mutex_unlock(&gt->tlb_inval.seqno_lock);
-
-	return ret;
+	return xe_tlb_inval_issue(tlb_inval, fence, tlb_inval->ops->ppgtt,
+				  start, end, asid);
 }
 
 /**
- * xe_tlb_inval_vm - Issue a TLB invalidation on this GT for a VM
+ * xe_tlb_inval_vm() - Issue a TLB invalidation for a VM
  * @tlb_inval: TLB invalidation client
  * @vm: VM to invalidate
  *
@@ -509,27 +326,22 @@ void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm)
 {
 	struct xe_tlb_inval_fence fence;
 	u64 range = 1ull << vm->xe->info.va_bits;
-	int ret;
 
 	xe_tlb_inval_fence_init(tlb_inval, &fence, true);
-
-	ret = xe_tlb_inval_range(tlb_inval, &fence, 0, range, vm->usm.asid);
-	if (ret < 0)
-		return;
-
+	xe_tlb_inval_range(tlb_inval, &fence, 0, range, vm->usm.asid);
 	xe_tlb_inval_fence_wait(&fence);
 }
 
 /**
- * xe_tlb_inval_done_handler - TLB invalidation done handler
- * @gt: gt
+ * xe_tlb_inval_done_handler() - TLB invalidation done handler
+ * @tlb_inval: TLB invalidation client
  * @seqno: seqno of invalidation that is done
  *
  * Update recv seqno, signal any TLB invalidation fences, and restart TDR
  */
-static void xe_tlb_inval_done_handler(struct xe_gt *gt, int seqno)
+void xe_tlb_inval_done_handler(struct xe_tlb_inval *tlb_inval, int seqno)
 {
-	struct xe_device *xe = gt_to_xe(gt);
+	struct xe_device *xe = tlb_inval->xe;
 	struct xe_tlb_inval_fence *fence, *next;
 	unsigned long flags;
 
@@ -548,77 +360,53 @@ static void xe_tlb_inval_done_handler(struct xe_gt *gt, int seqno)
 	 * officially process the CT message like if racing against
 	 * process_g2h_msg().
 	 */
-	spin_lock_irqsave(&gt->tlb_inval.pending_lock, flags);
-	if (tlb_inval_seqno_past(gt, seqno)) {
-		spin_unlock_irqrestore(&gt->tlb_inval.pending_lock, flags);
+	spin_lock_irqsave(&tlb_inval->pending_lock, flags);
+	if (xe_tlb_inval_seqno_past(tlb_inval, seqno)) {
+		spin_unlock_irqrestore(&tlb_inval->pending_lock, flags);
 		return;
 	}
 
-	WRITE_ONCE(gt->tlb_inval.seqno_recv, seqno);
+	WRITE_ONCE(tlb_inval->seqno_recv, seqno);
 
 	list_for_each_entry_safe(fence, next,
-				 &gt->tlb_inval.pending_fences, link) {
+				 &tlb_inval->pending_fences, link) {
 		trace_xe_tlb_inval_fence_recv(xe, fence);
 
-		if (!tlb_inval_seqno_past(gt, fence->seqno))
+		if (!xe_tlb_inval_seqno_past(tlb_inval, fence->seqno))
 			break;
 
-		inval_fence_signal(xe, fence);
+		xe_tlb_inval_fence_signal(fence);
 	}
 
-	if (!list_empty(&gt->tlb_inval.pending_fences))
+	if (!list_empty(&tlb_inval->pending_fences))
 		mod_delayed_work(system_wq,
-				 &gt->tlb_inval.fence_tdr,
-				 tlb_timeout_jiffies(gt));
+				 &tlb_inval->fence_tdr,
+				 tlb_inval->ops->timeout_delay(tlb_inval));
 	else
-		cancel_delayed_work(&gt->tlb_inval.fence_tdr);
+		cancel_delayed_work(&tlb_inval->fence_tdr);
 
-	spin_unlock_irqrestore(&gt->tlb_inval.pending_lock, flags);
-}
-
-/**
- * xe_guc_tlb_inval_done_handler - TLB invalidation done handler
- * @guc: guc
- * @msg: message indicating TLB invalidation done
- * @len: length of message
- *
- * Parse seqno of TLB invalidation, wake any waiters for seqno, and signal any
- * invalidation fences for seqno. Algorithm for this depends on seqno being
- * received in-order and asserts this assumption.
- *
- * Return: 0 on success, -EPROTO for malformed messages.
- */
-int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len)
-{
-	struct xe_gt *gt = guc_to_gt(guc);
-
-	if (unlikely(len != 1))
-		return -EPROTO;
-
-	xe_tlb_inval_done_handler(gt, msg[0]);
-
-	return 0;
+	spin_unlock_irqrestore(&tlb_inval->pending_lock, flags);
 }
 
 static const char *
-inval_fence_get_driver_name(struct dma_fence *dma_fence)
+xe_inval_fence_get_driver_name(struct dma_fence *dma_fence)
 {
 	return "xe";
 }
 
 static const char *
-inval_fence_get_timeline_name(struct dma_fence *dma_fence)
+xe_inval_fence_get_timeline_name(struct dma_fence *dma_fence)
 {
-	return "inval_fence";
+	return "tlb_inval_fence";
 }
 
 static const struct dma_fence_ops inval_fence_ops = {
-	.get_driver_name = inval_fence_get_driver_name,
-	.get_timeline_name = inval_fence_get_timeline_name,
+	.get_driver_name = xe_inval_fence_get_driver_name,
+	.get_timeline_name = xe_inval_fence_get_timeline_name,
 };
 
 /**
- * xe_tlb_inval_fence_init - Initialize TLB invalidation fence
+ * xe_tlb_inval_fence_init() - Initialize TLB invalidation fence
  * @tlb_inval: TLB invalidation client
  * @fence: TLB invalidation fence to initialize
  * @stack: fence is stack variable
@@ -631,15 +419,12 @@ void xe_tlb_inval_fence_init(struct xe_tlb_inval *tlb_inval,
 			     struct xe_tlb_inval_fence *fence,
 			     bool stack)
 {
-	struct xe_gt *gt = tlb_inval->private;
+	xe_pm_runtime_get_noresume(tlb_inval->xe);
 
-	xe_pm_runtime_get_noresume(gt_to_xe(gt));
-
-	spin_lock_irq(&gt->tlb_inval.lock);
-	dma_fence_init(&fence->base, &inval_fence_ops,
-		       &gt->tlb_inval.lock,
+	spin_lock_irq(&tlb_inval->lock);
+	dma_fence_init(&fence->base, &inval_fence_ops, &tlb_inval->lock,
 		       dma_fence_context_alloc(1), 1);
-	spin_unlock_irq(&gt->tlb_inval.lock);
+	spin_unlock_irq(&tlb_inval->lock);
 	INIT_LIST_HEAD(&fence->link);
 	if (stack)
 		set_bit(FENCE_STACK_BIT, &fence->base.flags);
diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.h b/drivers/gpu/drm/xe/xe_tlb_inval.h
index ab6f769c50be..554634dfd4e2 100644
--- a/drivers/gpu/drm/xe/xe_tlb_inval.h
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.h
@@ -15,27 +15,32 @@ struct xe_guc;
 struct xe_vm;
 
 int xe_gt_tlb_inval_init_early(struct xe_gt *gt);
-void xe_gt_tlb_inval_fini(struct xe_gt *gt);
 
 void xe_tlb_inval_reset(struct xe_tlb_inval *tlb_inval);
-int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval);
-void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm);
 int xe_tlb_inval_all(struct xe_tlb_inval *tlb_inval,
 		     struct xe_tlb_inval_fence *fence);
+int xe_tlb_inval_ggtt(struct xe_tlb_inval *tlb_inval);
+void xe_tlb_inval_vm(struct xe_tlb_inval *tlb_inval, struct xe_vm *vm);
 int xe_tlb_inval_range(struct xe_tlb_inval *tlb_inval,
 		       struct xe_tlb_inval_fence *fence,
 		       u64 start, u64 end, u32 asid);
-int xe_guc_tlb_inval_done_handler(struct xe_guc *guc, u32 *msg, u32 len);
 
 void xe_tlb_inval_fence_init(struct xe_tlb_inval *tlb_inval,
 			     struct xe_tlb_inval_fence *fence,
 			     bool stack);
-void xe_tlb_inval_fence_signal(struct xe_tlb_inval_fence *fence);
 
+/**
+ * xe_tlb_inval_fence_wait() - TLB invalidiation fence wait
+ * @fence: TLB invalidation fence to wait on
+ *
+ * Wait on a TLB invalidiation fence until it signals, non interruptable
+ */
 static inline void
 xe_tlb_inval_fence_wait(struct xe_tlb_inval_fence *fence)
 {
 	dma_fence_wait(&fence->base, false);
 }
 
+void xe_tlb_inval_done_handler(struct xe_tlb_inval *tlb_inval, int seqno);
+
 #endif	/* _XE_TLB_INVAL_ */
diff --git a/drivers/gpu/drm/xe/xe_tlb_inval_types.h b/drivers/gpu/drm/xe/xe_tlb_inval_types.h
index 6d14b9f17b91..8f8b060e9005 100644
--- a/drivers/gpu/drm/xe/xe_tlb_inval_types.h
+++ b/drivers/gpu/drm/xe/xe_tlb_inval_types.h
@@ -9,10 +9,75 @@
 #include <linux/workqueue.h>
 #include <linux/dma-fence.h>
 
-/** struct xe_tlb_inval - TLB invalidation client */
+struct xe_tlb_inval;
+
+/** struct xe_tlb_inval_ops - TLB invalidation ops (backend) */
+struct xe_tlb_inval_ops {
+	/**
+	 * @all: Invalidate all TLBs
+	 * @tlb_inval: TLB invalidation client
+	 * @seqno: Seqno of TLB invalidation
+	 *
+	 * Return 0 on success, -ECANCELED if backend is mid-reset, error on
+	 * failure
+	 */
+	int (*all)(struct xe_tlb_inval *tlb_inval, u32 seqno);
+
+	/**
+	 * @ggtt: Invalidate global translation TLBs
+	 * @tlb_inval: TLB invalidation client
+	 * @seqno: Seqno of TLB invalidation
+	 *
+	 * Return 0 on success, -ECANCELED if backend is mid-reset, error on
+	 * failure
+	 */
+	int (*ggtt)(struct xe_tlb_inval *tlb_inval, u32 seqno);
+
+	/**
+	 * @ppgtt: Invalidate per-process translation TLBs
+	 * @tlb_inval: TLB invalidation client
+	 * @seqno: Seqno of TLB invalidation
+	 * @start: Start address
+	 * @end: End address
+	 * @asid: Address space ID
+	 *
+	 * Return 0 on success, -ECANCELED if backend is mid-reset, error on
+	 * failure
+	 */
+	int (*ppgtt)(struct xe_tlb_inval *tlb_inval, u32 seqno, u64 start,
+		     u64 end, u32 asid);
+
+	/**
+	 * @initialized: Backend is initialized
+	 * @tlb_inval: TLB invalidation client
+	 *
+	 * Return: True if back is initialized, False otherwise
+	 */
+	bool (*initialized)(struct xe_tlb_inval *tlb_inval);
+
+	/**
+	 * @flush: Flush pending TLB invalidations
+	 * @tlb_inval: TLB invalidation client
+	 */
+	void (*flush)(struct xe_tlb_inval *tlb_inval);
+
+	/**
+	 * @timeout_delay: Timeout delay for TLB invalidation
+	 * @tlb_inval: TLB invalidation client
+	 *
+	 * Return: Timeout delay for TLB invalidation in jiffies
+	 */
+	long (*timeout_delay)(struct xe_tlb_inval *tlb_inval);
+};
+
+/** struct xe_tlb_inval - TLB invalidation client (frontend) */
 struct xe_tlb_inval {
 	/** @private: Backend private pointer */
 	void *private;
+	/** @xe: Pointer to Xe device */
+	struct xe_device *xe;
+	/** @ops: TLB invalidation ops */
+	const struct xe_tlb_inval_ops *ops;
 	/** @tlb_inval.seqno: TLB invalidation seqno, protected by CT lock */
 #define TLB_INVALIDATION_SEQNO_MAX	0x100000
 	int seqno;

From 5c76c794bf29399394ebacaa5af8436b8bed0d46 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Wed, 27 Aug 2025 15:57:31 +0300
Subject: [PATCH 0706/1455] HID: i2c-hid: Fix test in
 i2c_hid_core_register_panel_follower()

Bitwise AND was intended instead of OR.  With the current code the
condition is always true.

Fixes: cbdd16b818ee ("HID: i2c-hid: Make elan touch controllers power on after panel is enabled")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Pin-yen Lin <treapking@chromium.org>
Acked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/aK8Au3CgZSTvfEJ6@stanley.mountain
---
 drivers/hid/i2c-hid/i2c-hid-core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/hid/i2c-hid/i2c-hid-core.c b/drivers/hid/i2c-hid/i2c-hid-core.c
index 99ce6386176c..30ebde1273be 100644
--- a/drivers/hid/i2c-hid/i2c-hid-core.c
+++ b/drivers/hid/i2c-hid/i2c-hid-core.c
@@ -1189,7 +1189,7 @@ static int i2c_hid_core_register_panel_follower(struct i2c_hid *ihid)
 	struct device *dev = &ihid->client->dev;
 	int ret;
 
-	if (ihid->hid->initial_quirks | HID_QUIRK_POWER_ON_AFTER_BACKLIGHT)
+	if (ihid->hid->initial_quirks & HID_QUIRK_POWER_ON_AFTER_BACKLIGHT)
 		ihid->panel_follower.funcs = &i2c_hid_core_panel_follower_enable_funcs;
 	else
 		ihid->panel_follower.funcs = &i2c_hid_core_panel_follower_prepare_funcs;

From ec035aba7d4a9439d7fa6d54b9f9de6ce3fe1716 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Thu, 14 Aug 2025 14:36:27 +0100
Subject: [PATCH 0707/1455] drm/sched: Remove mention of indirect buffers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Indirect buffers are an AMD term describing essentialy a job submitted to
the scheduler, just a lower level one. Since scheduler was promoted to be
generic long ago, lets replace those references with jobs.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Danilo Krummrich <dakr@kernel.org>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Philipp Stanner <phasta@kernel.org>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250814133627.2550-1-tvrtko.ursulin@igalia.com
---
 drivers/gpu/drm/scheduler/sched_entity.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 8867b95ab089..254932b9050e 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -285,9 +285,9 @@ long drm_sched_entity_flush(struct drm_sched_entity *entity, long timeout)
 		return 0;
 
 	sched = entity->rq->sched;
-	/**
-	 * The client will not queue more IBs during this fini, consume existing
-	 * queued IBs or discard them on SIGKILL
+	/*
+	 * The client will not queue more jobs during this fini - consume
+	 * existing queued ones, or discard them on SIGKILL.
 	 */
 	if (current->flags & PF_EXITING) {
 		if (timeout)
@@ -300,7 +300,7 @@ long drm_sched_entity_flush(struct drm_sched_entity *entity, long timeout)
 				    drm_sched_entity_is_idle(entity));
 	}
 
-	/* For killed process disable any more IBs enqueue right now */
+	/* For a killed process disallow further enqueueing of jobs. */
 	last_user = cmpxchg(&entity->last_user, current->group_leader, NULL);
 	if ((!last_user || last_user == current->group_leader) &&
 	    (current->flags & PF_EXITING) && (current->exit_code == SIGKILL))
@@ -324,9 +324,9 @@ EXPORT_SYMBOL(drm_sched_entity_flush);
 void drm_sched_entity_fini(struct drm_sched_entity *entity)
 {
 	/*
-	 * If consumption of existing IBs wasn't completed. Forcefully remove
-	 * them here. Also makes sure that the scheduler won't touch this entity
-	 * any more.
+	 * If consumption of existing jobs wasn't completed forcefully remove
+	 * them. Also makes sure that the scheduler won't touch this entity any
+	 * more.
 	 */
 	drm_sched_entity_kill(entity);
 

From 77a62e557f54ecf6305e7ed6fb05d02e8748bffb Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Tue, 19 Aug 2025 22:26:24 +0800
Subject: [PATCH 0708/1455] drm/sched/tests: Remove redundant header files

The header file <linux/atomic.h> is already included on line 8. Remove the
redundant include.

Fixes: 5a99350794fec ("drm/sched: Add scheduler unit testing infrastructure and some basic tests")
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250819142630.368796-1-liaoyuanhong@vivo.com
---
 drivers/gpu/drm/scheduler/tests/sched_tests.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/scheduler/tests/sched_tests.h b/drivers/gpu/drm/scheduler/tests/sched_tests.h
index 5b262126b776..7f31d35780cc 100644
--- a/drivers/gpu/drm/scheduler/tests/sched_tests.h
+++ b/drivers/gpu/drm/scheduler/tests/sched_tests.h
@@ -11,7 +11,6 @@
 #include <linux/hrtimer.h>
 #include <linux/ktime.h>
 #include <linux/list.h>
-#include <linux/atomic.h>
 #include <linux/mutex.h>
 #include <linux/types.h>
 

From f4c75f975cf50fa2e1fd96c5aafe5aa62e55fbe4 Mon Sep 17 00:00:00 2001
From: Philipp Stanner <phasta@kernel.org>
Date: Wed, 13 Aug 2025 10:56:55 +0200
Subject: [PATCH 0709/1455] drm/sched: Document race condition in
 drm_sched_fini()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In drm_sched_fini() all entities are marked as stopped - without taking
the appropriate lock, because that would deadlock. That means that
drm_sched_fini() and drm_sched_entity_push_job() can race against each
other.

This should most likely be fixed by establishing the rule that all
entities associated with a scheduler must be torn down first. Then,
however, the locking should be removed from drm_sched_fini() alltogether
with an appropriate comment.

Reported-by: James Flowers <bold.zone2373@fastmail.com>
Link: https://lore.kernel.org/dri-devel/20250720235748.2798-1-bold.zone2373@fastmail.com/
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250813085654.102504-2-phasta@kernel.org
---
 drivers/gpu/drm/scheduler/sched_main.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 5a550fd76bf0..46119aacb809 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -1424,6 +1424,22 @@ void drm_sched_fini(struct drm_gpu_scheduler *sched)
 			 * Prevents reinsertion and marks job_queue as idle,
 			 * it will be removed from the rq in drm_sched_entity_fini()
 			 * eventually
+			 *
+			 * FIXME:
+			 * This lacks the proper spin_lock(&s_entity->lock) and
+			 * is, therefore, a race condition. Most notably, it
+			 * can race with drm_sched_entity_push_job(). The lock
+			 * cannot be taken here, however, because this would
+			 * lead to lock inversion -> deadlock.
+			 *
+			 * The best solution probably is to enforce the life
+			 * time rule of all entities having to be torn down
+			 * before their scheduler. Then, however, locking could
+			 * be dropped alltogether from this function.
+			 *
+			 * For now, this remains a potential race in all
+			 * drivers that keep entities alive for longer than
+			 * the scheduler.
 			 */
 			s_entity->stopped = true;
 		spin_unlock(&rq->lock);

From 81a45cb7ea31a59777af294de590eea66ab80691 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Fri, 8 Aug 2025 12:04:53 +0100
Subject: [PATCH 0710/1455] drm/xe/migrate: make MI_TLB_INVALIDATE conditional
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When clearing VRAM we should be able to skip invalidating the TLBs if we
are only using the identity map to access VRAM (which is the common
case), since no modifications are made to PTEs on the fly. Also since we
use huge 1G entries within the identity map, there should be a pretty
decent chance that the next packet(s) (if also clears) can avoid a tree
walk if we don't shoot down the TLBs, like if we have to process a long
stream of clears.

For normal moves/copies, we usually always end up with the src or dst
being system memory, meaning we can't only rely on the identity map and
will also need to emit PTEs and so will always require a TLB flush.

v2:
  - Update commit to explain the situation for normal copies (Matt B)
  - Rebase on latest changes

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250808110452.467513-2-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c  | 18 +++++++++++-------
 drivers/gpu/drm/xe/xe_ring_ops.c | 10 +++++-----
 2 files changed, 16 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 57e6d5a8ac39..9643442ef101 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -904,7 +904,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 			goto err;
 		}
 
-		xe_sched_job_add_migrate_flush(job, flush_flags);
+		xe_sched_job_add_migrate_flush(job, flush_flags | MI_INVALIDATE_TLB);
 		if (!fence) {
 			err = xe_sched_job_add_deps(job, src_bo->ttm.base.resv,
 						    DMA_RESV_USAGE_BOOKKEEP);
@@ -1288,11 +1288,13 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m,
 
 		size -= clear_L0;
 		/* Preemption is enabled again by the ring ops. */
-		if (clear_vram && xe_migrate_allow_identity(clear_L0, &src_it))
+		if (clear_vram && xe_migrate_allow_identity(clear_L0, &src_it)) {
 			xe_res_next(&src_it, clear_L0);
-		else
-			emit_pte(m, bb, clear_L0_pt, clear_vram, clear_only_system_ccs,
-				 &src_it, clear_L0, dst);
+		} else {
+			emit_pte(m, bb, clear_L0_pt, clear_vram,
+				 clear_only_system_ccs, &src_it, clear_L0, dst);
+			flush_flags |= MI_INVALIDATE_TLB;
+		}
 
 		bb->cs[bb->len++] = MI_BATCH_BUFFER_END;
 		update_idx = bb->len;
@@ -1303,7 +1305,7 @@ struct dma_fence *xe_migrate_clear(struct xe_migrate *m,
 		if (xe_migrate_needs_ccs_emit(xe)) {
 			emit_copy_ccs(gt, bb, clear_L0_ofs, true,
 				      m->cleared_mem_ofs, false, clear_L0);
-			flush_flags = MI_FLUSH_DW_CCS;
+			flush_flags |= MI_FLUSH_DW_CCS;
 		}
 
 		job = xe_bb_create_migration_job(m->q, bb,
@@ -1638,6 +1640,8 @@ next_cmd:
 		goto err_sa;
 	}
 
+	xe_sched_job_add_migrate_flush(job, MI_INVALIDATE_TLB);
+
 	if (ops->pre_commit) {
 		pt_update->job = job;
 		err = ops->pre_commit(pt_update);
@@ -1863,7 +1867,7 @@ static struct dma_fence *xe_migrate_vram(struct xe_migrate *m,
 		goto err;
 	}
 
-	xe_sched_job_add_migrate_flush(job, 0);
+	xe_sched_job_add_migrate_flush(job, MI_INVALIDATE_TLB);
 
 	mutex_lock(&m->job_mutex);
 	xe_sched_job_arm(job);
diff --git a/drivers/gpu/drm/xe/xe_ring_ops.c b/drivers/gpu/drm/xe/xe_ring_ops.c
index 5f15360d14bf..d71837773d6c 100644
--- a/drivers/gpu/drm/xe/xe_ring_ops.c
+++ b/drivers/gpu/drm/xe/xe_ring_ops.c
@@ -110,10 +110,10 @@ static int emit_bb_start(u64 batch_addr, u32 ppgtt_flag, u32 *dw, int i)
 	return i;
 }
 
-static int emit_flush_invalidate(u32 addr, u32 val, u32 *dw, int i)
+static int emit_flush_invalidate(u32 addr, u32 val, u32 flush_flags, u32 *dw, int i)
 {
-	dw[i++] = MI_FLUSH_DW | MI_INVALIDATE_TLB | MI_FLUSH_DW_OP_STOREDW |
-		  MI_FLUSH_IMM_DW;
+	dw[i++] = MI_FLUSH_DW | MI_FLUSH_DW_OP_STOREDW |
+		  MI_FLUSH_IMM_DW | (flush_flags & MI_INVALIDATE_TLB) ?: 0;
 
 	dw[i++] = addr | MI_FLUSH_DW_USE_GTT;
 	dw[i++] = 0;
@@ -410,13 +410,13 @@ static void emit_migration_job_gen12(struct xe_sched_job *job,
 	i = emit_bb_start(job->ptrs[0].batch_addr, BIT(8), dw, i);
 
 	dw[i++] = preparser_disable(true);
-	i = emit_flush_invalidate(saddr, seqno, dw, i);
+	i = emit_flush_invalidate(saddr, seqno, job->migrate_flush_flags, dw, i);
 	dw[i++] = preparser_disable(false);
 
 	i = emit_bb_start(job->ptrs[1].batch_addr, BIT(8), dw, i);
 
 	i = emit_flush_imm_ggtt(xe_lrc_seqno_ggtt_addr(lrc), seqno,
-				MI_INVALIDATE_TLB | job->migrate_flush_flags,
+				job->migrate_flush_flags,
 				dw, i);
 
 	i = emit_user_interrupt(dw, i);

From 9df8043a546d2eb3adfaba920c027c0d701c73a1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 26 Aug 2025 15:18:57 +0300
Subject: [PATCH 0711/1455] iopoll: Generalize read_poll_timeout() into
 poll_timeout_us()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

While read_poll_timeout() & co. were originally introduced just
for simple I/O usage scenarios they have since been generalized to
be useful in more cases.

However the interface is very cumbersome to use in the general case.
Attempt to make it more flexible by combining the 'op', 'var' and
'args' parameter into just a single 'op' that the caller can fully
specify.

For example i915 has one case where one might currently
have to write something like:
	ret = read_poll_timeout(drm_dp_dpcd_read_byte, err,
				err || (status & mask),
				0 * 1000, 200 * 1000, false,
				aux, DP_FEC_STATUS, &status);
which is practically illegible, but with the adjusted macro
we do:
	ret = poll_timeout_us(err = drm_dp_dpcd_read_byte(aux, DP_FEC_STATUS, &status),
			      err || (status & mask),
			      0 * 1000, 200 * 1000, false);
which much easier to understand.

One could even combine the 'op' and 'cond'  parameters into
one, but that might make the caller a bit too unwieldly with
assignments and checks being done on the same statement.

This makes poll_timeout_us() closer to the i915 __wait_for()
macro, with the main difference being that __wait_for() uses
expenential backoff as opposed to the fixed polling interval
used by poll_timeout_us(). Eventually we might be able to switch
(at least most of) i915 to use poll_timeout_us().

v2: Fix typos (Jani)
    Fix delay_us docs for poll_timeout_us_atomic() (Jani)

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: David Laight <david.laight.linux@gmail.com>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Matt Wagantall <mattw@codeaurora.org>
Cc: Dejin Zheng <zhengdejin5@gmail.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: intel-xe@lists.freedesktop.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Simona Vetter <simona.vetter@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250826121859.15497-1-ville.syrjala@linux.intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 include/linux/iopoll.h | 158 ++++++++++++++++++++++++++---------------
 1 file changed, 102 insertions(+), 56 deletions(-)

diff --git a/include/linux/iopoll.h b/include/linux/iopoll.h
index 91324c331a4b..440aca5b4b59 100644
--- a/include/linux/iopoll.h
+++ b/include/linux/iopoll.h
@@ -13,9 +13,104 @@
 #include <linux/errno.h>
 #include <linux/io.h>
 
+/**
+ * poll_timeout_us - Periodically poll and perform an operation until
+ *                   a condition is met or a timeout occurs
+ *
+ * @op: Operation
+ * @cond: Break condition
+ * @sleep_us: Maximum time to sleep between operations in us (0 tight-loops).
+ *            Please read usleep_range() function description for details and
+ *            limitations.
+ * @timeout_us: Timeout in us, 0 means never timeout
+ * @sleep_before_op: if it is true, sleep @sleep_us before operation.
+ *
+ * When available, you'll probably want to use one of the specialized
+ * macros defined below rather than this macro directly.
+ *
+ * Returns: 0 on success and -ETIMEDOUT upon a timeout. Must not
+ * be called from atomic context if sleep_us or timeout_us are used.
+ */
+#define poll_timeout_us(op, cond, sleep_us, timeout_us, sleep_before_op) \
+({ \
+	u64 __timeout_us = (timeout_us); \
+	unsigned long __sleep_us = (sleep_us); \
+	ktime_t __timeout = ktime_add_us(ktime_get(), __timeout_us); \
+	might_sleep_if((__sleep_us) != 0); \
+	if ((sleep_before_op) && __sleep_us) \
+		usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
+	for (;;) { \
+		op; \
+		if (cond) \
+			break; \
+		if (__timeout_us && \
+		    ktime_compare(ktime_get(), __timeout) > 0) { \
+			op; \
+			break; \
+		} \
+		if (__sleep_us) \
+			usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
+		cpu_relax(); \
+	} \
+	(cond) ? 0 : -ETIMEDOUT; \
+})
+
+/**
+ * poll_timeout_us_atomic - Periodically poll and perform an operation until
+ *                          a condition is met or a timeout occurs
+ *
+ * @op: Operation
+ * @cond: Break condition
+ * @delay_us: Time to udelay between operations in us (0 tight-loops).
+ *            Please read udelay() function description for details and
+ *            limitations.
+ * @timeout_us: Timeout in us, 0 means never timeout
+ * @delay_before_op: if it is true, delay @delay_us before operation.
+ *
+ * This macro does not rely on timekeeping.  Hence it is safe to call even when
+ * timekeeping is suspended, at the expense of an underestimation of wall clock
+ * time, which is rather minimal with a non-zero delay_us.
+ *
+ * When available, you'll probably want to use one of the specialized
+ * macros defined below rather than this macro directly.
+ *
+ * Returns: 0 on success and -ETIMEDOUT upon a timeout.
+ */
+#define poll_timeout_us_atomic(op, cond, delay_us, timeout_us, \
+			       delay_before_op) \
+({ \
+	u64 __timeout_us = (timeout_us); \
+	s64 __left_ns = __timeout_us * NSEC_PER_USEC; \
+	unsigned long __delay_us = (delay_us); \
+	u64 __delay_ns = __delay_us * NSEC_PER_USEC; \
+	if ((delay_before_op) && __delay_us) { \
+		udelay(__delay_us); \
+		if (__timeout_us) \
+			__left_ns -= __delay_ns; \
+	} \
+	for (;;) { \
+		op; \
+		if (cond) \
+			break; \
+		if (__timeout_us && __left_ns < 0) { \
+			op; \
+			break; \
+		} \
+		if (__delay_us) { \
+			udelay(__delay_us); \
+			if (__timeout_us) \
+				__left_ns -= __delay_ns; \
+		} \
+		cpu_relax(); \
+		if (__timeout_us) \
+			__left_ns--; \
+	} \
+	(cond) ? 0 : -ETIMEDOUT; \
+})
+
 /**
  * read_poll_timeout - Periodically poll an address until a condition is
- *			met or a timeout occurs
+ *                     met or a timeout occurs
  * @op: accessor function (takes @args as its arguments)
  * @val: Variable to read the value into
  * @cond: Break condition (usually involving @val)
@@ -34,33 +129,12 @@
  * be called from atomic context if sleep_us or timeout_us are used.
  */
 #define read_poll_timeout(op, val, cond, sleep_us, timeout_us, \
-				sleep_before_read, args...) \
-({ \
-	u64 __timeout_us = (timeout_us); \
-	unsigned long __sleep_us = (sleep_us); \
-	ktime_t __timeout = ktime_add_us(ktime_get(), __timeout_us); \
-	might_sleep_if((__sleep_us) != 0); \
-	if (sleep_before_read && __sleep_us) \
-		usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
-	for (;;) { \
-		(val) = op(args); \
-		if (cond) \
-			break; \
-		if (__timeout_us && \
-		    ktime_compare(ktime_get(), __timeout) > 0) { \
-			(val) = op(args); \
-			break; \
-		} \
-		if (__sleep_us) \
-			usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
-		cpu_relax(); \
-	} \
-	(cond) ? 0 : -ETIMEDOUT; \
-})
+			  sleep_before_read, args...) \
+	poll_timeout_us((val) = op(args), cond, sleep_us, timeout_us, sleep_before_read)
 
 /**
  * read_poll_timeout_atomic - Periodically poll an address until a condition is
- * 				met or a timeout occurs
+ *                            met or a timeout occurs
  * @op: accessor function (takes @args as its arguments)
  * @val: Variable to read the value into
  * @cond: Break condition (usually involving @val)
@@ -81,37 +155,9 @@
  * Returns: 0 on success and -ETIMEDOUT upon a timeout. In either
  * case, the last read value at @args is stored in @val.
  */
-#define read_poll_timeout_atomic(op, val, cond, delay_us, timeout_us, \
-					delay_before_read, args...) \
-({ \
-	u64 __timeout_us = (timeout_us); \
-	s64 __left_ns = __timeout_us * NSEC_PER_USEC; \
-	unsigned long __delay_us = (delay_us); \
-	u64 __delay_ns = __delay_us * NSEC_PER_USEC; \
-	if (delay_before_read && __delay_us) { \
-		udelay(__delay_us); \
-		if (__timeout_us) \
-			__left_ns -= __delay_ns; \
-	} \
-	for (;;) { \
-		(val) = op(args); \
-		if (cond) \
-			break; \
-		if (__timeout_us && __left_ns < 0) { \
-			(val) = op(args); \
-			break; \
-		} \
-		if (__delay_us) { \
-			udelay(__delay_us); \
-			if (__timeout_us) \
-				__left_ns -= __delay_ns; \
-		} \
-		cpu_relax(); \
-		if (__timeout_us) \
-			__left_ns--; \
-	} \
-	(cond) ? 0 : -ETIMEDOUT; \
-})
+#define read_poll_timeout_atomic(op, val, cond, sleep_us, timeout_us, \
+				 sleep_before_read, args...) \
+	poll_timeout_us_atomic((val) = op(args), cond, sleep_us, timeout_us, sleep_before_read)
 
 /**
  * readx_poll_timeout - Periodically poll an address until a condition is met or a timeout occurs

From 563e5eca4ea3b6e1901cbc7cd6dc42731a8d2999 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 26 Aug 2025 15:18:58 +0300
Subject: [PATCH 0712/1455] iopoll: Avoid evaluating 'cond' twice in
 poll_timeout_us()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently poll_timeout_us() evaluates 'cond' twice at the end
of the success case. This not desirable in case 'cond' itself
is expensive.

Avoid the double evaluation by tracking the return value in
a variable. Need to use a triple undescore '___ret' name to
avoid a conflict with an existing double undescore '__ret'
variable in the regmap code.

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: David Laight <david.laight.linux@gmail.com>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Matt Wagantall <mattw@codeaurora.org>
Cc: Dejin Zheng <zhengdejin5@gmail.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: intel-xe@lists.freedesktop.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Simona Vetter <simona.vetter@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250826121859.15497-2-ville.syrjala@linux.intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 include/linux/iopoll.h | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/include/linux/iopoll.h b/include/linux/iopoll.h
index 440aca5b4b59..d8c801ad68fa 100644
--- a/include/linux/iopoll.h
+++ b/include/linux/iopoll.h
@@ -36,23 +36,30 @@
 	u64 __timeout_us = (timeout_us); \
 	unsigned long __sleep_us = (sleep_us); \
 	ktime_t __timeout = ktime_add_us(ktime_get(), __timeout_us); \
+	int ___ret; \
 	might_sleep_if((__sleep_us) != 0); \
 	if ((sleep_before_op) && __sleep_us) \
 		usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
 	for (;;) { \
 		op; \
-		if (cond) \
+		if (cond) { \
+			___ret = 0; \
 			break; \
+		} \
 		if (__timeout_us && \
 		    ktime_compare(ktime_get(), __timeout) > 0) { \
 			op; \
+			if (cond) \
+				___ret = 0; \
+			else \
+				___ret = -ETIMEDOUT; \
 			break; \
 		} \
 		if (__sleep_us) \
 			usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
 		cpu_relax(); \
 	} \
-	(cond) ? 0 : -ETIMEDOUT; \
+	___ret; \
 })
 
 /**
@@ -83,6 +90,7 @@
 	s64 __left_ns = __timeout_us * NSEC_PER_USEC; \
 	unsigned long __delay_us = (delay_us); \
 	u64 __delay_ns = __delay_us * NSEC_PER_USEC; \
+	int ___ret; \
 	if ((delay_before_op) && __delay_us) { \
 		udelay(__delay_us); \
 		if (__timeout_us) \
@@ -90,10 +98,16 @@
 	} \
 	for (;;) { \
 		op; \
-		if (cond) \
+		if (cond) { \
+			___ret = 0; \
 			break; \
+		} \
 		if (__timeout_us && __left_ns < 0) { \
 			op; \
+			if (cond) \
+				___ret = 0; \
+			else \
+				___ret = -ETIMEDOUT; \
 			break; \
 		} \
 		if (__delay_us) { \
@@ -105,7 +119,7 @@
 		if (__timeout_us) \
 			__left_ns--; \
 	} \
-	(cond) ? 0 : -ETIMEDOUT; \
+	___ret; \
 })
 
 /**

From 3b6f62b6b5777f56e3fedc45041c2f61645b5204 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 26 Aug 2025 15:18:59 +0300
Subject: [PATCH 0713/1455] iopoll: Reorder the timeout handling in
 poll_timeout_us()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently poll_timeout_us() evaluates 'op' and 'cond' twice
within the loop, once at the start, and a second time after
the timeout check. While it's probably not a big deal to do
it twice almost back to back, it does make the macro a bit messy.

Simplify the implementation to evaluate the timeout at the
very start, then follow up with 'op'/'cond', and finally
check if the timeout did in fact happen or not.

For good measure throw in a compiler barrier between the timeout
and 'op'/'cond' evaluations to make sure the compiler can't reoder
the operations (which could cause false positive timeouts).
The similar i915 __wait_for() macro already has the barrier, though
there it is between the 'op' and 'cond' evaluations, which seems
like it could still allow 'op' and the timeout evaluations to get
reordered incorrectly. I suppose the ktime_get() might itself act
as a sufficient barrier here, but better safe than sorry I guess.

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Dibin Moolakadan Subrahmanian <dibin.moolakadan.subrahmanian@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: David Laight <david.laight.linux@gmail.com>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Matt Wagantall <mattw@codeaurora.org>
Cc: Dejin Zheng <zhengdejin5@gmail.com>
Cc: intel-gfx@lists.freedesktop.org
Cc: intel-xe@lists.freedesktop.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Acked-by: Simona Vetter <simona.vetter@ffwll.ch>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250826121859.15497-3-ville.syrjala@linux.intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 include/linux/iopoll.h | 24 +++++++++++-------------
 1 file changed, 11 insertions(+), 13 deletions(-)

diff --git a/include/linux/iopoll.h b/include/linux/iopoll.h
index d8c801ad68fa..bdd2e0652bc3 100644
--- a/include/linux/iopoll.h
+++ b/include/linux/iopoll.h
@@ -41,18 +41,17 @@
 	if ((sleep_before_op) && __sleep_us) \
 		usleep_range((__sleep_us >> 2) + 1, __sleep_us); \
 	for (;;) { \
+		bool __expired = __timeout_us && \
+			ktime_compare(ktime_get(), __timeout) > 0; \
+		/* guarantee 'op' and 'cond' are evaluated after timeout expired */ \
+		barrier(); \
 		op; \
 		if (cond) { \
 			___ret = 0; \
 			break; \
 		} \
-		if (__timeout_us && \
-		    ktime_compare(ktime_get(), __timeout) > 0) { \
-			op; \
-			if (cond) \
-				___ret = 0; \
-			else \
-				___ret = -ETIMEDOUT; \
+		if (__expired) { \
+			___ret = -ETIMEDOUT; \
 			break; \
 		} \
 		if (__sleep_us) \
@@ -97,17 +96,16 @@
 			__left_ns -= __delay_ns; \
 	} \
 	for (;;) { \
+		bool __expired = __timeout_us && __left_ns < 0; \
+		/* guarantee 'op' and 'cond' are evaluated after timeout expired */ \
+		barrier(); \
 		op; \
 		if (cond) { \
 			___ret = 0; \
 			break; \
 		} \
-		if (__timeout_us && __left_ns < 0) { \
-			op; \
-			if (cond) \
-				___ret = 0; \
-			else \
-				___ret = -ETIMEDOUT; \
+		if (__expired) { \
+			___ret = -ETIMEDOUT; \
 			break; \
 		} \
 		if (__delay_us) { \

From bddf32f16534be69cff99180d16a1c129bfcc226 Mon Sep 17 00:00:00 2001
From: Chen Ni <nichen@iscas.ac.cn>
Date: Thu, 28 Aug 2025 16:57:08 +0800
Subject: [PATCH 0714/1455] drm/vesadrm: Remove unneeded semicolon

Remove unnecessary semicolons reported by Coccinelle/coccicheck and the
semantic patch at scripts/coccinelle/misc/semicolon.cocci.

Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250828085708.3383760-1-nichen@iscas.ac.cn
---
 drivers/gpu/drm/sysfb/vesadrm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
index 1170e71e4208..16a4b52d45c6 100644
--- a/drivers/gpu/drm/sysfb/vesadrm.c
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -289,7 +289,7 @@ static int vesadrm_primary_plane_helper_atomic_check(struct drm_plane *plane,
 			break;
 		}
 		break;
-	};
+	}
 
 	return 0;
 }

From e7fa80e2932c68c17e7003fdfd01addc123567f7 Mon Sep 17 00:00:00 2001
From: Alice Ryhl <aliceryhl@google.com>
Date: Wed, 27 Aug 2025 13:38:37 +0000
Subject: [PATCH 0715/1455] drm_gem: add mutex to drm_gem_object.gpuva

There are two main ways that GPUVM might be used:

* staged mode, where VM_BIND ioctls update the GPUVM immediately so that
  the GPUVM reflects the state of the VM *including* staged changes that
  are not yet applied to the GPU's virtual address space.
* immediate mode, where the GPUVM state is updated during run_job(),
  i.e., in the DMA fence signalling critical path, to ensure that the
  GPUVM and the GPU's virtual address space has the same state at all
  times.

Currently, only Panthor uses GPUVM in immediate mode, but the Rust
drivers Tyr and Nova will also use GPUVM in immediate mode, so it is
worth to support both staged and immediate mode well in GPUVM. To use
immediate mode, the GEMs gpuva list must be modified during the fence
signalling path, which means that it must be protected by a lock that is
fence signalling safe.

For this reason, a mutex is added to struct drm_gem_object that is
intended to achieve this purpose. Adding it directly in the GEM object
both makes it easier to use GPUVM in immediate mode, but also makes it
possible to take the gpuva lock from core drm code.

As a follow-up, another change that should probably be made to support
immediate mode is a mechanism to postpone cleanup of vm_bo objects, as
dropping a vm_bo object in the fence signalling path is problematic for
two reasons:

* When using DRM_GPUVM_RESV_PROTECTED, you cannot remove the vm_bo from
  the extobj/evicted lists during the fence signalling path.
* Dropping a vm_bo could lead to the GEM object getting destroyed.
  The requirement that GEM object cleanup is fence signalling safe is
  dubious and likely to be violated in practice.

Panthor already has its own custom implementation of postponing vm_bo
cleanup.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250827-gpuva-mutex-in-gem-v3-1-bd89f5a82c0d@google.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/drm_gem.c |  2 ++
 include/drm/drm_gem.h     | 24 ++++++++++++++++++------
 2 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
index 4a89b6acb6af..8d25cc65707d 100644
--- a/drivers/gpu/drm/drm_gem.c
+++ b/drivers/gpu/drm/drm_gem.c
@@ -187,6 +187,7 @@ void drm_gem_private_object_init(struct drm_device *dev,
 	kref_init(&obj->refcount);
 	obj->handle_count = 0;
 	obj->size = size;
+	mutex_init(&obj->gpuva.lock);
 	dma_resv_init(&obj->_resv);
 	if (!obj->resv)
 		obj->resv = &obj->_resv;
@@ -210,6 +211,7 @@ void drm_gem_private_object_fini(struct drm_gem_object *obj)
 	WARN_ON(obj->dma_buf);
 
 	dma_resv_fini(&obj->_resv);
+	mutex_destroy(&obj->gpuva.lock);
 }
 EXPORT_SYMBOL(drm_gem_private_object_fini);
 
diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
index d3a7b43e2c63..a995c0c1b63c 100644
--- a/include/drm/drm_gem.h
+++ b/include/drm/drm_gem.h
@@ -398,16 +398,28 @@ struct drm_gem_object {
 	struct dma_resv _resv;
 
 	/**
-	 * @gpuva:
-	 *
-	 * Provides the list of GPU VAs attached to this GEM object.
-	 *
-	 * Drivers should lock list accesses with the GEMs &dma_resv lock
-	 * (&drm_gem_object.resv) or a custom lock if one is provided.
+	 * @gpuva: Fields used by GPUVM to manage mappings pointing to this GEM object.
 	 */
 	struct {
+		/**
+		 * @gpuva.list: list of GPUVM mappings attached to this GEM object.
+		 *
+		 * Drivers should lock list accesses with either the GEMs
+		 * &dma_resv lock (&drm_gem_object.resv) or the
+		 * &drm_gem_object.gpuva.lock mutex.
+		 */
 		struct list_head list;
 
+		/**
+		 * @gpuva.lock: lock protecting access to &drm_gem_object.gpuva.list
+		 * when the resv lock can't be used.
+		 *
+		 * Should only be used when the VM is being modified in a fence
+		 * signalling path, otherwise you should use &drm_gem_object.resv to
+		 * protect accesses to &drm_gem_object.gpuva.list.
+		 */
+		struct mutex lock;
+
 #ifdef CONFIG_LOCKDEP
 		struct lockdep_map *lock_dep_map;
 #endif

From 69013f52b4b6782db6941f83b6ff1be123bfa9e5 Mon Sep 17 00:00:00 2001
From: Alice Ryhl <aliceryhl@google.com>
Date: Wed, 27 Aug 2025 13:38:38 +0000
Subject: [PATCH 0716/1455] panthor: use drm_gem_object.gpuva.lock instead of
 gpuva_list_lock

Now that drm_gem_object has a dedicated mutex for the gpuva list that is
intended to be used in cases that must be fence signalling safe, use it
in Panthor.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250827-gpuva-mutex-in-gem-v3-2-bd89f5a82c0d@google.com
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/panthor/panthor_gem.c |  4 +---
 drivers/gpu/drm/panthor/panthor_gem.h | 12 ------------
 drivers/gpu/drm/panthor/panthor_mmu.c | 16 ++++++++--------
 3 files changed, 9 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_gem.c b/drivers/gpu/drm/panthor/panthor_gem.c
index a123bc740ba1..c654a3377903 100644
--- a/drivers/gpu/drm/panthor/panthor_gem.c
+++ b/drivers/gpu/drm/panthor/panthor_gem.c
@@ -74,7 +74,6 @@ static void panthor_gem_free_object(struct drm_gem_object *obj)
 	mutex_destroy(&bo->label.lock);
 
 	drm_gem_free_mmap_offset(&bo->base.base);
-	mutex_destroy(&bo->gpuva_list_lock);
 	drm_gem_shmem_free(&bo->base);
 	drm_gem_object_put(vm_root_gem);
 }
@@ -246,8 +245,7 @@ struct drm_gem_object *panthor_gem_create_object(struct drm_device *ddev, size_t
 
 	obj->base.base.funcs = &panthor_gem_funcs;
 	obj->base.map_wc = !ptdev->coherent;
-	mutex_init(&obj->gpuva_list_lock);
-	drm_gem_gpuva_set_lock(&obj->base.base, &obj->gpuva_list_lock);
+	drm_gem_gpuva_set_lock(&obj->base.base, &obj->base.base.gpuva.lock);
 	mutex_init(&obj->label.lock);
 
 	panthor_gem_debugfs_bo_init(obj);
diff --git a/drivers/gpu/drm/panthor/panthor_gem.h b/drivers/gpu/drm/panthor/panthor_gem.h
index 8fc7215e9b90..80c6e24112d0 100644
--- a/drivers/gpu/drm/panthor/panthor_gem.h
+++ b/drivers/gpu/drm/panthor/panthor_gem.h
@@ -79,18 +79,6 @@ struct panthor_gem_object {
 	 */
 	struct drm_gem_object *exclusive_vm_root_gem;
 
-	/**
-	 * @gpuva_list_lock: Custom GPUVA lock.
-	 *
-	 * Used to protect insertion of drm_gpuva elements to the
-	 * drm_gem_object.gpuva.list list.
-	 *
-	 * We can't use the GEM resv for that, because drm_gpuva_link() is
-	 * called in a dma-signaling path, where we're not allowed to take
-	 * resv locks.
-	 */
-	struct mutex gpuva_list_lock;
-
 	/** @flags: Combination of drm_panthor_bo_flags flags. */
 	u32 flags;
 
diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index 2003b91a8409..2881942ab511 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -1107,9 +1107,9 @@ static void panthor_vm_bo_put(struct drm_gpuvm_bo *vm_bo)
 	 * GEM vm_bo list.
 	 */
 	dma_resv_lock(drm_gpuvm_resv(vm), NULL);
-	mutex_lock(&bo->gpuva_list_lock);
+	mutex_lock(&bo->base.base.gpuva.lock);
 	unpin = drm_gpuvm_bo_put(vm_bo);
-	mutex_unlock(&bo->gpuva_list_lock);
+	mutex_unlock(&bo->base.base.gpuva.lock);
 	dma_resv_unlock(drm_gpuvm_resv(vm));
 
 	/* If the vm_bo object was destroyed, release the pin reference that
@@ -1282,9 +1282,9 @@ static int panthor_vm_prepare_map_op_ctx(struct panthor_vm_op_ctx *op_ctx,
 	 * calling this function.
 	 */
 	dma_resv_lock(panthor_vm_resv(vm), NULL);
-	mutex_lock(&bo->gpuva_list_lock);
+	mutex_lock(&bo->base.base.gpuva.lock);
 	op_ctx->map.vm_bo = drm_gpuvm_bo_obtain_prealloc(preallocated_vm_bo);
-	mutex_unlock(&bo->gpuva_list_lock);
+	mutex_unlock(&bo->base.base.gpuva.lock);
 	dma_resv_unlock(panthor_vm_resv(vm));
 
 	/* If the a vm_bo for this <VM,BO> combination exists, it already
@@ -2036,10 +2036,10 @@ static void panthor_vma_link(struct panthor_vm *vm,
 {
 	struct panthor_gem_object *bo = to_panthor_bo(vma->base.gem.obj);
 
-	mutex_lock(&bo->gpuva_list_lock);
+	mutex_lock(&bo->base.base.gpuva.lock);
 	drm_gpuva_link(&vma->base, vm_bo);
 	drm_WARN_ON(&vm->ptdev->base, drm_gpuvm_bo_put(vm_bo));
-	mutex_unlock(&bo->gpuva_list_lock);
+	mutex_unlock(&bo->base.base.gpuva.lock);
 }
 
 static void panthor_vma_unlink(struct panthor_vm *vm,
@@ -2048,9 +2048,9 @@ static void panthor_vma_unlink(struct panthor_vm *vm,
 	struct panthor_gem_object *bo = to_panthor_bo(vma->base.gem.obj);
 	struct drm_gpuvm_bo *vm_bo = drm_gpuvm_bo_get(vma->base.vm_bo);
 
-	mutex_lock(&bo->gpuva_list_lock);
+	mutex_lock(&bo->base.base.gpuva.lock);
 	drm_gpuva_unlink(&vma->base);
-	mutex_unlock(&bo->gpuva_list_lock);
+	mutex_unlock(&bo->base.base.gpuva.lock);
 
 	/* drm_gpuva_unlink() release the vm_bo, but we manually retained it
 	 * when entering this function, so we can implement deferred VMA

From 3c8d31b8937a7ee6e5de74f0274810b8705d77ea Mon Sep 17 00:00:00 2001
From: Alice Ryhl <aliceryhl@google.com>
Date: Wed, 27 Aug 2025 13:38:39 +0000
Subject: [PATCH 0717/1455] gpuvm: remove gem.gpuva.lock_dep_map

Since all users of gem.gpuva.lock_dep_map now rely on the mutex directly
in gpuva, we may remove it. Whether the mutex is used is now tracked by
a flag in gpuvm rather than by whether lock_dep_map is null.

Note that a GEM object may not be pushed to multiple gpuvms that
disagree on the value of this new flag. But that's okay because a single
driver should use the same locking scheme everywhere, and a GEM object
is driver specific (when a GEM is exported with prime, a new GEM object
instance is created from the backing dma-buf).

The flag is present even with CONFIG_LOCKDEP=n because the intent is
that the flag will also cause vm_bo cleanup to become deferred. However,
that will happen in a follow-up patch.

Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250827-gpuva-mutex-in-gem-v3-3-bd89f5a82c0d@google.com
[ Use lockdep_is_held() instead of lock_is_held(). - Danilo ]
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/drm_gpuvm.c           | 30 +++++++++-----------
 drivers/gpu/drm/panthor/panthor_gem.c |  1 -
 drivers/gpu/drm/panthor/panthor_mmu.c |  5 ++--
 include/drm/drm_gem.h                 | 41 ++++++++++-----------------
 include/drm/drm_gpuvm.h               | 30 ++++++++++++++++++--
 5 files changed, 59 insertions(+), 48 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpuvm.c b/drivers/gpu/drm/drm_gpuvm.c
index d6bea8a4fffd..78a1a4f09509 100644
--- a/drivers/gpu/drm/drm_gpuvm.c
+++ b/drivers/gpu/drm/drm_gpuvm.c
@@ -497,8 +497,7 @@
  * DRM GPUVM also does not take care of the locking of the backing
  * &drm_gem_object buffers GPU VA lists and &drm_gpuvm_bo abstractions by
  * itself; drivers are responsible to enforce mutual exclusion using either the
- * GEMs dma_resv lock or alternatively a driver specific external lock. For the
- * latter see also drm_gem_gpuva_set_lock().
+ * GEMs dma_resv lock or the GEMs gpuva.lock mutex.
  *
  * However, DRM GPUVM contains lockdep checks to ensure callers of its API hold
  * the corresponding lock whenever the &drm_gem_objects GPU VA list is accessed
@@ -1582,7 +1581,7 @@ drm_gpuvm_bo_destroy(struct kref *kref)
 	drm_gpuvm_bo_list_del(vm_bo, extobj, lock);
 	drm_gpuvm_bo_list_del(vm_bo, evict, lock);
 
-	drm_gem_gpuva_assert_lock_held(obj);
+	drm_gem_gpuva_assert_lock_held(gpuvm, obj);
 	list_del(&vm_bo->list.entry.gem);
 
 	if (ops && ops->vm_bo_free)
@@ -1603,7 +1602,8 @@ drm_gpuvm_bo_destroy(struct kref *kref)
  * If the reference count drops to zero, the &gpuvm_bo is destroyed, which
  * includes removing it from the GEMs gpuva list. Hence, if a call to this
  * function can potentially let the reference count drop to zero the caller must
- * hold the dma-resv or driver specific GEM gpuva lock.
+ * hold the lock that the GEM uses for its gpuva list (either the GEM's
+ * dma-resv or gpuva.lock mutex).
  *
  * This function may only be called from non-atomic context.
  *
@@ -1627,7 +1627,7 @@ __drm_gpuvm_bo_find(struct drm_gpuvm *gpuvm,
 {
 	struct drm_gpuvm_bo *vm_bo;
 
-	drm_gem_gpuva_assert_lock_held(obj);
+	drm_gem_gpuva_assert_lock_held(gpuvm, obj);
 	drm_gem_for_each_gpuvm_bo(vm_bo, obj)
 		if (vm_bo->vm == gpuvm)
 			return vm_bo;
@@ -1686,7 +1686,7 @@ drm_gpuvm_bo_obtain(struct drm_gpuvm *gpuvm,
 	if (!vm_bo)
 		return ERR_PTR(-ENOMEM);
 
-	drm_gem_gpuva_assert_lock_held(obj);
+	drm_gem_gpuva_assert_lock_held(gpuvm, obj);
 	list_add_tail(&vm_bo->list.entry.gem, &obj->gpuva.list);
 
 	return vm_bo;
@@ -1722,7 +1722,7 @@ drm_gpuvm_bo_obtain_prealloc(struct drm_gpuvm_bo *__vm_bo)
 		return vm_bo;
 	}
 
-	drm_gem_gpuva_assert_lock_held(obj);
+	drm_gem_gpuva_assert_lock_held(gpuvm, obj);
 	list_add_tail(&__vm_bo->list.entry.gem, &obj->gpuva.list);
 
 	return __vm_bo;
@@ -1894,8 +1894,7 @@ EXPORT_SYMBOL_GPL(drm_gpuva_remove);
  * reference of the latter is taken.
  *
  * This function expects the caller to protect the GEM's GPUVA list against
- * concurrent access using either the GEMs dma_resv lock or a driver specific
- * lock set through drm_gem_gpuva_set_lock().
+ * concurrent access using either the GEM's dma-resv or gpuva.lock mutex.
  */
 void
 drm_gpuva_link(struct drm_gpuva *va, struct drm_gpuvm_bo *vm_bo)
@@ -1910,7 +1909,7 @@ drm_gpuva_link(struct drm_gpuva *va, struct drm_gpuvm_bo *vm_bo)
 
 	va->vm_bo = drm_gpuvm_bo_get(vm_bo);
 
-	drm_gem_gpuva_assert_lock_held(obj);
+	drm_gem_gpuva_assert_lock_held(gpuvm, obj);
 	list_add_tail(&va->gem.entry, &vm_bo->list.gpuva);
 }
 EXPORT_SYMBOL_GPL(drm_gpuva_link);
@@ -1930,8 +1929,7 @@ EXPORT_SYMBOL_GPL(drm_gpuva_link);
  * the latter is dropped.
  *
  * This function expects the caller to protect the GEM's GPUVA list against
- * concurrent access using either the GEMs dma_resv lock or a driver specific
- * lock set through drm_gem_gpuva_set_lock().
+ * concurrent access using either the GEM's dma-resv or gpuva.lock mutex.
  */
 void
 drm_gpuva_unlink(struct drm_gpuva *va)
@@ -1942,7 +1940,7 @@ drm_gpuva_unlink(struct drm_gpuva *va)
 	if (unlikely(!obj))
 		return;
 
-	drm_gem_gpuva_assert_lock_held(obj);
+	drm_gem_gpuva_assert_lock_held(va->vm, obj);
 	list_del_init(&va->gem.entry);
 
 	va->vm_bo = NULL;
@@ -2943,8 +2941,8 @@ EXPORT_SYMBOL_GPL(drm_gpuvm_prefetch_ops_create);
  * After the caller finished processing the returned &drm_gpuva_ops, they must
  * be freed with &drm_gpuva_ops_free.
  *
- * It is the callers responsibility to protect the GEMs GPUVA list against
- * concurrent access using the GEMs dma_resv lock.
+ * This function expects the caller to protect the GEM's GPUVA list against
+ * concurrent access using either the GEM's dma-resv or gpuva.lock mutex.
  *
  * Returns: a pointer to the &drm_gpuva_ops on success, an ERR_PTR on failure
  */
@@ -2956,7 +2954,7 @@ drm_gpuvm_bo_unmap_ops_create(struct drm_gpuvm_bo *vm_bo)
 	struct drm_gpuva *va;
 	int ret;
 
-	drm_gem_gpuva_assert_lock_held(vm_bo->obj);
+	drm_gem_gpuva_assert_lock_held(vm_bo->vm, vm_bo->obj);
 
 	ops = kzalloc(sizeof(*ops), GFP_KERNEL);
 	if (!ops)
diff --git a/drivers/gpu/drm/panthor/panthor_gem.c b/drivers/gpu/drm/panthor/panthor_gem.c
index c654a3377903..156c7a0b62a2 100644
--- a/drivers/gpu/drm/panthor/panthor_gem.c
+++ b/drivers/gpu/drm/panthor/panthor_gem.c
@@ -245,7 +245,6 @@ struct drm_gem_object *panthor_gem_create_object(struct drm_device *ddev, size_t
 
 	obj->base.base.funcs = &panthor_gem_funcs;
 	obj->base.map_wc = !ptdev->coherent;
-	drm_gem_gpuva_set_lock(&obj->base.base, &obj->base.base.gpuva.lock);
 	mutex_init(&obj->label.lock);
 
 	panthor_gem_debugfs_bo_init(obj);
diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index 2881942ab511..ee9e94448b76 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -2420,8 +2420,9 @@ panthor_vm_create(struct panthor_device *ptdev, bool for_mcu,
 	 * to be handled the same way user VMAs are.
 	 */
 	drm_gpuvm_init(&vm->base, for_mcu ? "panthor-MCU-VM" : "panthor-GPU-VM",
-		       DRM_GPUVM_RESV_PROTECTED, &ptdev->base, dummy_gem,
-		       min_va, va_range, 0, 0, &panthor_gpuvm_ops);
+		       DRM_GPUVM_RESV_PROTECTED | DRM_GPUVM_IMMEDIATE_MODE,
+		       &ptdev->base, dummy_gem, min_va, va_range, 0, 0,
+		       &panthor_gpuvm_ops);
 	drm_gem_object_put(dummy_gem);
 	return vm;
 
diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h
index a995c0c1b63c..8d48d2af2649 100644
--- a/include/drm/drm_gem.h
+++ b/include/drm/drm_gem.h
@@ -399,6 +399,12 @@ struct drm_gem_object {
 
 	/**
 	 * @gpuva: Fields used by GPUVM to manage mappings pointing to this GEM object.
+	 *
+	 * When DRM_GPUVM_IMMEDIATE_MODE is set, this list is protected by the
+	 * mutex. Otherwise, the list is protected by the GEMs &dma_resv lock.
+	 *
+	 * Note that all entries in this list must agree on whether
+	 * DRM_GPUVM_IMMEDIATE_MODE is set.
 	 */
 	struct {
 		/**
@@ -412,17 +418,14 @@ struct drm_gem_object {
 
 		/**
 		 * @gpuva.lock: lock protecting access to &drm_gem_object.gpuva.list
-		 * when the resv lock can't be used.
+		 * when DRM_GPUVM_IMMEDIATE_MODE is used.
 		 *
-		 * Should only be used when the VM is being modified in a fence
-		 * signalling path, otherwise you should use &drm_gem_object.resv to
-		 * protect accesses to &drm_gem_object.gpuva.list.
+		 * Only used when DRM_GPUVM_IMMEDIATE_MODE is set. It should be
+		 * safe to take this mutex during the fence signalling path, so
+		 * do not allocate memory while holding this lock. Otherwise,
+		 * the &dma_resv lock should be used.
 		 */
 		struct mutex lock;
-
-#ifdef CONFIG_LOCKDEP
-		struct lockdep_map *lock_dep_map;
-#endif
 	} gpuva;
 
 	/**
@@ -607,26 +610,12 @@ static inline bool drm_gem_is_imported(const struct drm_gem_object *obj)
 }
 
 #ifdef CONFIG_LOCKDEP
-/**
- * drm_gem_gpuva_set_lock() - Set the lock protecting accesses to the gpuva list.
- * @obj: the &drm_gem_object
- * @lock: the lock used to protect the gpuva list. The locking primitive
- * must contain a dep_map field.
- *
- * Call this if you're not proctecting access to the gpuva list with the
- * dma-resv lock, but with a custom lock.
- */
-#define drm_gem_gpuva_set_lock(obj, lock) \
-	if (!WARN((obj)->gpuva.lock_dep_map, \
-		  "GEM GPUVA lock should be set only once.")) \
-		(obj)->gpuva.lock_dep_map = &(lock)->dep_map
-#define drm_gem_gpuva_assert_lock_held(obj) \
-	lockdep_assert((obj)->gpuva.lock_dep_map ? \
-		       lock_is_held((obj)->gpuva.lock_dep_map) : \
+#define drm_gem_gpuva_assert_lock_held(gpuvm, obj) \
+	lockdep_assert(drm_gpuvm_immediate_mode(gpuvm) ? \
+		       lockdep_is_held(&(obj)->gpuva.lock) : \
 		       dma_resv_held((obj)->resv))
 #else
-#define drm_gem_gpuva_set_lock(obj, lock) do {} while (0)
-#define drm_gem_gpuva_assert_lock_held(obj) do {} while (0)
+#define drm_gem_gpuva_assert_lock_held(gpuvm, obj) do {} while (0)
 #endif
 
 /**
diff --git a/include/drm/drm_gpuvm.h b/include/drm/drm_gpuvm.h
index 4a22b9d848f7..727b8f336fad 100644
--- a/include/drm/drm_gpuvm.h
+++ b/include/drm/drm_gpuvm.h
@@ -196,10 +196,20 @@ enum drm_gpuvm_flags {
 	 */
 	DRM_GPUVM_RESV_PROTECTED = BIT(0),
 
+	/**
+	 * @DRM_GPUVM_IMMEDIATE_MODE: use the locking scheme for GEMs designed
+	 * for modifying the GPUVM during the fence signalling path
+	 *
+	 * When set, gpuva.lock is used to protect gpuva.list in all GEM
+	 * objects associated with this GPUVM. Otherwise, the GEMs dma-resv is
+	 * used.
+	 */
+	DRM_GPUVM_IMMEDIATE_MODE = BIT(1),
+
 	/**
 	 * @DRM_GPUVM_USERBITS: user defined bits
 	 */
-	DRM_GPUVM_USERBITS = BIT(1),
+	DRM_GPUVM_USERBITS = BIT(2),
 };
 
 /**
@@ -369,6 +379,19 @@ drm_gpuvm_resv_protected(struct drm_gpuvm *gpuvm)
 	return gpuvm->flags & DRM_GPUVM_RESV_PROTECTED;
 }
 
+/**
+ * drm_gpuvm_immediate_mode() - indicates whether &DRM_GPUVM_IMMEDIATE_MODE is
+ * set
+ * @gpuvm: the &drm_gpuvm
+ *
+ * Returns: true if &DRM_GPUVM_IMMEDIATE_MODE is set, false otherwise.
+ */
+static inline bool
+drm_gpuvm_immediate_mode(struct drm_gpuvm *gpuvm)
+{
+	return gpuvm->flags & DRM_GPUVM_IMMEDIATE_MODE;
+}
+
 /**
  * drm_gpuvm_resv() - returns the &drm_gpuvm's &dma_resv
  * @gpuvm__: the &drm_gpuvm
@@ -742,9 +765,10 @@ drm_gpuvm_bo_gem_evict(struct drm_gem_object *obj, bool evict)
 {
 	struct drm_gpuvm_bo *vm_bo;
 
-	drm_gem_gpuva_assert_lock_held(obj);
-	drm_gem_for_each_gpuvm_bo(vm_bo, obj)
+	drm_gem_for_each_gpuvm_bo(vm_bo, obj) {
+		drm_gem_gpuva_assert_lock_held(vm_bo->vm, obj);
 		drm_gpuvm_bo_evict(vm_bo, evict);
+	}
 }
 
 void drm_gpuvm_bo_extobj_add(struct drm_gpuvm_bo *vm_bo);

From 24f028b91ed3329b259874e28b54aa715b008153 Mon Sep 17 00:00:00 2001
From: Steven Price <steven.price@arm.com>
Date: Fri, 15 Aug 2025 14:42:24 +0100
Subject: [PATCH 0718/1455] drm/panthor: Simplify mmu_hw_do_operation_locked

The only callers to mmu_hw_do_operation_locked() pass an 'op' of either
AS_COMAND_FLUSH_MEM or AS_COMMAND_FLUSH_PT. This means the code paths
after that are dead. Removing those paths means the
mmu_hw_do_flush_on_gpu_ctrl() function might has well be inlined.

Simplify everything by having a switch statement for the type of 'op'
(warning if we get an unexpected value) and removing the dead cases.

Suggested-by: Daniel Stone <daniel@fooishbar.org>
Signed-off-by: Steven Price <steven.price@arm.com>
Reviewed-by: Karunika Choo <karunika.choo@arm.com>
Link: https://lore.kernel.org/r/20250815134226.57703-1-steven.price@arm.com
---
 drivers/gpu/drm/panthor/panthor_mmu.c | 57 ++++++++++++---------------
 1 file changed, 26 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index ee9e94448b76..bad720f23eb7 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -569,15 +569,37 @@ static void lock_region(struct panthor_device *ptdev, u32 as_nr,
 	write_cmd(ptdev, as_nr, AS_COMMAND_LOCK);
 }
 
-static int mmu_hw_do_flush_on_gpu_ctrl(struct panthor_device *ptdev, int as_nr,
-				       u32 op)
+static int mmu_hw_do_operation_locked(struct panthor_device *ptdev, int as_nr,
+				      u64 iova, u64 size, u32 op)
 {
 	const u32 l2_flush_op = CACHE_CLEAN | CACHE_INV;
-	u32 lsc_flush_op = 0;
+	u32 lsc_flush_op;
 	int ret;
 
-	if (op == AS_COMMAND_FLUSH_MEM)
+	lockdep_assert_held(&ptdev->mmu->as.slots_lock);
+
+	switch (op) {
+	case AS_COMMAND_FLUSH_MEM:
 		lsc_flush_op = CACHE_CLEAN | CACHE_INV;
+		break;
+	case AS_COMMAND_FLUSH_PT:
+		lsc_flush_op = 0;
+		break;
+	default:
+		drm_WARN(&ptdev->base, 1, "Unexpected AS_COMMAND: %d", op);
+		return -EINVAL;
+	}
+
+	if (as_nr < 0)
+		return 0;
+
+	/*
+	 * If the AS number is greater than zero, then we can be sure
+	 * the device is up and running, so we don't need to explicitly
+	 * power it up
+	 */
+
+	lock_region(ptdev, as_nr, iova, size);
 
 	ret = wait_ready(ptdev, as_nr);
 	if (ret)
@@ -598,33 +620,6 @@ static int mmu_hw_do_flush_on_gpu_ctrl(struct panthor_device *ptdev, int as_nr,
 	return wait_ready(ptdev, as_nr);
 }
 
-static int mmu_hw_do_operation_locked(struct panthor_device *ptdev, int as_nr,
-				      u64 iova, u64 size, u32 op)
-{
-	lockdep_assert_held(&ptdev->mmu->as.slots_lock);
-
-	if (as_nr < 0)
-		return 0;
-
-	/*
-	 * If the AS number is greater than zero, then we can be sure
-	 * the device is up and running, so we don't need to explicitly
-	 * power it up
-	 */
-
-	if (op != AS_COMMAND_UNLOCK)
-		lock_region(ptdev, as_nr, iova, size);
-
-	if (op == AS_COMMAND_FLUSH_MEM || op == AS_COMMAND_FLUSH_PT)
-		return mmu_hw_do_flush_on_gpu_ctrl(ptdev, as_nr, op);
-
-	/* Run the MMU operation */
-	write_cmd(ptdev, as_nr, op);
-
-	/* Wait for the flush to complete */
-	return wait_ready(ptdev, as_nr);
-}
-
 static int mmu_hw_do_operation(struct panthor_vm *vm,
 			       u64 iova, u64 size, u32 op)
 {

From 12267f831fa6fe679019d09c48089941908b7cf3 Mon Sep 17 00:00:00 2001
From: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
Date: Thu, 14 Aug 2025 16:59:50 +0200
Subject: [PATCH 0719/1455] drm/xe/tests: Make cross-device dma-buf BOs
 CPU-visible on small BAR
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Small-BAR systems (e.g., SR-IOV VFs in VMs) expose only a subset of
VRAM via PCI/BAR. Exporting a BO outside that window fails, and the
selftests also do CPU fill/verify.

Set XE_BO_FLAG_NEEDS_CPU_ACCESS for cross-device variants to force
CPU-mappable placement and keep tests reliable. Large-BAR/P2P setups
are unaffected.

Suggested-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Marcin Bernatowicz <marcin.bernatowicz@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250814145950.430231-1-marcin.bernatowicz@linux.intel.com
Signed-off-by: Michał Winiarski <michal.winiarski@intel.com>
---
 drivers/gpu/drm/xe/tests/xe_dma_buf.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
index cde9530bef8c..3c5ad8cc65a0 100644
--- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
@@ -204,7 +204,7 @@ static const struct dma_buf_attach_ops nop2p_attach_ops = {
 static const struct dma_buf_test_params test_params[] = {
 	{.mem_mask = XE_BO_FLAG_VRAM0,
 	 .attach_ops = &xe_dma_buf_attach_ops},
-	{.mem_mask = XE_BO_FLAG_VRAM0,
+	{.mem_mask = XE_BO_FLAG_VRAM0 | XE_BO_FLAG_NEEDS_CPU_ACCESS,
 	 .attach_ops = &xe_dma_buf_attach_ops,
 	 .force_different_devices = true},
 
@@ -236,7 +236,8 @@ static const struct dma_buf_test_params test_params[] = {
 
 	{.mem_mask = XE_BO_FLAG_SYSTEM | XE_BO_FLAG_VRAM0,
 	 .attach_ops = &xe_dma_buf_attach_ops},
-	{.mem_mask = XE_BO_FLAG_SYSTEM | XE_BO_FLAG_VRAM0,
+	{.mem_mask = XE_BO_FLAG_SYSTEM | XE_BO_FLAG_VRAM0 |
+		     XE_BO_FLAG_NEEDS_CPU_ACCESS,
 	 .attach_ops = &xe_dma_buf_attach_ops,
 	 .force_different_devices = true},
 

From 72031905cf2e36519a98b5becbe62dc10ebb4b97 Mon Sep 17 00:00:00 2001
From: "Christian S. Lima" <christiansantoslima21@gmail.com>
Date: Sun, 24 Aug 2025 18:31:33 -0300
Subject: [PATCH 0720/1455] rust: transmute: Add methods for FromBytes trait

The two methods added take a slice of bytes and return those bytes in
a specific type. These methods are useful when we need to transform
the stream of bytes into specific type.

Since the `is_aligned` method for pointer types has been stabilized in
`1.79` version and is being used in this patch, I'm enabling the
feature. In this case, using this method is useful to check the
alignment and avoid a giant boilerplate, such as `(foo.as_ptr() as
usize) % core::mem::align_of::<T>() == 0`.

Also add `#[allow(clippy::incompatible_msrv)]` where needed until the
MSRV is updated to `1.79`.

Suggested-by: Benno Lossin <benno.lossin@proton.me>
Link: https://github.com/Rust-for-Linux/linux/issues/1119
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Signed-off-by: Christian S. Lima <christiansantoslima21@gmail.com>
Link: https://lore.kernel.org/r/20250824213134.27079-1-christiansantoslima21@gmail.com
Acked-by: Miguel Ojeda <ojeda@kernel.org>
[acourbot@nvidia.com: minor rewording of commit messages and doccomments]
[acourbot@nvidia.com: revert slice implementation removal]
[acourbot@nvidia.com: move incompatible_msrv clippy allow closer to site of need]
[acourbot@nvidia.com: call the doctest method]
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 rust/kernel/lib.rs       |  1 +
 rust/kernel/transmute.rs | 69 +++++++++++++++++++++++++++++++++++++++-
 2 files changed, 69 insertions(+), 1 deletion(-)

diff --git a/rust/kernel/lib.rs b/rust/kernel/lib.rs
index ed53169e795c..c859a8984bae 100644
--- a/rust/kernel/lib.rs
+++ b/rust/kernel/lib.rs
@@ -18,6 +18,7 @@
 //
 // Stable since Rust 1.79.0.
 #![feature(inline_const)]
+#![feature(pointer_is_aligned)]
 //
 // Stable since Rust 1.81.0.
 #![feature(lint_reasons)]
diff --git a/rust/kernel/transmute.rs b/rust/kernel/transmute.rs
index 517a432ac693..5fc7d4ed76b4 100644
--- a/rust/kernel/transmute.rs
+++ b/rust/kernel/transmute.rs
@@ -2,6 +2,8 @@
 
 //! Traits for transmuting types.
 
+use core::mem::size_of;
+
 /// Types for which any bit pattern is valid.
 ///
 /// Not all types are valid for all values. For example, a `bool` must be either zero or one, so
@@ -9,10 +11,75 @@
 ///
 /// It's okay for the type to have padding, as initializing those bytes has no effect.
 ///
+/// # Examples
+///
+/// ```
+/// use kernel::transmute::FromBytes;
+///
+/// # fn test() -> Option<()> {
+/// let raw = [1, 2, 3, 4];
+///
+/// let result = u32::from_bytes(&raw)?;
+///
+/// #[cfg(target_endian = "little")]
+/// assert_eq!(*result, 0x4030201);
+///
+/// #[cfg(target_endian = "big")]
+/// assert_eq!(*result, 0x1020304);
+///
+/// # Some(()) }
+/// # test().ok_or(EINVAL)?;
+/// # Ok::<(), Error>(())
+/// ```
+///
 /// # Safety
 ///
 /// All bit-patterns must be valid for this type. This type must not have interior mutability.
-pub unsafe trait FromBytes {}
+pub unsafe trait FromBytes {
+    /// Converts a slice of bytes to a reference to `Self`.
+    ///
+    /// Succeeds if the reference is properly aligned, and the size of `bytes` is equal to that of
+    /// `T` and different from zero.
+    ///
+    /// Otherwise, returns [`None`].
+    fn from_bytes(bytes: &[u8]) -> Option<&Self>
+    where
+        Self: Sized,
+    {
+        let slice_ptr = bytes.as_ptr().cast::<Self>();
+        let size = size_of::<Self>();
+
+        #[allow(clippy::incompatible_msrv)]
+        if bytes.len() == size && slice_ptr.is_aligned() {
+            // SAFETY: Size and alignment were just checked.
+            unsafe { Some(&*slice_ptr) }
+        } else {
+            None
+        }
+    }
+
+    /// Converts a mutable slice of bytes to a reference to `Self`.
+    ///
+    /// Succeeds if the reference is properly aligned, and the size of `bytes` is equal to that of
+    /// `T` and different from zero.
+    ///
+    /// Otherwise, returns [`None`].
+    fn from_bytes_mut(bytes: &mut [u8]) -> Option<&mut Self>
+    where
+        Self: AsBytes + Sized,
+    {
+        let slice_ptr = bytes.as_mut_ptr().cast::<Self>();
+        let size = size_of::<Self>();
+
+        #[allow(clippy::incompatible_msrv)]
+        if bytes.len() == size && slice_ptr.is_aligned() {
+            // SAFETY: Size and alignment were just checked.
+            unsafe { Some(&mut *slice_ptr) }
+        } else {
+            None
+        }
+    }
+}
 
 macro_rules! impl_frombytes {
     ($($({$($generics:tt)*})? $t:ty, )*) => {

From 2b036b13c8c2cadf7a27036f911ec99742504fce Mon Sep 17 00:00:00 2001
From: Andy Yan <andy.yan@rock-chips.com>
Date: Fri, 22 Aug 2025 14:39:45 +0800
Subject: [PATCH 0721/1455] dt-bindings: display: rockchip: Add schema for
 RK3588 DPTX Controller

The Rockchip RK3588 SoC integrates the Synopsys DesignWare DPTX
controller. And this DPTX controller need share a USBDP PHY with
the USB 3.0 OTG controller during operation.

Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250822063959.692098-2-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../display/rockchip/rockchip,dw-dp.yaml      | 150 ++++++++++++++++++
 1 file changed, 150 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/rockchip/rockchip,dw-dp.yaml

diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-dp.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-dp.yaml
new file mode 100644
index 000000000000..a8a008717997
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,dw-dp.yaml
@@ -0,0 +1,150 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/rockchip/rockchip,dw-dp.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Rockchip DW DisplayPort Transmitter
+
+maintainers:
+  - Andy Yan <andy.yan@rock-chips.com>
+
+description: |
+  The Rockchip RK3588 SoC integrates the Synopsys DesignWare DPTX controller
+  which is compliant with the DisplayPort Specification Version 1.4 with the
+  following features:
+
+  * DisplayPort 1.4a
+  * Main Link: 1/2/4 lanes
+  * Main Link Support 1.62Gbps, 2.7Gbps, 5.4Gbps and 8.1Gbps
+  * AUX channel 1Mbps
+  * Single Stream Transport(SST)
+  * Multistream Transport (MST)
+  * Type-C support (alternate mode)
+  * HDCP 2.2, HDCP 1.3
+  * Supports up to 8/10 bits per color component
+  * Supports RBG, YCbCr4:4:4, YCbCr4:2:2, YCbCr4:2:0
+  * Pixel clock up to 594MHz
+  * I2S, SPDIF audio interface
+
+allOf:
+  - $ref: /schemas/sound/dai-common.yaml#
+
+properties:
+  compatible:
+    enum:
+      - rockchip,rk3588-dp
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  clocks:
+    items:
+      - description: Peripheral/APB bus clock
+      - description: DisplayPort AUX clock
+      - description: HDCP clock
+      - description: I2S interface clock
+      - description: SPDIF interfce clock
+
+  clock-names:
+    items:
+      - const: apb
+      - const: aux
+      - const: hdcp
+      - const: i2s
+      - const: spdif
+
+  phys:
+    maxItems: 1
+
+  ports:
+    $ref: /schemas/graph.yaml#/properties/ports
+
+    properties:
+      port@0:
+        $ref: /schemas/graph.yaml#/properties/port
+        description: Video port for RGB/YUV input.
+
+      port@1:
+        $ref: /schemas/graph.yaml#/properties/port
+        description: Video port for DP output.
+
+    required:
+      - port@0
+      - port@1
+
+  power-domains:
+    maxItems: 1
+
+  resets:
+    maxItems: 1
+
+  "#sound-dai-cells":
+    const: 0
+
+required:
+  - compatible
+  - reg
+  - clocks
+  - clock-names
+  - interrupts
+  - phys
+  - ports
+  - resets
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/rockchip,rk3588-cru.h>
+    #include <dt-bindings/phy/phy.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/interrupt-controller/irq.h>
+    #include <dt-bindings/power/rk3588-power.h>
+    #include <dt-bindings/reset/rockchip,rk3588-cru.h>
+
+    soc {
+      #address-cells = <2>;
+      #size-cells = <2>;
+
+      dp@fde50000 {
+        compatible = "rockchip,rk3588-dp";
+        reg = <0x0 0xfde50000 0x0 0x4000>;
+        interrupts = <GIC_SPI 161 IRQ_TYPE_LEVEL_HIGH 0>;
+        clocks = <&cru PCLK_DP0>, <&cru CLK_AUX16M_0>,
+        <&cru CLK_DP0>, <&cru MCLK_I2S4_8CH_TX>,
+        <&cru MCLK_SPDIF2_DP0>;
+        clock-names = "apb", "aux", "hdcp", "i2s", "spdif";
+        assigned-clocks = <&cru CLK_AUX16M_0>;
+        assigned-clock-rates = <16000000>;
+        resets = <&cru SRST_DP0>;
+        phys = <&usbdp_phy0 PHY_TYPE_DP>;
+        power-domains = <&power RK3588_PD_VO0>;
+        #sound-dai-cells = <0>;
+
+
+        ports {
+          #address-cells = <1>;
+          #size-cells = <0>;
+
+          port@0 {
+            reg = <0>;
+
+            dp0_in_vp2: endpoint {
+              remote-endpoint = <&vp2_out_dp0>;
+            };
+          };
+
+          port@1 {
+            reg = <1>;
+
+            dp0_out_con0: endpoint {
+              remote-endpoint = <&dp_con0_in>;
+            };
+          };
+        };
+      };
+    };

From 86eecc3a9c2e06462f6a273fcd24150b6da787de Mon Sep 17 00:00:00 2001
From: Andy Yan <andy.yan@rock-chips.com>
Date: Fri, 22 Aug 2025 14:39:46 +0800
Subject: [PATCH 0722/1455] drm/bridge: synopsys: Add DW DPTX Controller
 support library

The DW DP TX Controller is compliant with the DisplayPort Specification
Version 1.4 with the following features:

* DisplayPort 1.4a
* Main Link: 1/2/4 lanes
* Main Link Support 1.62Gbps, 2.7Gbps, 5.4Gbps and 8.1Gbps
* AUX channel 1Mbps
* Single Stream Transport(SST)
* Multistream Transport (MST)
* Type-C support (alternate mode)
* HDCP 2.2, HDCP 1.3
* Supports up to 8/10 bits per color component
* Supports RBG, YCbCr4:4:4, YCbCr4:2:2, YCbCr4:2:0
* Pixel clock up to 594MHz
* I2S, SPDIF audio interface

Add library with common helpers to make it can be shared with
other SoC.

Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Tested-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Link: https://lore.kernel.org/r/20250822063959.692098-3-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/synopsys/Kconfig  |    7 +
 drivers/gpu/drm/bridge/synopsys/Makefile |    1 +
 drivers/gpu/drm/bridge/synopsys/dw-dp.c  | 2095 ++++++++++++++++++++++
 include/drm/bridge/dw_dp.h               |   20 +
 4 files changed, 2123 insertions(+)
 create mode 100644 drivers/gpu/drm/bridge/synopsys/dw-dp.c
 create mode 100644 include/drm/bridge/dw_dp.h

diff --git a/drivers/gpu/drm/bridge/synopsys/Kconfig b/drivers/gpu/drm/bridge/synopsys/Kconfig
index f3ab2f985f8c..2c5e532410de 100644
--- a/drivers/gpu/drm/bridge/synopsys/Kconfig
+++ b/drivers/gpu/drm/bridge/synopsys/Kconfig
@@ -1,4 +1,11 @@
 # SPDX-License-Identifier: GPL-2.0-only
+config DRM_DW_DP
+	tristate
+	select DRM_DISPLAY_HELPER
+	select DRM_DISPLAY_DP_HELPER
+	select DRM_KMS_HELPER
+	select REGMAP_MMIO
+
 config DRM_DW_HDMI
 	tristate
 	select DRM_DISPLAY_HDMI_HELPER
diff --git a/drivers/gpu/drm/bridge/synopsys/Makefile b/drivers/gpu/drm/bridge/synopsys/Makefile
index 9dc376d220ad..4dada44029ac 100644
--- a/drivers/gpu/drm/bridge/synopsys/Makefile
+++ b/drivers/gpu/drm/bridge/synopsys/Makefile
@@ -1,4 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0-only
+obj-$(CONFIG_DRM_DW_DP) += dw-dp.o
 obj-$(CONFIG_DRM_DW_HDMI) += dw-hdmi.o
 obj-$(CONFIG_DRM_DW_HDMI_AHB_AUDIO) += dw-hdmi-ahb-audio.o
 obj-$(CONFIG_DRM_DW_HDMI_GP_AUDIO) += dw-hdmi-gp-audio.o
diff --git a/drivers/gpu/drm/bridge/synopsys/dw-dp.c b/drivers/gpu/drm/bridge/synopsys/dw-dp.c
new file mode 100644
index 000000000000..9bbfe8da3de0
--- /dev/null
+++ b/drivers/gpu/drm/bridge/synopsys/dw-dp.c
@@ -0,0 +1,2095 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Synopsys DesignWare Cores DisplayPort Transmitter Controller
+ *
+ * Copyright (c) 2025 Rockchip Electronics Co., Ltd.
+ *
+ * Author: Andy Yan <andy.yan@rock-chips.com>
+ */
+#include <linux/bitfield.h>
+#include <linux/clk.h>
+#include <linux/iopoll.h>
+#include <linux/irq.h>
+#include <linux/media-bus-format.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/reset.h>
+#include <linux/phy/phy.h>
+#include <linux/unaligned.h>
+
+#include <drm/bridge/dw_dp.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/display/drm_dp_helper.h>
+#include <drm/drm_edid.h>
+#include <drm/drm_of.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_simple_kms_helper.h>
+
+#define DW_DP_VERSION_NUMBER			0x0000
+#define DW_DP_VERSION_TYPE			0x0004
+#define DW_DP_ID				0x0008
+
+#define DW_DP_CONFIG_REG1			0x0100
+#define DW_DP_CONFIG_REG2			0x0104
+#define DW_DP_CONFIG_REG3			0x0108
+
+#define DW_DP_CCTL				0x0200
+#define FORCE_HPD				BIT(4)
+#define DEFAULT_FAST_LINK_TRAIN_EN		BIT(2)
+#define ENHANCE_FRAMING_EN			BIT(1)
+#define SCRAMBLE_DIS				BIT(0)
+#define DW_DP_SOFT_RESET_CTRL			0x0204
+#define VIDEO_RESET				BIT(5)
+#define AUX_RESET				BIT(4)
+#define AUDIO_SAMPLER_RESET			BIT(3)
+#define HDCP_MODULE_RESET			BIT(2)
+#define PHY_SOFT_RESET				BIT(1)
+#define CONTROLLER_RESET			BIT(0)
+
+#define DW_DP_VSAMPLE_CTRL			0x0300
+#define PIXEL_MODE_SELECT			GENMASK(22, 21)
+#define VIDEO_MAPPING				GENMASK(20, 16)
+#define VIDEO_STREAM_ENABLE			BIT(5)
+
+#define DW_DP_VSAMPLE_STUFF_CTRL1		0x0304
+
+#define DW_DP_VSAMPLE_STUFF_CTRL2		0x0308
+
+#define DW_DP_VINPUT_POLARITY_CTRL		0x030c
+#define DE_IN_POLARITY				BIT(2)
+#define HSYNC_IN_POLARITY			BIT(1)
+#define VSYNC_IN_POLARITY			BIT(0)
+
+#define DW_DP_VIDEO_CONFIG1			0x0310
+#define HACTIVE					GENMASK(31, 16)
+#define HBLANK					GENMASK(15, 2)
+#define I_P					BIT(1)
+#define R_V_BLANK_IN_OSC			BIT(0)
+
+#define DW_DP_VIDEO_CONFIG2			0x0314
+#define VBLANK					GENMASK(31, 16)
+#define VACTIVE					GENMASK(15, 0)
+
+#define DW_DP_VIDEO_CONFIG3			0x0318
+#define H_SYNC_WIDTH				GENMASK(31, 16)
+#define H_FRONT_PORCH				GENMASK(15, 0)
+
+#define DW_DP_VIDEO_CONFIG4			0x031c
+#define V_SYNC_WIDTH				GENMASK(31, 16)
+#define V_FRONT_PORCH				GENMASK(15, 0)
+
+#define DW_DP_VIDEO_CONFIG5			0x0320
+#define INIT_THRESHOLD_HI			GENMASK(22, 21)
+#define AVERAGE_BYTES_PER_TU_FRAC		GENMASK(19, 16)
+#define INIT_THRESHOLD				GENMASK(13, 7)
+#define AVERAGE_BYTES_PER_TU			GENMASK(6, 0)
+
+#define DW_DP_VIDEO_MSA1			0x0324
+#define VSTART					GENMASK(31, 16)
+#define HSTART					GENMASK(15, 0)
+
+#define DW_DP_VIDEO_MSA2			0x0328
+#define MISC0					GENMASK(31, 24)
+
+#define DW_DP_VIDEO_MSA3			0x032c
+#define MISC1					GENMASK(31, 24)
+
+#define DW_DP_VIDEO_HBLANK_INTERVAL		0x0330
+#define HBLANK_INTERVAL_EN			BIT(16)
+#define HBLANK_INTERVAL				GENMASK(15, 0)
+
+#define DW_DP_AUD_CONFIG1			0x0400
+#define AUDIO_TIMESTAMP_VERSION_NUM		GENMASK(29, 24)
+#define AUDIO_PACKET_ID				GENMASK(23, 16)
+#define AUDIO_MUTE				BIT(15)
+#define NUM_CHANNELS				GENMASK(14, 12)
+#define HBR_MODE_ENABLE				BIT(10)
+#define AUDIO_DATA_WIDTH			GENMASK(9, 5)
+#define AUDIO_DATA_IN_EN			GENMASK(4, 1)
+#define AUDIO_INF_SELECT			BIT(0)
+
+#define DW_DP_SDP_VERTICAL_CTRL			0x0500
+#define EN_VERTICAL_SDP				BIT(2)
+#define EN_AUDIO_STREAM_SDP			BIT(1)
+#define EN_AUDIO_TIMESTAMP_SDP			BIT(0)
+#define DW_DP_SDP_HORIZONTAL_CTRL		0x0504
+#define EN_HORIZONTAL_SDP			BIT(2)
+#define DW_DP_SDP_STATUS_REGISTER		0x0508
+#define DW_DP_SDP_MANUAL_CTRL			0x050c
+#define DW_DP_SDP_STATUS_EN			0x0510
+
+#define DW_DP_SDP_REGISTER_BANK			0x0600
+#define SDP_REGS				GENMASK(31, 0)
+
+#define DW_DP_PHYIF_CTRL			0x0a00
+#define PHY_WIDTH				BIT(25)
+#define PHY_POWERDOWN				GENMASK(20, 17)
+#define PHY_BUSY				GENMASK(15, 12)
+#define SSC_DIS					BIT(16)
+#define XMIT_ENABLE				GENMASK(11, 8)
+#define PHY_LANES				GENMASK(7, 6)
+#define PHY_RATE				GENMASK(5, 4)
+#define TPS_SEL					GENMASK(3, 0)
+
+#define DW_DP_PHY_TX_EQ				0x0a04
+#define DW_DP_CUSTOMPAT0			0x0a08
+#define DW_DP_CUSTOMPAT1			0x0a0c
+#define DW_DP_CUSTOMPAT2			0x0a10
+#define DW_DP_HBR2_COMPLIANCE_SCRAMBLER_RESET	0x0a14
+#define DW_DP_PHYIF_PWRDOWN_CTRL		0x0a18
+
+#define DW_DP_AUX_CMD				0x0b00
+#define AUX_CMD_TYPE				GENMASK(31, 28)
+#define AUX_ADDR				GENMASK(27, 8)
+#define I2C_ADDR_ONLY				BIT(4)
+#define AUX_LEN_REQ				GENMASK(3, 0)
+
+#define DW_DP_AUX_STATUS			0x0b04
+#define AUX_TIMEOUT				BIT(17)
+#define AUX_BYTES_READ				GENMASK(23, 19)
+#define AUX_STATUS				GENMASK(7, 4)
+
+#define DW_DP_AUX_DATA0				0x0b08
+#define DW_DP_AUX_DATA1				0x0b0c
+#define DW_DP_AUX_DATA2				0x0b10
+#define DW_DP_AUX_DATA3				0x0b14
+
+#define DW_DP_GENERAL_INTERRUPT			0x0d00
+#define VIDEO_FIFO_OVERFLOW_STREAM0		BIT(6)
+#define AUDIO_FIFO_OVERFLOW_STREAM0		BIT(5)
+#define SDP_EVENT_STREAM0			BIT(4)
+#define AUX_CMD_INVALID				BIT(3)
+#define HDCP_EVENT				BIT(2)
+#define AUX_REPLY_EVENT				BIT(1)
+#define HPD_EVENT				BIT(0)
+
+#define DW_DP_GENERAL_INTERRUPT_ENABLE		0x0d04
+#define HDCP_EVENT_EN				BIT(2)
+#define AUX_REPLY_EVENT_EN			BIT(1)
+#define HPD_EVENT_EN				BIT(0)
+
+#define DW_DP_HPD_STATUS			0x0d08
+#define HPD_STATE				GENMASK(11, 9)
+#define HPD_STATUS				BIT(8)
+#define HPD_HOT_UNPLUG				BIT(2)
+#define HPD_HOT_PLUG				BIT(1)
+#define HPD_IRQ					BIT(0)
+
+#define DW_DP_HPD_INTERRUPT_ENABLE		0x0d0c
+#define HPD_UNPLUG_ERR_EN			BIT(3)
+#define HPD_UNPLUG_EN				BIT(2)
+#define HPD_PLUG_EN				BIT(1)
+#define HPD_IRQ_EN				BIT(0)
+
+#define DW_DP_HDCP_CFG				0x0e00
+#define DPCD12PLUS				BIT(7)
+#define CP_IRQ					BIT(6)
+#define BYPENCRYPTION				BIT(5)
+#define HDCP_LOCK				BIT(4)
+#define ENCRYPTIONDISABLE			BIT(3)
+#define ENABLE_HDCP_13				BIT(2)
+#define ENABLE_HDCP				BIT(1)
+
+#define DW_DP_HDCP_OBS				0x0e04
+#define HDCP22_RE_AUTHENTICATION_REQ		BIT(31)
+#define HDCP22_AUTHENTICATION_FAILED		BIT(30)
+#define HDCP22_AUTHENTICATION_SUCCESS		BIT(29)
+#define HDCP22_CAPABLE_SINK			BIT(28)
+#define HDCP22_SINK_CAP_CHECK_COMPLETE		BIT(27)
+#define HDCP22_STATE				GENMASK(26, 24)
+#define HDCP22_BOOTED				BIT(23)
+#define HDCP13_BSTATUS				GENMASK(22, 19)
+#define REPEATER				BIT(18)
+#define HDCP_CAPABLE				BIT(17)
+#define STATEE					GENMASK(16, 14)
+#define STATEOEG				GENMASK(13, 11)
+#define STATER					GENMASK(10, 8)
+#define STATEA					GENMASK(7, 4)
+#define SUBSTATEA				GENMASK(3, 1)
+#define HDCPENGAGED				BIT(0)
+
+#define DW_DP_HDCP_APIINTCLR			0x0e08
+#define DW_DP_HDCP_APIINTSTAT			0x0e0c
+#define DW_DP_HDCP_APIINTMSK			0x0e10
+#define HDCP22_GPIOINT				BIT(8)
+#define HDCP_ENGAGED				BIT(7)
+#define HDCP_FAILED				BIT(6)
+#define KSVSHA1CALCDONEINT			BIT(5)
+#define AUXRESPNACK7TIMES			BIT(4)
+#define AUXRESPTIMEOUT				BIT(3)
+#define AUXRESPDEFER7TIMES			BIT(2)
+#define KSVACCESSINT				BIT(0)
+
+#define DW_DP_HDCP_KSVMEMCTRL			0x0e18
+#define KSVSHA1STATUS				BIT(4)
+#define KSVMEMACCESS				BIT(1)
+#define KSVMEMREQUEST				BIT(0)
+
+#define DW_DP_HDCP_REG_BKSV0			0x3600
+#define DW_DP_HDCP_REG_BKSV1			0x3604
+#define DW_DP_HDCP_REG_ANCONF			0x3608
+#define AN_BYPASS				BIT(0)
+
+#define DW_DP_HDCP_REG_AN0			0x360c
+#define DW_DP_HDCP_REG_AN1			0x3610
+#define DW_DP_HDCP_REG_RMLCTL			0x3614
+#define ODPK_DECRYPT_ENABLE			BIT(0)
+
+#define DW_DP_HDCP_REG_RMLSTS			0x3618
+#define IDPK_WR_OK_STS				BIT(6)
+#define	IDPK_DATA_INDEX				GENMASK(5, 0)
+#define DW_DP_HDCP_REG_SEED			0x361c
+#define DW_DP_HDCP_REG_DPK0			0x3620
+#define DW_DP_HDCP_REG_DPK1			0x3624
+#define DW_DP_HDCP22_GPIOSTS			0x3628
+#define DW_DP_HDCP22_GPIOCHNGSTS		0x362c
+#define DW_DP_HDCP_REG_DPK_CRC			0x3630
+
+#define DW_DP_MAX_REGISTER			DW_DP_HDCP_REG_DPK_CRC
+
+#define SDP_REG_BANK_SIZE			16
+
+struct dw_dp_link_caps {
+	bool enhanced_framing;
+	bool tps3_supported;
+	bool tps4_supported;
+	bool fast_training;
+	bool channel_coding;
+	bool ssc;
+};
+
+struct dw_dp_link_train_set {
+	unsigned int voltage_swing[4];
+	unsigned int pre_emphasis[4];
+	bool voltage_max_reached[4];
+	bool pre_max_reached[4];
+};
+
+struct dw_dp_link_train {
+	struct dw_dp_link_train_set adjust;
+	bool clock_recovered;
+	bool channel_equalized;
+};
+
+struct dw_dp_link {
+	u8 dpcd[DP_RECEIVER_CAP_SIZE];
+	unsigned char revision;
+	unsigned int rate;
+	unsigned int lanes;
+	u8 sink_count;
+	u8 vsc_sdp_supported;
+	struct dw_dp_link_caps caps;
+	struct dw_dp_link_train train;
+	struct drm_dp_desc desc;
+};
+
+struct dw_dp_bridge_state {
+	struct drm_bridge_state base;
+	struct drm_display_mode mode;
+	u8 video_mapping;
+	u8 color_format;
+	u8 bpc;
+	u8 bpp;
+};
+
+struct dw_dp_sdp {
+	struct dp_sdp base;
+	unsigned long flags;
+};
+
+struct dw_dp_hotplug {
+	bool long_hpd;
+};
+
+struct dw_dp {
+	struct drm_bridge bridge;
+	struct device *dev;
+	struct regmap *regmap;
+	struct phy *phy;
+	struct clk *apb_clk;
+	struct clk *aux_clk;
+	struct clk *i2s_clk;
+	struct clk *spdif_clk;
+	struct clk *hdcp_clk;
+	struct reset_control *rstc;
+	struct completion complete;
+	int irq;
+	struct work_struct hpd_work;
+	struct dw_dp_hotplug hotplug;
+	/* Serialize hpd status access */
+	struct mutex irq_lock;
+
+	struct drm_dp_aux aux;
+
+	struct dw_dp_link link;
+	struct dw_dp_plat_data plat_data;
+	u8 pixel_mode;
+
+	DECLARE_BITMAP(sdp_reg_bank, SDP_REG_BANK_SIZE);
+};
+
+enum {
+	DW_DP_RGB_6BIT,
+	DW_DP_RGB_8BIT,
+	DW_DP_RGB_10BIT,
+	DW_DP_RGB_12BIT,
+	DW_DP_RGB_16BIT,
+	DW_DP_YCBCR444_8BIT,
+	DW_DP_YCBCR444_10BIT,
+	DW_DP_YCBCR444_12BIT,
+	DW_DP_YCBCR444_16BIT,
+	DW_DP_YCBCR422_8BIT,
+	DW_DP_YCBCR422_10BIT,
+	DW_DP_YCBCR422_12BIT,
+	DW_DP_YCBCR422_16BIT,
+	DW_DP_YCBCR420_8BIT,
+	DW_DP_YCBCR420_10BIT,
+	DW_DP_YCBCR420_12BIT,
+	DW_DP_YCBCR420_16BIT,
+};
+
+enum {
+	DW_DP_MP_SINGLE_PIXEL,
+	DW_DP_MP_DUAL_PIXEL,
+	DW_DP_MP_QUAD_PIXEL,
+};
+
+enum {
+	DW_DP_SDP_VERTICAL_INTERVAL = BIT(0),
+	DW_DP_SDP_HORIZONTAL_INTERVAL = BIT(1),
+};
+
+enum {
+	DW_DP_HPD_STATE_IDLE,
+	DW_DP_HPD_STATE_UNPLUG,
+	DP_DP_HPD_STATE_TIMEOUT = 4,
+	DW_DP_HPD_STATE_PLUG = 7
+};
+
+enum {
+	DW_DP_PHY_PATTERN_NONE,
+	DW_DP_PHY_PATTERN_TPS_1,
+	DW_DP_PHY_PATTERN_TPS_2,
+	DW_DP_PHY_PATTERN_TPS_3,
+	DW_DP_PHY_PATTERN_TPS_4,
+	DW_DP_PHY_PATTERN_SERM,
+	DW_DP_PHY_PATTERN_PBRS7,
+	DW_DP_PHY_PATTERN_CUSTOM_80BIT,
+	DW_DP_PHY_PATTERN_CP2520_1,
+	DW_DP_PHY_PATTERN_CP2520_2,
+};
+
+struct dw_dp_output_format {
+	u32 bus_format;
+	u32 color_format;
+	u8 video_mapping;
+	u8 bpc;
+	u8 bpp;
+};
+
+#define to_dw_dp_bridge_state(s) container_of(s, struct dw_dp_bridge_state, base)
+
+static const struct dw_dp_output_format dw_dp_output_formats[] = {
+	{ MEDIA_BUS_FMT_RGB101010_1X30, DRM_COLOR_FORMAT_RGB444, DW_DP_RGB_10BIT, 10, 30 },
+	{ MEDIA_BUS_FMT_RGB888_1X24, DRM_COLOR_FORMAT_RGB444, DW_DP_RGB_8BIT, 8, 24 },
+	{ MEDIA_BUS_FMT_YUV10_1X30, DRM_COLOR_FORMAT_YCBCR444, DW_DP_YCBCR444_10BIT, 10, 30 },
+	{ MEDIA_BUS_FMT_YUV8_1X24, DRM_COLOR_FORMAT_YCBCR444, DW_DP_YCBCR444_8BIT, 8, 24},
+	{ MEDIA_BUS_FMT_YUYV10_1X20, DRM_COLOR_FORMAT_YCBCR422, DW_DP_YCBCR422_10BIT, 10, 20 },
+	{ MEDIA_BUS_FMT_YUYV8_1X16, DRM_COLOR_FORMAT_YCBCR422, DW_DP_YCBCR422_8BIT, 8, 16 },
+	{ MEDIA_BUS_FMT_UYYVYY10_0_5X30, DRM_COLOR_FORMAT_YCBCR420, DW_DP_YCBCR420_10BIT, 10, 15 },
+	{ MEDIA_BUS_FMT_UYYVYY8_0_5X24, DRM_COLOR_FORMAT_YCBCR420, DW_DP_YCBCR420_8BIT, 8, 12 },
+	{ MEDIA_BUS_FMT_RGB666_1X24_CPADHI, DRM_COLOR_FORMAT_RGB444, DW_DP_RGB_6BIT, 6, 18 },
+};
+
+static const struct dw_dp_output_format *dw_dp_get_output_format(u32 bus_format)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(dw_dp_output_formats); i++)
+		if (dw_dp_output_formats[i].bus_format == bus_format)
+			return &dw_dp_output_formats[i];
+
+	return NULL;
+}
+
+static inline struct dw_dp *bridge_to_dp(struct drm_bridge *b)
+{
+	return container_of(b, struct dw_dp, bridge);
+}
+
+static struct dw_dp_bridge_state *dw_dp_get_bridge_state(struct dw_dp *dp)
+{
+	struct dw_dp_bridge_state *dw_bridge_state;
+	struct drm_bridge_state *state;
+
+	state = drm_priv_to_bridge_state(dp->bridge.base.state);
+	if (!state)
+		return  NULL;
+
+	dw_bridge_state = to_dw_dp_bridge_state(state);
+	if (!dw_bridge_state)
+		return NULL;
+
+	return dw_bridge_state;
+}
+
+static inline void dw_dp_phy_set_pattern(struct dw_dp *dp, u32 pattern)
+{
+	regmap_update_bits(dp->regmap, DW_DP_PHYIF_CTRL, TPS_SEL,
+			   FIELD_PREP(TPS_SEL, pattern));
+}
+
+static void dw_dp_phy_xmit_enable(struct dw_dp *dp, u32 lanes)
+{
+	u32 xmit_enable;
+
+	switch (lanes) {
+	case 4:
+	case 2:
+	case 1:
+		xmit_enable = GENMASK(lanes - 1, 0);
+		break;
+	case 0:
+	default:
+		xmit_enable = 0;
+		break;
+	}
+
+	regmap_update_bits(dp->regmap, DW_DP_PHYIF_CTRL, XMIT_ENABLE,
+			   FIELD_PREP(XMIT_ENABLE, xmit_enable));
+}
+
+static bool dw_dp_bandwidth_ok(struct dw_dp *dp,
+			       const struct drm_display_mode *mode, u32 bpp,
+			       unsigned int lanes, unsigned int rate)
+{
+	u32 max_bw, req_bw;
+
+	req_bw = mode->clock * bpp / 8;
+	max_bw = lanes * rate;
+	if (req_bw > max_bw)
+		return false;
+
+	return true;
+}
+
+static bool dw_dp_hpd_detect(struct dw_dp *dp)
+{
+	u32 value;
+
+	regmap_read(dp->regmap, DW_DP_HPD_STATUS, &value);
+
+	return FIELD_GET(HPD_STATE, value) == DW_DP_HPD_STATE_PLUG;
+}
+
+static void dw_dp_link_caps_reset(struct dw_dp_link_caps *caps)
+{
+	caps->enhanced_framing = false;
+	caps->tps3_supported = false;
+	caps->tps4_supported = false;
+	caps->fast_training = false;
+	caps->channel_coding = false;
+}
+
+static void dw_dp_link_reset(struct dw_dp_link *link)
+{
+	link->vsc_sdp_supported = 0;
+	link->sink_count = 0;
+	link->revision = 0;
+	link->rate = 0;
+	link->lanes = 0;
+
+	dw_dp_link_caps_reset(&link->caps);
+	memset(link->dpcd, 0, sizeof(link->dpcd));
+}
+
+static int dw_dp_link_parse(struct dw_dp *dp, struct drm_connector *connector)
+{
+	struct dw_dp_link *link = &dp->link;
+	int ret;
+
+	dw_dp_link_reset(link);
+
+	ret = drm_dp_read_dpcd_caps(&dp->aux, link->dpcd);
+	if (ret < 0)
+		return ret;
+
+	drm_dp_read_desc(&dp->aux, &link->desc, drm_dp_is_branch(link->dpcd));
+
+	if (drm_dp_read_sink_count_cap(connector, link->dpcd, &link->desc)) {
+		ret = drm_dp_read_sink_count(&dp->aux);
+		if (ret < 0)
+			return ret;
+
+		link->sink_count = ret;
+
+		/* Dongle connected, but no display */
+		if (!link->sink_count)
+			return -ENODEV;
+	}
+
+	link->vsc_sdp_supported = drm_dp_vsc_sdp_supported(&dp->aux, link->dpcd);
+
+	link->revision = link->dpcd[DP_DPCD_REV];
+	link->rate = min_t(u32, min(dp->plat_data.max_link_rate,
+				    dp->phy->attrs.max_link_rate * 100),
+			   drm_dp_max_link_rate(link->dpcd));
+	link->lanes = min_t(u8, phy_get_bus_width(dp->phy),
+			    drm_dp_max_lane_count(link->dpcd));
+
+	link->caps.enhanced_framing = drm_dp_enhanced_frame_cap(link->dpcd);
+	link->caps.tps3_supported = drm_dp_tps3_supported(link->dpcd);
+	link->caps.tps4_supported = drm_dp_tps4_supported(link->dpcd);
+	link->caps.fast_training = drm_dp_fast_training_cap(link->dpcd);
+	link->caps.channel_coding = drm_dp_channel_coding_supported(link->dpcd);
+	link->caps.ssc = !!(link->dpcd[DP_MAX_DOWNSPREAD] & DP_MAX_DOWNSPREAD_0_5);
+
+	return 0;
+}
+
+static int dw_dp_link_train_update_vs_emph(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	struct dw_dp_link_train_set *train_set = &link->train.adjust;
+	unsigned int lanes = dp->link.lanes;
+	union phy_configure_opts phy_cfg;
+	unsigned int *vs, *pe;
+	int i, ret;
+	u8 buf[4];
+
+	vs = train_set->voltage_swing;
+	pe = train_set->pre_emphasis;
+
+	for (i = 0; i < lanes; i++) {
+		phy_cfg.dp.voltage[i] = vs[i];
+		phy_cfg.dp.pre[i] = pe[i];
+	}
+
+	phy_cfg.dp.set_lanes = false;
+	phy_cfg.dp.set_rate = false;
+	phy_cfg.dp.set_voltages = true;
+
+	ret = phy_configure(dp->phy, &phy_cfg);
+	if (ret)
+		return ret;
+
+	for (i = 0; i < lanes; i++) {
+		buf[i] = (vs[i] << DP_TRAIN_VOLTAGE_SWING_SHIFT) |
+			 (pe[i] << DP_TRAIN_PRE_EMPHASIS_SHIFT);
+		if (train_set->voltage_max_reached[i])
+			buf[i] |= DP_TRAIN_MAX_SWING_REACHED;
+		if (train_set->pre_max_reached[i])
+			buf[i] |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED;
+	}
+
+	ret = drm_dp_dpcd_write(&dp->aux, DP_TRAINING_LANE0_SET, buf, lanes);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int dw_dp_phy_configure(struct dw_dp *dp, unsigned int rate,
+			       unsigned int lanes, bool ssc)
+{
+	union phy_configure_opts phy_cfg;
+	int ret;
+
+	/* Move PHY to P3 */
+	regmap_update_bits(dp->regmap, DW_DP_PHYIF_CTRL, PHY_POWERDOWN,
+			   FIELD_PREP(PHY_POWERDOWN, 0x3));
+
+	phy_cfg.dp.lanes = lanes;
+	phy_cfg.dp.link_rate = rate / 100;
+	phy_cfg.dp.ssc = ssc;
+	phy_cfg.dp.set_lanes = true;
+	phy_cfg.dp.set_rate = true;
+	phy_cfg.dp.set_voltages = false;
+	ret = phy_configure(dp->phy, &phy_cfg);
+	if (ret)
+		return ret;
+
+	regmap_update_bits(dp->regmap, DW_DP_PHYIF_CTRL, PHY_LANES,
+			   FIELD_PREP(PHY_LANES, lanes / 2));
+
+	/* Move PHY to P0 */
+	regmap_update_bits(dp->regmap, DW_DP_PHYIF_CTRL, PHY_POWERDOWN,
+			   FIELD_PREP(PHY_POWERDOWN, 0x0));
+
+	dw_dp_phy_xmit_enable(dp, lanes);
+
+	return 0;
+}
+
+static int dw_dp_link_configure(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	u8 buf[2];
+	int ret;
+
+	ret = dw_dp_phy_configure(dp, link->rate, link->lanes, link->caps.ssc);
+	if (ret)
+		return ret;
+
+	buf[0] = drm_dp_link_rate_to_bw_code(link->rate);
+	buf[1] = link->lanes;
+
+	if (link->caps.enhanced_framing) {
+		buf[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN;
+		regmap_update_bits(dp->regmap, DW_DP_CCTL, ENHANCE_FRAMING_EN,
+				   FIELD_PREP(ENHANCE_FRAMING_EN, 1));
+	} else {
+		regmap_update_bits(dp->regmap, DW_DP_CCTL, ENHANCE_FRAMING_EN,
+				   FIELD_PREP(ENHANCE_FRAMING_EN, 0));
+	}
+
+	ret = drm_dp_dpcd_write(&dp->aux, DP_LINK_BW_SET, buf, sizeof(buf));
+	if (ret < 0)
+		return ret;
+
+	buf[0] = link->caps.ssc ? DP_SPREAD_AMP_0_5 : 0;
+	buf[1] = link->caps.channel_coding ? DP_SET_ANSI_8B10B : 0;
+
+	ret = drm_dp_dpcd_write(&dp->aux, DP_DOWNSPREAD_CTRL, buf, sizeof(buf));
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static void dw_dp_link_train_init(struct dw_dp_link_train *train)
+{
+	struct dw_dp_link_train_set *adj = &train->adjust;
+	unsigned int i;
+
+	for (i = 0; i < 4; i++) {
+		adj->voltage_swing[i] = 0;
+		adj->pre_emphasis[i] = 0;
+		adj->voltage_max_reached[i] = false;
+		adj->pre_max_reached[i] = false;
+	}
+
+	train->clock_recovered = false;
+	train->channel_equalized = false;
+}
+
+static bool dw_dp_link_train_valid(const struct dw_dp_link_train *train)
+{
+	return train->clock_recovered && train->channel_equalized;
+}
+
+static int dw_dp_link_train_set_pattern(struct dw_dp *dp, u32 pattern)
+{
+	u8 buf = 0;
+	int ret;
+
+	if (pattern && pattern != DP_TRAINING_PATTERN_4) {
+		buf |= DP_LINK_SCRAMBLING_DISABLE;
+
+		regmap_update_bits(dp->regmap, DW_DP_CCTL, SCRAMBLE_DIS,
+				   FIELD_PREP(SCRAMBLE_DIS, 1));
+	} else {
+		regmap_update_bits(dp->regmap, DW_DP_CCTL, SCRAMBLE_DIS,
+				   FIELD_PREP(SCRAMBLE_DIS, 0));
+	}
+
+	switch (pattern) {
+	case DP_TRAINING_PATTERN_DISABLE:
+		dw_dp_phy_set_pattern(dp, DW_DP_PHY_PATTERN_NONE);
+		break;
+	case DP_TRAINING_PATTERN_1:
+		dw_dp_phy_set_pattern(dp, DW_DP_PHY_PATTERN_TPS_1);
+		break;
+	case DP_TRAINING_PATTERN_2:
+		dw_dp_phy_set_pattern(dp, DW_DP_PHY_PATTERN_TPS_2);
+		break;
+	case DP_TRAINING_PATTERN_3:
+		dw_dp_phy_set_pattern(dp, DW_DP_PHY_PATTERN_TPS_3);
+		break;
+	case DP_TRAINING_PATTERN_4:
+		dw_dp_phy_set_pattern(dp, DW_DP_PHY_PATTERN_TPS_4);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	ret = drm_dp_dpcd_writeb(&dp->aux, DP_TRAINING_PATTERN_SET,
+				 buf | pattern);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static u8 dw_dp_voltage_max(u8 preemph)
+{
+	switch (preemph & DP_TRAIN_PRE_EMPHASIS_MASK) {
+	case DP_TRAIN_PRE_EMPH_LEVEL_0:
+		return DP_TRAIN_VOLTAGE_SWING_LEVEL_3;
+	case DP_TRAIN_PRE_EMPH_LEVEL_1:
+		return DP_TRAIN_VOLTAGE_SWING_LEVEL_2;
+	case DP_TRAIN_PRE_EMPH_LEVEL_2:
+		return DP_TRAIN_VOLTAGE_SWING_LEVEL_1;
+	case DP_TRAIN_PRE_EMPH_LEVEL_3:
+	default:
+		return DP_TRAIN_VOLTAGE_SWING_LEVEL_0;
+	}
+}
+
+static bool dw_dp_link_get_adjustments(struct dw_dp_link *link,
+				       u8 status[DP_LINK_STATUS_SIZE])
+{
+	struct dw_dp_link_train_set *adj = &link->train.adjust;
+	unsigned int i;
+	bool changed = false;
+	u8 v = 0;
+	u8 p = 0;
+
+	for (i = 0; i < link->lanes; i++) {
+		v = drm_dp_get_adjust_request_voltage(status, i);
+		v >>= DP_TRAIN_VOLTAGE_SWING_SHIFT;
+		p = drm_dp_get_adjust_request_pre_emphasis(status, i);
+		p >>= DP_TRAIN_PRE_EMPHASIS_SHIFT;
+
+		if (v != adj->voltage_swing[i] || p != adj->pre_emphasis[i])
+			changed = true;
+
+		if (p >=  (DP_TRAIN_PRE_EMPH_LEVEL_3 >> DP_TRAIN_PRE_EMPHASIS_SHIFT)) {
+			adj->pre_emphasis[i] = DP_TRAIN_PRE_EMPH_LEVEL_3 >>
+					       DP_TRAIN_PRE_EMPHASIS_SHIFT;
+			adj->pre_max_reached[i] = true;
+		} else {
+			adj->pre_emphasis[i] = p;
+			adj->pre_max_reached[i] = false;
+		}
+
+		v = min(v, dw_dp_voltage_max(p));
+		if (v >= (DP_TRAIN_VOLTAGE_SWING_LEVEL_3 >> DP_TRAIN_VOLTAGE_SWING_SHIFT)) {
+			adj->voltage_swing[i] = DP_TRAIN_VOLTAGE_SWING_LEVEL_3 >>
+						DP_TRAIN_VOLTAGE_SWING_SHIFT;
+			adj->voltage_max_reached[i] = true;
+		} else {
+			adj->voltage_swing[i] = v;
+			adj->voltage_max_reached[i] = false;
+		}
+	}
+
+	return changed;
+}
+
+static int dw_dp_link_clock_recovery(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	u8 status[DP_LINK_STATUS_SIZE];
+	unsigned int tries = 0;
+	int ret;
+	bool adj_changed;
+
+	ret = dw_dp_link_train_set_pattern(dp, DP_TRAINING_PATTERN_1);
+	if (ret)
+		return ret;
+
+	for (;;) {
+		ret = dw_dp_link_train_update_vs_emph(dp);
+		if (ret)
+			return ret;
+
+		drm_dp_link_train_clock_recovery_delay(&dp->aux, link->dpcd);
+
+		ret = drm_dp_dpcd_read_link_status(&dp->aux, status);
+		if (ret < 0) {
+			dev_err(dp->dev, "failed to read link status: %d\n", ret);
+			return ret;
+		}
+
+		if (drm_dp_clock_recovery_ok(status, link->lanes)) {
+			link->train.clock_recovered = true;
+			break;
+		}
+
+		/*
+		 * According to DP spec 1.4, if current ADJ is the same
+		 * with previous REQ, we need to retry 5 times.
+		 */
+		adj_changed = dw_dp_link_get_adjustments(link, status);
+		if (!adj_changed)
+			tries++;
+		else
+			tries = 0;
+
+		if (tries == 5)
+			break;
+	}
+
+	return 0;
+}
+
+static int dw_dp_link_channel_equalization(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	u8 status[DP_LINK_STATUS_SIZE], pattern;
+	unsigned int tries;
+	int ret;
+
+	if (link->caps.tps4_supported)
+		pattern = DP_TRAINING_PATTERN_4;
+	else if (link->caps.tps3_supported)
+		pattern = DP_TRAINING_PATTERN_3;
+	else
+		pattern = DP_TRAINING_PATTERN_2;
+	ret = dw_dp_link_train_set_pattern(dp, pattern);
+	if (ret)
+		return ret;
+
+	for (tries = 1; tries < 5; tries++) {
+		ret = dw_dp_link_train_update_vs_emph(dp);
+		if (ret)
+			return ret;
+
+		drm_dp_link_train_channel_eq_delay(&dp->aux, link->dpcd);
+
+		ret = drm_dp_dpcd_read_link_status(&dp->aux, status);
+		if (ret < 0)
+			return ret;
+
+		if (!drm_dp_clock_recovery_ok(status, link->lanes)) {
+			dev_err(dp->dev, "clock recovery lost while equalizing channel\n");
+			link->train.clock_recovered = false;
+			break;
+		}
+
+		if (drm_dp_channel_eq_ok(status, link->lanes)) {
+			link->train.channel_equalized = true;
+			break;
+		}
+
+		dw_dp_link_get_adjustments(link, status);
+	}
+
+	return 0;
+}
+
+static int dw_dp_link_downgrade(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	struct dw_dp_bridge_state *state;
+
+	state = dw_dp_get_bridge_state(dp);
+
+	switch (link->rate) {
+	case 162000:
+		return -EINVAL;
+	case 270000:
+		link->rate = 162000;
+		break;
+	case 540000:
+		link->rate = 270000;
+		break;
+	case 810000:
+		link->rate = 540000;
+		break;
+	}
+
+	if (!dw_dp_bandwidth_ok(dp, &state->mode, state->bpp, link->lanes,
+				link->rate))
+		return -E2BIG;
+
+	return 0;
+}
+
+static int dw_dp_link_train_full(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	int ret;
+
+retry:
+	dw_dp_link_train_init(&link->train);
+
+	dev_dbg(dp->dev, "full-training link: %u lane%s at %u MHz\n",
+		link->lanes, (link->lanes > 1) ? "s" : "", link->rate / 100);
+
+	ret = dw_dp_link_configure(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "failed to configure DP link: %d\n", ret);
+		return ret;
+	}
+
+	ret = dw_dp_link_clock_recovery(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "clock recovery failed: %d\n", ret);
+		goto out;
+	}
+
+	if (!link->train.clock_recovered) {
+		dev_err(dp->dev, "clock recovery failed, downgrading link\n");
+
+		ret = dw_dp_link_downgrade(dp);
+		if (ret < 0)
+			goto out;
+		else
+			goto retry;
+	}
+
+	dev_dbg(dp->dev, "clock recovery succeeded\n");
+
+	ret = dw_dp_link_channel_equalization(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "channel equalization failed: %d\n", ret);
+		goto out;
+	}
+
+	if (!link->train.channel_equalized) {
+		dev_err(dp->dev, "channel equalization failed, downgrading link\n");
+
+		ret = dw_dp_link_downgrade(dp);
+		if (ret < 0)
+			goto out;
+		else
+			goto retry;
+	}
+
+	dev_dbg(dp->dev, "channel equalization succeeded\n");
+
+out:
+	dw_dp_link_train_set_pattern(dp, DP_TRAINING_PATTERN_DISABLE);
+	return ret;
+}
+
+static int dw_dp_link_train_fast(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	int ret;
+	u8 status[DP_LINK_STATUS_SIZE];
+	u8 pattern;
+
+	dw_dp_link_train_init(&link->train);
+
+	dev_dbg(dp->dev, "fast-training link: %u lane%s at %u MHz\n",
+		link->lanes, (link->lanes > 1) ? "s" : "", link->rate / 100);
+
+	ret = dw_dp_link_configure(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "failed to configure DP link: %d\n", ret);
+		return ret;
+	}
+
+	ret = dw_dp_link_train_set_pattern(dp, DP_TRAINING_PATTERN_1);
+	if (ret)
+		goto out;
+
+	usleep_range(500, 1000);
+
+	if (link->caps.tps4_supported)
+		pattern = DP_TRAINING_PATTERN_4;
+	else if (link->caps.tps3_supported)
+		pattern = DP_TRAINING_PATTERN_3;
+	else
+		pattern = DP_TRAINING_PATTERN_2;
+	ret = dw_dp_link_train_set_pattern(dp, pattern);
+	if (ret)
+		goto out;
+
+	usleep_range(500, 1000);
+
+	ret = drm_dp_dpcd_read_link_status(&dp->aux, status);
+	if (ret < 0) {
+		dev_err(dp->dev, "failed to read link status: %d\n", ret);
+		goto out;
+	}
+
+	if (!drm_dp_clock_recovery_ok(status, link->lanes)) {
+		dev_err(dp->dev, "clock recovery failed\n");
+		ret = -EIO;
+		goto out;
+	}
+
+	if (!drm_dp_channel_eq_ok(status, link->lanes)) {
+		dev_err(dp->dev, "channel equalization failed\n");
+		ret = -EIO;
+		goto out;
+	}
+
+out:
+	dw_dp_link_train_set_pattern(dp, DP_TRAINING_PATTERN_DISABLE);
+	return ret;
+}
+
+static int dw_dp_link_train(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	int ret;
+
+	if (link->caps.fast_training) {
+		if (dw_dp_link_train_valid(&link->train)) {
+			ret = dw_dp_link_train_fast(dp);
+			if (ret < 0)
+				dev_err(dp->dev, "fast link training failed: %d\n", ret);
+			else
+				return 0;
+		}
+	}
+
+	ret = dw_dp_link_train_full(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "full link training failed: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int dw_dp_send_sdp(struct dw_dp *dp, struct dw_dp_sdp *sdp)
+{
+	const u8 *payload = sdp->base.db;
+	u32 reg;
+	int i, nr;
+
+	nr = find_first_zero_bit(dp->sdp_reg_bank, SDP_REG_BANK_SIZE);
+	if (nr < SDP_REG_BANK_SIZE)
+		set_bit(nr, dp->sdp_reg_bank);
+	else
+		return -EBUSY;
+
+	reg = DW_DP_SDP_REGISTER_BANK + nr * 9 * 4;
+
+	/* SDP header */
+	regmap_write(dp->regmap, reg, get_unaligned_le32(&sdp->base.sdp_header));
+
+	/* SDP data payload */
+	for (i = 1; i < 9; i++, payload += 4)
+		regmap_write(dp->regmap, reg + i * 4,
+			     FIELD_PREP(SDP_REGS, get_unaligned_le32(payload)));
+
+	if (sdp->flags & DW_DP_SDP_VERTICAL_INTERVAL)
+		regmap_update_bits(dp->regmap, DW_DP_SDP_VERTICAL_CTRL,
+				   EN_VERTICAL_SDP << nr,
+				   EN_VERTICAL_SDP << nr);
+
+	if (sdp->flags & DW_DP_SDP_HORIZONTAL_INTERVAL)
+		regmap_update_bits(dp->regmap, DW_DP_SDP_HORIZONTAL_CTRL,
+				   EN_HORIZONTAL_SDP << nr,
+				   EN_HORIZONTAL_SDP << nr);
+
+	return 0;
+}
+
+static int dw_dp_send_vsc_sdp(struct dw_dp *dp)
+{
+	struct dw_dp_bridge_state *state;
+	struct dw_dp_sdp sdp = {};
+	struct drm_dp_vsc_sdp vsc = {};
+
+	state = dw_dp_get_bridge_state(dp);
+	if (!state)
+		return -EINVAL;
+
+	vsc.bpc = state->bpc;
+
+	vsc.sdp_type = DP_SDP_VSC;
+	vsc.revision = 0x5;
+	vsc.length = 0x13;
+	vsc.content_type = DP_CONTENT_TYPE_NOT_DEFINED;
+
+	sdp.flags = DW_DP_SDP_VERTICAL_INTERVAL;
+
+	switch (state->color_format) {
+	case DRM_COLOR_FORMAT_YCBCR444:
+		vsc.pixelformat = DP_PIXELFORMAT_YUV444;
+		break;
+	case DRM_COLOR_FORMAT_YCBCR420:
+		vsc.pixelformat = DP_PIXELFORMAT_YUV420;
+		break;
+	case DRM_COLOR_FORMAT_YCBCR422:
+		vsc.pixelformat = DP_PIXELFORMAT_YUV422;
+		break;
+	case DRM_COLOR_FORMAT_RGB444:
+	default:
+		vsc.pixelformat = DP_PIXELFORMAT_RGB;
+		break;
+	}
+
+	if (state->color_format == DRM_COLOR_FORMAT_RGB444) {
+		vsc.colorimetry = DP_COLORIMETRY_DEFAULT;
+		vsc.dynamic_range = DP_DYNAMIC_RANGE_VESA;
+	} else {
+		vsc.colorimetry = DP_COLORIMETRY_BT709_YCC;
+		vsc.dynamic_range = DP_DYNAMIC_RANGE_CTA;
+	}
+
+	drm_dp_vsc_sdp_pack(&vsc, &sdp.base);
+
+	return dw_dp_send_sdp(dp, &sdp);
+}
+
+static int dw_dp_video_set_pixel_mode(struct dw_dp *dp)
+{
+	switch (dp->pixel_mode) {
+	case DW_DP_MP_SINGLE_PIXEL:
+	case DW_DP_MP_DUAL_PIXEL:
+	case DW_DP_MP_QUAD_PIXEL:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	regmap_update_bits(dp->regmap, DW_DP_VSAMPLE_CTRL, PIXEL_MODE_SELECT,
+			   FIELD_PREP(PIXEL_MODE_SELECT, dp->pixel_mode));
+
+	return 0;
+}
+
+static bool dw_dp_video_need_vsc_sdp(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	struct dw_dp_bridge_state *state;
+
+	state = dw_dp_get_bridge_state(dp);
+	if (!state)
+		return -EINVAL;
+
+	if (!link->vsc_sdp_supported)
+		return false;
+
+	if (state->color_format == DRM_COLOR_FORMAT_YCBCR420)
+		return true;
+
+	return false;
+}
+
+static int dw_dp_video_set_msa(struct dw_dp *dp, u8 color_format, u8 bpc,
+			       u16 vstart, u16 hstart)
+{
+	u16 misc = 0;
+
+	if (dw_dp_video_need_vsc_sdp(dp))
+		misc |= DP_MSA_MISC_COLOR_VSC_SDP;
+
+	switch (color_format) {
+	case DRM_COLOR_FORMAT_RGB444:
+		misc |= DP_MSA_MISC_COLOR_RGB;
+		break;
+	case DRM_COLOR_FORMAT_YCBCR444:
+		misc |= DP_MSA_MISC_COLOR_YCBCR_444_BT709;
+		break;
+	case DRM_COLOR_FORMAT_YCBCR422:
+		misc |= DP_MSA_MISC_COLOR_YCBCR_422_BT709;
+		break;
+	case DRM_COLOR_FORMAT_YCBCR420:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	switch (bpc) {
+	case 6:
+		misc |= DP_MSA_MISC_6_BPC;
+		break;
+	case 8:
+		misc |= DP_MSA_MISC_8_BPC;
+		break;
+	case 10:
+		misc |= DP_MSA_MISC_10_BPC;
+		break;
+	case 12:
+		misc |= DP_MSA_MISC_12_BPC;
+		break;
+	case 16:
+		misc |= DP_MSA_MISC_16_BPC;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	regmap_write(dp->regmap, DW_DP_VIDEO_MSA1,
+		     FIELD_PREP(VSTART, vstart) | FIELD_PREP(HSTART, hstart));
+	regmap_write(dp->regmap, DW_DP_VIDEO_MSA2, FIELD_PREP(MISC0, misc));
+	regmap_write(dp->regmap, DW_DP_VIDEO_MSA3, FIELD_PREP(MISC1, misc >> 8));
+
+	return 0;
+}
+
+static void dw_dp_video_disable(struct dw_dp *dp)
+{
+	regmap_update_bits(dp->regmap, DW_DP_VSAMPLE_CTRL, VIDEO_STREAM_ENABLE,
+			   FIELD_PREP(VIDEO_STREAM_ENABLE, 0));
+}
+
+static int dw_dp_video_enable(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	struct dw_dp_bridge_state *state;
+	struct drm_display_mode *mode;
+	u8 color_format, bpc, bpp;
+	u8 init_threshold, vic;
+	u32 hstart, hactive, hblank, h_sync_width, h_front_porch;
+	u32 vstart, vactive, vblank, v_sync_width, v_front_porch;
+	u32 peak_stream_bandwidth, link_bandwidth;
+	u32 average_bytes_per_tu, average_bytes_per_tu_frac;
+	u32 ts, hblank_interval;
+	u32 value;
+	int ret;
+
+	state = dw_dp_get_bridge_state(dp);
+	if (!state)
+		return -EINVAL;
+
+	bpc = state->bpc;
+	bpp = state->bpp;
+	color_format = state->color_format;
+	mode = &state->mode;
+
+	vstart = mode->vtotal - mode->vsync_start;
+	hstart = mode->htotal - mode->hsync_start;
+
+	ret = dw_dp_video_set_pixel_mode(dp);
+	if (ret)
+		return ret;
+
+	ret = dw_dp_video_set_msa(dp, color_format, bpc, vstart, hstart);
+	if (ret)
+		return ret;
+
+	regmap_update_bits(dp->regmap, DW_DP_VSAMPLE_CTRL, VIDEO_MAPPING,
+			   FIELD_PREP(VIDEO_MAPPING, state->video_mapping));
+
+	/* Configure DW_DP_VINPUT_POLARITY_CTRL register */
+	value = 0;
+	if (mode->flags & DRM_MODE_FLAG_PHSYNC)
+		value |= FIELD_PREP(HSYNC_IN_POLARITY, 1);
+	if (mode->flags & DRM_MODE_FLAG_PVSYNC)
+		value |= FIELD_PREP(VSYNC_IN_POLARITY, 1);
+	regmap_write(dp->regmap, DW_DP_VINPUT_POLARITY_CTRL, value);
+
+	/* Configure DW_DP_VIDEO_CONFIG1 register */
+	hactive = mode->hdisplay;
+	hblank = mode->htotal - mode->hdisplay;
+	value = FIELD_PREP(HACTIVE, hactive) | FIELD_PREP(HBLANK, hblank);
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		value |= FIELD_PREP(I_P, 1);
+	vic = drm_match_cea_mode(mode);
+	if (vic == 5 || vic == 6 || vic == 7 ||
+	    vic == 10 || vic == 11 || vic == 20 ||
+	    vic == 21 || vic == 22 || vic == 39 ||
+	    vic == 25 || vic == 26 || vic == 40 ||
+	    vic == 44 || vic == 45 || vic == 46 ||
+	    vic == 50 || vic == 51 || vic == 54 ||
+	    vic == 55 || vic == 58 || vic  == 59)
+		value |= R_V_BLANK_IN_OSC;
+	regmap_write(dp->regmap, DW_DP_VIDEO_CONFIG1, value);
+
+	/* Configure DW_DP_VIDEO_CONFIG2 register */
+	vblank = mode->vtotal - mode->vdisplay;
+	vactive = mode->vdisplay;
+	regmap_write(dp->regmap, DW_DP_VIDEO_CONFIG2,
+		     FIELD_PREP(VBLANK, vblank) | FIELD_PREP(VACTIVE, vactive));
+
+	/* Configure DW_DP_VIDEO_CONFIG3 register */
+	h_sync_width = mode->hsync_end - mode->hsync_start;
+	h_front_porch = mode->hsync_start - mode->hdisplay;
+	regmap_write(dp->regmap, DW_DP_VIDEO_CONFIG3,
+		     FIELD_PREP(H_SYNC_WIDTH, h_sync_width) |
+		     FIELD_PREP(H_FRONT_PORCH, h_front_porch));
+
+	/* Configure DW_DP_VIDEO_CONFIG4 register */
+	v_sync_width = mode->vsync_end - mode->vsync_start;
+	v_front_porch = mode->vsync_start - mode->vdisplay;
+	regmap_write(dp->regmap, DW_DP_VIDEO_CONFIG4,
+		     FIELD_PREP(V_SYNC_WIDTH, v_sync_width) |
+		     FIELD_PREP(V_FRONT_PORCH, v_front_porch));
+
+	/* Configure DW_DP_VIDEO_CONFIG5 register */
+	peak_stream_bandwidth = mode->clock * bpp / 8;
+	link_bandwidth = (link->rate / 1000) * link->lanes;
+	ts = peak_stream_bandwidth * 64 / link_bandwidth;
+	average_bytes_per_tu = ts / 1000;
+	average_bytes_per_tu_frac = ts / 100 - average_bytes_per_tu * 10;
+	if (dp->pixel_mode == DW_DP_MP_SINGLE_PIXEL) {
+		if (average_bytes_per_tu < 6)
+			init_threshold = 32;
+		else if (hblank <= 80 && color_format != DRM_COLOR_FORMAT_YCBCR420)
+			init_threshold = 12;
+		else if (hblank <= 40 && color_format == DRM_COLOR_FORMAT_YCBCR420)
+			init_threshold = 3;
+		else
+			init_threshold = 16;
+	} else {
+		u32 t1 = 0, t2 = 0, t3 = 0;
+
+		switch (bpc) {
+		case 6:
+			t1 = (4 * 1000 / 9) * link->lanes;
+			break;
+		case 8:
+			if (color_format == DRM_COLOR_FORMAT_YCBCR422) {
+				t1 = (1000 / 2) * link->lanes;
+			} else {
+				if (dp->pixel_mode == DW_DP_MP_DUAL_PIXEL)
+					t1 = (1000 / 3) * link->lanes;
+				else
+					t1 = (3000 / 16) * link->lanes;
+			}
+			break;
+		case 10:
+			if (color_format == DRM_COLOR_FORMAT_YCBCR422)
+				t1 = (2000 / 5) * link->lanes;
+			else
+				t1 = (4000 / 15) * link->lanes;
+			break;
+		case 12:
+			if (color_format == DRM_COLOR_FORMAT_YCBCR422) {
+				if (dp->pixel_mode == DW_DP_MP_DUAL_PIXEL)
+					t1 = (1000 / 6) * link->lanes;
+				else
+					t1 = (1000 / 3) * link->lanes;
+			} else {
+				t1 = (2000 / 9) * link->lanes;
+			}
+			break;
+		case 16:
+			if (color_format != DRM_COLOR_FORMAT_YCBCR422 &&
+			    dp->pixel_mode == DW_DP_MP_DUAL_PIXEL)
+				t1 = (1000 / 6) * link->lanes;
+			else
+				t1 = (1000 / 4) * link->lanes;
+			break;
+		default:
+			return -EINVAL;
+		}
+
+		if (color_format == DRM_COLOR_FORMAT_YCBCR420)
+			t2 = (link->rate / 4) * 1000 / (mode->clock / 2);
+		else
+			t2 = (link->rate / 4) * 1000 / mode->clock;
+
+		if (average_bytes_per_tu_frac)
+			t3 = average_bytes_per_tu + 1;
+		else
+			t3 = average_bytes_per_tu;
+		init_threshold = t1 * t2 * t3 / (1000 * 1000);
+		if (init_threshold <= 16 || average_bytes_per_tu < 10)
+			init_threshold = 40;
+	}
+
+	regmap_write(dp->regmap, DW_DP_VIDEO_CONFIG5,
+		     FIELD_PREP(INIT_THRESHOLD_HI, init_threshold >> 6) |
+		     FIELD_PREP(AVERAGE_BYTES_PER_TU_FRAC, average_bytes_per_tu_frac) |
+		     FIELD_PREP(INIT_THRESHOLD, init_threshold) |
+		     FIELD_PREP(AVERAGE_BYTES_PER_TU, average_bytes_per_tu));
+
+	/* Configure DW_DP_VIDEO_HBLANK_INTERVAL register */
+	hblank_interval = hblank * (link->rate / 4) / mode->clock;
+	regmap_write(dp->regmap, DW_DP_VIDEO_HBLANK_INTERVAL,
+		     FIELD_PREP(HBLANK_INTERVAL_EN, 1) |
+		     FIELD_PREP(HBLANK_INTERVAL, hblank_interval));
+
+	/* Video stream enable */
+	regmap_update_bits(dp->regmap, DW_DP_VSAMPLE_CTRL, VIDEO_STREAM_ENABLE,
+			   FIELD_PREP(VIDEO_STREAM_ENABLE, 1));
+
+	if (dw_dp_video_need_vsc_sdp(dp))
+		dw_dp_send_vsc_sdp(dp);
+
+	return 0;
+}
+
+static void dw_dp_hpd_init(struct dw_dp *dp)
+{
+	/* Enable all HPD interrupts */
+	regmap_update_bits(dp->regmap, DW_DP_HPD_INTERRUPT_ENABLE,
+			   HPD_UNPLUG_EN | HPD_PLUG_EN | HPD_IRQ_EN,
+			   FIELD_PREP(HPD_UNPLUG_EN, 1) |
+			   FIELD_PREP(HPD_PLUG_EN, 1) |
+			   FIELD_PREP(HPD_IRQ_EN, 1));
+
+	/* Enable all top-level interrupts */
+	regmap_update_bits(dp->regmap, DW_DP_GENERAL_INTERRUPT_ENABLE,
+			   HPD_EVENT_EN, FIELD_PREP(HPD_EVENT_EN, 1));
+}
+
+static void dw_dp_aux_init(struct dw_dp *dp)
+{
+	regmap_update_bits(dp->regmap, DW_DP_GENERAL_INTERRUPT_ENABLE,
+			   AUX_REPLY_EVENT_EN, FIELD_PREP(AUX_REPLY_EVENT_EN, 1));
+}
+
+static void dw_dp_init_hw(struct dw_dp *dp)
+{
+	regmap_update_bits(dp->regmap, DW_DP_CCTL, DEFAULT_FAST_LINK_TRAIN_EN,
+			   FIELD_PREP(DEFAULT_FAST_LINK_TRAIN_EN, 0));
+
+	dw_dp_hpd_init(dp);
+	dw_dp_aux_init(dp);
+}
+
+static int dw_dp_aux_write_data(struct dw_dp *dp, const u8 *buffer, size_t size)
+{
+	size_t i, j;
+
+	for (i = 0; i < DIV_ROUND_UP(size, 4); i++) {
+		size_t num = min_t(size_t, size - i * 4, 4);
+		u32 value = 0;
+
+		for (j = 0; j < num; j++)
+			value |= buffer[i * 4 + j] << (j * 8);
+
+		regmap_write(dp->regmap, DW_DP_AUX_DATA0 + i * 4, value);
+	}
+
+	return size;
+}
+
+static int dw_dp_aux_read_data(struct dw_dp *dp, u8 *buffer, size_t size)
+{
+	size_t i, j;
+
+	for (i = 0; i < DIV_ROUND_UP(size, 4); i++) {
+		size_t num = min_t(size_t, size - i * 4, 4);
+		u32 value;
+
+		regmap_read(dp->regmap, DW_DP_AUX_DATA0 + i * 4, &value);
+
+		for (j = 0; j < num; j++)
+			buffer[i * 4 + j] = value >> (j * 8);
+	}
+
+	return size;
+}
+
+static ssize_t dw_dp_aux_transfer(struct drm_dp_aux *aux,
+				  struct drm_dp_aux_msg *msg)
+{
+	struct dw_dp *dp = container_of(aux, struct dw_dp, aux);
+	unsigned long timeout = msecs_to_jiffies(10);
+	u32 status, value;
+	ssize_t ret = 0;
+
+	if (WARN_ON(msg->size > 16))
+		return -E2BIG;
+
+	switch (msg->request & ~DP_AUX_I2C_MOT) {
+	case DP_AUX_NATIVE_WRITE:
+	case DP_AUX_I2C_WRITE:
+	case DP_AUX_I2C_WRITE_STATUS_UPDATE:
+		ret = dw_dp_aux_write_data(dp, msg->buffer, msg->size);
+		if (ret < 0)
+			return ret;
+		break;
+	case DP_AUX_NATIVE_READ:
+	case DP_AUX_I2C_READ:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (msg->size > 0)
+		value = FIELD_PREP(AUX_LEN_REQ, msg->size - 1);
+	else
+		value = FIELD_PREP(I2C_ADDR_ONLY, 1);
+	value |= FIELD_PREP(AUX_CMD_TYPE, msg->request);
+	value |= FIELD_PREP(AUX_ADDR, msg->address);
+	regmap_write(dp->regmap, DW_DP_AUX_CMD, value);
+
+	status = wait_for_completion_timeout(&dp->complete, timeout);
+	if (!status) {
+		dev_err(dp->dev, "timeout waiting for AUX reply\n");
+		return -ETIMEDOUT;
+	}
+
+	regmap_read(dp->regmap, DW_DP_AUX_STATUS, &value);
+	if (value & AUX_TIMEOUT)
+		return -ETIMEDOUT;
+
+	msg->reply = FIELD_GET(AUX_STATUS, value);
+
+	if (msg->size > 0 && msg->reply == DP_AUX_NATIVE_REPLY_ACK) {
+		if (msg->request & DP_AUX_I2C_READ) {
+			size_t count = FIELD_GET(AUX_BYTES_READ, value) - 1;
+
+			if (count != msg->size)
+				return -EBUSY;
+
+			ret = dw_dp_aux_read_data(dp, msg->buffer, count);
+			if (ret < 0)
+				return ret;
+		}
+	}
+
+	return ret;
+}
+
+/*
+ * Limits for the video timing for DP:
+ * 1. the hfp should be 2 pixels aligned;
+ * 2. the minimum hsync should be 9 pixel;
+ * 3. the minimum hbp should be 16 pixel;
+ */
+static int dw_dp_bridge_atomic_check(struct drm_bridge *bridge,
+				     struct drm_bridge_state *bridge_state,
+				     struct drm_crtc_state *crtc_state,
+				     struct drm_connector_state *conn_state)
+{
+	struct drm_display_mode *adjusted_mode = &crtc_state->adjusted_mode;
+	struct dw_dp *dp = bridge_to_dp(bridge);
+	struct dw_dp_bridge_state *state;
+	const struct dw_dp_output_format *fmt;
+	struct drm_display_mode *mode;
+	int min_hbp = 16;
+	int min_hsync = 9;
+
+	state = to_dw_dp_bridge_state(bridge_state);
+	mode = &state->mode;
+
+	fmt = dw_dp_get_output_format(bridge_state->output_bus_cfg.format);
+	if (!fmt)
+		return -EINVAL;
+
+	state->video_mapping = fmt->video_mapping;
+	state->color_format = fmt->color_format;
+	state->bpc = fmt->bpc;
+	state->bpp = fmt->bpp;
+
+	if ((adjusted_mode->hsync_start - adjusted_mode->hdisplay) & 0x1) {
+		adjusted_mode->hsync_start += 1;
+		dev_warn(dp->dev, "hfp is not 2 pixeel aligned, fixup to aligned hfp\n");
+	}
+
+	if (adjusted_mode->hsync_end - adjusted_mode->hsync_start < min_hsync) {
+		adjusted_mode->hsync_end = adjusted_mode->hsync_start + min_hsync;
+		dev_warn(dp->dev, "hsync is too narrow, fixup to min hsync:%d\n", min_hsync);
+	}
+
+	if (adjusted_mode->htotal - adjusted_mode->hsync_end < min_hbp) {
+		adjusted_mode->htotal = adjusted_mode->hsync_end + min_hbp;
+		dev_warn(dp->dev, "hbp is too narrow, fixup to min hbp:%d\n", min_hbp);
+	}
+
+	drm_mode_copy(mode, adjusted_mode);
+
+	return 0;
+}
+
+static enum drm_mode_status dw_dp_bridge_mode_valid(struct drm_bridge *bridge,
+						    const struct drm_display_info *info,
+						    const struct drm_display_mode *mode)
+{
+	struct dw_dp *dp = bridge_to_dp(bridge);
+	struct dw_dp_link *link = &dp->link;
+	u32 min_bpp;
+
+	if (info->color_formats & DRM_COLOR_FORMAT_YCBCR420 &&
+	    link->vsc_sdp_supported &&
+	    (drm_mode_is_420_only(info, mode) || drm_mode_is_420_also(info, mode)))
+		min_bpp = 12;
+	else if (info->color_formats & DRM_COLOR_FORMAT_YCBCR422)
+		min_bpp = 16;
+	else if (info->color_formats & DRM_COLOR_FORMAT_RGB444)
+		min_bpp = 18;
+	else
+		min_bpp = 24;
+
+	if (!link->vsc_sdp_supported &&
+	    drm_mode_is_420_only(info, mode))
+		return MODE_NO_420;
+
+	if (!dw_dp_bandwidth_ok(dp, mode, min_bpp, link->lanes, link->rate))
+		return MODE_CLOCK_HIGH;
+
+	return MODE_OK;
+}
+
+static bool dw_dp_needs_link_retrain(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+	u8 link_status[DP_LINK_STATUS_SIZE];
+
+	if (!dw_dp_link_train_valid(&link->train))
+		return false;
+
+	if (drm_dp_dpcd_read_link_status(&dp->aux, link_status) < 0)
+		return false;
+
+	/* Retrain if Channel EQ or CR not ok */
+	return !drm_dp_channel_eq_ok(link_status, dp->link.lanes);
+}
+
+static void dw_dp_link_disable(struct dw_dp *dp)
+{
+	struct dw_dp_link *link = &dp->link;
+
+	if (dw_dp_hpd_detect(dp))
+		drm_dp_link_power_down(&dp->aux, dp->link.revision);
+
+	dw_dp_phy_xmit_enable(dp, 0);
+
+	phy_power_off(dp->phy);
+
+	link->train.clock_recovered = false;
+	link->train.channel_equalized = false;
+}
+
+static int dw_dp_link_enable(struct dw_dp *dp)
+{
+	int ret;
+
+	ret = phy_power_on(dp->phy);
+	if (ret)
+		return ret;
+
+	ret = drm_dp_link_power_up(&dp->aux, dp->link.revision);
+	if (ret < 0)
+		return ret;
+
+	ret = dw_dp_link_train(dp);
+
+	return ret;
+}
+
+static void dw_dp_bridge_atomic_enable(struct drm_bridge *bridge,
+				       struct drm_atomic_state *state)
+{
+	struct dw_dp *dp = bridge_to_dp(bridge);
+	struct drm_connector *connector;
+	struct drm_connector_state *conn_state;
+	int ret;
+
+	connector = drm_atomic_get_new_connector_for_encoder(state, bridge->encoder);
+	if (!connector) {
+		dev_err(dp->dev, "failed to get connector\n");
+		return;
+	}
+
+	conn_state = drm_atomic_get_new_connector_state(state, connector);
+	if (!conn_state) {
+		dev_err(dp->dev, "failed to get connector state\n");
+		return;
+	}
+
+	set_bit(0, dp->sdp_reg_bank);
+
+	ret = dw_dp_link_enable(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "failed to enable link: %d\n", ret);
+		return;
+	}
+
+	ret = dw_dp_video_enable(dp);
+	if (ret < 0) {
+		dev_err(dp->dev, "failed to enable video: %d\n", ret);
+		return;
+	}
+}
+
+static void dw_dp_reset(struct dw_dp *dp)
+{
+	int val;
+
+	disable_irq(dp->irq);
+	regmap_update_bits(dp->regmap, DW_DP_SOFT_RESET_CTRL, CONTROLLER_RESET,
+			   FIELD_PREP(CONTROLLER_RESET, 1));
+	usleep_range(10, 20);
+	regmap_update_bits(dp->regmap, DW_DP_SOFT_RESET_CTRL, CONTROLLER_RESET,
+			   FIELD_PREP(CONTROLLER_RESET, 0));
+
+	dw_dp_init_hw(dp);
+	regmap_read_poll_timeout(dp->regmap, DW_DP_HPD_STATUS, val,
+				 FIELD_GET(HPD_HOT_PLUG, val), 200, 200000);
+	regmap_write(dp->regmap, DW_DP_HPD_STATUS, HPD_HOT_PLUG);
+	enable_irq(dp->irq);
+}
+
+static void dw_dp_bridge_atomic_disable(struct drm_bridge *bridge,
+					struct drm_atomic_state *state)
+{
+	struct dw_dp *dp = bridge_to_dp(bridge);
+
+	dw_dp_video_disable(dp);
+	dw_dp_link_disable(dp);
+	bitmap_zero(dp->sdp_reg_bank, SDP_REG_BANK_SIZE);
+	dw_dp_reset(dp);
+}
+
+static bool dw_dp_hpd_detect_link(struct dw_dp *dp, struct drm_connector *connector)
+{
+	int ret;
+
+	ret = phy_power_on(dp->phy);
+	if (ret < 0)
+		return false;
+	ret = dw_dp_link_parse(dp, connector);
+	phy_power_off(dp->phy);
+
+	return !ret;
+}
+
+static enum drm_connector_status dw_dp_bridge_detect(struct drm_bridge *bridge,
+						     struct drm_connector *connector)
+{
+	struct dw_dp *dp = bridge_to_dp(bridge);
+
+	if (!dw_dp_hpd_detect(dp))
+		return connector_status_disconnected;
+
+	if (!dw_dp_hpd_detect_link(dp, connector))
+		return connector_status_disconnected;
+
+	return connector_status_connected;
+}
+
+static const struct drm_edid *dw_dp_bridge_edid_read(struct drm_bridge *bridge,
+						     struct drm_connector *connector)
+{
+	struct dw_dp *dp = bridge_to_dp(bridge);
+	const struct drm_edid *edid;
+	int ret;
+
+	ret = phy_power_on(dp->phy);
+	if (ret)
+		return NULL;
+
+	edid = drm_edid_read_ddc(connector, &dp->aux.ddc);
+
+	phy_power_off(dp->phy);
+
+	return edid;
+}
+
+static u32 *dw_dp_bridge_atomic_get_output_bus_fmts(struct drm_bridge *bridge,
+						    struct drm_bridge_state *bridge_state,
+						    struct drm_crtc_state *crtc_state,
+						    struct drm_connector_state *conn_state,
+						    unsigned int *num_output_fmts)
+{
+	struct dw_dp *dp = bridge_to_dp(bridge);
+	struct dw_dp_link *link = &dp->link;
+	struct drm_display_info *di = &conn_state->connector->display_info;
+	struct drm_display_mode mode = crtc_state->mode;
+	const struct dw_dp_output_format *fmt;
+	u32 i, j = 0;
+	u32 *output_fmts;
+
+	*num_output_fmts = 0;
+
+	output_fmts = kcalloc(ARRAY_SIZE(dw_dp_output_formats), sizeof(*output_fmts), GFP_KERNEL);
+	if (!output_fmts)
+		return NULL;
+
+	for (i = 0; i < ARRAY_SIZE(dw_dp_output_formats); i++) {
+		fmt = &dw_dp_output_formats[i];
+
+		if (fmt->bpc > conn_state->max_bpc)
+			continue;
+
+		if (!(fmt->color_format & di->color_formats))
+			continue;
+
+		if (fmt->color_format == DRM_COLOR_FORMAT_YCBCR420 &&
+		    !link->vsc_sdp_supported)
+			continue;
+
+		if (fmt->color_format != DRM_COLOR_FORMAT_YCBCR420 &&
+		    drm_mode_is_420_only(di, &mode))
+			continue;
+
+		if (!dw_dp_bandwidth_ok(dp, &mode, fmt->bpp, link->lanes, link->rate))
+			continue;
+
+		output_fmts[j++] = fmt->bus_format;
+	}
+
+	*num_output_fmts = j;
+
+	return output_fmts;
+}
+
+static struct drm_bridge_state *dw_dp_bridge_atomic_duplicate_state(struct drm_bridge *bridge)
+{
+	struct dw_dp_bridge_state *state;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return NULL;
+
+	__drm_atomic_helper_bridge_duplicate_state(bridge, &state->base);
+
+	return &state->base;
+}
+
+static const struct drm_bridge_funcs dw_dp_bridge_funcs = {
+	.atomic_duplicate_state = dw_dp_bridge_atomic_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_bridge_destroy_state,
+	.atomic_reset = drm_atomic_helper_bridge_reset,
+	.atomic_get_input_bus_fmts = drm_atomic_helper_bridge_propagate_bus_fmt,
+	.atomic_get_output_bus_fmts = dw_dp_bridge_atomic_get_output_bus_fmts,
+	.atomic_check = dw_dp_bridge_atomic_check,
+	.mode_valid = dw_dp_bridge_mode_valid,
+	.atomic_enable = dw_dp_bridge_atomic_enable,
+	.atomic_disable = dw_dp_bridge_atomic_disable,
+	.detect = dw_dp_bridge_detect,
+	.edid_read = dw_dp_bridge_edid_read,
+};
+
+static int dw_dp_link_retrain(struct dw_dp *dp)
+{
+	struct drm_device *dev = dp->bridge.dev;
+	struct drm_modeset_acquire_ctx ctx;
+	int ret;
+
+	if (!dw_dp_needs_link_retrain(dp))
+		return 0;
+
+	dev_dbg(dp->dev, "Retraining link\n");
+
+	drm_modeset_acquire_init(&ctx, 0);
+	for (;;) {
+		ret = drm_modeset_lock(&dev->mode_config.connection_mutex, &ctx);
+		if (ret != -EDEADLK)
+			break;
+
+		drm_modeset_backoff(&ctx);
+	}
+
+	if (!ret)
+		ret = dw_dp_link_train(dp);
+
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+
+	return ret;
+}
+
+static void dw_dp_hpd_work(struct work_struct *work)
+{
+	struct dw_dp *dp = container_of(work, struct dw_dp, hpd_work);
+	bool long_hpd;
+	int ret;
+
+	mutex_lock(&dp->irq_lock);
+	long_hpd = dp->hotplug.long_hpd;
+	mutex_unlock(&dp->irq_lock);
+
+	dev_dbg(dp->dev, "[drm] Get hpd irq - %s\n", long_hpd ? "long" : "short");
+
+	if (!long_hpd) {
+		if (dw_dp_needs_link_retrain(dp)) {
+			ret = dw_dp_link_retrain(dp);
+			if (ret)
+				dev_warn(dp->dev, "Retrain link failed\n");
+		}
+	} else {
+		drm_helper_hpd_irq_event(dp->bridge.dev);
+	}
+}
+
+static void dw_dp_handle_hpd_event(struct dw_dp *dp)
+{
+	u32 value;
+
+	mutex_lock(&dp->irq_lock);
+	regmap_read(dp->regmap, DW_DP_HPD_STATUS, &value);
+
+	if (value & HPD_IRQ) {
+		dev_dbg(dp->dev, "IRQ from the HPD\n");
+		dp->hotplug.long_hpd = false;
+		regmap_write(dp->regmap, DW_DP_HPD_STATUS, HPD_IRQ);
+	}
+
+	if (value & HPD_HOT_PLUG) {
+		dev_dbg(dp->dev, "Hot plug detected\n");
+		dp->hotplug.long_hpd = true;
+		regmap_write(dp->regmap, DW_DP_HPD_STATUS, HPD_HOT_PLUG);
+	}
+
+	if (value & HPD_HOT_UNPLUG) {
+		dev_dbg(dp->dev, "Unplug detected\n");
+		dp->hotplug.long_hpd = true;
+		regmap_write(dp->regmap, DW_DP_HPD_STATUS, HPD_HOT_UNPLUG);
+	}
+	mutex_unlock(&dp->irq_lock);
+
+	schedule_work(&dp->hpd_work);
+}
+
+static irqreturn_t dw_dp_irq(int irq, void *data)
+{
+	struct dw_dp *dp = data;
+	u32 value;
+
+	regmap_read(dp->regmap, DW_DP_GENERAL_INTERRUPT, &value);
+	if (!value)
+		return IRQ_NONE;
+
+	if (value & HPD_EVENT)
+		dw_dp_handle_hpd_event(dp);
+
+	if (value & AUX_REPLY_EVENT) {
+		regmap_write(dp->regmap, DW_DP_GENERAL_INTERRUPT, AUX_REPLY_EVENT);
+		complete(&dp->complete);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static const struct regmap_range dw_dp_readable_ranges[] = {
+	regmap_reg_range(DW_DP_VERSION_NUMBER, DW_DP_ID),
+	regmap_reg_range(DW_DP_CONFIG_REG1, DW_DP_CONFIG_REG3),
+	regmap_reg_range(DW_DP_CCTL, DW_DP_SOFT_RESET_CTRL),
+	regmap_reg_range(DW_DP_VSAMPLE_CTRL, DW_DP_VIDEO_HBLANK_INTERVAL),
+	regmap_reg_range(DW_DP_AUD_CONFIG1, DW_DP_AUD_CONFIG1),
+	regmap_reg_range(DW_DP_SDP_VERTICAL_CTRL, DW_DP_SDP_STATUS_EN),
+	regmap_reg_range(DW_DP_PHYIF_CTRL, DW_DP_PHYIF_PWRDOWN_CTRL),
+	regmap_reg_range(DW_DP_AUX_CMD, DW_DP_AUX_DATA3),
+	regmap_reg_range(DW_DP_GENERAL_INTERRUPT, DW_DP_HPD_INTERRUPT_ENABLE),
+};
+
+static const struct regmap_access_table dw_dp_readable_table = {
+	.yes_ranges     = dw_dp_readable_ranges,
+	.n_yes_ranges   = ARRAY_SIZE(dw_dp_readable_ranges),
+};
+
+static const struct regmap_config dw_dp_regmap_config = {
+	.reg_bits = 32,
+	.reg_stride = 4,
+	.val_bits = 32,
+	.fast_io = true,
+	.max_register = DW_DP_MAX_REGISTER,
+	.rd_table = &dw_dp_readable_table,
+};
+
+static void dw_dp_phy_exit(void *data)
+{
+	struct dw_dp *dp = data;
+
+	phy_exit(dp->phy);
+}
+
+struct dw_dp *dw_dp_bind(struct device *dev, struct drm_encoder *encoder,
+			 const struct dw_dp_plat_data *plat_data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct dw_dp *dp;
+	struct drm_bridge *bridge;
+	void __iomem *res;
+	int ret;
+
+	dp = devm_kzalloc(dev, sizeof(*dp), GFP_KERNEL);
+	if (!dp)
+		return ERR_PTR(-ENOMEM);
+
+	dp = devm_drm_bridge_alloc(dev, struct dw_dp, bridge, &dw_dp_bridge_funcs);
+	if (IS_ERR(dp))
+		return ERR_CAST(dp);
+
+	dp->dev = dev;
+	dp->pixel_mode = DW_DP_MP_QUAD_PIXEL;
+
+	dp->plat_data.max_link_rate = plat_data->max_link_rate;
+	bridge = &dp->bridge;
+	mutex_init(&dp->irq_lock);
+	INIT_WORK(&dp->hpd_work, dw_dp_hpd_work);
+	init_completion(&dp->complete);
+
+	res = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(res))
+		return ERR_CAST(res);
+
+	dp->regmap = devm_regmap_init_mmio(dev, res, &dw_dp_regmap_config);
+	if (IS_ERR(dp->regmap)) {
+		dev_err_probe(dev, PTR_ERR(dp->regmap), "failed to create regmap\n");
+		return ERR_CAST(dp->regmap);
+	}
+
+	dp->phy = devm_of_phy_get(dev, dev->of_node, NULL);
+	if (IS_ERR(dp->phy)) {
+		dev_err_probe(dev, PTR_ERR(dp->phy), "failed to get phy\n");
+		return ERR_CAST(dp->phy);
+	}
+
+	dp->apb_clk = devm_clk_get_enabled(dev, "apb");
+	if (IS_ERR(dp->apb_clk)) {
+		dev_err_probe(dev, PTR_ERR(dp->apb_clk), "failed to get apb clock\n");
+		return ERR_CAST(dp->apb_clk);
+	}
+
+	dp->aux_clk = devm_clk_get_enabled(dev, "aux");
+	if (IS_ERR(dp->aux_clk)) {
+		dev_err_probe(dev, PTR_ERR(dp->aux_clk), "failed to get aux clock\n");
+		return ERR_CAST(dp->aux_clk);
+	}
+
+	dp->i2s_clk = devm_clk_get(dev, "i2s");
+	if (IS_ERR(dp->i2s_clk)) {
+		dev_err_probe(dev, PTR_ERR(dp->i2s_clk), "failed to get i2s clock\n");
+		return ERR_CAST(dp->i2s_clk);
+	}
+
+	dp->spdif_clk = devm_clk_get(dev, "spdif");
+	if (IS_ERR(dp->spdif_clk)) {
+		dev_err_probe(dev, PTR_ERR(dp->spdif_clk), "failed to get spdif clock\n");
+		return ERR_CAST(dp->spdif_clk);
+	}
+
+	dp->hdcp_clk = devm_clk_get(dev, "hdcp");
+	if (IS_ERR(dp->hdcp_clk)) {
+		dev_err_probe(dev, PTR_ERR(dp->hdcp_clk), "failed to get hdcp clock\n");
+		return ERR_CAST(dp->hdcp_clk);
+	}
+
+	dp->rstc = devm_reset_control_get(dev, NULL);
+	if (IS_ERR(dp->rstc)) {
+		dev_err_probe(dev, PTR_ERR(dp->rstc), "failed to get reset control\n");
+		return ERR_CAST(dp->rstc);
+	}
+
+	bridge->of_node = dev->of_node;
+	bridge->ops = DRM_BRIDGE_OP_DETECT | DRM_BRIDGE_OP_EDID | DRM_BRIDGE_OP_HPD;
+	bridge->type = DRM_MODE_CONNECTOR_DisplayPort;
+	bridge->ycbcr_420_allowed = true;
+
+	dp->aux.dev = dev;
+	dp->aux.drm_dev = encoder->dev;
+	dp->aux.name = dev_name(dev);
+	dp->aux.transfer = dw_dp_aux_transfer;
+	ret = drm_dp_aux_register(&dp->aux);
+	if (ret) {
+		dev_err_probe(dev, ret, "Aux register failed\n");
+		return ERR_PTR(ret);
+	}
+
+	ret = drm_bridge_attach(encoder, bridge, NULL, DRM_BRIDGE_ATTACH_NO_CONNECTOR);
+	if (ret)
+		dev_err_probe(dev, ret, "Failed to attach bridge\n");
+
+	dw_dp_init_hw(dp);
+
+	ret = phy_init(dp->phy);
+	if (ret) {
+		dev_err_probe(dev, ret, "phy init failed\n");
+		return ERR_PTR(ret);
+	}
+
+	ret = devm_add_action_or_reset(dev, dw_dp_phy_exit, dp);
+	if (ret)
+		return ERR_PTR(ret);
+
+	dp->irq = platform_get_irq(pdev, 0);
+	if (dp->irq < 0)
+		return ERR_PTR(ret);
+
+	ret = devm_request_threaded_irq(dev, dp->irq, NULL, dw_dp_irq,
+					IRQF_ONESHOT, dev_name(dev), dp);
+	if (ret) {
+		dev_err_probe(dev, ret, "failed to request irq\n");
+		return ERR_PTR(ret);
+	}
+
+	return dp;
+}
+EXPORT_SYMBOL_GPL(dw_dp_bind);
+
+MODULE_AUTHOR("Andy Yan <andyshrk@163.com>");
+MODULE_DESCRIPTION("DW DP Core Library");
+MODULE_LICENSE("GPL");
diff --git a/include/drm/bridge/dw_dp.h b/include/drm/bridge/dw_dp.h
new file mode 100644
index 000000000000..d05df49fd884
--- /dev/null
+++ b/include/drm/bridge/dw_dp.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright (c) 2025 Rockchip Electronics Co., Ltd.
+ */
+
+#ifndef __DW_DP__
+#define __DW_DP__
+
+#include <linux/device.h>
+
+struct drm_encoder;
+struct dw_dp;
+
+struct dw_dp_plat_data {
+	u32 max_link_rate;
+};
+
+struct dw_dp *dw_dp_bind(struct device *dev, struct drm_encoder *encoder,
+			 const struct dw_dp_plat_data *plat_data);
+#endif /* __DW_DP__ */

From d68ba7bac9555d05e2f5b310c898b2a5c7eff174 Mon Sep 17 00:00:00 2001
From: Andy Yan <andy.yan@rock-chips.com>
Date: Fri, 22 Aug 2025 14:39:47 +0800
Subject: [PATCH 0723/1455] drm/rockchip: Add RK3588 DPTX output support

Add driver extension for Synopsys DesignWare DPTX IP used
on Rockchip RK3588 SoC.

Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Acked-by: Dmitry Baryshkov <lumag@kernel.org>
Tested-by: Nicolas Frattaroli <nicolas.frattaroli@collabora.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250822063959.692098-4-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/rockchip/Kconfig            |   9 ++
 drivers/gpu/drm/rockchip/Makefile           |   1 +
 drivers/gpu/drm/rockchip/dw_dp-rockchip.c   | 150 ++++++++++++++++++++
 drivers/gpu/drm/rockchip/rockchip_drm_drv.c |   1 +
 drivers/gpu/drm/rockchip/rockchip_drm_drv.h |   1 +
 5 files changed, 162 insertions(+)
 create mode 100644 drivers/gpu/drm/rockchip/dw_dp-rockchip.c

diff --git a/drivers/gpu/drm/rockchip/Kconfig b/drivers/gpu/drm/rockchip/Kconfig
index ab525668939a..616c0bc1ccee 100644
--- a/drivers/gpu/drm/rockchip/Kconfig
+++ b/drivers/gpu/drm/rockchip/Kconfig
@@ -10,6 +10,7 @@ config DRM_ROCKCHIP
 	select VIDEOMODE_HELPERS
 	select DRM_ANALOGIX_DP if ROCKCHIP_ANALOGIX_DP
 	select DRM_DISPLAY_DP_AUX_BUS if ROCKCHIP_ANALOGIX_DP
+	select DRM_DW_DP if ROCKCHIP_DW_DP
 	select DRM_DW_HDMI if ROCKCHIP_DW_HDMI
 	select DRM_DW_HDMI_QP if ROCKCHIP_DW_HDMI_QP
 	select DRM_DW_MIPI_DSI if ROCKCHIP_DW_MIPI_DSI
@@ -60,6 +61,14 @@ config ROCKCHIP_CDN_DP
 	  RK3399 based SoC, you should select this
 	  option.
 
+config ROCKCHIP_DW_DP
+	bool "Rockchip specific extensions for Synopsys DW DP"
+	help
+	  This selects support for Rockchip SoC specific extensions
+	  to enable Synopsys DesignWare Cores based DisplayPort transmit
+	  controller support on Rockchip SoC, If you want to enable DP on
+	  rk3588 based SoC, you should select this option.
+
 config ROCKCHIP_DW_HDMI
 	bool "Rockchip specific extensions for Synopsys DW HDMI"
 	help
diff --git a/drivers/gpu/drm/rockchip/Makefile b/drivers/gpu/drm/rockchip/Makefile
index 2b867cebbc12..097f062399c7 100644
--- a/drivers/gpu/drm/rockchip/Makefile
+++ b/drivers/gpu/drm/rockchip/Makefile
@@ -14,6 +14,7 @@ rockchipdrm-$(CONFIG_ROCKCHIP_DW_HDMI) += dw_hdmi-rockchip.o
 rockchipdrm-$(CONFIG_ROCKCHIP_DW_HDMI_QP) += dw_hdmi_qp-rockchip.o
 rockchipdrm-$(CONFIG_ROCKCHIP_DW_MIPI_DSI) += dw-mipi-dsi-rockchip.o
 rockchipdrm-$(CONFIG_ROCKCHIP_DW_MIPI_DSI2) += dw-mipi-dsi2-rockchip.o
+rockchipdrm-$(CONFIG_ROCKCHIP_DW_DP) += dw_dp-rockchip.o
 rockchipdrm-$(CONFIG_ROCKCHIP_INNO_HDMI) += inno_hdmi.o
 rockchipdrm-$(CONFIG_ROCKCHIP_LVDS) += rockchip_lvds.o
 rockchipdrm-$(CONFIG_ROCKCHIP_RGB) += rockchip_rgb.o
diff --git a/drivers/gpu/drm/rockchip/dw_dp-rockchip.c b/drivers/gpu/drm/rockchip/dw_dp-rockchip.c
new file mode 100644
index 000000000000..25ab4e46301e
--- /dev/null
+++ b/drivers/gpu/drm/rockchip/dw_dp-rockchip.c
@@ -0,0 +1,150 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2020 Rockchip Electronics Co., Ltd.
+ *
+ * Author: Zhang Yubing <yubing.zhang@rock-chips.com>
+ * Author: Andy Yan <andy.yan@rock-chips.com>
+ */
+
+#include <linux/component.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <drm/bridge/dw_dp.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_bridge.h>
+#include <drm/drm_bridge_connector.h>
+#include <drm/drm_of.h>
+#include <drm/drm_print.h>
+#include <drm/drm_probe_helper.h>
+#include <drm/drm_simple_kms_helper.h>
+
+#include <linux/media-bus-format.h>
+#include <linux/videodev2.h>
+
+#include "rockchip_drm_drv.h"
+#include "rockchip_drm_vop.h"
+
+struct rockchip_dw_dp {
+	struct dw_dp *base;
+	struct device *dev;
+	struct rockchip_encoder encoder;
+};
+
+static int dw_dp_encoder_atomic_check(struct drm_encoder *encoder,
+				      struct drm_crtc_state *crtc_state,
+				      struct drm_connector_state *conn_state)
+{
+	struct rockchip_crtc_state *s = to_rockchip_crtc_state(crtc_state);
+	struct drm_atomic_state *state = conn_state->state;
+	struct drm_display_info *di = &conn_state->connector->display_info;
+	struct drm_bridge *bridge  = drm_bridge_chain_get_first_bridge(encoder);
+	struct drm_bridge_state *bridge_state = drm_atomic_get_new_bridge_state(state, bridge);
+	u32 bus_format = bridge_state->input_bus_cfg.format;
+
+	switch (bus_format) {
+	case MEDIA_BUS_FMT_UYYVYY10_0_5X30:
+	case MEDIA_BUS_FMT_UYYVYY8_0_5X24:
+		s->output_mode = ROCKCHIP_OUT_MODE_YUV420;
+		break;
+	case MEDIA_BUS_FMT_YUYV10_1X20:
+	case MEDIA_BUS_FMT_YUYV8_1X16:
+		s->output_mode = ROCKCHIP_OUT_MODE_S888_DUMMY;
+		break;
+	case MEDIA_BUS_FMT_RGB101010_1X30:
+	case MEDIA_BUS_FMT_RGB888_1X24:
+	case MEDIA_BUS_FMT_RGB666_1X24_CPADHI:
+	case MEDIA_BUS_FMT_YUV10_1X30:
+	case MEDIA_BUS_FMT_YUV8_1X24:
+	default:
+		s->output_mode = ROCKCHIP_OUT_MODE_AAAA;
+		break;
+	}
+
+	s->output_type = DRM_MODE_CONNECTOR_DisplayPort;
+	s->bus_format = bus_format;
+	s->bus_flags = di->bus_flags;
+	s->color_space = V4L2_COLORSPACE_DEFAULT;
+
+	return 0;
+}
+
+static const struct drm_encoder_helper_funcs dw_dp_encoder_helper_funcs = {
+	.atomic_check		= dw_dp_encoder_atomic_check,
+};
+
+static int dw_dp_rockchip_bind(struct device *dev, struct device *master, void *data)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct dw_dp_plat_data plat_data;
+	struct drm_device *drm_dev = data;
+	struct rockchip_dw_dp *dp;
+	struct drm_encoder *encoder;
+	struct drm_connector *connector;
+	int ret;
+
+	dp = devm_kzalloc(dev, sizeof(*dp), GFP_KERNEL);
+	if (!dp)
+		return -ENOMEM;
+
+	dp->dev = dev;
+	platform_set_drvdata(pdev, dp);
+
+	plat_data.max_link_rate = 810000;
+	encoder = &dp->encoder.encoder;
+	encoder->possible_crtcs = drm_of_find_possible_crtcs(drm_dev, dev->of_node);
+	rockchip_drm_encoder_set_crtc_endpoint_id(&dp->encoder, dev->of_node, 0, 0);
+
+	ret = drmm_encoder_init(drm_dev, encoder, NULL, DRM_MODE_ENCODER_TMDS, NULL);
+	if (ret)
+		return ret;
+	drm_encoder_helper_add(encoder, &dw_dp_encoder_helper_funcs);
+
+	dp->base = dw_dp_bind(dev, encoder, &plat_data);
+	if (IS_ERR(dp->base)) {
+		ret = PTR_ERR(dp->base);
+		return ret;
+	}
+
+	connector = drm_bridge_connector_init(drm_dev, encoder);
+	if (IS_ERR(connector)) {
+		ret = PTR_ERR(connector);
+		return dev_err_probe(dev, ret, "Failed to init bridge connector");
+	}
+
+	drm_connector_attach_encoder(connector, encoder);
+
+	return 0;
+}
+
+static const struct component_ops dw_dp_rockchip_component_ops = {
+	.bind = dw_dp_rockchip_bind,
+};
+
+static int dw_dp_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+
+	return component_add(dev, &dw_dp_rockchip_component_ops);
+}
+
+static void dw_dp_remove(struct platform_device *pdev)
+{
+	struct rockchip_dw_dp *dp = platform_get_drvdata(pdev);
+
+	component_del(dp->dev, &dw_dp_rockchip_component_ops);
+}
+
+static const struct of_device_id dw_dp_of_match[] = {
+	{ .compatible = "rockchip,rk3588-dp", },
+	{}
+};
+MODULE_DEVICE_TABLE(of, dw_dp_of_match);
+
+struct platform_driver dw_dp_driver = {
+	.probe	= dw_dp_probe,
+	.remove = dw_dp_remove,
+	.driver = {
+		.name = "dw-dp",
+		.of_match_table = dw_dp_of_match,
+	},
+};
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index 180fad5d49ad..eb77bde9f628 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -529,6 +529,7 @@ static int __init rockchip_drm_init(void)
 	ADD_ROCKCHIP_SUB_DRIVER(rockchip_dp_driver,
 				CONFIG_ROCKCHIP_ANALOGIX_DP);
 	ADD_ROCKCHIP_SUB_DRIVER(cdn_dp_driver, CONFIG_ROCKCHIP_CDN_DP);
+	ADD_ROCKCHIP_SUB_DRIVER(dw_dp_driver, CONFIG_ROCKCHIP_DW_DP);
 	ADD_ROCKCHIP_SUB_DRIVER(dw_hdmi_rockchip_pltfm_driver,
 				CONFIG_ROCKCHIP_DW_HDMI);
 	ADD_ROCKCHIP_SUB_DRIVER(dw_hdmi_qp_rockchip_pltfm_driver,
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
index c183e82a42a5..2e86ad00979c 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
@@ -87,6 +87,7 @@ int rockchip_drm_encoder_set_crtc_endpoint_id(struct rockchip_encoder *rencoder,
 					      struct device_node *np, int port, int reg);
 int rockchip_drm_endpoint_is_subdriver(struct device_node *ep);
 extern struct platform_driver cdn_dp_driver;
+extern struct platform_driver dw_dp_driver;
 extern struct platform_driver dw_hdmi_rockchip_pltfm_driver;
 extern struct platform_driver dw_hdmi_qp_rockchip_pltfm_driver;
 extern struct platform_driver dw_mipi_dsi_rockchip_driver;

From 0b6da6d3a8d5391ff9da2358ac5afd8b7badd943 Mon Sep 17 00:00:00 2001
From: Andy Yan <andy.yan@rock-chips.com>
Date: Fri, 22 Aug 2025 14:39:48 +0800
Subject: [PATCH 0724/1455] MAINTAINERS: Add entry for DW DPTX Controller
 bridge

Add an entry for the DW DPTX Controller bridge driver.

Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20250822063959.692098-5-andyshrk@163.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 MAINTAINERS | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index cfa28b3470ab..0ae7d13c2b3c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7999,6 +7999,14 @@ S:	Maintained
 F:	Documentation/devicetree/bindings/display/panel/synaptics,r63353.yaml
 F:	drivers/gpu/drm/panel/panel-synaptics-r63353.c
 
+DRM DRIVER FOR SYNOPSYS DESIGNWARE DISPLAYPORT BRIDGE
+M:	Andy Yan <andy.yan@rock-chips.com>
+S:	Maintained
+T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
+F:	Documentation/devicetree/bindings/display/rockchip/rockchip,dw-dp.yaml
+F:	drivers/gpu/drm/bridge/synopsys/dw-dp.c
+F:	include/drm/bridge/dw_dp.h
+
 DRM DRIVER FOR TI DLPC3433 MIPI DSI TO DMD BRIDGE
 M:	Jagan Teki <jagan@amarulasolutions.com>
 S:	Maintained

From 09f90256e8902793f594517ef440698585eb3595 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Tue, 26 Aug 2025 13:07:37 +0900
Subject: [PATCH 0725/1455] rust: transmute: add `from_bytes_copy` method to
 `FromBytes` trait

`FromBytes::from_bytes` comes with a few practical limitations:

- It requires the bytes slice to have the same alignment as the returned
  type, which might not be guaranteed in the case of a byte stream,
- It returns a reference, requiring the returned type to implement
  `Clone` if one wants to keep the value for longer than the lifetime of
  the slice.

To overcome these when needed, add a `from_bytes_copy` with a default
implementation in the trait. `from_bytes_copy` returns an owned value
that is populated using an unaligned read, removing the lifetime
constraint and making it usable even on non-aligned byte slices.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Acked-by: Miguel Ojeda <ojeda@kernel.org>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Link: https://lore.kernel.org/r/20250826-nova_firmware-v2-1-93566252fe3a@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 rust/kernel/transmute.rs | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/rust/kernel/transmute.rs b/rust/kernel/transmute.rs
index 5fc7d4ed76b4..cfc37d81adf2 100644
--- a/rust/kernel/transmute.rs
+++ b/rust/kernel/transmute.rs
@@ -79,6 +79,24 @@ pub unsafe trait FromBytes {
             None
         }
     }
+
+    /// Creates an owned instance of `Self` by copying `bytes`.
+    ///
+    /// Unlike [`FromBytes::from_bytes`], which requires aligned input, this method can be used on
+    /// non-aligned data at the cost of a copy.
+    fn from_bytes_copy(bytes: &[u8]) -> Option<Self>
+    where
+        Self: Sized,
+    {
+        if bytes.len() == size_of::<Self>() {
+            // SAFETY: we just verified that `bytes` has the same size as `Self`, and per the
+            // invariants of `FromBytes`, any byte sequence of the correct length is a valid value
+            // for `Self`.
+            Some(unsafe { core::ptr::read_unaligned(bytes.as_ptr().cast::<Self>()) })
+        } else {
+            None
+        }
+    }
 }
 
 macro_rules! impl_frombytes {

From ed906b7076d072fc0f1b358c868ff69cab69607c Mon Sep 17 00:00:00 2001
From: Riana Tauro <riana.tauro@intel.com>
Date: Mon, 25 Aug 2025 16:05:37 +0530
Subject: [PATCH 0726/1455] drm/xe/nvm: Use root tile mmio

To allow initialization of nvm during early probe for future usecases,
use root tile instead of root gt to access mmios, as gt is not
yet initialized at early probe.

v2: fix commit message (Lucas)

Signed-off-by: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250825103537.2551837-1-riana.tauro@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_nvm.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_nvm.c b/drivers/gpu/drm/xe/xe_nvm.c
index 61b0a1531a53..9ca4a5ae1693 100644
--- a/drivers/gpu/drm/xe/xe_nvm.c
+++ b/drivers/gpu/drm/xe/xe_nvm.c
@@ -39,17 +39,17 @@ static void xe_nvm_release_dev(struct device *dev)
 
 static bool xe_nvm_non_posted_erase(struct xe_device *xe)
 {
-	struct xe_gt *gt = xe_root_mmio_gt(xe);
+	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
 
 	if (xe->info.platform != XE_BATTLEMAGE)
 		return false;
-	return !(xe_mmio_read32(&gt->mmio, XE_REG(GEN12_CNTL_PROTECTED_NVM_REG)) &
+	return !(xe_mmio_read32(mmio, XE_REG(GEN12_CNTL_PROTECTED_NVM_REG)) &
 		 NVM_NON_POSTED_ERASE_CHICKEN_BIT);
 }
 
 static bool xe_nvm_writable_override(struct xe_device *xe)
 {
-	struct xe_gt *gt = xe_root_mmio_gt(xe);
+	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
 	bool writable_override;
 	resource_size_t base;
 
@@ -72,7 +72,7 @@ static bool xe_nvm_writable_override(struct xe_device *xe)
 	}
 
 	writable_override =
-		!(xe_mmio_read32(&gt->mmio, HECI_FWSTS2(base)) &
+		!(xe_mmio_read32(mmio, HECI_FWSTS2(base)) &
 		  HECI_FW_STATUS_2_NVM_ACCESS_MODE);
 	if (writable_override)
 		drm_info(&xe->drm, "NVM access overridden by jumper\n");

From eeb8117f5f1c2e8e625e5cb39dbccd21d395caad Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 28 Aug 2025 12:45:16 +0530
Subject: [PATCH 0727/1455] drm/xe/uapi: Fix kernel-doc formatting for madvise
 and vma_query

Correct kernel-doc formatting issues in the UAPI definitions for
madvise and VMA query interfaces to resolve docutils warnings during
documentation build.

Fixes: 418807860e94 ("drm/xe/uapi: Add UAPI for querying VMA count and memory attributes")
Fixes: 231bb0ee7aa5 ("drm/xe/uapi: Add madvise interface")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250828071516.3838110-1-himal.prasad.ghimiray@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 include/uapi/drm/xe_drm.h | 43 ++++++++++++++++++++-------------------
 1 file changed, 22 insertions(+), 21 deletions(-)

diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index 7dedd45ab995..40ff19f52a8d 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -1997,23 +1997,23 @@ struct drm_xe_query_eu_stall {
  * union member is used to provide additional parameters for @type.
  *
  * Supported attribute types:
- * - DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC: Set preferred memory location.
- * - DRM_XE_MEM_RANGE_ATTR_ATOMIC: Set atomic access policy.
- * - DRM_XE_MEM_RANGE_ATTR_PAT: Set page attribute table index.
+ *  - DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC: Set preferred memory location.
+ *  - DRM_XE_MEM_RANGE_ATTR_ATOMIC: Set atomic access policy.
+ *  - DRM_XE_MEM_RANGE_ATTR_PAT: Set page attribute table index.
  *
  * Example:
  *
  * .. code-block:: C
  *
- * struct drm_xe_madvise madvise = {
- *          .vm_id = vm_id,
- *          .start = 0x100000,
- *          .range = 0x2000,
- *          .type = DRM_XE_MEM_RANGE_ATTR_ATOMIC,
- *          .atomic_val = DRM_XE_ATOMIC_DEVICE,
- *         };
+ *    struct drm_xe_madvise madvise = {
+ *         .vm_id = vm_id,
+ *         .start = 0x100000,
+ *         .range = 0x2000,
+ *         .type = DRM_XE_MEM_RANGE_ATTR_ATOMIC,
+ *         .atomic_val = DRM_XE_ATOMIC_DEVICE,
+ *    };
  *
- * ioctl(fd, DRM_IOCTL_XE_MADVISE, &madvise);
+ *    ioctl(fd, DRM_IOCTL_XE_MADVISE, &madvise);
  *
  */
 struct drm_xe_madvise {
@@ -2042,12 +2042,12 @@ struct drm_xe_madvise {
 		 * Used when @type == DRM_XE_MEM_RANGE_ATTR_PREFERRED_LOC
 		 *
 		 * Supported values for @preferred_mem_loc.devmem_fd:
-		 * - DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE: set vram of faulting tile as preferred loc
-		 * - DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM: set smem as preferred loc
+		 *  - DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE: set vram of fault tile as preferred loc
+		 *  - DRM_XE_PREFERRED_LOC_DEFAULT_SYSTEM: set smem as preferred loc
 		 *
 		 * Supported values for @preferred_mem_loc.migration_policy:
-		 * - DRM_XE_MIGRATE_ALL_PAGES
-		 * - DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES
+		 *  - DRM_XE_MIGRATE_ALL_PAGES
+		 *  - DRM_XE_MIGRATE_ONLY_SYSTEM_PAGES
 		 */
 		struct {
 #define DRM_XE_PREFERRED_LOC_DEFAULT_DEVICE	0
@@ -2073,12 +2073,12 @@ struct drm_xe_madvise {
 		 * Used when @type == DRM_XE_MEM_RANGE_ATTR_ATOMIC.
 		 *
 		 * Supported values for @atomic.val:
-		 * - DRM_XE_ATOMIC_UNDEFINED: Undefined or default behaviour
-		 *   Support both GPU and CPU atomic operations for system allocator
-		 *   Support GPU atomic operations for normal(bo) allocator
-		 * - DRM_XE_ATOMIC_DEVICE: Support GPU atomic operations
-		 * - DRM_XE_ATOMIC_GLOBAL: Support both GPU and CPU atomic operations
-		 * - DRM_XE_ATOMIC_CPU: Support CPU atomic
+		 *  - DRM_XE_ATOMIC_UNDEFINED: Undefined or default behaviour.
+		 *    Support both GPU and CPU atomic operations for system allocator.
+		 *    Support GPU atomic operations for normal(bo) allocator.
+		 *  - DRM_XE_ATOMIC_DEVICE: Support GPU atomic operations.
+		 *  - DRM_XE_ATOMIC_GLOBAL: Support both GPU and CPU atomic operations.
+		 *  - DRM_XE_ATOMIC_CPU: Support CPU atomic only, no GPU atomics supported.
 		 */
 		struct {
 #define DRM_XE_ATOMIC_UNDEFINED	0
@@ -2196,6 +2196,7 @@ struct drm_xe_mem_range_attr {
  * Example:
  *
  * .. code-block:: C
+ *
  *    struct drm_xe_vm_query_mem_range_attr query = {
  *         .vm_id = vm_id,
  *         .start = 0x100000,

From 806d4fec0b80be6964b69947aad928a322fe77e4 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:10 +0200
Subject: [PATCH 0728/1455] dt-bindings: display: st: add two new compatibles
 to LTDC device

The new STMicroelectronics SoC features a display controller similar to
the one used in previous SoCs.  Because there is additional registers,
and different mandatory clocks it is incompatible with existing IPs.  On
STM32MP251, the device only needs two clocks while on STM32MP255 it
needs four.

Add the new names to the list of compatible string and handle each
quirks accordingly.

Acked-by: Philippe Cornu <philippe.cornu@foss.st.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-1-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 .../bindings/display/st,stm32-ltdc.yaml       | 52 ++++++++++++++++++-
 1 file changed, 50 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml b/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml
index d6ea4d62a2cf..1233345a1630 100644
--- a/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml
+++ b/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml
@@ -12,7 +12,10 @@ maintainers:
 
 properties:
   compatible:
-    const: st,stm32-ltdc
+    enum:
+      - st,stm32-ltdc
+      - st,stm32mp251-ltdc
+      - st,stm32mp255-ltdc
 
   reg:
     maxItems: 1
@@ -24,11 +27,16 @@ properties:
     minItems: 1
 
   clocks:
-    maxItems: 1
+    minItems: 1
+    maxItems: 4
 
   clock-names:
     items:
       - const: lcd
+      - const: bus
+      - const: ref
+      - const: lvds
+    minItems: 1
 
   resets:
     maxItems: 1
@@ -51,6 +59,46 @@ required:
   - resets
   - port
 
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - st,stm32-ltdc
+    then:
+      properties:
+        clocks:
+          maxItems: 1
+        clock-names:
+          maxItems: 1
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - st,stm32mp251-ltdc
+    then:
+      properties:
+        clocks:
+          minItems: 2
+          maxItems: 2
+        clock-names:
+          minItems: 2
+          maxItems: 2
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - st,stm32mp255-ltdc
+    then:
+      properties:
+        clocks:
+          minItems: 4
+        clock-names:
+          minItems: 4
+
 additionalProperties: false
 
 examples:

From 81d678332c404384e5e83300376d9be375f6d928 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:11 +0200
Subject: [PATCH 0729/1455] dt-bindings: display: st,stm32-ltdc: add
 access-controllers property

access-controllers is an optional property that allows a peripheral to
refer to one or more domain access controller(s).

This property is added when the peripheral is under the STM32 firewall
controller.  It allows an accurate representation of the hardware, where
the peripheral is connected to a firewall bus.  The firewall can then check
the peripheral accesses before allowing its device to probe.

Acked-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-2-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml b/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml
index 1233345a1630..77058a5ccf68 100644
--- a/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml
+++ b/Documentation/devicetree/bindings/display/st,stm32-ltdc.yaml
@@ -41,6 +41,9 @@ properties:
   resets:
     maxItems: 1
 
+  access-controllers:
+    maxItems: 1
+
   port:
     $ref: /schemas/graph.yaml#/properties/port
     description: |

From 6b96ffeb81be8e43da7c4ddd0802739e22459519 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:12 +0200
Subject: [PATCH 0730/1455] dt-bindings: display: st: add new compatible to
 LVDS device

Update the compatible to accept both "st,stm32mp255-lvds" and
st,stm32mp25-lvds" respectively.  Default will fall back to
"st,stm32mp25-lvds".

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-3-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 .../devicetree/bindings/display/st,stm32mp25-lvds.yaml     | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml b/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
index 6736f93256b5..74e61d95370c 100644
--- a/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
+++ b/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
@@ -31,7 +31,12 @@ description: |
 
 properties:
   compatible:
-    const: st,stm32mp25-lvds
+    oneOf:
+      - items:
+          - enum:
+              - st,stm32mp255-lvds
+          - const: st,stm32mp25-lvds
+      - const: st,stm32mp25-lvds
 
   "#clock-cells":
     const: 0

From 922e28eec67390686d944d03a12068df1197b393 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:13 +0200
Subject: [PATCH 0731/1455] dt-bindings: display: st,stm32mp25-lvds: add
 access-controllers property

access-controllers is an optional property that allows a peripheral to
refer to one or more domain access controller(s).

This property is added when the peripheral is under the STM32 firewall
controller.  It allows an accurate representation of the hardware, where
the peripheral is connected to a firewall bus.  The firewall can then
check the peripheral accesses before allowing its device to probe.

Acked-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-4-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 .../devicetree/bindings/display/st,stm32mp25-lvds.yaml         | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml b/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
index 74e61d95370c..05a73bbc246a 100644
--- a/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
+++ b/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
@@ -59,6 +59,9 @@ properties:
   resets:
     maxItems: 1
 
+  access-controllers:
+    maxItems: 1
+
   ports:
     $ref: /schemas/graph.yaml#/properties/ports
 

From 08f6613e274f6cfe4d7d09ac5050d632dbba9342 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:14 +0200
Subject: [PATCH 0732/1455] dt-bindings: display: st,stm32mp25-lvds: add
 power-domains property

STM32 LVDS peripheral may be in a power domain.  Allow an optional
single 'power-domains' entry for STM32 LVDS devices.

Acked-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-5-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 .../devicetree/bindings/display/st,stm32mp25-lvds.yaml         | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml b/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
index 05a73bbc246a..14e042156179 100644
--- a/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
+++ b/Documentation/devicetree/bindings/display/st,stm32mp25-lvds.yaml
@@ -62,6 +62,9 @@ properties:
   access-controllers:
     maxItems: 1
 
+  power-domains:
+    maxItems: 1
+
   ports:
     $ref: /schemas/graph.yaml#/properties/ports
 

From 1b9482f8dfce82af44e261c78777a37f32efe889 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:15 +0200
Subject: [PATCH 0733/1455] dt-bindings: arm: stm32: add required #clock-cells
 property

On STM32MP25 SoC, the syscfg peripheral provides a clock to the display
subsystem through a multiplexer.  Since it only provides a single clock,
the cell value is 0.

Doing so allows the clock consumers to reach the peripheral and gate the
clock accordingly.

Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Christophe Roullier <christophe.roullier@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-6-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 .../bindings/arm/stm32/st,stm32-syscon.yaml   | 31 +++++++++++++------
 1 file changed, 21 insertions(+), 10 deletions(-)

diff --git a/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml b/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml
index ed97652c8492..95d2319afe23 100644
--- a/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml
+++ b/Documentation/devicetree/bindings/arm/stm32/st,stm32-syscon.yaml
@@ -36,20 +36,31 @@ properties:
   clocks:
     maxItems: 1
 
+  "#clock-cells":
+    const: 0
+
 required:
   - compatible
   - reg
 
-if:
-  properties:
-    compatible:
-      contains:
-        enum:
-          - st,stm32mp157-syscfg
-          - st,stm32f4-gcan
-then:
-  required:
-    - clocks
+allOf:
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              - st,stm32mp157-syscfg
+              - st,stm32f4-gcan
+    then:
+      required:
+        - clocks
+  - if:
+      properties:
+        compatible:
+          const: st,stm32mp25-syscfg
+    then:
+      required:
+        - "#clock-cells"
 
 additionalProperties: false
 

From d80667642b0f43580efd489d627dd24aa10ab98f Mon Sep 17 00:00:00 2001
From: Yannick Fertre <yannick.fertre@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:16 +0200
Subject: [PATCH 0734/1455] drm/stm: ltdc: support new hardware version for
 STM32MP25 SoC

STM32MP25 SoC features a new version of the LTDC IP.  Add its compatible
to the list of device to probe and implement its quirks.

This hardware supports a pad frequency of 150MHz and a peripheral bus
clock.

Signed-off-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Philippe Cornu <philippe.cornu@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-7-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 drivers/gpu/drm/stm/drv.c  | 12 +++++++++++-
 drivers/gpu/drm/stm/ltdc.c | 38 +++++++++++++++++++++++++++++++++++---
 drivers/gpu/drm/stm/ltdc.h |  5 +++++
 3 files changed, 51 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/stm/drv.c b/drivers/gpu/drm/stm/drv.c
index 8ebcaf953782..ab00d1a6140c 100644
--- a/drivers/gpu/drm/stm/drv.c
+++ b/drivers/gpu/drm/stm/drv.c
@@ -236,8 +236,18 @@ static void stm_drm_platform_shutdown(struct platform_device *pdev)
 	drm_atomic_helper_shutdown(platform_get_drvdata(pdev));
 }
 
+static struct ltdc_plat_data stm_drm_plat_data = {
+	.pad_max_freq_hz = 90000000,
+};
+
+static struct ltdc_plat_data stm_drm_plat_data_mp25 = {
+	.pad_max_freq_hz = 150000000,
+};
+
 static const struct of_device_id drv_dt_ids[] = {
-	{ .compatible = "st,stm32-ltdc"},
+	{ .compatible = "st,stm32-ltdc", .data = &stm_drm_plat_data, },
+	{ .compatible = "st,stm32mp251-ltdc", .data = &stm_drm_plat_data_mp25, },
+	{ .compatible = "st,stm32mp255-ltdc", .data = &stm_drm_plat_data_mp25, },
 	{ /* end node */ },
 };
 MODULE_DEVICE_TABLE(of, drv_dt_ids);
diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c
index b9477fbec1c1..9970bcb748ea 100644
--- a/drivers/gpu/drm/stm/ltdc.c
+++ b/drivers/gpu/drm/stm/ltdc.c
@@ -14,6 +14,7 @@
 #include <linux/interrupt.h>
 #include <linux/media-bus-format.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/of_graph.h>
 #include <linux/pinctrl/consumer.h>
 #include <linux/platform_device.h>
@@ -51,6 +52,7 @@
 #define HWVER_10300 0x010300
 #define HWVER_20101 0x020101
 #define HWVER_40100 0x040100
+#define HWVER_40101 0x040101
 
 /*
  * The address of some registers depends on the HW version: such registers have
@@ -1780,6 +1782,7 @@ static int ltdc_get_caps(struct drm_device *ddev)
 {
 	struct ltdc_device *ldev = ddev->dev_private;
 	u32 bus_width_log2, lcr, gc2r;
+	const struct ltdc_plat_data *pdata = of_device_get_match_data(ddev->dev);
 
 	/*
 	 * at least 1 layer must be managed & the number of layers
@@ -1795,6 +1798,8 @@ static int ltdc_get_caps(struct drm_device *ddev)
 	ldev->caps.bus_width = 8 << bus_width_log2;
 	regmap_read(ldev->regmap, LTDC_IDR, &ldev->caps.hw_version);
 
+	ldev->caps.pad_max_freq_hz = pdata->pad_max_freq_hz;
+
 	switch (ldev->caps.hw_version) {
 	case HWVER_10200:
 	case HWVER_10300:
@@ -1812,7 +1817,6 @@ static int ltdc_get_caps(struct drm_device *ddev)
 		 * does not work on 2nd layer.
 		 */
 		ldev->caps.non_alpha_only_l1 = true;
-		ldev->caps.pad_max_freq_hz = 90000000;
 		if (ldev->caps.hw_version == HWVER_10200)
 			ldev->caps.pad_max_freq_hz = 65000000;
 		ldev->caps.nb_irq = 2;
@@ -1843,6 +1847,7 @@ static int ltdc_get_caps(struct drm_device *ddev)
 		ldev->caps.fifo_threshold = false;
 		break;
 	case HWVER_40100:
+	case HWVER_40101:
 		ldev->caps.layer_ofs = LAY_OFS_1;
 		ldev->caps.layer_regs = ltdc_layer_regs_a2;
 		ldev->caps.pix_fmt_hw = ltdc_pix_fmt_a2;
@@ -1850,7 +1855,6 @@ static int ltdc_get_caps(struct drm_device *ddev)
 		ldev->caps.pix_fmt_nb = ARRAY_SIZE(ltdc_drm_fmt_a2);
 		ldev->caps.pix_fmt_flex = true;
 		ldev->caps.non_alpha_only_l1 = false;
-		ldev->caps.pad_max_freq_hz = 90000000;
 		ldev->caps.nb_irq = 2;
 		ldev->caps.ycbcr_input = true;
 		ldev->caps.ycbcr_output = true;
@@ -1873,6 +1877,8 @@ void ltdc_suspend(struct drm_device *ddev)
 
 	drm_dbg_driver(ddev, "\n");
 	clk_disable_unprepare(ldev->pixel_clk);
+	if (ldev->bus_clk)
+		clk_disable_unprepare(ldev->bus_clk);
 }
 
 int ltdc_resume(struct drm_device *ddev)
@@ -1888,7 +1894,13 @@ int ltdc_resume(struct drm_device *ddev)
 		return ret;
 	}
 
-	return 0;
+	if (ldev->bus_clk) {
+		ret = clk_prepare_enable(ldev->bus_clk);
+		if (ret)
+			drm_err(ddev, "failed to enable bus clock (%d)\n", ret);
+	}
+
+	return ret;
 }
 
 int ltdc_load(struct drm_device *ddev)
@@ -1923,6 +1935,20 @@ int ltdc_load(struct drm_device *ddev)
 		return -ENODEV;
 	}
 
+	if (of_device_is_compatible(np, "st,stm32mp251-ltdc") ||
+	    of_device_is_compatible(np, "st,stm32mp255-ltdc")) {
+		ldev->bus_clk = devm_clk_get(dev, "bus");
+		if (IS_ERR(ldev->bus_clk))
+			return dev_err_probe(dev, PTR_ERR(ldev->bus_clk),
+					     "Unable to get bus clock\n");
+
+		ret = clk_prepare_enable(ldev->bus_clk);
+		if (ret) {
+			drm_err(ddev, "Unable to prepare bus clock\n");
+			return ret;
+		}
+	}
+
 	/* Get endpoints if any */
 	for (i = 0; i < nb_endpoints; i++) {
 		ret = drm_of_find_panel_or_bridge(np, 0, i, &panel, &bridge);
@@ -2035,6 +2061,9 @@ int ltdc_load(struct drm_device *ddev)
 
 	clk_disable_unprepare(ldev->pixel_clk);
 
+	if (ldev->bus_clk)
+		clk_disable_unprepare(ldev->bus_clk);
+
 	pinctrl_pm_select_sleep_state(ddev->dev);
 
 	pm_runtime_enable(ddev->dev);
@@ -2043,6 +2072,9 @@ int ltdc_load(struct drm_device *ddev)
 err:
 	clk_disable_unprepare(ldev->pixel_clk);
 
+	if (ldev->bus_clk)
+		clk_disable_unprepare(ldev->bus_clk);
+
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/stm/ltdc.h b/drivers/gpu/drm/stm/ltdc.h
index 9d488043ffdb..ddfa8ae61a7b 100644
--- a/drivers/gpu/drm/stm/ltdc.h
+++ b/drivers/gpu/drm/stm/ltdc.h
@@ -40,10 +40,15 @@ struct fps_info {
 	ktime_t last_timestamp;
 };
 
+struct ltdc_plat_data {
+	int pad_max_freq_hz;	/* max frequency supported by pad */
+};
+
 struct ltdc_device {
 	void __iomem *regs;
 	struct regmap *regmap;
 	struct clk *pixel_clk;	/* lcd pixel clock */
+	struct clk *bus_clk;	/* bus clock */
 	struct mutex err_lock;	/* protecting error_status */
 	struct ltdc_caps caps;
 	u32 irq_status;

From b7243862f82a088d6fb7e22a25c7c024818e631d Mon Sep 17 00:00:00 2001
From: Yannick Fertre <yannick.fertre@foss.st.com>
Date: Fri, 22 Aug 2025 16:34:17 +0200
Subject: [PATCH 0735/1455] drm/stm: ltdc: handle lvds pixel clock

Handle LVDS pixel clock.

The LTDC operates with multiple clock domains for register access,
requiring all clocks to be provided during read/write operations.  This
imposes a dependency between the LVDS and LTDC to access correctly all
LTDC registers.  And because both IPs' pixel rates must be synchronized,
the LTDC has to handle the LVDS clock.

Signed-off-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Yannick Fertre <yannick.fertre@foss.st.com>
Acked-by: Philippe Cornu <philippe.cornu@foss.st.com>
Link: https://lore.kernel.org/r/20250822-drm-misc-next-v5-8-9c825e28f733@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 drivers/gpu/drm/stm/ltdc.c | 22 +++++++++++++++++++++-
 drivers/gpu/drm/stm/ltdc.h |  1 +
 2 files changed, 22 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/stm/ltdc.c b/drivers/gpu/drm/stm/ltdc.c
index 9970bcb748ea..d1501e86a5b1 100644
--- a/drivers/gpu/drm/stm/ltdc.c
+++ b/drivers/gpu/drm/stm/ltdc.c
@@ -838,6 +838,12 @@ ltdc_crtc_mode_valid(struct drm_crtc *crtc,
 	int target_max = target + CLK_TOLERANCE_HZ;
 	int result;
 
+	if (ldev->lvds_clk) {
+		result = clk_round_rate(ldev->lvds_clk, target);
+		drm_dbg_driver(crtc->dev, "lvds pixclk rate target %d, available %d\n",
+			       target, result);
+	}
+
 	result = clk_round_rate(ldev->pixel_clk, target);
 
 	drm_dbg_driver(crtc->dev, "clk rate target %d, available %d\n", target, result);
@@ -1879,6 +1885,8 @@ void ltdc_suspend(struct drm_device *ddev)
 	clk_disable_unprepare(ldev->pixel_clk);
 	if (ldev->bus_clk)
 		clk_disable_unprepare(ldev->bus_clk);
+	if (ldev->lvds_clk)
+		clk_disable_unprepare(ldev->lvds_clk);
 }
 
 int ltdc_resume(struct drm_device *ddev)
@@ -1896,8 +1904,16 @@ int ltdc_resume(struct drm_device *ddev)
 
 	if (ldev->bus_clk) {
 		ret = clk_prepare_enable(ldev->bus_clk);
-		if (ret)
+		if (ret) {
 			drm_err(ddev, "failed to enable bus clock (%d)\n", ret);
+			return ret;
+		}
+	}
+
+	if (ldev->lvds_clk) {
+		ret = clk_prepare_enable(ldev->lvds_clk);
+		if (ret)
+			drm_err(ddev, "failed to prepare lvds clock\n");
 	}
 
 	return ret;
@@ -1982,6 +1998,10 @@ int ltdc_load(struct drm_device *ddev)
 		}
 	}
 
+	ldev->lvds_clk = devm_clk_get(dev, "lvds");
+	if (IS_ERR(ldev->lvds_clk))
+		ldev->lvds_clk = NULL;
+
 	rstc = devm_reset_control_get_exclusive(dev, NULL);
 
 	mutex_init(&ldev->err_lock);
diff --git a/drivers/gpu/drm/stm/ltdc.h b/drivers/gpu/drm/stm/ltdc.h
index ddfa8ae61a7b..17b51a7ce28e 100644
--- a/drivers/gpu/drm/stm/ltdc.h
+++ b/drivers/gpu/drm/stm/ltdc.h
@@ -48,6 +48,7 @@ struct ltdc_device {
 	void __iomem *regs;
 	struct regmap *regmap;
 	struct clk *pixel_clk;	/* lcd pixel clock */
+	struct clk *lvds_clk;	/* lvds pixel clock */
 	struct clk *bus_clk;	/* bus clock */
 	struct mutex err_lock;	/* protecting error_status */
 	struct ltdc_caps caps;

From bffa3a03e787d3e83193c8f3ef0b31d8bf7fb39c Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Fri, 27 Jun 2025 19:18:03 +0200
Subject: [PATCH 0736/1455] drm/msm/mdp4: Consistently use the "mdp4_"
 namespace

Functions and other stuff all start with "mdp4_", except a few ones that
start with "mpd4_" (d and p switched)

Make things consistent and use "mdp4_" everywhere.

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/661306/
Link: https://lore.kernel.org/r/6b9076268548c52ec371e9ed35fee0dd8fcb46ef.1751044672.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h      |  2 +-
 .../gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c |  2 +-
 drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c | 38 +++++++++----------
 3 files changed, 21 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
index fb348583dc84..06458d4ee48c 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.h
@@ -202,6 +202,6 @@ static inline struct drm_encoder *mdp4_dsi_encoder_init(struct drm_device *dev)
 }
 #endif
 
-struct clk *mpd4_get_lcdc_clock(struct drm_device *dev);
+struct clk *mdp4_get_lcdc_clock(struct drm_device *dev);
 
 #endif /* __MDP4_KMS_H__ */
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
index 06a307c1272d..1051873057f6 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c
@@ -375,7 +375,7 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev)
 
 	drm_encoder_helper_add(encoder, &mdp4_lcdc_encoder_helper_funcs);
 
-	mdp4_lcdc_encoder->lcdc_clk = mpd4_get_lcdc_clock(dev);
+	mdp4_lcdc_encoder->lcdc_clk = mdp4_get_lcdc_clock(dev);
 	if (IS_ERR(mdp4_lcdc_encoder->lcdc_clk)) {
 		DRM_DEV_ERROR(dev->dev, "failed to get lvds_clk\n");
 		return ERR_CAST(mdp4_lcdc_encoder->lcdc_clk);
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
index fa2c29470510..4612886f0e49 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
@@ -54,7 +54,7 @@ static const struct pll_rate *find_rate(unsigned long rate)
 	return &freqtbl[i-1];
 }
 
-static int mpd4_lvds_pll_enable(struct clk_hw *hw)
+static int mdp4_lvds_pll_enable(struct clk_hw *hw)
 {
 	struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw);
 	struct mdp4_kms *mdp4_kms = get_kms(lvds_pll);
@@ -80,7 +80,7 @@ static int mpd4_lvds_pll_enable(struct clk_hw *hw)
 	return 0;
 }
 
-static void mpd4_lvds_pll_disable(struct clk_hw *hw)
+static void mdp4_lvds_pll_disable(struct clk_hw *hw)
 {
 	struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw);
 	struct mdp4_kms *mdp4_kms = get_kms(lvds_pll);
@@ -91,21 +91,21 @@ static void mpd4_lvds_pll_disable(struct clk_hw *hw)
 	mdp4_write(mdp4_kms, REG_MDP4_LVDS_PHY_PLL_CTRL_0, 0x0);
 }
 
-static unsigned long mpd4_lvds_pll_recalc_rate(struct clk_hw *hw,
+static unsigned long mdp4_lvds_pll_recalc_rate(struct clk_hw *hw,
 				unsigned long parent_rate)
 {
 	struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw);
 	return lvds_pll->pixclk;
 }
 
-static long mpd4_lvds_pll_round_rate(struct clk_hw *hw, unsigned long rate,
+static long mdp4_lvds_pll_round_rate(struct clk_hw *hw, unsigned long rate,
 		unsigned long *parent_rate)
 {
 	const struct pll_rate *pll_rate = find_rate(rate);
 	return pll_rate->rate;
 }
 
-static int mpd4_lvds_pll_set_rate(struct clk_hw *hw, unsigned long rate,
+static int mdp4_lvds_pll_set_rate(struct clk_hw *hw, unsigned long rate,
 		unsigned long parent_rate)
 {
 	struct mdp4_lvds_pll *lvds_pll = to_mdp4_lvds_pll(hw);
@@ -114,26 +114,26 @@ static int mpd4_lvds_pll_set_rate(struct clk_hw *hw, unsigned long rate,
 }
 
 
-static const struct clk_ops mpd4_lvds_pll_ops = {
-	.enable = mpd4_lvds_pll_enable,
-	.disable = mpd4_lvds_pll_disable,
-	.recalc_rate = mpd4_lvds_pll_recalc_rate,
-	.round_rate = mpd4_lvds_pll_round_rate,
-	.set_rate = mpd4_lvds_pll_set_rate,
+static const struct clk_ops mdp4_lvds_pll_ops = {
+	.enable = mdp4_lvds_pll_enable,
+	.disable = mdp4_lvds_pll_disable,
+	.recalc_rate = mdp4_lvds_pll_recalc_rate,
+	.round_rate = mdp4_lvds_pll_round_rate,
+	.set_rate = mdp4_lvds_pll_set_rate,
 };
 
-static const struct clk_parent_data mpd4_lvds_pll_parents[] = {
+static const struct clk_parent_data mdp4_lvds_pll_parents[] = {
 	{ .fw_name = "pxo", .name = "pxo", },
 };
 
 static struct clk_init_data pll_init = {
-	.name = "mpd4_lvds_pll",
-	.ops = &mpd4_lvds_pll_ops,
-	.parent_data = mpd4_lvds_pll_parents,
-	.num_parents = ARRAY_SIZE(mpd4_lvds_pll_parents),
+	.name = "mdp4_lvds_pll",
+	.ops = &mdp4_lvds_pll_ops,
+	.parent_data = mdp4_lvds_pll_parents,
+	.num_parents = ARRAY_SIZE(mdp4_lvds_pll_parents),
 };
 
-static struct clk_hw *mpd4_lvds_pll_init(struct drm_device *dev)
+static struct clk_hw *mdp4_lvds_pll_init(struct drm_device *dev)
 {
 	struct mdp4_lvds_pll *lvds_pll;
 	int ret;
@@ -156,14 +156,14 @@ static struct clk_hw *mpd4_lvds_pll_init(struct drm_device *dev)
 	return &lvds_pll->pll_hw;
 }
 
-struct clk *mpd4_get_lcdc_clock(struct drm_device *dev)
+struct clk *mdp4_get_lcdc_clock(struct drm_device *dev)
 {
 	struct clk_hw *hw;
 	struct clk *clk;
 
 
 	/* TODO: do we need different pll in other cases? */
-	hw = mpd4_lvds_pll_init(dev);
+	hw = mdp4_lvds_pll_init(dev);
 	if (IS_ERR(hw)) {
 		DRM_DEV_ERROR(dev->dev, "failed to register LVDS PLL\n");
 		return ERR_CAST(hw);

From 86d39a046b59583675a8c9701acc86c02768a433 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Sat, 28 Jun 2025 06:02:35 +0300
Subject: [PATCH 0737/1455] dt-bindings: display/msm/gpu: account for 7xx GPUs
 in clocks conditions

Handle two cases for Adreno 7xx:
- Adreno 702 follows A610 and A619 example and has clocks in the GPU
  node.
- Newer 7xx GPUs use a different pattern for the compatibles and did not
  match currently.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/661351/
Link: https://lore.kernel.org/r/20250628-rework-msm-gpu-schema-v1-1-89f818c51b6a@oss.qualcomm.com
---
 Documentation/devicetree/bindings/display/msm/gpu.yaml | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/msm/gpu.yaml b/Documentation/devicetree/bindings/display/msm/gpu.yaml
index 6ddc72fd85b0..97254f909260 100644
--- a/Documentation/devicetree/bindings/display/msm/gpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gpu.yaml
@@ -187,6 +187,7 @@ allOf:
             enum:
               - qcom,adreno-610.0
               - qcom,adreno-619.1
+              - qcom,adreno-07000200
     then:
       properties:
         clocks:
@@ -222,7 +223,9 @@ allOf:
         properties:
           compatible:
             contains:
-              pattern: '^qcom,adreno-[67][0-9][0-9]\.[0-9]+$'
+              oneOf:
+                - pattern: '^qcom,adreno-[67][0-9][0-9]\.[0-9]+$'
+                - pattern: '^qcom,adreno-[0-9a-f]{8}$'
 
       then: # Starting with A6xx, the clocks are usually defined in the GMU node
         properties:

From cef3a60b30d1697ac796cf22f41531c916329db1 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Sat, 28 Jun 2025 06:02:36 +0300
Subject: [PATCH 0738/1455] dt-bindings: display/msm/gpu: describe alwayson
 clock

Adreno A506 and A510 have one extra clock, alwayson. Describe it in the
schema.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/661352/
Link: https://lore.kernel.org/r/20250628-rework-msm-gpu-schema-v1-2-89f818c51b6a@oss.qualcomm.com
---
 Documentation/devicetree/bindings/display/msm/gpu.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/gpu.yaml b/Documentation/devicetree/bindings/display/msm/gpu.yaml
index 97254f909260..af43527188ce 100644
--- a/Documentation/devicetree/bindings/display/msm/gpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gpu.yaml
@@ -173,6 +173,8 @@ allOf:
                 description: GPU RBBM Timer for Adreno 5xx series
               - const: rbcpr
                 description: GPU RB Core Power Reduction clock
+              - const: alwayson
+                description: GPU AON clock
           minItems: 2
           maxItems: 7
 

From f3b649d7a513f71a31c451bd0825657211366d7f Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Sat, 28 Jun 2025 06:02:37 +0300
Subject: [PATCH 0739/1455] dt-bindings: display/msm/gpu: describe clocks for
 each Adreno GPU type

Rather than having a single list with all possible clocks for A3xx-A5xx
define individual Adreno GPU types and corresponding clock lists.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/661348/
Link: https://lore.kernel.org/r/20250628-rework-msm-gpu-schema-v1-3-89f818c51b6a@oss.qualcomm.com
---
 .../devicetree/bindings/display/msm/gpu.yaml  | 226 +++++++++++++++---
 1 file changed, 197 insertions(+), 29 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/gpu.yaml b/Documentation/devicetree/bindings/display/msm/gpu.yaml
index af43527188ce..7ef80f9fac8b 100644
--- a/Documentation/devicetree/bindings/display/msm/gpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gpu.yaml
@@ -146,41 +146,209 @@ allOf:
       properties:
         compatible:
           contains:
-            pattern: '^qcom,adreno-[3-5][0-9][0-9]\.[0-9]+$'
-
+            oneOf:
+              - pattern: '^qcom,adreno-305\.[0-9]+$'
+              - pattern: '^qcom,adreno-330\.[0-9]+$'
     then:
       properties:
         clocks:
-          minItems: 2
-          maxItems: 7
-
+          minItems: 3
+          maxItems: 3
         clock-names:
           items:
-            anyOf:
-              - const: core
-                description: GPU Core clock
-              - const: iface
-                description: GPU Interface clock
-              - const: mem
-                description: GPU Memory clock
-              - const: mem_iface
-                description: GPU Memory Interface clock
-              - const: alt_mem_iface
-                description: GPU Alternative Memory Interface clock
-              - const: gfx3d
-                description: GPU 3D engine clock
-              - const: rbbmtimer
-                description: GPU RBBM Timer for Adreno 5xx series
-              - const: rbcpr
-                description: GPU RB Core Power Reduction clock
-              - const: alwayson
-                description: GPU AON clock
-          minItems: 2
-          maxItems: 7
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
 
-      required:
-        - clocks
-        - clock-names
+  - if:
+      properties:
+        compatible:
+          contains:
+            pattern: '^qcom,adreno-306\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 5
+          maxItems: 6
+        clock-names:
+          oneOf:
+            - items:
+                - const: core
+                  description: GPU Core clock
+                - const: iface
+                  description: GPU Interface clock
+                - const: mem_iface
+                  description: GPU Memory Interface clock
+                - const: alt_mem_iface
+                  description: GPU Alternative Memory Interface clock
+                - const: gfx3d
+                  description: GPU 3D engine clock
+            - items:
+                - const: core
+                  description: GPU Core clock
+                - const: iface
+                  description: GPU Interface clock
+                - const: mem
+                  description: GPU Memory clock
+                - const: mem_iface
+                  description: GPU Memory Interface clock
+                - const: alt_mem_iface
+                  description: GPU Alternative Memory Interface clock
+                - const: gfx3d
+                  description: GPU 3D engine clock
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            pattern: '^qcom,adreno-320\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 4
+          maxItems: 4
+        clock-names:
+          items:
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: mem
+              description: GPU Memory clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            pattern: '^qcom,adreno-405\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 7
+          maxItems: 7
+        clock-names:
+          items:
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: mem
+              description: GPU Memory clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
+            - const: alt_mem_iface
+              description: GPU Alternative Memory Interface clock
+            - const: gfx3d
+              description: GPU 3D engine clock
+            - const: rbbmtimer
+              description: GPU RBBM Timer for Adreno 5xx series
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            pattern: '^qcom,adreno-506\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 6
+          maxItems: 6
+        clock-names:
+          items:
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
+            - const: alt_mem_iface
+              description: GPU Alternative Memory Interface clock
+            - const: rbbmtimer
+              description: GPU RBBM Timer for Adreno 5xx series
+            - const: alwayson
+              description: GPU AON clock
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            oneOf:
+              - pattern: '^qcom,adreno-508\.[0-9]+$'
+              - pattern: '^qcom,adreno-509\.[0-9]+$'
+              - pattern: '^qcom,adreno-512\.[0-9]+$'
+              - pattern: '^qcom,adreno-540\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 6
+          maxItems: 6
+        clock-names:
+          items:
+            - const: iface
+              description: GPU Interface clock
+            - const: rbbmtimer
+              description: GPU RBBM Timer for Adreno 5xx series
+            - const: mem
+              description: GPU Memory clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
+            - const: rbcpr
+              description: GPU RB Core Power Reduction clock
+            - const: core
+              description: GPU Core clock
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            pattern: '^qcom,adreno-510\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 6
+          maxItems: 6
+        clock-names:
+          items:
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: mem
+              description: GPU Memory clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
+            - const: rbbmtimer
+              description: GPU RBBM Timer for Adreno 5xx series
+            - const: alwayson
+              description: GPU AON clock
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            pattern: '^qcom,adreno-530\.[0-9]+$'
+    then:
+      properties:
+        clocks:
+          minItems: 5
+          maxItems: 5
+        clock-names:
+          items:
+            - const: core
+              description: GPU Core clock
+            - const: iface
+              description: GPU Interface clock
+            - const: rbbmtimer
+              description: GPU RBBM Timer for Adreno 5xx series
+            - const: mem
+              description: GPU Memory clock
+            - const: mem_iface
+              description: GPU Memory Interface clock
 
   - if:
       properties:

From fb53e8f09fc1e1a343fd08ea4f353f81613975d7 Mon Sep 17 00:00:00 2001
From: "Rob Herring (Arm)" <robh@kernel.org>
Date: Thu, 3 Jul 2025 13:34:41 -0500
Subject: [PATCH 0740/1455] drm/msm: Use of_reserved_mem_region_to_resource()
 for "memory-region"

Use the newly added of_reserved_mem_region_to_resource() function to
handle "memory-region" properties.

The original code did not set 'zap_available' to false if
of_address_to_resource() failed which seems like an oversight.

Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/662275/
Link: https://lore.kernel.org/r/20250703183442.2073717-1-robh@kernel.org
[DB: dropped part related to VRAM, no longer applicable]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/adreno_gpu.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index f1230465bf0d..8c6336b007dc 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -10,7 +10,7 @@
 #include <linux/interconnect.h>
 #include <linux/firmware/qcom/qcom_scm.h>
 #include <linux/kernel.h>
-#include <linux/of_address.h>
+#include <linux/of_reserved_mem.h>
 #include <linux/pm_opp.h>
 #include <linux/slab.h>
 #include <linux/soc/qcom/mdt_loader.h>
@@ -33,7 +33,7 @@ static int zap_shader_load_mdt(struct msm_gpu *gpu, const char *fwname,
 	struct device *dev = &gpu->pdev->dev;
 	const struct firmware *fw;
 	const char *signed_fwname = NULL;
-	struct device_node *np, *mem_np;
+	struct device_node *np;
 	struct resource r;
 	phys_addr_t mem_phys;
 	ssize_t mem_size;
@@ -51,18 +51,11 @@ static int zap_shader_load_mdt(struct msm_gpu *gpu, const char *fwname,
 		return -ENODEV;
 	}
 
-	mem_np = of_parse_phandle(np, "memory-region", 0);
-	of_node_put(np);
-	if (!mem_np) {
+	ret = of_reserved_mem_region_to_resource(np, 0, &r);
+	if (ret) {
 		zap_available = false;
-		return -EINVAL;
-	}
-
-	ret = of_address_to_resource(mem_np, 0, &r);
-	of_node_put(mem_np);
-	if (ret)
 		return ret;
-
+	}
 	mem_phys = r.start;
 
 	/*

From 505441083afb9dde61d6b82bfa9d7cc6ad331e77 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Fri, 4 Jul 2025 19:31:53 +0300
Subject: [PATCH 0741/1455] dt-bindings: display/msm: dsi-controller-main: add
 SC8180X

Describe the SC8180X-specific compatible for the DSI controller persent
on the SoC. While the current DT for SC8180X doesn't use this
compatible, all other platforms were updated to have one. This change
makes SC8180X follow the lead.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/662495/
Link: https://lore.kernel.org/r/20250704-mdss-schema-v1-1-e978e4e73e14@oss.qualcomm.com
---
 .../devicetree/bindings/display/msm/dsi-controller-main.yaml    | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
index d4bb65c660af..4400d4cce072 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dsi-controller-main.yaml
@@ -27,6 +27,7 @@ properties:
               - qcom,sar2130p-dsi-ctrl
               - qcom,sc7180-dsi-ctrl
               - qcom,sc7280-dsi-ctrl
+              - qcom,sc8180x-dsi-ctrl
               - qcom,sdm660-dsi-ctrl
               - qcom,sdm670-dsi-ctrl
               - qcom,sdm845-dsi-ctrl
@@ -332,6 +333,7 @@ allOf:
               - qcom,sar2130p-dsi-ctrl
               - qcom,sc7180-dsi-ctrl
               - qcom,sc7280-dsi-ctrl
+              - qcom,sc8180x-dsi-ctrl
               - qcom,sdm845-dsi-ctrl
               - qcom,sm6115-dsi-ctrl
               - qcom,sm6125-dsi-ctrl

From 275af16d7294e7f8a36370f274d3aa9dd01ad76a Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Fri, 4 Jul 2025 19:31:54 +0300
Subject: [PATCH 0742/1455] dt-bindings: display/msm: describe DPU on SC8180X

Describe the Display Processing Unit (DPU) as present on the SC8180X
platform.

Reported-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/662502/
Link: https://lore.kernel.org/r/20250704-mdss-schema-v1-2-e978e4e73e14@oss.qualcomm.com
---
 .../display/msm/qcom,sc8180x-dpu.yaml         | 103 ++++++++++++++++++
 1 file changed, 103 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/msm/qcom,sc8180x-dpu.yaml

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sc8180x-dpu.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sc8180x-dpu.yaml
new file mode 100644
index 000000000000..a411126708b8
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sc8180x-dpu.yaml
@@ -0,0 +1,103 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sc8180x-dpu.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SC8180X Display DPU
+
+maintainers:
+  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
+
+$ref: /schemas/display/msm/dpu-common.yaml#
+
+properties:
+  compatible:
+    const: qcom,sc8180x-dpu
+
+  reg:
+    items:
+      - description: Address offset and size for mdp register set
+      - description: Address offset and size for vbif register set
+
+  reg-names:
+    items:
+      - const: mdp
+      - const: vbif
+
+  clocks:
+    items:
+      - description: Display AHB clock
+      - description: Display HF AXI clock
+      - description: Display core clock
+      - description: Display vsync clock
+      - description: Display rotator clock
+      - description: Display LUT clock
+
+  clock-names:
+    items:
+      - const: iface
+      - const: bus
+      - const: core
+      - const: vsync
+      - const: rot
+      - const: lut
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,dispcc-sm8250.h>
+    #include <dt-bindings/clock/qcom,gcc-sc8180x.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/interconnect/qcom,sc8180x.h>
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    display-controller@ae01000 {
+        compatible = "qcom,sc8180x-dpu";
+        reg = <0x0ae01000 0x8f000>,
+              <0x0aeb0000 0x2008>;
+        reg-names = "mdp", "vbif";
+
+        clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                 <&gcc GCC_DISP_HF_AXI_CLK>,
+                 <&dispcc DISP_CC_MDSS_MDP_CLK>,
+                 <&dispcc DISP_CC_MDSS_VSYNC_CLK>,
+                 <&dispcc DISP_CC_MDSS_ROT_CLK>,
+                 <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>;
+        clock-names = "iface",
+                      "bus",
+                      "core",
+                      "vsync",
+                      "rot",
+                      "lut";
+
+        assigned-clocks = <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+        assigned-clock-rates = <19200000>;
+
+        operating-points-v2 = <&mdp_opp_table>;
+        power-domains = <&rpmhpd SC8180X_MMCX>;
+
+        interrupt-parent = <&mdss>;
+        interrupts = <0>;
+
+        ports {
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            port@0 {
+                reg = <0>;
+                endpoint {
+                    remote-endpoint = <&dsi0_in>;
+                };
+            };
+
+            port@1 {
+                reg = <1>;
+                endpoint {
+                    remote-endpoint = <&dsi1_in>;
+                };
+            };
+        };
+    };
+...

From fd5c9b3c99da614a71119fce1848e66650a6633f Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Fri, 4 Jul 2025 19:31:55 +0300
Subject: [PATCH 0743/1455] dt-bindings: display/msm: describe MDSS on SC8180X

Describe the Mobile Display SubSystem (MDSS) unit as present on the
SC8180X platform.

Reported-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/662498/
Link: https://lore.kernel.org/r/20250704-mdss-schema-v1-3-e978e4e73e14@oss.qualcomm.com
---
 .../display/msm/qcom,sc8180x-mdss.yaml        | 359 ++++++++++++++++++
 1 file changed, 359 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/msm/qcom,sc8180x-mdss.yaml

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sc8180x-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sc8180x-mdss.yaml
new file mode 100644
index 000000000000..00e82bdbbcc7
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sc8180x-mdss.yaml
@@ -0,0 +1,359 @@
+# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/msm/qcom,sc8180x-mdss.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm SC8180X Display MDSS
+
+maintainers:
+  - Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
+
+description:
+  Device tree bindings for MSM Mobile Display Subsystem(MDSS) that encapsulates
+  sub-blocks like DPU display controller, DSI and DP interfaces etc. Device tree
+  bindings of MDSS are mentioned for SC8180X target.
+
+$ref: /schemas/display/msm/mdss-common.yaml#
+
+properties:
+  compatible:
+    items:
+      - const: qcom,sc8180x-mdss
+
+  clocks:
+    items:
+      - description: Display AHB clock from gcc
+      - description: Display hf axi clock
+      - description: Display sf axi clock
+      - description: Display core clock
+
+  clock-names:
+    items:
+      - const: iface
+      - const: bus
+      - const: nrt_bus
+      - const: core
+
+  iommus:
+    maxItems: 1
+
+  interconnects:
+    maxItems: 3
+
+  interconnect-names:
+    maxItems: 3
+
+patternProperties:
+  "^display-controller@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+
+    properties:
+      compatible:
+        const: qcom,sc8180x-dpu
+
+  "^displayport-controller@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+
+    properties:
+      compatible:
+        enum:
+          - qcom,sc8180x-dp
+          - qcom,sc8180x-edp
+
+  "^dsi@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+
+    properties:
+      compatible:
+        contains:
+          const: qcom,sc8180x-dsi-ctrl
+
+  "^phy@[0-9a-f]+$":
+    type: object
+    additionalProperties: true
+
+    properties:
+      compatible:
+        const: qcom,dsi-phy-7nm
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/clock/qcom,dispcc-sm8250.h>
+    #include <dt-bindings/clock/qcom,gcc-sc8180x.h>
+    #include <dt-bindings/clock/qcom,rpmh.h>
+    #include <dt-bindings/interrupt-controller/arm-gic.h>
+    #include <dt-bindings/interconnect/qcom,sc8180x.h>
+    #include <dt-bindings/power/qcom-rpmpd.h>
+
+    display-subsystem@ae00000 {
+        compatible = "qcom,sc8180x-mdss";
+        reg = <0x0ae00000 0x1000>;
+        reg-names = "mdss";
+
+        interconnects = <&mmss_noc MASTER_MDP_PORT0 &mc_virt SLAVE_EBI_CH0>,
+                        <&mmss_noc MASTER_MDP_PORT1 &mc_virt SLAVE_EBI_CH0>,
+                        <&gem_noc MASTER_AMPSS_M0 &config_noc SLAVE_DISPLAY_CFG>;
+        interconnect-names = "mdp0-mem",
+                             "mdp1-mem",
+                             "cpu-cfg";
+
+        power-domains = <&dispcc MDSS_GDSC>;
+
+        clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                 <&gcc GCC_DISP_HF_AXI_CLK>,
+                 <&gcc GCC_DISP_SF_AXI_CLK>,
+                 <&dispcc DISP_CC_MDSS_MDP_CLK>;
+        clock-names = "iface", "bus", "nrt_bus", "core";
+
+        interrupts = <GIC_SPI 83 IRQ_TYPE_LEVEL_HIGH>;
+        interrupt-controller;
+        #interrupt-cells = <1>;
+
+        iommus = <&apps_smmu 0x800 0x420>;
+
+        #address-cells = <1>;
+        #size-cells = <1>;
+        ranges;
+
+        display-controller@ae01000 {
+            compatible = "qcom,sc8180x-dpu";
+            reg = <0x0ae01000 0x8f000>,
+                  <0x0aeb0000 0x2008>;
+            reg-names = "mdp", "vbif";
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_HF_AXI_CLK>,
+                     <&dispcc DISP_CC_MDSS_MDP_CLK>,
+                     <&dispcc DISP_CC_MDSS_VSYNC_CLK>,
+                     <&dispcc DISP_CC_MDSS_ROT_CLK>,
+                     <&dispcc DISP_CC_MDSS_MDP_LUT_CLK>;
+            clock-names = "iface",
+                          "bus",
+                          "core",
+                          "vsync",
+                          "rot",
+                          "lut";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_VSYNC_CLK>;
+            assigned-clock-rates = <19200000>;
+
+            operating-points-v2 = <&mdp_opp_table>;
+            power-domains = <&rpmhpd SC8180X_MMCX>;
+
+            interrupt-parent = <&mdss>;
+            interrupts = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dpu_intf1_out: endpoint {
+                        remote-endpoint = <&dsi0_in>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dpu_intf2_out: endpoint {
+                        remote-endpoint = <&dsi1_in>;
+                    };
+                };
+            };
+
+            mdp_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-171428571 {
+                    opp-hz = /bits/ 64 <171428571>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-300000000 {
+                    opp-hz = /bits/ 64 <300000000>;
+                    required-opps = <&rpmhpd_opp_svs>;
+                };
+
+                opp-345000000 {
+                    opp-hz = /bits/ 64 <345000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+
+                opp-460000000 {
+                    opp-hz = /bits/ 64 <460000000>;
+                    required-opps = <&rpmhpd_opp_nom>;
+                };
+            };
+        };
+
+        dsi@ae94000 {
+            compatible = "qcom,sc8180x-dsi-ctrl",
+                         "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae94000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <4>;
+
+            clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK>,
+                     <&dispcc DISP_CC_MDSS_BYTE0_INTF_CLK>,
+                     <&dispcc DISP_CC_MDSS_PCLK0_CLK>,
+                     <&dispcc DISP_CC_MDSS_ESC0_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_HF_AXI_CLK>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE0_CLK_SRC>,
+                              <&dispcc DISP_CC_MDSS_PCLK0_CLK_SRC>;
+            assigned-clock-parents = <&dsi0_phy 0>, <&dsi0_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd SC8180X_MMCX>;
+
+            phys = <&dsi0_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dsi0_in: endpoint {
+                        remote-endpoint = <&dpu_intf1_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dsi0_out: endpoint {
+                    };
+                };
+            };
+
+            dsi_opp_table: opp-table {
+                compatible = "operating-points-v2";
+
+                opp-187500000 {
+                    opp-hz = /bits/ 64 <187500000>;
+                    required-opps = <&rpmhpd_opp_low_svs>;
+                };
+
+                opp-300000000 {
+                    opp-hz = /bits/ 64 <300000000>;
+                    required-opps = <&rpmhpd_opp_svs>;
+                };
+
+                opp-358000000 {
+                    opp-hz = /bits/ 64 <358000000>;
+                    required-opps = <&rpmhpd_opp_svs_l1>;
+                };
+            };
+        };
+
+        dsi0_phy: phy@ae94400 {
+            compatible = "qcom,dsi-phy-7nm";
+            reg = <0x0ae94400 0x200>,
+                  <0x0ae94600 0x280>,
+                  <0x0ae94900 0x260>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&rpmhcc RPMH_CXO_CLK>;
+            clock-names = "iface", "ref";
+            vdds-supply = <&vreg_dsi_phy>;
+        };
+
+        dsi@ae96000 {
+            compatible = "qcom,sc8180x-dsi-ctrl",
+                         "qcom,mdss-dsi-ctrl";
+            reg = <0x0ae96000 0x400>;
+            reg-names = "dsi_ctrl";
+
+            interrupt-parent = <&mdss>;
+            interrupts = <5>;
+
+            clocks = <&dispcc DISP_CC_MDSS_BYTE1_CLK>,
+                     <&dispcc DISP_CC_MDSS_BYTE1_INTF_CLK>,
+                     <&dispcc DISP_CC_MDSS_PCLK1_CLK>,
+                     <&dispcc DISP_CC_MDSS_ESC1_CLK>,
+                     <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&gcc GCC_DISP_HF_AXI_CLK>;
+            clock-names = "byte",
+                          "byte_intf",
+                          "pixel",
+                          "core",
+                          "iface",
+                          "bus";
+
+            assigned-clocks = <&dispcc DISP_CC_MDSS_BYTE1_CLK_SRC>,
+                              <&dispcc DISP_CC_MDSS_PCLK1_CLK_SRC>;
+            assigned-clock-parents = <&dsi1_phy 0>, <&dsi1_phy 1>;
+
+            operating-points-v2 = <&dsi_opp_table>;
+            power-domains = <&rpmhpd SC8180X_MMCX>;
+
+            phys = <&dsi1_phy>;
+            phy-names = "dsi";
+
+            #address-cells = <1>;
+            #size-cells = <0>;
+
+            ports {
+                #address-cells = <1>;
+                #size-cells = <0>;
+
+                port@0 {
+                    reg = <0>;
+                    dsi1_in: endpoint {
+                        remote-endpoint = <&dpu_intf2_out>;
+                    };
+                };
+
+                port@1 {
+                    reg = <1>;
+                    dsi1_out: endpoint {
+                    };
+                };
+            };
+        };
+
+        dsi1_phy: phy@ae96400 {
+            compatible = "qcom,dsi-phy-7nm";
+            reg = <0x0ae96400 0x200>,
+                  <0x0ae96600 0x280>,
+                  <0x0ae96900 0x260>;
+            reg-names = "dsi_phy",
+                        "dsi_phy_lane",
+                        "dsi_pll";
+
+            #clock-cells = <1>;
+            #phy-cells = <0>;
+
+            clocks = <&dispcc DISP_CC_MDSS_AHB_CLK>,
+                     <&rpmhcc RPMH_CXO_CLK>;
+            clock-names = "iface", "ref";
+            vdds-supply = <&vreg_dsi_phy>;
+        };
+    };
+...

From 12c3c6c44d1b0fd17f18522089b12ad269769c00 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Fri, 18 Jul 2025 14:34:04 +0100
Subject: [PATCH 0744/1455] drm/msm: remove extraneous semicolon after a
 statement

There is a statement that has an extraneous semicolon; remove it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/664675/
Link: https://lore.kernel.org/r/20250718133404.2047673-1-colin.i.king@gmail.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/msm_gpu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c
index 26c5ce897cbb..17759abc46d7 100644
--- a/drivers/gpu/drm/msm/msm_gpu.c
+++ b/drivers/gpu/drm/msm/msm_gpu.c
@@ -304,7 +304,7 @@ static void crashstate_get_bos(struct msm_gpu_state *state, struct msm_gem_submi
 			sizeof(struct msm_gpu_state_bo), GFP_KERNEL);
 
 		for (int i = 0; state->bos && i < submit->nr_bos; i++) {
-			struct drm_gem_object *obj = submit->bos[i].obj;;
+			struct drm_gem_object *obj = submit->bos[i].obj;
 			bool dump = rd_full || (submit->bos[i].flags & MSM_SUBMIT_BO_DUMP);
 
 			msm_gem_lock(obj);

From 62b7d68352881609e237b303fa391410ebc583a5 Mon Sep 17 00:00:00 2001
From: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Date: Tue, 6 May 2025 18:38:39 -0700
Subject: [PATCH 0745/1455] drm/msm/dpu: Filter modes based on adjusted mode
 clock

Filter out modes that have a clock rate greater than the max core clock
rate when adjusted for the perf clock factor

This is especially important for chipsets such as QCS615 that have lower
limits for the MDP max core clock.

Since the core CRTC clock is at least the mode clock (adjusted for the
perf clock factor) [1], the modes supported by the driver should be less
than the max core clock rate.

[1] https://elixir.bootlin.com/linux/v6.12.4/source/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c#L83

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/652041/
Link: https://lore.kernel.org/r/20250506-filter-modes-v2-1-c20a0b7aa241@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c | 35 +++++++++++++------
 drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h |  3 ++
 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c      | 12 +++++++
 3 files changed, 39 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c
index 0fb5789c60d0..13cc658065c5 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c
@@ -31,6 +31,26 @@ enum dpu_perf_mode {
 	DPU_PERF_MODE_MAX
 };
 
+/**
+ * dpu_core_perf_adjusted_mode_clk - Adjust given mode clock rate according to
+ *   the perf clock factor.
+ * @crtc_clk_rate - Unadjusted mode clock rate
+ * @perf_cfg: performance configuration
+ */
+u64 dpu_core_perf_adjusted_mode_clk(u64 mode_clk_rate,
+				    const struct dpu_perf_cfg *perf_cfg)
+{
+	u32 clk_factor;
+
+	clk_factor = perf_cfg->clk_inefficiency_factor;
+	if (clk_factor) {
+		mode_clk_rate *= clk_factor;
+		do_div(mode_clk_rate, 100);
+	}
+
+	return mode_clk_rate;
+}
+
 /**
  * _dpu_core_perf_calc_bw() - to calculate BW per crtc
  * @perf_cfg: performance configuration
@@ -75,28 +95,21 @@ static u64 _dpu_core_perf_calc_clk(const struct dpu_perf_cfg *perf_cfg,
 	struct drm_plane *plane;
 	struct dpu_plane_state *pstate;
 	struct drm_display_mode *mode;
-	u64 crtc_clk;
-	u32 clk_factor;
+	u64 mode_clk;
 
 	mode = &state->adjusted_mode;
 
-	crtc_clk = (u64)mode->vtotal * mode->hdisplay * drm_mode_vrefresh(mode);
+	mode_clk = (u64)mode->vtotal * mode->hdisplay * drm_mode_vrefresh(mode);
 
 	drm_atomic_crtc_for_each_plane(plane, crtc) {
 		pstate = to_dpu_plane_state(plane->state);
 		if (!pstate)
 			continue;
 
-		crtc_clk = max(pstate->plane_clk, crtc_clk);
+		mode_clk = max(pstate->plane_clk, mode_clk);
 	}
 
-	clk_factor = perf_cfg->clk_inefficiency_factor;
-	if (clk_factor) {
-		crtc_clk *= clk_factor;
-		do_div(crtc_clk, 100);
-	}
-
-	return crtc_clk;
+	return dpu_core_perf_adjusted_mode_clk(mode_clk, perf_cfg);
 }
 
 static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc)
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h
index d2f21d34e501..3740bc97422c 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h
@@ -54,6 +54,9 @@ struct dpu_core_perf {
 	u32 fix_core_ab_vote;
 };
 
+u64 dpu_core_perf_adjusted_mode_clk(u64 clk_rate,
+				    const struct dpu_perf_cfg *perf_cfg);
+
 int dpu_core_perf_crtc_check(struct drm_crtc *crtc,
 		struct drm_crtc_state *state);
 
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index 94912b4708fb..d59512e45af0 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -1534,6 +1534,7 @@ static enum drm_mode_status dpu_crtc_mode_valid(struct drm_crtc *crtc,
 						const struct drm_display_mode *mode)
 {
 	struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc);
+	u64 adjusted_mode_clk;
 
 	/* if there is no 3d_mux block we cannot merge LMs so we cannot
 	 * split the large layer into 2 LMs, filter out such modes
@@ -1541,6 +1542,17 @@ static enum drm_mode_status dpu_crtc_mode_valid(struct drm_crtc *crtc,
 	if (!dpu_kms->catalog->caps->has_3d_merge &&
 	    mode->hdisplay > dpu_kms->catalog->caps->max_mixer_width)
 		return MODE_BAD_HVALUE;
+
+	adjusted_mode_clk = dpu_core_perf_adjusted_mode_clk(mode->clock,
+							    dpu_kms->perf.perf_cfg);
+
+	/*
+	 * The given mode, adjusted for the perf clock factor, should not exceed
+	 * the max core clock rate
+	 */
+	if (dpu_kms->perf.max_core_clk_rate < adjusted_mode_clk * 1000)
+		return MODE_CLOCK_HIGH;
+
 	/*
 	 * max crtc width is equal to the max mixer width * 2 and max height is 4K
 	 */

From a7d17b4f8bc862dd9222b4506d8e3db0b895a1c7 Mon Sep 17 00:00:00 2001
From: "Jiri Slaby (SUSE)" <jirislaby@kernel.org>
Date: Wed, 23 Jul 2025 07:55:12 +0200
Subject: [PATCH 0746/1455] drm/msm: use dev_fwnode()

irq_domain_create_simple() takes fwnode as the first argument. It can be
extracted from the struct device using dev_fwnode() helper instead of
using of_node with of_fwnode_handle().

So use the dev_fwnode() helper.

Signed-off-by: Jiri Slaby (SUSE) <jirislaby@kernel.org>
Cc: Rob Clark <robdclark@gmail.com>
Cc: Abhinav Kumar <quic_abhinavk@quicinc.com>
Cc: Dmitry Baryshkov <lumag@kernel.org>
Cc: Sean Paul <sean@poorly.run>
Cc: Marijn Suijten <marijn.suijten@somainline.org>
Cc: David Airlie <airlied@gmail.com>
Cc: Simona Vetter <simona@ffwll.ch>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/665227/
Link: https://lore.kernel.org/r/20250723055512.1800438-1-jirislaby@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/msm_mdss.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_mdss.c b/drivers/gpu/drm/msm/msm_mdss.c
index 39885b333910..2d0e3e784c04 100644
--- a/drivers/gpu/drm/msm/msm_mdss.c
+++ b/drivers/gpu/drm/msm/msm_mdss.c
@@ -154,8 +154,7 @@ static int _msm_mdss_irq_domain_add(struct msm_mdss *msm_mdss)
 
 	dev = msm_mdss->dev;
 
-	domain = irq_domain_create_linear(of_fwnode_handle(dev->of_node), 32,
-			&msm_mdss_irqdomain_ops, msm_mdss);
+	domain = irq_domain_create_linear(dev_fwnode(dev), 32, &msm_mdss_irqdomain_ops, msm_mdss);
 	if (!domain) {
 		dev_err(dev, "failed to add irq_domain\n");
 		return -EINVAL;

From fe3190a3915f3de62ea3bba2dc1d80de0592fa41 Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:25 -0400
Subject: [PATCH 0747/1455] drm/msm/dsi_phy_10nm: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667866/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-1-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c
index af2e30f3f842..ec486ff02c9b 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c
@@ -444,21 +444,19 @@ static unsigned long dsi_pll_10nm_vco_recalc_rate(struct clk_hw *hw,
 	return (unsigned long)vco_rate;
 }
 
-static long dsi_pll_10nm_clk_round_rate(struct clk_hw *hw,
-		unsigned long rate, unsigned long *parent_rate)
+static int dsi_pll_10nm_clk_determine_rate(struct clk_hw *hw,
+					   struct clk_rate_request *req)
 {
 	struct dsi_pll_10nm *pll_10nm = to_pll_10nm(hw);
 
-	if      (rate < pll_10nm->phy->cfg->min_pll_rate)
-		return  pll_10nm->phy->cfg->min_pll_rate;
-	else if (rate > pll_10nm->phy->cfg->max_pll_rate)
-		return  pll_10nm->phy->cfg->max_pll_rate;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate,
+			    pll_10nm->phy->cfg->min_pll_rate, pll_10nm->phy->cfg->max_pll_rate);
+
+	return 0;
 }
 
 static const struct clk_ops clk_ops_dsi_pll_10nm_vco = {
-	.round_rate = dsi_pll_10nm_clk_round_rate,
+	.determine_rate = dsi_pll_10nm_clk_determine_rate,
 	.set_rate = dsi_pll_10nm_vco_set_rate,
 	.recalc_rate = dsi_pll_10nm_vco_recalc_rate,
 	.prepare = dsi_pll_10nm_vco_prepare,

From cc41f29a6b04d8b7ee7cefd77334aeba2598646f Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:26 -0400
Subject: [PATCH 0748/1455] drm/msm/dsi_phy_14nm: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667869/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-2-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c | 34 +++++++++++-----------
 1 file changed, 17 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c
index 3a1c8ece6657..fdefcbd9c284 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c
@@ -578,21 +578,19 @@ static void dsi_pll_14nm_vco_unprepare(struct clk_hw *hw)
 	pll_14nm->phy->pll_on = false;
 }
 
-static long dsi_pll_14nm_clk_round_rate(struct clk_hw *hw,
-		unsigned long rate, unsigned long *parent_rate)
+static int dsi_pll_14nm_clk_determine_rate(struct clk_hw *hw,
+					   struct clk_rate_request *req)
 {
 	struct dsi_pll_14nm *pll_14nm = to_pll_14nm(hw);
 
-	if      (rate < pll_14nm->phy->cfg->min_pll_rate)
-		return  pll_14nm->phy->cfg->min_pll_rate;
-	else if (rate > pll_14nm->phy->cfg->max_pll_rate)
-		return  pll_14nm->phy->cfg->max_pll_rate;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate,
+			    pll_14nm->phy->cfg->min_pll_rate, pll_14nm->phy->cfg->max_pll_rate);
+
+	return 0;
 }
 
 static const struct clk_ops clk_ops_dsi_pll_14nm_vco = {
-	.round_rate = dsi_pll_14nm_clk_round_rate,
+	.determine_rate = dsi_pll_14nm_clk_determine_rate,
 	.set_rate = dsi_pll_14nm_vco_set_rate,
 	.recalc_rate = dsi_pll_14nm_vco_recalc_rate,
 	.prepare = dsi_pll_14nm_vco_prepare,
@@ -622,18 +620,20 @@ static unsigned long dsi_pll_14nm_postdiv_recalc_rate(struct clk_hw *hw,
 				   postdiv->flags, width);
 }
 
-static long dsi_pll_14nm_postdiv_round_rate(struct clk_hw *hw,
-					    unsigned long rate,
-					    unsigned long *prate)
+static int dsi_pll_14nm_postdiv_determine_rate(struct clk_hw *hw,
+					       struct clk_rate_request *req)
 {
 	struct dsi_pll_14nm_postdiv *postdiv = to_pll_14nm_postdiv(hw);
 	struct dsi_pll_14nm *pll_14nm = postdiv->pll;
 
-	DBG("DSI%d PLL parent rate=%lu", pll_14nm->phy->id, rate);
+	DBG("DSI%d PLL parent rate=%lu", pll_14nm->phy->id, req->rate);
 
-	return divider_round_rate(hw, rate, prate, NULL,
-				  postdiv->width,
-				  postdiv->flags);
+	req->rate = divider_round_rate(hw, req->rate, &req->best_parent_rate,
+				       NULL,
+				       postdiv->width,
+				       postdiv->flags);
+
+	return 0;
 }
 
 static int dsi_pll_14nm_postdiv_set_rate(struct clk_hw *hw, unsigned long rate,
@@ -680,7 +680,7 @@ static int dsi_pll_14nm_postdiv_set_rate(struct clk_hw *hw, unsigned long rate,
 
 static const struct clk_ops clk_ops_dsi_pll_14nm_postdiv = {
 	.recalc_rate = dsi_pll_14nm_postdiv_recalc_rate,
-	.round_rate = dsi_pll_14nm_postdiv_round_rate,
+	.determine_rate = dsi_pll_14nm_postdiv_determine_rate,
 	.set_rate = dsi_pll_14nm_postdiv_set_rate,
 };
 

From 267c0a2dfbdb4ba27c7333c37fc56f3d5f05919e Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:27 -0400
Subject: [PATCH 0749/1455] drm/msm/dsi_phy_28nm_8960: convert from
 round_rate() to determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667870/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-3-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c   | 32 +++++++++----------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c
index f3643320ff2f..8dcce9581dc3 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c
@@ -231,21 +231,19 @@ static void dsi_pll_28nm_vco_unprepare(struct clk_hw *hw)
 	pll_28nm->phy->pll_on = false;
 }
 
-static long dsi_pll_28nm_clk_round_rate(struct clk_hw *hw,
-		unsigned long rate, unsigned long *parent_rate)
+static int dsi_pll_28nm_clk_determine_rate(struct clk_hw *hw,
+					   struct clk_rate_request *req)
 {
 	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(hw);
 
-	if      (rate < pll_28nm->phy->cfg->min_pll_rate)
-		return  pll_28nm->phy->cfg->min_pll_rate;
-	else if (rate > pll_28nm->phy->cfg->max_pll_rate)
-		return  pll_28nm->phy->cfg->max_pll_rate;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate,
+			    pll_28nm->phy->cfg->min_pll_rate, pll_28nm->phy->cfg->max_pll_rate);
+
+	return 0;
 }
 
 static const struct clk_ops clk_ops_dsi_pll_28nm_vco = {
-	.round_rate = dsi_pll_28nm_clk_round_rate,
+	.determine_rate = dsi_pll_28nm_clk_determine_rate,
 	.set_rate = dsi_pll_28nm_clk_set_rate,
 	.recalc_rate = dsi_pll_28nm_clk_recalc_rate,
 	.prepare = dsi_pll_28nm_vco_prepare,
@@ -296,18 +294,20 @@ static unsigned int get_vco_mul_factor(unsigned long byte_clk_rate)
 		return 8;
 }
 
-static long clk_bytediv_round_rate(struct clk_hw *hw, unsigned long rate,
-				   unsigned long *prate)
+static int clk_bytediv_determine_rate(struct clk_hw *hw,
+				      struct clk_rate_request *req)
 {
 	unsigned long best_parent;
 	unsigned int factor;
 
-	factor = get_vco_mul_factor(rate);
+	factor = get_vco_mul_factor(req->rate);
 
-	best_parent = rate * factor;
-	*prate = clk_hw_round_rate(clk_hw_get_parent(hw), best_parent);
+	best_parent = req->rate * factor;
+	req->best_parent_rate = clk_hw_round_rate(clk_hw_get_parent(hw), best_parent);
 
-	return *prate / factor;
+	req->rate = req->best_parent_rate / factor;
+
+	return 0;
 }
 
 static int clk_bytediv_set_rate(struct clk_hw *hw, unsigned long rate,
@@ -328,7 +328,7 @@ static int clk_bytediv_set_rate(struct clk_hw *hw, unsigned long rate,
 
 /* Our special byte clock divider ops */
 static const struct clk_ops clk_bytediv_ops = {
-	.round_rate = clk_bytediv_round_rate,
+	.determine_rate = clk_bytediv_determine_rate,
 	.set_rate = clk_bytediv_set_rate,
 	.recalc_rate = clk_bytediv_recalc_rate,
 };

From e04baa0b175627428424398623034c13f9decfb3 Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:28 -0400
Subject: [PATCH 0750/1455] drm/msm/dsi_phy_28nm: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667872/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-4-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
index 90348a2af3e9..d00e415b9a99 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c
@@ -533,21 +533,20 @@ static void dsi_pll_28nm_vco_unprepare(struct clk_hw *hw)
 	pll_28nm->phy->pll_on = false;
 }
 
-static long dsi_pll_28nm_clk_round_rate(struct clk_hw *hw,
-		unsigned long rate, unsigned long *parent_rate)
+static int dsi_pll_28nm_clk_determine_rate(struct clk_hw *hw,
+					   struct clk_rate_request *req)
 {
 	struct dsi_pll_28nm *pll_28nm = to_pll_28nm(hw);
 
-	if      (rate < pll_28nm->phy->cfg->min_pll_rate)
-		return  pll_28nm->phy->cfg->min_pll_rate;
-	else if (rate > pll_28nm->phy->cfg->max_pll_rate)
-		return  pll_28nm->phy->cfg->max_pll_rate;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate,
+			    pll_28nm->phy->cfg->min_pll_rate,
+			    pll_28nm->phy->cfg->max_pll_rate);
+
+	return 0;
 }
 
 static const struct clk_ops clk_ops_dsi_pll_28nm_vco_hpm = {
-	.round_rate = dsi_pll_28nm_clk_round_rate,
+	.determine_rate = dsi_pll_28nm_clk_determine_rate,
 	.set_rate = dsi_pll_28nm_clk_set_rate,
 	.recalc_rate = dsi_pll_28nm_clk_recalc_rate,
 	.prepare = dsi_pll_28nm_vco_prepare_hpm,
@@ -556,7 +555,7 @@ static const struct clk_ops clk_ops_dsi_pll_28nm_vco_hpm = {
 };
 
 static const struct clk_ops clk_ops_dsi_pll_28nm_vco_lp = {
-	.round_rate = dsi_pll_28nm_clk_round_rate,
+	.determine_rate = dsi_pll_28nm_clk_determine_rate,
 	.set_rate = dsi_pll_28nm_clk_set_rate,
 	.recalc_rate = dsi_pll_28nm_clk_recalc_rate,
 	.prepare = dsi_pll_28nm_vco_prepare_lp,
@@ -565,7 +564,7 @@ static const struct clk_ops clk_ops_dsi_pll_28nm_vco_lp = {
 };
 
 static const struct clk_ops clk_ops_dsi_pll_28nm_vco_8226 = {
-	.round_rate = dsi_pll_28nm_clk_round_rate,
+	.determine_rate = dsi_pll_28nm_clk_determine_rate,
 	.set_rate = dsi_pll_28nm_clk_set_rate,
 	.recalc_rate = dsi_pll_28nm_clk_recalc_rate,
 	.prepare = dsi_pll_28nm_vco_prepare_8226,

From 6e5173e7bf80bdd127b659b210e773695d78d1ea Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:29 -0400
Subject: [PATCH 0751/1455] drm/msm/dsi_phy_7nm: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667874/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-5-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
index 8c98f91a5930..c3bd3f89434e 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -561,21 +561,19 @@ static unsigned long dsi_pll_7nm_vco_recalc_rate(struct clk_hw *hw,
 	return (unsigned long)vco_rate;
 }
 
-static long dsi_pll_7nm_clk_round_rate(struct clk_hw *hw,
-		unsigned long rate, unsigned long *parent_rate)
+static int dsi_pll_7nm_clk_determine_rate(struct clk_hw *hw,
+					  struct clk_rate_request *req)
 {
 	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(hw);
 
-	if      (rate < pll_7nm->phy->cfg->min_pll_rate)
-		return  pll_7nm->phy->cfg->min_pll_rate;
-	else if (rate > pll_7nm->phy->cfg->max_pll_rate)
-		return  pll_7nm->phy->cfg->max_pll_rate;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate,
+			    pll_7nm->phy->cfg->min_pll_rate, pll_7nm->phy->cfg->max_pll_rate);
+
+	return 0;
 }
 
 static const struct clk_ops clk_ops_dsi_pll_7nm_vco = {
-	.round_rate = dsi_pll_7nm_clk_round_rate,
+	.determine_rate = dsi_pll_7nm_clk_determine_rate,
 	.set_rate = dsi_pll_7nm_vco_set_rate,
 	.recalc_rate = dsi_pll_7nm_vco_recalc_rate,
 	.prepare = dsi_pll_7nm_vco_prepare,

From 5beef1fec2afb5414ddadac4a3ab7d00039dba60 Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:30 -0400
Subject: [PATCH 0752/1455] drm/msm/hdmi_phy_8996: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667876/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-6-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c
index 8c8d80b59573..36e928b0fd5a 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c
@@ -629,16 +629,12 @@ static int hdmi_8996_pll_prepare(struct clk_hw *hw)
 	return 0;
 }
 
-static long hdmi_8996_pll_round_rate(struct clk_hw *hw,
-				     unsigned long rate,
-				     unsigned long *parent_rate)
+static int hdmi_8996_pll_determine_rate(struct clk_hw *hw,
+					struct clk_rate_request *req)
 {
-	if (rate < HDMI_PCLK_MIN_FREQ)
-		return HDMI_PCLK_MIN_FREQ;
-	else if (rate > HDMI_PCLK_MAX_FREQ)
-		return HDMI_PCLK_MAX_FREQ;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate, HDMI_PCLK_MIN_FREQ, HDMI_PCLK_MAX_FREQ);
+
+	return 0;
 }
 
 static unsigned long hdmi_8996_pll_recalc_rate(struct clk_hw *hw,
@@ -684,7 +680,7 @@ static int hdmi_8996_pll_is_enabled(struct clk_hw *hw)
 
 static const struct clk_ops hdmi_8996_pll_ops = {
 	.set_rate = hdmi_8996_pll_set_clk_rate,
-	.round_rate = hdmi_8996_pll_round_rate,
+	.determine_rate = hdmi_8996_pll_determine_rate,
 	.recalc_rate = hdmi_8996_pll_recalc_rate,
 	.prepare = hdmi_8996_pll_prepare,
 	.unprepare = hdmi_8996_pll_unprepare,

From a38d1fbbc90f5f2ddf632103064efa83d8f332fe Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Sun, 10 Aug 2025 18:57:31 -0400
Subject: [PATCH 0753/1455] drm/msm/hdmi_phy_8998: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series. The change to use clamp_t() was
done manually.

Signed-off-by: Brian Masney <bmasney@redhat.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/667878/
Link: https://lore.kernel.org/r/20250810-drm-msm-phy-clk-round-rate-v2-7-0fd1f7979c83@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c
index 33bb48ae58a2..a86ff3706369 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8998.c
@@ -646,16 +646,12 @@ static int hdmi_8998_pll_prepare(struct clk_hw *hw)
 	return 0;
 }
 
-static long hdmi_8998_pll_round_rate(struct clk_hw *hw,
-				     unsigned long rate,
-				     unsigned long *parent_rate)
+static int hdmi_8998_pll_determine_rate(struct clk_hw *hw,
+					struct clk_rate_request *req)
 {
-	if (rate < HDMI_PCLK_MIN_FREQ)
-		return HDMI_PCLK_MIN_FREQ;
-	else if (rate > HDMI_PCLK_MAX_FREQ)
-		return HDMI_PCLK_MAX_FREQ;
-	else
-		return rate;
+	req->rate = clamp_t(unsigned long, req->rate, HDMI_PCLK_MIN_FREQ, HDMI_PCLK_MAX_FREQ);
+
+	return 0;
 }
 
 static unsigned long hdmi_8998_pll_recalc_rate(struct clk_hw *hw,
@@ -688,7 +684,7 @@ static int hdmi_8998_pll_is_enabled(struct clk_hw *hw)
 
 static const struct clk_ops hdmi_8998_pll_ops = {
 	.set_rate = hdmi_8998_pll_set_clk_rate,
-	.round_rate = hdmi_8998_pll_round_rate,
+	.determine_rate = hdmi_8998_pll_determine_rate,
 	.recalc_rate = hdmi_8998_pll_recalc_rate,
 	.prepare = hdmi_8998_pll_prepare,
 	.unprepare = hdmi_8998_pll_unprepare,

From 470085357f77b922c834d38d25e6e2fa290b7cdd Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Mon, 11 Aug 2025 06:56:07 -0400
Subject: [PATCH 0754/1455] drm/msm/disp/mdp4/mdp4_lvds_pll: convert from
 round_rate() to determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series.

Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Brian Masney <bmasney@redhat.com>
Patchwork: https://patchwork.freedesktop.org/patch/667980/
Link: https://lore.kernel.org/r/20250811-drm-clk-round-rate-v2-3-4a91ccf239cf@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
index 4612886f0e49..04c49bf3d854 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lvds_pll.c
@@ -98,11 +98,14 @@ static unsigned long mdp4_lvds_pll_recalc_rate(struct clk_hw *hw,
 	return lvds_pll->pixclk;
 }
 
-static long mdp4_lvds_pll_round_rate(struct clk_hw *hw, unsigned long rate,
-		unsigned long *parent_rate)
+static int mdp4_lvds_pll_determine_rate(struct clk_hw *hw,
+					struct clk_rate_request *req)
 {
-	const struct pll_rate *pll_rate = find_rate(rate);
-	return pll_rate->rate;
+	const struct pll_rate *pll_rate = find_rate(req->rate);
+
+	req->rate = pll_rate->rate;
+
+	return 0;
 }
 
 static int mdp4_lvds_pll_set_rate(struct clk_hw *hw, unsigned long rate,
@@ -118,7 +121,7 @@ static const struct clk_ops mdp4_lvds_pll_ops = {
 	.enable = mdp4_lvds_pll_enable,
 	.disable = mdp4_lvds_pll_disable,
 	.recalc_rate = mdp4_lvds_pll_recalc_rate,
-	.round_rate = mdp4_lvds_pll_round_rate,
+	.determine_rate = mdp4_lvds_pll_determine_rate,
 	.set_rate = mdp4_lvds_pll_set_rate,
 };
 

From 0d7db6f377c4b2bbee5000d886f635d29cc2bdfd Mon Sep 17 00:00:00 2001
From: Brian Masney <bmasney@redhat.com>
Date: Mon, 11 Aug 2025 06:56:08 -0400
Subject: [PATCH 0755/1455] drm/msm/hdmi_pll_8960: convert from round_rate() to
 determine_rate()

The round_rate() clk ops is deprecated, so migrate this driver from
round_rate() to determine_rate() using the Coccinelle semantic patch
on the cover letter of this series.

Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Signed-off-by: Brian Masney <bmasney@redhat.com>
Patchwork: https://patchwork.freedesktop.org/patch/667982/
Link: https://lore.kernel.org/r/20250811-drm-clk-round-rate-v2-4-4a91ccf239cf@redhat.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c b/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c
index 83c8781fcc3f..6ba6bbdb7e05 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c
@@ -373,12 +373,14 @@ static unsigned long hdmi_pll_recalc_rate(struct clk_hw *hw,
 	return pll->pixclk;
 }
 
-static long hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate,
-				unsigned long *parent_rate)
+static int hdmi_pll_determine_rate(struct clk_hw *hw,
+				   struct clk_rate_request *req)
 {
-	const struct pll_rate *pll_rate = find_rate(rate);
+	const struct pll_rate *pll_rate = find_rate(req->rate);
 
-	return pll_rate->rate;
+	req->rate = pll_rate->rate;
+
+	return 0;
 }
 
 static int hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate,
@@ -402,7 +404,7 @@ static const struct clk_ops hdmi_pll_ops = {
 	.enable = hdmi_pll_enable,
 	.disable = hdmi_pll_disable,
 	.recalc_rate = hdmi_pll_recalc_rate,
-	.round_rate = hdmi_pll_round_rate,
+	.determine_rate = hdmi_pll_determine_rate,
 	.set_rate = hdmi_pll_set_rate,
 };
 

From 0b37ac63fc9db7779168ad18d7e6e8ab9df6a40b Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Tue, 19 Aug 2025 23:32:58 +0300
Subject: [PATCH 0756/1455] drm/msm/dpu: use drmm_writeback_connector_init()

Use drmm_writeback_connector_init() in order to initialize writeback
connector instance.

Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/669506/
Link: https://lore.kernel.org/r/20250819-wb-drop-encoder-v3-4-b48a6af7903b@oss.qualcomm.com
[DB: fixed commit message]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c
index 8ff496082902..cd73468e369a 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_writeback.c
@@ -80,7 +80,6 @@ static int dpu_wb_conn_atomic_check(struct drm_connector *connector,
 static const struct drm_connector_funcs dpu_wb_conn_funcs = {
 	.reset = drm_atomic_helper_connector_reset,
 	.fill_modes = drm_helper_probe_single_connector_modes,
-	.destroy = drm_connector_cleanup,
 	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
@@ -131,12 +130,9 @@ int dpu_writeback_init(struct drm_device *dev, struct drm_encoder *enc,
 
 	drm_connector_helper_add(&dpu_wb_conn->base.base, &dpu_wb_conn_helper_funcs);
 
-	/* DPU initializes the encoder and sets it up completely for writeback
-	 * cases and hence should use the new API drm_writeback_connector_init_with_encoder
-	 * to initialize the writeback connector
-	 */
-	rc = drm_writeback_connector_init_with_encoder(dev, &dpu_wb_conn->base, enc,
-			&dpu_wb_conn_funcs, format_list, num_formats);
+	rc = drmm_writeback_connector_init(dev, &dpu_wb_conn->base,
+					   &dpu_wb_conn_funcs, enc,
+					   format_list, num_formats);
 
 	if (!rc)
 		dpu_wb_conn->wb_enc = enc;

From f4028ef65dc4fe9010bcc017597fe0fea6479097 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 27 Aug 2025 16:29:04 +0300
Subject: [PATCH 0757/1455] drm/bridge: adv7511: provide SPD and HDMI
 infoframes

ADV75xx hardware supports sending SPD InfoFrame over the HDMI link. Also
it provides support for two generic (Spare) InfoFrames. Use those
capabilities to be able to send SPD and HDMI Vendor-Specific Infoframes.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-adv7511-infoframes-v1-1-f89b9690f89c@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/adv7511/adv7511.h     |  6 ++--
 drivers/gpu/drm/bridge/adv7511/adv7511_drv.c | 30 ++++++++++++++++++++
 2 files changed, 34 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511.h b/drivers/gpu/drm/bridge/adv7511/adv7511.h
index 85ebead9809c..8be7266fd4f4 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511.h
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511.h
@@ -195,13 +195,14 @@
 #define ADV7511_I2S_IEC958_DIRECT		3
 
 #define ADV7511_PACKET(p, x)	    ((p) * 0x20 + (x))
-#define ADV7511_PACKET_SDP(x)	    ADV7511_PACKET(0, x)
+#define ADV7511_PACKET_SPD(x)	    ADV7511_PACKET(0, x)
 #define ADV7511_PACKET_MPEG(x)	    ADV7511_PACKET(1, x)
 #define ADV7511_PACKET_ACP(x)	    ADV7511_PACKET(2, x)
 #define ADV7511_PACKET_ISRC1(x)	    ADV7511_PACKET(3, x)
 #define ADV7511_PACKET_ISRC2(x)	    ADV7511_PACKET(4, x)
 #define ADV7511_PACKET_GM(x)	    ADV7511_PACKET(5, x)
-#define ADV7511_PACKET_SPARE(x)	    ADV7511_PACKET(6, x)
+#define ADV7511_PACKET_SPARE1(x)    ADV7511_PACKET(6, x)
+#define ADV7511_PACKET_SPARE2(x)    ADV7511_PACKET(7, x)
 
 #define ADV7511_REG_CEC_TX_FRAME_HDR	0x00
 #define ADV7511_REG_CEC_TX_FRAME_DATA0	0x01
@@ -348,6 +349,7 @@ struct adv7511 {
 	struct i2c_client *i2c_cec;
 
 	struct regmap *regmap;
+	struct regmap *regmap_packet;
 	struct regmap *regmap_cec;
 	enum drm_connector_status status;
 	bool powered;
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index 00d6417c177b..26f8ef482423 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -132,6 +132,13 @@ static const struct regmap_config adv7511_regmap_config = {
 	.volatile_reg = adv7511_register_volatile,
 };
 
+static const struct regmap_config adv7511_packet_config = {
+	.reg_bits = 8,
+	.val_bits = 8,
+
+	.max_register = 0xff,
+};
+
 /* -----------------------------------------------------------------------------
  * Hardware configuration
  */
@@ -889,6 +896,12 @@ static int adv7511_bridge_hdmi_clear_infoframe(struct drm_bridge *bridge,
 	case HDMI_INFOFRAME_TYPE_AVI:
 		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_AVI_INFOFRAME);
 		break;
+	case HDMI_INFOFRAME_TYPE_SPD:
+		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_SPD);
+		break;
+	case HDMI_INFOFRAME_TYPE_VENDOR:
+		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_SPARE1);
+		break;
 	default:
 		drm_dbg_driver(adv7511->bridge.dev, "Unsupported HDMI InfoFrame %x\n", type);
 		break;
@@ -913,6 +926,16 @@ static int adv7511_bridge_hdmi_write_infoframe(struct drm_bridge *bridge,
 
 		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_AVI_INFOFRAME);
 		break;
+	case HDMI_INFOFRAME_TYPE_SPD:
+		regmap_bulk_write(adv7511->regmap_packet, ADV7511_PACKET_SPD(0),
+				  buffer, len);
+		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_SPD);
+		break;
+	case HDMI_INFOFRAME_TYPE_VENDOR:
+		regmap_bulk_write(adv7511->regmap_packet, ADV7511_PACKET_SPARE1(0),
+				  buffer, len);
+		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_SPARE1);
+		break;
 	default:
 		drm_dbg_driver(adv7511->bridge.dev, "Unsupported HDMI InfoFrame %x\n", type);
 		break;
@@ -1242,6 +1265,13 @@ static int adv7511_probe(struct i2c_client *i2c)
 		goto err_i2c_unregister_edid;
 	}
 
+	adv7511->regmap_packet = devm_regmap_init_i2c(adv7511->i2c_packet,
+						      &adv7511_packet_config);
+	if (IS_ERR(adv7511->regmap_packet)) {
+		ret = PTR_ERR(adv7511->regmap_packet);
+		goto err_i2c_unregister_packet;
+	}
+
 	regmap_write(adv7511->regmap, ADV7511_REG_PACKET_I2C_ADDR,
 		     adv7511->i2c_packet->addr << 1);
 

From e5e3d7bff7645d2577c56ccc8b110b9b6978aa08 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 26 Aug 2025 11:18:58 -0300
Subject: [PATCH 0758/1455] drm/v3d: Store a pointer to `struct v3d_file_priv`
 inside each job
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of storing a pointer to the DRM file data, store a pointer
directly to the private V3D file struct. No functional change, this
commit only avoids multiple levels of pointer indirection and makes
the code more straightforward.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-1-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_drv.h    |  4 ++--
 drivers/gpu/drm/v3d/v3d_sched.c  | 10 +++++-----
 drivers/gpu/drm/v3d/v3d_submit.c |  2 +-
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index 82d84a96235f..951a302336ce 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -327,9 +327,9 @@ struct v3d_job {
 	struct v3d_perfmon *perfmon;
 
 	/* File descriptor of the process that submitted the job that could be used
-	 * for collecting stats by process of GPU usage.
+	 * to collect per-process information about the GPU.
 	 */
-	struct drm_file *file;
+	struct v3d_file_priv *file_priv;
 
 	/* Callback for the freeing of the job on refcount going to 0. */
 	void (*free)(struct kref *ref);
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index f9d9a198d718..1846030c5f3a 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -139,7 +139,7 @@ static void
 v3d_job_start_stats(struct v3d_job *job, enum v3d_queue queue)
 {
 	struct v3d_dev *v3d = job->v3d;
-	struct v3d_file_priv *file = job->file->driver_priv;
+	struct v3d_file_priv *file = job->file_priv;
 	struct v3d_stats *global_stats = &v3d->queue[queue].stats;
 	struct v3d_stats *local_stats = &file->stats[queue];
 	u64 now = local_clock();
@@ -197,7 +197,7 @@ void
 v3d_job_update_stats(struct v3d_job *job, enum v3d_queue queue)
 {
 	struct v3d_dev *v3d = job->v3d;
-	struct v3d_file_priv *file = job->file->driver_priv;
+	struct v3d_file_priv *file = job->file_priv;
 	struct v3d_stats *global_stats = &v3d->queue[queue].stats;
 	u64 now = local_clock();
 	unsigned long flags;
@@ -574,7 +574,7 @@ static void
 v3d_reset_performance_queries(struct v3d_cpu_job *job)
 {
 	struct v3d_performance_query_info *performance_query = &job->performance_query;
-	struct v3d_file_priv *v3d_priv = job->base.file->driver_priv;
+	struct v3d_file_priv *v3d_priv = job->base.file_priv;
 	struct v3d_dev *v3d = job->base.v3d;
 	struct v3d_perfmon *perfmon;
 
@@ -604,7 +604,7 @@ v3d_write_performance_query_result(struct v3d_cpu_job *job, void *data,
 {
 	struct v3d_performance_query_info *performance_query =
 						&job->performance_query;
-	struct v3d_file_priv *v3d_priv = job->base.file->driver_priv;
+	struct v3d_file_priv *v3d_priv = job->base.file_priv;
 	struct v3d_performance_query *perf_query =
 			&performance_query->queries[query];
 	struct v3d_dev *v3d = job->base.v3d;
@@ -722,7 +722,7 @@ static enum drm_gpu_sched_stat
 v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 {
 	struct v3d_job *job = to_v3d_job(sched_job);
-	struct v3d_file_priv *v3d_priv = job->file->driver_priv;
+	struct v3d_file_priv *v3d_priv = job->file_priv;
 	enum v3d_queue q;
 
 	mutex_lock(&v3d->reset_lock);
diff --git a/drivers/gpu/drm/v3d/v3d_submit.c b/drivers/gpu/drm/v3d/v3d_submit.c
index 5171ffe9012d..f3652e90683c 100644
--- a/drivers/gpu/drm/v3d/v3d_submit.c
+++ b/drivers/gpu/drm/v3d/v3d_submit.c
@@ -166,7 +166,7 @@ v3d_job_init(struct v3d_dev *v3d, struct drm_file *file_priv,
 
 	job->v3d = v3d;
 	job->free = free;
-	job->file = file_priv;
+	job->file_priv = v3d_priv;
 
 	ret = drm_sched_job_init(&job->base, &v3d_priv->sched_entity[queue],
 				 1, v3d_priv, file_priv->client_id);

From 0d3768826d38c0ac740f8b45cd13346630535f2b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 26 Aug 2025 11:18:59 -0300
Subject: [PATCH 0759/1455] drm/v3d: Store the active job inside the queue's
 state
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Instead of storing the queue's active job in four different variables,
store the active job inside the queue's state. This way, it's possible
to access all active jobs using an index based in `enum v3d_queue`.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-2-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_drv.h   |  8 ++---
 drivers/gpu/drm/v3d/v3d_gem.c   |  7 ++--
 drivers/gpu/drm/v3d/v3d_irq.c   | 62 +++++++++++++--------------------
 drivers/gpu/drm/v3d/v3d_sched.c | 16 ++++-----
 4 files changed, 38 insertions(+), 55 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index 951a302336ce..d557caca5c4b 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -58,6 +58,9 @@ struct v3d_queue_state {
 
 	/* Stores the GPU stats for this queue in the global context. */
 	struct v3d_stats stats;
+
+	/* Currently active job for this queue */
+	struct v3d_job *active_job;
 };
 
 /* Performance monitor object. The perform lifetime is controlled by userspace
@@ -159,11 +162,6 @@ struct v3d_dev {
 
 	struct work_struct overflow_mem_work;
 
-	struct v3d_bin_job *bin_job;
-	struct v3d_render_job *render_job;
-	struct v3d_tfu_job *tfu_job;
-	struct v3d_csd_job *csd_job;
-
 	struct v3d_queue_state queue[V3D_MAX_QUEUES];
 
 	/* Spinlock used to synchronize the overflow memory
diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
index 37bf5eecdd2c..cfa09b73c1ed 100644
--- a/drivers/gpu/drm/v3d/v3d_gem.c
+++ b/drivers/gpu/drm/v3d/v3d_gem.c
@@ -324,6 +324,7 @@ void
 v3d_gem_destroy(struct drm_device *dev)
 {
 	struct v3d_dev *v3d = to_v3d_dev(dev);
+	enum v3d_queue q;
 
 	v3d_sched_fini(v3d);
 	v3d_gemfs_fini(v3d);
@@ -331,10 +332,8 @@ v3d_gem_destroy(struct drm_device *dev)
 	/* Waiting for jobs to finish would need to be done before
 	 * unregistering V3D.
 	 */
-	WARN_ON(v3d->bin_job);
-	WARN_ON(v3d->render_job);
-	WARN_ON(v3d->tfu_job);
-	WARN_ON(v3d->csd_job);
+	for (q = 0; q < V3D_MAX_QUEUES; q++)
+		WARN_ON(v3d->queue[q].active_job);
 
 	drm_mm_takedown(&v3d->mm);
 
diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c
index a515a301e480..6605ec200828 100644
--- a/drivers/gpu/drm/v3d/v3d_irq.c
+++ b/drivers/gpu/drm/v3d/v3d_irq.c
@@ -42,6 +42,8 @@ v3d_overflow_mem_work(struct work_struct *work)
 		container_of(work, struct v3d_dev, overflow_mem_work);
 	struct drm_device *dev = &v3d->drm;
 	struct v3d_bo *bo = v3d_bo_create(dev, NULL /* XXX: GMP */, 256 * 1024);
+	struct v3d_queue_state *queue = &v3d->queue[V3D_BIN];
+	struct v3d_bin_job *bin_job;
 	struct drm_gem_object *obj;
 	unsigned long irqflags;
 
@@ -61,13 +63,15 @@ v3d_overflow_mem_work(struct work_struct *work)
 	 * some binner pool anyway.
 	 */
 	spin_lock_irqsave(&v3d->job_lock, irqflags);
-	if (!v3d->bin_job) {
+	bin_job = (struct v3d_bin_job *)queue->active_job;
+
+	if (!bin_job) {
 		spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 		goto out;
 	}
 
 	drm_gem_object_get(obj);
-	list_add_tail(&bo->unref_head, &v3d->bin_job->render->unref_list);
+	list_add_tail(&bo->unref_head, &bin_job->render->unref_list);
 	spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 
 	v3d_mmu_flush_all(v3d);
@@ -79,6 +83,20 @@ out:
 	drm_gem_object_put(obj);
 }
 
+static void
+v3d_irq_signal_fence(struct v3d_dev *v3d, enum v3d_queue q,
+		     void (*trace_irq)(struct drm_device *, uint64_t))
+{
+	struct v3d_queue_state *queue = &v3d->queue[q];
+	struct v3d_fence *fence = to_v3d_fence(queue->active_job->irq_fence);
+
+	v3d_job_update_stats(queue->active_job, q);
+	trace_irq(&v3d->drm, fence->seqno);
+
+	queue->active_job = NULL;
+	dma_fence_signal(&fence->base);
+}
+
 static irqreturn_t
 v3d_irq(int irq, void *arg)
 {
@@ -102,41 +120,17 @@ v3d_irq(int irq, void *arg)
 	}
 
 	if (intsts & V3D_INT_FLDONE) {
-		struct v3d_fence *fence =
-			to_v3d_fence(v3d->bin_job->base.irq_fence);
-
-		v3d_job_update_stats(&v3d->bin_job->base, V3D_BIN);
-		trace_v3d_bcl_irq(&v3d->drm, fence->seqno);
-
-		v3d->bin_job = NULL;
-		dma_fence_signal(&fence->base);
-
+		v3d_irq_signal_fence(v3d, V3D_BIN, trace_v3d_bcl_irq);
 		status = IRQ_HANDLED;
 	}
 
 	if (intsts & V3D_INT_FRDONE) {
-		struct v3d_fence *fence =
-			to_v3d_fence(v3d->render_job->base.irq_fence);
-
-		v3d_job_update_stats(&v3d->render_job->base, V3D_RENDER);
-		trace_v3d_rcl_irq(&v3d->drm, fence->seqno);
-
-		v3d->render_job = NULL;
-		dma_fence_signal(&fence->base);
-
+		v3d_irq_signal_fence(v3d, V3D_RENDER, trace_v3d_rcl_irq);
 		status = IRQ_HANDLED;
 	}
 
 	if (intsts & V3D_INT_CSDDONE(v3d->ver)) {
-		struct v3d_fence *fence =
-			to_v3d_fence(v3d->csd_job->base.irq_fence);
-
-		v3d_job_update_stats(&v3d->csd_job->base, V3D_CSD);
-		trace_v3d_csd_irq(&v3d->drm, fence->seqno);
-
-		v3d->csd_job = NULL;
-		dma_fence_signal(&fence->base);
-
+		v3d_irq_signal_fence(v3d, V3D_CSD, trace_v3d_csd_irq);
 		status = IRQ_HANDLED;
 	}
 
@@ -168,15 +162,7 @@ v3d_hub_irq(int irq, void *arg)
 	V3D_WRITE(V3D_HUB_INT_CLR, intsts);
 
 	if (intsts & V3D_HUB_INT_TFUC) {
-		struct v3d_fence *fence =
-			to_v3d_fence(v3d->tfu_job->base.irq_fence);
-
-		v3d_job_update_stats(&v3d->tfu_job->base, V3D_TFU);
-		trace_v3d_tfu_irq(&v3d->drm, fence->seqno);
-
-		v3d->tfu_job = NULL;
-		dma_fence_signal(&fence->base);
-
+		v3d_irq_signal_fence(v3d, V3D_TFU, trace_v3d_tfu_irq);
 		status = IRQ_HANDLED;
 	}
 
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 1846030c5f3a..65375e2d4922 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -232,7 +232,7 @@ static struct dma_fence *v3d_bin_job_run(struct drm_sched_job *sched_job)
 
 	if (unlikely(job->base.base.s_fence->finished.error)) {
 		spin_lock_irqsave(&v3d->job_lock, irqflags);
-		v3d->bin_job = NULL;
+		v3d->queue[V3D_BIN].active_job = NULL;
 		spin_unlock_irqrestore(&v3d->job_lock, irqflags);
 		return NULL;
 	}
@@ -241,7 +241,7 @@ static struct dma_fence *v3d_bin_job_run(struct drm_sched_job *sched_job)
 	 * v3d_overflow_mem_work().
 	 */
 	spin_lock_irqsave(&v3d->job_lock, irqflags);
-	v3d->bin_job = job;
+	v3d->queue[V3D_BIN].active_job = &job->base;
 	/* Clear out the overflow allocation, so we don't
 	 * reuse the overflow attached to a previous job.
 	 */
@@ -290,11 +290,11 @@ static struct dma_fence *v3d_render_job_run(struct drm_sched_job *sched_job)
 	struct dma_fence *fence;
 
 	if (unlikely(job->base.base.s_fence->finished.error)) {
-		v3d->render_job = NULL;
+		v3d->queue[V3D_RENDER].active_job = NULL;
 		return NULL;
 	}
 
-	v3d->render_job = job;
+	v3d->queue[V3D_RENDER].active_job = &job->base;
 
 	/* Can we avoid this flush?  We need to be careful of
 	 * scheduling, though -- imagine job0 rendering to texture and
@@ -338,11 +338,11 @@ v3d_tfu_job_run(struct drm_sched_job *sched_job)
 	struct dma_fence *fence;
 
 	if (unlikely(job->base.base.s_fence->finished.error)) {
-		v3d->tfu_job = NULL;
+		v3d->queue[V3D_TFU].active_job = NULL;
 		return NULL;
 	}
 
-	v3d->tfu_job = job;
+	v3d->queue[V3D_TFU].active_job = &job->base;
 
 	fence = v3d_fence_create(v3d, V3D_TFU);
 	if (IS_ERR(fence))
@@ -386,11 +386,11 @@ v3d_csd_job_run(struct drm_sched_job *sched_job)
 	int i, csd_cfg0_reg;
 
 	if (unlikely(job->base.base.s_fence->finished.error)) {
-		v3d->csd_job = NULL;
+		v3d->queue[V3D_CSD].active_job = NULL;
 		return NULL;
 	}
 
-	v3d->csd_job = job;
+	v3d->queue[V3D_CSD].active_job = &job->base;
 
 	v3d_invalidate_caches(v3d);
 

From e9d8e027483267a9505f5b6b7f0476b5299743af Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 26 Aug 2025 11:19:00 -0300
Subject: [PATCH 0760/1455] drm/v3d: Replace a global spinlock with a per-queue
 spinlock
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Each V3D queue works independently and all the dependencies between the
jobs are handled through the DRM scheduler. Therefore, there is no need
to use one single lock for all queues. Using it, creates unnecessary
contention between different queues that can operate independently.

Replace the global spinlock with per-queue locks to improve parallelism
and reduce contention between different V3D queues (BIN, RENDER, TFU,
CSD). This allows independent queues to operate concurrently while
maintaining proper synchronization within each queue.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-3-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_drv.h   |  8 ++------
 drivers/gpu/drm/v3d/v3d_fence.c | 11 ++++++-----
 drivers/gpu/drm/v3d/v3d_gem.c   |  3 ++-
 drivers/gpu/drm/v3d/v3d_irq.c   |  6 +++---
 drivers/gpu/drm/v3d/v3d_sched.c | 13 +++++++------
 5 files changed, 20 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index d557caca5c4b..cfc2f9c123aa 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -61,6 +61,7 @@ struct v3d_queue_state {
 
 	/* Currently active job for this queue */
 	struct v3d_job *active_job;
+	spinlock_t queue_lock;
 };
 
 /* Performance monitor object. The perform lifetime is controlled by userspace
@@ -164,11 +165,6 @@ struct v3d_dev {
 
 	struct v3d_queue_state queue[V3D_MAX_QUEUES];
 
-	/* Spinlock used to synchronize the overflow memory
-	 * management against bin job submission.
-	 */
-	spinlock_t job_lock;
-
 	/* Used to track the active perfmon if any. */
 	struct v3d_perfmon *active_perfmon;
 
@@ -568,7 +564,7 @@ void v3d_get_stats(const struct v3d_stats *stats, u64 timestamp,
 
 /* v3d_fence.c */
 extern const struct dma_fence_ops v3d_fence_ops;
-struct dma_fence *v3d_fence_create(struct v3d_dev *v3d, enum v3d_queue queue);
+struct dma_fence *v3d_fence_create(struct v3d_dev *v3d, enum v3d_queue q);
 
 /* v3d_gem.c */
 int v3d_gem_init(struct drm_device *dev);
diff --git a/drivers/gpu/drm/v3d/v3d_fence.c b/drivers/gpu/drm/v3d/v3d_fence.c
index 89840ed212c0..8f8471adae34 100644
--- a/drivers/gpu/drm/v3d/v3d_fence.c
+++ b/drivers/gpu/drm/v3d/v3d_fence.c
@@ -3,8 +3,9 @@
 
 #include "v3d_drv.h"
 
-struct dma_fence *v3d_fence_create(struct v3d_dev *v3d, enum v3d_queue queue)
+struct dma_fence *v3d_fence_create(struct v3d_dev *v3d, enum v3d_queue q)
 {
+	struct v3d_queue_state *queue = &v3d->queue[q];
 	struct v3d_fence *fence;
 
 	fence = kzalloc(sizeof(*fence), GFP_KERNEL);
@@ -12,10 +13,10 @@ struct dma_fence *v3d_fence_create(struct v3d_dev *v3d, enum v3d_queue queue)
 		return ERR_PTR(-ENOMEM);
 
 	fence->dev = &v3d->drm;
-	fence->queue = queue;
-	fence->seqno = ++v3d->queue[queue].emit_seqno;
-	dma_fence_init(&fence->base, &v3d_fence_ops, &v3d->job_lock,
-		       v3d->queue[queue].fence_context, fence->seqno);
+	fence->queue = q;
+	fence->seqno = ++queue->emit_seqno;
+	dma_fence_init(&fence->base, &v3d_fence_ops, &queue->queue_lock,
+		       queue->fence_context, fence->seqno);
 
 	return &fence->base;
 }
diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
index cfa09b73c1ed..c77d90aa9b82 100644
--- a/drivers/gpu/drm/v3d/v3d_gem.c
+++ b/drivers/gpu/drm/v3d/v3d_gem.c
@@ -271,10 +271,11 @@ v3d_gem_init(struct drm_device *dev)
 		queue->fence_context = dma_fence_context_alloc(1);
 		memset(&queue->stats, 0, sizeof(queue->stats));
 		seqcount_init(&queue->stats.lock);
+
+		spin_lock_init(&queue->queue_lock);
 	}
 
 	spin_lock_init(&v3d->mm_lock);
-	spin_lock_init(&v3d->job_lock);
 	ret = drmm_mutex_init(dev, &v3d->bo_lock);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c
index 6605ec200828..31ecc5b4ba5a 100644
--- a/drivers/gpu/drm/v3d/v3d_irq.c
+++ b/drivers/gpu/drm/v3d/v3d_irq.c
@@ -62,17 +62,17 @@ v3d_overflow_mem_work(struct work_struct *work)
 	 * bin job got scheduled, that's fine.  We'll just give them
 	 * some binner pool anyway.
 	 */
-	spin_lock_irqsave(&v3d->job_lock, irqflags);
+	spin_lock_irqsave(&queue->queue_lock, irqflags);
 	bin_job = (struct v3d_bin_job *)queue->active_job;
 
 	if (!bin_job) {
-		spin_unlock_irqrestore(&v3d->job_lock, irqflags);
+		spin_unlock_irqrestore(&queue->queue_lock, irqflags);
 		goto out;
 	}
 
 	drm_gem_object_get(obj);
 	list_add_tail(&bo->unref_head, &bin_job->render->unref_list);
-	spin_unlock_irqrestore(&v3d->job_lock, irqflags);
+	spin_unlock_irqrestore(&queue->queue_lock, irqflags);
 
 	v3d_mmu_flush_all(v3d);
 
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 65375e2d4922..1ae447e58ac2 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -226,27 +226,28 @@ static struct dma_fence *v3d_bin_job_run(struct drm_sched_job *sched_job)
 {
 	struct v3d_bin_job *job = to_bin_job(sched_job);
 	struct v3d_dev *v3d = job->base.v3d;
+	struct v3d_queue_state *queue = &v3d->queue[V3D_BIN];
 	struct drm_device *dev = &v3d->drm;
 	struct dma_fence *fence;
 	unsigned long irqflags;
 
 	if (unlikely(job->base.base.s_fence->finished.error)) {
-		spin_lock_irqsave(&v3d->job_lock, irqflags);
-		v3d->queue[V3D_BIN].active_job = NULL;
-		spin_unlock_irqrestore(&v3d->job_lock, irqflags);
+		spin_lock_irqsave(&queue->queue_lock, irqflags);
+		queue->active_job = NULL;
+		spin_unlock_irqrestore(&queue->queue_lock, irqflags);
 		return NULL;
 	}
 
 	/* Lock required around bin_job update vs
 	 * v3d_overflow_mem_work().
 	 */
-	spin_lock_irqsave(&v3d->job_lock, irqflags);
-	v3d->queue[V3D_BIN].active_job = &job->base;
+	spin_lock_irqsave(&queue->queue_lock, irqflags);
+	queue->active_job = &job->base;
 	/* Clear out the overflow allocation, so we don't
 	 * reuse the overflow attached to a previous job.
 	 */
 	V3D_CORE_WRITE(0, V3D_PTB_BPOS, 0);
-	spin_unlock_irqrestore(&v3d->job_lock, irqflags);
+	spin_unlock_irqrestore(&queue->queue_lock, irqflags);
 
 	v3d_invalidate_caches(v3d);
 

From fa6a20c8747041df406d184321a2a49e82d46298 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 26 Aug 2025 11:19:01 -0300
Subject: [PATCH 0761/1455] drm/v3d: Address race-condition between per-fd GPU
 stats and fd release
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When the file descriptor is closed while a job is still running,
there's a race condition between the job completion callback and the
file descriptor cleanup. This can lead to accessing freed memory when
updating per-fd GPU stats, such as the following example:

[56120.512903] Unable to handle kernel paging request at virtual address 0000330a92b9688a
[56120.520881] Mem abort info:
[56120.523687] ESR = 0x0000000096000005
[56120.527454] EC = 0x25: DABT (current EL), IL = 32 bits
[56120.532785] SET = 0, FnV = 0
[56120.535847] EA = 0, S1PTW = 0
[56120.538995] FSC = 0x05: level 1 translation fault
[56120.543891] Data abort info:
[56120.546778] ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000
[56120.552289] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[56120.557362] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[56120.562690] user pgtable: 16k pages, 47-bit VAs, pgdp=0000000023f54000
[56120.569239] [0000330a92b9688a] pgd=0000000000000000, p4d=0000000000000000, pud=0000000000000000
[56120.577975] Internal error: Oops: 0000000096000005 [#1] PREEMPT SMP
 	       CPU: 0 UID: 1000 PID: 1497409 Comm: mpv Not tainted 6.12.37-ncvm5+ #1
 	       Hardware name: Raspberry Pi 5 Model B Rev 1.0 (DT)
 	       pstate: 604000c9 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
 	       pc : v3d_job_update_stats+0x64/0x168 [v3d]
 	       lr : v3d_job_update_stats+0x40/0x168 [v3d]
 	       sp : ffffc00080003e60
 	       x29: ffffc00080003e60 x28: ffff800002860000 x27: 0000000000000000
 	       x26: 0000000000000000 x25: ffff800002860000 x24: ffff800002630800
 	       x23: ffff800060786000 x22: 0000330a933c31fb x21: 0000000000000001
 	       x20: 0000330a92b96302 x19: ffff800060786b10 x18: 0000000000000000
 	       x17: ffffaf90506a0000 x16: ffffd06fce57c360 x15: 0000000000000000
 	       x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000
 	       x11: 0000000000000000 x10: 0000000000000000 x9 : ffffd06f5d0fec40
 	       x8 : 0000000000000000 x7 : 0000000000000000 x6 : 000002978dbd535a
 	       x5 : 00ffffffffffffff x4 : 0000000000000015 x3 : 0000300001fddf88
 	       x2 : 0000000000000020 x1 : 0000000000010001 x0 : 0000330a92b96872
 	       Call trace:
		 v3d_job_update_stats+0x64/0x168 [v3d]
		 v3d_irq+0x118/0x2e0 [v3d]
		 __handle_irq_event_percpu+0x60/0x220

Fix such an issue by protecting all accesses to `job->file_priv` with
the queue's lock. With that, we can clear `job->file_priv` before the
V3D per-fd structure is freed and assure that `job->file_priv` exists
during the per-fd GPU stats updates.

Fixes: e1bc3a13bd77 ("drm/v3d: Avoid NULL pointer dereference in `v3d_job_update_stats()`")
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-4-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_drv.c   | 14 +++++++++++++-
 drivers/gpu/drm/v3d/v3d_drv.h   |  2 +-
 drivers/gpu/drm/v3d/v3d_sched.c | 14 +++++++-------
 3 files changed, 21 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_drv.c b/drivers/gpu/drm/v3d/v3d_drv.c
index 2def155ce496..c5a3bbbc74c5 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.c
+++ b/drivers/gpu/drm/v3d/v3d_drv.c
@@ -157,12 +157,24 @@ v3d_open(struct drm_device *dev, struct drm_file *file)
 static void
 v3d_postclose(struct drm_device *dev, struct drm_file *file)
 {
+	struct v3d_dev *v3d = to_v3d_dev(dev);
 	struct v3d_file_priv *v3d_priv = file->driver_priv;
+	unsigned long irqflags;
 	enum v3d_queue q;
 
-	for (q = 0; q < V3D_MAX_QUEUES; q++)
+	for (q = 0; q < V3D_MAX_QUEUES; q++) {
+		struct v3d_queue_state *queue = &v3d->queue[q];
+		struct v3d_job *job = queue->active_job;
+
 		drm_sched_entity_destroy(&v3d_priv->sched_entity[q]);
 
+		if (job && job->base.entity == &v3d_priv->sched_entity[q]) {
+			spin_lock_irqsave(&queue->queue_lock, irqflags);
+			job->file_priv = NULL;
+			spin_unlock_irqrestore(&queue->queue_lock, irqflags);
+		}
+	}
+
 	v3d_perfmon_close_file(v3d_priv);
 	kfree(v3d_priv);
 }
diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
index cfc2f9c123aa..0317f3d7452a 100644
--- a/drivers/gpu/drm/v3d/v3d_drv.h
+++ b/drivers/gpu/drm/v3d/v3d_drv.h
@@ -608,7 +608,7 @@ void v3d_timestamp_query_info_free(struct v3d_timestamp_query_info *query_info,
 				   unsigned int count);
 void v3d_performance_query_info_free(struct v3d_performance_query_info *query_info,
 				     unsigned int count);
-void v3d_job_update_stats(struct v3d_job *job, enum v3d_queue queue);
+void v3d_job_update_stats(struct v3d_job *job, enum v3d_queue q);
 int v3d_sched_init(struct v3d_dev *v3d);
 void v3d_sched_fini(struct v3d_dev *v3d);
 
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 1ae447e58ac2..ad72be7c5d50 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -194,11 +194,11 @@ v3d_stats_update(struct v3d_stats *stats, u64 now)
 }
 
 void
-v3d_job_update_stats(struct v3d_job *job, enum v3d_queue queue)
+v3d_job_update_stats(struct v3d_job *job, enum v3d_queue q)
 {
 	struct v3d_dev *v3d = job->v3d;
-	struct v3d_file_priv *file = job->file_priv;
-	struct v3d_stats *global_stats = &v3d->queue[queue].stats;
+	struct v3d_queue_state *queue = &v3d->queue[q];
+	struct v3d_stats *global_stats = &queue->stats;
 	u64 now = local_clock();
 	unsigned long flags;
 
@@ -209,10 +209,10 @@ v3d_job_update_stats(struct v3d_job *job, enum v3d_queue queue)
 		preempt_disable();
 
 	/* Don't update the local stats if the file context has already closed */
-	if (file)
-		v3d_stats_update(&file->stats[queue], now);
-	else
-		drm_dbg(&v3d->drm, "The file descriptor was closed before job completion\n");
+	spin_lock(&queue->queue_lock);
+	if (job->file_priv)
+		v3d_stats_update(&job->file_priv->stats[q], now);
+	spin_unlock(&queue->queue_lock);
 
 	v3d_stats_update(global_stats, now);
 

From 78fe02d090d38393cc00de9d1929e59426f202a4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 26 Aug 2025 11:19:02 -0300
Subject: [PATCH 0762/1455] drm/v3d: Synchronous operations can't timeout
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

CPU jobs and CACHE CLEAN jobs execute synchronously once the DRM
scheduler starts running them. Therefore, there is no fence to wait on,
neither are those jobs able to timeout.

Hence, remove the `timedout_job` hook from the CPU and CACHE CLEAN
scheduler ops.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Reviewed-by: Melissa Wen <mwen@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-5-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_sched.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index ad72be7c5d50..b8984e608547 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -701,6 +701,7 @@ v3d_cpu_job_run(struct drm_sched_job *sched_job)
 	trace_v3d_cpu_job_end(&v3d->drm, job->job_type);
 	v3d_job_update_stats(&job->base, V3D_CPU);
 
+	/* Synchronous operation, so no fence to wait on. */
 	return NULL;
 }
 
@@ -716,6 +717,7 @@ v3d_cache_clean_job_run(struct drm_sched_job *sched_job)
 
 	v3d_job_update_stats(job, V3D_CACHE_CLEAN);
 
+	/* Synchronous operation, so no fence to wait on. */
 	return NULL;
 }
 
@@ -797,7 +799,7 @@ v3d_render_job_timedout(struct drm_sched_job *sched_job)
 }
 
 static enum drm_gpu_sched_stat
-v3d_generic_job_timedout(struct drm_sched_job *sched_job)
+v3d_tfu_job_timedout(struct drm_sched_job *sched_job)
 {
 	struct v3d_job *job = to_v3d_job(sched_job);
 
@@ -837,7 +839,7 @@ static const struct drm_sched_backend_ops v3d_render_sched_ops = {
 
 static const struct drm_sched_backend_ops v3d_tfu_sched_ops = {
 	.run_job = v3d_tfu_job_run,
-	.timedout_job = v3d_generic_job_timedout,
+	.timedout_job = v3d_tfu_job_timedout,
 	.free_job = v3d_sched_job_free,
 };
 
@@ -849,13 +851,11 @@ static const struct drm_sched_backend_ops v3d_csd_sched_ops = {
 
 static const struct drm_sched_backend_ops v3d_cache_clean_sched_ops = {
 	.run_job = v3d_cache_clean_job_run,
-	.timedout_job = v3d_generic_job_timedout,
 	.free_job = v3d_sched_job_free
 };
 
 static const struct drm_sched_backend_ops v3d_cpu_sched_ops = {
 	.run_job = v3d_cpu_job_run,
-	.timedout_job = v3d_generic_job_timedout,
 	.free_job = v3d_cpu_job_free
 };
 

From 7d9bc9bee2778e2da7147aeb3a81427487598493 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ma=C3=ADra=20Canal?= <mcanal@igalia.com>
Date: Tue, 26 Aug 2025 11:19:03 -0300
Subject: [PATCH 0763/1455] drm/v3d: Protect per-fd reset counter against fd
 release
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The per-fd reset counter tracks GPU resets caused by jobs submitted
through a specific file descriptor. However, there's a race condition
where the file descriptor can be closed while jobs are still running,
leading to potential access to freed memory when updating the reset
counter.

Ensure that the per-fd reset counter is only updated when the file
descriptor is still valid, preventing use-after-free scenarios during
GPU reset handling.

Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Link: https://lore.kernel.org/r/20250826-v3d-queue-lock-v3-6-979efc43e490@igalia.com
Signed-off-by: Maíra Canal <mcanal@igalia.com>
---
 drivers/gpu/drm/v3d/v3d_sched.c | 30 +++++++++++++++++-------------
 1 file changed, 17 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index b8984e608547..0ec06bfbbebb 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -722,17 +722,19 @@ v3d_cache_clean_job_run(struct drm_sched_job *sched_job)
 }
 
 static enum drm_gpu_sched_stat
-v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
+v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job,
+			  enum v3d_queue q)
 {
 	struct v3d_job *job = to_v3d_job(sched_job);
 	struct v3d_file_priv *v3d_priv = job->file_priv;
-	enum v3d_queue q;
+	unsigned long irqflags;
+	enum v3d_queue i;
 
 	mutex_lock(&v3d->reset_lock);
 
 	/* block scheduler */
-	for (q = 0; q < V3D_MAX_QUEUES; q++)
-		drm_sched_stop(&v3d->queue[q].sched, sched_job);
+	for (i = 0; i < V3D_MAX_QUEUES; i++)
+		drm_sched_stop(&v3d->queue[i].sched, sched_job);
 
 	if (sched_job)
 		drm_sched_increase_karma(sched_job);
@@ -741,15 +743,17 @@ v3d_gpu_reset_for_timeout(struct v3d_dev *v3d, struct drm_sched_job *sched_job)
 	v3d_reset(v3d);
 
 	v3d->reset_counter++;
-	v3d_priv->reset_counter++;
+	spin_lock_irqsave(&v3d->queue[q].queue_lock, irqflags);
+	if (v3d_priv)
+		v3d_priv->reset_counter++;
+	spin_unlock_irqrestore(&v3d->queue[q].queue_lock, irqflags);
 
-	for (q = 0; q < V3D_MAX_QUEUES; q++)
-		drm_sched_resubmit_jobs(&v3d->queue[q].sched);
+	for (i = 0; i < V3D_MAX_QUEUES; i++)
+		drm_sched_resubmit_jobs(&v3d->queue[i].sched);
 
 	/* Unblock schedulers and restart their jobs. */
-	for (q = 0; q < V3D_MAX_QUEUES; q++) {
-		drm_sched_start(&v3d->queue[q].sched, 0);
-	}
+	for (i = 0; i < V3D_MAX_QUEUES; i++)
+		drm_sched_start(&v3d->queue[i].sched, 0);
 
 	mutex_unlock(&v3d->reset_lock);
 
@@ -777,7 +781,7 @@ v3d_cl_job_timedout(struct drm_sched_job *sched_job, enum v3d_queue q,
 		return DRM_GPU_SCHED_STAT_NO_HANG;
 	}
 
-	return v3d_gpu_reset_for_timeout(v3d, sched_job);
+	return v3d_gpu_reset_for_timeout(v3d, sched_job, q);
 }
 
 static enum drm_gpu_sched_stat
@@ -803,7 +807,7 @@ v3d_tfu_job_timedout(struct drm_sched_job *sched_job)
 {
 	struct v3d_job *job = to_v3d_job(sched_job);
 
-	return v3d_gpu_reset_for_timeout(job->v3d, sched_job);
+	return v3d_gpu_reset_for_timeout(job->v3d, sched_job, V3D_TFU);
 }
 
 static enum drm_gpu_sched_stat
@@ -822,7 +826,7 @@ v3d_csd_job_timedout(struct drm_sched_job *sched_job)
 		return DRM_GPU_SCHED_STAT_NO_HANG;
 	}
 
-	return v3d_gpu_reset_for_timeout(v3d, sched_job);
+	return v3d_gpu_reset_for_timeout(v3d, sched_job, V3D_CSD);
 }
 
 static const struct drm_sched_backend_ops v3d_bin_sched_ops = {

From 1047bd82794a1eab64d643f196d09171ce983f44 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 28 Aug 2025 15:48:37 +0200
Subject: [PATCH 0764/1455] drm/xe: Fix incorrect migration of backed-up object
 to VRAM
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If an object is backed up to shmem it is incorrectly identified
as not having valid data by the move code. This means moving
to VRAM skips the -EMULTIHOP step and the bo is cleared. This
causes all sorts of weird behaviour on DGFX if an already evicted
object is targeted by the shrinker.

Fix this by using ttm_tt_is_swapped() to identify backed-up
objects.

Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5996
Fixes: 00c8efc3180f ("drm/xe: Add a shrinker for xe bos")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: <stable@vger.kernel.org> # v6.15+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250828134837.5709-1-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_bo.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 7d1ff642b02a..4faf15d5fa6d 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -823,8 +823,7 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,
 		return ret;
 	}
 
-	tt_has_data = ttm && (ttm_tt_is_populated(ttm) ||
-			      (ttm->page_flags & TTM_TT_FLAG_SWAPPED));
+	tt_has_data = ttm && (ttm_tt_is_populated(ttm) || ttm_tt_is_swapped(ttm));
 
 	move_lacks_source = !old_mem || (handle_system_ccs ? (!bo->ccs_cleared) :
 					 (!mem_type_is_vram(old_mem_type) && !tt_has_data));

From 03c69957c46ccdd042741c818460332c650f3ccf Mon Sep 17 00:00:00 2001
From: Yugansh Mittal <mittalyugansh1@gmail.com>
Date: Sun, 24 Aug 2025 17:20:51 +0530
Subject: [PATCH 0765/1455] drm/amdgpu: atomfirmware.h: fix multiple spelling
 mistakes
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch corrects several typographical errors in atomfirmware.h.
The fixes improve readability and maintain consistency in the codebase.
No functional changes are introduced.

Corrected terms include:
- aligment    → alignment
- Offest      → Offset
- defintion   → definition
- swithing    → switching
- calcualted  → calculated
- compability → compatibility
- intenal     → internal
- sequece     → sequence
- indiate     → indicate
- stucture    → structure
- regiser     → register

Signed-off-by: Yugansh Mittal <mittalyugansh1@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/include/atomfirmware.h | 30 +++++++++++-----------
 1 file changed, 15 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h
index 5c86423c2e92..3d083010e734 100644
--- a/drivers/gpu/drm/amd/include/atomfirmware.h
+++ b/drivers/gpu/drm/amd/include/atomfirmware.h
@@ -211,7 +211,7 @@ atom_bios_string          = "ATOM"
 };
 */
 
-#pragma pack(1)                          /* BIOS data must use byte aligment*/
+#pragma pack(1)                          /* BIOS data must use byte alignment*/
 
 enum atombios_image_offset{
   OFFSET_TO_ATOM_ROM_HEADER_POINTER          = 0x00000048,
@@ -255,8 +255,8 @@ struct atom_rom_header_v2_2
   uint16_t subsystem_vendor_id;
   uint16_t subsystem_id;
   uint16_t pci_info_offset;
-  uint16_t masterhwfunction_offset;      //Offest for SW to get all command function offsets, Don't change the position
-  uint16_t masterdatatable_offset;       //Offest for SW to get all data table offsets, Don't change the position
+  uint16_t masterhwfunction_offset;      //Offset for SW to get all command function offsets, Don't change the position
+  uint16_t masterdatatable_offset;       //Offset for SW to get all data table offsets, Don't change the position
   uint16_t reserved;
   uint32_t pspdirtableoffset;
 };
@@ -453,7 +453,7 @@ struct atom_dtd_format
   uint8_t   refreshrate;
 };
 
-/* atom_dtd_format.modemiscinfo defintion */
+/* atom_dtd_format.modemiscinfo definition */
 enum atom_dtd_format_modemiscinfo{
   ATOM_HSYNC_POLARITY    = 0x0002,
   ATOM_VSYNC_POLARITY    = 0x0004,
@@ -678,7 +678,7 @@ struct lcd_info_v2_1
   uint32_t reserved1[8];
 };
 
-/* lcd_info_v2_1.panel_misc defintion */
+/* lcd_info_v2_1.panel_misc definition */
 enum atom_lcd_info_panel_misc{
   ATOM_PANEL_MISC_FPDI            =0x0002,
 };
@@ -716,7 +716,7 @@ enum atom_gpio_pin_assignment_gpio_id {
   /* gpio_id pre-define id for multiple usage */
   /* GPIO use to control PCIE_VDDC in certain SLT board */
   PCIE_VDDC_CONTROL_GPIO_PINID = 56,
-  /* if PP_AC_DC_SWITCH_GPIO_PINID in Gpio_Pin_LutTable, AC/DC swithing feature is enable */
+  /* if PP_AC_DC_SWITCH_GPIO_PINID in Gpio_Pin_LutTable, AC/DC switching feature is enable */
   PP_AC_DC_SWITCH_GPIO_PINID = 60,
   /* VDDC_REGULATOR_VRHOT_GPIO_PINID in Gpio_Pin_LutTable, VRHot feature is enable */
   VDDC_VRHOT_GPIO_PINID = 61,
@@ -734,7 +734,7 @@ enum atom_gpio_pin_assignment_gpio_id {
 struct atom_gpio_pin_lut_v2_1
 {
   struct  atom_common_table_header  table_header;
-  /*the real number of this included in the structure is calcualted by using the (whole structure size - the header size)/size of atom_gpio_pin_lut  */
+  /*the real number of this included in the structure is calculated by using the (whole structure size - the header size)/size of atom_gpio_pin_lut  */
   struct  atom_gpio_pin_assignment  gpio_pin[];
 };
 
@@ -997,7 +997,7 @@ enum atom_connector_layout_info_mini_type_def {
 
 enum atom_display_device_tag_def{
   ATOM_DISPLAY_LCD1_SUPPORT            = 0x0002, //an embedded display is either an LVDS or eDP signal type of display
-  ATOM_DISPLAY_LCD2_SUPPORT			       = 0x0020, //second edp device tag 0x0020 for backward compability
+  ATOM_DISPLAY_LCD2_SUPPORT            = 0x0020, //second edp device tag 0x0020 for backward compatibility
   ATOM_DISPLAY_DFP1_SUPPORT            = 0x0008,
   ATOM_DISPLAY_DFP2_SUPPORT            = 0x0080,
   ATOM_DISPLAY_DFP3_SUPPORT            = 0x0200,
@@ -1011,7 +1011,7 @@ struct atom_display_object_path_v2
 {
   uint16_t display_objid;                  //Connector Object ID or Misc Object ID
   uint16_t disp_recordoffset;
-  uint16_t encoderobjid;                   //first encoder closer to the connector, could be either an external or intenal encoder
+  uint16_t encoderobjid;                   //first encoder closer to the connector, could be either an external or internal encoder
   uint16_t extencoderobjid;                //2nd encoder after the first encoder, from the connector point of view;
   uint16_t encoder_recordoffset;
   uint16_t extencoder_recordoffset;
@@ -1023,7 +1023,7 @@ struct atom_display_object_path_v2
 struct atom_display_object_path_v3 {
 	uint16_t display_objid; //Connector Object ID or Misc Object ID
 	uint16_t disp_recordoffset;
-	uint16_t encoderobjid; //first encoder closer to the connector, could be either an external or intenal encoder
+	uint16_t encoderobjid; //first encoder closer to the connector, could be either an external or internal encoder
 	uint16_t reserved1; //only on USBC case, otherwise always = 0
 	uint16_t reserved2; //reserved and always = 0
 	uint16_t reserved3; //reserved and always = 0
@@ -3547,7 +3547,7 @@ struct atom_voltage_object_header_v4{
 enum atom_voltage_object_mode 
 {
    VOLTAGE_OBJ_GPIO_LUT              =  0,        //VOLTAGE and GPIO Lookup table ->atom_gpio_voltage_object_v4
-   VOLTAGE_OBJ_VR_I2C_INIT_SEQ       =  3,        //VOLTAGE REGULATOR INIT sequece through I2C -> atom_i2c_voltage_object_v4
+   VOLTAGE_OBJ_VR_I2C_INIT_SEQ       =  3,        //VOLTAGE REGULATOR INIT sequence through I2C -> atom_i2c_voltage_object_v4
    VOLTAGE_OBJ_PHASE_LUT             =  4,        //Set Vregulator Phase lookup table ->atom_gpio_voltage_object_v4
    VOLTAGE_OBJ_SVID2                 =  7,        //Indicate voltage control by SVID2 ->atom_svid2_voltage_object_v4
    VOLTAGE_OBJ_EVV                   =  8, 
@@ -3585,7 +3585,7 @@ struct atom_gpio_voltage_object_v4
 {
    struct atom_voltage_object_header_v4 header;  // voltage mode = VOLTAGE_OBJ_GPIO_LUT or VOLTAGE_OBJ_PHASE_LUT
    uint8_t  gpio_control_id;                     // default is 0 which indicate control through CG VID mode 
-   uint8_t  gpio_entry_num;                      // indiate the entry numbers of Votlage/Gpio value Look up table
+   uint8_t  gpio_entry_num;                      // indicate the entry numbers of Votlage/Gpio value Look up table
    uint8_t  phase_delay_us;                      // phase delay in unit of micro second
    uint8_t  reserved;   
    uint32_t gpio_mask_val;                         // GPIO Mask value
@@ -4507,8 +4507,8 @@ struct amd_acpi_description_header{
 struct uefi_acpi_vfct{
   struct   amd_acpi_description_header sheader;
   uint8_t  tableUUID[16];    //0x24
-  uint32_t vbiosimageoffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the stucture.
-  uint32_t lib1Imageoffset;  //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the stucture.
+  uint32_t vbiosimageoffset; //0x34. Offset to the first GOP_VBIOS_CONTENT block from the beginning of the structure.
+  uint32_t lib1Imageoffset;  //0x38. Offset to the first GOP_LIB1_CONTENT block from the beginning of the structure.
   uint32_t reserved[4];      //0x3C
 };
 
@@ -4540,7 +4540,7 @@ struct gop_lib1_content {
 /* 
   ***************************************************************************
                    Scratch Register definitions
-  Each number below indicates which scratch regiser request, Active and 
+  Each number below indicates which scratch register request, Active and
   Connect all share the same definitions as display_device_tag defines
   *************************************************************************** 
 */   

From 7563e21a977f1e77212ecbe081d632b70689b9f7 Mon Sep 17 00:00:00 2001
From: "Kavithesh A.S" <kavitheshnitt@gmail.com>
Date: Sun, 24 Aug 2025 01:55:40 +0530
Subject: [PATCH 0766/1455] drm/amd/display: Document num_rmcm_3dluts in
 mpc_color_caps

Fix a kernel-doc warning by documenting the num_rmcm_3dluts member of struct mpc_color_caps.

v2: improve comment (Melissa)

Signed-off-by: Kavithesh A.S <kavitheshnitt@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 274ab744a331..c2bf583f30a7 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -234,6 +234,7 @@ struct lut3d_caps {
  * @ogam_ram: programmable out gamma LUT
  * @ocsc: output color space conversion matrix
  * @num_3dluts: MPC 3D LUT; always assumes a preceding shaper LUT
+ * @num_rmcm_3dluts: number of RMCM 3D LUTS; always assumes a preceding shaper LUT
  * @shared_3d_lut: shared 3D LUT flag. Can be either DPP or MPC, but single
  * instance
  * @ogam_rom_caps: pre-definied curve caps for regamma 1D LUT

From 14570c649aef0e80126b7e12710c65564283aa5f Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Sat, 16 Aug 2025 22:23:05 +0800
Subject: [PATCH 0767/1455] drm/amd/display: use max() to improve code

Use max() to reduce the code and improve readability.

No functional changes.

Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c   | 7 ++-----
 .../gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c | 6 +-----
 .../amd/display/dc/dml2/dml21/dml21_translation_helper.c   | 6 +-----
 3 files changed, 4 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index 084994c650c4..8376e2b0e73d 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -1047,11 +1047,8 @@ static void dcn32_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 			&num_entries_per_clk->num_fclk_levels);
 	clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK);
 
-	if (num_entries_per_clk->num_memclk_levels >= num_entries_per_clk->num_fclk_levels) {
-		num_levels = num_entries_per_clk->num_memclk_levels;
-	} else {
-		num_levels = num_entries_per_clk->num_fclk_levels;
-	}
+	num_levels = max(num_entries_per_clk->num_memclk_levels, num_entries_per_clk->num_fclk_levels);
+
 	clk_mgr_base->bw_params->max_memclk_mhz =
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_memclk_levels - 1].memclk_mhz;
 	clk_mgr_base->bw_params->clk_table.num_entries = num_levels ? num_levels : 1;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
index b59703467128..47ff4c965d76 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
@@ -1404,11 +1404,7 @@ static void dcn401_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_fclk_levels - 1].fclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = 0;
 
-	if (num_entries_per_clk->num_memclk_levels >= num_entries_per_clk->num_fclk_levels) {
-		num_levels = num_entries_per_clk->num_memclk_levels;
-	} else {
-		num_levels = num_entries_per_clk->num_fclk_levels;
-	}
+	num_levels = max(num_entries_per_clk->num_memclk_levels, num_entries_per_clk->num_fclk_levels);
 
 	clk_mgr_base->bw_params->clk_table.num_entries = num_levels ? num_levels : 1;
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
index 23fdb17f851a..d6c77d96e4f7 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
@@ -463,11 +463,7 @@ static void populate_dml21_timing_config_from_stream_state(struct dml2_timing_cf
 				(stream->timing.h_total * (long long)calc_max_hardware_v_total(stream)));
 	}
 
-	if (stream->timing.min_refresh_in_uhz > min_hardware_refresh_in_uhz) {
-		timing->drr_config.min_refresh_uhz = stream->timing.min_refresh_in_uhz;
-	} else {
-		timing->drr_config.min_refresh_uhz = min_hardware_refresh_in_uhz;
-	}
+	timing->drr_config.min_refresh_uhz = max(stream->timing.min_refresh_in_uhz, min_hardware_refresh_in_uhz);
 
 	if (dml_ctx->config.callbacks.get_max_flickerless_instant_vtotal_increase &&
 			stream->ctx->dc->config.enable_fpo_flicker_detection == 1)

From 010219ccec738e5cd03ea65878d6a61ce404d560 Mon Sep 17 00:00:00 2001
From: "David (Ming Qiang) Wu" <David.Wu3@amd.com>
Date: Mon, 25 Aug 2025 15:15:42 -0400
Subject: [PATCH 0768/1455] drm/amdgpu/vcn: remove unused code in vcn_v4_0.c

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 1785786a72f8..d0d27790b73b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1621,7 +1621,6 @@ static int vcn_v4_0_stop(struct amdgpu_vcn_inst *vinst)
 
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
 		vcn_v4_0_stop_dpg_mode(vinst);
-		r = 0;
 		goto done;
 	}
 

From b1d83546cf55816c60d6e953ebd0dd008f17be9e Mon Sep 17 00:00:00 2001
From: "David (Ming Qiang) Wu" <David.Wu3@amd.com>
Date: Mon, 25 Aug 2025 15:42:23 -0400
Subject: [PATCH 0769/1455] drm/amdgpu/vcn: add instance number to VCN version
 message

For multiple VCN instances case we get multiple lines of the same
message like below:

  amdgpu 0000:43:00.0: amdgpu: Found VCN firmware Version ENC: 1.24 DEC: 9 VEP: 0 Revision: 11
  amdgpu 0000:43:00.0: amdgpu: Found VCN firmware Version ENC: 1.24 DEC: 9 VEP: 0 Revision: 11

By adding instance number to the log message for multiple VCN instances,
each line will clearly indicate which VCN instance it refers to.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index fd8ebf4b5a82..5a90abcea0ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -185,16 +185,16 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev, int i)
 		dec_ver = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xf;
 		vep = (le32_to_cpu(hdr->ucode_version) >> 28) & 0xf;
 		dev_info(adev->dev,
-			 "Found VCN firmware Version ENC: %u.%u DEC: %u VEP: %u Revision: %u\n",
-			 enc_major, enc_minor, dec_ver, vep, fw_rev);
+			 "[VCN instance %d] Found VCN firmware Version ENC: %u.%u DEC: %u VEP: %u Revision: %u\n",
+			 i, enc_major, enc_minor, dec_ver, vep, fw_rev);
 	} else {
 		unsigned int version_major, version_minor, family_id;
 
 		family_id = le32_to_cpu(hdr->ucode_version) & 0xff;
 		version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff;
 		version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff;
-		dev_info(adev->dev, "Found VCN firmware Version: %u.%u Family ID: %u\n",
-			 version_major, version_minor, family_id);
+		dev_info(adev->dev, "[VCN instance %d] Found VCN firmware Version: %u.%u Family ID: %u\n",
+			 i, version_major, version_minor, family_id);
 	}
 
 	bo_size = AMDGPU_VCN_STACK_SIZE + AMDGPU_VCN_CONTEXT_SIZE;

From c8d6e90abe50377110f92702fbebc6efdd22391d Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Wed, 27 Aug 2025 09:40:27 +0800
Subject: [PATCH 0770/1455] drm/amdgpu: Notify pmfw bad page threshold exceeded

Notify pmfw when bad page threshold is exceeded, no matter the module
parameter 'bad_page_threshold' is set or not.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
index e3e6b32e4830..3eb3fb55ccb0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras_eeprom.c
@@ -777,9 +777,10 @@ amdgpu_ras_eeprom_update_header(struct amdgpu_ras_eeprom_control *control)
 				control->tbl_rai.health_percent = 0;
 			}
 			ras->is_rma = true;
-			/* ignore the -ENOTSUPP return value */
-			amdgpu_dpm_send_rma_reason(adev);
 		}
+
+		/* ignore the -ENOTSUPP return value */
+		amdgpu_dpm_send_rma_reason(adev);
 	}
 
 	if (control->tbl_hdr.version >= RAS_TABLE_VER_V2_1)

From aab8b689aded255425db3d80c0030d1ba02fe2ef Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 27 Aug 2025 13:29:17 +0800
Subject: [PATCH 0771/1455] drm/amdgpu/sdma: bump firmware version checks for
 user queue support

Using the previous firmware could lead to problems with
PROTECTED_FENCE_SIGNAL commands, specifically causing register
conflicts between MCU_DBG0 and MCU_DBG1.

The updated firmware versions ensure proper alignment
and unification of the SDMA_SUBOP_PROTECTED_FENCE_SIGNAL value with SDMA 7.x,
resolving these hardware coordination issues

Fixes: e8cca30d8b34 ("drm/amdgpu/sdma6: add ucode version checks for userq support")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
index e6d8eddda2bf..db6e41967f12 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c
@@ -1377,7 +1377,7 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 
 	switch (amdgpu_ip_version(adev, SDMA0_HWIP, 0)) {
 	case IP_VERSION(6, 0, 0):
-		if ((adev->sdma.instance[0].fw_version >= 24) && !adev->sdma.disable_uq)
+		if ((adev->sdma.instance[0].fw_version >= 27) && !adev->sdma.disable_uq)
 			adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
 		break;
 	case IP_VERSION(6, 0, 1):
@@ -1385,11 +1385,11 @@ static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 			adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
 		break;
 	case IP_VERSION(6, 0, 2):
-		if ((adev->sdma.instance[0].fw_version >= 21) && !adev->sdma.disable_uq)
+		if ((adev->sdma.instance[0].fw_version >= 23) && !adev->sdma.disable_uq)
 			adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
 		break;
 	case IP_VERSION(6, 0, 3):
-		if ((adev->sdma.instance[0].fw_version >= 25) && !adev->sdma.disable_uq)
+		if ((adev->sdma.instance[0].fw_version >= 27) && !adev->sdma.disable_uq)
 			adev->userq_funcs[AMDGPU_HW_IP_DMA] = &userq_mes_funcs;
 		break;
 	case IP_VERSION(6, 1, 0):

From 89c3503bc6b65493696dd202e0d9f037450d4cbf Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Mon, 25 Aug 2025 09:38:32 +0800
Subject: [PATCH 0772/1455] drm/amd/amdgpu: unified amdgpu ip block name

v1:
1. Unified amdgpu ip block name print with format
   "{ip_type}_v{major}_{minor}_{rev}"

2. Avoid IP block name conflicts for SMU/PSP ip block

v2:
Update IP block print format to keep legacy IP block name (Alex)
"{ip_type}_v{major}_{minor}_{rev} ({funcs->name})"

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 36 ++++++++++++++++++++--
 1 file changed, 34 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 9c373bdadb39..e7878a32e43c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2447,6 +2447,33 @@ int amdgpu_device_ip_block_version_cmp(struct amdgpu_device *adev,
 	return 1;
 }
 
+static const char *ip_block_names[] = {
+	[AMD_IP_BLOCK_TYPE_COMMON] = "common",
+	[AMD_IP_BLOCK_TYPE_GMC] = "gmc",
+	[AMD_IP_BLOCK_TYPE_IH] = "ih",
+	[AMD_IP_BLOCK_TYPE_SMC] = "smu",
+	[AMD_IP_BLOCK_TYPE_PSP] = "psp",
+	[AMD_IP_BLOCK_TYPE_DCE] = "dce",
+	[AMD_IP_BLOCK_TYPE_GFX] = "gfx",
+	[AMD_IP_BLOCK_TYPE_SDMA] = "sdma",
+	[AMD_IP_BLOCK_TYPE_UVD] = "uvd",
+	[AMD_IP_BLOCK_TYPE_VCE] = "vce",
+	[AMD_IP_BLOCK_TYPE_ACP] = "acp",
+	[AMD_IP_BLOCK_TYPE_VCN] = "vcn",
+	[AMD_IP_BLOCK_TYPE_MES] = "mes",
+	[AMD_IP_BLOCK_TYPE_JPEG] = "jpeg",
+	[AMD_IP_BLOCK_TYPE_VPE] = "vpe",
+	[AMD_IP_BLOCK_TYPE_UMSCH_MM] = "umsch_mm",
+	[AMD_IP_BLOCK_TYPE_ISP] = "isp",
+};
+
+static const char *ip_block_name(struct amdgpu_device *adev, enum amd_ip_block_type type)
+{
+	int idx = (int)type;
+
+	return idx < ARRAY_SIZE(ip_block_names) ? ip_block_names[idx] : "unknown";
+}
+
 /**
  * amdgpu_device_ip_block_add
  *
@@ -2475,8 +2502,13 @@ int amdgpu_device_ip_block_add(struct amdgpu_device *adev,
 		break;
 	}
 
-	dev_info(adev->dev, "detected ip block number %d <%s>\n",
-		 adev->num_ip_blocks, ip_block_version->funcs->name);
+	dev_info(adev->dev, "detected ip block number %d <%s_v%d_%d_%d> (%s)\n",
+		 adev->num_ip_blocks,
+		 ip_block_name(adev, ip_block_version->type),
+		 ip_block_version->major,
+		 ip_block_version->minor,
+		 ip_block_version->rev,
+		 ip_block_version->funcs->name);
 
 	adev->ip_blocks[adev->num_ip_blocks].adev = adev;
 

From 01152c30eef972c5ca3b3eeb14f2984fa48d18c2 Mon Sep 17 00:00:00 2001
From: Mangesh Gadre <Mangesh.Gadre@amd.com>
Date: Mon, 25 Aug 2025 21:18:42 +0800
Subject: [PATCH 0773/1455] drm/amdgpu: Avoid jpeg v5.0.1 poison irq call trace
 on sriov guest

Sriov guest side doesn't init ras feature hence the poison irq shouldn't
be put during hw fini

Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
index a2d781898767..47bde62b3909 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -315,7 +315,7 @@ static int jpeg_v5_0_1_hw_fini(struct amdgpu_ip_block *ip_block)
 			ret = jpeg_v5_0_1_set_powergating_state(ip_block, AMD_PG_STATE_GATE);
 	}
 
-	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG))
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__JPEG) && !amdgpu_sriov_vf(adev))
 		amdgpu_irq_put(adev, &adev->jpeg.inst->ras_poison_irq, 0);
 
 	return ret;

From 37551277dfed796b6749e4fa52bdb62403cfdb42 Mon Sep 17 00:00:00 2001
From: Mangesh Gadre <Mangesh.Gadre@amd.com>
Date: Mon, 25 Aug 2025 21:22:30 +0800
Subject: [PATCH 0774/1455] drm/amdgpu: Avoid vcn v5.0.1 poison irq call trace
 on sriov guest

Sriov guest side doesn't init ras feature hence the poison irq shouldn't
be put during hw fini

Signed-off-by: Mangesh Gadre <Mangesh.Gadre@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 7cb21e2b4eb0..4b01e35ad7ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -326,7 +326,7 @@ static int vcn_v5_0_1_hw_fini(struct amdgpu_ip_block *ip_block)
 			vinst->set_pg_state(vinst, AMD_PG_STATE_GATE);
 	}
 
-	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN))
+	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__VCN) && !amdgpu_sriov_vf(adev))
 		amdgpu_irq_put(adev, &adev->vcn.inst->ras_poison_irq, 0);
 
 	return 0;

From 9c0442286f84a5036958b3d8eb00bf0bb070dbd1 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 14 Aug 2025 13:52:50 +0530
Subject: [PATCH 0775/1455] drm/amdgpu: Check vcn state before profile switch

The patch uses power state of VCN instances for requesting video
profile.

In idle worker of a vcn instance, when there is no outstanding
submisssion or fence, the instance is put to power gated state. When
all instances are powered off that means video profile is no longer
required. A request is made to turn off video profile.

A job submission starts with begin_use of ring, and at that time
vcn instance state is changed to power on. Subsequently a check is
made for active video profile, and if not active, a request is made.

Fixes: 3b669df92c85 ("drm/amdgpu/vcn: adjust workload profile handling")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 80 ++++++++++++++++---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h |  3 +
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c   | 27 +--------
 3 files changed, 56 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 5a90abcea0ac..ad415203d245 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -408,6 +408,54 @@ int amdgpu_vcn_resume(struct amdgpu_device *adev, int i)
 	return 0;
 }
 
+void amdgpu_vcn_get_profile(struct amdgpu_device *adev)
+{
+	int r;
+
+	mutex_lock(&adev->vcn.workload_profile_mutex);
+
+	if (adev->vcn.workload_profile_active) {
+		mutex_unlock(&adev->vcn.workload_profile_mutex);
+		return;
+	}
+	r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
+					    true);
+	if (r)
+		dev_warn(adev->dev,
+			 "(%d) failed to enable video power profile mode\n", r);
+	else
+		adev->vcn.workload_profile_active = true;
+	mutex_unlock(&adev->vcn.workload_profile_mutex);
+}
+
+void amdgpu_vcn_put_profile(struct amdgpu_device *adev)
+{
+	bool pg = true;
+	int r, i;
+
+	mutex_lock(&adev->vcn.workload_profile_mutex);
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
+		if (adev->vcn.inst[i].cur_state != AMD_PG_STATE_GATE) {
+			pg = false;
+			break;
+		}
+	}
+
+	if (pg) {
+		r = amdgpu_dpm_switch_power_profile(
+			adev, PP_SMC_POWER_PROFILE_VIDEO, false);
+		if (r)
+			dev_warn(
+				adev->dev,
+				"(%d) failed to disable video power profile mode\n",
+				r);
+		else
+			adev->vcn.workload_profile_active = false;
+	}
+
+	mutex_unlock(&adev->vcn.workload_profile_mutex);
+}
+
 static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 {
 	struct amdgpu_vcn_inst *vcn_inst =
@@ -415,7 +463,6 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 	struct amdgpu_device *adev = vcn_inst->adev;
 	unsigned int fences = 0, fence[AMDGPU_MAX_VCN_INSTANCES] = {0};
 	unsigned int i = vcn_inst->inst, j;
-	int r = 0;
 
 	if (adev->vcn.harvest_config & (1 << i))
 		return;
@@ -444,15 +491,8 @@ static void amdgpu_vcn_idle_work_handler(struct work_struct *work)
 		mutex_lock(&vcn_inst->vcn_pg_lock);
 		vcn_inst->set_pg_state(vcn_inst, AMD_PG_STATE_GATE);
 		mutex_unlock(&vcn_inst->vcn_pg_lock);
-		mutex_lock(&adev->vcn.workload_profile_mutex);
-		if (adev->vcn.workload_profile_active) {
-			r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
-							    false);
-			if (r)
-				dev_warn(adev->dev, "(%d) failed to disable video power profile mode\n", r);
-			adev->vcn.workload_profile_active = false;
-		}
-		mutex_unlock(&adev->vcn.workload_profile_mutex);
+		amdgpu_vcn_put_profile(adev);
+
 	} else {
 		schedule_delayed_work(&vcn_inst->idle_work, VCN_IDLE_TIMEOUT);
 	}
@@ -462,30 +502,11 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_vcn_inst *vcn_inst = &adev->vcn.inst[ring->me];
-	int r = 0;
 
 	atomic_inc(&vcn_inst->total_submission_cnt);
 
 	cancel_delayed_work_sync(&vcn_inst->idle_work);
 
-	/* We can safely return early here because we've cancelled the
-	 * the delayed work so there is no one else to set it to false
-	 * and we don't care if someone else sets it to true.
-	 */
-	if (adev->vcn.workload_profile_active)
-		goto pg_lock;
-
-	mutex_lock(&adev->vcn.workload_profile_mutex);
-	if (!adev->vcn.workload_profile_active) {
-		r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
-						    true);
-		if (r)
-			dev_warn(adev->dev, "(%d) failed to switch to video power profile mode\n", r);
-		adev->vcn.workload_profile_active = true;
-	}
-	mutex_unlock(&adev->vcn.workload_profile_mutex);
-
-pg_lock:
 	mutex_lock(&vcn_inst->vcn_pg_lock);
 	vcn_inst->set_pg_state(vcn_inst, AMD_PG_STATE_UNGATE);
 
@@ -513,6 +534,7 @@ pg_lock:
 		vcn_inst->pause_dpg_mode(vcn_inst, &new_state);
 	}
 	mutex_unlock(&vcn_inst->vcn_pg_lock);
+	amdgpu_vcn_get_profile(adev);
 }
 
 void amdgpu_vcn_ring_end_use(struct amdgpu_ring *ring)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index b3fb1d0e43fc..6d9acd36041d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -565,4 +565,7 @@ int amdgpu_vcn_reg_dump_init(struct amdgpu_device *adev,
 			     const struct amdgpu_hwip_reg_entry *reg, u32 count);
 void amdgpu_vcn_dump_ip_state(struct amdgpu_ip_block *ip_block);
 void amdgpu_vcn_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p);
+void amdgpu_vcn_get_profile(struct amdgpu_device *adev);
+void amdgpu_vcn_put_profile(struct amdgpu_device *adev);
+
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index 3a7c137a83ef..904b94bc8693 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -116,7 +116,6 @@ static void vcn_v2_5_idle_work_handler(struct work_struct *work)
 	struct amdgpu_device *adev = vcn_inst->adev;
 	unsigned int fences = 0, fence[AMDGPU_MAX_VCN_INSTANCES] = {0};
 	unsigned int i, j;
-	int r = 0;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
 		struct amdgpu_vcn_inst *v = &adev->vcn.inst[i];
@@ -149,15 +148,7 @@ static void vcn_v2_5_idle_work_handler(struct work_struct *work)
 	if (!fences && !atomic_read(&adev->vcn.inst[0].total_submission_cnt)) {
 		amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
 						       AMD_PG_STATE_GATE);
-		mutex_lock(&adev->vcn.workload_profile_mutex);
-		if (adev->vcn.workload_profile_active) {
-			r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
-							    false);
-			if (r)
-				dev_warn(adev->dev, "(%d) failed to disable video power profile mode\n", r);
-			adev->vcn.workload_profile_active = false;
-		}
-		mutex_unlock(&adev->vcn.workload_profile_mutex);
+		amdgpu_vcn_put_profile(adev);
 	} else {
 		schedule_delayed_work(&adev->vcn.inst[0].idle_work, VCN_IDLE_TIMEOUT);
 	}
@@ -167,7 +158,6 @@ static void vcn_v2_5_ring_begin_use(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_vcn_inst *v = &adev->vcn.inst[ring->me];
-	int r = 0;
 
 	atomic_inc(&adev->vcn.inst[0].total_submission_cnt);
 
@@ -177,20 +167,6 @@ static void vcn_v2_5_ring_begin_use(struct amdgpu_ring *ring)
 	 * the delayed work so there is no one else to set it to false
 	 * and we don't care if someone else sets it to true.
 	 */
-	if (adev->vcn.workload_profile_active)
-		goto pg_lock;
-
-	mutex_lock(&adev->vcn.workload_profile_mutex);
-	if (!adev->vcn.workload_profile_active) {
-		r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
-						    true);
-		if (r)
-			dev_warn(adev->dev, "(%d) failed to switch to video power profile mode\n", r);
-		adev->vcn.workload_profile_active = true;
-	}
-	mutex_unlock(&adev->vcn.workload_profile_mutex);
-
-pg_lock:
 	mutex_lock(&adev->vcn.inst[0].vcn_pg_lock);
 	amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
 					       AMD_PG_STATE_UNGATE);
@@ -218,6 +194,7 @@ pg_lock:
 		v->pause_dpg_mode(v, &new_state);
 	}
 	mutex_unlock(&adev->vcn.inst[0].vcn_pg_lock);
+	amdgpu_vcn_get_profile(adev);
 }
 
 static void vcn_v2_5_ring_end_use(struct amdgpu_ring *ring)

From edc9d17ea17b9d4eb7b6474a922364bb23652f9a Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Mon, 18 Aug 2025 11:44:28 +0530
Subject: [PATCH 0776/1455] drm/amd/pm: Make use of __free for cleanup

Use __free(kfree) for memory alloc cleanups in SMUv13.0.6

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 63 ++++++++-----------
 1 file changed, 25 insertions(+), 38 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index fe2d48dc47af..e37b7b5358ea 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -538,6 +538,9 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
 	struct smu_table *tables = smu_table->tables;
+	void *gpu_metrics_table __free(kfree) = NULL;
+	void *driver_pptable __free(kfree) = NULL;
+	void *metrics_table __free(kfree) = NULL;
 	struct amdgpu_device *adev = smu->adev;
 	int gpu_metrcs_size = METRICS_TABLE_SIZE;
 	int ret;
@@ -556,37 +559,32 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 		       PAGE_SIZE,
 		       AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT);
 
-	smu_table->metrics_table = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL);
-	if (!smu_table->metrics_table)
+	metrics_table = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL);
+	if (!metrics_table)
 		return -ENOMEM;
 	smu_table->metrics_time = 0;
 
 	smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_8);
-	smu_table->gpu_metrics_table =
+	gpu_metrics_table =
 		kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL);
-	if (!smu_table->gpu_metrics_table) {
-		kfree(smu_table->metrics_table);
+	if (!gpu_metrics_table)
 		return -ENOMEM;
-	}
 
-	smu_table->driver_pptable =
-		kzalloc(sizeof(struct PPTable_t), GFP_KERNEL);
-	if (!smu_table->driver_pptable) {
-		kfree(smu_table->metrics_table);
-		kfree(smu_table->gpu_metrics_table);
+	driver_pptable = kzalloc(sizeof(struct PPTable_t), GFP_KERNEL);
+	if (!driver_pptable)
 		return -ENOMEM;
-	}
 
 	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) ==
 	    IP_VERSION(13, 0, 12)) {
 		ret = smu_v13_0_12_tables_init(smu);
-		if (ret) {
-			kfree(smu_table->metrics_table);
-			kfree(smu_table->gpu_metrics_table);
+		if (ret)
 			return ret;
-		}
 	}
 
+	smu_table->gpu_metrics_table = no_free_ptr(gpu_metrics_table);
+	smu_table->metrics_table = no_free_ptr(metrics_table);
+	smu_table->driver_pptable = no_free_ptr(driver_pptable);
+
 	return 0;
 }
 
@@ -2622,9 +2620,9 @@ static ssize_t smu_v13_0_6_get_xcp_metrics(struct smu_context *smu, int xcp_id,
 	const u8 num_jpeg_rings = AMDGPU_MAX_JPEG_RINGS_4_0_3;
 	int version = smu_v13_0_6_get_metrics_version(smu);
 	struct amdgpu_partition_metrics_v1_0 *xcp_metrics;
+	MetricsTableV0_t *metrics_v0 __free(kfree) = NULL;
 	struct amdgpu_device *adev = smu->adev;
 	int ret, inst, i, j, k, idx;
-	MetricsTableV0_t *metrics_v0;
 	MetricsTableV1_t *metrics_v1;
 	MetricsTableV2_t *metrics_v2;
 	struct amdgpu_xcp *xcp;
@@ -2649,17 +2647,14 @@ static ssize_t smu_v13_0_6_get_xcp_metrics(struct smu_context *smu, int xcp_id,
 		return -ENOMEM;
 
 	ret = smu_v13_0_6_get_metrics_table(smu, metrics_v0, false);
-	if (ret) {
-		kfree(metrics_v0);
+	if (ret)
 		return ret;
-	}
 
 	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) ==
 		    IP_VERSION(13, 0, 12) &&
-	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) {
-		ret = smu_v13_0_12_get_xcp_metrics(smu, xcp, table, metrics_v0);
-		goto out;
-	}
+	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
+		return smu_v13_0_12_get_xcp_metrics(smu, xcp, table,
+						    metrics_v0);
 
 	metrics_v1 = (MetricsTableV1_t *)metrics_v0;
 	metrics_v2 = (MetricsTableV2_t *)metrics_v0;
@@ -2730,8 +2725,6 @@ static ssize_t smu_v13_0_6_get_xcp_metrics(struct smu_context *smu, int xcp_id,
 			idx++;
 		}
 	}
-out:
-	kfree(metrics_v0);
 
 	return sizeof(*xcp_metrics);
 }
@@ -2742,31 +2735,26 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 	struct gpu_metrics_v1_8 *gpu_metrics =
 		(struct gpu_metrics_v1_8 *)smu_table->gpu_metrics_table;
 	int version = smu_v13_0_6_get_metrics_version(smu);
+	MetricsTableV0_t *metrics_v0 __free(kfree) = NULL;
 	int ret = 0, xcc_id, inst, i, j, k, idx;
 	struct amdgpu_device *adev = smu->adev;
-	MetricsTableV0_t *metrics_v0;
 	MetricsTableV1_t *metrics_v1;
 	MetricsTableV2_t *metrics_v2;
 	struct amdgpu_xcp *xcp;
 	u16 link_width_level;
-	ssize_t num_bytes;
 	u8 num_jpeg_rings;
 	u32 inst_mask;
 	bool per_inst;
 
 	metrics_v0 = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL);
 	ret = smu_v13_0_6_get_metrics_table(smu, metrics_v0, false);
-	if (ret) {
-		kfree(metrics_v0);
+	if (ret)
 		return ret;
-	}
 
-	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 12) &&
-	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS))) {
-		num_bytes = smu_v13_0_12_get_gpu_metrics(smu, table, metrics_v0);
-		kfree(metrics_v0);
-		return num_bytes;
-	}
+	if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) ==
+		    IP_VERSION(13, 0, 12) &&
+	    smu_v13_0_6_cap_supported(smu, SMU_CAP(STATIC_METRICS)))
+		return smu_v13_0_12_get_gpu_metrics(smu, table, metrics_v0);
 
 	metrics_v1 = (MetricsTableV1_t *)metrics_v0;
 	metrics_v2 = (MetricsTableV2_t *)metrics_v0;
@@ -2952,7 +2940,6 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table
 	gpu_metrics->firmware_timestamp = GET_METRIC_FIELD(Timestamp, version);
 
 	*table = (void *)gpu_metrics;
-	kfree(metrics_v0);
 
 	return sizeof(*gpu_metrics);
 }

From 9f28af76fab0948b59673f69c10aeec47de11c60 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 27 Aug 2025 14:24:31 -0400
Subject: [PATCH 0777/1455] drm/amdgpu/mes11: make MES_MISC_OP_CHANGE_CONFIG
 failure non-fatal

If the firmware is too old, just warn and return success.

Fixes: 27b791514789 ("drm/amdgpu/mes: keep enforce isolation up to date")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4414
Cc: shaoyun.Liu@amd.com
Reviewed-by: Shaoyun.liu <Shaoyun.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 28eb846280dd..3f6a828cad8a 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -641,8 +641,9 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes,
 		break;
 	case MES_MISC_OP_CHANGE_CONFIG:
 		if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) < 0x63) {
-			dev_err(mes->adev->dev, "MES FW version must be larger than 0x63 to support limit single process feature.\n");
-			return -EINVAL;
+			dev_warn_once(mes->adev->dev,
+				      "MES FW version must be larger than 0x63 to support limit single process feature.\n");
+			return 0;
 		}
 		misc_pkt.opcode = MESAPI_MISC__CHANGE_CONFIG;
 		misc_pkt.change_config.opcode =

From c2fbf72fe3c2d08856e834ca43328a8829a261d8 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 28 Aug 2025 09:37:05 -0500
Subject: [PATCH 0778/1455] drm/amd: Re-enable common modes for eDP and LVDS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
Although compositors will add their own modes, Xorg won't use it's own
modes and will only stick to modes advertised by the driver. This mean a
user that used to pick 1024x768 could no longer access it unless the
panel's native resolution was 1024x768.

[How]
Revert commit 6d396e7ac1ce3 ("drm/amd/display: Disable common modes for
LVDS") and commit 7948afb46af92 ("drm/amd/display: Disable common modes
for eDP").

The panel will still use scaling for any non-native modes due to
commit 978fa2f6d0b12 ("drm/amd/display: Use scaling for non-native
resolutions on eDP")

Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4538
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250828140856.2887993-1-superm1@kernel.org
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index b944abea306d..12eb1a91a4e9 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8433,8 +8433,7 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector)
 				drm_add_modes_noedid(connector, 1920, 1080);
 	} else {
 		amdgpu_dm_connector_ddc_get_modes(connector, drm_edid);
-		if (encoder && (connector->connector_type != DRM_MODE_CONNECTOR_eDP) &&
-		    (connector->connector_type != DRM_MODE_CONNECTOR_LVDS))
+		if (encoder)
 			amdgpu_dm_connector_add_common_modes(encoder, connector);
 		amdgpu_dm_connector_add_freesync_modes(connector, drm_edid);
 	}

From 5eeb16ca727f11278b2917fd4311a7d7efb0bbd6 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 6 Aug 2025 10:47:50 -0400
Subject: [PATCH 0779/1455] drm/amdgpu: drop hw access in non-DC audio fini

We already disable the audio pins in hw_fini so
there is no need to do it again in sw_fini.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4481
Cc: oushixiong <oushixiong1025@163.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/dce_v10_0.c | 5 -----
 drivers/gpu/drm/amd/amdgpu/dce_v11_0.c | 5 -----
 drivers/gpu/drm/amd/amdgpu/dce_v6_0.c  | 5 -----
 drivers/gpu/drm/amd/amdgpu/dce_v8_0.c  | 5 -----
 4 files changed, 20 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index bf7c22f81cda..ba73518f5cdf 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -1462,17 +1462,12 @@ static int dce_v10_0_audio_init(struct amdgpu_device *adev)
 
 static void dce_v10_0_audio_fini(struct amdgpu_device *adev)
 {
-	int i;
-
 	if (!amdgpu_audio)
 		return;
 
 	if (!adev->mode_info.audio.enabled)
 		return;
 
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++)
-		dce_v10_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-
 	adev->mode_info.audio.enabled = false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 47e05783c4a0..b01d88d078fa 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -1511,17 +1511,12 @@ static int dce_v11_0_audio_init(struct amdgpu_device *adev)
 
 static void dce_v11_0_audio_fini(struct amdgpu_device *adev)
 {
-	int i;
-
 	if (!amdgpu_audio)
 		return;
 
 	if (!adev->mode_info.audio.enabled)
 		return;
 
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++)
-		dce_v11_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-
 	adev->mode_info.audio.enabled = false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index 276c025c4c03..81760a26f2ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -1451,17 +1451,12 @@ static int dce_v6_0_audio_init(struct amdgpu_device *adev)
 
 static void dce_v6_0_audio_fini(struct amdgpu_device *adev)
 {
-	int i;
-
 	if (!amdgpu_audio)
 		return;
 
 	if (!adev->mode_info.audio.enabled)
 		return;
 
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++)
-		dce_v6_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-
 	adev->mode_info.audio.enabled = false;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index e62ccf9eb73d..19a265bd4d19 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -1443,17 +1443,12 @@ static int dce_v8_0_audio_init(struct amdgpu_device *adev)
 
 static void dce_v8_0_audio_fini(struct amdgpu_device *adev)
 {
-	int i;
-
 	if (!amdgpu_audio)
 		return;
 
 	if (!adev->mode_info.audio.enabled)
 		return;
 
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++)
-		dce_v8_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-
 	adev->mode_info.audio.enabled = false;
 }
 

From 002a612023c8b105bd3829d81862dee04368d6de Mon Sep 17 00:00:00 2001
From: Relja Vojvodic <rvojvodi@amd.com>
Date: Thu, 14 Aug 2025 11:33:22 -0400
Subject: [PATCH 0780/1455] drm/amd/display: Increase minimum clock for TMDS
 420 with pipe splitting

[Why]
-Pipe splitting allows for clocks to be reduced, but when using TMDS 420,
reduced clocks lead to missed clocks cycles on clock resyncing

[How]
-Impose a minimum clock when using TMDS 420

Reviewed-by: Chris Park <chris.park@amd.com>
Signed-off-by: Relja Vojvodic <rvojvodi@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../src/dml2_core/dml2_core_dcn4_calcs.c      | 28 +++++++++++++------
 1 file changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c
index b9cff2198511..bf62d42b3f78 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4_calcs.c
@@ -1238,18 +1238,27 @@ static void CalculateDETBufferSize(
 
 static double CalculateRequiredDispclk(
 	enum dml2_odm_mode ODMMode,
-	double PixelClock)
+	double PixelClock,
+	bool isTMDS420)
 {
+	double DispClk;
 
 	if (ODMMode == dml2_odm_mode_combine_4to1) {
-		return PixelClock / 4.0;
+		DispClk = PixelClock / 4.0;
 	} else if (ODMMode == dml2_odm_mode_combine_3to1) {
-		return PixelClock / 3.0;
+		DispClk = PixelClock / 3.0;
 	} else if (ODMMode == dml2_odm_mode_combine_2to1) {
-		return PixelClock / 2.0;
+		DispClk = PixelClock / 2.0;
 	} else {
-		return PixelClock;
+		DispClk = PixelClock;
 	}
+
+	if (isTMDS420) {
+		double TMDS420MinPixClock = PixelClock / 2.0;
+		DispClk = math_max2(DispClk, TMDS420MinPixClock);
+	}
+
+	return DispClk;
 }
 
 static double TruncToValidBPP(
@@ -4122,11 +4131,12 @@ static noinline_for_stack void CalculateODMMode(
 	bool success;
 	bool UseDSC = DSCEnable && (NumberOfDSCSlices > 0);
 	enum dml2_odm_mode DecidedODMMode;
+	bool isTMDS420 = (OutFormat == dml2_420 && Output == dml2_hdmi);
 
-	SurfaceRequiredDISPCLKWithoutODMCombine = CalculateRequiredDispclk(dml2_odm_mode_bypass, PixelClock);
-	SurfaceRequiredDISPCLKWithODMCombineTwoToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_2to1, PixelClock);
-	SurfaceRequiredDISPCLKWithODMCombineThreeToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_3to1, PixelClock);
-	SurfaceRequiredDISPCLKWithODMCombineFourToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_4to1, PixelClock);
+	SurfaceRequiredDISPCLKWithoutODMCombine = CalculateRequiredDispclk(dml2_odm_mode_bypass, PixelClock, isTMDS420);
+	SurfaceRequiredDISPCLKWithODMCombineTwoToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_2to1, PixelClock, isTMDS420);
+	SurfaceRequiredDISPCLKWithODMCombineThreeToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_3to1, PixelClock, isTMDS420);
+	SurfaceRequiredDISPCLKWithODMCombineFourToOne = CalculateRequiredDispclk(dml2_odm_mode_combine_4to1, PixelClock, isTMDS420);
 #ifdef __DML_VBA_DEBUG__
 	DML_LOG_VERBOSE("DML::%s: ODMUse = %d\n", __func__, ODMUse);
 	DML_LOG_VERBOSE("DML::%s: Output = %d\n", __func__, Output);

From de63b05593573cb0106875452c706ebaf6faa8f7 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Tue, 19 Aug 2025 11:29:05 -0500
Subject: [PATCH 0781/1455] drm/amd/display: Optimize custom brightness curve
 interpolation

[Why]
Custom brightness curve works by walking through all data points one
by one.  When the brightness value is at either extreme this is a lot
of data points to walk.  This is especially noticeable when moving a
brightness slider around how it can lag.

[How]
Bisect the data points to find the closest for interpolation.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 64 +++++++++++--------
 1 file changed, 38 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 12eb1a91a4e9..242f98564261 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -4817,8 +4817,8 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap
 				      uint32_t *user_brightness)
 {
 	u32 brightness = scale_input_to_fw(min, max, *user_brightness);
-	u8 prev_signal = 0, prev_lum = 0;
-	int i = 0;
+	u8 lower_signal, upper_signal, upper_lum, lower_lum, lum;
+	int left, right;
 
 	if (amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)
 		return;
@@ -4826,32 +4826,44 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap
 	if (!caps->data_points)
 		return;
 
-	/* choose start to run less interpolation steps */
-	if (caps->luminance_data[caps->data_points/2].input_signal > brightness)
-		i = caps->data_points/2;
-	do {
-		u8 signal = caps->luminance_data[i].input_signal;
-		u8 lum = caps->luminance_data[i].luminance;
+	left = 0;
+	right = caps->data_points - 1;
+	while (left <= right) {
+		int mid = left + (right - left) / 2;
+		u8 signal = caps->luminance_data[mid].input_signal;
 
-		/*
-		 * brightness == signal: luminance is percent numerator
-		 * brightness < signal: interpolate between previous and current luminance numerator
-		 * brightness > signal: find next data point
-		 */
-		if (brightness > signal) {
-			prev_signal = signal;
-			prev_lum = lum;
-			i++;
-			continue;
+		/* Exact match found */
+		if (signal == brightness) {
+			lum = caps->luminance_data[mid].luminance;
+			goto scale;
 		}
-		if (brightness < signal)
-			lum = prev_lum + DIV_ROUND_CLOSEST((lum - prev_lum) *
-							   (brightness - prev_signal),
-							   signal - prev_signal);
-		*user_brightness = scale_fw_to_input(min, max,
-						     DIV_ROUND_CLOSEST(lum * brightness, 101));
-		return;
-	} while (i < caps->data_points);
+
+		if (signal < brightness)
+			left = mid + 1;
+		else
+			right = mid - 1;
+	}
+
+	/* verify bound */
+	if (left >= caps->data_points)
+		left = caps->data_points - 1;
+
+	/* At this point, left > right */
+	lower_signal = caps->luminance_data[right].input_signal;
+	upper_signal = caps->luminance_data[left].input_signal;
+	lower_lum = caps->luminance_data[right].luminance;
+	upper_lum = caps->luminance_data[left].luminance;
+
+	/* interpolate */
+	if (right == left || !lower_lum)
+		lum = upper_lum;
+	else
+		lum = lower_lum + DIV_ROUND_CLOSEST((upper_lum - lower_lum) *
+						    (brightness - lower_signal),
+						    upper_signal - lower_signal);
+scale:
+	*user_brightness = scale_fw_to_input(min, max,
+					     DIV_ROUND_CLOSEST(lum * brightness, 101));
 }
 
 static u32 convert_brightness_from_user(const struct amdgpu_dm_backlight_caps *caps,

From cf49a454e16675fad59c1080f909a7e51192a494 Mon Sep 17 00:00:00 2001
From: Austin Zheng <Austin.Zheng@amd.com>
Date: Thu, 14 Aug 2025 09:54:45 -0400
Subject: [PATCH 0782/1455] drm/amd/display: Add Component To Handle Bounding
 Box Values and IP Caps

[Why]
Bounding box values can be stored in multiple locations. (e.g. PMFW, VBIOS, DMUB).
The source and interpretation of these values can vary with DCN revision
so there should be a component that can gather these values and translate
them accordingly

[How]
Have component start with the statically defined values as a base.
Then update them as needed with DCN-specific logic
Guard this component with FPU flags since values need to be in float point.

Reviewed-by: Jun Lei <jun.lei@amd.com>
Signed-off-by: Austin Zheng <Austin.Zheng@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/Makefile          |   1 +
 drivers/gpu/drm/amd/display/dc/Makefile       |   1 +
 drivers/gpu/drm/amd/display/dc/core/dc.c      |   8 +-
 drivers/gpu/drm/amd/display/dc/dc.h           |   2 +-
 .../dc/dml2/dml21/dml21_translation_helper.c  | 372 +-----------------
 .../amd/display/dc/dml2/dml21/dml21_wrapper.c |   2 +-
 .../display/dc/inc/soc_and_ip_translator.h    |  24 ++
 .../display/dc/soc_and_ip_translator/Makefile |  19 +
 .../dcn401/dcn401_soc_and_ip_translator.c     | 304 ++++++++++++++
 .../dcn401/dcn401_soc_and_ip_translator.h     |  22 ++
 .../dcn42/dcn42_soc_and_ip_translator.c       |  27 ++
 .../dcn42/dcn42_soc_and_ip_translator.h       |  16 +
 .../soc_and_ip_translator.c                   |  37 ++
 13 files changed, 477 insertions(+), 358 deletions(-)
 create mode 100644 drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h
 create mode 100644 drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile
 create mode 100644 drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c
 create mode 100644 drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h
 create mode 100644 drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c
 create mode 100644 drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h
 create mode 100644 drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c

diff --git a/drivers/gpu/drm/amd/display/Makefile b/drivers/gpu/drm/amd/display/Makefile
index 89d605de0595..0084a8d55254 100644
--- a/drivers/gpu/drm/amd/display/Makefile
+++ b/drivers/gpu/drm/amd/display/Makefile
@@ -44,6 +44,7 @@ subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/mmhubbub
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/mpc
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/opp
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/pg
+subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/dc/soc_and_ip_translator
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/inc
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/freesync
 subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/color
diff --git a/drivers/gpu/drm/amd/display/dc/Makefile b/drivers/gpu/drm/amd/display/dc/Makefile
index 3c9ecea7eebc..dc943abd6dba 100644
--- a/drivers/gpu/drm/amd/display/dc/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/Makefile
@@ -37,6 +37,7 @@ DC_LIBS += dcn301
 DC_LIBS += dcn31
 DC_LIBS += dml
 DC_LIBS += dml2
+DC_LIBS += soc_and_ip_translator
 endif
 
 DC_LIBS += dce120
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index c30d9ee51c83..261b7d43e91d 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -84,6 +84,7 @@
 
 #if defined(CONFIG_DRM_AMD_DC_FP)
 #include "dml2/dml2_internal_types.h"
+#include "soc_and_ip_translator.h"
 #endif
 
 #include "dce/dmub_outbox.h"
@@ -949,7 +950,9 @@ static void dc_destruct(struct dc *dc)
 	}
 
 	dc_destroy_resource_pool(dc);
-
+#ifdef CONFIG_DRM_AMD_DC_FP
+	dc_destroy_soc_and_ip_translator(&dc->soc_and_ip_translator);
+#endif
 	if (dc->link_srv)
 		link_destroy_link_service(&dc->link_srv);
 
@@ -1153,6 +1156,9 @@ static bool dc_construct(struct dc *dc,
 		dc->res_pool->funcs->update_bw_bounding_box(dc, dc->clk_mgr->bw_params);
 		DC_FP_END();
 	}
+	dc->soc_and_ip_translator = dc_create_soc_and_ip_translator(dc_ctx->dce_version);
+	if (!dc->soc_and_ip_translator)
+		goto fail;
 #endif
 
 	if (!create_links(dc, init_params->num_virtual_links))
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index c2bf583f30a7..e81b488cd72d 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1776,7 +1776,7 @@ struct dc {
 	struct dml2_configuration_options dml2_options;
 	struct dml2_configuration_options dml2_dc_power_options;
 	enum dc_acpi_cm_power_state power_state;
-
+	struct soc_and_ip_translator *soc_and_ip_translator;
 };
 
 struct dc_scaling_info {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
index d6c77d96e4f7..f6879e622271 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
@@ -8,7 +8,7 @@
 #include "dml2_internal_types.h"
 #include "dml21_utils.h"
 #include "dml21_translation_helper.h"
-#include "bounding_boxes/dcn4_soc_bb.h"
+#include "soc_and_ip_translator.h"
 
 static void dml21_populate_pmo_options(struct dml2_pmo_options *pmo_options,
 		const struct dc *in_dc,
@@ -38,375 +38,37 @@ static void dml21_populate_pmo_options(struct dml2_pmo_options *pmo_options,
 	pmo_options->disable_drr_clamped_when_var_active = in_dc->debug.disable_fams_gaming == INGAME_FAMS_DISABLE;
 }
 
-/*
- * Populate dml_init based on default static values in soc bb. The default
- * values are for reference and support at least minimal operation of current
- * SoC and DCN hardware. The values could be modifed by subsequent override
- * functions to reflect our true hardware capability.
- */
-static void populate_default_dml_init_params(struct dml2_initialize_instance_in_out *dml_init,
-		const struct dml2_configuration_options *config,
-		const struct dc *in_dc)
+static enum dml2_project_id dml21_dcn_revision_to_dml2_project_id(enum dce_version dcn_version)
 {
-	switch (in_dc->ctx->dce_version) {
+	enum dml2_project_id project_id;
+	switch (dcn_version) {
 	case DCN_VERSION_4_01:
-		dml_init->options.project_id = dml2_project_dcn4x_stage2_auto_drr_svp;
-		dml21_populate_pmo_options(&dml_init->options.pmo_options, in_dc, config);
-		dml_init->soc_bb = dml2_socbb_dcn401;
-		dml_init->soc_bb.qos_parameters = dml_dcn4_variant_a_soc_qos_params;
-		dml_init->ip_caps = dml2_dcn401_max_ip_caps;
+		project_id = dml2_project_dcn4x_stage2_auto_drr_svp;
 		break;
 	default:
-		memset(dml_init, 0, sizeof(*dml_init));
+		project_id = dml2_project_invalid;
 		DC_ERR("unsupported dcn version for DML21!");
-		return;
-	}
-}
-
-static void override_dml_init_with_values_from_hardware_default(struct dml2_initialize_instance_in_out *dml_init,
-		const struct dml2_configuration_options *config,
-		const struct dc *in_dc)
-{
-	dml_init->soc_bb.dchub_refclk_mhz = in_dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000;
-	dml_init->soc_bb.dprefclk_mhz = in_dc->clk_mgr->dprefclk_khz / 1000;
-	dml_init->soc_bb.dispclk_dppclk_vco_speed_mhz = in_dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
-}
-
-/*
- * SMU stands for System Management Unit. It is a power management processor.
- * It owns the initialization of dc's clock table and programming of clock values
- * based on dc's requests.
- * Our clock values in base soc bb is a dummy placeholder. The real clock values
- * are retrieved from SMU firmware to dc clock table at runtime.
- * This function overrides our dummy placeholder values with real values in dc
- * clock table.
- */
-static void override_dml_init_with_values_from_smu(
-		struct dml2_initialize_instance_in_out *dml_init,
-		const struct dml2_configuration_options *config,
-		const struct dc *in_dc)
-{
-	int i;
-	const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params;
-	const struct clk_limit_table *dc_clk_table = &dc_bw_params->clk_table;
-	struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table;
-
-	if (!in_dc->clk_mgr->funcs->is_smu_present ||
-			!in_dc->clk_mgr->funcs->is_smu_present(in_dc->clk_mgr))
-		/* skip if smu is not present */
-		return;
-
-	/* dcfclk */
-	if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) {
-		dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->dcfclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz &&
-						dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) {
-						dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000;
-						dml_clk_table->dcfclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->dcfclk.clk_values_khz[i] = 0;
-						dml_clk_table->dcfclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->dcfclk.clk_values_khz[i] = 0;
-			}
-		}
+		break;
 	}
 
-	/* fclk */
-	if (dc_clk_table->num_entries_per_clk.num_fclk_levels) {
-		dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->fclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz &&
-						dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) {
-						dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000;
-						dml_clk_table->fclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->fclk.clk_values_khz[i] = 0;
-						dml_clk_table->fclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->fclk.clk_values_khz[i] = 0;
-			}
-		}
-	}
-
-	/* uclk */
-	if (dc_clk_table->num_entries_per_clk.num_memclk_levels) {
-		dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->uclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz &&
-						dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) {
-						dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000;
-						dml_clk_table->uclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->uclk.clk_values_khz[i] = 0;
-						dml_clk_table->uclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->uclk.clk_values_khz[i] = 0;
-			}
-		}
-	}
-
-	/* dispclk */
-	if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) {
-		dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->dispclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz &&
-						dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) {
-						dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000;
-						dml_clk_table->dispclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->dispclk.clk_values_khz[i] = 0;
-						dml_clk_table->dispclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->dispclk.clk_values_khz[i] = 0;
-			}
-		}
-	}
-
-	/* dppclk */
-	if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) {
-		dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->dppclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz &&
-						dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) {
-						dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000;
-						dml_clk_table->dppclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->dppclk.clk_values_khz[i] = 0;
-						dml_clk_table->dppclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->dppclk.clk_values_khz[i] = 0;
-			}
-		}
-	}
-
-	/* dtbclk */
-	if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) {
-		dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->dtbclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz &&
-						dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) {
-						dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000;
-						dml_clk_table->dtbclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->dtbclk.clk_values_khz[i] = 0;
-						dml_clk_table->dtbclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->dtbclk.clk_values_khz[i] = 0;
-			}
-		}
-	}
-
-	/* socclk */
-	if (dc_clk_table->num_entries_per_clk.num_socclk_levels) {
-		dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels;
-		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
-			if (i < dml_clk_table->socclk.num_clk_values) {
-				if (config->use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz &&
-						dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) {
-					if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) {
-						dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000;
-						dml_clk_table->socclk.num_clk_values = i + 1;
-					} else {
-						dml_clk_table->socclk.clk_values_khz[i] = 0;
-						dml_clk_table->socclk.num_clk_values = i;
-					}
-				} else {
-					dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000;
-				}
-			} else {
-				dml_clk_table->socclk.clk_values_khz[i] = 0;
-			}
-		}
-	}
-}
-
-static void override_dml_init_with_values_from_vbios(
-		struct dml2_initialize_instance_in_out *dml_init,
-		const struct dml2_configuration_options *config,
-		const struct dc *in_dc)
-{
-	const struct clk_bw_params *dc_bw_params = in_dc->clk_mgr->bw_params;
-	struct dml2_soc_bb *dml_soc_bb = &dml_init->soc_bb;
-	struct dml2_soc_state_table *dml_clk_table = &dml_init->soc_bb.clk_table;
-
-	if (in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns > 0)
-		dml_soc_bb->power_management_parameters.dram_clk_change_blackout_us =
-				(in_dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns + 9) / 10;
-
-	if (in_dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns > 0)
-		dml_soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us =
-				(in_dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns + 9) / 10;
-
-	if (in_dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns > 0)
-		dml_soc_bb->power_management_parameters.stutter_exit_latency_us =
-			(in_dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns + 9) / 10;
-
-	if (dc_bw_params->num_channels) {
-		dml_clk_table->dram_config.channel_count = dc_bw_params->num_channels;
-		dml_soc_bb->mall_allocated_for_dcn_mbytes = in_dc->caps.mall_size_total / 1048576;
-	} else if (in_dc->ctx->dc_bios->vram_info.num_chans) {
-		dml_clk_table->dram_config.channel_count = in_dc->ctx->dc_bios->vram_info.num_chans;
-		dml_soc_bb->mall_allocated_for_dcn_mbytes = in_dc->caps.mall_size_total / 1048576;
-	}
-
-	if (dc_bw_params->dram_channel_width_bytes) {
-		dml_clk_table->dram_config.channel_width_bytes = dc_bw_params->dram_channel_width_bytes;
-	} else if (in_dc->ctx->dc_bios->vram_info.dram_channel_width_bytes) {
-		dml_clk_table->dram_config.channel_width_bytes = in_dc->ctx->dc_bios->vram_info.dram_channel_width_bytes;
-	}
-
-	dml_init->soc_bb.xtalclk_mhz = in_dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000;
-}
-
-
-static void override_dml_init_with_values_from_dmub(struct dml2_initialize_instance_in_out *dml_init,
-		const struct dml2_configuration_options *config,
-		const struct dc *in_dc)
-{
-	/*
-	 * TODO - There seems to be overlaps between the values overriden from
-	 * dmub and vbios. Investigate and identify the values that DMUB needs
-	 * to own.
-	 */
-// 	const struct dmub_soc_bb_params *dmub_bb_params =
-// 			(const struct dmub_soc_bb_params *)config->bb_from_dmub;
-
-// 	if (dmub_bb_params == NULL)
-// 		return;
-
-// 	if (dmub_bb_params->dram_clk_change_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us =
-// 			(double) dmub_bb_params->dram_clk_change_blackout_ns / 1000.0;
-// 	if (dmub_bb_params->dram_clk_change_read_only_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.dram_clk_change_read_only_us =
-// 			(double) dmub_bb_params->dram_clk_change_read_only_ns / 1000.0;
-// 	if (dmub_bb_params->dram_clk_change_write_only_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.dram_clk_change_write_only_us =
-// 			(double) dmub_bb_params->dram_clk_change_write_only_ns / 1000.0;
-// 	if (dmub_bb_params->fclk_change_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us =
-// 			(double) dmub_bb_params->fclk_change_blackout_ns / 1000.0;
-// 	if (dmub_bb_params->g7_ppt_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.g7_ppt_blackout_us =
-// 			(double) dmub_bb_params->g7_ppt_blackout_ns / 1000.0;
-// 	if (dmub_bb_params->stutter_enter_plus_exit_latency_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us =
-// 			(double) dmub_bb_params->stutter_enter_plus_exit_latency_ns / 1000.0;
-// 	if (dmub_bb_params->stutter_exit_latency_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us =
-// 			(double) dmub_bb_params->stutter_exit_latency_ns / 1000.0;
-// 	if (dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.z8_stutter_enter_plus_exit_latency_us =
-// 			(double) dmub_bb_params->z8_stutter_enter_plus_exit_latency_ns / 1000.0;
-// 	if (dmub_bb_params->z8_stutter_exit_latency_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.z8_stutter_exit_latency_us =
-// 			(double) dmub_bb_params->z8_stutter_exit_latency_ns / 1000.0;
-// 	if (dmub_bb_params->z8_min_idle_time_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.z8_min_idle_time =
-// 			(double) dmub_bb_params->z8_min_idle_time_ns / 1000.0;
-// #ifndef TRIM_DML2_DCN6B_IP_SENSITIVE
-// 	if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.lpddr5_dram_clk_change_blackout_us =
-// 			(double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0;
-// 	if (dmub_bb_params->type_b_ppt_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.lpddr5_ppt_blackout_us =
-// 			(double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0;
-// #else
-// 	if (dmub_bb_params->type_b_dram_clk_change_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.type_b_dram_clk_change_blackout_us =
-// 			(double) dmub_bb_params->type_b_dram_clk_change_blackout_ns / 1000.0;
-// 	if (dmub_bb_params->type_b_ppt_blackout_ns > 0)
-// 		dml_init->soc_bb.power_management_parameters.type_b_ppt_blackout_us =
-// 			(double) dmub_bb_params->type_b_ppt_blackout_ns / 1000.0;
-// #endif
-// 	if (dmub_bb_params->vmin_limit_dispclk_khz > 0)
-// 		dml_init->soc_bb.vmin_limit.dispclk_khz = dmub_bb_params->vmin_limit_dispclk_khz;
-// 	if (dmub_bb_params->vmin_limit_dcfclk_khz > 0)
-// 		dml_init->soc_bb.vmin_limit.dcfclk_khz = dmub_bb_params->vmin_limit_dcfclk_khz;
-//	if (dmub_bb_params->g7_temperature_read_blackout_ns > 0)
-//		dml_init->soc_bb.power_management_parameters.g7_temperature_read_blackout_us =
-//				(double) dmub_bb_params->g7_temperature_read_blackout_ns / 1000.0;
-}
-
-static void override_dml_init_with_values_from_software_policy(struct dml2_initialize_instance_in_out *dml_init,
-		const struct dml2_configuration_options *config,
-		const struct dc *in_dc)
-{
-	if (!config->use_native_soc_bb_construction) {
-		dml_init->soc_bb = config->external_socbb_ip_params->soc_bb;
-		dml_init->ip_caps = config->external_socbb_ip_params->ip_params;
-	}
-
-	if (in_dc->bb_overrides.sr_exit_time_ns)
-		dml_init->soc_bb.power_management_parameters.stutter_exit_latency_us =
-				in_dc->bb_overrides.sr_exit_time_ns / 1000.0;
-
-	if (in_dc->bb_overrides.sr_enter_plus_exit_time_ns)
-		dml_init->soc_bb.power_management_parameters.stutter_enter_plus_exit_latency_us =
-				in_dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
-
-	if (in_dc->bb_overrides.dram_clock_change_latency_ns)
-		dml_init->soc_bb.power_management_parameters.dram_clk_change_blackout_us =
-				in_dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
-
-	if (in_dc->bb_overrides.fclk_clock_change_latency_ns)
-		dml_init->soc_bb.power_management_parameters.fclk_change_blackout_us =
-				in_dc->bb_overrides.fclk_clock_change_latency_ns / 1000.0;
+	return project_id;
 }
 
 void dml21_populate_dml_init_params(struct dml2_initialize_instance_in_out *dml_init,
 		const struct dml2_configuration_options *config,
 		const struct dc *in_dc)
 {
-	populate_default_dml_init_params(dml_init, config, in_dc);
+	dml_init->options.project_id = dml21_dcn_revision_to_dml2_project_id(in_dc->ctx->dce_version);
 
-	override_dml_init_with_values_from_hardware_default(dml_init, config, in_dc);
+	if (config->use_native_soc_bb_construction) {
+		in_dc->soc_and_ip_translator->translator_funcs->get_soc_bb(&dml_init->soc_bb, in_dc, config);
+		in_dc->soc_and_ip_translator->translator_funcs->get_ip_caps(&dml_init->ip_caps);
+	} else {
+		dml_init->soc_bb = config->external_socbb_ip_params->soc_bb;
+		dml_init->ip_caps = config->external_socbb_ip_params->ip_params;
+	}
 
-	override_dml_init_with_values_from_smu(dml_init, config, in_dc);
-
-	override_dml_init_with_values_from_vbios(dml_init, config, in_dc);
-
-	override_dml_init_with_values_from_dmub(dml_init, config, in_dc);
-
-	override_dml_init_with_values_from_software_policy(dml_init, config, in_dc);
+	dml21_populate_pmo_options(&dml_init->options.pmo_options, in_dc, config);
 }
 
 static unsigned int calc_max_hardware_v_total(const struct dc_stream_state *stream)
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
index 03de3cf06ae5..798abb2b2e67 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
@@ -60,7 +60,7 @@ static void dml21_init(const struct dc *in_dc, struct dml2_context *dml_ctx, con
 
 	DC_FP_START();
 
-	dml21_populate_dml_init_params(&dml_ctx->v21.dml_init, config, in_dc);
+	dml21_populate_dml_init_params(&dml_ctx->v21.dml_init, &dml_ctx->config, in_dc);
 
 	dml2_initialize_instance(&dml_ctx->v21.dml_init);
 
diff --git a/drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h b/drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h
new file mode 100644
index 000000000000..23daf98b8aa8
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/inc/soc_and_ip_translator.h
@@ -0,0 +1,24 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#ifndef __SOC_AND_IP_TRANSLATOR_H__
+#define __SOC_AND_IP_TRANSLATOR_H__
+
+#include "dc.h"
+#include "dml_top_soc_parameter_types.h"
+
+struct soc_and_ip_translator_funcs {
+	void (*get_soc_bb)(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config);
+	void (*get_ip_caps)(struct dml2_ip_capabilities *dml_ip_caps);
+};
+
+struct soc_and_ip_translator {
+	const struct soc_and_ip_translator_funcs *translator_funcs;
+};
+
+struct soc_and_ip_translator *dc_create_soc_and_ip_translator(enum dce_version dc_version);
+void dc_destroy_soc_and_ip_translator(struct soc_and_ip_translator **soc_and_ip_translator);
+
+
+#endif // __SOC_AND_IP_TRANSLATOR_H__
diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile
new file mode 100644
index 000000000000..bc93356a0b5b
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/Makefile
@@ -0,0 +1,19 @@
+# SPDX-License-Identifier: MIT
+#
+# Copyright 2025 Advanced Micro Devices, Inc.
+# Makefile for bounding box component.
+# Floating point required due to nature of bounding box values
+
+soc_and_ip_translator_ccflags := $(CC_FLAGS_FPU)
+soc_and_ip_translator_rcflags := $(CC_FLAGS_NO_FPU)
+
+CFLAGS_$(AMDDALPATH)/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.o := $(soc_and_ip_translator_ccflags)
+
+CFLAGS_REMOVE_$(AMDDALPATH)/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.o := $(soc_and_ip_translator_rcflags)
+
+soc_and_ip_translator := soc_and_ip_translator.o
+soc_and_ip_translator += dcn401/dcn401_soc_and_ip_translator.o
+
+AMD_DAL_soc_and_ip_translator := $(addprefix $(AMDDALPATH)/dc/soc_and_ip_translator/, $(soc_and_ip_translator))
+
+AMD_DISPLAY_FILES += $(AMD_DAL_soc_and_ip_translator)
diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c
new file mode 100644
index 000000000000..3190c76eb482
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.c
@@ -0,0 +1,304 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#include "dcn401_soc_and_ip_translator.h"
+#include "bounding_boxes/dcn4_soc_bb.h"
+
+/* soc_and_ip_translator component used to get up-to-date values for bounding box.
+ * Bounding box values are stored in several locations and locations can vary with DCN revision.
+ * This component provides an interface to get DCN-specific bounding box values.
+ */
+
+static void get_default_soc_bb(struct dml2_soc_bb *soc_bb)
+{
+	memcpy(soc_bb, &dml2_socbb_dcn401, sizeof(struct dml2_soc_bb));
+	memcpy(&soc_bb->qos_parameters, &dml_dcn4_variant_a_soc_qos_params, sizeof(struct dml2_soc_qos_parameters));
+}
+
+/*
+ * DC clock table is obtained from SMU during runtime.
+ * SMU stands for System Management Unit. It is a power management processor.
+ * It owns the initialization of dc's clock table and programming of clock values
+ * based on dc's requests.
+ * Our clock values in base soc bb is a dummy placeholder. The real clock values
+ * are retrieved from SMU firmware to dc clock table at runtime.
+ * This function overrides our dummy placeholder values with real values in dc
+ * clock table.
+ */
+static void dcn401_convert_dc_clock_table_to_soc_bb_clock_table(
+		struct dml2_soc_state_table *dml_clk_table,
+		const struct clk_bw_params *dc_bw_params,
+		bool use_clock_dc_limits)
+{
+	int i;
+	const struct clk_limit_table *dc_clk_table;
+
+	if (dc_bw_params == NULL)
+		/* skip if bw params could not be obtained from smu */
+		return;
+
+	dc_clk_table = &dc_bw_params->clk_table;
+
+	/* dcfclk */
+	if (dc_clk_table->num_entries_per_clk.num_dcfclk_levels) {
+		dml_clk_table->dcfclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dcfclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->dcfclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dcfclk_mhz &&
+						dc_clk_table->entries[i].dcfclk_mhz > dc_bw_params->dc_mode_limit.dcfclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].dcfclk_mhz < dc_bw_params->dc_mode_limit.dcfclk_mhz) {
+						dml_clk_table->dcfclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dcfclk_mhz * 1000;
+						dml_clk_table->dcfclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->dcfclk.clk_values_khz[i] = 0;
+						dml_clk_table->dcfclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->dcfclk.clk_values_khz[i] = dc_clk_table->entries[i].dcfclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->dcfclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* fclk */
+	if (dc_clk_table->num_entries_per_clk.num_fclk_levels) {
+		dml_clk_table->fclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_fclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->fclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.fclk_mhz &&
+						dc_clk_table->entries[i].fclk_mhz > dc_bw_params->dc_mode_limit.fclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].fclk_mhz < dc_bw_params->dc_mode_limit.fclk_mhz) {
+						dml_clk_table->fclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.fclk_mhz * 1000;
+						dml_clk_table->fclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->fclk.clk_values_khz[i] = 0;
+						dml_clk_table->fclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->fclk.clk_values_khz[i] = dc_clk_table->entries[i].fclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->fclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* uclk */
+	if (dc_clk_table->num_entries_per_clk.num_memclk_levels) {
+		dml_clk_table->uclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_memclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->uclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.memclk_mhz &&
+						dc_clk_table->entries[i].memclk_mhz > dc_bw_params->dc_mode_limit.memclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].memclk_mhz < dc_bw_params->dc_mode_limit.memclk_mhz) {
+						dml_clk_table->uclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.memclk_mhz * 1000;
+						dml_clk_table->uclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->uclk.clk_values_khz[i] = 0;
+						dml_clk_table->uclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->uclk.clk_values_khz[i] = dc_clk_table->entries[i].memclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->uclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* dispclk */
+	if (dc_clk_table->num_entries_per_clk.num_dispclk_levels) {
+		dml_clk_table->dispclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dispclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->dispclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dispclk_mhz &&
+						dc_clk_table->entries[i].dispclk_mhz > dc_bw_params->dc_mode_limit.dispclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].dispclk_mhz < dc_bw_params->dc_mode_limit.dispclk_mhz) {
+						dml_clk_table->dispclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dispclk_mhz * 1000;
+						dml_clk_table->dispclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->dispclk.clk_values_khz[i] = 0;
+						dml_clk_table->dispclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->dispclk.clk_values_khz[i] = dc_clk_table->entries[i].dispclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->dispclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* dppclk */
+	if (dc_clk_table->num_entries_per_clk.num_dppclk_levels) {
+		dml_clk_table->dppclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dppclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->dppclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dppclk_mhz &&
+						dc_clk_table->entries[i].dppclk_mhz > dc_bw_params->dc_mode_limit.dppclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].dppclk_mhz < dc_bw_params->dc_mode_limit.dppclk_mhz) {
+						dml_clk_table->dppclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dppclk_mhz * 1000;
+						dml_clk_table->dppclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->dppclk.clk_values_khz[i] = 0;
+						dml_clk_table->dppclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->dppclk.clk_values_khz[i] = dc_clk_table->entries[i].dppclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->dppclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* dtbclk */
+	if (dc_clk_table->num_entries_per_clk.num_dtbclk_levels) {
+		dml_clk_table->dtbclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_dtbclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->dtbclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.dtbclk_mhz &&
+						dc_clk_table->entries[i].dtbclk_mhz > dc_bw_params->dc_mode_limit.dtbclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].dtbclk_mhz < dc_bw_params->dc_mode_limit.dtbclk_mhz) {
+						dml_clk_table->dtbclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.dtbclk_mhz * 1000;
+						dml_clk_table->dtbclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->dtbclk.clk_values_khz[i] = 0;
+						dml_clk_table->dtbclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->dtbclk.clk_values_khz[i] = dc_clk_table->entries[i].dtbclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->dtbclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* socclk */
+	if (dc_clk_table->num_entries_per_clk.num_socclk_levels) {
+		dml_clk_table->socclk.num_clk_values = dc_clk_table->num_entries_per_clk.num_socclk_levels;
+		for (i = 0; i < min(DML_MAX_CLK_TABLE_SIZE, MAX_NUM_DPM_LVL); i++) {
+			if (i < dml_clk_table->socclk.num_clk_values) {
+				if (use_clock_dc_limits && dc_bw_params->dc_mode_limit.socclk_mhz &&
+						dc_clk_table->entries[i].socclk_mhz > dc_bw_params->dc_mode_limit.socclk_mhz) {
+					if (i == 0 || dc_clk_table->entries[i-1].socclk_mhz < dc_bw_params->dc_mode_limit.socclk_mhz) {
+						dml_clk_table->socclk.clk_values_khz[i] = dc_bw_params->dc_mode_limit.socclk_mhz * 1000;
+						dml_clk_table->socclk.num_clk_values = i + 1;
+					} else {
+						dml_clk_table->socclk.clk_values_khz[i] = 0;
+						dml_clk_table->socclk.num_clk_values = i;
+					}
+				} else {
+					dml_clk_table->socclk.clk_values_khz[i] = dc_clk_table->entries[i].socclk_mhz * 1000;
+				}
+			} else {
+				dml_clk_table->socclk.clk_values_khz[i] = 0;
+			}
+		}
+	}
+
+	/* dram config */
+	dml_clk_table->dram_config.channel_count = dc_bw_params->num_channels;
+	dml_clk_table->dram_config.channel_width_bytes = dc_bw_params->dram_channel_width_bytes;
+}
+
+void dcn401_update_soc_bb_with_values_from_clk_mgr(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config)
+{
+	soc_bb->dprefclk_mhz = dc->clk_mgr->dprefclk_khz / 1000;
+	soc_bb->dispclk_dppclk_vco_speed_mhz = dc->clk_mgr->dentist_vco_freq_khz / 1000.0;
+	soc_bb->mall_allocated_for_dcn_mbytes = dc->caps.mall_size_total / (1024 * 1024);
+
+	if (dc->clk_mgr->funcs->is_smu_present &&
+			dc->clk_mgr->funcs->is_smu_present(dc->clk_mgr)) {
+		dcn401_convert_dc_clock_table_to_soc_bb_clock_table(&soc_bb->clk_table,
+			dc->clk_mgr->bw_params,
+			config->use_clock_dc_limits);
+	}
+}
+
+void dcn401_update_soc_bb_with_values_from_vbios(struct dml2_soc_bb *soc_bb, const struct dc *dc)
+{
+	soc_bb->dchub_refclk_mhz = dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000;
+	soc_bb->xtalclk_mhz = dc->ctx->dc_bios->fw_info.pll_info.crystal_frequency / 1000;
+
+	/* latencies in vbios are platform specific and should be used if provided */
+	if (dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns)
+		soc_bb->power_management_parameters.dram_clk_change_blackout_us =
+				dc->ctx->dc_bios->bb_info.dram_clock_change_latency_100ns / 10.0;
+
+	if (dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns)
+		soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us =
+				dc->ctx->dc_bios->bb_info.dram_sr_enter_exit_latency_100ns / 10.0;
+
+	if (dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns)
+		soc_bb->power_management_parameters.stutter_exit_latency_us =
+			dc->ctx->dc_bios->bb_info.dram_sr_exit_latency_100ns / 10.0;
+}
+
+void dcn401_update_soc_bb_with_values_from_software_policy(struct dml2_soc_bb *soc_bb, const struct dc *dc)
+{
+	/* set if the value is provided */
+	if (dc->bb_overrides.sr_exit_time_ns)
+		soc_bb->power_management_parameters.stutter_exit_latency_us =
+				dc->bb_overrides.sr_exit_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_enter_plus_exit_time_ns)
+		soc_bb->power_management_parameters.stutter_enter_plus_exit_latency_us =
+				dc->bb_overrides.sr_enter_plus_exit_time_ns / 1000.0;
+
+	if (dc->bb_overrides.dram_clock_change_latency_ns)
+		soc_bb->power_management_parameters.dram_clk_change_blackout_us =
+				dc->bb_overrides.dram_clock_change_latency_ns / 1000.0;
+
+	if (dc->bb_overrides.fclk_clock_change_latency_ns)
+		soc_bb->power_management_parameters.fclk_change_blackout_us =
+				dc->bb_overrides.fclk_clock_change_latency_ns / 1000.0;
+
+	//Z8 values not expected nor used on DCN401 but still added for completeness
+	if (dc->bb_overrides.sr_exit_z8_time_ns)
+		soc_bb->power_management_parameters.z8_stutter_exit_latency_us =
+				dc->bb_overrides.sr_exit_z8_time_ns / 1000.0;
+
+	if (dc->bb_overrides.sr_enter_plus_exit_z8_time_ns)
+		soc_bb->power_management_parameters.z8_stutter_enter_plus_exit_latency_us =
+				dc->bb_overrides.sr_enter_plus_exit_z8_time_ns / 1000.0;
+}
+
+static void apply_soc_bb_updates(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config)
+{
+	/* Individual modification can be overwritten even if it was obtained by a previous function.
+	 * Modifications are acquired in order of priority (lowest to highest).
+	 */
+	dc_assert_fp_enabled();
+
+	dcn401_update_soc_bb_with_values_from_clk_mgr(soc_bb, dc, config);
+	dcn401_update_soc_bb_with_values_from_vbios(soc_bb, dc);
+	dcn401_update_soc_bb_with_values_from_software_policy(soc_bb, dc);
+}
+
+void dcn401_get_soc_bb(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config)
+{
+	//get default soc_bb with static values
+	get_default_soc_bb(soc_bb);
+	//update soc_bb values with more accurate values
+	apply_soc_bb_updates(soc_bb, dc, config);
+}
+
+static void dcn401_get_ip_caps(struct dml2_ip_capabilities *ip_caps)
+{
+	*ip_caps = dml2_dcn401_max_ip_caps;
+}
+
+static struct soc_and_ip_translator_funcs dcn401_translator_funcs = {
+	.get_soc_bb = dcn401_get_soc_bb,
+	.get_ip_caps = dcn401_get_ip_caps,
+};
+
+void dcn401_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator)
+{
+	soc_and_ip_translator->translator_funcs = &dcn401_translator_funcs;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h
new file mode 100644
index 000000000000..21d842857601
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h
@@ -0,0 +1,22 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#ifndef _DCN401_SOC_AND_IP_TRANSLATOR_H_
+#define _DCN401_SOC_AND_IP_TRANSLATOR_H_
+
+#include "core_types.h"
+#include "dc.h"
+#include "clk_mgr.h"
+#include "soc_and_ip_translator.h"
+#include "dml2/dml21/inc/dml_top_soc_parameter_types.h"
+
+void dcn401_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator);
+
+/* Functions that can be re-used by higher DCN revisions of this component */
+void dcn401_get_soc_bb(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config);
+void dcn401_update_soc_bb_with_values_from_clk_mgr(struct dml2_soc_bb *soc_bb, const struct dc *dc, const struct dml2_configuration_options *config);
+void dcn401_update_soc_bb_with_values_from_vbios(struct dml2_soc_bb *soc_bb, const struct dc *dc);
+void dcn401_update_soc_bb_with_values_from_software_policy(struct dml2_soc_bb *soc_bb, const struct dc *dc);
+
+#endif /* _DCN401_SOC_AND_IP_TRANSLATOR_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c
new file mode 100644
index 000000000000..c9e224d262c9
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.c
@@ -0,0 +1,27 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#include "dcn42_soc_and_ip_translator.h"
+#include "soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h"
+#include "bounding_boxes/dcn42_soc_bb.h"
+
+/* soc_and_ip_translator component used to get up-to-date values for bounding box.
+ * Bounding box values are stored in several locations and locations can vary with DCN revision.
+ * This component provides an interface to get DCN-specific bounding box values.
+ */
+
+static void dcn42_get_ip_caps(struct dml2_ip_capabilities *ip_caps)
+{
+	*ip_caps = dml2_dcn42_max_ip_caps;
+}
+
+static struct soc_and_ip_translator_funcs dcn42_translator_funcs = {
+	.get_soc_bb = dcn401_get_soc_bb,
+	.get_ip_caps = dcn42_get_ip_caps,
+};
+
+void dcn42_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator)
+{
+	soc_and_ip_translator->translator_funcs = &dcn42_translator_funcs;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h
new file mode 100644
index 000000000000..914dcbb369a7
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/dcn42/dcn42_soc_and_ip_translator.h
@@ -0,0 +1,16 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#ifndef _DCN42_SOC_AND_IP_TRANSLATOR_H_
+#define _DCN42_SOC_AND_IP_TRANSLATOR_H_
+
+#include "core_types.h"
+#include "dc.h"
+#include "clk_mgr.h"
+#include "dml_top_soc_parameter_types.h"
+#include "soc_and_ip_translator.h"
+
+void dcn42_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator);
+
+#endif /* _DCN42_SOC_AND_IP_TRANSLATOR_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c
new file mode 100644
index 000000000000..0fc0e5a6c171
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/soc_and_ip_translator/soc_and_ip_translator.c
@@ -0,0 +1,37 @@
+// SPDX-License-Identifier: MIT
+//
+// Copyright 2025 Advanced Micro Devices, Inc.
+
+#include "soc_and_ip_translator.h"
+#include "soc_and_ip_translator/dcn401/dcn401_soc_and_ip_translator.h"
+
+static void dc_construct_soc_and_ip_translator(struct soc_and_ip_translator *soc_and_ip_translator,
+		enum dce_version dc_version)
+{
+	switch (dc_version) {
+	case DCN_VERSION_4_01:
+		dcn401_construct_soc_and_ip_translator(soc_and_ip_translator);
+		break;
+	default:
+		break;
+	}
+}
+
+struct soc_and_ip_translator *dc_create_soc_and_ip_translator(enum dce_version dc_version)
+{
+	struct soc_and_ip_translator *soc_and_ip_translator;
+
+	soc_and_ip_translator = kzalloc(sizeof(*soc_and_ip_translator), GFP_KERNEL);
+	if (!soc_and_ip_translator)
+		return NULL;
+
+	dc_construct_soc_and_ip_translator(soc_and_ip_translator, dc_version);
+
+	return soc_and_ip_translator;
+}
+
+void dc_destroy_soc_and_ip_translator(struct soc_and_ip_translator **soc_and_ip_translator)
+{
+	kfree(*soc_and_ip_translator);
+	*soc_and_ip_translator = NULL;
+}

From 306cbcc6f687d791ab3cc8fbbe30f5286fd0d1e5 Mon Sep 17 00:00:00 2001
From: Clay King <clayking@amd.com>
Date: Wed, 20 Aug 2025 15:04:29 -0400
Subject: [PATCH 0783/1455] drm/amd/display: incorrect conditions for failing
 dto calculations

[Why & How]
Previously, when calculating dto phase, we would incorrectly fail when phase
<=0 without additionally checking for the integer value. This meant that
calculations would incorrectly fail when the desired pixel clock was an exact
multiple of the reference clock.

Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Clay King <clayking@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c
index 668ee2d405fd..0b8ed9b94d3c 100644
--- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c
@@ -619,7 +619,7 @@ void dccg401_set_dp_dto(
 		dto_integer = div_u64(params->pixclk_hz, dto_modulo_hz);
 		dto_phase_hz = params->pixclk_hz - dto_integer * dto_modulo_hz;
 
-		if (dto_phase_hz <= 0) {
+		if (dto_phase_hz <= 0 && dto_integer <= 0) {
 			/* negative pixel rate should never happen */
 			BREAK_TO_DEBUGGER();
 			return;

From 645f74f1dc119dad5a2c7bbc05cc315e76883011 Mon Sep 17 00:00:00 2001
From: Ivan Lipski <ivan.lipski@amd.com>
Date: Wed, 20 Aug 2025 15:46:52 -0400
Subject: [PATCH 0784/1455] drm/amd/display: Clear the CUR_ENABLE register on
 DCN314 w/out DPP PG

[Why&How]
ON DCN314, clearing DPP SW structure without power gating it can cause a
double cursor in full screen with non-native scaling.

A W/A that clears CURSOR0_CONTROL cursor_enable flag if
dcn10_plane_atomic_power_down is called and DPP power gating is disabled.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4168
Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c  |  9 +++
 .../drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h  |  2 +
 .../drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c  |  1 +
 .../amd/display/dc/hwss/dcn314/dcn314_hwseq.c | 72 +++++++++++++++++++
 .../amd/display/dc/hwss/dcn314/dcn314_hwseq.h |  2 +
 .../amd/display/dc/hwss/dcn314/dcn314_init.c  |  1 +
 drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h   |  3 +
 7 files changed, 90 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c
index 75fb77bca83b..01480a04f85e 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.c
@@ -520,6 +520,15 @@ void dpp1_dppclk_control(
 		REG_UPDATE(DPP_CONTROL, DPP_CLOCK_ENABLE, 0);
 }
 
+void dpp_force_disable_cursor(struct dpp *dpp_base)
+{
+	struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base);
+
+	/* Force disable cursor */
+	REG_UPDATE(CURSOR0_CONTROL, CUR0_ENABLE, 0);
+	dpp_base->pos.cur0_ctl.bits.cur0_enable = 0;
+}
+
 static const struct dpp_funcs dcn10_dpp_funcs = {
 		.dpp_read_state = dpp_read_state,
 		.dpp_reset = dpp_reset,
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h
index c48139bed11f..f466182963f7 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn10/dcn10_dpp.h
@@ -1525,4 +1525,6 @@ void dpp1_construct(struct dcn10_dpp *dpp1,
 
 void dpp1_cm_get_gamut_remap(struct dpp *dpp_base,
 			     struct dpp_grph_csc_adjustment *adjust);
+void dpp_force_disable_cursor(struct dpp *dpp_base);
+
 #endif
diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c
index 2d70586cef40..09be2a90cc79 100644
--- a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c
+++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.c
@@ -1494,6 +1494,7 @@ static struct dpp_funcs dcn30_dpp_funcs = {
 	.dpp_dppclk_control		= dpp1_dppclk_control,
 	.dpp_set_hdr_multiplier		= dpp3_set_hdr_multiplier,
 	.dpp_get_gamut_remap		= dpp3_cm_get_gamut_remap,
+	.dpp_force_disable_cursor 	= dpp_force_disable_cursor,
 };
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
index e68f21fd5f0f..560984533950 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
@@ -528,3 +528,75 @@ void dcn314_disable_link_output(struct dc_link *link,
 
 	apply_symclk_on_tx_off_wa(link);
 }
+
+/**
+ * dcn314_dpp_pg_control - DPP power gate control.
+ *
+ * @hws: dce_hwseq reference.
+ * @dpp_inst: DPP instance reference.
+ * @power_on: true if we want to enable power gate, false otherwise.
+ *
+ * Enable or disable power gate in the specific DPP instance.
+ * If power gating is disabled, will force disable cursor in the DPP instance.
+ */
+void dcn314_dpp_pg_control(
+		struct dce_hwseq *hws,
+		unsigned int dpp_inst,
+		bool power_on)
+{
+	uint32_t power_gate = power_on ? 0 : 1;
+	uint32_t pwr_status = power_on ? 0 : 2;
+
+
+	if (hws->ctx->dc->debug.disable_dpp_power_gate) {
+		/* Workaround for DCN314 with disabled power gating */
+		if (!power_on) {
+
+			/* Force disable cursor if power gating is disabled */
+			struct dpp *dpp = hws->ctx->dc->res_pool->dpps[dpp_inst];
+			if (dpp && dpp->funcs->dpp_force_disable_cursor)
+				dpp->funcs->dpp_force_disable_cursor(dpp);
+		}
+		return;
+	}
+	if (REG(DOMAIN1_PG_CONFIG) == 0)
+		return;
+
+	switch (dpp_inst) {
+	case 0: /* DPP0 */
+		REG_UPDATE(DOMAIN1_PG_CONFIG,
+				DOMAIN1_POWER_GATE, power_gate);
+
+		REG_WAIT(DOMAIN1_PG_STATUS,
+				DOMAIN1_PGFSM_PWR_STATUS, pwr_status,
+				1, 1000);
+		break;
+	case 1: /* DPP1 */
+		REG_UPDATE(DOMAIN3_PG_CONFIG,
+				DOMAIN3_POWER_GATE, power_gate);
+
+		REG_WAIT(DOMAIN3_PG_STATUS,
+				DOMAIN3_PGFSM_PWR_STATUS, pwr_status,
+				1, 1000);
+		break;
+	case 2: /* DPP2 */
+		REG_UPDATE(DOMAIN5_PG_CONFIG,
+				DOMAIN5_POWER_GATE, power_gate);
+
+		REG_WAIT(DOMAIN5_PG_STATUS,
+				DOMAIN5_PGFSM_PWR_STATUS, pwr_status,
+				1, 1000);
+		break;
+	case 3: /* DPP3 */
+		REG_UPDATE(DOMAIN7_PG_CONFIG,
+				DOMAIN7_POWER_GATE, power_gate);
+
+		REG_WAIT(DOMAIN7_PG_STATUS,
+				DOMAIN7_PGFSM_PWR_STATUS, pwr_status,
+				1, 1000);
+		break;
+	default:
+		BREAK_TO_DEBUGGER();
+		break;
+	}
+}
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h
index 2305ad282f21..6c072d0274ea 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.h
@@ -47,4 +47,6 @@ void dcn314_dpp_root_clock_control(struct dce_hwseq *hws, unsigned int dpp_inst,
 
 void dcn314_disable_link_output(struct dc_link *link, const struct link_resource *link_res, enum signal_type signal);
 
+void dcn314_dpp_pg_control(struct dce_hwseq *hws, unsigned int dpp_inst, bool power_on);
+
 #endif /* __DC_HWSS_DCN314_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c
index a99145a30230..79faab1125d4 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_init.c
@@ -142,6 +142,7 @@ static const struct hwseq_private_funcs dcn314_private_funcs = {
 	.enable_power_gating_plane = dcn314_enable_power_gating_plane,
 	.dpp_root_clock_control = dcn314_dpp_root_clock_control,
 	.hubp_pg_control = dcn31_hubp_pg_control,
+	.dpp_pg_control = dcn314_dpp_pg_control,
 	.program_all_writeback_pipes_in_tree = dcn30_program_all_writeback_pipes_in_tree,
 	.update_odm = dcn314_update_odm,
 	.dsc_pg_control = dcn314_dsc_pg_control,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h
index 0c5675d1c593..1b7c085dc2cc 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dpp.h
@@ -349,6 +349,9 @@ struct dpp_funcs {
 		struct dpp *dpp_base,
 		enum dc_color_space color_space,
 		struct dc_csc_transform cursor_csc_color_matrix);
+
+	void (*dpp_force_disable_cursor)(struct dpp *dpp_base);
+
 };
 
 

From 16f2e14ee9429586dd58dc7ac1f5fc0755a0a0c6 Mon Sep 17 00:00:00 2001
From: Fangzhi Zuo <Jerry.Zuo@amd.com>
Date: Thu, 14 Aug 2025 14:41:44 -0400
Subject: [PATCH 0785/1455] drm/amd/display: Skip Check Runtime Link Setting
 for Specific Branch Device

[why]
Read link setting inside mode validation is not always the final downlink setting.
It is found true in Synaptics branch device.

At bootup, the preferred mode being set right after 1080p is set. It occurred
before graphic load. That modeset switch in a short period of time makes
the branch device switch back and forth from lower and higher link rate,
observed at Synaptics branch device.
DP2 RTK hub on the other hand, sticks to highest available downlink rate after bootup.

Existing check of runtime downlink setting in mode validation shows asynchronous at
branch device link switch, i.e., downlink switch to higher link rate not yet complete
when the mode validation tries to probe the downlink setting. That makes mode validation
checking downlink setting making wrong decision by pruning modes that should pass the
validation after the downlink setting switch is complete.

[how]
If Synaptics is found at the last branch, skip checking downlink setting
at mode validation.

Reviewed-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  3 ++
 .../display/amdgpu_dm/amdgpu_dm_mst_types.c   | 40 ++++++++++++++++++-
 2 files changed, 41 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index cbd107493f8a..ce74125c713e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -768,6 +768,9 @@ struct amdgpu_dm_connector {
 	uint16_t vc_full_pbn;
 	struct mutex handle_mst_msg_ready;
 
+	/* branch device specific data */
+	uint32_t branch_ieee_oui;
+
 	/* TODO see if we can merge with ddc_bus or make a dm_connector */
 	struct amdgpu_i2c_adapter *i2c;
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 32281bf62f70..6a817508c826 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -330,6 +330,34 @@ static bool retrieve_downstream_port_device(struct amdgpu_dm_connector *aconnect
 	return true;
 }
 
+static bool retrieve_branch_specific_data(struct amdgpu_dm_connector *aconnector)
+{
+	struct drm_connector *connector = &aconnector->base;
+	struct drm_dp_mst_port *port = aconnector->mst_output_port;
+	struct drm_dp_mst_port *port_parent;
+	struct drm_dp_aux *immediate_upstream_aux;
+	struct drm_dp_desc branch_desc;
+
+	if (!port->parent)
+		return false;
+
+	port_parent = port->parent->port_parent;
+
+	immediate_upstream_aux = port_parent ? &port_parent->aux : port->mgr->aux;
+
+	if (drm_dp_read_desc(immediate_upstream_aux, &branch_desc, true))
+		return false;
+
+	aconnector->branch_ieee_oui = (branch_desc.ident.oui[0] << 16) +
+				      (branch_desc.ident.oui[1] << 8) +
+				      (branch_desc.ident.oui[2]);
+
+	drm_dbg_dp(port->aux.drm_dev, "MST branch oui 0x%x detected at %s\n",
+		   aconnector->branch_ieee_oui, connector->name);
+
+	return true;
+}
+
 static int dm_dp_mst_get_modes(struct drm_connector *connector)
 {
 	struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
@@ -669,6 +697,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
 
 	drm_connector_set_path_property(connector, pathprop);
 
+	if (!retrieve_branch_specific_data(aconnector))
+		aconnector->branch_ieee_oui = 0;
+
 	/*
 	 * Initialize connector state before adding the connectror to drm and
 	 * framebuffer lists
@@ -1774,6 +1805,9 @@ static bool dp_get_link_current_set_bw(struct drm_dp_aux *aux, uint32_t *cur_lin
 	link_bw_set = data[DP_LINK_BW_SET - DP_LINK_BW_SET];
 	lane_count.raw = data[DP_LANE_COUNT_SET - DP_LINK_BW_SET];
 
+	drm_dbg_dp(aux->drm_dev, "MST_DSC downlink setting: %d, 0x%x x %d\n",
+		   dp_link_encoding, link_bw_set, lane_count.bits.LANE_COUNT_SET);
+
 	switch (dp_link_encoding) {
 	case DP_8b_10b_ENCODING:
 		link_rate = link_bw_set;
@@ -1870,8 +1904,10 @@ enum dc_status dm_dp_mst_is_port_support_mode(
 					end_link_bw = aconnector->mst_local_bw;
 				}
 
-				if (end_link_bw > 0 && stream_kbps > end_link_bw) {
-					DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link."
+				if (end_link_bw > 0 &&
+				    stream_kbps > end_link_bw &&
+				    aconnector->branch_ieee_oui != DP_BRANCH_DEVICE_ID_90CC24) {
+					DRM_DEBUG_DRIVER("MST_DSC dsc decode at last link. "
 							 "Mode required bw can't fit into last link\n");
 					return DC_FAIL_BANDWIDTH_VALIDATE;
 				}

From 5e610ed48a0df10aabd4fa17d7cee012dddf21e4 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 22 Aug 2025 16:13:17 -0400
Subject: [PATCH 0786/1455] drm/amd/display: [FW Promotion] Release 0.1.25.0

Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index b7d49a117fa7..02a4a20e3560 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -2364,6 +2364,7 @@ struct dmub_cmd_fams2_global_config {
 	union dmub_fams2_global_feature_config features;
 	uint32_t recovery_timeout_us;
 	uint32_t hwfq_flip_programming_delay_us;
+	uint32_t max_allow_to_target_delta_us; // how early DCN could assert P-State allow compared to the P-State target
 };
 
 union dmub_cmd_fams2_config {
@@ -4170,6 +4171,12 @@ enum replay_version {
  * Data passed from driver to FW in a DMUB_CMD___SET_REPLAY_VERSION command.
  */
 struct dmub_cmd_replay_set_version_data {
+	/**
+	 * Panel Instance.
+	 * Panel instance to identify which psr_state to use
+	 * Currently the support is only for 0 or 1
+	 */
+	uint8_t panel_inst;
 	/**
 	 * PSR version that FW should implement.
 	 */
@@ -4178,12 +4185,6 @@ struct dmub_cmd_replay_set_version_data {
 	 * PSR control version.
 	 */
 	uint8_t cmd_version;
-	/**
-	 * Panel Instance.
-	 * Panel instance to identify which psr_state to use
-	 * Currently the support is only for 0 or 1
-	 */
-	uint8_t panel_inst;
 	/**
 	 * Explicit padding to 4 byte boundary.
 	 */

From 220b2bd529fa660d7df802a30730373af9345dd8 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 22 Aug 2025 16:17:11 -0500
Subject: [PATCH 0787/1455] drm/amd/display: Promote DC to 3.2.348

Summary:

* Refactor bounding box values handling
* Fix incorrect condition to fail dto clk calculation
* Skip check downlink setting for a certain MST branch device
* Fix double cursor issue on dcn314

Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Alex Hung <alex.hung@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index e81b488cd72d..b41e66c31e6a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.347"
+#define DC_VER "3.2.348"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 585b2f685c56c5095cc22c7202bf74d8e9a73cdd Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 16:50:36 +0200
Subject: [PATCH 0788/1455] drm/amdgpu: Respect max pixel clock for HDMI and
 DVI-D (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Update the legacy (non-DC) display code to respect the maximum
pixel clock for HDMI and DVI-D. Reject modes that would require
a higher pixel clock than can be supported.

Also update the maximum supported HDMI clock value depending on
the ASIC type.

For reference, see the DC code:
check max_hdmi_pixel_clock in dce*_resource.c

v2:
Fix maximum clocks for DVI-D and DVI/HDMI adapters.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_connectors.c    | 57 ++++++++++++++-----
 1 file changed, 44 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 5e375e9c4f5d..a381de8648e5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -1195,29 +1195,60 @@ static void amdgpu_connector_dvi_force(struct drm_connector *connector)
 		amdgpu_connector->use_digital = true;
 }
 
+/**
+ * Returns the maximum supported HDMI (TMDS) pixel clock in KHz.
+ */
+static int amdgpu_max_hdmi_pixel_clock(const struct amdgpu_device *adev)
+{
+	if (adev->asic_type >= CHIP_POLARIS10)
+		return 600000;
+	else if (adev->asic_type >= CHIP_TONGA)
+		return 300000;
+	else
+		return 297000;
+}
+
+/**
+ * Validates the given display mode on DVI and HDMI connectors,
+ * including analog signals on DVI-I.
+ */
 static enum drm_mode_status amdgpu_connector_dvi_mode_valid(struct drm_connector *connector,
 					    const struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
+	const int max_hdmi_pixel_clock = amdgpu_max_hdmi_pixel_clock(adev);
+	const int max_dvi_single_link_pixel_clock = 165000;
+	int max_digital_pixel_clock_khz;
 
 	/* XXX check mode bandwidth */
 
-	if (amdgpu_connector->use_digital && (mode->clock > 165000)) {
-		if ((amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_I) ||
-		    (amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D) ||
-		    (amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_HDMI_TYPE_B)) {
-			return MODE_OK;
-		} else if (connector->display_info.is_hdmi) {
-			/* HDMI 1.3+ supports max clock of 340 Mhz */
-			if (mode->clock > 340000)
-				return MODE_CLOCK_HIGH;
-			else
-				return MODE_OK;
-		} else {
-			return MODE_CLOCK_HIGH;
+	if (amdgpu_connector->use_digital) {
+		switch (amdgpu_connector->connector_object_id) {
+		case CONNECTOR_OBJECT_ID_HDMI_TYPE_A:
+			max_digital_pixel_clock_khz = max_hdmi_pixel_clock;
+			break;
+		case CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_I:
+		case CONNECTOR_OBJECT_ID_SINGLE_LINK_DVI_D:
+			max_digital_pixel_clock_khz = max_dvi_single_link_pixel_clock;
+			break;
+		case CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_I:
+		case CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D:
+		case CONNECTOR_OBJECT_ID_HDMI_TYPE_B:
+			max_digital_pixel_clock_khz = max_dvi_single_link_pixel_clock * 2;
+			break;
 		}
+
+		/* When the display EDID claims that it's an HDMI display,
+		 * we use the HDMI encoder mode of the display HW,
+		 * so we should verify against the max HDMI clock here.
+		 */
+		if (connector->display_info.is_hdmi)
+			max_digital_pixel_clock_khz = max_hdmi_pixel_clock;
+
+		if (mode->clock > max_digital_pixel_clock_khz)
+			return MODE_CLOCK_HIGH;
 	}
 
 	/* check against the max pixel clock */

From 9ca595f501e48cec41c6c56becda6468a409b8ba Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Fri, 29 Aug 2025 09:55:39 +0200
Subject: [PATCH 0789/1455] drm/test: drm_exec: use kzalloc() to allocate GEM
 objects
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since commit e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
it is possible for test_prepare_array() to exceed a stack frame size of
2048 bytes depending on the exact configuration of the kernel.

  drivers/gpu/drm/tests/drm_exec_test.c: In function ‘test_prepare_array’:
  drivers/gpu/drm/tests/drm_exec_test.c:171:1: error: the frame size of 2128 bytes is larger than 2048 bytes [-Werror=frame-larger-than=]
    171 | }
        | ^
  cc1: all warnings being treated as errors
  make[6]: *** [scripts/Makefile.build:287: drivers/gpu/drm/tests/drm_exec_test.o] Error 1
  make[6]: *** Waiting for unfinished jobs....

In order to fix this, allocate the GEM objects in test_prepare_array()
with kzalloc(), rather than placing them on the stack.

Cc: Alice Ryhl <aliceryhl@google.com>
Cc: Christian König <christian.koenig@amd.com>
Fixes: e7fa80e2932c ("drm_gem: add mutex to drm_gem_object.gpuva")
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Nirmoy Das <nirmoyd@nvidia.com>
Link: https://lore.kernel.org/r/20250829075633.2306-1-dakr@kernel.org
[ Use kunit_kzalloc() instead of kzalloc(). - Danilo ]
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/drm/tests/drm_exec_test.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/tests/drm_exec_test.c b/drivers/gpu/drm/tests/drm_exec_test.c
index d6c4dd1194a0..3a20c788c51f 100644
--- a/drivers/gpu/drm/tests/drm_exec_test.c
+++ b/drivers/gpu/drm/tests/drm_exec_test.c
@@ -150,14 +150,22 @@ static void test_prepare(struct kunit *test)
 static void test_prepare_array(struct kunit *test)
 {
 	struct drm_exec_priv *priv = test->priv;
-	struct drm_gem_object gobj1 = { };
-	struct drm_gem_object gobj2 = { };
-	struct drm_gem_object *array[] = { &gobj1, &gobj2 };
+	struct drm_gem_object *gobj1;
+	struct drm_gem_object *gobj2;
+	struct drm_gem_object *array[] = {
+		(gobj1 = kunit_kzalloc(test, sizeof(*gobj1), GFP_KERNEL)),
+		(gobj2 = kunit_kzalloc(test, sizeof(*gobj2), GFP_KERNEL)),
+	};
 	struct drm_exec exec;
 	int ret;
 
-	drm_gem_private_object_init(priv->drm, &gobj1, PAGE_SIZE);
-	drm_gem_private_object_init(priv->drm, &gobj2, PAGE_SIZE);
+	if (!gobj1 || !gobj2) {
+		KUNIT_FAIL(test, "Failed to allocate GEM objects.\n");
+		return;
+	}
+
+	drm_gem_private_object_init(priv->drm, gobj1, PAGE_SIZE);
+	drm_gem_private_object_init(priv->drm, gobj2, PAGE_SIZE);
 
 	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
 	drm_exec_until_all_locked(&exec)
@@ -166,8 +174,8 @@ static void test_prepare_array(struct kunit *test)
 	KUNIT_EXPECT_EQ(test, ret, 0);
 	drm_exec_fini(&exec);
 
-	drm_gem_private_object_fini(&gobj1);
-	drm_gem_private_object_fini(&gobj2);
+	drm_gem_private_object_fini(gobj1);
+	drm_gem_private_object_fini(gobj2);
 }
 
 static void test_multiple_loops(struct kunit *test)

From 24de3daf6179bce3710527b8292d7ee6f1b56393 Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Thu, 28 Aug 2025 11:39:17 +0800
Subject: [PATCH 0790/1455] accel/amdxdna: Use int instead of u32 to store
 error codes

Change the 'ret' variable from u32 to int to store -EINVAL.  Storing the
negative error codes in unsigned type, doesn't cause an issue at runtime
but it's ugly as pants.

Additionally, assigning -EINVAL to u32 ret (i.e., u32 ret = -EINVAL) may
trigger a GCC warning when the -Wsign-conversion flag is enabled.

Fixes: aac243092b70 ("accel/amdxdna: Add command execution")
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Reviewed-by: Lizhi Hou <lizhi.hou@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250828033917.113364-1-rongqianfeng@vivo.com
---
 drivers/accel/amdxdna/aie2_ctx.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index 420467a5325c..e9f9b1fa5dc1 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -199,7 +199,7 @@ aie2_sched_resp_handler(void *handle, void __iomem *data, size_t size)
 {
 	struct amdxdna_sched_job *job = handle;
 	struct amdxdna_gem_obj *cmd_abo;
-	u32 ret = 0;
+	int ret = 0;
 	u32 status;
 
 	cmd_abo = job->cmd_bo;
@@ -229,7 +229,7 @@ static int
 aie2_sched_nocmd_resp_handler(void *handle, void __iomem *data, size_t size)
 {
 	struct amdxdna_sched_job *job = handle;
-	u32 ret = 0;
+	int ret = 0;
 	u32 status;
 
 	if (unlikely(!data))
@@ -257,7 +257,7 @@ aie2_sched_cmdlist_resp_handler(void *handle, void __iomem *data, size_t size)
 	u32 fail_cmd_status;
 	u32 fail_cmd_idx;
 	u32 cmd_status;
-	u32 ret = 0;
+	int ret = 0;
 
 	cmd_abo = job->cmd_bo;
 	if (unlikely(!data) || unlikely(size != sizeof(u32) * 3)) {

From f3e82936857b3bd77b824ecd2fa7839dd99ec0c6 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Fri, 15 Aug 2025 21:36:43 +0800
Subject: [PATCH 0791/1455] drm/nouveau: Replace redundant return value
 judgment with PTR_ERR_OR_ZERO()

Replace redundant return value judgment with PTR_ERR_OR_ZERO() to
enhance code readability.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250815133643.418089-1-liaoyuanhong@vivo.com
---
 drivers/gpu/drm/nouveau/nouveau_platform.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c
index a5ce8eb4a3be..8d5853deeee4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_platform.c
+++ b/drivers/gpu/drm/nouveau/nouveau_platform.c
@@ -30,10 +30,7 @@ static int nouveau_platform_probe(struct platform_device *pdev)
 	func = of_device_get_match_data(&pdev->dev);
 
 	drm = nouveau_platform_device_create(func, pdev, &device);
-	if (IS_ERR(drm))
-		return PTR_ERR(drm);
-
-	return 0;
+	return PTR_ERR_OR_ZERO(drm);
 }
 
 static void nouveau_platform_remove(struct platform_device *pdev)

From 256576ed6895b81290690ae3e2b9f62eb7c642fc Mon Sep 17 00:00:00 2001
From: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Date: Wed, 4 Jun 2025 14:28:23 +0200
Subject: [PATCH 0792/1455] drm/amdgpu: give each kernel job a unique id

Userspace jobs have drm_file.client_id as a unique identifier
as job's owners. For kernel jobs, we can allocate arbitrary
values - the risk of overlap with userspace ids is small (given
that it's a u64 value).
In the unlikely case the overlap happens, it'll only impact
trace events.

Since this ID is traced in the gpu_scheduler trace events, this
allows to determine the source of each job sent to the hardware.

To make grepping easier, the IDs are defined as they will appear
in the trace output.

Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Link: https://lore.kernel.org/r/20250604122827.2191-1-pierre-eric.pelloux-prayer@amd.com
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c     |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.c     |  5 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_job.h     | 19 +++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c    |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c  |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c     | 28 +++++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h     |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c     |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c     |  5 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c     |  8 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c      |  6 +++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h      |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c  |  4 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c   |  4 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c | 12 +++++----
 drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c       |  6 +++--
 drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c       |  6 +++--
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c    |  3 ++-
 19 files changed, 84 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index c80c8f543532..98aa99b314c9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -1474,7 +1474,8 @@ static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring)
 	owner = (void *)(unsigned long)atomic_inc_return(&counter);
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, &entity, owner,
-				     64, 0, &job);
+				     64, 0, &job,
+				     AMDGPU_KERNEL_JOB_ID_CLEANER_SHADER);
 	if (r)
 		goto err;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index 97b562a79ea8..9dcf51991b5b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -690,7 +690,7 @@ void amdgpu_gmc_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid,
 	r = amdgpu_job_alloc_with_ib(ring->adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     16 * 4, AMDGPU_IB_POOL_IMMEDIATE,
-				     &job);
+				     &job, AMDGPU_KERNEL_JOB_ID_FLUSH_GPU_TLB);
 	if (r)
 		goto error_alloc;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index 9b1c55115921..d020a890a0ea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -209,11 +209,12 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev,
 			     struct drm_sched_entity *entity, void *owner,
 			     size_t size, enum amdgpu_ib_pool_type pool_type,
-			     struct amdgpu_job **job)
+			     struct amdgpu_job **job, u64 k_job_id)
 {
 	int r;
 
-	r = amdgpu_job_alloc(adev, NULL, entity, owner, 1, job, 0);
+	r = amdgpu_job_alloc(adev, NULL, entity, owner, 1, job,
+			     k_job_id);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index 2f302266662b..4a6487eb6cb5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -44,6 +44,22 @@
 struct amdgpu_fence;
 enum amdgpu_ib_pool_type;
 
+/* Internal kernel job ids. (decreasing values, starting from U64_MAX). */
+#define AMDGPU_KERNEL_JOB_ID_VM_UPDATE              (18446744073709551615ULL)
+#define AMDGPU_KERNEL_JOB_ID_VM_UPDATE_PDES         (18446744073709551614ULL)
+#define AMDGPU_KERNEL_JOB_ID_VM_UPDATE_RANGE        (18446744073709551613ULL)
+#define AMDGPU_KERNEL_JOB_ID_VM_PT_CLEAR            (18446744073709551612ULL)
+#define AMDGPU_KERNEL_JOB_ID_TTM_MAP_BUFFER         (18446744073709551611ULL)
+#define AMDGPU_KERNEL_JOB_ID_TTM_ACCESS_MEMORY_SDMA (18446744073709551610ULL)
+#define AMDGPU_KERNEL_JOB_ID_TTM_COPY_BUFFER        (18446744073709551609ULL)
+#define AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE       (18446744073709551608ULL)
+#define AMDGPU_KERNEL_JOB_ID_MOVE_BLIT              (18446744073709551607ULL)
+#define AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER       (18446744073709551606ULL)
+#define AMDGPU_KERNEL_JOB_ID_CLEANER_SHADER         (18446744073709551605ULL)
+#define AMDGPU_KERNEL_JOB_ID_FLUSH_GPU_TLB          (18446744073709551604ULL)
+#define AMDGPU_KERNEL_JOB_ID_KFD_GART_MAP           (18446744073709551603ULL)
+#define AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST          (18446744073709551602ULL)
+
 struct amdgpu_job {
 	struct drm_sched_job    base;
 	struct amdgpu_vm	*vm;
@@ -96,7 +112,8 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev,
 			     struct drm_sched_entity *entity, void *owner,
 			     size_t size, enum amdgpu_ib_pool_type pool_type,
-			     struct amdgpu_job **job);
+			     struct amdgpu_job **job,
+			     u64 k_job_id);
 void amdgpu_job_set_resources(struct amdgpu_job *job, struct amdgpu_bo *gds,
 			      struct amdgpu_bo *gws, struct amdgpu_bo *oa);
 void amdgpu_job_free_resources(struct amdgpu_job *job);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index 82d58ac7afb0..4980595bcddd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -194,7 +194,8 @@ static int amdgpu_jpeg_dec_set_reg(struct amdgpu_ring *ring, uint32_t handle,
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL, ib_size_dw * 4,
-				     AMDGPU_IB_POOL_DIRECT, &job);
+				     AMDGPU_IB_POOL_DIRECT, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 122a88294883..d18bade9c98f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -1313,7 +1313,8 @@ void amdgpu_bo_release_notify(struct ttm_buffer_object *bo)
 	if (r)
 		goto out;
 
-	r = amdgpu_fill_buffer(abo, 0, &bo->base._resv, &fence, true);
+	r = amdgpu_fill_buffer(abo, 0, &bo->base._resv, &fence, true,
+			       AMDGPU_KERNEL_JOB_ID_CLEAR_ON_RELEASE);
 	if (WARN_ON(r))
 		goto out;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 27ab4e754b2a..428265046815 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -226,7 +226,8 @@ static int amdgpu_ttm_map_buffer(struct ttm_buffer_object *bo,
 	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4 + num_bytes,
-				     AMDGPU_IB_POOL_DELAYED, &job);
+				     AMDGPU_IB_POOL_DELAYED, &job,
+				     AMDGPU_KERNEL_JOB_ID_TTM_MAP_BUFFER);
 	if (r)
 		return r;
 
@@ -406,7 +407,7 @@ static int amdgpu_move_blit(struct ttm_buffer_object *bo,
 		struct dma_fence *wipe_fence = NULL;
 
 		r = amdgpu_fill_buffer(abo, 0, NULL, &wipe_fence,
-				       false);
+				       false, AMDGPU_KERNEL_JOB_ID_MOVE_BLIT);
 		if (r) {
 			goto error;
 		} else if (wipe_fence) {
@@ -1510,7 +1511,8 @@ static int amdgpu_ttm_access_memory_sdma(struct ttm_buffer_object *bo,
 	r = amdgpu_job_alloc_with_ib(adev, &adev->mman.high_pr,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4, AMDGPU_IB_POOL_DELAYED,
-				     &job);
+				     &job,
+				     AMDGPU_KERNEL_JOB_ID_TTM_ACCESS_MEMORY_SDMA);
 	if (r)
 		goto out;
 
@@ -2167,7 +2169,7 @@ static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev,
 				  struct dma_resv *resv,
 				  bool vm_needs_flush,
 				  struct amdgpu_job **job,
-				  bool delayed)
+				  bool delayed, u64 k_job_id)
 {
 	enum amdgpu_ib_pool_type pool = direct_submit ?
 		AMDGPU_IB_POOL_DIRECT :
@@ -2177,7 +2179,7 @@ static int amdgpu_ttm_prepare_job(struct amdgpu_device *adev,
 						    &adev->mman.high_pr;
 	r = amdgpu_job_alloc_with_ib(adev, entity,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
-				     num_dw * 4, pool, job);
+				     num_dw * 4, pool, job, k_job_id);
 	if (r)
 		return r;
 
@@ -2217,7 +2219,8 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
 	num_loops = DIV_ROUND_UP(byte_count, max_bytes);
 	num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->copy_num_dw, 8);
 	r = amdgpu_ttm_prepare_job(adev, direct_submit, num_dw,
-				   resv, vm_needs_flush, &job, false);
+				   resv, vm_needs_flush, &job, false,
+				   AMDGPU_KERNEL_JOB_ID_TTM_COPY_BUFFER);
 	if (r)
 		return r;
 
@@ -2252,7 +2255,8 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data,
 			       uint64_t dst_addr, uint32_t byte_count,
 			       struct dma_resv *resv,
 			       struct dma_fence **fence,
-			       bool vm_needs_flush, bool delayed)
+			       bool vm_needs_flush, bool delayed,
+			       u64 k_job_id)
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned int num_loops, num_dw;
@@ -2265,7 +2269,7 @@ static int amdgpu_ttm_fill_mem(struct amdgpu_ring *ring, uint32_t src_data,
 	num_loops = DIV_ROUND_UP_ULL(byte_count, max_bytes);
 	num_dw = ALIGN(num_loops * adev->mman.buffer_funcs->fill_num_dw, 8);
 	r = amdgpu_ttm_prepare_job(adev, false, num_dw, resv, vm_needs_flush,
-				   &job, delayed);
+				   &job, delayed, k_job_id);
 	if (r)
 		return r;
 
@@ -2335,7 +2339,8 @@ int amdgpu_ttm_clear_buffer(struct amdgpu_bo *bo,
 			goto err;
 
 		r = amdgpu_ttm_fill_mem(ring, 0, addr, size, resv,
-					&next, true, true);
+					&next, true, true,
+					AMDGPU_KERNEL_JOB_ID_TTM_CLEAR_BUFFER);
 		if (r)
 			goto err;
 
@@ -2354,7 +2359,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 			uint32_t src_data,
 			struct dma_resv *resv,
 			struct dma_fence **f,
-			bool delayed)
+			bool delayed,
+			u64 k_job_id)
 {
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
@@ -2384,7 +2390,7 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 			goto error;
 
 		r = amdgpu_ttm_fill_mem(ring, src_data, to, cur_size, resv,
-					&next, true, delayed);
+					&next, true, delayed, k_job_id);
 		if (r)
 			goto error;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 2309df3f68a9..d82d107fdcc6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -182,7 +182,8 @@ int amdgpu_fill_buffer(struct amdgpu_bo *bo,
 			uint32_t src_data,
 			struct dma_resv *resv,
 			struct dma_fence **fence,
-			bool delayed);
+			bool delayed,
+			u64 k_job_id);
 
 int amdgpu_ttm_alloc_gart(struct ttm_buffer_object *bo);
 void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index 74758b5ffc6c..5c38f0d30c87 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -1136,7 +1136,8 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo,
 	r = amdgpu_job_alloc_with_ib(ring->adev, &adev->uvd.entity,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     64, direct ? AMDGPU_IB_POOL_DIRECT :
-				     AMDGPU_IB_POOL_DELAYED, &job);
+				     AMDGPU_IB_POOL_DELAYED, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index b9060bcd4806..ce318f5de047 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -449,7 +449,7 @@ static int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 	r = amdgpu_job_alloc_with_ib(ring->adev, &ring->adev->vce.entity,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     ib_size_dw * 4, AMDGPU_IB_POOL_DIRECT,
-				     &job);
+				     &job, AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
@@ -540,7 +540,8 @@ static int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     ib_size_dw * 4,
 				     direct ? AMDGPU_IB_POOL_DIRECT :
-				     AMDGPU_IB_POOL_DELAYED, &job);
+				     AMDGPU_IB_POOL_DELAYED, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index f1f67521c29c..d287c44ddbc1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -601,7 +601,7 @@ static int amdgpu_vcn_dec_send_msg(struct amdgpu_ring *ring,
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
 				     64, AMDGPU_IB_POOL_DIRECT,
-				     &job);
+				     &job, AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		goto err;
 
@@ -781,7 +781,7 @@ static int amdgpu_vcn_dec_sw_send_msg(struct amdgpu_ring *ring,
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
 				     ib_size_dw * 4, AMDGPU_IB_POOL_DIRECT,
-				     &job);
+				     &job, AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		goto err;
 
@@ -911,7 +911,7 @@ static int amdgpu_vcn_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t hand
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
 				     ib_size_dw * 4, AMDGPU_IB_POOL_DIRECT,
-				     &job);
+				     &job, AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
@@ -978,7 +978,7 @@ static int amdgpu_vcn_enc_get_destroy_msg(struct amdgpu_ring *ring, uint32_t han
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL,
 				     ib_size_dw * 4, AMDGPU_IB_POOL_DIRECT,
-				     &job);
+				     &job, AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 0b87798daebd..314f11e6c78b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -977,7 +977,8 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 	params.vm = vm;
 	params.immediate = immediate;
 
-	r = vm->update_funcs->prepare(&params, NULL);
+	r = vm->update_funcs->prepare(&params, NULL,
+				      AMDGPU_KERNEL_JOB_ID_VM_UPDATE_PDES);
 	if (r)
 		goto error;
 
@@ -1146,7 +1147,8 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		dma_fence_put(tmp);
 	}
 
-	r = vm->update_funcs->prepare(&params, sync);
+	r = vm->update_funcs->prepare(&params, sync,
+				      AMDGPU_KERNEL_JOB_ID_VM_UPDATE_RANGE);
 	if (r)
 		goto error_free;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index fd086efd8457..ff2c8c6f4d7c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -308,7 +308,7 @@ struct amdgpu_vm_update_params {
 struct amdgpu_vm_update_funcs {
 	int (*map_table)(struct amdgpu_bo_vm *bo);
 	int (*prepare)(struct amdgpu_vm_update_params *p,
-		       struct amdgpu_sync *sync);
+		       struct amdgpu_sync *sync, u64 k_job_id);
 	int (*update)(struct amdgpu_vm_update_params *p,
 		      struct amdgpu_bo_vm *bo, uint64_t pe, uint64_t addr,
 		      unsigned count, uint32_t incr, uint64_t flags);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
index 0c1ef5850a5e..22e2e5b47341 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
@@ -40,12 +40,14 @@ static int amdgpu_vm_cpu_map_table(struct amdgpu_bo_vm *table)
  *
  * @p: see amdgpu_vm_update_params definition
  * @sync: sync obj with fences to wait on
+ * @k_job_id: the id for tracing/debug purposes
  *
  * Returns:
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p,
-				 struct amdgpu_sync *sync)
+				 struct amdgpu_sync *sync,
+				 u64 k_job_id)
 {
 	if (!sync)
 		return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index 30022123b0bf..f794fb1cc06e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -26,6 +26,7 @@
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
 #include "amdgpu_vm.h"
+#include "amdgpu_job.h"
 
 /*
  * amdgpu_vm_pt_cursor - state for for_each_amdgpu_vm_pt
@@ -395,7 +396,8 @@ int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	params.vm = vm;
 	params.immediate = immediate;
 
-	r = vm->update_funcs->prepare(&params, NULL);
+	r = vm->update_funcs->prepare(&params, NULL,
+				      AMDGPU_KERNEL_JOB_ID_VM_PT_CLEAR);
 	if (r)
 		goto exit;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
index 46d9fb433ab2..36805dcfa159 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@@ -40,7 +40,7 @@ static int amdgpu_vm_sdma_map_table(struct amdgpu_bo_vm *table)
 
 /* Allocate a new job for @count PTE updates */
 static int amdgpu_vm_sdma_alloc_job(struct amdgpu_vm_update_params *p,
-				    unsigned int count)
+				    unsigned int count, u64 k_job_id)
 {
 	enum amdgpu_ib_pool_type pool = p->immediate ? AMDGPU_IB_POOL_IMMEDIATE
 		: AMDGPU_IB_POOL_DELAYED;
@@ -56,7 +56,7 @@ static int amdgpu_vm_sdma_alloc_job(struct amdgpu_vm_update_params *p,
 	ndw = min(ndw, AMDGPU_VM_SDMA_MAX_NUM_DW);
 
 	r = amdgpu_job_alloc_with_ib(p->adev, entity, AMDGPU_FENCE_OWNER_VM,
-				     ndw * 4, pool, &p->job);
+				     ndw * 4, pool, &p->job, k_job_id);
 	if (r)
 		return r;
 
@@ -69,16 +69,17 @@ static int amdgpu_vm_sdma_alloc_job(struct amdgpu_vm_update_params *p,
  *
  * @p: see amdgpu_vm_update_params definition
  * @sync: amdgpu_sync object with fences to wait for
+ * @k_job_id: identifier of the job, for tracing purpose
  *
  * Returns:
  * Negativ errno, 0 for success.
  */
 static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
-				  struct amdgpu_sync *sync)
+				  struct amdgpu_sync *sync, u64 k_job_id)
 {
 	int r;
 
-	r = amdgpu_vm_sdma_alloc_job(p, 0);
+	r = amdgpu_vm_sdma_alloc_job(p, 0, k_job_id);
 	if (r)
 		return r;
 
@@ -249,7 +250,8 @@ static int amdgpu_vm_sdma_update(struct amdgpu_vm_update_params *p,
 			if (r)
 				return r;
 
-			r = amdgpu_vm_sdma_alloc_job(p, count);
+			r = amdgpu_vm_sdma_alloc_job(p, count,
+						     AMDGPU_KERNEL_JOB_ID_VM_UPDATE);
 			if (r)
 				return r;
 		}
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index 1c07b701d0e4..ceb94bbb03a4 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -217,7 +217,8 @@ static int uvd_v6_0_enc_get_create_msg(struct amdgpu_ring *ring, uint32_t handle
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL, ib_size_dw * 4,
-				     AMDGPU_IB_POOL_DIRECT, &job);
+				     AMDGPU_IB_POOL_DIRECT, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
@@ -281,7 +282,8 @@ static int uvd_v6_0_enc_get_destroy_msg(struct amdgpu_ring *ring,
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL, ib_size_dw * 4,
-				     AMDGPU_IB_POOL_DIRECT, &job);
+				     AMDGPU_IB_POOL_DIRECT, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index 9d237b5937fb..1f8866f3f63c 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -225,7 +225,8 @@ static int uvd_v7_0_enc_get_create_msg(struct amdgpu_ring *ring, u32 handle,
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL, ib_size_dw * 4,
-				     AMDGPU_IB_POOL_DIRECT, &job);
+				     AMDGPU_IB_POOL_DIRECT, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
@@ -288,7 +289,8 @@ static int uvd_v7_0_enc_get_destroy_msg(struct amdgpu_ring *ring, u32 handle,
 	int i, r;
 
 	r = amdgpu_job_alloc_with_ib(ring->adev, NULL, NULL, ib_size_dw * 4,
-				     AMDGPU_IB_POOL_DIRECT, &job);
+				     AMDGPU_IB_POOL_DIRECT, &job,
+				     AMDGPU_KERNEL_JOB_ID_VCN_RING_TEST);
 	if (r)
 		return r;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 79251f22b702..683ff02c45af 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -68,7 +68,8 @@ svm_migrate_gart_map(struct amdgpu_ring *ring, uint64_t npages,
 				     AMDGPU_FENCE_OWNER_UNDEFINED,
 				     num_dw * 4 + num_bytes,
 				     AMDGPU_IB_POOL_DELAYED,
-				     &job);
+				     &job,
+				     AMDGPU_KERNEL_JOB_ID_KFD_GART_MAP);
 	if (r)
 		return r;
 

From 73cfd166e045769a1b42d36897accaa6e06b8102 Mon Sep 17 00:00:00 2001
From: Ruben Wauters <rubenru09@aol.com>
Date: Mon, 18 Aug 2025 20:35:26 +0100
Subject: [PATCH 0793/1455] drm/gud: Replace simple display pipe with DRM
 atomic helpers

The simple display pipe is obsolete and the atomic helpers allow for
more control over the rendering process. As such, this patch replaces
the old simple display pipe system with the newer atomic helpers.

As the code is mainly the same, merely replaced with the new atomic
system, there should be no change in functionality.

Signed-off-by: Ruben Wauters <rubenru09@aol.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250818193553.2162-1-rubenru09@aol.com
---
 drivers/gpu/drm/gud/gud_connector.c | 25 +++++------
 drivers/gpu/drm/gud/gud_drv.c       | 52 ++++++++++++++++++-----
 drivers/gpu/drm/gud/gud_internal.h  | 13 +++---
 drivers/gpu/drm/gud/gud_pipe.c      | 64 +++++++++++++++++++----------
 4 files changed, 99 insertions(+), 55 deletions(-)

diff --git a/drivers/gpu/drm/gud/gud_connector.c b/drivers/gpu/drm/gud/gud_connector.c
index 0f07d77c5d52..4a15695fa933 100644
--- a/drivers/gpu/drm/gud/gud_connector.c
+++ b/drivers/gpu/drm/gud/gud_connector.c
@@ -16,7 +16,6 @@
 #include <drm/drm_modeset_helper_vtables.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
-#include <drm/drm_simple_kms_helper.h>
 #include <drm/gud.h>
 
 #include "gud_internal.h"
@@ -607,13 +606,16 @@ int gud_connector_fill_properties(struct drm_connector_state *connector_state,
 	return gconn->num_properties;
 }
 
+static const struct drm_encoder_funcs gud_drm_simple_encoder_funcs_cleanup = {
+	.destroy = drm_encoder_cleanup,
+};
+
 static int gud_connector_create(struct gud_device *gdrm, unsigned int index,
 				struct gud_connector_descriptor_req *desc)
 {
 	struct drm_device *drm = &gdrm->drm;
 	struct gud_connector *gconn;
 	struct drm_connector *connector;
-	struct drm_encoder *encoder;
 	int ret, connector_type;
 	u32 flags;
 
@@ -681,20 +683,13 @@ static int gud_connector_create(struct gud_device *gdrm, unsigned int index,
 		return ret;
 	}
 
-	/* The first connector is attached to the existing simple pipe encoder */
-	if (!connector->index) {
-		encoder = &gdrm->pipe.encoder;
-	} else {
-		encoder = &gconn->encoder;
+	gconn->encoder.possible_crtcs = drm_crtc_mask(&gdrm->crtc);
+	ret = drm_encoder_init(drm, &gconn->encoder, &gud_drm_simple_encoder_funcs_cleanup,
+			       DRM_MODE_ENCODER_NONE, NULL);
+	if (ret)
+		return ret;
 
-		ret = drm_simple_encoder_init(drm, encoder, DRM_MODE_ENCODER_NONE);
-		if (ret)
-			return ret;
-
-		encoder->possible_crtcs = 1;
-	}
-
-	return drm_connector_attach_encoder(connector, encoder);
+	return drm_connector_attach_encoder(connector, &gconn->encoder);
 }
 
 int gud_get_connectors(struct gud_device *gdrm)
diff --git a/drivers/gpu/drm/gud/gud_drv.c b/drivers/gpu/drm/gud/gud_drv.c
index b52a12cbba3e..b7345c8d823d 100644
--- a/drivers/gpu/drm/gud/gud_drv.c
+++ b/drivers/gpu/drm/gud/gud_drv.c
@@ -16,6 +16,7 @@
 #include <drm/clients/drm_client_setup.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_blend.h>
+#include <drm/drm_crtc_helper.h>
 #include <drm/drm_damage_helper.h>
 #include <drm/drm_debugfs.h>
 #include <drm/drm_drv.h>
@@ -27,7 +28,6 @@
 #include <drm/drm_managed.h>
 #include <drm/drm_print.h>
 #include <drm/drm_probe_helper.h>
-#include <drm/drm_simple_kms_helper.h>
 #include <drm/gud.h>
 
 #include "gud_internal.h"
@@ -289,7 +289,7 @@ static int gud_get_properties(struct gud_device *gdrm)
 			 * but mask out any additions on future devices.
 			 */
 			val &= GUD_ROTATION_MASK;
-			ret = drm_plane_create_rotation_property(&gdrm->pipe.plane,
+			ret = drm_plane_create_rotation_property(&gdrm->plane,
 								 DRM_MODE_ROTATE_0, val);
 			break;
 		default:
@@ -338,10 +338,30 @@ static int gud_stats_debugfs(struct seq_file *m, void *data)
 	return 0;
 }
 
-static const struct drm_simple_display_pipe_funcs gud_pipe_funcs = {
-	.check      = gud_pipe_check,
-	.update	    = gud_pipe_update,
-	DRM_GEM_SIMPLE_DISPLAY_PIPE_SHADOW_PLANE_FUNCS
+static const struct drm_crtc_helper_funcs gud_crtc_helper_funcs = {
+	.atomic_check = drm_crtc_helper_atomic_check
+};
+
+static const struct drm_crtc_funcs gud_crtc_funcs = {
+	.reset = drm_atomic_helper_crtc_reset,
+	.destroy = drm_crtc_cleanup,
+	.set_config = drm_atomic_helper_set_config,
+	.page_flip = drm_atomic_helper_page_flip,
+	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+};
+
+static const struct drm_plane_helper_funcs gud_plane_helper_funcs = {
+	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
+	.atomic_check = gud_plane_atomic_check,
+	.atomic_update = gud_plane_atomic_update,
+};
+
+static const struct drm_plane_funcs gud_plane_funcs = {
+	.update_plane = drm_atomic_helper_update_plane,
+	.disable_plane = drm_atomic_helper_disable_plane,
+	.destroy = drm_plane_cleanup,
+	DRM_GEM_SHADOW_PLANE_FUNCS,
 };
 
 static const struct drm_mode_config_funcs gud_mode_config_funcs = {
@@ -350,7 +370,7 @@ static const struct drm_mode_config_funcs gud_mode_config_funcs = {
 	.atomic_commit = drm_atomic_helper_commit,
 };
 
-static const u64 gud_pipe_modifiers[] = {
+static const u64 gud_plane_modifiers[] = {
 	DRM_FORMAT_MOD_LINEAR,
 	DRM_FORMAT_MOD_INVALID
 };
@@ -567,12 +587,17 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id)
 			return -ENOMEM;
 	}
 
-	ret = drm_simple_display_pipe_init(drm, &gdrm->pipe, &gud_pipe_funcs,
-					   formats, num_formats,
-					   gud_pipe_modifiers, NULL);
+	ret = drm_universal_plane_init(drm, &gdrm->plane, 0,
+				       &gud_plane_funcs,
+				       formats, num_formats,
+				       gud_plane_modifiers,
+				       DRM_PLANE_TYPE_PRIMARY, NULL);
 	if (ret)
 		return ret;
 
+	drm_plane_helper_add(&gdrm->plane, &gud_plane_helper_funcs);
+	drm_plane_enable_fb_damage_clips(&gdrm->plane);
+
 	devm_kfree(dev, formats);
 	devm_kfree(dev, formats_dev);
 
@@ -582,7 +607,12 @@ static int gud_probe(struct usb_interface *intf, const struct usb_device_id *id)
 		return ret;
 	}
 
-	drm_plane_enable_fb_damage_clips(&gdrm->pipe.plane);
+	ret = drm_crtc_init_with_planes(drm, &gdrm->crtc, &gdrm->plane, NULL,
+					&gud_crtc_funcs, NULL);
+	if (ret)
+		return ret;
+
+	drm_crtc_helper_add(&gdrm->crtc, &gud_crtc_helper_funcs);
 
 	ret = gud_get_connectors(gdrm);
 	if (ret) {
diff --git a/drivers/gpu/drm/gud/gud_internal.h b/drivers/gpu/drm/gud/gud_internal.h
index d6fb25388722..d27c31648341 100644
--- a/drivers/gpu/drm/gud/gud_internal.h
+++ b/drivers/gpu/drm/gud/gud_internal.h
@@ -11,11 +11,11 @@
 #include <uapi/drm/drm_fourcc.h>
 
 #include <drm/drm_modes.h>
-#include <drm/drm_simple_kms_helper.h>
 
 struct gud_device {
 	struct drm_device drm;
-	struct drm_simple_display_pipe pipe;
+	struct drm_plane plane;
+	struct drm_crtc crtc;
 	struct work_struct work;
 	u32 flags;
 	const struct drm_format_info *xrgb8888_emulation_format;
@@ -62,11 +62,10 @@ int gud_usb_set_u8(struct gud_device *gdrm, u8 request, u8 val);
 
 void gud_clear_damage(struct gud_device *gdrm);
 void gud_flush_work(struct work_struct *work);
-int gud_pipe_check(struct drm_simple_display_pipe *pipe,
-		   struct drm_plane_state *new_plane_state,
-		   struct drm_crtc_state *new_crtc_state);
-void gud_pipe_update(struct drm_simple_display_pipe *pipe,
-		     struct drm_plane_state *old_state);
+int gud_plane_atomic_check(struct drm_plane *plane,
+			   struct drm_atomic_state *state);
+void gud_plane_atomic_update(struct drm_plane *plane,
+			     struct drm_atomic_state *atomic_state);
 int gud_connector_fill_properties(struct drm_connector_state *connector_state,
 				  struct gud_property_req *properties);
 int gud_get_connectors(struct gud_device *gdrm);
diff --git a/drivers/gpu/drm/gud/gud_pipe.c b/drivers/gpu/drm/gud/gud_pipe.c
index 8d548d08f127..54d9aa9998e5 100644
--- a/drivers/gpu/drm/gud/gud_pipe.c
+++ b/drivers/gpu/drm/gud/gud_pipe.c
@@ -20,7 +20,6 @@
 #include <drm/drm_gem_framebuffer_helper.h>
 #include <drm/drm_print.h>
 #include <drm/drm_rect.h>
-#include <drm/drm_simple_kms_helper.h>
 #include <drm/gud.h>
 
 #include "gud_internal.h"
@@ -451,14 +450,15 @@ static void gud_fb_handle_damage(struct gud_device *gdrm, struct drm_framebuffer
 	gud_flush_damage(gdrm, fb, src, !fb->obj[0]->import_attach, damage);
 }
 
-int gud_pipe_check(struct drm_simple_display_pipe *pipe,
-		   struct drm_plane_state *new_plane_state,
-		   struct drm_crtc_state *new_crtc_state)
+int gud_plane_atomic_check(struct drm_plane *plane,
+			   struct drm_atomic_state *state)
 {
-	struct gud_device *gdrm = to_gud_device(pipe->crtc.dev);
-	struct drm_plane_state *old_plane_state = pipe->plane.state;
-	const struct drm_display_mode *mode = &new_crtc_state->mode;
-	struct drm_atomic_state *state = new_plane_state->state;
+	struct gud_device *gdrm = to_gud_device(plane->dev);
+	struct drm_plane_state *old_plane_state = drm_atomic_get_old_plane_state(state, plane);
+	struct drm_plane_state *new_plane_state = drm_atomic_get_new_plane_state(state, plane);
+	struct drm_crtc *crtc = new_plane_state->crtc;
+	struct drm_crtc_state *crtc_state;
+	const struct drm_display_mode *mode;
 	struct drm_framebuffer *old_fb = old_plane_state->fb;
 	struct drm_connector_state *connector_state = NULL;
 	struct drm_framebuffer *fb = new_plane_state->fb;
@@ -469,20 +469,37 @@ int gud_pipe_check(struct drm_simple_display_pipe *pipe,
 	int idx, ret;
 	size_t len;
 
-	if (WARN_ON_ONCE(!fb))
+	if (drm_WARN_ON_ONCE(plane->dev, !fb))
 		return -EINVAL;
 
+	if (drm_WARN_ON_ONCE(plane->dev, !crtc))
+		return -EINVAL;
+
+	crtc_state = drm_atomic_get_new_crtc_state(state, crtc);
+
+	mode = &crtc_state->mode;
+
+	ret = drm_atomic_helper_check_plane_state(new_plane_state, crtc_state,
+						  DRM_PLANE_NO_SCALING,
+						  DRM_PLANE_NO_SCALING,
+						  false, false);
+	if (ret)
+		return ret;
+
+	if (!new_plane_state->visible)
+		return 0;
+
 	if (old_plane_state->rotation != new_plane_state->rotation)
-		new_crtc_state->mode_changed = true;
+		crtc_state->mode_changed = true;
 
 	if (old_fb && old_fb->format != format)
-		new_crtc_state->mode_changed = true;
+		crtc_state->mode_changed = true;
 
-	if (!new_crtc_state->mode_changed && !new_crtc_state->connectors_changed)
+	if (!crtc_state->mode_changed && !crtc_state->connectors_changed)
 		return 0;
 
 	/* Only one connector is supported */
-	if (hweight32(new_crtc_state->connector_mask) != 1)
+	if (hweight32(crtc_state->connector_mask) != 1)
 		return -EINVAL;
 
 	if (format->format == DRM_FORMAT_XRGB8888 && gdrm->xrgb8888_emulation_format)
@@ -500,7 +517,7 @@ int gud_pipe_check(struct drm_simple_display_pipe *pipe,
 	if (!connector_state) {
 		struct drm_connector_list_iter conn_iter;
 
-		drm_connector_list_iter_begin(pipe->crtc.dev, &conn_iter);
+		drm_connector_list_iter_begin(plane->dev, &conn_iter);
 		drm_for_each_connector_iter(connector, &conn_iter) {
 			if (connector->state->crtc) {
 				connector_state = connector->state;
@@ -567,16 +584,18 @@ out:
 	return ret;
 }
 
-void gud_pipe_update(struct drm_simple_display_pipe *pipe,
-		     struct drm_plane_state *old_state)
+void gud_plane_atomic_update(struct drm_plane *plane,
+			     struct drm_atomic_state *atomic_state)
 {
-	struct drm_device *drm = pipe->crtc.dev;
+	struct drm_device *drm = plane->dev;
 	struct gud_device *gdrm = to_gud_device(drm);
-	struct drm_plane_state *state = pipe->plane.state;
-	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(state);
-	struct drm_framebuffer *fb = state->fb;
-	struct drm_crtc *crtc = &pipe->crtc;
+	struct drm_plane_state *old_state = drm_atomic_get_old_plane_state(atomic_state, plane);
+	struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(atomic_state, plane);
+	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(new_state);
+	struct drm_framebuffer *fb = new_state->fb;
+	struct drm_crtc *crtc = new_state->crtc;
 	struct drm_rect damage;
+	struct drm_atomic_helper_damage_iter iter;
 	int ret, idx;
 
 	if (crtc->state->mode_changed || !crtc->state->enable) {
@@ -611,7 +630,8 @@ void gud_pipe_update(struct drm_simple_display_pipe *pipe,
 	if (ret)
 		goto ctrl_disable;
 
-	if (drm_atomic_helper_damage_merged(old_state, state, &damage))
+	drm_atomic_helper_damage_iter_init(&iter, old_state, new_state);
+	drm_atomic_for_each_plane_damage(&iter, &damage)
 		gud_fb_handle_damage(gdrm, fb, &shadow_plane_state->data[0], &damage);
 
 	drm_gem_fb_end_cpu_access(fb, DMA_FROM_DEVICE);

From d42ce9ea571e528645fcab48a8ef7280331f5872 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:32 +0200
Subject: [PATCH 0794/1455] drm/tidss: dispc: Remove unused OVR_REG_GET

The OVR_REG_GET function in the dispc driver is not used anywhere. Let's
drop it.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-1-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 3f6cff2ab1b2..71031c3ad60e 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -653,13 +653,6 @@ static void VP_REG_FLD_MOD(struct dispc_device *dispc, u32 vp, u32 idx, u32 val,
 					       val, start, end));
 }
 
-__maybe_unused
-static u32 OVR_REG_GET(struct dispc_device *dispc, u32 ovr, u32 idx,
-		       u32 start, u32 end)
-{
-	return FLD_GET(dispc_ovr_read(dispc, ovr, idx), start, end);
-}
-
 static void OVR_REG_FLD_MOD(struct dispc_device *dispc, u32 ovr, u32 idx,
 			    u32 val, u32 start, u32 end)
 {

From e8e202bcbd06b9a199e431102af1a8ea2088f847 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:33 +0200
Subject: [PATCH 0795/1455] drm/tidss: dispc: Convert accessors to macros

The dispc driver uses upper-cased, inlined, functions to provide
macro-like accessors to the dispc registers.

This is confusing, since upper-case is usually used by macros, and that
pattern will create gcc errors later on in this series.

Let's switch to macros to make it more consistent, and prevent those
errors down the line.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-2-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 122 +++++++++++++++-------------
 1 file changed, 66 insertions(+), 56 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 71031c3ad60e..6231e1aa07fd 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -594,72 +594,82 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
  * number. For example 7:0
  */
 
-static u32 FLD_MASK(u32 start, u32 end)
-{
-	return ((1 << (start - end + 1)) - 1) << end;
-}
+#define FLD_MASK(start, end)						\
+	({								\
+		int _end_inner = (end);					\
+		u32 _mask = ((1 << ((start) - _end_inner + 1)) - 1) << _end_inner; \
+		_mask;							\
+	})
 
-static u32 FLD_VAL(u32 val, u32 start, u32 end)
-{
-	return (val << end) & FLD_MASK(start, end);
-}
+#define FLD_VAL(val, start, end)					\
+	({								\
+		int _end_inner = (end);					\
+		u32 _new_val = ((val) << _end_inner) & FLD_MASK((start), _end_inner); \
+		_new_val;						\
+	})
 
-static u32 FLD_GET(u32 val, u32 start, u32 end)
-{
-	return (val & FLD_MASK(start, end)) >> end;
-}
+#define FLD_GET(val, start, end)					\
+	({								\
+		int _end = (end);					\
+		u32 _ret_val = ((val) & FLD_MASK((start), _end)) >> _end; \
+		_ret_val;						\
+	})
 
-static u32 FLD_MOD(u32 orig, u32 val, u32 start, u32 end)
-{
-	return (orig & ~FLD_MASK(start, end)) | FLD_VAL(val, start, end);
-}
+#define FLD_MOD(orig, val, start, end)					\
+	({								\
+		int _start = (start), _end = (end);			\
+		u32 _masked_val = (orig) & ~FLD_MASK(_start, _end);	\
+		u32 _new_val = _masked_val | FLD_VAL((val), _start, _end); \
+		_new_val;						\
+	})
 
-static u32 REG_GET(struct dispc_device *dispc, u32 idx, u32 start, u32 end)
-{
-	return FLD_GET(dispc_read(dispc, idx), start, end);
-}
+#define REG_GET(dispc, idx, start, end)					\
+	((u32)FLD_GET(dispc_read((dispc), (idx)), (start), (end)))
 
-static void REG_FLD_MOD(struct dispc_device *dispc, u32 idx, u32 val,
-			u32 start, u32 end)
-{
-	dispc_write(dispc, idx, FLD_MOD(dispc_read(dispc, idx), val,
-					start, end));
-}
+#define REG_FLD_MOD(dispc, idx, val, start, end)			\
+	({								\
+		struct dispc_device *_dispc = (dispc);			\
+		u32 _idx = (idx);					\
+		u32 _curr = dispc_read(_dispc, _idx);			\
+		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
+		dispc_write(_dispc, _idx, _new);			\
+	})
 
-static u32 VID_REG_GET(struct dispc_device *dispc, u32 hw_plane, u32 idx,
-		       u32 start, u32 end)
-{
-	return FLD_GET(dispc_vid_read(dispc, hw_plane, idx), start, end);
-}
+#define VID_REG_GET(dispc, hw_plane, idx, start, end)			\
+	((u32)FLD_GET(dispc_vid_read((dispc), (hw_plane), (idx)), (start), (end)))
 
-static void VID_REG_FLD_MOD(struct dispc_device *dispc, u32 hw_plane, u32 idx,
-			    u32 val, u32 start, u32 end)
-{
-	dispc_vid_write(dispc, hw_plane, idx,
-			FLD_MOD(dispc_vid_read(dispc, hw_plane, idx),
-				val, start, end));
-}
+#define VID_REG_FLD_MOD(dispc, hw_plane, idx, val, start, end)		\
+	({								\
+		struct dispc_device *_dispc = (dispc);			\
+		u32 _hw_plane = (hw_plane);				\
+		u32 _idx = (idx);					\
+		u32 _curr = dispc_vid_read(_dispc, _hw_plane, _idx);	\
+		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
+		dispc_vid_write(_dispc, _hw_plane, _idx, _new);		\
+	})
 
-static u32 VP_REG_GET(struct dispc_device *dispc, u32 vp, u32 idx,
-		      u32 start, u32 end)
-{
-	return FLD_GET(dispc_vp_read(dispc, vp, idx), start, end);
-}
+#define VP_REG_GET(dispc, vp, idx, start, end)				\
+	((u32)FLD_GET(dispc_vp_read((dispc), (vp), (idx)), (start), (end)))
 
-static void VP_REG_FLD_MOD(struct dispc_device *dispc, u32 vp, u32 idx, u32 val,
-			   u32 start, u32 end)
-{
-	dispc_vp_write(dispc, vp, idx, FLD_MOD(dispc_vp_read(dispc, vp, idx),
-					       val, start, end));
-}
+#define VP_REG_FLD_MOD(dispc, vp, idx, val, start, end)			\
+	({								\
+		struct dispc_device *_dispc = (dispc);			\
+		u32 _vp = (vp);						\
+		u32 _idx = (idx);					\
+		u32 _curr = dispc_vp_read(_dispc, _vp, _idx);		\
+		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
+		dispc_vp_write(_dispc, _vp, _idx, _new);		\
+	})
 
-static void OVR_REG_FLD_MOD(struct dispc_device *dispc, u32 ovr, u32 idx,
-			    u32 val, u32 start, u32 end)
-{
-	dispc_ovr_write(dispc, ovr, idx,
-			FLD_MOD(dispc_ovr_read(dispc, ovr, idx),
-				val, start, end));
-}
+#define OVR_REG_FLD_MOD(dispc, ovr, idx, val, start, end)		\
+	({								\
+		struct dispc_device *_dispc = (dispc);			\
+		u32 _ovr = (ovr);						\
+		u32 _idx = (idx);					\
+		u32 _curr = dispc_ovr_read(_dispc, _ovr, _idx);		\
+		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
+		dispc_ovr_write(_dispc, _ovr, _idx, _new);		\
+	})
 
 static dispc_irq_t dispc_vp_irq_from_raw(u32 stat, u32 hw_videoport)
 {

From db7c13ebf2d4570efcb8bed0d6e772d3805bbb45 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:34 +0200
Subject: [PATCH 0796/1455] drm/tidss: dispc: Switch to GENMASK instead of
 FLD_MASK

The dispc FLD_MASK function is an exact equivalent of the GENMASK macro.
Let's convert the dispc driver to the latter.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-3-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 6231e1aa07fd..b5f09ed2f370 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -594,31 +594,24 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
  * number. For example 7:0
  */
 
-#define FLD_MASK(start, end)						\
-	({								\
-		int _end_inner = (end);					\
-		u32 _mask = ((1 << ((start) - _end_inner + 1)) - 1) << _end_inner; \
-		_mask;							\
-	})
-
 #define FLD_VAL(val, start, end)					\
 	({								\
 		int _end_inner = (end);					\
-		u32 _new_val = ((val) << _end_inner) & FLD_MASK((start), _end_inner); \
+		u32 _new_val = ((val) << _end_inner) & GENMASK((start), _end_inner); \
 		_new_val;						\
 	})
 
 #define FLD_GET(val, start, end)					\
 	({								\
 		int _end = (end);					\
-		u32 _ret_val = ((val) & FLD_MASK((start), _end)) >> _end; \
+		u32 _ret_val = ((val) & GENMASK((start), _end)) >> _end; \
 		_ret_val;						\
 	})
 
 #define FLD_MOD(orig, val, start, end)					\
 	({								\
 		int _start = (start), _end = (end);			\
-		u32 _masked_val = (orig) & ~FLD_MASK(_start, _end);	\
+		u32 _masked_val = (orig) & ~GENMASK(_start, _end);	\
 		u32 _new_val = _masked_val | FLD_VAL((val), _start, _end); \
 		_new_val;						\
 	})

From 9accc8b10de87a8989d382231d417827aca61c8f Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:35 +0200
Subject: [PATCH 0797/1455] drm/tidss: dispc: Get rid of FLD_VAL

The FLD_VAL function is an equivalent to what FIELD_PREP + GENMASK
provide, so let's drop it and switch to the latter.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-4-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 58 +++++++++++++----------------
 1 file changed, 26 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index b5f09ed2f370..43f8078a1eff 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -594,13 +594,6 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
  * number. For example 7:0
  */
 
-#define FLD_VAL(val, start, end)					\
-	({								\
-		int _end_inner = (end);					\
-		u32 _new_val = ((val) << _end_inner) & GENMASK((start), _end_inner); \
-		_new_val;						\
-	})
-
 #define FLD_GET(val, start, end)					\
 	({								\
 		int _end = (end);					\
@@ -612,7 +605,7 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 	({								\
 		int _start = (start), _end = (end);			\
 		u32 _masked_val = (orig) & ~GENMASK(_start, _end);	\
-		u32 _new_val = _masked_val | FLD_VAL((val), _start, _end); \
+		u32 _new_val = _masked_val | FIELD_PREP(GENMASK(_start, _end), (val)); \
 		_new_val;						\
 	})
 
@@ -1220,14 +1213,14 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 	vbp = mode->crtc_vtotal - mode->crtc_vsync_end;
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_TIMING_H,
-		       FLD_VAL(hsw - 1, 7, 0) |
-		       FLD_VAL(hfp - 1, 19, 8) |
-		       FLD_VAL(hbp - 1, 31, 20));
+		       FIELD_PREP(GENMASK(7, 0), hsw - 1) |
+		       FIELD_PREP(GENMASK(19, 8), hfp - 1) |
+		       FIELD_PREP(GENMASK(31, 20), hbp - 1));
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_TIMING_V,
-		       FLD_VAL(vsw - 1, 7, 0) |
-		       FLD_VAL(vfp, 19, 8) |
-		       FLD_VAL(vbp, 31, 20));
+		       FIELD_PREP(GENMASK(7, 0), vsw - 1) |
+		       FIELD_PREP(GENMASK(19, 8), vfp) |
+		       FIELD_PREP(GENMASK(31, 20), vbp));
 
 	ivs = !!(mode->flags & DRM_MODE_FLAG_NVSYNC);
 
@@ -1250,17 +1243,17 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 		ieo = false;
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_POL_FREQ,
-		       FLD_VAL(align, 18, 18) |
-		       FLD_VAL(onoff, 17, 17) |
-		       FLD_VAL(rf, 16, 16) |
-		       FLD_VAL(ieo, 15, 15) |
-		       FLD_VAL(ipc, 14, 14) |
-		       FLD_VAL(ihs, 13, 13) |
-		       FLD_VAL(ivs, 12, 12));
+		       FIELD_PREP(GENMASK(18, 18), align) |
+		       FIELD_PREP(GENMASK(17, 17), onoff) |
+		       FIELD_PREP(GENMASK(16, 16), rf) |
+		       FIELD_PREP(GENMASK(15, 15), ieo) |
+		       FIELD_PREP(GENMASK(14, 14), ipc) |
+		       FIELD_PREP(GENMASK(13, 13), ihs) |
+		       FIELD_PREP(GENMASK(12, 12), ivs));
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_SIZE_SCREEN,
-		       FLD_VAL(mode->crtc_hdisplay - 1, 11, 0) |
-		       FLD_VAL(mode->crtc_vdisplay - 1, 27, 16));
+		       FIELD_PREP(GENMASK(11, 0), mode->crtc_hdisplay - 1) |
+		       FIELD_PREP(GENMASK(27, 16), mode->crtc_vdisplay - 1));
 
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1, 0, 0);
 }
@@ -1576,14 +1569,14 @@ struct dispc_csc_coef {
 static
 void dispc_csc_offset_regval(const struct dispc_csc_coef *csc, u32 *regval)
 {
-#define OVAL(x, y) (FLD_VAL(x, 15, 3) | FLD_VAL(y, 31, 19))
+#define OVAL(x, y) (FIELD_PREP(GENMASK(15, 3), x) | FIELD_PREP(GENMASK(31, 19), y))
 	regval[5] = OVAL(csc->preoffset[0], csc->preoffset[1]);
 	regval[6] = OVAL(csc->preoffset[2], csc->postoffset[0]);
 	regval[7] = OVAL(csc->postoffset[1], csc->postoffset[2]);
 #undef OVAL
 }
 
-#define CVAL(x, y) (FLD_VAL(x, 10, 0) | FLD_VAL(y, 26, 16))
+#define CVAL(x, y) (FIELD_PREP(GENMASK(10, 0), x) | FIELD_PREP(GENMASK(26, 16), y))
 static
 void dispc_csc_yuv2rgb_regval(const struct dispc_csc_coef *csc, u32 *regval)
 {
@@ -1822,7 +1815,8 @@ static void dispc_vid_write_fir_coefs(struct dispc_device *dispc,
 
 		c1 = coefs->c1[phase];
 		c2 = coefs->c2[phase];
-		c12 = FLD_VAL(c1, 19, 10) | FLD_VAL(c2, 29, 20);
+		c12 = FIELD_PREP(GENMASK(19, 10), c1) | FIELD_PREP(GENMASK(29, 20),
+								   c2);
 
 		dispc_vid_write(dispc, hw_plane, reg, c12);
 	}
@@ -2320,14 +2314,14 @@ static void dispc_vid_set_mflag_threshold(struct dispc_device *dispc,
 					  u32 hw_plane, u32 low, u32 high)
 {
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_MFLAG_THRESHOLD,
-			FLD_VAL(high, 31, 16) | FLD_VAL(low, 15, 0));
+			FIELD_PREP(GENMASK(31, 16), high) | FIELD_PREP(GENMASK(15, 0), low));
 }
 
 static void dispc_vid_set_buf_threshold(struct dispc_device *dispc,
 					u32 hw_plane, u32 low, u32 high)
 {
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_BUF_THRESHOLD,
-			FLD_VAL(high, 31, 16) | FLD_VAL(low, 15, 0));
+			FIELD_PREP(GENMASK(31, 16), high) | FIELD_PREP(GENMASK(15, 0), low));
 }
 
 static void dispc_k2g_plane_init(struct dispc_device *dispc)
@@ -2468,8 +2462,8 @@ static void dispc_initial_config(struct dispc_device *dispc)
 	/* Note: Hardcoded DPI routing on J721E for now */
 	if (dispc->feat->subrev == DISPC_J721E) {
 		dispc_write(dispc, DISPC_CONNECTIONS,
-			    FLD_VAL(2, 3, 0) |		/* VP1 to DPI0 */
-			    FLD_VAL(8, 7, 4)		/* VP3 to DPI1 */
+			    FIELD_PREP(GENMASK(3, 0), 2) |		/* VP1 to DPI0 */
+			    FIELD_PREP(GENMASK(7, 4), 8)		/* VP3 to DPI1 */
 			);
 	}
 }
@@ -2647,8 +2641,8 @@ static void dispc_k2g_cpr_from_ctm(const struct drm_color_ctm *ctm,
 	cpr->m[CSC_BB] = dispc_S31_32_to_s2_8(ctm->matrix[8]);
 }
 
-#define CVAL(xR, xG, xB) (FLD_VAL(xR, 9, 0) | FLD_VAL(xG, 20, 11) |	\
-			  FLD_VAL(xB, 31, 22))
+#define CVAL(xR, xG, xB) (FIELD_PREP(GENMASK(9, 0), xR) | FIELD_PREP(GENMASK(20, 11), xG) |	\
+			  FIELD_PREP(GENMASK(31, 22), xB))
 
 static void dispc_k2g_vp_csc_cpr_regval(const struct dispc_csc_coef *csc,
 					u32 *regval)

From 6e50765b829b68d34cbf45415173b1b1f429b233 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:36 +0200
Subject: [PATCH 0798/1455] drm/tidss: dispc: Get rid of FLD_GET

The FLD_GET function is an equivalent to what FIELD_GET + GENMASK
provide, so let's drop it and switch to the latter.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-5-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 43f8078a1eff..b4ba342ac241 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -594,13 +594,6 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
  * number. For example 7:0
  */
 
-#define FLD_GET(val, start, end)					\
-	({								\
-		int _end = (end);					\
-		u32 _ret_val = ((val) & GENMASK((start), _end)) >> _end; \
-		_ret_val;						\
-	})
-
 #define FLD_MOD(orig, val, start, end)					\
 	({								\
 		int _start = (start), _end = (end);			\
@@ -610,7 +603,8 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 	})
 
 #define REG_GET(dispc, idx, start, end)					\
-	((u32)FLD_GET(dispc_read((dispc), (idx)), (start), (end)))
+	((u32)FIELD_GET(GENMASK((start), (end)),			\
+			dispc_read((dispc), (idx))))
 
 #define REG_FLD_MOD(dispc, idx, val, start, end)			\
 	({								\
@@ -622,7 +616,8 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 	})
 
 #define VID_REG_GET(dispc, hw_plane, idx, start, end)			\
-	((u32)FLD_GET(dispc_vid_read((dispc), (hw_plane), (idx)), (start), (end)))
+	((u32)FIELD_GET(GENMASK((start), (end)),			\
+			dispc_vid_read((dispc), (hw_plane), (idx))))
 
 #define VID_REG_FLD_MOD(dispc, hw_plane, idx, val, start, end)		\
 	({								\
@@ -635,7 +630,8 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 	})
 
 #define VP_REG_GET(dispc, vp, idx, start, end)				\
-	((u32)FLD_GET(dispc_vp_read((dispc), (vp), (idx)), (start), (end)))
+	((u32)FIELD_GET(GENMASK((start), (end)),			\
+			dispc_vp_read((dispc), (vp), (idx))))
 
 #define VP_REG_FLD_MOD(dispc, vp, idx, val, start, end)			\
 	({								\

From 13925ccb0d6de6e8a75928ea4ac74610c629093a Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:37 +0200
Subject: [PATCH 0799/1455] drm/tidss: dispc: Get rid of FLD_MOD

The FLD_MOD function is an equivalent to what FIELD_MODIFY + GENMASK
provide, so let's drop it and switch to the latter.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-6-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 36 +++++++++++------------------
 1 file changed, 14 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index b4ba342ac241..4eeeae6e0a0e 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -594,14 +594,6 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
  * number. For example 7:0
  */
 
-#define FLD_MOD(orig, val, start, end)					\
-	({								\
-		int _start = (start), _end = (end);			\
-		u32 _masked_val = (orig) & ~GENMASK(_start, _end);	\
-		u32 _new_val = _masked_val | FIELD_PREP(GENMASK(_start, _end), (val)); \
-		_new_val;						\
-	})
-
 #define REG_GET(dispc, idx, start, end)					\
 	((u32)FIELD_GET(GENMASK((start), (end)),			\
 			dispc_read((dispc), (idx))))
@@ -610,9 +602,9 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 	({								\
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _idx = (idx);					\
-		u32 _curr = dispc_read(_dispc, _idx);			\
-		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
-		dispc_write(_dispc, _idx, _new);			\
+		u32 _reg = dispc_read(_dispc, _idx);			\
+		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		dispc_write(_dispc, _idx, _reg);			\
 	})
 
 #define VID_REG_GET(dispc, hw_plane, idx, start, end)			\
@@ -624,9 +616,9 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _hw_plane = (hw_plane);				\
 		u32 _idx = (idx);					\
-		u32 _curr = dispc_vid_read(_dispc, _hw_plane, _idx);	\
-		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
-		dispc_vid_write(_dispc, _hw_plane, _idx, _new);		\
+		u32 _reg = dispc_vid_read(_dispc, _hw_plane, _idx);	\
+		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		dispc_vid_write(_dispc, _hw_plane, _idx, _reg);		\
 	})
 
 #define VP_REG_GET(dispc, vp, idx, start, end)				\
@@ -638,19 +630,19 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _vp = (vp);						\
 		u32 _idx = (idx);					\
-		u32 _curr = dispc_vp_read(_dispc, _vp, _idx);		\
-		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
-		dispc_vp_write(_dispc, _vp, _idx, _new);		\
+		u32 _reg = dispc_vp_read(_dispc, _vp, _idx);		\
+		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		dispc_vp_write(_dispc, _vp, _idx, _reg);		\
 	})
 
 #define OVR_REG_FLD_MOD(dispc, ovr, idx, val, start, end)		\
 	({								\
 		struct dispc_device *_dispc = (dispc);			\
-		u32 _ovr = (ovr);						\
+		u32 _ovr = (ovr);					\
 		u32 _idx = (idx);					\
-		u32 _curr = dispc_ovr_read(_dispc, _ovr, _idx);		\
-		u32 _new = FLD_MOD(_curr, (val), (start), (end));	\
-		dispc_ovr_write(_dispc, _ovr, _idx, _new);		\
+		u32 _reg = dispc_ovr_read(_dispc, _ovr, _idx);		\
+		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		dispc_ovr_write(_dispc, _ovr, _idx, _reg);		\
 	})
 
 static dispc_irq_t dispc_vp_irq_from_raw(u32 stat, u32 hw_videoport)
@@ -1147,7 +1139,7 @@ static void dispc_enable_am65x_oldi(struct dispc_device *dispc, u32 hw_videoport
 
 	oldi_cfg |= BIT(7); /* DEPOL */
 
-	oldi_cfg = FLD_MOD(oldi_cfg, fmt->am65x_oldi_mode_reg_val, 3, 1);
+	FIELD_MODIFY(GENMASK(3, 1), &oldi_cfg, fmt->am65x_oldi_mode_reg_val);
 
 	oldi_cfg |= BIT(12); /* SOFTRST */
 

From 8bd839a3ec2305a2a29706dcbd27632b6d63d9ac Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:38 +0200
Subject: [PATCH 0800/1455] drm/tidss: dispc: Switch REG_GET to using a mask

The REG_GET function takes the start and end bits as parameter and will
generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change REG_GET to take the mask as an argument instead, and let
the caller create the mask. Eventually, this mask will be moved to a
define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-7-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 4eeeae6e0a0e..48985f0bbbac 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -594,9 +594,8 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
  * number. For example 7:0
  */
 
-#define REG_GET(dispc, idx, start, end)					\
-	((u32)FIELD_GET(GENMASK((start), (end)),			\
-			dispc_read((dispc), (idx))))
+#define REG_GET(dispc, idx, mask)					\
+	((u32)FIELD_GET((mask), dispc_read((dispc), (idx))))
 
 #define REG_FLD_MOD(dispc, idx, val, start, end)			\
 	({								\
@@ -2794,26 +2793,26 @@ int dispc_runtime_resume(struct dispc_device *dispc)
 
 	clk_prepare_enable(dispc->fclk);
 
-	if (REG_GET(dispc, DSS_SYSSTATUS, 0, 0) == 0)
+	if (REG_GET(dispc, DSS_SYSSTATUS, GENMASK(0, 0)) == 0)
 		dev_warn(dispc->dev, "DSS FUNC RESET not done!\n");
 
 	dev_dbg(dispc->dev, "OMAP DSS7 rev 0x%x\n",
 		dispc_read(dispc, DSS_REVISION));
 
 	dev_dbg(dispc->dev, "VP RESETDONE %d,%d,%d\n",
-		REG_GET(dispc, DSS_SYSSTATUS, 1, 1),
-		REG_GET(dispc, DSS_SYSSTATUS, 2, 2),
-		REG_GET(dispc, DSS_SYSSTATUS, 3, 3));
+		REG_GET(dispc, DSS_SYSSTATUS, GENMASK(1, 1)),
+		REG_GET(dispc, DSS_SYSSTATUS, GENMASK(2, 2)),
+		REG_GET(dispc, DSS_SYSSTATUS, GENMASK(3, 3)));
 
 	if (dispc->feat->subrev == DISPC_AM625 ||
 	    dispc->feat->subrev == DISPC_AM65X)
 		dev_dbg(dispc->dev, "OLDI RESETDONE %d,%d,%d\n",
-			REG_GET(dispc, DSS_SYSSTATUS, 5, 5),
-			REG_GET(dispc, DSS_SYSSTATUS, 6, 6),
-			REG_GET(dispc, DSS_SYSSTATUS, 7, 7));
+			REG_GET(dispc, DSS_SYSSTATUS, GENMASK(5, 5)),
+			REG_GET(dispc, DSS_SYSSTATUS, GENMASK(6, 6)),
+			REG_GET(dispc, DSS_SYSSTATUS, GENMASK(7, 7)));
 
 	dev_dbg(dispc->dev, "DISPC IDLE %d\n",
-		REG_GET(dispc, DSS_SYSSTATUS, 9, 9));
+		REG_GET(dispc, DSS_SYSSTATUS, GENMASK(9, 9)));
 
 	dispc_initial_config(dispc);
 

From e6b571e6ff6b46c5859ebd3364c98ccc4c90fbdb Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:39 +0200
Subject: [PATCH 0801/1455] drm/tidss: dispc: Switch REG_FLD_MOD to using a
 mask

The REG_FLD_MOD function takes the start and end bits as parameter and
will generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change REG_FLD_MOD to take the mask as an argument instead, and
let the caller create the mask. Eventually, this mask will be moved to a
define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-8-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 48985f0bbbac..82b038285acc 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -597,12 +597,12 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 #define REG_GET(dispc, idx, mask)					\
 	((u32)FIELD_GET((mask), dispc_read((dispc), (idx))))
 
-#define REG_FLD_MOD(dispc, idx, val, start, end)			\
+#define REG_FLD_MOD(dispc, idx, val, mask)				\
 	({								\
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _idx = (idx);					\
 		u32 _reg = dispc_read(_dispc, _idx);			\
-		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		FIELD_MODIFY((mask), &_reg, (val));			\
 		dispc_write(_dispc, _idx, _reg);			\
 	})
 
@@ -2318,9 +2318,9 @@ static void dispc_k2g_plane_init(struct dispc_device *dispc)
 	dev_dbg(dispc->dev, "%s()\n", __func__);
 
 	/* MFLAG_CTRL = ENABLED */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2, 1, 0);
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2, GENMASK(1, 0));
 	/* MFLAG_START = MFLAGNORMALSTARTMODE */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0, 6, 6);
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0, GENMASK(6, 6));
 
 	for (hw_plane = 0; hw_plane < dispc->feat->num_vids; hw_plane++) {
 		u32 size = dispc_vid_get_fifo_size(dispc, hw_plane);
@@ -2369,13 +2369,13 @@ static void dispc_k3_plane_init(struct dispc_device *dispc)
 
 	dev_dbg(dispc->dev, "%s()\n", __func__);
 
-	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_lo_pri, 2, 0);
-	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_hi_pri, 5, 3);
+	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_lo_pri, GENMASK(2, 0));
+	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_hi_pri, GENMASK(5, 3));
 
 	/* MFLAG_CTRL = ENABLED */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2, 1, 0);
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2, GENMASK(1, 0));
 	/* MFLAG_START = MFLAGNORMALSTARTMODE */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0, 6, 6);
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0, GENMASK(6, 6));
 
 	for (hw_plane = 0; hw_plane < dispc->feat->num_vids; hw_plane++) {
 		u32 size = dispc_vid_get_fifo_size(dispc, hw_plane);
@@ -2903,7 +2903,7 @@ static int dispc_softreset(struct dispc_device *dispc)
 	}
 
 	/* Soft reset */
-	REG_FLD_MOD(dispc, DSS_SYSCONFIG, 1, 1, 1);
+	REG_FLD_MOD(dispc, DSS_SYSCONFIG, 1, GENMASK(1, 1));
 	/* Wait for reset to complete */
 	ret = readl_poll_timeout(dispc->base_common + DSS_SYSSTATUS,
 				 val, val & 1, 100, 5000);

From 990e6f28e78fc0a760fb2bd8575e3eaf5067306d Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:40 +0200
Subject: [PATCH 0802/1455] drm/tidss: dispc: Switch VID_REG_GET to using a
 mask

The VID_REG_GET function takes the start and end bits as parameter and
will generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change VID_REG_GET to take the mask as an argument instead, and
let the caller create the mask. Eventually, this mask will be moved to a
define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-9-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 82b038285acc..8e2c8ecb9f18 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -606,9 +606,8 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 		dispc_write(_dispc, _idx, _reg);			\
 	})
 
-#define VID_REG_GET(dispc, hw_plane, idx, start, end)			\
-	((u32)FIELD_GET(GENMASK((start), (end)),			\
-			dispc_vid_read((dispc), (hw_plane), (idx))))
+#define VID_REG_GET(dispc, hw_plane, idx, mask)				\
+	((u32)FIELD_GET((mask), dispc_vid_read((dispc), (hw_plane), (idx))))
 
 #define VID_REG_FLD_MOD(dispc, hw_plane, idx, val, start, end)		\
 	({								\
@@ -2294,7 +2293,8 @@ void dispc_plane_enable(struct dispc_device *dispc, u32 hw_plane, bool enable)
 
 static u32 dispc_vid_get_fifo_size(struct dispc_device *dispc, u32 hw_plane)
 {
-	return VID_REG_GET(dispc, hw_plane, DISPC_VID_BUF_SIZE_STATUS, 15, 0);
+	return VID_REG_GET(dispc, hw_plane, DISPC_VID_BUF_SIZE_STATUS,
+			   GENMASK(15, 0));
 }
 
 static void dispc_vid_set_mflag_threshold(struct dispc_device *dispc,

From aeaef1ba6bba76eb256123505bbc01032bfd5caa Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:41 +0200
Subject: [PATCH 0803/1455] drm/tidss: dispc: Switch VID_REG_FLD_MOD to using a
 mask

The VID_REG_FLD_MOD function takes the start and end bits as parameter
and will generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change VID_REG_FLD_MOD to take the mask as an argument instead,
and let the caller create the mask. Eventually, this mask will be moved
to a define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-10-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 32 +++++++++++++++--------------
 1 file changed, 17 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 8e2c8ecb9f18..27ac57d77032 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -609,13 +609,13 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 #define VID_REG_GET(dispc, hw_plane, idx, mask)				\
 	((u32)FIELD_GET((mask), dispc_vid_read((dispc), (hw_plane), (idx))))
 
-#define VID_REG_FLD_MOD(dispc, hw_plane, idx, val, start, end)		\
+#define VID_REG_FLD_MOD(dispc, hw_plane, idx, val, mask)		\
 	({								\
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _hw_plane = (hw_plane);				\
 		u32 _idx = (idx);					\
 		u32 _reg = dispc_vid_read(_dispc, _hw_plane, _idx);	\
-		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		FIELD_MODIFY((mask), &_reg, (val));			\
 		dispc_vid_write(_dispc, _hw_plane, _idx, _reg);		\
 	})
 
@@ -1742,7 +1742,8 @@ static void dispc_vid_csc_setup(struct dispc_device *dispc, u32 hw_plane,
 static void dispc_vid_csc_enable(struct dispc_device *dispc, u32 hw_plane,
 				 bool enable)
 {
-	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable, 9, 9);
+	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable,
+			GENMASK(9, 9));
 }
 
 /* SCALER */
@@ -1999,20 +2000,20 @@ static void dispc_vid_set_scaling(struct dispc_device *dispc,
 				  u32 fourcc)
 {
 	/* HORIZONTAL RESIZE ENABLE */
-	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES,
-			sp->scale_x, 7, 7);
+	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, sp->scale_x,
+			GENMASK(7, 7));
 
 	/* VERTICAL RESIZE ENABLE */
-	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES,
-			sp->scale_y, 8, 8);
+	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, sp->scale_y,
+			GENMASK(8, 8));
 
 	/* Skip the rest if no scaling is used */
 	if (!sp->scale_x && !sp->scale_y)
 		return;
 
 	/* VERTICAL 5-TAPS  */
-	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES,
-			sp->five_taps, 21, 21);
+	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, sp->five_taps,
+			GENMASK(21, 21));
 
 	if (dispc_fourcc_is_yuv(fourcc)) {
 		if (sp->scale_x) {
@@ -2102,7 +2103,7 @@ static void dispc_plane_set_pixel_format(struct dispc_device *dispc,
 		if (dispc_color_formats[i].fourcc == fourcc) {
 			VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES,
 					dispc_color_formats[i].dss_code,
-					6, 1);
+					GENMASK(6, 1));
 			return;
 		}
 	}
@@ -2280,15 +2281,16 @@ void dispc_plane_setup(struct dispc_device *dispc, u32 hw_plane,
 
 	if (state->pixel_blend_mode == DRM_MODE_BLEND_PREMULTI)
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 1,
-				28, 28);
+				GENMASK(28, 28));
 	else
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 0,
-				28, 28);
+				GENMASK(28, 28));
 }
 
 void dispc_plane_enable(struct dispc_device *dispc, u32 hw_plane, bool enable)
 {
-	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable, 0, 0);
+	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable,
+			GENMASK(0, 0));
 }
 
 static u32 dispc_vid_get_fifo_size(struct dispc_device *dispc, u32 hw_plane)
@@ -2357,7 +2359,7 @@ static void dispc_k2g_plane_init(struct dispc_device *dispc)
 		 * register is ignored.
 		 */
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 1,
-				19, 19);
+				GENMASK(19, 19));
 	}
 }
 
@@ -2408,7 +2410,7 @@ static void dispc_k3_plane_init(struct dispc_device *dispc)
 
 		/* Prefech up to PRELOAD value */
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 0,
-				19, 19);
+				GENMASK(19, 19));
 	}
 }
 

From b695ff1e7ad418b7525f76fa442f8511ec12f6d9 Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:42 +0200
Subject: [PATCH 0804/1455] drm/tidss: dispc: Switch VP_REG_GET to using a mask

The VP_REG_GET function takes the start and end bits as parameter and
will generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change VP_REG_GET to take the mask as an argument instead, and let
the caller create the mask. Eventually, this mask will be moved to a
define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-11-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 27ac57d77032..3d807b129c09 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -619,9 +619,8 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 		dispc_vid_write(_dispc, _hw_plane, _idx, _reg);		\
 	})
 
-#define VP_REG_GET(dispc, vp, idx, start, end)				\
-	((u32)FIELD_GET(GENMASK((start), (end)),			\
-			dispc_vp_read((dispc), (vp), (idx))))
+#define VP_REG_GET(dispc, vp, idx, mask)				\
+	((u32)FIELD_GET((mask), dispc_vp_read((dispc), (vp), (idx))))
 
 #define VP_REG_FLD_MOD(dispc, vp, idx, val, start, end)			\
 	({								\
@@ -1260,12 +1259,13 @@ void dispc_vp_unprepare(struct dispc_device *dispc, u32 hw_videoport)
 
 bool dispc_vp_go_busy(struct dispc_device *dispc, u32 hw_videoport)
 {
-	return VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL, 5, 5);
+	return VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL,
+			  GENMASK(5, 5));
 }
 
 void dispc_vp_go(struct dispc_device *dispc, u32 hw_videoport)
 {
-	WARN_ON(VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL, 5, 5));
+	WARN_ON(VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL, GENMASK(5, 5)));
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1, 5, 5);
 }
 

From 68f7fa24776e2821896dd95ba0cf1a77a302575f Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:43 +0200
Subject: [PATCH 0805/1455] drm/tidss: dispc: Switch VP_REG_FLD_MOD to using a
 mask

The VP_REG_FLD_MOD function takes the start and end bits as parameter
and will generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change VP_REG_FLD_MOD to take the mask as an argument instead, and
let the caller create the mask. Eventually, this mask will be moved to a
define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-12-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 29 +++++++++++++++++------------
 1 file changed, 17 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 3d807b129c09..0255f7156f46 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -622,13 +622,13 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 #define VP_REG_GET(dispc, vp, idx, mask)				\
 	((u32)FIELD_GET((mask), dispc_vp_read((dispc), (vp), (idx))))
 
-#define VP_REG_FLD_MOD(dispc, vp, idx, val, start, end)			\
+#define VP_REG_FLD_MOD(dispc, vp, idx, val, mask)			\
 	({								\
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _vp = (vp);						\
 		u32 _idx = (idx);					\
 		u32 _reg = dispc_vp_read(_dispc, _vp, _idx);		\
-		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		FIELD_MODIFY((mask), &_reg, (val));			\
 		dispc_vp_write(_dispc, _vp, _idx, _reg);		\
 	})
 
@@ -1113,7 +1113,8 @@ static void dispc_set_num_datalines(struct dispc_device *dispc,
 		v = 3;
 	}
 
-	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, v, 10, 8);
+	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, v,
+		       GENMASK(10, 8));
 }
 
 static void dispc_enable_am65x_oldi(struct dispc_device *dispc, u32 hw_videoport,
@@ -1240,12 +1241,14 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 		       FIELD_PREP(GENMASK(11, 0), mode->crtc_hdisplay - 1) |
 		       FIELD_PREP(GENMASK(27, 16), mode->crtc_vdisplay - 1));
 
-	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1, 0, 0);
+	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1,
+		       GENMASK(0, 0));
 }
 
 void dispc_vp_disable(struct dispc_device *dispc, u32 hw_videoport)
 {
-	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 0, 0, 0);
+	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 0,
+		       GENMASK(0, 0));
 }
 
 void dispc_vp_unprepare(struct dispc_device *dispc, u32 hw_videoport)
@@ -1266,7 +1269,8 @@ bool dispc_vp_go_busy(struct dispc_device *dispc, u32 hw_videoport)
 void dispc_vp_go(struct dispc_device *dispc, u32 hw_videoport)
 {
 	WARN_ON(VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL, GENMASK(5, 5)));
-	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1, 5, 5);
+	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1,
+		       GENMASK(5, 5));
 }
 
 enum c8_to_c12_mode { C8_TO_C12_REPLICATE, C8_TO_C12_MAX, C8_TO_C12_MIN };
@@ -2440,7 +2444,7 @@ static void dispc_vp_init(struct dispc_device *dispc)
 
 	/* Enable the gamma Shadow bit-field for all VPs*/
 	for (i = 0; i < dispc->feat->num_vps; i++)
-		VP_REG_FLD_MOD(dispc, i, DISPC_VP_CONFIG, 1, 2, 2);
+		VP_REG_FLD_MOD(dispc, i, DISPC_VP_CONFIG, 1, GENMASK(2, 2));
 }
 
 static void dispc_initial_config(struct dispc_device *dispc)
@@ -2673,8 +2677,8 @@ static void dispc_k2g_vp_set_ctm(struct dispc_device *dispc, u32 hw_videoport,
 		cprenable = 1;
 	}
 
-	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONFIG,
-		       cprenable, 15, 15);
+	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONFIG, cprenable,
+		       GENMASK(15, 15));
 }
 
 static s16 dispc_S31_32_to_s3_8(s64 coef)
@@ -2739,8 +2743,8 @@ static void dispc_k3_vp_set_ctm(struct dispc_device *dispc, u32 hw_videoport,
 		colorconvenable = 1;
 	}
 
-	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONFIG,
-		       colorconvenable, 24, 24);
+	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONFIG, colorconvenable,
+		       GENMASK(24, 24));
 }
 
 static void dispc_vp_set_color_mgmt(struct dispc_device *dispc,
@@ -2891,7 +2895,8 @@ static void dispc_softreset_k2g(struct dispc_device *dispc)
 	spin_unlock_irqrestore(&dispc->tidss->irq_lock, flags);
 
 	for (unsigned int vp_idx = 0; vp_idx < dispc->feat->num_vps; ++vp_idx)
-		VP_REG_FLD_MOD(dispc, vp_idx, DISPC_VP_CONTROL, 0, 0, 0);
+		VP_REG_FLD_MOD(dispc, vp_idx, DISPC_VP_CONTROL, 0,
+			       GENMASK(0, 0));
 }
 
 static int dispc_softreset(struct dispc_device *dispc)

From 9b74ce7a691750b04ec7d2d7b08490134681c3eb Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:44 +0200
Subject: [PATCH 0806/1455] drm/tidss: dispc: Switch OVR_REG_FLD_MOD to using a
 mask

The OVR_REG_FLD_MOD function takes the start and end bits as parameter
and will generate a mask out of them.

This makes it difficult to share the masks between callers, since we now
need two arguments and to keep them consistent.

Let's change OVR_REG_FLD_MOD to take the mask as an argument instead,
and let the caller create the mask. Eventually, this mask will be moved
to a define.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-13-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 0255f7156f46..9ecbea60d37c 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -632,13 +632,13 @@ void tidss_disable_oldi(struct tidss_device *tidss, u32 hw_videoport)
 		dispc_vp_write(_dispc, _vp, _idx, _reg);		\
 	})
 
-#define OVR_REG_FLD_MOD(dispc, ovr, idx, val, start, end)		\
+#define OVR_REG_FLD_MOD(dispc, ovr, idx, val, mask)			\
 	({								\
 		struct dispc_device *_dispc = (dispc);			\
 		u32 _ovr = (ovr);					\
 		u32 _idx = (idx);					\
 		u32 _reg = dispc_ovr_read(_dispc, _ovr, _idx);		\
-		FIELD_MODIFY(GENMASK((start), (end)), &_reg, (val));	\
+		FIELD_MODIFY((mask), &_reg, (val));			\
 		dispc_ovr_write(_dispc, _ovr, _idx, _reg);		\
 	})
 
@@ -1470,11 +1470,11 @@ static void dispc_am65x_ovr_set_plane(struct dispc_device *dispc,
 	u32 hw_id = dispc->feat->vid_info[hw_plane].hw_id;
 
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			hw_id, 4, 1);
-	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			x, 17, 6);
-	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			y, 30, 19);
+			hw_id, GENMASK(4, 1));
+	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), x,
+			GENMASK(17, 6));
+	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), y,
+			GENMASK(30, 19));
 }
 
 static void dispc_j721e_ovr_set_plane(struct dispc_device *dispc,
@@ -1484,11 +1484,11 @@ static void dispc_j721e_ovr_set_plane(struct dispc_device *dispc,
 	u32 hw_id = dispc->feat->vid_info[hw_plane].hw_id;
 
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			hw_id, 4, 1);
-	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer),
-			x, 13, 0);
-	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer),
-			y, 29, 16);
+			hw_id, GENMASK(4, 1));
+	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer), x,
+			GENMASK(13, 0));
+	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer), y,
+			GENMASK(29, 16));
 }
 
 void dispc_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane,
@@ -1523,7 +1523,7 @@ void dispc_ovr_enable_layer(struct dispc_device *dispc,
 		return;
 
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			!!enable, 0, 0);
+			!!enable, GENMASK(0, 0));
 }
 
 /* CSC */

From 7287177e5abe2472c03f1c78099352c64d1b86fc Mon Sep 17 00:00:00 2001
From: Maxime Ripard <mripard@kernel.org>
Date: Wed, 27 Aug 2025 17:12:45 +0200
Subject: [PATCH 0807/1455] drm/tidss: dispc: Define field masks being used

Now that we have all the accessors taking masks, we can create defines
for them and reuse them as needed.

It makes the driver easier to read, less prone to consistency issues,
and allows to reuse defines when needed.

Signed-off-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250827-drm-tidss-field-api-v3-14-7689b664cc63@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c      | 139 +++++++++++++----------
 drivers/gpu/drm/tidss/tidss_dispc_regs.h |  76 +++++++++++++
 2 files changed, 154 insertions(+), 61 deletions(-)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 9ecbea60d37c..97e2cd7b1e20 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -1114,7 +1114,7 @@ static void dispc_set_num_datalines(struct dispc_device *dispc,
 	}
 
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, v,
-		       GENMASK(10, 8));
+		       DISPC_VP_CONTROL_DATALINES_MASK);
 }
 
 static void dispc_enable_am65x_oldi(struct dispc_device *dispc, u32 hw_videoport,
@@ -1137,7 +1137,8 @@ static void dispc_enable_am65x_oldi(struct dispc_device *dispc, u32 hw_videoport
 
 	oldi_cfg |= BIT(7); /* DEPOL */
 
-	FIELD_MODIFY(GENMASK(3, 1), &oldi_cfg, fmt->am65x_oldi_mode_reg_val);
+	FIELD_MODIFY(DISPC_VP_DSS_OLDI_CFG_MAP_MASK, &oldi_cfg,
+		     fmt->am65x_oldi_mode_reg_val);
 
 	oldi_cfg |= BIT(12); /* SOFTRST */
 
@@ -1199,14 +1200,14 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 	vbp = mode->crtc_vtotal - mode->crtc_vsync_end;
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_TIMING_H,
-		       FIELD_PREP(GENMASK(7, 0), hsw - 1) |
-		       FIELD_PREP(GENMASK(19, 8), hfp - 1) |
-		       FIELD_PREP(GENMASK(31, 20), hbp - 1));
+		       FIELD_PREP(DISPC_VP_TIMING_H_SYNC_PULSE_MASK, hsw - 1) |
+		       FIELD_PREP(DISPC_VP_TIMING_H_FRONT_PORCH_MASK, hfp - 1) |
+		       FIELD_PREP(DISPC_VP_TIMING_H_BACK_PORCH_MASK, hbp - 1));
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_TIMING_V,
-		       FIELD_PREP(GENMASK(7, 0), vsw - 1) |
-		       FIELD_PREP(GENMASK(19, 8), vfp) |
-		       FIELD_PREP(GENMASK(31, 20), vbp));
+		       FIELD_PREP(DISPC_VP_TIMING_V_SYNC_PULSE_MASK, vsw - 1) |
+		       FIELD_PREP(DISPC_VP_TIMING_V_FRONT_PORCH_MASK, vfp) |
+		       FIELD_PREP(DISPC_VP_TIMING_V_BACK_PORCH_MASK, vbp));
 
 	ivs = !!(mode->flags & DRM_MODE_FLAG_NVSYNC);
 
@@ -1229,26 +1230,28 @@ void dispc_vp_enable(struct dispc_device *dispc, u32 hw_videoport,
 		ieo = false;
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_POL_FREQ,
-		       FIELD_PREP(GENMASK(18, 18), align) |
-		       FIELD_PREP(GENMASK(17, 17), onoff) |
-		       FIELD_PREP(GENMASK(16, 16), rf) |
-		       FIELD_PREP(GENMASK(15, 15), ieo) |
-		       FIELD_PREP(GENMASK(14, 14), ipc) |
-		       FIELD_PREP(GENMASK(13, 13), ihs) |
-		       FIELD_PREP(GENMASK(12, 12), ivs));
+		       FIELD_PREP(DISPC_VP_POL_FREQ_ALIGN_MASK, align) |
+		       FIELD_PREP(DISPC_VP_POL_FREQ_ONOFF_MASK, onoff) |
+		       FIELD_PREP(DISPC_VP_POL_FREQ_RF_MASK, rf) |
+		       FIELD_PREP(DISPC_VP_POL_FREQ_IEO_MASK, ieo) |
+		       FIELD_PREP(DISPC_VP_POL_FREQ_IPC_MASK, ipc) |
+		       FIELD_PREP(DISPC_VP_POL_FREQ_IHS_MASK, ihs) |
+		       FIELD_PREP(DISPC_VP_POL_FREQ_IVS_MASK, ivs));
 
 	dispc_vp_write(dispc, hw_videoport, DISPC_VP_SIZE_SCREEN,
-		       FIELD_PREP(GENMASK(11, 0), mode->crtc_hdisplay - 1) |
-		       FIELD_PREP(GENMASK(27, 16), mode->crtc_vdisplay - 1));
+		       FIELD_PREP(DISPC_VP_SIZE_SCREEN_HDISPLAY_MASK,
+				  mode->crtc_hdisplay - 1) |
+		       FIELD_PREP(DISPC_VP_SIZE_SCREEN_VDISPLAY_MASK,
+				  mode->crtc_vdisplay - 1));
 
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1,
-		       GENMASK(0, 0));
+		       DISPC_VP_CONTROL_ENABLE_MASK);
 }
 
 void dispc_vp_disable(struct dispc_device *dispc, u32 hw_videoport)
 {
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 0,
-		       GENMASK(0, 0));
+		       DISPC_VP_CONTROL_ENABLE_MASK);
 }
 
 void dispc_vp_unprepare(struct dispc_device *dispc, u32 hw_videoport)
@@ -1263,14 +1266,15 @@ void dispc_vp_unprepare(struct dispc_device *dispc, u32 hw_videoport)
 bool dispc_vp_go_busy(struct dispc_device *dispc, u32 hw_videoport)
 {
 	return VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL,
-			  GENMASK(5, 5));
+			  DISPC_VP_CONTROL_GOBIT_MASK);
 }
 
 void dispc_vp_go(struct dispc_device *dispc, u32 hw_videoport)
 {
-	WARN_ON(VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL, GENMASK(5, 5)));
+	WARN_ON(VP_REG_GET(dispc, hw_videoport, DISPC_VP_CONTROL,
+			   DISPC_VP_CONTROL_GOBIT_MASK));
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, 1,
-		       GENMASK(5, 5));
+		       DISPC_VP_CONTROL_GOBIT_MASK);
 }
 
 enum c8_to_c12_mode { C8_TO_C12_REPLICATE, C8_TO_C12_MAX, C8_TO_C12_MIN };
@@ -1470,11 +1474,11 @@ static void dispc_am65x_ovr_set_plane(struct dispc_device *dispc,
 	u32 hw_id = dispc->feat->vid_info[hw_plane].hw_id;
 
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			hw_id, GENMASK(4, 1));
+			hw_id, DISPC_OVR_ATTRIBUTES_CHANNELIN_MASK);
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), x,
-			GENMASK(17, 6));
+			DISPC_OVR_ATTRIBUTES_POSX_MASK);
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer), y,
-			GENMASK(30, 19));
+			DISPC_OVR_ATTRIBUTES_POSY_MASK);
 }
 
 static void dispc_j721e_ovr_set_plane(struct dispc_device *dispc,
@@ -1484,11 +1488,11 @@ static void dispc_j721e_ovr_set_plane(struct dispc_device *dispc,
 	u32 hw_id = dispc->feat->vid_info[hw_plane].hw_id;
 
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			hw_id, GENMASK(4, 1));
+			hw_id, DISPC_OVR_ATTRIBUTES_CHANNELIN_MASK);
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer), x,
-			GENMASK(13, 0));
+			DISPC_OVR_ATTRIBUTES2_POSX_MASK);
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES2(layer), y,
-			GENMASK(29, 16));
+			DISPC_OVR_ATTRIBUTES2_POSY_MASK);
 }
 
 void dispc_ovr_set_plane(struct dispc_device *dispc, u32 hw_plane,
@@ -1523,7 +1527,7 @@ void dispc_ovr_enable_layer(struct dispc_device *dispc,
 		return;
 
 	OVR_REG_FLD_MOD(dispc, hw_videoport, DISPC_OVR_ATTRIBUTES(layer),
-			!!enable, GENMASK(0, 0));
+			!!enable, DISPC_OVR_ATTRIBUTES_ENABLE_MASK);
 }
 
 /* CSC */
@@ -1747,7 +1751,7 @@ static void dispc_vid_csc_enable(struct dispc_device *dispc, u32 hw_plane,
 				 bool enable)
 {
 	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable,
-			GENMASK(9, 9));
+			DISPC_VID_ATTRIBUTES_COLORCONVENABLE_MASK);
 }
 
 /* SCALER */
@@ -2005,11 +2009,11 @@ static void dispc_vid_set_scaling(struct dispc_device *dispc,
 {
 	/* HORIZONTAL RESIZE ENABLE */
 	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, sp->scale_x,
-			GENMASK(7, 7));
+			DISPC_VID_ATTRIBUTES_HRESIZEENABLE_MASK);
 
 	/* VERTICAL RESIZE ENABLE */
 	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, sp->scale_y,
-			GENMASK(8, 8));
+			DISPC_VID_ATTRIBUTES_VRESIZEENABLE_MASK);
 
 	/* Skip the rest if no scaling is used */
 	if (!sp->scale_x && !sp->scale_y)
@@ -2017,7 +2021,7 @@ static void dispc_vid_set_scaling(struct dispc_device *dispc,
 
 	/* VERTICAL 5-TAPS  */
 	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, sp->five_taps,
-			GENMASK(21, 21));
+			DISPC_VID_ATTRIBUTES_VERTICALTAPS_MASK);
 
 	if (dispc_fourcc_is_yuv(fourcc)) {
 		if (sp->scale_x) {
@@ -2107,7 +2111,7 @@ static void dispc_plane_set_pixel_format(struct dispc_device *dispc,
 		if (dispc_color_formats[i].fourcc == fourcc) {
 			VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES,
 					dispc_color_formats[i].dss_code,
-					GENMASK(6, 1));
+					DISPC_VID_ATTRIBUTES_FORMAT_MASK);
 			return;
 		}
 	}
@@ -2229,7 +2233,8 @@ void dispc_plane_setup(struct dispc_device *dispc, u32 hw_plane,
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_BA_EXT_1, (u64)dma_addr >> 32);
 
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_PICTURE_SIZE,
-			(scale.in_w - 1) | ((scale.in_h - 1) << 16));
+			FIELD_PREP(DISPC_VID_PICTURE_SIZE_MEMSIZEY_MASK, scale.in_h - 1) |
+			FIELD_PREP(DISPC_VID_PICTURE_SIZE_MEMSIZEX_MASK, scale.in_w - 1));
 
 	/* For YUV422 format we use the macropixel size for pixel inc */
 	if (fourcc == DRM_FORMAT_YUYV || fourcc == DRM_FORMAT_UYVY)
@@ -2266,8 +2271,10 @@ void dispc_plane_setup(struct dispc_device *dispc, u32 hw_plane,
 
 	if (!lite) {
 		dispc_vid_write(dispc, hw_plane, DISPC_VID_SIZE,
-				(state->crtc_w - 1) |
-				((state->crtc_h - 1) << 16));
+				FIELD_PREP(DISPC_VID_SIZE_SIZEY_MASK,
+					   state->crtc_h - 1) |
+				FIELD_PREP(DISPC_VID_SIZE_SIZEX_MASK,
+					   state->crtc_w - 1));
 
 		dispc_vid_set_scaling(dispc, hw_plane, &scale, fourcc);
 	}
@@ -2281,40 +2288,45 @@ void dispc_plane_setup(struct dispc_device *dispc, u32 hw_plane,
 	}
 
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_GLOBAL_ALPHA,
-			0xFF & (state->alpha >> 8));
+			FIELD_PREP(DISPC_VID_GLOBAL_ALPHA_GLOBALALPHA_MASK,
+				   state->alpha >> 8));
 
 	if (state->pixel_blend_mode == DRM_MODE_BLEND_PREMULTI)
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 1,
-				GENMASK(28, 28));
+				DISPC_VID_ATTRIBUTES_PREMULTIPLYALPHA_MASK);
 	else
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 0,
-				GENMASK(28, 28));
+				DISPC_VID_ATTRIBUTES_PREMULTIPLYALPHA_MASK);
 }
 
 void dispc_plane_enable(struct dispc_device *dispc, u32 hw_plane, bool enable)
 {
 	VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, !!enable,
-			GENMASK(0, 0));
+			DISPC_VID_ATTRIBUTES_ENABLE_MASK);
 }
 
 static u32 dispc_vid_get_fifo_size(struct dispc_device *dispc, u32 hw_plane)
 {
 	return VID_REG_GET(dispc, hw_plane, DISPC_VID_BUF_SIZE_STATUS,
-			   GENMASK(15, 0));
+			   DISPC_VID_BUF_SIZE_STATUS_BUFSIZE_MASK);
 }
 
 static void dispc_vid_set_mflag_threshold(struct dispc_device *dispc,
 					  u32 hw_plane, u32 low, u32 high)
 {
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_MFLAG_THRESHOLD,
-			FIELD_PREP(GENMASK(31, 16), high) | FIELD_PREP(GENMASK(15, 0), low));
+			FIELD_PREP(DISPC_VID_MFLAG_THRESHOLD_HT_MFLAG_MASK, high) |
+			FIELD_PREP(DISPC_VID_MFLAG_THRESHOLD_LT_MFLAG_MASK, low));
 }
 
 static void dispc_vid_set_buf_threshold(struct dispc_device *dispc,
 					u32 hw_plane, u32 low, u32 high)
 {
 	dispc_vid_write(dispc, hw_plane, DISPC_VID_BUF_THRESHOLD,
-			FIELD_PREP(GENMASK(31, 16), high) | FIELD_PREP(GENMASK(15, 0), low));
+			FIELD_PREP(DISPC_VID_BUF_THRESHOLD_BUFHIGHTHRESHOLD_MASK,
+				   high) |
+			FIELD_PREP(DISPC_VID_BUF_THRESHOLD_BUFLOWTHRESHOLD_MASK,
+				   low));
 }
 
 static void dispc_k2g_plane_init(struct dispc_device *dispc)
@@ -2324,9 +2336,11 @@ static void dispc_k2g_plane_init(struct dispc_device *dispc)
 	dev_dbg(dispc->dev, "%s()\n", __func__);
 
 	/* MFLAG_CTRL = ENABLED */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2, GENMASK(1, 0));
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2,
+		    DISPC_GLOBAL_MFLAG_ATTRIBUTE_MFLAG_CTRL_MASK);
 	/* MFLAG_START = MFLAGNORMALSTARTMODE */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0, GENMASK(6, 6));
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0,
+		    DISPC_GLOBAL_MFLAG_ATTRIBUTE_MFLAG_START_MASK);
 
 	for (hw_plane = 0; hw_plane < dispc->feat->num_vids; hw_plane++) {
 		u32 size = dispc_vid_get_fifo_size(dispc, hw_plane);
@@ -2363,7 +2377,7 @@ static void dispc_k2g_plane_init(struct dispc_device *dispc)
 		 * register is ignored.
 		 */
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 1,
-				GENMASK(19, 19));
+				DISPC_VID_ATTRIBUTES_BUFPRELOAD_MASK);
 	}
 }
 
@@ -2375,13 +2389,15 @@ static void dispc_k3_plane_init(struct dispc_device *dispc)
 
 	dev_dbg(dispc->dev, "%s()\n", __func__);
 
-	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_lo_pri, GENMASK(2, 0));
-	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_hi_pri, GENMASK(5, 3));
+	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_lo_pri, DSS_CBA_CFG_PRI_LO_MASK);
+	REG_FLD_MOD(dispc, DSS_CBA_CFG, cba_hi_pri, DSS_CBA_CFG_PRI_HI_MASK);
 
 	/* MFLAG_CTRL = ENABLED */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2, GENMASK(1, 0));
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 2,
+		    DISPC_GLOBAL_MFLAG_ATTRIBUTE_MFLAG_CTRL_MASK);
 	/* MFLAG_START = MFLAGNORMALSTARTMODE */
-	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0, GENMASK(6, 6));
+	REG_FLD_MOD(dispc, DISPC_GLOBAL_MFLAG_ATTRIBUTE, 0,
+		    DISPC_GLOBAL_MFLAG_ATTRIBUTE_MFLAG_START_MASK);
 
 	for (hw_plane = 0; hw_plane < dispc->feat->num_vids; hw_plane++) {
 		u32 size = dispc_vid_get_fifo_size(dispc, hw_plane);
@@ -2414,7 +2430,7 @@ static void dispc_k3_plane_init(struct dispc_device *dispc)
 
 		/* Prefech up to PRELOAD value */
 		VID_REG_FLD_MOD(dispc, hw_plane, DISPC_VID_ATTRIBUTES, 0,
-				GENMASK(19, 19));
+				DISPC_VID_ATTRIBUTES_BUFPRELOAD_MASK);
 	}
 }
 
@@ -2444,7 +2460,8 @@ static void dispc_vp_init(struct dispc_device *dispc)
 
 	/* Enable the gamma Shadow bit-field for all VPs*/
 	for (i = 0; i < dispc->feat->num_vps; i++)
-		VP_REG_FLD_MOD(dispc, i, DISPC_VP_CONFIG, 1, GENMASK(2, 2));
+		VP_REG_FLD_MOD(dispc, i, DISPC_VP_CONFIG, 1,
+			       DISPC_VP_CONFIG_GAMMAENABLE_MASK);
 }
 
 static void dispc_initial_config(struct dispc_device *dispc)
@@ -2455,8 +2472,8 @@ static void dispc_initial_config(struct dispc_device *dispc)
 	/* Note: Hardcoded DPI routing on J721E for now */
 	if (dispc->feat->subrev == DISPC_J721E) {
 		dispc_write(dispc, DISPC_CONNECTIONS,
-			    FIELD_PREP(GENMASK(3, 0), 2) |		/* VP1 to DPI0 */
-			    FIELD_PREP(GENMASK(7, 4), 8)		/* VP3 to DPI1 */
+			    FIELD_PREP(DISPC_CONNECTIONS_DPI_0_CONN_MASK, 2) |		/* VP1 to DPI0 */
+			    FIELD_PREP(DISPC_CONNECTIONS_DPI_1_CONN_MASK, 8)		/* VP3 to DPI1 */
 			);
 	}
 }
@@ -2678,7 +2695,7 @@ static void dispc_k2g_vp_set_ctm(struct dispc_device *dispc, u32 hw_videoport,
 	}
 
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONFIG, cprenable,
-		       GENMASK(15, 15));
+		       DISPC_VP_CONFIG_CPR_MASK);
 }
 
 static s16 dispc_S31_32_to_s3_8(s64 coef)
@@ -2744,7 +2761,7 @@ static void dispc_k3_vp_set_ctm(struct dispc_device *dispc, u32 hw_videoport,
 	}
 
 	VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONFIG, colorconvenable,
-		       GENMASK(24, 24));
+		       DISPC_VP_CONFIG_COLORCONVENABLE_MASK);
 }
 
 static void dispc_vp_set_color_mgmt(struct dispc_device *dispc,
@@ -2799,7 +2816,7 @@ int dispc_runtime_resume(struct dispc_device *dispc)
 
 	clk_prepare_enable(dispc->fclk);
 
-	if (REG_GET(dispc, DSS_SYSSTATUS, GENMASK(0, 0)) == 0)
+	if (REG_GET(dispc, DSS_SYSSTATUS, DSS_SYSSTATUS_DISPC_FUNC_RESETDONE) == 0)
 		dev_warn(dispc->dev, "DSS FUNC RESET not done!\n");
 
 	dev_dbg(dispc->dev, "OMAP DSS7 rev 0x%x\n",
@@ -2818,7 +2835,7 @@ int dispc_runtime_resume(struct dispc_device *dispc)
 			REG_GET(dispc, DSS_SYSSTATUS, GENMASK(7, 7)));
 
 	dev_dbg(dispc->dev, "DISPC IDLE %d\n",
-		REG_GET(dispc, DSS_SYSSTATUS, GENMASK(9, 9)));
+		REG_GET(dispc, DSS_SYSSTATUS, DSS_SYSSTATUS_DISPC_IDLE_STATUS));
 
 	dispc_initial_config(dispc);
 
@@ -2896,7 +2913,7 @@ static void dispc_softreset_k2g(struct dispc_device *dispc)
 
 	for (unsigned int vp_idx = 0; vp_idx < dispc->feat->num_vps; ++vp_idx)
 		VP_REG_FLD_MOD(dispc, vp_idx, DISPC_VP_CONTROL, 0,
-			       GENMASK(0, 0));
+			       DISPC_VP_CONTROL_ENABLE_MASK);
 }
 
 static int dispc_softreset(struct dispc_device *dispc)
@@ -2910,7 +2927,7 @@ static int dispc_softreset(struct dispc_device *dispc)
 	}
 
 	/* Soft reset */
-	REG_FLD_MOD(dispc, DSS_SYSCONFIG, 1, GENMASK(1, 1));
+	REG_FLD_MOD(dispc, DSS_SYSCONFIG, 1, DSS_SYSCONFIG_SOFTRESET_MASK);
 	/* Wait for reset to complete */
 	ret = readl_poll_timeout(dispc->base_common + DSS_SYSSTATUS,
 				 val, val & 1, 100, 5000);
diff --git a/drivers/gpu/drm/tidss/tidss_dispc_regs.h b/drivers/gpu/drm/tidss/tidss_dispc_regs.h
index 50a3f28250ef..382027dddce8 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc_regs.h
+++ b/drivers/gpu/drm/tidss/tidss_dispc_regs.h
@@ -56,7 +56,12 @@ enum dispc_common_regs {
 
 #define DSS_REVISION			REG(DSS_REVISION)
 #define DSS_SYSCONFIG			REG(DSS_SYSCONFIG)
+#define DSS_SYSCONFIG_SOFTRESET_MASK		GENMASK(1, 1)
+
 #define DSS_SYSSTATUS			REG(DSS_SYSSTATUS)
+#define DSS_SYSSTATUS_DISPC_IDLE_STATUS		GENMASK(9, 9)
+#define DSS_SYSSTATUS_DISPC_FUNC_RESETDONE	GENMASK(0, 0)
+
 #define DISPC_IRQ_EOI			REG(DISPC_IRQ_EOI)
 #define DISPC_IRQSTATUS_RAW		REG(DISPC_IRQSTATUS_RAW)
 #define DISPC_IRQSTATUS			REG(DISPC_IRQSTATUS)
@@ -70,9 +75,15 @@ enum dispc_common_regs {
 #define WB_IRQSTATUS			REG(WB_IRQSTATUS)
 
 #define DISPC_GLOBAL_MFLAG_ATTRIBUTE	REG(DISPC_GLOBAL_MFLAG_ATTRIBUTE)
+#define DISPC_GLOBAL_MFLAG_ATTRIBUTE_MFLAG_START_MASK	GENMASK(6, 6)
+#define DISPC_GLOBAL_MFLAG_ATTRIBUTE_MFLAG_CTRL_MASK	GENMASK(1, 0)
+
 #define DISPC_GLOBAL_OUTPUT_ENABLE	REG(DISPC_GLOBAL_OUTPUT_ENABLE)
 #define DISPC_GLOBAL_BUFFER		REG(DISPC_GLOBAL_BUFFER)
 #define DSS_CBA_CFG			REG(DSS_CBA_CFG)
+#define DSS_CBA_CFG_PRI_HI_MASK			GENMASK(5, 3)
+#define DSS_CBA_CFG_PRI_LO_MASK			GENMASK(2, 0)
+
 #define DISPC_DBG_CONTROL		REG(DISPC_DBG_CONTROL)
 #define DISPC_DBG_STATUS		REG(DISPC_DBG_STATUS)
 #define DISPC_CLKGATING_DISABLE		REG(DISPC_CLKGATING_DISABLE)
@@ -88,6 +99,9 @@ enum dispc_common_regs {
 #define FBDC_CONSTANT_COLOR_0		REG(FBDC_CONSTANT_COLOR_0)
 #define FBDC_CONSTANT_COLOR_1		REG(FBDC_CONSTANT_COLOR_1)
 #define DISPC_CONNECTIONS		REG(DISPC_CONNECTIONS)
+#define DISPC_CONNECTIONS_DPI_1_CONN_MASK	GENMASK(7, 4)
+#define DISPC_CONNECTIONS_DPI_0_CONN_MASK	GENMASK(3, 0)
+
 #define DISPC_MSS_VP1			REG(DISPC_MSS_VP1)
 #define DISPC_MSS_VP3			REG(DISPC_MSS_VP3)
 
@@ -102,13 +116,27 @@ enum dispc_common_regs {
 #define DISPC_VID_ACCUV2_0		0x18
 #define DISPC_VID_ACCUV2_1		0x1c
 #define DISPC_VID_ATTRIBUTES		0x20
+#define DISPC_VID_ATTRIBUTES_PREMULTIPLYALPHA_MASK	GENMASK(28, 28)
+#define DISPC_VID_ATTRIBUTES_VERTICALTAPS_MASK		GENMASK(21, 21)
+#define DISPC_VID_ATTRIBUTES_BUFPRELOAD_MASK		GENMASK(19, 19)
+#define DISPC_VID_ATTRIBUTES_COLORCONVENABLE_MASK	GENMASK(9, 9)
+#define DISPC_VID_ATTRIBUTES_VRESIZEENABLE_MASK		GENMASK(8, 8)
+#define DISPC_VID_ATTRIBUTES_HRESIZEENABLE_MASK		GENMASK(7, 7)
+#define DISPC_VID_ATTRIBUTES_FORMAT_MASK		GENMASK(6, 1)
+#define DISPC_VID_ATTRIBUTES_ENABLE_MASK		GENMASK(0, 0)
+
 #define DISPC_VID_ATTRIBUTES2		0x24
 #define DISPC_VID_BA_0			0x28
 #define DISPC_VID_BA_1			0x2c
 #define DISPC_VID_BA_UV_0		0x30
 #define DISPC_VID_BA_UV_1		0x34
 #define DISPC_VID_BUF_SIZE_STATUS	0x38
+#define DISPC_VID_BUF_SIZE_STATUS_BUFSIZE_MASK	GENMASK(15, 0)
+
 #define DISPC_VID_BUF_THRESHOLD		0x3c
+#define DISPC_VID_BUF_THRESHOLD_BUFHIGHTHRESHOLD_MASK	GENMASK(31, 16)
+#define DISPC_VID_BUF_THRESHOLD_BUFLOWTHRESHOLD_MASK	GENMASK(15, 0)
+
 #define DISPC_VID_CSC_COEF(n)		(0x40 + (n) * 4)
 
 #define DISPC_VID_FIRH			0x5c
@@ -137,15 +165,26 @@ enum dispc_common_regs {
 #define DISPC_VID_FIR_COEF_V12_C(phase)	(0x1bc + (phase) * 4)
 
 #define DISPC_VID_GLOBAL_ALPHA		0x1fc
+#define DISPC_VID_GLOBAL_ALPHA_GLOBALALPHA_MASK	GENMASK(7, 0)
+
 #define DISPC_VID_K2G_IRQENABLE		0x200 /* K2G */
 #define DISPC_VID_K2G_IRQSTATUS		0x204 /* K2G */
 #define DISPC_VID_MFLAG_THRESHOLD	0x208
+#define DISPC_VID_MFLAG_THRESHOLD_HT_MFLAG_MASK	GENMASK(31, 16)
+#define DISPC_VID_MFLAG_THRESHOLD_LT_MFLAG_MASK	GENMASK(15, 0)
+
 #define DISPC_VID_PICTURE_SIZE		0x20c
+#define DISPC_VID_PICTURE_SIZE_MEMSIZEY_MASK	GENMASK(27, 16)
+#define DISPC_VID_PICTURE_SIZE_MEMSIZEX_MASK	GENMASK(11, 0)
+
 #define DISPC_VID_PIXEL_INC		0x210
 #define DISPC_VID_K2G_POSITION		0x214 /* K2G */
 #define DISPC_VID_PRELOAD		0x218
 #define DISPC_VID_ROW_INC		0x21c
 #define DISPC_VID_SIZE			0x220
+#define DISPC_VID_SIZE_SIZEY_MASK		GENMASK(27, 16)
+#define DISPC_VID_SIZE_SIZEX_MASK		GENMASK(11, 0)
+
 #define DISPC_VID_BA_EXT_0		0x22c
 #define DISPC_VID_BA_EXT_1		0x230
 #define DISPC_VID_BA_UV_EXT_0		0x234
@@ -173,11 +212,27 @@ enum dispc_common_regs {
 #define DISPC_OVR_TRANS_COLOR_MIN	0x18
 #define DISPC_OVR_TRANS_COLOR_MIN2	0x1c
 #define DISPC_OVR_ATTRIBUTES(n)		(0x20 + (n) * 4)
+#define DISPC_OVR_ATTRIBUTES_POSY_MASK		GENMASK(30, 19)
+#define DISPC_OVR_ATTRIBUTES_POSX_MASK		GENMASK(17, 6)
+#define DISPC_OVR_ATTRIBUTES_CHANNELIN_MASK	GENMASK(4, 1)
+#define DISPC_OVR_ATTRIBUTES_ENABLE_MASK	GENMASK(0, 0)
+
 #define DISPC_OVR_ATTRIBUTES2(n)	(0x34 + (n) * 4) /* J721E */
+#define DISPC_OVR_ATTRIBUTES2_POSY_MASK		GENMASK(29, 16)
+#define DISPC_OVR_ATTRIBUTES2_POSX_MASK		GENMASK(13, 0)
+
 /* VP */
 
 #define DISPC_VP_CONFIG				0x0
+#define DISPC_VP_CONFIG_COLORCONVENABLE_MASK		GENMASK(24, 24)
+#define DISPC_VP_CONFIG_CPR_MASK			GENMASK(15, 15)
+#define DISPC_VP_CONFIG_GAMMAENABLE_MASK		GENMASK(2, 2)
+
 #define DISPC_VP_CONTROL			0x4
+#define DISPC_VP_CONTROL_DATALINES_MASK			GENMASK(10, 8)
+#define DISPC_VP_CONTROL_GOBIT_MASK			GENMASK(5, 5)
+#define DISPC_VP_CONTROL_ENABLE_MASK			GENMASK(0, 0)
+
 #define DISPC_VP_CSC_COEF0			0x8
 #define DISPC_VP_CSC_COEF1			0xc
 #define DISPC_VP_CSC_COEF2			0x10
@@ -189,9 +244,28 @@ enum dispc_common_regs {
 #define DISPC_VP_DATA_CYCLE_2			0x1c
 #define DISPC_VP_LINE_NUMBER			0x44
 #define DISPC_VP_POL_FREQ			0x4c
+#define DISPC_VP_POL_FREQ_ALIGN_MASK			GENMASK(18, 18)
+#define DISPC_VP_POL_FREQ_ONOFF_MASK			GENMASK(17, 17)
+#define DISPC_VP_POL_FREQ_RF_MASK			GENMASK(16, 16)
+#define DISPC_VP_POL_FREQ_IEO_MASK			GENMASK(15, 15)
+#define DISPC_VP_POL_FREQ_IPC_MASK			GENMASK(14, 14)
+#define DISPC_VP_POL_FREQ_IHS_MASK			GENMASK(13, 13)
+#define DISPC_VP_POL_FREQ_IVS_MASK			GENMASK(12, 12)
+
 #define DISPC_VP_SIZE_SCREEN			0x50
+#define DISPC_VP_SIZE_SCREEN_HDISPLAY_MASK		GENMASK(11, 0)
+#define DISPC_VP_SIZE_SCREEN_VDISPLAY_MASK		GENMASK(27, 16)
+
 #define DISPC_VP_TIMING_H			0x54
+#define DISPC_VP_TIMING_H_SYNC_PULSE_MASK		GENMASK(7, 0)
+#define DISPC_VP_TIMING_H_FRONT_PORCH_MASK		GENMASK(19, 8)
+#define DISPC_VP_TIMING_H_BACK_PORCH_MASK		GENMASK(31, 20)
+
 #define DISPC_VP_TIMING_V			0x58
+#define DISPC_VP_TIMING_V_SYNC_PULSE_MASK		GENMASK(7, 0)
+#define DISPC_VP_TIMING_V_FRONT_PORCH_MASK		GENMASK(19, 8)
+#define DISPC_VP_TIMING_V_BACK_PORCH_MASK		GENMASK(31, 20)
+
 #define DISPC_VP_CSC_COEF3			0x5c
 #define DISPC_VP_CSC_COEF4			0x60
 #define DISPC_VP_CSC_COEF5			0x64
@@ -220,6 +294,8 @@ enum dispc_common_regs {
 #define DISPC_VP_SAFETY_LFSR_SEED		0x110
 #define DISPC_VP_GAMMA_TABLE			0x120
 #define DISPC_VP_DSS_OLDI_CFG			0x160
+#define DISPC_VP_DSS_OLDI_CFG_MAP_MASK			GENMASK(3, 1)
+
 #define DISPC_VP_DSS_OLDI_STATUS		0x164
 #define DISPC_VP_DSS_OLDI_LB			0x168
 #define DISPC_VP_DSS_MERGE_SPLIT		0x16c /* J721E */

From 1c936e85926dbd417f8ce4c290acee52d085fe0d Mon Sep 17 00:00:00 2001
From: Javier Martinez Canillas <javierm@redhat.com>
Date: Fri, 18 Jul 2025 17:25:25 +0200
Subject: [PATCH 0808/1455] drm/sitronix/st7571-i2c: Make st7571_panel_data
 variables static const

The kernel test robot reported that sparse gives the following warnings:

make C=2 M=drivers/gpu/drm/sitronix/
  CC [M]  st7571-i2c.o
  CHECK   st7571-i2c.c
st7571-i2c.c:1027:26: warning: symbol 'st7567_config' was not declared. Should it be static?
st7571-i2c.c:1039:26: warning: symbol 'st7571_config' was not declared. Should it be static?
  MODPOST Module.symvers
  LD [M]  st7571-i2c.ko

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202507180503.nfyD9uRv-lkp@intel.com
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250718152534.729770-1-javierm@redhat.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/gpu/drm/sitronix/st7571-i2c.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
index 453eb7e045e5..dda99cbe5372 100644
--- a/drivers/gpu/drm/sitronix/st7571-i2c.c
+++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
@@ -1024,7 +1024,7 @@ static void st7571_remove(struct i2c_client *client)
 	drm_dev_unplug(&st7571->dev);
 }
 
-struct st7571_panel_data st7567_config = {
+static const struct st7571_panel_data st7567_config = {
 	.init = st7567_lcd_init,
 	.parse_dt = st7567_parse_dt,
 	.constraints = {
@@ -1036,7 +1036,7 @@ struct st7571_panel_data st7567_config = {
 	},
 };
 
-struct st7571_panel_data st7571_config = {
+static const struct st7571_panel_data st7571_config = {
 	.init = st7571_lcd_init,
 	.parse_dt = st7571_parse_dt,
 	.constraints = {

From d83f1d19c898ac1b54ae64d1c950f5beff801982 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marek.vasut+renesas@mailbox.org>
Date: Wed, 13 Aug 2025 23:08:13 +0200
Subject: [PATCH 0809/1455] drm/rcar-du: dsi: Fix 1/2/3 lane support

Remove fixed PPI lane count setup. The R-Car DSI host is capable
of operating in 1..4 DSI lane mode. Remove the hard-coded 4-lane
configuration from PPI register settings and instead configure
the PPI lane count according to lane count information already
obtained by this driver instance.

Configure TXSETR register to match PPI lane count. The R-Car V4H
Reference Manual R19UH0186EJ0121 Rev.1.21 section 67.2.2.3 Tx Set
Register (TXSETR), field LANECNT description indicates that the
TXSETR register LANECNT bitfield lane count must be configured
such, that it matches lane count configuration in PPISETR register
DLEN bitfield. Make sure the LANECNT and DLEN bitfields are
configured to match.

Fixes: 155358310f01 ("drm: rcar-du: Add R-Car DSI driver")
Cc: stable@vger.kernel.org
Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Reviewed-by: Tomi Valkeinen <tomi.valkeinen+renesas@ideasonboard.com>
Link: https://lore.kernel.org/r/20250813210840.97621-1-marek.vasut+renesas@mailbox.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c      | 5 ++++-
 drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h | 8 ++++----
 2 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
index 1af4c73f7a88..952c3efb74da 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
@@ -576,7 +576,10 @@ static int rcar_mipi_dsi_startup(struct rcar_mipi_dsi *dsi,
 	udelay(10);
 	rcar_mipi_dsi_clr(dsi, CLOCKSET1, CLOCKSET1_UPDATEPLL);
 
-	ppisetr = PPISETR_DLEN_3 | PPISETR_CLEN;
+	rcar_mipi_dsi_clr(dsi, TXSETR, TXSETR_LANECNT_MASK);
+	rcar_mipi_dsi_set(dsi, TXSETR, dsi->lanes - 1);
+
+	ppisetr = ((BIT(dsi->lanes) - 1) & PPISETR_DLEN_MASK) | PPISETR_CLEN;
 	rcar_mipi_dsi_write(dsi, PPISETR, ppisetr);
 
 	rcar_mipi_dsi_set(dsi, PHYSETUP, PHYSETUP_SHUTDOWNZ);
diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
index a6b276f1d6ee..a54c7eb4113b 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
@@ -12,6 +12,9 @@
 #define LINKSR_LPBUSY			(1 << 1)
 #define LINKSR_HSBUSY			(1 << 0)
 
+#define TXSETR				0x100
+#define TXSETR_LANECNT_MASK		(0x3 << 0)
+
 /*
  * Video Mode Register
  */
@@ -80,10 +83,7 @@
  * PHY-Protocol Interface (PPI) Registers
  */
 #define PPISETR				0x700
-#define PPISETR_DLEN_0			(0x1 << 0)
-#define PPISETR_DLEN_1			(0x3 << 0)
-#define PPISETR_DLEN_2			(0x7 << 0)
-#define PPISETR_DLEN_3			(0xf << 0)
+#define PPISETR_DLEN_MASK		(0xf << 0)
 #define PPISETR_CLEN			(1 << 8)
 
 #define PPICLCR				0x710

From c4d24127aaf1c50753c8bcdf050ed39b7c709869 Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Thu, 14 Aug 2025 13:35:18 +0200
Subject: [PATCH 0810/1455] accel/rocket: Fix indentation of Kconfig entry

The general indentation for the Kconfig lines is one tab, so adapt the
lines accordingly.

The description is correctly indented (1 tab + 2 spaces) so doesn't need
changes.

Fixes: ed98261b4168 ("accel/rocket: Add a new driver for Rockchip's NPU")
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250814113519.1551855-2-heiko@sntech.de
---
 drivers/accel/rocket/Kconfig | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/accel/rocket/Kconfig b/drivers/accel/rocket/Kconfig
index 43d6cd98ec8e..af6cda273892 100644
--- a/drivers/accel/rocket/Kconfig
+++ b/drivers/accel/rocket/Kconfig
@@ -1,14 +1,14 @@
 # SPDX-License-Identifier: GPL-2.0-only
 
 config DRM_ACCEL_ROCKET
-       tristate "Rocket (support for Rockchip NPUs)"
-       depends on DRM
-       depends on (ARCH_ROCKCHIP && ARM64) || COMPILE_TEST
-       depends on ROCKCHIP_IOMMU || COMPILE_TEST
-       depends on MMU
-       select DRM_SCHED
-       select DRM_GEM_SHMEM_HELPER
-       help
+	tristate "Rocket (support for Rockchip NPUs)"
+	depends on DRM
+	depends on (ARCH_ROCKCHIP && ARM64) || COMPILE_TEST
+	depends on ROCKCHIP_IOMMU || COMPILE_TEST
+	depends on MMU
+	select DRM_SCHED
+	select DRM_GEM_SHMEM_HELPER
+	help
 	  Choose this option if you have a Rockchip SoC that contains a
 	  compatible Neural Processing Unit (NPU), such as the RK3588. Called by
 	  Rockchip either RKNN or RKNPU, it accelerates inference of neural

From 503b0ed38ca60cfb5ba101517a680beebd38d377 Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Thu, 14 Aug 2025 13:35:19 +0200
Subject: [PATCH 0811/1455] accel/rocket: Depend on DRM_ACCEL not just DRM

With the current dependency on only DRM, a config of

CONFIG_DRM_ACCEL_ROCKET=y

is possible, but of course wrong, because without DRM_ACCEL the build-
system will never even enter drivers/accel/* .

So depend on DRM_ACCEL instead of just DRM.

Fixes: ed98261b4168 ("accel/rocket: Add a new driver for Rockchip's NPU")
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250814113519.1551855-3-heiko@sntech.de
---
 drivers/accel/rocket/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/rocket/Kconfig b/drivers/accel/rocket/Kconfig
index af6cda273892..16465abe0660 100644
--- a/drivers/accel/rocket/Kconfig
+++ b/drivers/accel/rocket/Kconfig
@@ -2,7 +2,7 @@
 
 config DRM_ACCEL_ROCKET
 	tristate "Rocket (support for Rockchip NPUs)"
-	depends on DRM
+	depends on DRM_ACCEL
 	depends on (ARCH_ROCKCHIP && ARM64) || COMPILE_TEST
 	depends on ROCKCHIP_IOMMU || COMPILE_TEST
 	depends on MMU

From ce6b656b5d64d612e9fc1c495eba6a362051c2ca Mon Sep 17 00:00:00 2001
From: Brigham Campbell <me@brighamcampbell.com>
Date: Wed, 13 Aug 2025 10:02:37 -0600
Subject: [PATCH 0812/1455] accel/rocket: Fix usages of kfree() and sizeof()

Replace usages of kfree() with kvfree() for pointers which were
allocated using kvmalloc(), as required by the kernel memory management
API.

Use sizeof() on the type that a pointer references instead of the
pointer itself. In this case, scheds and *scheds both happen to be
pointers, so sizeof() will expand to the same value in either case, but
using *scheds is more technically correct since scheds is an array of
drm_gpu_scheduler *.

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Julia Lawall <julia.lawall@inria.fr>
Closes: https://lore.kernel.org/r/202508120730.PLbjlKbI-lkp@intel.com/
Signed-off-by: Brigham Campbell <me@brighamcampbell.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250813-rocket-free-fix-v1-1-51f00a7a1271@brighamcampbell.com
Fixes: 0810d5ad88a1 ("accel/rocket: Add job submission IOCTL")
---
 drivers/accel/rocket/rocket_job.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/accel/rocket/rocket_job.c b/drivers/accel/rocket/rocket_job.c
index 5d4afd692306..f6fe1a6d9264 100644
--- a/drivers/accel/rocket/rocket_job.c
+++ b/drivers/accel/rocket/rocket_job.c
@@ -222,7 +222,7 @@ static int rocket_job_push(struct rocket_job *job)
 err_unlock:
 	drm_gem_unlock_reservations(bos, job->in_bo_count + job->out_bo_count, &acquire_ctx);
 err:
-	kfree(bos);
+	kvfree(bos);
 
 	return ret;
 }
@@ -496,7 +496,8 @@ void rocket_job_fini(struct rocket_core *core)
 int rocket_job_open(struct rocket_file_priv *rocket_priv)
 {
 	struct rocket_device *rdev = rocket_priv->rdev;
-	struct drm_gpu_scheduler **scheds = kmalloc_array(rdev->num_cores, sizeof(scheds),
+	struct drm_gpu_scheduler **scheds = kmalloc_array(rdev->num_cores,
+							  sizeof(*scheds),
 							  GFP_KERNEL);
 	unsigned int core;
 	int ret;
@@ -630,7 +631,7 @@ int rocket_ioctl_submit(struct drm_device *dev, void *data, struct drm_file *fil
 		rocket_ioctl_submit_job(dev, file, &jobs[i]);
 
 exit:
-	kfree(jobs);
+	kvfree(jobs);
 
 	return ret;
 }

From 78e399955d352a2387a4b22713385a0c5d737685 Mon Sep 17 00:00:00 2001
From: Heiko Stuebner <heiko@sntech.de>
Date: Mon, 18 Aug 2025 20:56:58 +0200
Subject: [PATCH 0813/1455] accel/rocket: Check the correct DMA irq status to
 warn about

Right now, the code checks the DMA_READ_ERROR state 2 times, while
I guess it was supposed to warn about both read and write errors.

Change the 2nd check to look at the write-error flag.

Fixes: 0810d5ad88a1 ("accel/rocket: Add job submission IOCTL")
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250818185658.2585696-1-heiko@sntech.de
---
 drivers/accel/rocket/rocket_job.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/rocket/rocket_job.c b/drivers/accel/rocket/rocket_job.c
index f6fe1a6d9264..acd606160dc9 100644
--- a/drivers/accel/rocket/rocket_job.c
+++ b/drivers/accel/rocket/rocket_job.c
@@ -422,7 +422,7 @@ static irqreturn_t rocket_job_irq_handler(int irq, void *data)
 	u32 raw_status = rocket_pc_readl(core, INTERRUPT_RAW_STATUS);
 
 	WARN_ON(raw_status & PC_INTERRUPT_RAW_STATUS_DMA_READ_ERROR);
-	WARN_ON(raw_status & PC_INTERRUPT_RAW_STATUS_DMA_READ_ERROR);
+	WARN_ON(raw_status & PC_INTERRUPT_RAW_STATUS_DMA_WRITE_ERROR);
 
 	if (!(raw_status & PC_INTERRUPT_RAW_STATUS_DPU_0 ||
 	      raw_status & PC_INTERRUPT_RAW_STATUS_DPU_1))

From 09e6d39cc83b18a17857b8d4d6a8902bc87dfbc2 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 21 Aug 2025 15:30:19 +0300
Subject: [PATCH 0814/1455] accel/rocket: Fix some error checking in
 rocket_core_init()

The problem is that pm_runtime_get_sync() can return 1 on success so
checking for zero doesn't work.  Use the pm_runtime_resume_and_get()
function instead.  The pm_runtime_resume_and_get() function does
additional cleanup as well so that's a bonus as well.

Fixes: 0810d5ad88a1 ("accel/rocket: Add job submission IOCTL")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/aKcRW6fsRP_o5C_y@stanley.mountain
---
 drivers/accel/rocket/rocket_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/rocket/rocket_core.c b/drivers/accel/rocket/rocket_core.c
index 72fb5e5798fa..abe7719c1db4 100644
--- a/drivers/accel/rocket/rocket_core.c
+++ b/drivers/accel/rocket/rocket_core.c
@@ -74,7 +74,7 @@ int rocket_core_init(struct rocket_core *core)
 
 	pm_runtime_enable(dev);
 
-	err = pm_runtime_get_sync(dev);
+	err = pm_runtime_resume_and_get(dev);
 	if (err) {
 		rocket_job_fini(core);
 		return err;

From c51f6fe14cd7e2e0a0e1094d38805dcbbd04a058 Mon Sep 17 00:00:00 2001
From: Lukas Bulwahn <lukas.bulwahn@redhat.com>
Date: Tue, 26 Aug 2025 08:32:48 +0200
Subject: [PATCH 0815/1455] MAINTAINERS: adjust file entry in DRM ACCEL DRIVER
 FOR ROCKCHIP NPU

Commit a7352c849492 ("dt-bindings: npu: rockchip,rknn: Add bindings") adds
the device-tree binding rockchip,rk3588-rknn-core.yaml, whereas the commit
ed98261b4168 ("accel/rocket: Add a new driver for Rockchip's NPU") adds the
section DRM ACCEL DRIVER FOR ROCKCHIP NPU in MAINTAINERS with a file entry
referring to rockchip,rknn-core.yaml. Note that the file entry is missing
the part rk3588, compared to the added file above, which it intends to
refer to.

Adjust the file entry to the intended file name.

Fixes: ed98261b4168 ("accel/rocket: Add a new driver for Rockchip's NPU")
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@redhat.com>
Signed-off-by: Tomeu Vizoso <tomeu@tomeuvizoso.net>
Link: https://lore.kernel.org/r/20250826063248.32153-1-lukas.bulwahn@redhat.com
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 0ae7d13c2b3c..88851907b672 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7505,7 +7505,7 @@ L:	dri-devel@lists.freedesktop.org
 S:	Supported
 T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/accel/rocket/
-F:	Documentation/devicetree/bindings/npu/rockchip,rknn-core.yaml
+F:	Documentation/devicetree/bindings/npu/rockchip,rk3588-rknn-core.yaml
 F:	drivers/accel/rocket/
 F:	include/uapi/drm/rocket_accel.h
 

From 78f859673c85cc7c18ffdeb659f29cad3f13b580 Mon Sep 17 00:00:00 2001
From: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Date: Fri, 8 Aug 2025 13:10:44 +0200
Subject: [PATCH 0816/1455] accel/ivpu: Remove unused PLL_CONFIG_DEFAULT

This change removes the unnecessary condition, makes the code clearer,
and silences clang-tidy warning.

Reviewed-by: Lizhi Hou <lizhi.hou@amd.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://lore.kernel.org/r/20250808111044.328800-1-jacek.lawrynowicz@linux.intel.com
---
 drivers/accel/ivpu/ivpu_hw_btrs.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c
index b236c7234daa..afdb3b2aa72a 100644
--- a/drivers/accel/ivpu/ivpu_hw_btrs.c
+++ b/drivers/accel/ivpu/ivpu_hw_btrs.c
@@ -33,7 +33,6 @@
 
 #define PLL_CDYN_DEFAULT               0x80
 #define PLL_EPP_DEFAULT                0x80
-#define PLL_CONFIG_DEFAULT             0x0
 #define PLL_REF_CLK_FREQ               50000000ull
 #define PLL_RATIO_TO_FREQ(x)           ((x) * PLL_REF_CLK_FREQ)
 
@@ -303,7 +302,7 @@ static void prepare_wp_request(struct ivpu_device *vdev, struct wp_request *wp,
 		wp->epp = 0;
 	} else {
 		wp->target = hw->pll.pn_ratio;
-		wp->cfg = enable ? PLL_CONFIG_DEFAULT : 0;
+		wp->cfg = 0;
 		wp->cdyn = enable ? PLL_CDYN_DEFAULT : 0;
 		wp->epp = enable ? PLL_EPP_DEFAULT : 0;
 	}

From cf87f93847dea607e8a35983cb006ef8493f8065 Mon Sep 17 00:00:00 2001
From: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Date: Fri, 8 Aug 2025 13:10:14 +0200
Subject: [PATCH 0817/1455] accel/ivpu: Make function parameter names
 consistent

Make ivpu_hw_btrs_dct_set_status() and ivpu_fw_boot_params_setup()
declaration and definition parameter names consistent.

Reviewed-by: Lizhi Hou <lizhi.hou@amd.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://lore.kernel.org/r/20250808111014.328607-1-jacek.lawrynowicz@linux.intel.com
---
 drivers/accel/ivpu/ivpu_fw.h      | 2 +-
 drivers/accel/ivpu/ivpu_hw_btrs.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/ivpu/ivpu_fw.h b/drivers/accel/ivpu/ivpu_fw.h
index 9a3935be1c05..7081913fb0dd 100644
--- a/drivers/accel/ivpu/ivpu_fw.h
+++ b/drivers/accel/ivpu/ivpu_fw.h
@@ -45,7 +45,7 @@ struct ivpu_fw_info {
 int ivpu_fw_init(struct ivpu_device *vdev);
 void ivpu_fw_fini(struct ivpu_device *vdev);
 void ivpu_fw_load(struct ivpu_device *vdev);
-void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params *bp);
+void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params *boot_params);
 
 static inline bool ivpu_fw_is_cold_boot(struct ivpu_device *vdev)
 {
diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.h b/drivers/accel/ivpu/ivpu_hw_btrs.h
index d2d82651976d..032c384ac3d4 100644
--- a/drivers/accel/ivpu/ivpu_hw_btrs.h
+++ b/drivers/accel/ivpu/ivpu_hw_btrs.h
@@ -36,7 +36,7 @@ u32 ivpu_hw_btrs_dpu_freq_get(struct ivpu_device *vdev);
 bool ivpu_hw_btrs_irq_handler_mtl(struct ivpu_device *vdev, int irq);
 bool ivpu_hw_btrs_irq_handler_lnl(struct ivpu_device *vdev, int irq);
 int ivpu_hw_btrs_dct_get_request(struct ivpu_device *vdev, bool *enable);
-void ivpu_hw_btrs_dct_set_status(struct ivpu_device *vdev, bool enable, u32 dct_percent);
+void ivpu_hw_btrs_dct_set_status(struct ivpu_device *vdev, bool enable, u32 active_percent);
 u32 ivpu_hw_btrs_telemetry_offset_get(struct ivpu_device *vdev);
 u32 ivpu_hw_btrs_telemetry_size_get(struct ivpu_device *vdev);
 u32 ivpu_hw_btrs_telemetry_enable_get(struct ivpu_device *vdev);

From e38e8391f30b41c5a24bb46dc6ef4161921e782d Mon Sep 17 00:00:00 2001
From: Michal Wilczynski <m.wilczynski@samsung.com>
Date: Fri, 22 Aug 2025 00:20:15 +0200
Subject: [PATCH 0818/1455] drm/imagination: Use pwrseq for TH1520 GPU power
 management

Update the Imagination PVR DRM driver to leverage the pwrseq framework
for managing the complex power sequence of the GPU on the T-HEAD TH1520
SoC.

To cleanly separate platform-specific logic from the generic driver,
this patch introduces an `init` callback to the `pwr_power_sequence_ops`
struct. This allows for different power management strategies to be
selected at probe time based on the device's compatible string.

A `pvr_device_data` struct, associated with each compatible in the
of_device_id table, points to the appropriate ops table (manual or
pwrseq).

At probe time, the driver now calls the `->init()` op. For pwrseq-based
platforms, this callback calls `devm_pwrseq_get("gpu-power")`, deferring
probe if the sequencer is not yet available. For other platforms, it
falls back to the existing manual clock and reset handling. The runtime
PM callbacks continue to call the appropriate functions via the ops
table.

Signed-off-by: Michal Wilczynski <m.wilczynski@samsung.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Matt Coster <matt.coster@imgtec.com>
Link: https://lore.kernel.org/r/20250822-apr_14_for_sending-v13-1-af656f7cc6c3@samsung.com
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
---
 drivers/gpu/drm/imagination/pvr_device.c |  22 +--
 drivers/gpu/drm/imagination/pvr_device.h |  17 +++
 drivers/gpu/drm/imagination/pvr_drv.c    |  23 +++-
 drivers/gpu/drm/imagination/pvr_power.c  | 168 +++++++++++++++++------
 drivers/gpu/drm/imagination/pvr_power.h  |  15 ++
 5 files changed, 181 insertions(+), 64 deletions(-)

diff --git a/drivers/gpu/drm/imagination/pvr_device.c b/drivers/gpu/drm/imagination/pvr_device.c
index 8b9ba4983c4c..294b6019b415 100644
--- a/drivers/gpu/drm/imagination/pvr_device.c
+++ b/drivers/gpu/drm/imagination/pvr_device.c
@@ -23,6 +23,7 @@
 #include <linux/firmware.h>
 #include <linux/gfp.h>
 #include <linux/interrupt.h>
+#include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/reset.h>
@@ -121,21 +122,6 @@ static int pvr_device_clk_init(struct pvr_device *pvr_dev)
 	return 0;
 }
 
-static int pvr_device_reset_init(struct pvr_device *pvr_dev)
-{
-	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
-	struct reset_control *reset;
-
-	reset = devm_reset_control_get_optional_exclusive(drm_dev->dev, NULL);
-	if (IS_ERR(reset))
-		return dev_err_probe(drm_dev->dev, PTR_ERR(reset),
-				     "failed to get gpu reset line\n");
-
-	pvr_dev->reset = reset;
-
-	return 0;
-}
-
 /**
  * pvr_device_process_active_queues() - Process all queue related events.
  * @pvr_dev: PowerVR device to check
@@ -618,6 +604,9 @@ pvr_device_init(struct pvr_device *pvr_dev)
 	struct device *dev = drm_dev->dev;
 	int err;
 
+	/* Get the platform-specific data based on the compatible string. */
+	pvr_dev->device_data = of_device_get_match_data(dev);
+
 	/*
 	 * Setup device parameters. We do this first in case other steps
 	 * depend on them.
@@ -631,8 +620,7 @@ pvr_device_init(struct pvr_device *pvr_dev)
 	if (err)
 		return err;
 
-	/* Get the reset line for the GPU */
-	err = pvr_device_reset_init(pvr_dev);
+	err = pvr_dev->device_data->pwr_ops->init(pvr_dev);
 	if (err)
 		return err;
 
diff --git a/drivers/gpu/drm/imagination/pvr_device.h b/drivers/gpu/drm/imagination/pvr_device.h
index 7cb01c38d2a9..ab8f56ae15df 100644
--- a/drivers/gpu/drm/imagination/pvr_device.h
+++ b/drivers/gpu/drm/imagination/pvr_device.h
@@ -37,6 +37,9 @@ struct clk;
 /* Forward declaration from <linux/firmware.h>. */
 struct firmware;
 
+/* Forward declaration from <linux/pwrseq/consumer.h> */
+struct pwrseq_desc;
+
 /**
  * struct pvr_gpu_id - Hardware GPU ID information for a PowerVR device
  * @b: Branch ID.
@@ -57,6 +60,14 @@ struct pvr_fw_version {
 	u16 major, minor;
 };
 
+/**
+ * struct pvr_device_data - Platform specific data associated with a compatible string.
+ * @pwr_ops: Pointer to a structure with platform-specific power functions.
+ */
+struct pvr_device_data {
+	const struct pvr_power_sequence_ops *pwr_ops;
+};
+
 /**
  * struct pvr_device - powervr-specific wrapper for &struct drm_device
  */
@@ -98,6 +109,9 @@ struct pvr_device {
 	/** @fw_version: Firmware version detected at runtime. */
 	struct pvr_fw_version fw_version;
 
+	/** @device_data: Pointer to platform-specific data. */
+	const struct pvr_device_data *device_data;
+
 	/** @regs_resource: Resource representing device control registers. */
 	struct resource *regs_resource;
 
@@ -148,6 +162,9 @@ struct pvr_device {
 	 */
 	struct reset_control *reset;
 
+	/** @pwrseq: Pointer to a power sequencer, if one is used. */
+	struct pwrseq_desc *pwrseq;
+
 	/** @irq: IRQ number. */
 	int irq;
 
diff --git a/drivers/gpu/drm/imagination/pvr_drv.c b/drivers/gpu/drm/imagination/pvr_drv.c
index b058ec183bb3..916b40ced7eb 100644
--- a/drivers/gpu/drm/imagination/pvr_drv.c
+++ b/drivers/gpu/drm/imagination/pvr_drv.c
@@ -1480,15 +1480,33 @@ static void pvr_remove(struct platform_device *plat_dev)
 	pvr_power_domains_fini(pvr_dev);
 }
 
+static const struct pvr_device_data pvr_device_data_manual = {
+	.pwr_ops = &pvr_power_sequence_ops_manual,
+};
+
+static const struct pvr_device_data pvr_device_data_pwrseq = {
+	.pwr_ops = &pvr_power_sequence_ops_pwrseq,
+};
+
 static const struct of_device_id dt_match[] = {
-	{ .compatible = "img,img-rogue", .data = NULL },
+	{
+		.compatible = "thead,th1520-gpu",
+		.data = &pvr_device_data_pwrseq,
+	},
+	{
+		.compatible = "img,img-rogue",
+		.data = &pvr_device_data_manual,
+	},
 
 	/*
 	 * This legacy compatible string was introduced early on before the more generic
 	 * "img,img-rogue" was added. Keep it around here for compatibility, but never use
 	 * "img,img-axe" in new devicetrees.
 	 */
-	{ .compatible = "img,img-axe", .data = NULL },
+	{
+		.compatible = "img,img-axe",
+		.data = &pvr_device_data_manual,
+	},
 	{}
 };
 MODULE_DEVICE_TABLE(of, dt_match);
@@ -1513,4 +1531,5 @@ MODULE_DESCRIPTION(PVR_DRIVER_DESC);
 MODULE_LICENSE("Dual MIT/GPL");
 MODULE_IMPORT_NS("DMA_BUF");
 MODULE_FIRMWARE("powervr/rogue_33.15.11.3_v1.fw");
+MODULE_FIRMWARE("powervr/rogue_36.52.104.182_v1.fw");
 MODULE_FIRMWARE("powervr/rogue_36.53.104.796_v1.fw");
diff --git a/drivers/gpu/drm/imagination/pvr_power.c b/drivers/gpu/drm/imagination/pvr_power.c
index 187a07e0bd9a..c6e7ff9e935d 100644
--- a/drivers/gpu/drm/imagination/pvr_power.c
+++ b/drivers/gpu/drm/imagination/pvr_power.c
@@ -18,6 +18,7 @@
 #include <linux/platform_device.h>
 #include <linux/pm_domain.h>
 #include <linux/pm_runtime.h>
+#include <linux/pwrseq/consumer.h>
 #include <linux/reset.h>
 #include <linux/timer.h>
 #include <linux/types.h>
@@ -234,51 +235,28 @@ pvr_watchdog_init(struct pvr_device *pvr_dev)
 	return 0;
 }
 
-int
-pvr_power_device_suspend(struct device *dev)
+static int pvr_power_init_manual(struct pvr_device *pvr_dev)
 {
-	struct platform_device *plat_dev = to_platform_device(dev);
-	struct drm_device *drm_dev = platform_get_drvdata(plat_dev);
-	struct pvr_device *pvr_dev = to_pvr_device(drm_dev);
-	int err = 0;
-	int idx;
+	struct drm_device *drm_dev = from_pvr_device(pvr_dev);
+	struct reset_control *reset;
 
-	if (!drm_dev_enter(drm_dev, &idx))
-		return -EIO;
+	reset = devm_reset_control_get_optional_exclusive(drm_dev->dev, NULL);
+	if (IS_ERR(reset))
+		return dev_err_probe(drm_dev->dev, PTR_ERR(reset),
+				     "failed to get gpu reset line\n");
 
-	if (pvr_dev->fw_dev.booted) {
-		err = pvr_power_fw_disable(pvr_dev, false);
-		if (err)
-			goto err_drm_dev_exit;
-	}
+	pvr_dev->reset = reset;
 
-	clk_disable_unprepare(pvr_dev->mem_clk);
-	clk_disable_unprepare(pvr_dev->sys_clk);
-	clk_disable_unprepare(pvr_dev->core_clk);
-
-	err = reset_control_assert(pvr_dev->reset);
-
-err_drm_dev_exit:
-	drm_dev_exit(idx);
-
-	return err;
+	return 0;
 }
 
-int
-pvr_power_device_resume(struct device *dev)
+static int pvr_power_on_sequence_manual(struct pvr_device *pvr_dev)
 {
-	struct platform_device *plat_dev = to_platform_device(dev);
-	struct drm_device *drm_dev = platform_get_drvdata(plat_dev);
-	struct pvr_device *pvr_dev = to_pvr_device(drm_dev);
-	int idx;
 	int err;
 
-	if (!drm_dev_enter(drm_dev, &idx))
-		return -EIO;
-
 	err = clk_prepare_enable(pvr_dev->core_clk);
 	if (err)
-		goto err_drm_dev_exit;
+		return err;
 
 	err = clk_prepare_enable(pvr_dev->sys_clk);
 	if (err)
@@ -302,19 +280,8 @@ pvr_power_device_resume(struct device *dev)
 	if (err)
 		goto err_mem_clk_disable;
 
-	if (pvr_dev->fw_dev.booted) {
-		err = pvr_power_fw_enable(pvr_dev);
-		if (err)
-			goto err_reset_assert;
-	}
-
-	drm_dev_exit(idx);
-
 	return 0;
 
-err_reset_assert:
-	reset_control_assert(pvr_dev->reset);
-
 err_mem_clk_disable:
 	clk_disable_unprepare(pvr_dev->mem_clk);
 
@@ -324,6 +291,117 @@ err_sys_clk_disable:
 err_core_clk_disable:
 	clk_disable_unprepare(pvr_dev->core_clk);
 
+	return err;
+}
+
+static int pvr_power_off_sequence_manual(struct pvr_device *pvr_dev)
+{
+	int err;
+
+	err = reset_control_assert(pvr_dev->reset);
+
+	clk_disable_unprepare(pvr_dev->mem_clk);
+	clk_disable_unprepare(pvr_dev->sys_clk);
+	clk_disable_unprepare(pvr_dev->core_clk);
+
+	return err;
+}
+
+const struct pvr_power_sequence_ops pvr_power_sequence_ops_manual = {
+	.init = pvr_power_init_manual,
+	.power_on = pvr_power_on_sequence_manual,
+	.power_off = pvr_power_off_sequence_manual,
+};
+
+static int pvr_power_init_pwrseq(struct pvr_device *pvr_dev)
+{
+	struct device *dev = from_pvr_device(pvr_dev)->dev;
+
+	pvr_dev->pwrseq = devm_pwrseq_get(dev, "gpu-power");
+	if (IS_ERR(pvr_dev->pwrseq)) {
+		/*
+		 * This platform requires a sequencer. If we can't get it, we
+		 * must return the error (including -EPROBE_DEFER to wait for
+		 * the provider to appear)
+		 */
+		return dev_err_probe(dev, PTR_ERR(pvr_dev->pwrseq),
+				     "Failed to get required power sequencer\n");
+	}
+
+	return 0;
+}
+
+static int pvr_power_on_sequence_pwrseq(struct pvr_device *pvr_dev)
+{
+	return pwrseq_power_on(pvr_dev->pwrseq);
+}
+
+static int pvr_power_off_sequence_pwrseq(struct pvr_device *pvr_dev)
+{
+	return pwrseq_power_off(pvr_dev->pwrseq);
+}
+
+const struct pvr_power_sequence_ops pvr_power_sequence_ops_pwrseq = {
+	.init = pvr_power_init_pwrseq,
+	.power_on = pvr_power_on_sequence_pwrseq,
+	.power_off = pvr_power_off_sequence_pwrseq,
+};
+
+int
+pvr_power_device_suspend(struct device *dev)
+{
+	struct platform_device *plat_dev = to_platform_device(dev);
+	struct drm_device *drm_dev = platform_get_drvdata(plat_dev);
+	struct pvr_device *pvr_dev = to_pvr_device(drm_dev);
+	int err = 0;
+	int idx;
+
+	if (!drm_dev_enter(drm_dev, &idx))
+		return -EIO;
+
+	if (pvr_dev->fw_dev.booted) {
+		err = pvr_power_fw_disable(pvr_dev, false);
+		if (err)
+			goto err_drm_dev_exit;
+	}
+
+	err = pvr_dev->device_data->pwr_ops->power_off(pvr_dev);
+
+err_drm_dev_exit:
+	drm_dev_exit(idx);
+
+	return err;
+}
+
+int
+pvr_power_device_resume(struct device *dev)
+{
+	struct platform_device *plat_dev = to_platform_device(dev);
+	struct drm_device *drm_dev = platform_get_drvdata(plat_dev);
+	struct pvr_device *pvr_dev = to_pvr_device(drm_dev);
+	int idx;
+	int err;
+
+	if (!drm_dev_enter(drm_dev, &idx))
+		return -EIO;
+
+	err = pvr_dev->device_data->pwr_ops->power_on(pvr_dev);
+	if (err)
+		goto err_drm_dev_exit;
+
+	if (pvr_dev->fw_dev.booted) {
+		err = pvr_power_fw_enable(pvr_dev);
+		if (err)
+			goto err_power_off;
+	}
+
+	drm_dev_exit(idx);
+
+	return 0;
+
+err_power_off:
+	pvr_dev->device_data->pwr_ops->power_off(pvr_dev);
+
 err_drm_dev_exit:
 	drm_dev_exit(idx);
 
diff --git a/drivers/gpu/drm/imagination/pvr_power.h b/drivers/gpu/drm/imagination/pvr_power.h
index ada85674a7ca..b853d092242c 100644
--- a/drivers/gpu/drm/imagination/pvr_power.h
+++ b/drivers/gpu/drm/imagination/pvr_power.h
@@ -41,4 +41,19 @@ pvr_power_put(struct pvr_device *pvr_dev)
 int pvr_power_domains_init(struct pvr_device *pvr_dev);
 void pvr_power_domains_fini(struct pvr_device *pvr_dev);
 
+/**
+ * struct pvr_power_sequence_ops - Platform specific power sequence operations.
+ * @init: Pointer to the platform-specific initialization function.
+ * @power_on: Pointer to the platform-specific power on function.
+ * @power_off: Pointer to the platform-specific power off function.
+ */
+struct pvr_power_sequence_ops {
+	int (*init)(struct pvr_device *pvr_dev);
+	int (*power_on)(struct pvr_device *pvr_dev);
+	int (*power_off)(struct pvr_device *pvr_dev);
+};
+
+extern const struct pvr_power_sequence_ops pvr_power_sequence_ops_manual;
+extern const struct pvr_power_sequence_ops pvr_power_sequence_ops_pwrseq;
+
 #endif /* PVR_POWER_H */

From 337ebfda8a4f2627bf52e200cacf6f3a2f5ccf48 Mon Sep 17 00:00:00 2001
From: Michal Wilczynski <m.wilczynski@samsung.com>
Date: Fri, 22 Aug 2025 00:20:16 +0200
Subject: [PATCH 0819/1455] dt-bindings: gpu: img,powervr-rogue: Add TH1520 GPU
 support

Rework the PowerVR Rogue GPU binding to use an explicit, per variant
style for defining power domain properties and add support for the
T-HEAD TH1520 SoC's GPU.

To improve clarity and precision, the binding is refactored so that
power domain items are listed explicitly for each variant [1]. The
previous method relied on an implicit, positional mapping between the
`power-domains` and `power-domain-names` properties. This change
replaces the generic rules with self contained if/then blocks for each
GPU variant, making the relationship between power domains and their
names explicit and unambiguous.

The generic if block for img,img-rogue, which previously required
power-domains and power-domain-names for all variants, is removed.
Instead, each specific GPU variant now defines its own power domain
requirements within a self-contained if/then block, making the schema
more explicit.

This new structure is then used to add support for the
`thead,th1520-gpu`. While its BXM-4-64 IP has two conceptual power
domains, the TH1520 SoC integrates them behind a single power gate. The
new binding models this with a specific rule that enforces a single
`power-domains` entry and disallows the `power-domain-names` property.

Link: https://lore.kernel.org/all/4d79c8dd-c5fb-442c-ac65-37e7176b0cdd@linaro.org/ [1]

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Matt Coster <matt.coster@imgtec.com>
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Michal Wilczynski <m.wilczynski@samsung.com>
Link: https://lore.kernel.org/r/20250822-apr_14_for_sending-v13-2-af656f7cc6c3@samsung.com
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
---
 .../bindings/gpu/img,powervr-rogue.yaml       | 49 +++++++++++++------
 1 file changed, 35 insertions(+), 14 deletions(-)

diff --git a/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml b/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
index 4450e2e73b3c..c87d7bece0ec 100644
--- a/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
+++ b/Documentation/devicetree/bindings/gpu/img,powervr-rogue.yaml
@@ -21,6 +21,11 @@ properties:
           # work with newer dts.
           - const: img,img-axe
           - const: img,img-rogue
+      - items:
+          - enum:
+              - thead,th1520-gpu
+          - const: img,img-bxm-4-64
+          - const: img,img-rogue
       - items:
           - enum:
               - ti,j721s2-gpu
@@ -77,18 +82,6 @@ required:
 additionalProperties: false
 
 allOf:
-  # Constraints added alongside the new compatible strings that would otherwise
-  # create an ABI break.
-  - if:
-      properties:
-        compatible:
-          contains:
-            const: img,img-rogue
-    then:
-      required:
-        - power-domains
-        - power-domain-names
-
   - if:
       properties:
         compatible:
@@ -97,9 +90,32 @@ allOf:
     then:
       properties:
         power-domains:
-          maxItems: 1
+          items:
+            - description: Power domain A
         power-domain-names:
           maxItems: 1
+      required:
+        - power-domains
+        - power-domain-names
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: thead,th1520-gpu
+    then:
+      properties:
+        clocks:
+          minItems: 3
+        clock-names:
+          minItems: 3
+        power-domains:
+          items:
+            - description: The single, unified power domain for the GPU on the
+                TH1520 SoC, integrating all internal IP power domains.
+        power-domain-names: false
+      required:
+        - power-domains
 
   - if:
       properties:
@@ -109,9 +125,14 @@ allOf:
     then:
       properties:
         power-domains:
-          minItems: 2
+          items:
+            - description: Power domain A
+            - description: Power domain B
         power-domain-names:
           minItems: 2
+      required:
+        - power-domains
+        - power-domain-names
 
   - if:
       properties:

From 6b53cf48d9339c75fa51927b0a67d8a6751066bd Mon Sep 17 00:00:00 2001
From: Michal Wilczynski <m.wilczynski@samsung.com>
Date: Fri, 22 Aug 2025 00:20:18 +0200
Subject: [PATCH 0820/1455] drm/imagination: Enable PowerVR driver for RISC-V

Several RISC-V boards feature Imagination GPUs that are compatible with
the PowerVR driver. An example is the IMG BXM-4-64 GPU on the Lichee Pi
4A board. This commit adjusts the driver's Kconfig dependencies to allow
the PowerVR driver to be compiled on the RISC-V architecture.

By enabling compilation on RISC-V, we expand support for these GPUs,
providing graphics acceleration capabilities and enhancing hardware
compatibility on RISC-V platforms.

The RISC-V support is restricted to 64-bit systems (RISCV && 64BIT) as
the driver currently has an implicit dependency on a 64-bit platform.

Add a dependency on MMU to fix a build warning on RISC-V configurations
without an MMU.

Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Reviewed-by: Matt Coster <matt.coster@imgtec.com>
Signed-off-by: Michal Wilczynski <m.wilczynski@samsung.com>
Link: https://lore.kernel.org/r/20250822-apr_14_for_sending-v13-4-af656f7cc6c3@samsung.com
Signed-off-by: Matt Coster <matt.coster@imgtec.com>
---
 drivers/gpu/drm/imagination/Kconfig | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/imagination/Kconfig b/drivers/gpu/drm/imagination/Kconfig
index 3bfa2ac212dc..682dd2633d0c 100644
--- a/drivers/gpu/drm/imagination/Kconfig
+++ b/drivers/gpu/drm/imagination/Kconfig
@@ -3,8 +3,9 @@
 
 config DRM_POWERVR
 	tristate "Imagination Technologies PowerVR (Series 6 and later) & IMG Graphics"
-	depends on ARM64
+	depends on (ARM64 || RISCV && 64BIT)
 	depends on DRM
+	depends on MMU
 	depends on PM
 	select DRM_EXEC
 	select DRM_GEM_SHMEM_HELPER

From dff11511d1967b9fd82a5b4b60cf5045eaa545b6 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 8 Aug 2025 11:46:41 +0900
Subject: [PATCH 0821/1455] gpu: nova-core: vbios: replace pci::Device with
 device::Device

The passed pci::Device is exclusively used for logging purposes, so it
can be replaced by a regular device::Device, which allows us to remove
the `as_ref()` indirections at each logging site.

Reviewed-by: Joel Fernandes <joelagnelf@nvidia.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250808-vbios_device-v1-1-834bbbab6471@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/gpu.rs   |   2 +-
 drivers/gpu/nova-core/vbios.rs | 135 ++++++++++++++-------------------
 2 files changed, 57 insertions(+), 80 deletions(-)

diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index 8caecaf7dfb4..4d0f759610ec 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -298,7 +298,7 @@ impl Gpu {
         let fb_layout = FbLayout::new(spec.chipset, bar)?;
         dev_dbg!(pdev.as_ref(), "{:#x?}\n", fb_layout);
 
-        let bios = Vbios::new(pdev, bar)?;
+        let bios = Vbios::new(pdev.as_ref(), bar)?;
 
         Self::run_fwsec_frts(pdev.as_ref(), &gsp_falcon, bar, &bios, &fb_layout)?;
 
diff --git a/drivers/gpu/nova-core/vbios.rs b/drivers/gpu/nova-core/vbios.rs
index 50fbea69fc00..ce32ab4cd5b2 100644
--- a/drivers/gpu/nova-core/vbios.rs
+++ b/drivers/gpu/nova-core/vbios.rs
@@ -8,7 +8,6 @@ use crate::firmware::FalconUCodeDescV3;
 use core::convert::TryFrom;
 use kernel::device;
 use kernel::error::Result;
-use kernel::pci;
 use kernel::prelude::*;
 
 /// The offset of the VBIOS ROM in the BAR0 space.
@@ -31,7 +30,7 @@ const FALCON_UCODE_ENTRY_APPID_FWSEC_PROD: u8 = 0x85;
 
 /// Vbios Reader for constructing the VBIOS data.
 struct VbiosIterator<'a> {
-    pdev: &'a pci::Device,
+    dev: &'a device::Device,
     bar0: &'a Bar0,
     /// VBIOS data vector: As BIOS images are scanned, they are added to this vector for reference
     /// or copying into other data structures. It is the entire scanned contents of the VBIOS which
@@ -46,9 +45,9 @@ struct VbiosIterator<'a> {
 }
 
 impl<'a> VbiosIterator<'a> {
-    fn new(pdev: &'a pci::Device, bar0: &'a Bar0) -> Result<Self> {
+    fn new(dev: &'a device::Device, bar0: &'a Bar0) -> Result<Self> {
         Ok(Self {
-            pdev,
+            dev,
             bar0,
             data: KVec::new(),
             current_offset: 0,
@@ -64,7 +63,7 @@ impl<'a> VbiosIterator<'a> {
         // Ensure length is a multiple of 4 for 32-bit reads
         if len % core::mem::size_of::<u32>() != 0 {
             dev_err!(
-                self.pdev.as_ref(),
+                self.dev,
                 "VBIOS read length {} is not a multiple of 4\n",
                 len
             );
@@ -89,7 +88,7 @@ impl<'a> VbiosIterator<'a> {
     /// Read bytes at a specific offset, filling any gap.
     fn read_more_at_offset(&mut self, offset: usize, len: usize) -> Result {
         if offset > BIOS_MAX_SCAN_LEN {
-            dev_err!(self.pdev.as_ref(), "Error: exceeded BIOS scan limit.\n");
+            dev_err!(self.dev, "Error: exceeded BIOS scan limit.\n");
             return Err(EINVAL);
         }
 
@@ -115,7 +114,7 @@ impl<'a> VbiosIterator<'a> {
         if offset + len > data_len {
             self.read_more_at_offset(offset, len).inspect_err(|e| {
                 dev_err!(
-                    self.pdev.as_ref(),
+                    self.dev,
                     "Failed to read more at offset {:#x}: {:?}\n",
                     offset,
                     e
@@ -123,9 +122,9 @@ impl<'a> VbiosIterator<'a> {
             })?;
         }
 
-        BiosImage::new(self.pdev, &self.data[offset..offset + len]).inspect_err(|err| {
+        BiosImage::new(self.dev, &self.data[offset..offset + len]).inspect_err(|err| {
             dev_err!(
-                self.pdev.as_ref(),
+                self.dev,
                 "Failed to {} at offset {:#x}: {:?}\n",
                 context,
                 offset,
@@ -146,10 +145,7 @@ impl<'a> Iterator for VbiosIterator<'a> {
         }
 
         if self.current_offset > BIOS_MAX_SCAN_LEN {
-            dev_err!(
-                self.pdev.as_ref(),
-                "Error: exceeded BIOS scan limit, stopping scan\n"
-            );
+            dev_err!(self.dev, "Error: exceeded BIOS scan limit, stopping scan\n");
             return None;
         }
 
@@ -192,18 +188,18 @@ impl Vbios {
     /// Probe for VBIOS extraction.
     ///
     /// Once the VBIOS object is built, `bar0` is not read for [`Vbios`] purposes anymore.
-    pub(crate) fn new(pdev: &pci::Device, bar0: &Bar0) -> Result<Vbios> {
+    pub(crate) fn new(dev: &device::Device, bar0: &Bar0) -> Result<Vbios> {
         // Images to extract from iteration
         let mut pci_at_image: Option<PciAtBiosImage> = None;
         let mut first_fwsec_image: Option<FwSecBiosBuilder> = None;
         let mut second_fwsec_image: Option<FwSecBiosBuilder> = None;
 
         // Parse all VBIOS images in the ROM
-        for image_result in VbiosIterator::new(pdev, bar0)? {
+        for image_result in VbiosIterator::new(dev, bar0)? {
             let full_image = image_result?;
 
             dev_dbg!(
-                pdev.as_ref(),
+                dev,
                 "Found BIOS image: size: {:#x}, type: {}, last: {}\n",
                 full_image.image_size_bytes(),
                 full_image.image_type_str(),
@@ -234,14 +230,14 @@ impl Vbios {
             (second_fwsec_image, first_fwsec_image, pci_at_image)
         {
             second
-                .setup_falcon_data(pdev, &pci_at, &first)
-                .inspect_err(|e| dev_err!(pdev.as_ref(), "Falcon data setup failed: {:?}\n", e))?;
+                .setup_falcon_data(dev, &pci_at, &first)
+                .inspect_err(|e| dev_err!(dev, "Falcon data setup failed: {:?}\n", e))?;
             Ok(Vbios {
-                fwsec_image: second.build(pdev)?,
+                fwsec_image: second.build(dev)?,
             })
         } else {
             dev_err!(
-                pdev.as_ref(),
+                dev,
                 "Missing required images for falcon data setup, skipping\n"
             );
             Err(EINVAL)
@@ -284,9 +280,9 @@ struct PcirStruct {
 }
 
 impl PcirStruct {
-    fn new(pdev: &pci::Device, data: &[u8]) -> Result<Self> {
+    fn new(dev: &device::Device, data: &[u8]) -> Result<Self> {
         if data.len() < core::mem::size_of::<PcirStruct>() {
-            dev_err!(pdev.as_ref(), "Not enough data for PcirStruct\n");
+            dev_err!(dev, "Not enough data for PcirStruct\n");
             return Err(EINVAL);
         }
 
@@ -295,11 +291,7 @@ impl PcirStruct {
 
         // Signature should be "PCIR" (0x52494350) or "NPDS" (0x5344504e).
         if &signature != b"PCIR" && &signature != b"NPDS" {
-            dev_err!(
-                pdev.as_ref(),
-                "Invalid signature for PcirStruct: {:?}\n",
-                signature
-            );
+            dev_err!(dev, "Invalid signature for PcirStruct: {:?}\n", signature);
             return Err(EINVAL);
         }
 
@@ -308,7 +300,7 @@ impl PcirStruct {
 
         let image_len = u16::from_le_bytes([data[16], data[17]]);
         if image_len == 0 {
-            dev_err!(pdev.as_ref(), "Invalid image length: 0\n");
+            dev_err!(dev, "Invalid image length: 0\n");
             return Err(EINVAL);
         }
 
@@ -467,7 +459,7 @@ struct PciRomHeader {
 }
 
 impl PciRomHeader {
-    fn new(pdev: &pci::Device, data: &[u8]) -> Result<Self> {
+    fn new(dev: &device::Device, data: &[u8]) -> Result<Self> {
         if data.len() < 26 {
             // Need at least 26 bytes to read pciDataStrucPtr and sizeOfBlock.
             return Err(EINVAL);
@@ -479,7 +471,7 @@ impl PciRomHeader {
         match signature {
             0xAA55 | 0xBB77 | 0x4E56 => {}
             _ => {
-                dev_err!(pdev.as_ref(), "ROM signature unknown {:#x}\n", signature);
+                dev_err!(dev, "ROM signature unknown {:#x}\n", signature);
                 return Err(EINVAL);
             }
         }
@@ -538,9 +530,9 @@ struct NpdeStruct {
 }
 
 impl NpdeStruct {
-    fn new(pdev: &pci::Device, data: &[u8]) -> Option<Self> {
+    fn new(dev: &device::Device, data: &[u8]) -> Option<Self> {
         if data.len() < core::mem::size_of::<Self>() {
-            dev_dbg!(pdev.as_ref(), "Not enough data for NpdeStruct\n");
+            dev_dbg!(dev, "Not enough data for NpdeStruct\n");
             return None;
         }
 
@@ -549,17 +541,13 @@ impl NpdeStruct {
 
         // Signature should be "NPDE" (0x4544504E).
         if &signature != b"NPDE" {
-            dev_dbg!(
-                pdev.as_ref(),
-                "Invalid signature for NpdeStruct: {:?}\n",
-                signature
-            );
+            dev_dbg!(dev, "Invalid signature for NpdeStruct: {:?}\n", signature);
             return None;
         }
 
         let subimage_len = u16::from_le_bytes([data[8], data[9]]);
         if subimage_len == 0 {
-            dev_dbg!(pdev.as_ref(), "Invalid subimage length: 0\n");
+            dev_dbg!(dev, "Invalid subimage length: 0\n");
             return None;
         }
 
@@ -584,7 +572,7 @@ impl NpdeStruct {
 
     /// Try to find NPDE in the data, the NPDE is right after the PCIR.
     fn find_in_data(
-        pdev: &pci::Device,
+        dev: &device::Device,
         data: &[u8],
         rom_header: &PciRomHeader,
         pcir: &PcirStruct,
@@ -596,12 +584,12 @@ impl NpdeStruct {
 
         // Check if we have enough data
         if npde_start + core::mem::size_of::<Self>() > data.len() {
-            dev_dbg!(pdev.as_ref(), "Not enough data for NPDE\n");
+            dev_dbg!(dev, "Not enough data for NPDE\n");
             return None;
         }
 
         // Try to create NPDE from the data
-        NpdeStruct::new(pdev, &data[npde_start..])
+        NpdeStruct::new(dev, &data[npde_start..])
     }
 }
 
@@ -669,10 +657,10 @@ impl BiosImage {
 
     /// Create a [`BiosImageBase`] from a byte slice and convert it to a [`BiosImage`] which
     /// triggers the constructor of the specific BiosImage enum variant.
-    fn new(pdev: &pci::Device, data: &[u8]) -> Result<Self> {
-        let base = BiosImageBase::new(pdev, data)?;
+    fn new(dev: &device::Device, data: &[u8]) -> Result<Self> {
+        let base = BiosImageBase::new(dev, data)?;
         let image = base.into_image().inspect_err(|e| {
-            dev_err!(pdev.as_ref(), "Failed to create BiosImage: {:?}\n", e);
+            dev_err!(dev, "Failed to create BiosImage: {:?}\n", e);
         })?;
 
         Ok(image)
@@ -773,16 +761,16 @@ impl BiosImageBase {
     }
 
     /// Creates a new BiosImageBase from raw byte data.
-    fn new(pdev: &pci::Device, data: &[u8]) -> Result<Self> {
+    fn new(dev: &device::Device, data: &[u8]) -> Result<Self> {
         // Ensure we have enough data for the ROM header.
         if data.len() < 26 {
-            dev_err!(pdev.as_ref(), "Not enough data for ROM header\n");
+            dev_err!(dev, "Not enough data for ROM header\n");
             return Err(EINVAL);
         }
 
         // Parse the ROM header.
-        let rom_header = PciRomHeader::new(pdev, &data[0..26])
-            .inspect_err(|e| dev_err!(pdev.as_ref(), "Failed to create PciRomHeader: {:?}\n", e))?;
+        let rom_header = PciRomHeader::new(dev, &data[0..26])
+            .inspect_err(|e| dev_err!(dev, "Failed to create PciRomHeader: {:?}\n", e))?;
 
         // Get the PCI Data Structure using the pointer from the ROM header.
         let pcir_offset = rom_header.pci_data_struct_offset as usize;
@@ -791,22 +779,22 @@ impl BiosImageBase {
             .ok_or(EINVAL)
             .inspect_err(|_| {
                 dev_err!(
-                    pdev.as_ref(),
+                    dev,
                     "PCIR offset {:#x} out of bounds (data length: {})\n",
                     pcir_offset,
                     data.len()
                 );
                 dev_err!(
-                    pdev.as_ref(),
+                    dev,
                     "Consider reading more data for construction of BiosImage\n"
                 );
             })?;
 
-        let pcir = PcirStruct::new(pdev, pcir_data)
-            .inspect_err(|e| dev_err!(pdev.as_ref(), "Failed to create PcirStruct: {:?}\n", e))?;
+        let pcir = PcirStruct::new(dev, pcir_data)
+            .inspect_err(|e| dev_err!(dev, "Failed to create PcirStruct: {:?}\n", e))?;
 
         // Look for NPDE structure if this is not an NBSI image (type != 0x70).
-        let npde = NpdeStruct::find_in_data(pdev, data, &rom_header, &pcir);
+        let npde = NpdeStruct::find_in_data(dev, data, &rom_header, &pcir);
 
         // Create a copy of the data.
         let mut data_copy = KVec::new();
@@ -848,7 +836,7 @@ impl PciAtBiosImage {
     ///
     /// This is just a 4 byte structure that contains a pointer to the Falcon data in the FWSEC
     /// image.
-    fn falcon_data_ptr(&self, pdev: &pci::Device) -> Result<u32> {
+    fn falcon_data_ptr(&self, dev: &device::Device) -> Result<u32> {
         let token = self.get_bit_token(BIT_TOKEN_ID_FALCON_DATA)?;
 
         // Make sure we don't go out of bounds
@@ -859,14 +847,14 @@ impl PciAtBiosImage {
         // read the 4 bytes at the offset specified in the token
         let offset = token.data_offset as usize;
         let bytes: [u8; 4] = self.base.data[offset..offset + 4].try_into().map_err(|_| {
-            dev_err!(pdev.as_ref(), "Failed to convert data slice to array");
+            dev_err!(dev, "Failed to convert data slice to array");
             EINVAL
         })?;
 
         let data_ptr = u32::from_le_bytes(bytes);
 
         if (data_ptr as usize) < self.base.data.len() {
-            dev_err!(pdev.as_ref(), "Falcon data pointer out of bounds\n");
+            dev_err!(dev, "Falcon data pointer out of bounds\n");
             return Err(EINVAL);
         }
 
@@ -928,7 +916,7 @@ struct PmuLookupTable {
 }
 
 impl PmuLookupTable {
-    fn new(pdev: &pci::Device, data: &[u8]) -> Result<Self> {
+    fn new(dev: &device::Device, data: &[u8]) -> Result<Self> {
         if data.len() < 4 {
             return Err(EINVAL);
         }
@@ -940,10 +928,7 @@ impl PmuLookupTable {
         let required_bytes = header_len + (entry_count * entry_len);
 
         if data.len() < required_bytes {
-            dev_err!(
-                pdev.as_ref(),
-                "PmuLookupTable data length less than required\n"
-            );
+            dev_err!(dev, "PmuLookupTable data length less than required\n");
             return Err(EINVAL);
         }
 
@@ -956,11 +941,7 @@ impl PmuLookupTable {
 
         // Debug logging of entries (dumps the table data to dmesg)
         for i in (header_len..required_bytes).step_by(entry_len) {
-            dev_dbg!(
-                pdev.as_ref(),
-                "PMU entry: {:02x?}\n",
-                &data[i..][..entry_len]
-            );
+            dev_dbg!(dev, "PMU entry: {:02x?}\n", &data[i..][..entry_len]);
         }
 
         Ok(PmuLookupTable {
@@ -997,11 +978,11 @@ impl PmuLookupTable {
 impl FwSecBiosBuilder {
     fn setup_falcon_data(
         &mut self,
-        pdev: &pci::Device,
+        dev: &device::Device,
         pci_at_image: &PciAtBiosImage,
         first_fwsec: &FwSecBiosBuilder,
     ) -> Result {
-        let mut offset = pci_at_image.falcon_data_ptr(pdev)? as usize;
+        let mut offset = pci_at_image.falcon_data_ptr(dev)? as usize;
         let mut pmu_in_first_fwsec = false;
 
         // The falcon data pointer assumes that the PciAt and FWSEC images
@@ -1025,9 +1006,9 @@ impl FwSecBiosBuilder {
 
         if pmu_in_first_fwsec {
             self.pmu_lookup_table =
-                Some(PmuLookupTable::new(pdev, &first_fwsec.base.data[offset..])?);
+                Some(PmuLookupTable::new(dev, &first_fwsec.base.data[offset..])?);
         } else {
-            self.pmu_lookup_table = Some(PmuLookupTable::new(pdev, &self.base.data[offset..])?);
+            self.pmu_lookup_table = Some(PmuLookupTable::new(dev, &self.base.data[offset..])?);
         }
 
         match self
@@ -1040,18 +1021,14 @@ impl FwSecBiosBuilder {
                 let mut ucode_offset = entry.data as usize;
                 ucode_offset -= pci_at_image.base.data.len();
                 if ucode_offset < first_fwsec.base.data.len() {
-                    dev_err!(pdev.as_ref(), "Falcon Ucode offset not in second Fwsec.\n");
+                    dev_err!(dev, "Falcon Ucode offset not in second Fwsec.\n");
                     return Err(EINVAL);
                 }
                 ucode_offset -= first_fwsec.base.data.len();
                 self.falcon_ucode_offset = Some(ucode_offset);
             }
             Err(e) => {
-                dev_err!(
-                    pdev.as_ref(),
-                    "PmuLookupTableEntry not found, error: {:?}\n",
-                    e
-                );
+                dev_err!(dev, "PmuLookupTableEntry not found, error: {:?}\n", e);
                 return Err(EINVAL);
             }
         }
@@ -1059,7 +1036,7 @@ impl FwSecBiosBuilder {
     }
 
     /// Build the final FwSecBiosImage from this builder
-    fn build(self, pdev: &pci::Device) -> Result<FwSecBiosImage> {
+    fn build(self, dev: &device::Device) -> Result<FwSecBiosImage> {
         let ret = FwSecBiosImage {
             base: self.base,
             falcon_ucode_offset: self.falcon_ucode_offset.ok_or(EINVAL)?,
@@ -1067,8 +1044,8 @@ impl FwSecBiosBuilder {
 
         if cfg!(debug_assertions) {
             // Print the desc header for debugging
-            let desc = ret.header(pdev.as_ref())?;
-            dev_dbg!(pdev.as_ref(), "PmuLookupTableEntry desc: {:#?}\n", desc);
+            let desc = ret.header(dev)?;
+            dev_dbg!(dev, "PmuLookupTableEntry desc: {:#?}\n", desc);
         }
 
         Ok(ret)

From 93296e9d9528f0d87f2cf3fee494599060a0f14a Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Fri, 8 Aug 2025 11:46:42 +0900
Subject: [PATCH 0822/1455] gpu: nova-core: vbios: store reference to Device
 where relevant

Now that the vbios code uses a non-bound `Device` instance, store an
`ARef` to it at construction time so we can use it for logging without
having to carry an extra argument on every method for that sole purpose.

Reviewed-by: Joel Fernandes <joelagnelf@nvidia.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250808-vbios_device-v1-2-834bbbab6471@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/firmware/fwsec.rs |  8 +--
 drivers/gpu/nova-core/vbios.rs          | 69 +++++++++++++++----------
 2 files changed, 46 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/nova-core/firmware/fwsec.rs b/drivers/gpu/nova-core/firmware/fwsec.rs
index 47f5e4524072..8edbb5c0572c 100644
--- a/drivers/gpu/nova-core/firmware/fwsec.rs
+++ b/drivers/gpu/nova-core/firmware/fwsec.rs
@@ -246,8 +246,8 @@ impl FalconFirmware for FwsecFirmware {
 
 impl FirmwareDmaObject<FwsecFirmware, Unsigned> {
     fn new_fwsec(dev: &Device<device::Bound>, bios: &Vbios, cmd: FwsecCommand) -> Result<Self> {
-        let desc = bios.fwsec_image().header(dev)?;
-        let ucode = bios.fwsec_image().ucode(dev, desc)?;
+        let desc = bios.fwsec_image().header()?;
+        let ucode = bios.fwsec_image().ucode(desc)?;
         let mut dma_object = DmaObject::from_data(dev, ucode)?;
 
         let hdr_offset = (desc.imem_load_size + desc.interface_offset) as usize;
@@ -336,7 +336,7 @@ impl FwsecFirmware {
         let ucode_dma = FirmwareDmaObject::<Self, _>::new_fwsec(dev, bios, cmd)?;
 
         // Patch signature if needed.
-        let desc = bios.fwsec_image().header(dev)?;
+        let desc = bios.fwsec_image().header()?;
         let ucode_signed = if desc.signature_count != 0 {
             let sig_base_img = (desc.imem_load_size + desc.pkc_data_offset) as usize;
             let desc_sig_versions = u32::from(desc.signature_versions);
@@ -375,7 +375,7 @@ impl FwsecFirmware {
             dev_dbg!(dev, "patching signature with index {}\n", signature_idx);
             let signature = bios
                 .fwsec_image()
-                .sigs(dev, desc)
+                .sigs(desc)
                 .and_then(|sigs| sigs.get(signature_idx).ok_or(EINVAL))?;
 
             ucode_dma.patch_signature(signature, sig_base_img)?
diff --git a/drivers/gpu/nova-core/vbios.rs b/drivers/gpu/nova-core/vbios.rs
index ce32ab4cd5b2..e6a060714205 100644
--- a/drivers/gpu/nova-core/vbios.rs
+++ b/drivers/gpu/nova-core/vbios.rs
@@ -9,6 +9,7 @@ use core::convert::TryFrom;
 use kernel::device;
 use kernel::error::Result;
 use kernel::prelude::*;
+use kernel::types::ARef;
 
 /// The offset of the VBIOS ROM in the BAR0 space.
 const ROM_OFFSET: usize = 0x300000;
@@ -230,10 +231,10 @@ impl Vbios {
             (second_fwsec_image, first_fwsec_image, pci_at_image)
         {
             second
-                .setup_falcon_data(dev, &pci_at, &first)
+                .setup_falcon_data(&pci_at, &first)
                 .inspect_err(|e| dev_err!(dev, "Falcon data setup failed: {:?}\n", e))?;
             Ok(Vbios {
-                fwsec_image: second.build(dev)?,
+                fwsec_image: second.build()?,
             })
         } else {
             dev_err!(
@@ -742,9 +743,10 @@ impl TryFrom<BiosImageBase> for BiosImage {
 ///
 /// Each BiosImage type has a BiosImageBase type along with other image-specific fields. Note that
 /// Rust favors composition of types over inheritance.
-#[derive(Debug)]
 #[expect(dead_code)]
 struct BiosImageBase {
+    /// Used for logging.
+    dev: ARef<device::Device>,
     /// PCI ROM Expansion Header
     rom_header: PciRomHeader,
     /// PCI Data Structure
@@ -801,6 +803,7 @@ impl BiosImageBase {
         data_copy.extend_from_slice(data, GFP_KERNEL)?;
 
         Ok(BiosImageBase {
+            dev: dev.into(),
             rom_header,
             pcir,
             npde,
@@ -836,7 +839,7 @@ impl PciAtBiosImage {
     ///
     /// This is just a 4 byte structure that contains a pointer to the Falcon data in the FWSEC
     /// image.
-    fn falcon_data_ptr(&self, dev: &device::Device) -> Result<u32> {
+    fn falcon_data_ptr(&self) -> Result<u32> {
         let token = self.get_bit_token(BIT_TOKEN_ID_FALCON_DATA)?;
 
         // Make sure we don't go out of bounds
@@ -847,14 +850,14 @@ impl PciAtBiosImage {
         // read the 4 bytes at the offset specified in the token
         let offset = token.data_offset as usize;
         let bytes: [u8; 4] = self.base.data[offset..offset + 4].try_into().map_err(|_| {
-            dev_err!(dev, "Failed to convert data slice to array");
+            dev_err!(self.base.dev, "Failed to convert data slice to array");
             EINVAL
         })?;
 
         let data_ptr = u32::from_le_bytes(bytes);
 
         if (data_ptr as usize) < self.base.data.len() {
-            dev_err!(dev, "Falcon data pointer out of bounds\n");
+            dev_err!(self.base.dev, "Falcon data pointer out of bounds\n");
             return Err(EINVAL);
         }
 
@@ -978,11 +981,10 @@ impl PmuLookupTable {
 impl FwSecBiosBuilder {
     fn setup_falcon_data(
         &mut self,
-        dev: &device::Device,
         pci_at_image: &PciAtBiosImage,
         first_fwsec: &FwSecBiosBuilder,
     ) -> Result {
-        let mut offset = pci_at_image.falcon_data_ptr(dev)? as usize;
+        let mut offset = pci_at_image.falcon_data_ptr()? as usize;
         let mut pmu_in_first_fwsec = false;
 
         // The falcon data pointer assumes that the PciAt and FWSEC images
@@ -1005,10 +1007,15 @@ impl FwSecBiosBuilder {
         self.falcon_data_offset = Some(offset);
 
         if pmu_in_first_fwsec {
-            self.pmu_lookup_table =
-                Some(PmuLookupTable::new(dev, &first_fwsec.base.data[offset..])?);
+            self.pmu_lookup_table = Some(PmuLookupTable::new(
+                &self.base.dev,
+                &first_fwsec.base.data[offset..],
+            )?);
         } else {
-            self.pmu_lookup_table = Some(PmuLookupTable::new(dev, &self.base.data[offset..])?);
+            self.pmu_lookup_table = Some(PmuLookupTable::new(
+                &self.base.dev,
+                &self.base.data[offset..],
+            )?);
         }
 
         match self
@@ -1021,14 +1028,18 @@ impl FwSecBiosBuilder {
                 let mut ucode_offset = entry.data as usize;
                 ucode_offset -= pci_at_image.base.data.len();
                 if ucode_offset < first_fwsec.base.data.len() {
-                    dev_err!(dev, "Falcon Ucode offset not in second Fwsec.\n");
+                    dev_err!(self.base.dev, "Falcon Ucode offset not in second Fwsec.\n");
                     return Err(EINVAL);
                 }
                 ucode_offset -= first_fwsec.base.data.len();
                 self.falcon_ucode_offset = Some(ucode_offset);
             }
             Err(e) => {
-                dev_err!(dev, "PmuLookupTableEntry not found, error: {:?}\n", e);
+                dev_err!(
+                    self.base.dev,
+                    "PmuLookupTableEntry not found, error: {:?}\n",
+                    e
+                );
                 return Err(EINVAL);
             }
         }
@@ -1036,7 +1047,7 @@ impl FwSecBiosBuilder {
     }
 
     /// Build the final FwSecBiosImage from this builder
-    fn build(self, dev: &device::Device) -> Result<FwSecBiosImage> {
+    fn build(self) -> Result<FwSecBiosImage> {
         let ret = FwSecBiosImage {
             base: self.base,
             falcon_ucode_offset: self.falcon_ucode_offset.ok_or(EINVAL)?,
@@ -1044,8 +1055,8 @@ impl FwSecBiosBuilder {
 
         if cfg!(debug_assertions) {
             // Print the desc header for debugging
-            let desc = ret.header(dev)?;
-            dev_dbg!(dev, "PmuLookupTableEntry desc: {:#?}\n", desc);
+            let desc = ret.header()?;
+            dev_dbg!(ret.base.dev, "PmuLookupTableEntry desc: {:#?}\n", desc);
         }
 
         Ok(ret)
@@ -1054,13 +1065,16 @@ impl FwSecBiosBuilder {
 
 impl FwSecBiosImage {
     /// Get the FwSec header ([`FalconUCodeDescV3`]).
-    pub(crate) fn header(&self, dev: &device::Device) -> Result<&FalconUCodeDescV3> {
+    pub(crate) fn header(&self) -> Result<&FalconUCodeDescV3> {
         // Get the falcon ucode offset that was found in setup_falcon_data.
         let falcon_ucode_offset = self.falcon_ucode_offset;
 
         // Make sure the offset is within the data bounds.
         if falcon_ucode_offset + core::mem::size_of::<FalconUCodeDescV3>() > self.base.data.len() {
-            dev_err!(dev, "fwsec-frts header not contained within BIOS bounds\n");
+            dev_err!(
+                self.base.dev,
+                "fwsec-frts header not contained within BIOS bounds\n"
+            );
             return Err(ERANGE);
         }
 
@@ -1072,7 +1086,7 @@ impl FwSecBiosImage {
         let ver = (hdr & 0xff00) >> 8;
 
         if ver != 3 {
-            dev_err!(dev, "invalid fwsec firmware version: {:?}\n", ver);
+            dev_err!(self.base.dev, "invalid fwsec firmware version: {:?}\n", ver);
             return Err(EINVAL);
         }
 
@@ -1092,7 +1106,7 @@ impl FwSecBiosImage {
     }
 
     /// Get the ucode data as a byte slice
-    pub(crate) fn ucode(&self, dev: &device::Device, desc: &FalconUCodeDescV3) -> Result<&[u8]> {
+    pub(crate) fn ucode(&self, desc: &FalconUCodeDescV3) -> Result<&[u8]> {
         let falcon_ucode_offset = self.falcon_ucode_offset;
 
         // The ucode data follows the descriptor.
@@ -1104,15 +1118,16 @@ impl FwSecBiosImage {
             .data
             .get(ucode_data_offset..ucode_data_offset + size)
             .ok_or(ERANGE)
-            .inspect_err(|_| dev_err!(dev, "fwsec ucode data not contained within BIOS bounds\n"))
+            .inspect_err(|_| {
+                dev_err!(
+                    self.base.dev,
+                    "fwsec ucode data not contained within BIOS bounds\n"
+                )
+            })
     }
 
     /// Get the signatures as a byte slice
-    pub(crate) fn sigs(
-        &self,
-        dev: &device::Device,
-        desc: &FalconUCodeDescV3,
-    ) -> Result<&[Bcrt30Rsa3kSignature]> {
+    pub(crate) fn sigs(&self, desc: &FalconUCodeDescV3) -> Result<&[Bcrt30Rsa3kSignature]> {
         // The signatures data follows the descriptor.
         let sigs_data_offset = self.falcon_ucode_offset + core::mem::size_of::<FalconUCodeDescV3>();
         let sigs_size =
@@ -1121,7 +1136,7 @@ impl FwSecBiosImage {
         // Make sure the data is within bounds.
         if sigs_data_offset + sigs_size > self.base.data.len() {
             dev_err!(
-                dev,
+                self.base.dev,
                 "fwsec signatures data not contained within BIOS bounds\n"
             );
             return Err(ERANGE);

From c4baa2d987c025076db8732abf8c44ccb9bcd8df Mon Sep 17 00:00:00 2001
From: Marcus Folkesson <marcus.folkesson@gmail.com>
Date: Mon, 21 Jul 2025 12:43:31 +0200
Subject: [PATCH 0823/1455] drm/st7571-i2c: correct pixel data format
 description

The comment describes the pixel data format as stated in
the st7571 datasheet, which is not necessary the same
as for the connected display.

Instead, describe the expected pixel data format which is used for
R1/R2/XRGB8888.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250721-st7571-format-v2-1-159f4134098c@gmail.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
index dda99cbe5372..6f8142943428 100644
--- a/drivers/gpu/drm/sitronix/st7571-i2c.c
+++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
@@ -386,10 +386,10 @@ static int st7571_fb_update_rect_grayscale(struct drm_framebuffer *fb, struct dr
 			 * even if the format is monochrome.
 			 *
 			 * The bit values maps to the following grayscale:
-			 * 0 0 = White
-			 * 0 1 = Light gray
-			 * 1 0 = Dark gray
-			 * 1 1 = Black
+			 * 0 0 = Black
+			 * 0 1 = Dark gray
+			 * 1 0 = Light gray
+			 * 1 1 = White
 			 *
 			 * For monochrome formats, write the same value twice to get
 			 * either a black or white pixel.

From 2596e9414e5f5917070a6c524d8e307234d3cc18 Mon Sep 17 00:00:00 2001
From: Marcus Folkesson <marcus.folkesson@gmail.com>
Date: Mon, 21 Jul 2025 12:43:32 +0200
Subject: [PATCH 0824/1455] dt-bindings: display: sitronix,st7571: add optional
 inverted property

Depending on which display that is connected to the controller, an "1"
means either a black or a white pixel.

The supported formats (R1/R2/XRGB8888) expects the pixels
to map against (4bit):
00 => Black
01 => Dark Gray
10 => Light Gray
11 => White

If this is not what the display map against, the controller has support
to invert these values.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250721-st7571-format-v2-2-159f4134098c@gmail.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 .../devicetree/bindings/display/sitronix,st7571.yaml         | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/sitronix,st7571.yaml b/Documentation/devicetree/bindings/display/sitronix,st7571.yaml
index 4fea782fccd7..b83721eb4b7f 100644
--- a/Documentation/devicetree/bindings/display/sitronix,st7571.yaml
+++ b/Documentation/devicetree/bindings/display/sitronix,st7571.yaml
@@ -28,6 +28,11 @@ properties:
     description:
       Display supports 4-level grayscale.
 
+  sitronix,inverted:
+    type: boolean
+    description:
+      Display pixels are inverted, i.e. 0 is white and 1 is black.
+
   reset-gpios: true
   width-mm: true
   height-mm: true

From c68f78ae1d86a5eb9632b72f7b4f15912bff5b0f Mon Sep 17 00:00:00 2001
From: Marcus Folkesson <marcus.folkesson@gmail.com>
Date: Mon, 21 Jul 2025 12:43:33 +0200
Subject: [PATCH 0825/1455] dt-bindings: display: sitronix,st7567: add optional
 inverted property

Depending on which display that is connected to the controller, an "1"
means either a black or a white pixel.

The supported format (R1) expects the pixels to map against:
    0 => Black
    1 => White

If this is not what the display map against, the controller has support
to invert these values.

Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250721-st7571-format-v2-3-159f4134098c@gmail.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 .../devicetree/bindings/display/sitronix,st7567.yaml         | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/sitronix,st7567.yaml b/Documentation/devicetree/bindings/display/sitronix,st7567.yaml
index e8a5b8ad18fe..2eb6d00b5a25 100644
--- a/Documentation/devicetree/bindings/display/sitronix,st7567.yaml
+++ b/Documentation/devicetree/bindings/display/sitronix,st7567.yaml
@@ -23,6 +23,11 @@ properties:
   reg:
     maxItems: 1
 
+  sitronix,inverted:
+    type: boolean
+    description:
+      Display pixels are inverted, i.e. 0 is white and 1 is black.
+
   width-mm: true
   height-mm: true
   panel-timing: true

From e61c35157d32b4b422f0a4cbc3c40d04d883a9c9 Mon Sep 17 00:00:00 2001
From: Marcus Folkesson <marcus.folkesson@gmail.com>
Date: Mon, 21 Jul 2025 12:43:34 +0200
Subject: [PATCH 0826/1455] drm/st7571-i2c: add support for inverted pixel
 format

Depending on which display that is connected to the controller, an
"1" means either a black or a white pixel.

The supported formats (R1/R2/XRGB8888) expects the pixels
to map against (4bit):
    00 => Black
    01 => Dark Gray
    10 => Light Gray
    11 => White

If this is not what the display map against, make it possible to invert
the pixels.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250721-st7571-format-v2-4-159f4134098c@gmail.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/gpu/drm/sitronix/st7571-i2c.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
index 6f8142943428..807bea62d96e 100644
--- a/drivers/gpu/drm/sitronix/st7571-i2c.c
+++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
@@ -151,6 +151,7 @@ struct st7571_device {
 	bool ignore_nak;
 
 	bool grayscale;
+	bool inverted;
 	u32 height_mm;
 	u32 width_mm;
 	u32 startline;
@@ -792,6 +793,7 @@ static int st7567_parse_dt(struct st7571_device *st7567)
 
 	of_property_read_u32(np, "width-mm", &st7567->width_mm);
 	of_property_read_u32(np, "height-mm", &st7567->height_mm);
+	st7567->inverted = of_property_read_bool(np, "sitronix,inverted");
 
 	st7567->pformat = &st7571_monochrome;
 	st7567->bpp = 1;
@@ -819,6 +821,7 @@ static int st7571_parse_dt(struct st7571_device *st7571)
 	of_property_read_u32(np, "width-mm", &st7571->width_mm);
 	of_property_read_u32(np, "height-mm", &st7571->height_mm);
 	st7571->grayscale = of_property_read_bool(np, "sitronix,grayscale");
+	st7571->inverted = of_property_read_bool(np, "sitronix,inverted");
 
 	if (st7571->grayscale) {
 		st7571->pformat = &st7571_grayscale;
@@ -873,7 +876,7 @@ static int st7567_lcd_init(struct st7571_device *st7567)
 		ST7571_SET_POWER(0x6),	/* Power Control, VC: ON, VR: ON, VF: OFF */
 		ST7571_SET_POWER(0x7),	/* Power Control, VC: ON, VR: ON, VF: ON */
 
-		ST7571_SET_REVERSE(0),
+		ST7571_SET_REVERSE(st7567->inverted ? 1 : 0),
 		ST7571_SET_ENTIRE_DISPLAY_ON(0),
 	};
 
@@ -917,7 +920,7 @@ static int st7571_lcd_init(struct st7571_device *st7571)
 		ST7571_SET_COLOR_MODE(st7571->pformat->mode),
 		ST7571_COMMAND_SET_NORMAL,
 
-		ST7571_SET_REVERSE(0),
+		ST7571_SET_REVERSE(st7571->inverted ? 1 : 0),
 		ST7571_SET_ENTIRE_DISPLAY_ON(0),
 	};
 

From 8bde81ec684238587decd5ab6b1bf18041814937 Mon Sep 17 00:00:00 2001
From: Marcus Folkesson <marcus.folkesson@gmail.com>
Date: Mon, 21 Jul 2025 12:43:35 +0200
Subject: [PATCH 0827/1455] drm/format-helper: introduce
 drm_fb_xrgb8888_to_gray2()

Convert XRGB8888 to 2bit grayscale.

It uses drm_fb_xrgb8888_to_gray8() to convert the pixels to gray8 as an
intermediate step before converting to gray2.

Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250721-st7571-format-v2-5-159f4134098c@gmail.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/gpu/drm/drm_format_helper.c | 108 ++++++++++++++++++++++++++++
 include/drm/drm_format_helper.h     |   4 ++
 2 files changed, 112 insertions(+)

diff --git a/drivers/gpu/drm/drm_format_helper.c b/drivers/gpu/drm/drm_format_helper.c
index 37d5e0a7eb46..006836554cc2 100644
--- a/drivers/gpu/drm/drm_format_helper.c
+++ b/drivers/gpu/drm/drm_format_helper.c
@@ -1256,6 +1256,25 @@ int drm_fb_blit(struct iosys_map *dst, const unsigned int *dst_pitch, uint32_t d
 }
 EXPORT_SYMBOL(drm_fb_blit);
 
+static void drm_fb_gray8_to_gray2_line(void *dbuf, const void *sbuf, unsigned int pixels)
+{
+	u8 *dbuf8 = dbuf;
+	const u8 *sbuf8 = sbuf;
+	u8 px;
+
+	while (pixels) {
+		unsigned int i, bits = min(pixels, 4U);
+		u8 byte = 0;
+
+		for (i = 0; i < bits; i++, pixels--) {
+			byte >>= 2;
+			px = (*sbuf8++ * 3 + 127) / 255;
+			byte |= (px &= 0x03) << 6;
+		}
+		*dbuf8++ = byte;
+	}
+}
+
 static void drm_fb_gray8_to_mono_line(void *dbuf, const void *sbuf, unsigned int pixels)
 {
 	u8 *dbuf8 = dbuf;
@@ -1362,3 +1381,92 @@ void drm_fb_xrgb8888_to_mono(struct iosys_map *dst, const unsigned int *dst_pitc
 	}
 }
 EXPORT_SYMBOL(drm_fb_xrgb8888_to_mono);
+
+/**
+ * drm_fb_xrgb8888_to_gray2 - Convert XRGB8888 to gray2
+ * @dst: Array of gray2 destination buffer
+ * @dst_pitch: Array of numbers of bytes between the start of two consecutive scanlines
+ *             within @dst; can be NULL if scanlines are stored next to each other.
+ * @src: Array of XRGB8888 source buffers
+ * @fb: DRM framebuffer
+ * @clip: Clip rectangle area to copy
+ * @state: Transform and conversion state
+ *
+ * This function copies parts of a framebuffer to display memory and converts the
+ * color format during the process. Destination and framebuffer formats must match. The
+ * parameters @dst, @dst_pitch and @src refer to arrays. Each array must have at
+ * least as many entries as there are planes in @fb's format. Each entry stores the
+ * value for the format's respective color plane at the same index.
+ *
+ * This function does not apply clipping on @dst (i.e. the destination is at the
+ * top-left corner). The first pixel (upper left corner of the clip rectangle) will
+ * be converted and copied to the two first bits (LSB) in the first byte of the gray2
+ * destination buffer. If the caller requires that the first pixel in a byte must
+ * be located at an x-coordinate that is a multiple of 8, then the caller must take
+ * care itself of supplying a suitable clip rectangle.
+ *
+ * DRM doesn't have native gray2 support. Drivers can use this function for
+ * gray2 devices that don't support XRGB8888 natively. Such drivers can
+ * announce the commonly supported XR24 format to userspace and use this function
+ * to convert to the native format.
+ *
+ */
+void drm_fb_xrgb8888_to_gray2(struct iosys_map *dst, const unsigned int *dst_pitch,
+			      const struct iosys_map *src, const struct drm_framebuffer *fb,
+			      const struct drm_rect *clip, struct drm_format_conv_state *state)
+{
+	static const unsigned int default_dst_pitch[DRM_FORMAT_MAX_PLANES] = {
+		0, 0, 0, 0
+	};
+	unsigned int linepixels = drm_rect_width(clip);
+	unsigned int lines = drm_rect_height(clip);
+	unsigned int cpp = fb->format->cpp[0];
+	unsigned int len_src32 = linepixels * cpp;
+	struct drm_device *dev = fb->dev;
+	void *vaddr = src[0].vaddr;
+	unsigned int dst_pitch_0;
+	unsigned int y;
+	u8 *gray2 = dst[0].vaddr, *gray8;
+	u32 *src32;
+
+	if (drm_WARN_ON(dev, fb->format->format != DRM_FORMAT_XRGB8888))
+		return;
+
+	if (!dst_pitch)
+		dst_pitch = default_dst_pitch;
+	dst_pitch_0 = dst_pitch[0];
+
+	/*
+	 * The gray2 destination buffer contains 2 bit per pixel
+	 */
+	if (!dst_pitch_0)
+		dst_pitch_0 = DIV_ROUND_UP(linepixels, 4);
+
+	/*
+	 * The dma memory is write-combined so reads are uncached.
+	 * Speed up by fetching one line at a time.
+	 *
+	 * Also, format conversion from XR24 to gray2 are done
+	 * line-by-line but are converted to 8-bit grayscale as an
+	 * intermediate step.
+	 *
+	 * Allocate a buffer to be used for both copying from the cma
+	 * memory and to store the intermediate grayscale line pixels.
+	 */
+	src32 = drm_format_conv_state_reserve(state, len_src32 + linepixels, GFP_KERNEL);
+	if (!src32)
+		return;
+
+	gray8 = (u8 *)src32 + len_src32;
+
+	vaddr += clip_offset(clip, fb->pitches[0], cpp);
+	for (y = 0; y < lines; y++) {
+		src32 = memcpy(src32, vaddr, len_src32);
+		drm_fb_xrgb8888_to_gray8_line(gray8, src32, linepixels);
+		drm_fb_gray8_to_gray2_line(gray2, gray8, linepixels);
+		vaddr += fb->pitches[0];
+		gray2 += dst_pitch_0;
+	}
+}
+EXPORT_SYMBOL(drm_fb_xrgb8888_to_gray2);
+
diff --git a/include/drm/drm_format_helper.h b/include/drm/drm_format_helper.h
index 562bc383ece4..32d57d6c5327 100644
--- a/include/drm/drm_format_helper.h
+++ b/include/drm/drm_format_helper.h
@@ -136,4 +136,8 @@ void drm_fb_xrgb8888_to_mono(struct iosys_map *dst, const unsigned int *dst_pitc
 			     const struct iosys_map *src, const struct drm_framebuffer *fb,
 			     const struct drm_rect *clip, struct drm_format_conv_state *state);
 
+void drm_fb_xrgb8888_to_gray2(struct iosys_map *dst, const unsigned int *dst_pitch,
+			      const struct iosys_map *src, const struct drm_framebuffer *fb,
+			      const struct drm_rect *clip, struct drm_format_conv_state *state);
+
 #endif /* __LINUX_DRM_FORMAT_HELPER_H */

From 100d457c0feadd131083e3ebb64973ac0b8510d6 Mon Sep 17 00:00:00 2001
From: Marcus Folkesson <marcus.folkesson@gmail.com>
Date: Mon, 21 Jul 2025 12:43:36 +0200
Subject: [PATCH 0828/1455] drm/st7571-i2c: add support for 2bit grayscale for
 XRGB8888

Add support for 2bit grayscale and use it for XRGB8888 when grayscale is
supported.

Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Marcus Folkesson <marcus.folkesson@gmail.com>
Link: https://lore.kernel.org/r/20250721-st7571-format-v2-6-159f4134098c@gmail.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/gpu/drm/sitronix/st7571-i2c.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
index 807bea62d96e..a6c4a6738ded 100644
--- a/drivers/gpu/drm/sitronix/st7571-i2c.c
+++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
@@ -219,10 +219,11 @@ static int st7571_send_command_list(struct st7571_device *st7571,
 	return ret;
 }
 
-static inline u8 st7571_transform_xy(const char *p, int x, int y)
+static inline u8 st7571_transform_xy(const char *p, int x, int y, u8 bpp)
 {
 	int xrest = x % 8;
 	u8 result = 0;
+	u8 row_len = 16 * bpp;
 
 	/*
 	 * Transforms an (x, y) pixel coordinate into a vertical 8-bit
@@ -237,7 +238,7 @@ static inline u8 st7571_transform_xy(const char *p, int x, int y)
 
 	for (int i = 0; i < 8; i++) {
 		int row_idx = y + i;
-		u8 byte = p[row_idx * 16 + x];
+		u8 byte = p[row_idx * row_len + x];
 		u8 bit = (byte >> xrest) & 1;
 
 		result |= (bit << i);
@@ -304,11 +305,11 @@ static void st7571_prepare_buffer_grayscale(struct st7571_device *st7571,
 	struct iosys_map dst;
 
 	switch (fb->format->format) {
-	case DRM_FORMAT_XRGB8888: /* Only support XRGB8888 in monochrome mode */
-		dst_pitch = DIV_ROUND_UP(drm_rect_width(rect), 8);
+	case DRM_FORMAT_XRGB8888:
+		dst_pitch = DIV_ROUND_UP(drm_rect_width(rect), 4);
 		iosys_map_set_vaddr(&dst, st7571->hwbuf);
 
-		drm_fb_xrgb8888_to_mono(&dst, &dst_pitch, vmap, fb, rect, fmtcnv_state);
+		drm_fb_xrgb8888_to_gray2(&dst, &dst_pitch, vmap, fb, rect, fmtcnv_state);
 		break;
 
 	case DRM_FORMAT_R1:
@@ -334,7 +335,7 @@ static int st7571_fb_update_rect_monochrome(struct drm_framebuffer *fb, struct d
 
 	for (int y = rect->y1; y < rect->y2; y += ST7571_PAGE_HEIGHT) {
 		for (int x = rect->x1; x < rect->x2; x++)
-			row[x] = st7571_transform_xy(st7571->hwbuf, x, y);
+			row[x] = st7571_transform_xy(st7571->hwbuf, x, y, 1);
 
 		st7571_set_position(st7571, rect->x1, y);
 
@@ -359,14 +360,13 @@ static int st7571_fb_update_rect_grayscale(struct drm_framebuffer *fb, struct dr
 	rect->y2 = min_t(unsigned int, round_up(rect->y2, ST7571_PAGE_HEIGHT), st7571->nlines);
 
 	switch (format) {
-	case DRM_FORMAT_XRGB8888:
-		/* Threated as monochrome (R1) */
-		fallthrough;
 	case DRM_FORMAT_R1:
-		x1 = rect->x1;
-		x2 = rect->x2;
+		x1 = rect->x1 * 1;
+		x2 = rect->x2 * 1;
 		break;
 	case DRM_FORMAT_R2:
+		fallthrough;
+	case DRM_FORMAT_XRGB8888:
 		x1 = rect->x1 * 2;
 		x2 = rect->x2 * 2;
 		break;
@@ -374,7 +374,7 @@ static int st7571_fb_update_rect_grayscale(struct drm_framebuffer *fb, struct dr
 
 	for (int y = rect->y1; y < rect->y2; y += ST7571_PAGE_HEIGHT) {
 		for (int x = x1; x < x2; x++)
-			row[x] = st7571_transform_xy(st7571->hwbuf, x, y);
+			row[x] = st7571_transform_xy(st7571->hwbuf, x, y, 2);
 
 		st7571_set_position(st7571, rect->x1, y);
 
@@ -395,7 +395,7 @@ static int st7571_fb_update_rect_grayscale(struct drm_framebuffer *fb, struct dr
 			 * For monochrome formats, write the same value twice to get
 			 * either a black or white pixel.
 			 */
-			if (format == DRM_FORMAT_R1 || format == DRM_FORMAT_XRGB8888)
+			if (format == DRM_FORMAT_R1)
 				regmap_bulk_write(st7571->regmap, ST7571_DATA_MODE, row + x, 1);
 		}
 	}

From a69997be443dcc64e6f0d6588f40e06c9b3a39be Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 20 Aug 2025 21:14:15 +0800
Subject: [PATCH 0829/1455] drm/ssd130x: Remove the use of dev_err_probe()

Logging messages that show some type of "out of memory" error are generally
unnecessary as there is a generic message and a stack dump done by the
memory subsystem. These messages generally increase kernel size without
much added value[1].

The dev_err_probe() doesn't do anything when error is '-ENOMEM'. Therefore,
remove the useless call to dev_err_probe(), and just return the value
instead.

[1]: https://lore.kernel.org/lkml/1402419340.30479.18.camel@joe-AO725/

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Link: https://lore.kernel.org/r/20250820131416.500048-1-liaoyuanhong@vivo.com
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---
 drivers/gpu/drm/solomon/ssd130x-spi.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/solomon/ssd130x-spi.c b/drivers/gpu/drm/solomon/ssd130x-spi.c
index 7c935870f7d2..b52f5fd592a1 100644
--- a/drivers/gpu/drm/solomon/ssd130x-spi.c
+++ b/drivers/gpu/drm/solomon/ssd130x-spi.c
@@ -74,8 +74,7 @@ static int ssd130x_spi_probe(struct spi_device *spi)
 
 	t = devm_kzalloc(dev, sizeof(*t), GFP_KERNEL);
 	if (!t)
-		return dev_err_probe(dev, -ENOMEM,
-				     "Failed to allocate SPI transport data\n");
+		return -ENOMEM;
 
 	t->spi = spi;
 	t->dc = dc;

From e548e04e619d34c31c98f3b250c2792a0e535278 Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Tue, 19 Aug 2025 20:37:29 +0530
Subject: [PATCH 0830/1455] drm/i915/hdcp: Remove the encoder check in hdcp
 enable

Remove the intel_encoder NULL check from _intel_hdcp_enable. With
all the changes it has gone through this check has become unnecessary
since at this point the connector is supposed to have the encoder
in it.

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250819150729.88561-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 42202c8bb066..e11cf26ed0bf 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -2446,12 +2446,6 @@ static int _intel_hdcp_enable(struct intel_atomic_state *state,
 	if (!hdcp->shim)
 		return -ENOENT;
 
-	if (!connector->encoder) {
-		drm_err(display->drm, "[CONNECTOR:%d:%s] encoder is not initialized\n",
-			connector->base.base.id, connector->base.name);
-		return -ENODEV;
-	}
-
 	mutex_lock(&hdcp->mutex);
 	mutex_lock(&dig_port->hdcp.mutex);
 	drm_WARN_ON(display->drm,

From ee4b32220a6b41e71512e8804585325e685456ba Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Mon, 1 Sep 2025 12:55:41 +0530
Subject: [PATCH 0831/1455] drm/xe/guc: Add devm release action to safely tear
 down CT

When a buffer object (BO) is allocated with the XE_BO_FLAG_GGTT_INVALIDATE
flag, the driver initiates TLB invalidation requests via the CTB mechanism
while releasing the BO. However a premature release of the CTB BO can lead
to system crashes, as observed in:

Oops: Oops: 0000 [#1] SMP NOPTI
RIP: 0010:h2g_write+0x2f3/0x7c0 [xe]
Call Trace:
 guc_ct_send_locked+0x8b/0x670 [xe]
 xe_guc_ct_send_locked+0x19/0x60 [xe]
 send_tlb_invalidation+0xb4/0x460 [xe]
 xe_gt_tlb_invalidation_ggtt+0x15e/0x2e0 [xe]
 ggtt_invalidate_gt_tlb.part.0+0x16/0x90 [xe]
 ggtt_node_remove+0x110/0x140 [xe]
 xe_ggtt_node_remove+0x40/0xa0 [xe]
 xe_ggtt_remove_bo+0x87/0x250 [xe]

Introduce a devm-managed release action during xe_guc_ct_init() and
xe_guc_ct_init_post_hwconfig() to ensure proper CTB disablement before
resource deallocation, preventing the use-after-free scenario.

Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Summers Stuart <stuart.summers@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250901072541.31461-1-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/xe_guc.c    |  8 +++----
 drivers/gpu/drm/xe/xe_guc_ct.c | 41 +++++++++++++++++++++++++++++++++-
 drivers/gpu/drm/xe/xe_guc_ct.h |  1 +
 3 files changed, 45 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 37d06c51180c..b3a6408a5760 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -709,10 +709,6 @@ static int xe_guc_realloc_post_hwconfig(struct xe_guc *guc)
 	if (ret)
 		return ret;
 
-	ret = xe_managed_bo_reinit_in_vram(xe, tile, &guc->ct.bo);
-	if (ret)
-		return ret;
-
 	return 0;
 }
 
@@ -847,6 +843,10 @@ int xe_guc_init_post_hwconfig(struct xe_guc *guc)
 	if (ret)
 		return ret;
 
+	ret = xe_guc_ct_init_post_hwconfig(&guc->ct);
+	if (ret)
+		return ret;
+
 	guc_init_params_post_hwconfig(guc);
 
 	ret = xe_guc_submit_init(guc, ~0);
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 848065a25c44..e431ff73227c 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -39,6 +39,8 @@ static void receive_g2h(struct xe_guc_ct *ct);
 static void g2h_worker_func(struct work_struct *w);
 static void safe_mode_worker_func(struct work_struct *w);
 static void ct_exit_safe_mode(struct xe_guc_ct *ct);
+static void guc_ct_change_state(struct xe_guc_ct *ct,
+				enum xe_guc_ct_state state);
 
 #if IS_ENABLED(CONFIG_DRM_XE_DEBUG)
 enum {
@@ -252,6 +254,13 @@ int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct)
 }
 ALLOW_ERROR_INJECTION(xe_guc_ct_init_noalloc, ERRNO); /* See xe_pci_probe() */
 
+static void guc_action_disable_ct(void *arg)
+{
+	struct xe_guc_ct *ct = arg;
+
+	guc_ct_change_state(ct, XE_GUC_CT_STATE_DISABLED);
+}
+
 int xe_guc_ct_init(struct xe_guc_ct *ct)
 {
 	struct xe_device *xe = ct_to_xe(ct);
@@ -268,10 +277,40 @@ int xe_guc_ct_init(struct xe_guc_ct *ct)
 		return PTR_ERR(bo);
 
 	ct->bo = bo;
-	return 0;
+
+	return devm_add_action_or_reset(xe->drm.dev, guc_action_disable_ct, ct);
 }
 ALLOW_ERROR_INJECTION(xe_guc_ct_init, ERRNO); /* See xe_pci_probe() */
 
+/**
+ * xe_guc_ct_init_post_hwconfig - Reinitialize the GuC CTB in VRAM
+ * @ct: the &xe_guc_ct
+ *
+ * Allocate a new BO in VRAM and free the previous BO that was allocated
+ * in system memory (SMEM). Applicable only for DGFX products.
+ *
+ * Return: 0 on success, or a negative errno on failure.
+ */
+int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct)
+{
+	struct xe_device *xe = ct_to_xe(ct);
+	struct xe_gt *gt = ct_to_gt(ct);
+	struct xe_tile *tile = gt_to_tile(gt);
+	int ret;
+
+	xe_assert(xe, !xe_guc_ct_enabled(ct));
+
+	if (!IS_DGFX(xe))
+		return 0;
+
+	ret = xe_managed_bo_reinit_in_vram(xe, tile, &ct->bo);
+	if (ret)
+		return ret;
+
+	devm_release_action(xe->drm.dev, guc_action_disable_ct, ct);
+	return devm_add_action_or_reset(xe->drm.dev, guc_action_disable_ct, ct);
+}
+
 #define desc_read(xe_, guc_ctb__, field_)			\
 	xe_map_rd_field(xe_, &guc_ctb__->desc, 0,		\
 			struct guc_ct_buffer_desc, field_)
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.h b/drivers/gpu/drm/xe/xe_guc_ct.h
index 18d4225e6502..cf41210ab30a 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.h
+++ b/drivers/gpu/drm/xe/xe_guc_ct.h
@@ -13,6 +13,7 @@ struct xe_device;
 
 int xe_guc_ct_init_noalloc(struct xe_guc_ct *ct);
 int xe_guc_ct_init(struct xe_guc_ct *ct);
+int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct);
 int xe_guc_ct_enable(struct xe_guc_ct *ct);
 void xe_guc_ct_disable(struct xe_guc_ct *ct);
 void xe_guc_ct_stop(struct xe_guc_ct *ct);

From b2b8af21fec35be417a3199b5a6c354605dd222a Mon Sep 17 00:00:00 2001
From: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Date: Mon, 1 Sep 2025 14:40:32 +0200
Subject: [PATCH 0832/1455] drm/sched: Fix racy access to
 drm_sched_entity.dependency

The drm_sched_job_unschedulable trace point can access
entity->dependency after it was cleared by the callback
installed in drm_sched_entity_add_dependency_cb, causing:

BUG: kernel NULL pointer dereference, address: 0000000000000020
[...]
Workqueue: comp_1.1.0 drm_sched_run_job_work [gpu_sched]
RIP: 0010:trace_event_raw_event_drm_sched_job_unschedulable+0x70/0xd0 [gpu_sched]

To fix this we either need to keep a reference to the fence before
setting up the callbacks, or move the trace_drm_sched_job_unschedulable
calls into drm_sched_entity_add_dependency_cb where they can be
done earlier.

Fixes: 76d97c870f29 ("drm/sched: Trace dependencies for GPU jobs")

Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Philipp Stanner <phasta@kernel.org>
Link: https://lore.kernel.org/r/20250901124032.1955-1-pierre-eric.pelloux-prayer@amd.com
---
 drivers/gpu/drm/scheduler/sched_entity.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 254932b9050e..5a4697f636f2 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -391,7 +391,8 @@ EXPORT_SYMBOL(drm_sched_entity_set_priority);
  * Add a callback to the current dependency of the entity to wake up the
  * scheduler when the entity becomes available.
  */
-static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity)
+static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity,
+					       struct drm_sched_job *sched_job)
 {
 	struct drm_gpu_scheduler *sched = entity->rq->sched;
 	struct dma_fence *fence = entity->dependency;
@@ -421,6 +422,10 @@ static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity)
 		entity->dependency = fence;
 	}
 
+	if (trace_drm_sched_job_unschedulable_enabled() &&
+	    !test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &entity->dependency->flags))
+		trace_drm_sched_job_unschedulable(sched_job, entity->dependency);
+
 	if (!dma_fence_add_callback(entity->dependency, &entity->cb,
 				    drm_sched_entity_wakeup))
 		return true;
@@ -461,10 +466,8 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity)
 
 	while ((entity->dependency =
 			drm_sched_job_dependency(sched_job, entity))) {
-		if (drm_sched_entity_add_dependency_cb(entity)) {
-			trace_drm_sched_job_unschedulable(sched_job, entity->dependency);
+		if (drm_sched_entity_add_dependency_cb(entity, sched_job))
 			return NULL;
-		}
 	}
 
 	/* skip jobs from entity that marked guilty */

From 2a06126d1624f0661fac0833e7ee1dbdc636154e Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Tue, 19 Aug 2025 11:42:10 +0200
Subject: [PATCH 0833/1455] drm/debugfs: bridges_show: show refcount

Now that bridges are refcounted, exposing the refcount in debugfs can be
useful.

Suggested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250819-drm-bridge-debugfs-removed-v7-1-970702579978@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/drm_bridge.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c
index dd439d55177a..d031447eebc9 100644
--- a/drivers/gpu/drm/drm_bridge.c
+++ b/drivers/gpu/drm/drm_bridge.c
@@ -1435,6 +1435,9 @@ static void drm_bridge_debugfs_show_bridge(struct drm_printer *p,
 					   unsigned int idx)
 {
 	drm_printf(p, "bridge[%u]: %ps\n", idx, bridge->funcs);
+
+	drm_printf(p, "\trefcount: %u\n", kref_read(&bridge->refcount));
+
 	drm_printf(p, "\ttype: [%d] %s\n",
 		   bridge->type,
 		   drm_get_connector_type_name(bridge->type));

From cb86408b1fc2e3f6fe45ebe8509a5404060e01e0 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 1 Aug 2025 19:05:23 +0200
Subject: [PATCH 0834/1455] list: add list_last_entry_or_null()

Add an equivalent of list_first_entry_or_null() to obtain the last element
of a list.

Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20250801-drm-bridge-alloc-getput-drm_bridge_get_next_bridge-v2-1-888912b0be13@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 include/linux/list.h | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/include/linux/list.h b/include/linux/list.h
index e7e28afd28f8..7f7657e41620 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -636,6 +636,20 @@ static inline void list_splice_tail_init(struct list_head *list,
 	pos__ != head__ ? list_entry(pos__, type, member) : NULL; \
 })
 
+/**
+ * list_last_entry_or_null - get the last element from a list
+ * @ptr:	the list head to take the element from.
+ * @type:	the type of the struct this is embedded in.
+ * @member:	the name of the list_head within the struct.
+ *
+ * Note that if the list is empty, it returns NULL.
+ */
+#define list_last_entry_or_null(ptr, type, member) ({ \
+	struct list_head *head__ = (ptr); \
+	struct list_head *pos__ = READ_ONCE(head__->prev); \
+	pos__ != head__ ? list_entry(pos__, type, member) : NULL; \
+})
+
 /**
  * list_next_entry - get the next element in list
  * @pos:	the type * to cursor

From d77ad5178e90f5aa4ce6085510b3b2f742abc5f0 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 1 Aug 2025 19:05:24 +0200
Subject: [PATCH 0835/1455] drm/bridge: add drm_bridge_chain_get_last_bridge()

Add an equivalent of drm_bridge_chain_get_first_bridge() to get the last
bridge.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250801-drm-bridge-alloc-getput-drm_bridge_get_next_bridge-v2-2-888912b0be13@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 include/drm/drm_bridge.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index 8d9d4fd078e7..788517ab00d3 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -1410,6 +1410,24 @@ drm_bridge_chain_get_first_bridge(struct drm_encoder *encoder)
 						       struct drm_bridge, chain_node));
 }
 
+/**
+ * drm_bridge_chain_get_last_bridge() - Get the last bridge in the chain
+ * @encoder: encoder object
+ *
+ * The refcount of the returned bridge is incremented. Use drm_bridge_put()
+ * when done with it.
+ *
+ * RETURNS:
+ * the last bridge in the chain, or NULL if @encoder has no bridge attached
+ * to it.
+ */
+static inline struct drm_bridge *
+drm_bridge_chain_get_last_bridge(struct drm_encoder *encoder)
+{
+	return drm_bridge_get(list_last_entry_or_null(&encoder->bridge_chain,
+						      struct drm_bridge, chain_node));
+}
+
 /**
  * drm_for_each_bridge_in_chain() - Iterate over all bridges present in a chain
  * @encoder: the encoder to iterate bridges on

From 5849eff7f06709c98baa7989806165fe30bd0cbe Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 1 Aug 2025 19:05:25 +0200
Subject: [PATCH 0836/1455] drm/bridge: imx93-mipi-dsi: use
 drm_bridge_chain_get_last_bridge()

Use drm_bridge_chain_get_last_bridge() instead of open coding a loop with
two invocations of drm_bridge_get_next_bridge() per iteration.

Besides being cleaner and more efficient, this change is necessary in
preparation for drm_bridge_get_next_bridge() to get a reference to the
returned bridge.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Reviewed-by: Liu Ying <victor.liu@nxp.com>
Link: https://lore.kernel.org/r/20250801-drm-bridge-alloc-getput-drm_bridge_get_next_bridge-v2-3-888912b0be13@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/bridge/imx/imx93-mipi-dsi.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/bridge/imx/imx93-mipi-dsi.c b/drivers/gpu/drm/bridge/imx/imx93-mipi-dsi.c
index bea8346515b8..8f7a0d46601a 100644
--- a/drivers/gpu/drm/bridge/imx/imx93-mipi-dsi.c
+++ b/drivers/gpu/drm/bridge/imx/imx93-mipi-dsi.c
@@ -492,14 +492,12 @@ static int imx93_dsi_get_phy_configure_opts(struct imx93_dsi *dsi,
 static enum drm_mode_status
 imx93_dsi_validate_mode(struct imx93_dsi *dsi, const struct drm_display_mode *mode)
 {
-	struct drm_bridge *bridge = dw_mipi_dsi_get_bridge(dsi->dmd);
+	struct drm_bridge *dmd_bridge = dw_mipi_dsi_get_bridge(dsi->dmd);
+	struct drm_bridge *last_bridge __free(drm_bridge_put) =
+		drm_bridge_chain_get_last_bridge(dmd_bridge->encoder);
 
-	/* Get the last bridge */
-	while (drm_bridge_get_next_bridge(bridge))
-		bridge = drm_bridge_get_next_bridge(bridge);
-
-	if ((bridge->ops & DRM_BRIDGE_OP_DETECT) &&
-	    (bridge->ops & DRM_BRIDGE_OP_EDID)) {
+	if ((last_bridge->ops & DRM_BRIDGE_OP_DETECT) &&
+	    (last_bridge->ops & DRM_BRIDGE_OP_EDID)) {
 		unsigned long pixel_clock_rate = mode->clock * 1000;
 		unsigned long rounded_rate;
 

From c3183f71c3f1f004a0ca0b84c45c40d8b84dce3d Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 1 Aug 2025 19:05:26 +0200
Subject: [PATCH 0837/1455] drm/omapdrm: use drm_bridge_chain_get_last_bridge()

Use drm_bridge_chain_get_last_bridge() instead of open coding a loop with
two invocations of drm_bridge_get_next_bridge() per iteration.

Besides being cleaner and more efficient, this change is necessary in
preparation for drm_bridge_get_next_bridge() to get a reference to the
returned bridge.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250801-drm-bridge-alloc-getput-drm_bridge_get_next_bridge-v2-4-888912b0be13@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/omapdrm/omap_drv.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 054b71dba6a7..794267f0f007 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -378,10 +378,8 @@ static int omap_display_id(struct omap_dss_device *output)
 	struct device_node *node = NULL;
 
 	if (output->bridge) {
-		struct drm_bridge *bridge = output->bridge;
-
-		while (drm_bridge_get_next_bridge(bridge))
-			bridge = drm_bridge_get_next_bridge(bridge);
+		struct drm_bridge *bridge __free(drm_bridge_put) =
+			drm_bridge_chain_get_last_bridge(output->bridge->encoder);
 
 		node = bridge->of_node;
 	}

From e4cedfd4f0fc839bb3c783d7e827e9755e1af5cf Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 1 Aug 2025 19:05:27 +0200
Subject: [PATCH 0838/1455] drm/bridge: add drm_bridge_is_last()

Some code needing to know whether a bridge is the last in a chain currently
call drm_bridge_get_next_bridge(). However drm_bridge_get_next_bridge()
will soon increment the refcount of the returned bridge, which would make
such code more annoying to write.

In preparation for drm_bridge_get_next_bridge() to increment the refcount,
as well as to simplify such code, introduce a simple bool function to tell
whether a bridge is the last in the chain.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250801-drm-bridge-alloc-getput-drm_bridge_get_next_bridge-v2-5-888912b0be13@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 include/drm/drm_bridge.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/drm/drm_bridge.h b/include/drm/drm_bridge.h
index 788517ab00d3..76e05930f50e 100644
--- a/include/drm/drm_bridge.h
+++ b/include/drm/drm_bridge.h
@@ -1321,6 +1321,11 @@ static inline struct drm_bridge *of_drm_find_bridge(struct device_node *np)
 }
 #endif
 
+static inline bool drm_bridge_is_last(struct drm_bridge *bridge)
+{
+	return list_is_last(&bridge->chain_node, &bridge->encoder->bridge_chain);
+}
+
 /**
  * drm_bridge_get_current_state() - Get the current bridge state
  * @bridge: bridge object

From 306734a1c76ce35fa613ea823a31ddbcdbf1eef5 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 1 Aug 2025 19:05:28 +0200
Subject: [PATCH 0839/1455] drm/display: bridge_connector: use
 drm_bridge_is_last()

Simplify code to know whether a bridge is the last in the chain by using
drm_bridge_is_last().

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250801-drm-bridge-alloc-getput-drm_bridge_get_next_bridge-v2-6-888912b0be13@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/display/drm_bridge_connector.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c
index 091c5335355a..54f3f16d64c7 100644
--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -751,12 +751,11 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 				return ERR_PTR(-EINVAL);
 		}
 
-		if (!drm_bridge_get_next_bridge(bridge))
+		if (drm_bridge_is_last(bridge))
 			connector_type = bridge->type;
 
 #ifdef CONFIG_OF
-		if (!drm_bridge_get_next_bridge(bridge) &&
-		    bridge->of_node)
+		if (drm_bridge_is_last(bridge) && bridge->of_node)
 			connector->fwnode = fwnode_handle_get(of_fwnode_handle(bridge->of_node));
 #endif
 

From 51bfda2e7690805fab08453c8b057de224724fc0 Mon Sep 17 00:00:00 2001
From: Lee Shawn C <shawn.c.lee@intel.com>
Date: Mon, 1 Sep 2025 05:57:20 +0000
Subject: [PATCH 0840/1455] drm/i915/hdmi: add debugfs to contorl HDMI bpc

While performing HDMI compliance testing, test equipment may request
different bpc output for signal measurement. However, display driver
typically determines the maximum available bpc based on HW bandwidth.
This change leverages the existing debugfs (intel_force_link_bpp)
to manage HDMI bpc, and making it easier to pass HDMI CTS.

v2: Using exist variable max_requested_bpc.
v3: Extend intel_force_link_bpp to support HDMI as suggested by Imre.
v4: Update commit message suggested by Jani.
v5: Remove unused header file.

Cc: Shankar Uma <uma.shankar@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Vidya Srinivas <vidya.srinivas@intel.com>
Signed-off-by: Lee Shawn C <shawn.c.lee@intel.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250901055721.219995-2-shawn.c.lee@intel.com
---
 drivers/gpu/drm/i915/display/g4x_hdmi.c      | 6 +-----
 drivers/gpu/drm/i915/display/intel_hdmi.c    | 4 ++++
 drivers/gpu/drm/i915/display/intel_link_bw.c | 6 +-----
 3 files changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/g4x_hdmi.c b/drivers/gpu/drm/i915/display/g4x_hdmi.c
index 108ebd97f9e4..f6e2d1ed5639 100644
--- a/drivers/gpu/drm/i915/display/g4x_hdmi.c
+++ b/drivers/gpu/drm/i915/display/g4x_hdmi.c
@@ -20,7 +20,6 @@
 #include "intel_dp_aux.h"
 #include "intel_dpio_phy.h"
 #include "intel_encoder.h"
-#include "intel_fdi.h"
 #include "intel_fifo_underrun.h"
 #include "intel_hdmi.h"
 #include "intel_hotplug.h"
@@ -136,11 +135,8 @@ static int g4x_hdmi_compute_config(struct intel_encoder *encoder,
 	struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->uapi.state);
 	struct intel_crtc *crtc = to_intel_crtc(crtc_state->uapi.crtc);
 
-	if (HAS_PCH_SPLIT(display)) {
+	if (HAS_PCH_SPLIT(display))
 		crtc_state->has_pch_encoder = true;
-		if (!intel_fdi_compute_pipe_bpp(crtc_state))
-			return -EINVAL;
-	}
 
 	if (display->platform.g4x)
 		crtc_state->has_hdmi_sink = g4x_compute_has_hdmi_sink(state, crtc);
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index cbee628eb26b..027e8ed0cea8 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -55,6 +55,7 @@
 #include "intel_display_regs.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
+#include "intel_fdi.h"
 #include "intel_gmbus.h"
 #include "intel_hdcp.h"
 #include "intel_hdcp_regs.h"
@@ -2345,6 +2346,9 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK)
 		pipe_config->pixel_multiplier = 2;
 
+	if (!intel_fdi_compute_pipe_bpp(pipe_config))
+		return -EINVAL;
+
 	pipe_config->has_audio =
 		intel_hdmi_has_audio(encoder, pipe_config, conn_state) &&
 		intel_audio_compute_config(encoder, pipe_config, conn_state);
diff --git a/drivers/gpu/drm/i915/display/intel_link_bw.c b/drivers/gpu/drm/i915/display/intel_link_bw.c
index 3caef7f9c7c4..d194a366ff10 100644
--- a/drivers/gpu/drm/i915/display/intel_link_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_link_bw.c
@@ -449,6 +449,7 @@ void intel_link_bw_connector_debugfs_add(struct intel_connector *connector)
 	switch (connector->base.connector_type) {
 	case DRM_MODE_CONNECTOR_DisplayPort:
 	case DRM_MODE_CONNECTOR_eDP:
+	case DRM_MODE_CONNECTOR_HDMIA:
 		break;
 	case DRM_MODE_CONNECTOR_VGA:
 	case DRM_MODE_CONNECTOR_SVIDEO:
@@ -457,11 +458,6 @@ void intel_link_bw_connector_debugfs_add(struct intel_connector *connector)
 		if (HAS_FDI(display))
 			break;
 
-		return;
-	case DRM_MODE_CONNECTOR_HDMIA:
-		if (HAS_FDI(display) && !HAS_DDI(display))
-			break;
-
 		return;
 	default:
 		return;

From 5f9bab808180c47426346a4f8f494a642e9f69de Mon Sep 17 00:00:00 2001
From: Lee Shawn C <shawn.c.lee@intel.com>
Date: Mon, 1 Sep 2025 05:57:21 +0000
Subject: [PATCH 0841/1455] drm/i915: compute pipe bpp from link bandwidth
 management

Since intel_fdi_compute_pipe_bpp() is no longer FDI-specific and
now applies to all connectors. Move it to intel_link_bw.c,
and rename to intel_link_bw_compute_pipe_bpp().

v2: Remove unused header file.

Cc: Shankar Uma <uma.shankar@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Imre Deak <imre.deak@intel.com>
Cc: Vidya Srinivas <vidya.srinivas@intel.com>
Signed-off-by: Lee Shawn C <shawn.c.lee@intel.com>
[Imre: Remove unused intel_fdi.h include from intel_hdmi.c]
Reviewed-by: Imre Deak <imre.deak@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250901055721.219995-3-shawn.c.lee@intel.com
---
 drivers/gpu/drm/i915/display/intel_crt.c     |  5 ++--
 drivers/gpu/drm/i915/display/intel_fdi.c     | 28 --------------------
 drivers/gpu/drm/i915/display/intel_fdi.h     |  1 -
 drivers/gpu/drm/i915/display/intel_hdmi.c    |  4 +--
 drivers/gpu/drm/i915/display/intel_link_bw.c | 28 ++++++++++++++++++++
 drivers/gpu/drm/i915/display/intel_link_bw.h |  1 +
 drivers/gpu/drm/i915/display/intel_lvds.c    |  3 ++-
 drivers/gpu/drm/i915/display/intel_sdvo.c    |  4 +--
 8 files changed, 38 insertions(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_crt.c b/drivers/gpu/drm/i915/display/intel_crt.c
index 898c5d9e8f7a..31e68047f217 100644
--- a/drivers/gpu/drm/i915/display/intel_crt.c
+++ b/drivers/gpu/drm/i915/display/intel_crt.c
@@ -50,6 +50,7 @@
 #include "intel_gmbus.h"
 #include "intel_hotplug.h"
 #include "intel_hotplug_irq.h"
+#include "intel_link_bw.h"
 #include "intel_load_detect.h"
 #include "intel_pch_display.h"
 #include "intel_pch_refclk.h"
@@ -421,7 +422,7 @@ static int pch_crt_compute_config(struct intel_encoder *encoder,
 		return -EINVAL;
 
 	crtc_state->has_pch_encoder = true;
-	if (!intel_fdi_compute_pipe_bpp(crtc_state))
+	if (!intel_link_bw_compute_pipe_bpp(crtc_state))
 		return -EINVAL;
 
 	crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
@@ -446,7 +447,7 @@ static int hsw_crt_compute_config(struct intel_encoder *encoder,
 		return -EINVAL;
 
 	crtc_state->has_pch_encoder = true;
-	if (!intel_fdi_compute_pipe_bpp(crtc_state))
+	if (!intel_link_bw_compute_pipe_bpp(crtc_state))
 		return -EINVAL;
 
 	crtc_state->output_format = INTEL_OUTPUT_FORMAT_RGB;
diff --git a/drivers/gpu/drm/i915/display/intel_fdi.c b/drivers/gpu/drm/i915/display/intel_fdi.c
index 8039a84671cc..59a36b3a22c1 100644
--- a/drivers/gpu/drm/i915/display/intel_fdi.c
+++ b/drivers/gpu/drm/i915/display/intel_fdi.c
@@ -292,34 +292,6 @@ int intel_fdi_link_freq(struct intel_display *display,
 		return display->fdi.pll_freq;
 }
 
-/**
- * intel_fdi_compute_pipe_bpp - compute pipe bpp limited by max link bpp
- * @crtc_state: the crtc state
- *
- * Compute the pipe bpp limited by the CRTC's maximum link bpp. Encoders can
- * call this function during state computation in the simple case where the
- * link bpp will always match the pipe bpp. This is the case for all non-DP
- * encoders, while DP encoders will use a link bpp lower than pipe bpp in case
- * of DSC compression.
- *
- * Returns %true in case of success, %false if pipe bpp would need to be
- * reduced below its valid range.
- */
-bool intel_fdi_compute_pipe_bpp(struct intel_crtc_state *crtc_state)
-{
-	int pipe_bpp = min(crtc_state->pipe_bpp,
-			   fxp_q4_to_int(crtc_state->max_link_bpp_x16));
-
-	pipe_bpp = rounddown(pipe_bpp, 2 * 3);
-
-	if (pipe_bpp < 6 * 3)
-		return false;
-
-	crtc_state->pipe_bpp = pipe_bpp;
-
-	return true;
-}
-
 int ilk_fdi_compute_config(struct intel_crtc *crtc,
 			   struct intel_crtc_state *pipe_config)
 {
diff --git a/drivers/gpu/drm/i915/display/intel_fdi.h b/drivers/gpu/drm/i915/display/intel_fdi.h
index ad5e103c38a8..1cd08df9b0c2 100644
--- a/drivers/gpu/drm/i915/display/intel_fdi.h
+++ b/drivers/gpu/drm/i915/display/intel_fdi.h
@@ -20,7 +20,6 @@ struct intel_link_bw_limits;
 int intel_fdi_add_affected_crtcs(struct intel_atomic_state *state);
 int intel_fdi_link_freq(struct intel_display *display,
 			const struct intel_crtc_state *pipe_config);
-bool intel_fdi_compute_pipe_bpp(struct intel_crtc_state *crtc_state);
 int ilk_fdi_compute_config(struct intel_crtc *intel_crtc,
 			   struct intel_crtc_state *pipe_config);
 int intel_fdi_atomic_check_link(struct intel_atomic_state *state,
diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 027e8ed0cea8..0a8af7bb80ef 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -55,12 +55,12 @@
 #include "intel_display_regs.h"
 #include "intel_display_types.h"
 #include "intel_dp.h"
-#include "intel_fdi.h"
 #include "intel_gmbus.h"
 #include "intel_hdcp.h"
 #include "intel_hdcp_regs.h"
 #include "intel_hdcp_shim.h"
 #include "intel_hdmi.h"
+#include "intel_link_bw.h"
 #include "intel_lspcon.h"
 #include "intel_panel.h"
 #include "intel_pfit.h"
@@ -2346,7 +2346,7 @@ int intel_hdmi_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK)
 		pipe_config->pixel_multiplier = 2;
 
-	if (!intel_fdi_compute_pipe_bpp(pipe_config))
+	if (!intel_link_bw_compute_pipe_bpp(pipe_config))
 		return -EINVAL;
 
 	pipe_config->has_audio =
diff --git a/drivers/gpu/drm/i915/display/intel_link_bw.c b/drivers/gpu/drm/i915/display/intel_link_bw.c
index d194a366ff10..f52dee0ea412 100644
--- a/drivers/gpu/drm/i915/display/intel_link_bw.c
+++ b/drivers/gpu/drm/i915/display/intel_link_bw.c
@@ -164,6 +164,34 @@ int intel_link_bw_reduce_bpp(struct intel_atomic_state *state,
 	return ret;
 }
 
+/**
+ * intel_link_bw_compute_pipe_bpp - compute pipe bpp limited by max link bpp
+ * @crtc_state: the crtc state
+ *
+ * Compute the pipe bpp limited by the CRTC's maximum link bpp. Encoders can
+ * call this function during state computation in the simple case where the
+ * link bpp will always match the pipe bpp. This is the case for all non-DP
+ * encoders, while DP encoders will use a link bpp lower than pipe bpp in case
+ * of DSC compression.
+ *
+ * Returns %true in case of success, %false if pipe bpp would need to be
+ * reduced below its valid range.
+ */
+bool intel_link_bw_compute_pipe_bpp(struct intel_crtc_state *crtc_state)
+{
+	int pipe_bpp = min(crtc_state->pipe_bpp,
+			   fxp_q4_to_int(crtc_state->max_link_bpp_x16));
+
+	pipe_bpp = rounddown(pipe_bpp, 2 * 3);
+
+	if (pipe_bpp < 6 * 3)
+		return false;
+
+	crtc_state->pipe_bpp = pipe_bpp;
+
+	return true;
+}
+
 /**
  * intel_link_bw_set_bpp_limit_for_pipe - set link bpp limit for a pipe to its minimum
  * @state: atomic state
diff --git a/drivers/gpu/drm/i915/display/intel_link_bw.h b/drivers/gpu/drm/i915/display/intel_link_bw.h
index b499042e62b1..95ab7c50c61d 100644
--- a/drivers/gpu/drm/i915/display/intel_link_bw.h
+++ b/drivers/gpu/drm/i915/display/intel_link_bw.h
@@ -27,6 +27,7 @@ int intel_link_bw_reduce_bpp(struct intel_atomic_state *state,
 			     struct intel_link_bw_limits *limits,
 			     u8 pipe_mask,
 			     const char *reason);
+bool intel_link_bw_compute_pipe_bpp(struct intel_crtc_state *crtc_state);
 bool intel_link_bw_set_bpp_limit_for_pipe(struct intel_atomic_state *state,
 					  const struct intel_link_bw_limits *old_limits,
 					  struct intel_link_bw_limits *new_limits,
diff --git a/drivers/gpu/drm/i915/display/intel_lvds.c b/drivers/gpu/drm/i915/display/intel_lvds.c
index 7e48a235c99f..48f4d8ed4f15 100644
--- a/drivers/gpu/drm/i915/display/intel_lvds.c
+++ b/drivers/gpu/drm/i915/display/intel_lvds.c
@@ -48,6 +48,7 @@
 #include "intel_dpll.h"
 #include "intel_fdi.h"
 #include "intel_gmbus.h"
+#include "intel_link_bw.h"
 #include "intel_lvds.h"
 #include "intel_lvds_regs.h"
 #include "intel_panel.h"
@@ -433,7 +434,7 @@ static int intel_lvds_compute_config(struct intel_encoder *encoder,
 
 	if (HAS_PCH_SPLIT(display)) {
 		crtc_state->has_pch_encoder = true;
-		if (!intel_fdi_compute_pipe_bpp(crtc_state))
+		if (!intel_link_bw_compute_pipe_bpp(crtc_state))
 			return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 87aff2754f69..8848a33fe9cc 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -47,11 +47,11 @@
 #include "intel_display_driver.h"
 #include "intel_display_regs.h"
 #include "intel_display_types.h"
-#include "intel_fdi.h"
 #include "intel_fifo_underrun.h"
 #include "intel_gmbus.h"
 #include "intel_hdmi.h"
 #include "intel_hotplug.h"
+#include "intel_link_bw.h"
 #include "intel_panel.h"
 #include "intel_sdvo.h"
 #include "intel_sdvo_regs.h"
@@ -1367,7 +1367,7 @@ static int intel_sdvo_compute_config(struct intel_encoder *encoder,
 
 	if (HAS_PCH_SPLIT(display)) {
 		pipe_config->has_pch_encoder = true;
-		if (!intel_fdi_compute_pipe_bpp(pipe_config))
+		if (!intel_link_bw_compute_pipe_bpp(pipe_config))
 			return -EINVAL;
 	}
 

From b63f008f395ca5f6bc89123db97440bdc19981c4 Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Tue, 10 Jun 2025 16:05:44 +0200
Subject: [PATCH 0842/1455] drm/msm/dsi/phy: Toggle back buffer resync after
 preparing PLL

According to Hardware Programming Guide for DSI PHY, the retime buffer
resync should be done after PLL clock users (byte_clk and intf_byte_clk)
are enabled.  Downstream also does it as part of configuring the PLL.

Driver was only turning off the resync FIFO buffer, but never bringing it
on again.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/657823/
Link: https://lore.kernel.org/r/20250610-b4-sm8750-display-v6-6-ee633e3ddbff@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
index c3bd3f89434e..1538945eaeb1 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -491,6 +491,10 @@ static int dsi_pll_7nm_vco_prepare(struct clk_hw *hw)
 	if (pll_7nm->slave)
 		dsi_pll_enable_global_clk(pll_7nm->slave);
 
+	writel(0x1, pll_7nm->phy->base + REG_DSI_7nm_PHY_CMN_RBUF_CTRL);
+	if (pll_7nm->slave)
+		writel(0x1, pll_7nm->slave->phy->base + REG_DSI_7nm_PHY_CMN_RBUF_CTRL);
+
 error:
 	return rc;
 }

From 3bb6433ecb65ab91862e39a9a788beecd38e10de Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Tue, 10 Jun 2025 16:05:45 +0200
Subject: [PATCH 0843/1455] drm/msm/dsi/phy: Define PHY_CMN_CTRL_0 bitfields

Add bitfields for PHY_CMN_CTRL_0 registers to avoid hard-coding bit
masks and shifts and make the code a bit more readable.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/657818/
Link: https://lore.kernel.org/r/20250610-b4-sm8750-display-v6-7-ee633e3ddbff@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c        | 16 +++++++++++-----
 .../drm/msm/registers/display/dsi_phy_7nm.xml    | 11 ++++++++++-
 2 files changed, 21 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
index 1538945eaeb1..eb912aa3937d 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -385,18 +385,23 @@ static int dsi_pll_7nm_lock_status(struct dsi_pll_7nm *pll)
 
 static void dsi_pll_disable_pll_bias(struct dsi_pll_7nm *pll)
 {
-	u32 data = readl(pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	u32 data;
 
+	data = readl(pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	data &= ~DSI_7nm_PHY_CMN_CTRL_0_PLL_SHUTDOWNB;
 	writel(0, pll->phy->pll_base + REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES);
-	writel(data & ~BIT(5), pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	writel(data, pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 	ndelay(250);
 }
 
 static void dsi_pll_enable_pll_bias(struct dsi_pll_7nm *pll)
 {
-	u32 data = readl(pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	u32 data;
+
+	data = readl(pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	data |= DSI_7nm_PHY_CMN_CTRL_0_PLL_SHUTDOWNB;
+	writel(data, pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 
-	writel(data | BIT(5), pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 	writel(0xc0, pll->phy->pll_base + REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES);
 	ndelay(250);
 }
@@ -1036,7 +1041,8 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy,
 	}
 
 	/* de-assert digital and pll power down */
-	data = BIT(6) | BIT(5);
+	data = DSI_7nm_PHY_CMN_CTRL_0_DIGTOP_PWRDN_B |
+	       DSI_7nm_PHY_CMN_CTRL_0_PLL_SHUTDOWNB;
 	writel(data, base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 
 	/* Assert PLL core reset */
diff --git a/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml b/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml
index 4e5ac0f25dea..f41516dd0567 100644
--- a/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml
+++ b/drivers/gpu/drm/msm/registers/display/dsi_phy_7nm.xml
@@ -22,7 +22,16 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 	<reg32 offset="0x00018" name="GLBL_CTRL"/>
 	<reg32 offset="0x0001c" name="RBUF_CTRL"/>
 	<reg32 offset="0x00020" name="VREG_CTRL_0"/>
-	<reg32 offset="0x00024" name="CTRL_0"/>
+	<reg32 offset="0x00024" name="CTRL_0">
+		<bitfield name="CLKSL_SHUTDOWNB" pos="7" type="boolean"/>
+		<bitfield name="DIGTOP_PWRDN_B" pos="6" type="boolean"/>
+		<bitfield name="PLL_SHUTDOWNB" pos="5" type="boolean"/>
+		<bitfield name="DLN3_SHUTDOWNB" pos="4" type="boolean"/>
+		<bitfield name="DLN2_SHUTDOWNB" pos="3" type="boolean"/>
+		<bitfield name="CLK_SHUTDOWNB" pos="2" type="boolean"/>
+		<bitfield name="DLN1_SHUTDOWNB" pos="1" type="boolean"/>
+		<bitfield name="DLN0_SHUTDOWNB" pos="0" type="boolean"/>
+	</reg32>
 	<reg32 offset="0x00028" name="CTRL_1"/>
 	<reg32 offset="0x0002c" name="CTRL_2"/>
 	<reg32 offset="0x00030" name="CTRL_3"/>

From 5ddcb0cb9d10e6e70a68e0cb8f0b8e3a7eb8ccaf Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Tue, 10 Jun 2025 16:05:47 +0200
Subject: [PATCH 0844/1455] drm/msm/dsi/phy_7nm: Fix missing initial VCO rate

Driver unconditionally saves current state on first init in
dsi_pll_7nm_init(), but does not save the VCO rate, only some of the
divider registers.  The state is then restored during probe/enable via
msm_dsi_phy_enable() -> msm_dsi_phy_pll_restore_state() ->
dsi_7nm_pll_restore_state().

Restoring calls dsi_pll_7nm_vco_set_rate() with
pll_7nm->vco_current_rate=0, which basically overwrites existing rate of
VCO and messes with clock hierarchy, by setting frequency to 0 to clock
tree.  This makes anyway little sense - VCO rate was not saved, so
should not be restored.

If PLL was not configured configure it to minimum rate to avoid glitches
and configuring entire in clock hierarchy to 0 Hz.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/657827/
Link: https://lore.kernel.org/r/20250610-b4-sm8750-display-v6-9-ee633e3ddbff@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
index eb912aa3937d..2c2bbda46c78 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -846,6 +846,12 @@ static int dsi_pll_7nm_init(struct msm_dsi_phy *phy)
 
 	/* TODO: Remove this when we have proper display handover support */
 	msm_dsi_phy_pll_save_state(phy);
+	/*
+	 * Store also proper vco_current_rate, because its value will be used in
+	 * dsi_7nm_pll_restore_state().
+	 */
+	if (!dsi_pll_7nm_vco_recalc_rate(&pll_7nm->clk_hw, VCO_REF_CLK_RATE))
+		pll_7nm->vco_current_rate = pll_7nm->phy->cfg->min_pll_rate;
 
 	return 0;
 }

From be5590c384f3e57b1b556be0b88ec62246948315 Mon Sep 17 00:00:00 2001
From: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Date: Tue, 2 Sep 2025 16:02:57 +0530
Subject: [PATCH 0845/1455] drm/xe/vf: Enable CCS save/restore only on
 supported GUC versions

CCS save/restore is supported starting with GuC 70.48.0 (compatibility
version 1.23.0). Gate the feature on the GuC firmware version and keep it
disabled on older or unsupported versions.

Fixes: f3009272ff2e ("drm/xe/vf: Create contexts for CCS read write")
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Andi Shyti <andi.shyti@kernel.org>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250902103256.21658-2-satyanarayana.k.v.p@intel.com
---
 drivers/gpu/drm/xe/xe_device.c         |  2 +-
 drivers/gpu/drm/xe/xe_sriov.c          | 14 ++---
 drivers/gpu/drm/xe/xe_sriov.h          |  2 +-
 drivers/gpu/drm/xe/xe_sriov_vf.c       | 87 +++++++++++++++++++++++---
 drivers/gpu/drm/xe/xe_sriov_vf.h       |  4 ++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c   |  5 +-
 drivers/gpu/drm/xe/xe_sriov_vf_types.h |  5 ++
 7 files changed, 95 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 9e4773a17ef8..9e2952c9c06a 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -950,7 +950,7 @@ int xe_device_probe(struct xe_device *xe)
 
 	xe_vsec_init(xe);
 
-	err = xe_sriov_late_init(xe);
+	err = xe_sriov_init_late(xe);
 	if (err)
 		goto err_unregister_display;
 
diff --git a/drivers/gpu/drm/xe/xe_sriov.c b/drivers/gpu/drm/xe/xe_sriov.c
index 87911fb4eea7..7d2d6de2aabf 100644
--- a/drivers/gpu/drm/xe/xe_sriov.c
+++ b/drivers/gpu/drm/xe/xe_sriov.c
@@ -160,19 +160,15 @@ const char *xe_sriov_function_name(unsigned int n, char *buf, size_t size)
 }
 
 /**
- * xe_sriov_late_init() - SR-IOV late initialization functions.
+ * xe_sriov_init_late() - SR-IOV late initialization functions.
  * @xe: the &xe_device to initialize
  *
- * On VF this function will initialize code for CCS migration.
- *
  * Return: 0 on success or a negative error code on failure.
  */
-int xe_sriov_late_init(struct xe_device *xe)
+int xe_sriov_init_late(struct xe_device *xe)
 {
-	int err = 0;
+	if (IS_SRIOV_VF(xe))
+		return xe_sriov_vf_init_late(xe);
 
-	if (IS_VF_CCS_INIT_NEEDED(xe))
-		err = xe_sriov_vf_ccs_init(xe);
-
-	return err;
+	return 0;
 }
diff --git a/drivers/gpu/drm/xe/xe_sriov.h b/drivers/gpu/drm/xe/xe_sriov.h
index 0e0c1abf2d14..6db45df55615 100644
--- a/drivers/gpu/drm/xe/xe_sriov.h
+++ b/drivers/gpu/drm/xe/xe_sriov.h
@@ -18,7 +18,7 @@ const char *xe_sriov_function_name(unsigned int n, char *buf, size_t len);
 void xe_sriov_probe_early(struct xe_device *xe);
 void xe_sriov_print_info(struct xe_device *xe, struct drm_printer *p);
 int xe_sriov_init(struct xe_device *xe);
-int xe_sriov_late_init(struct xe_device *xe);
+int xe_sriov_init_late(struct xe_device *xe);
 
 static inline enum xe_sriov_mode xe_device_sriov_mode(const struct xe_device *xe)
 {
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index 5de81f213d83..cb062328551a 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -10,6 +10,7 @@
 #include "xe_gt.h"
 #include "xe_gt_sriov_printk.h"
 #include "xe_gt_sriov_vf.h"
+#include "xe_guc.h"
 #include "xe_guc_ct.h"
 #include "xe_guc_submit.h"
 #include "xe_irq.h"
@@ -18,6 +19,7 @@
 #include "xe_sriov.h"
 #include "xe_sriov_printk.h"
 #include "xe_sriov_vf.h"
+#include "xe_sriov_vf_ccs.h"
 #include "xe_tile_sriov_vf.h"
 
 /**
@@ -127,16 +129,66 @@
  *      |                               |                               |
  */
 
-static bool vf_migration_supported(struct xe_device *xe)
+/**
+ * xe_sriov_vf_migration_supported - Report whether SR-IOV VF migration is
+ * supported or not.
+ * @xe: the &xe_device to check
+ *
+ * Returns: true if VF migration is supported, false otherwise.
+ */
+bool xe_sriov_vf_migration_supported(struct xe_device *xe)
+{
+	xe_assert(xe, IS_SRIOV_VF(xe));
+	return xe->sriov.vf.migration.enabled;
+}
+
+static void vf_disable_migration(struct xe_device *xe, const char *fmt, ...)
+{
+	struct va_format vaf;
+	va_list va_args;
+
+	xe_assert(xe, IS_SRIOV_VF(xe));
+
+	va_start(va_args, fmt);
+	vaf.fmt = fmt;
+	vaf.va  = &va_args;
+	xe_sriov_notice(xe, "migration disabled: %pV\n", &vaf);
+	va_end(va_args);
+
+	xe->sriov.vf.migration.enabled = false;
+}
+
+static void migration_worker_func(struct work_struct *w);
+
+static void vf_migration_init_early(struct xe_device *xe)
 {
 	/*
 	 * TODO: Add conditions to allow specific platforms, when they're
 	 * supported at production quality.
 	 */
-	return IS_ENABLED(CONFIG_DRM_XE_DEBUG);
-}
+	if (!IS_ENABLED(CONFIG_DRM_XE_DEBUG))
+		return vf_disable_migration(xe,
+					    "experimental feature not available on production builds");
 
-static void migration_worker_func(struct work_struct *w);
+	if (GRAPHICS_VER(xe) < 20)
+		return vf_disable_migration(xe, "requires gfx version >= 20, but only %u found",
+					    GRAPHICS_VER(xe));
+
+	if (!IS_DGFX(xe)) {
+		struct xe_uc_fw_version guc_version;
+
+		xe_gt_sriov_vf_guc_versions(xe_device_get_gt(xe, 0), NULL, &guc_version);
+		if (MAKE_GUC_VER_STRUCT(guc_version) < MAKE_GUC_VER(1, 23, 0))
+			return vf_disable_migration(xe,
+						    "CCS migration requires GuC ABI >= 1.23 but only %u.%u found",
+						    guc_version.major, guc_version.minor);
+	}
+
+	INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
+
+	xe->sriov.vf.migration.enabled = true;
+	xe_sriov_dbg(xe, "migration support enabled\n");
+}
 
 /**
  * xe_sriov_vf_init_early - Initialize SR-IOV VF specific data.
@@ -144,10 +196,7 @@ static void migration_worker_func(struct work_struct *w);
  */
 void xe_sriov_vf_init_early(struct xe_device *xe)
 {
-	INIT_WORK(&xe->sriov.vf.migration.worker, migration_worker_func);
-
-	if (!vf_migration_supported(xe))
-		xe_sriov_info(xe, "migration not supported by this module version\n");
+	vf_migration_init_early(xe);
 }
 
 /**
@@ -302,8 +351,8 @@ static void vf_post_migration_recovery(struct xe_device *xe)
 	xe_pm_runtime_get(xe);
 	vf_post_migration_shutdown(xe);
 
-	if (!vf_migration_supported(xe)) {
-		xe_sriov_err(xe, "migration not supported by this module version\n");
+	if (!xe_sriov_vf_migration_supported(xe)) {
+		xe_sriov_err(xe, "migration is not supported\n");
 		err = -ENOTRECOVERABLE;
 		goto fail;
 	}
@@ -378,3 +427,21 @@ void xe_sriov_vf_start_migration_recovery(struct xe_device *xe)
 	drm_info(&xe->drm, "VF migration recovery %s\n", started ?
 		 "scheduled" : "already in progress");
 }
+
+/**
+ * xe_sriov_vf_init_late() - SR-IOV VF late initialization functions.
+ * @xe: the &xe_device to initialize
+ *
+ * This function initializes code for CCS migration.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+int xe_sriov_vf_init_late(struct xe_device *xe)
+{
+	int err = 0;
+
+	if (xe_sriov_vf_migration_supported(xe))
+		err = xe_sriov_vf_ccs_init(xe);
+
+	return err;
+}
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
index 7b8622cff2b7..3bf3364799ad 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
@@ -6,9 +6,13 @@
 #ifndef _XE_SRIOV_VF_H_
 #define _XE_SRIOV_VF_H_
 
+#include <linux/types.h>
+
 struct xe_device;
 
 void xe_sriov_vf_init_early(struct xe_device *xe);
+int xe_sriov_vf_init_late(struct xe_device *xe);
 void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
+bool xe_sriov_vf_migration_supported(struct xe_device *xe);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 4872e43eb440..908590fa79d4 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -264,9 +264,8 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 	u32 flags;
 	int err;
 
-	xe_assert(xe, IS_SRIOV_VF(xe));
-	xe_assert(xe, !IS_DGFX(xe));
-	xe_assert(xe, xe_device_has_flat_ccs(xe));
+	if (!IS_VF_CCS_INIT_NEEDED(xe))
+		return 0;
 
 	for_each_ccs_rw_ctx(ctx_id) {
 		ctx = &tile->sriov.vf.ccs[ctx_id];
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
index 24a873c50c49..2c94d1f92187 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
@@ -35,6 +35,11 @@ struct xe_device_vf {
 		struct work_struct worker;
 		/** @migration.gt_flags: Per-GT request flags for VF migration recovery */
 		unsigned long gt_flags;
+		/**
+		 * @migration.enabled: flag indicating if migration support
+		 * was enabled or not due to missing prerequisites
+		 */
+		bool enabled;
 	} migration;
 
 	/** @ccs: VF CCS state data */

From 0317e0e224bcda9706a70c8bbd5aeb8b10b6ea0a Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Tue, 12 Aug 2025 14:19:18 -0400
Subject: [PATCH 0846/1455] drm/amdgpu: Allow more flags to be set on gem
 create.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The GEM create flag AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE
specifies that gem memory contains sensitive information and
should be cleared to prevent snooping.

The COHERENT and UNCACHED gem create flags enable memory
features related to sharing memory across devices.

For CRIU we need to re-create KFD BOs through the
GEM_CREATE IOCTL, so allow those KFD specific flags here as well.
This will also aid us in the future and allows to move
the KFD components over using the render node for allocations.

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 10 +---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h | 14 ++++++++++++++
 2 files changed, 15 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index d5e685c5e28b..dff5f525ed1b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -442,15 +442,7 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	int r;
 
 	/* reject invalid gem flags */
-	if (flags & ~(AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED |
-		      AMDGPU_GEM_CREATE_NO_CPU_ACCESS |
-		      AMDGPU_GEM_CREATE_CPU_GTT_USWC |
-		      AMDGPU_GEM_CREATE_VRAM_CLEARED |
-		      AMDGPU_GEM_CREATE_VM_ALWAYS_VALID |
-		      AMDGPU_GEM_CREATE_EXPLICIT_SYNC |
-		      AMDGPU_GEM_CREATE_ENCRYPTED |
-		      AMDGPU_GEM_CREATE_GFX12_DCC |
-		      AMDGPU_GEM_CREATE_DISCARDABLE))
+	if (flags & ~AMDGPU_GEM_CREATE_SETTABLE_MASK)
 		return -EINVAL;
 
 	/* reject invalid gem domains */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
index b51e8f95ee86..b3047d73fe07 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
@@ -71,4 +71,18 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *filp);
 
+#define AMDGPU_GEM_CREATE_SETTABLE_MASK	(AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED | \
+	AMDGPU_GEM_CREATE_NO_CPU_ACCESS | \
+	AMDGPU_GEM_CREATE_CPU_GTT_USWC | \
+	AMDGPU_GEM_CREATE_VRAM_CLEARED | \
+	AMDGPU_GEM_CREATE_VM_ALWAYS_VALID | \
+	AMDGPU_GEM_CREATE_EXPLICIT_SYNC | \
+	AMDGPU_GEM_CREATE_VRAM_WIPE_ON_RELEASE | \
+	AMDGPU_GEM_CREATE_ENCRYPTED | \
+	AMDGPU_GEM_CREATE_GFX12_DCC | \
+	AMDGPU_GEM_CREATE_DISCARDABLE | \
+	AMDGPU_GEM_CREATE_COHERENT | \
+	AMDGPU_GEM_CREATE_UNCACHED | \
+	AMDGPU_GEM_CREATE_EXT_COHERENT)
+
 #endif

From f9db1fc52ceb42f4a18506693349316f5e209ba6 Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Mon, 16 Jun 2025 09:47:42 -0400
Subject: [PATCH 0847/1455] drm/amdgpu: Add ioctl to get all gem handles for a
 process
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add new ioctl DRM_IOCTL_AMDGPU_GEM_LIST_HANDLES.

This ioctl returns a list of bos with their handles, sizes,
and flags and domains.

This ioctl is meant to be used during CRIU checkpoint and
provide information needed to reconstruct the bos
in CRIU restore.

Userspace for this and the next change can be found at
https://github.com/checkpoint-restore/criu/pull/2613

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 79 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h |  2 +
 include/uapi/drm/amdgpu_drm.h           | 34 +++++++++++
 4 files changed, 116 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 17f754d1135d..b42a2a32b0b7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -3051,6 +3051,7 @@ const struct drm_ioctl_desc amdgpu_ioctls_kms[] = {
 	DRM_IOCTL_DEF_DRV(AMDGPU_USERQ, amdgpu_userq_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(AMDGPU_USERQ_SIGNAL, amdgpu_userq_signal_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(AMDGPU_USERQ_WAIT, amdgpu_userq_wait_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(AMDGPU_GEM_LIST_HANDLES, amdgpu_gem_list_handles_ioctl, DRM_AUTH|DRM_RENDER_ALLOW),
 };
 
 static const struct drm_driver amdgpu_kms_driver = {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index dff5f525ed1b..bddb8d7475b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -1024,6 +1024,85 @@ out:
 	return r;
 }
 
+/**
+ * drm_amdgpu_gem_list_handles_ioctl - get information about a process' buffer objects
+ *
+ * @dev: drm device pointer
+ * @data: drm_amdgpu_gem_list_handles
+ * @filp: drm file pointer
+ *
+ * num_entries is set as an input to the size of the entries array.
+ * num_entries is sent back as output as the number of bos in the process.
+ * If that number is larger than the size of the array, the ioctl must
+ * be retried.
+ *
+ * Returns:
+ * 0 for success, -errno for errors.
+ */
+int amdgpu_gem_list_handles_ioctl(struct drm_device *dev, void *data,
+				  struct drm_file *filp)
+{
+	struct drm_amdgpu_gem_list_handles *args = data;
+	struct drm_amdgpu_gem_list_handles_entry *bo_entries;
+	struct drm_gem_object *gobj;
+	int id, ret = 0;
+	int bo_index = 0;
+	int num_bos = 0;
+
+	spin_lock(&filp->table_lock);
+	idr_for_each_entry(&filp->object_idr, gobj, id)
+		num_bos += 1;
+	spin_unlock(&filp->table_lock);
+
+	if (args->num_entries < num_bos) {
+		args->num_entries = num_bos;
+		return 0;
+	}
+
+	if (num_bos == 0) {
+		args->num_entries = 0;
+		return 0;
+	}
+
+	bo_entries = kvcalloc(num_bos, sizeof(*bo_entries), GFP_KERNEL);
+	if (!bo_entries)
+		return -ENOMEM;
+
+	spin_lock(&filp->table_lock);
+	idr_for_each_entry(&filp->object_idr, gobj, id) {
+		struct amdgpu_bo *bo = gem_to_amdgpu_bo(gobj);
+		struct drm_amdgpu_gem_list_handles_entry *bo_entry;
+
+		if (bo_index >= num_bos) {
+			ret = -EAGAIN;
+			break;
+		}
+
+		bo_entry = &bo_entries[bo_index];
+
+		bo_entry->size = amdgpu_bo_size(bo);
+		bo_entry->alloc_flags = bo->flags & AMDGPU_GEM_CREATE_SETTABLE_MASK;
+		bo_entry->preferred_domains = bo->preferred_domains;
+		bo_entry->gem_handle = id;
+		bo_entry->alignment = bo->tbo.page_alignment;
+
+		if (bo->tbo.base.import_attach)
+			bo_entry->flags |= AMDGPU_GEM_LIST_HANDLES_FLAG_IS_IMPORT;
+
+		bo_index += 1;
+	}
+	spin_unlock(&filp->table_lock);
+
+	args->num_entries = bo_index;
+
+	if (!ret)
+		ret = copy_to_user(u64_to_user_ptr(args->entries), bo_entries, num_bos * sizeof(*bo_entries));
+
+	kvfree(bo_entries);
+
+	return ret;
+}
+
 static int amdgpu_gem_align_pitch(struct amdgpu_device *adev,
 				  int width,
 				  int cpp,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
index b3047d73fe07..b558336bc4c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h
@@ -67,6 +67,8 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 			  struct drm_file *filp);
 int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *filp);
+int amdgpu_gem_list_handles_ioctl(struct drm_device *dev, void *data,
+				  struct drm_file *filp);
 
 int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *filp);
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index bdedbaccf776..902e30263fcc 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -57,6 +57,7 @@ extern "C" {
 #define DRM_AMDGPU_USERQ		0x16
 #define DRM_AMDGPU_USERQ_SIGNAL		0x17
 #define DRM_AMDGPU_USERQ_WAIT		0x18
+#define DRM_AMDGPU_GEM_LIST_HANDLES	0x19
 
 #define DRM_IOCTL_AMDGPU_GEM_CREATE	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_GEM_CREATE, union drm_amdgpu_gem_create)
 #define DRM_IOCTL_AMDGPU_GEM_MMAP	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_GEM_MMAP, union drm_amdgpu_gem_mmap)
@@ -77,6 +78,7 @@ extern "C" {
 #define DRM_IOCTL_AMDGPU_USERQ		DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_USERQ, union drm_amdgpu_userq)
 #define DRM_IOCTL_AMDGPU_USERQ_SIGNAL	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_USERQ_SIGNAL, struct drm_amdgpu_userq_signal)
 #define DRM_IOCTL_AMDGPU_USERQ_WAIT	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_USERQ_WAIT, struct drm_amdgpu_userq_wait)
+#define DRM_IOCTL_AMDGPU_GEM_LIST_HANDLES DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDGPU_GEM_LIST_HANDLES, struct drm_amdgpu_gem_list_handles)
 
 /**
  * DOC: memory domains
@@ -811,6 +813,38 @@ struct drm_amdgpu_gem_op {
 	__u64	value;
 };
 
+#define AMDGPU_GEM_LIST_HANDLES_FLAG_IS_IMPORT	(1 << 0)
+
+struct drm_amdgpu_gem_list_handles {
+	/* User pointer to array of drm_amdgpu_gem_bo_info_entry */
+	__u64   entries;
+
+	/* Size of entries buffer / Number of handles in process (if larger than size of buffer, must retry) */
+	__u32   num_entries;
+
+	__u32 padding;
+};
+
+struct drm_amdgpu_gem_list_handles_entry {
+	/* gem handle of buffer object */
+	__u32 gem_handle;
+
+	/* Currently just one flag: IS_IMPORT */
+	__u32 flags;
+
+	/* Size of bo */
+	__u64 size;
+
+	/* Preferred domains for GEM_CREATE */
+	__u64 preferred_domains;
+
+	/* GEM_CREATE flags for re-creation of buffer */
+	__u64 alloc_flags;
+
+	/* physical start_addr alignment in bytes for some HW requirements */
+	__u64 alignment;
+};
+
 #define AMDGPU_VA_OP_MAP			1
 #define AMDGPU_VA_OP_UNMAP			2
 #define AMDGPU_VA_OP_CLEAR			3

From 4d82724f7f2b847eb0454b1aab5450545b39abd4 Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Mon, 16 Jun 2025 09:49:33 -0400
Subject: [PATCH 0848/1455] drm/amdgpu: Add mapping info option for GEM_OP
 ioctl
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add new GEM_OP_IOCTL option GET_MAPPING_INFO, which
returns a list of mappings associated with a given bo, along with
their positions and offsets.

Userspace for this and the previous change can be found at:
https://github.com/checkpoint-restore/criu/pull/2613

Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 90 +++++++++++++++++++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h  |  5 ++
 include/uapi/drm/amdgpu_drm.h           | 21 +++++-
 3 files changed, 101 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index bddb8d7475b2..e587e5ddeb00 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -955,17 +955,34 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 	struct drm_gem_object *gobj;
 	struct amdgpu_vm_bo_base *base;
 	struct amdgpu_bo *robj;
+	struct drm_exec exec;
+	struct amdgpu_fpriv *fpriv = filp->driver_priv;
 	int r;
 
+	if (args->padding)
+		return -EINVAL;
+
 	gobj = drm_gem_object_lookup(filp, args->handle);
 	if (!gobj)
 		return -ENOENT;
 
 	robj = gem_to_amdgpu_bo(gobj);
 
-	r = amdgpu_bo_reserve(robj, false);
-	if (unlikely(r))
-		goto out;
+	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT |
+			  DRM_EXEC_IGNORE_DUPLICATES, 0);
+	drm_exec_until_all_locked(&exec) {
+		r = drm_exec_lock_obj(&exec, gobj);
+		drm_exec_retry_on_contention(&exec);
+		if (r)
+			goto out_exec;
+
+		if (args->op == AMDGPU_GEM_OP_GET_MAPPING_INFO) {
+			r = amdgpu_vm_lock_pd(&fpriv->vm, &exec, 0);
+			drm_exec_retry_on_contention(&exec);
+			if (r)
+				goto out_exec;
+		}
+	}
 
 	switch (args->op) {
 	case AMDGPU_GEM_OP_GET_GEM_CREATE_INFO: {
@@ -976,7 +993,7 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 		info.alignment = robj->tbo.page_alignment << PAGE_SHIFT;
 		info.domains = robj->preferred_domains;
 		info.domain_flags = robj->flags;
-		amdgpu_bo_unreserve(robj);
+		drm_exec_fini(&exec);
 		if (copy_to_user(out, &info, sizeof(info)))
 			r = -EFAULT;
 		break;
@@ -985,20 +1002,17 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 		if (drm_gem_is_imported(&robj->tbo.base) &&
 		    args->value & AMDGPU_GEM_DOMAIN_VRAM) {
 			r = -EINVAL;
-			amdgpu_bo_unreserve(robj);
-			break;
+			goto out_exec;
 		}
 		if (amdgpu_ttm_tt_get_usermm(robj->tbo.ttm)) {
 			r = -EPERM;
-			amdgpu_bo_unreserve(robj);
-			break;
+			goto out_exec;
 		}
 		for (base = robj->vm_bo; base; base = base->next)
 			if (amdgpu_xgmi_same_hive(amdgpu_ttm_adev(robj->tbo.bdev),
 				amdgpu_ttm_adev(base->vm->root.bo->tbo.bdev))) {
 				r = -EINVAL;
-				amdgpu_bo_unreserve(robj);
-				goto out;
+				goto out_exec;
 			}
 
 
@@ -1011,15 +1025,63 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 
 		if (robj->flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID)
 			amdgpu_vm_bo_invalidate(robj, true);
-
-		amdgpu_bo_unreserve(robj);
+		drm_exec_fini(&exec);
 		break;
+	case AMDGPU_GEM_OP_GET_MAPPING_INFO: {
+		struct amdgpu_bo_va *bo_va = amdgpu_vm_bo_find(&fpriv->vm, robj);
+		struct drm_amdgpu_gem_vm_entry *vm_entries;
+		struct amdgpu_bo_va_mapping *mapping;
+		int num_mappings = 0;
+		/*
+		 * num_entries is set as an input to the size of the user-allocated array of
+		 * drm_amdgpu_gem_vm_entry stored at args->value.
+		 * num_entries is sent back as output as the number of mappings the bo has.
+		 * If that number is larger than the size of the array, the ioctl must
+		 * be retried.
+		 */
+		vm_entries = kvcalloc(args->num_entries, sizeof(*vm_entries), GFP_KERNEL);
+		if (!vm_entries)
+			return -ENOMEM;
+
+		amdgpu_vm_bo_va_for_each_valid_mapping(bo_va, mapping) {
+			if (num_mappings < args->num_entries) {
+				vm_entries[num_mappings].addr = mapping->start * AMDGPU_GPU_PAGE_SIZE;
+				vm_entries[num_mappings].size = (mapping->last - mapping->start + 1) * AMDGPU_GPU_PAGE_SIZE;
+				vm_entries[num_mappings].offset = mapping->offset;
+				vm_entries[num_mappings].flags = mapping->flags;
+			}
+			num_mappings += 1;
+		}
+
+		amdgpu_vm_bo_va_for_each_invalid_mapping(bo_va, mapping) {
+			if (num_mappings < args->num_entries) {
+				vm_entries[num_mappings].addr = mapping->start * AMDGPU_GPU_PAGE_SIZE;
+				vm_entries[num_mappings].size = (mapping->last - mapping->start + 1) * AMDGPU_GPU_PAGE_SIZE;
+				vm_entries[num_mappings].offset = mapping->offset;
+				vm_entries[num_mappings].flags = mapping->flags;
+			}
+			num_mappings += 1;
+		}
+
+		drm_exec_fini(&exec);
+
+		if (num_mappings > 0 && num_mappings <= args->num_entries)
+			r = copy_to_user(u64_to_user_ptr(args->value), vm_entries, num_mappings * sizeof(*vm_entries));
+
+		args->num_entries = num_mappings;
+
+		kvfree(vm_entries);
+		break;
+	}
 	default:
-		amdgpu_bo_unreserve(robj);
+		drm_exec_fini(&exec);
 		r = -EINVAL;
 	}
 
-out:
+	drm_gem_object_put(gobj);
+	return r;
+out_exec:
+	drm_exec_fini(&exec);
 	drm_gem_object_put(gobj);
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 3b4fa3246675..e045c1590d78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -670,4 +670,9 @@ void amdgpu_vm_tlb_fence_create(struct amdgpu_device *adev,
 void amdgpu_vm_print_task_info(struct amdgpu_device *adev,
 			       struct amdgpu_task_info *task_info);
 
+#define amdgpu_vm_bo_va_for_each_valid_mapping(bo_va, mapping) \
+		list_for_each_entry(mapping, &(bo_va)->valids, list)
+#define amdgpu_vm_bo_va_for_each_invalid_mapping(bo_va, mapping) \
+		list_for_each_entry(mapping, &(bo_va)->invalids, list)
+
 #endif
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 902e30263fcc..9cebd072a042 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -802,6 +802,21 @@ union drm_amdgpu_wait_fences {
 
 #define AMDGPU_GEM_OP_GET_GEM_CREATE_INFO	0
 #define AMDGPU_GEM_OP_SET_PLACEMENT		1
+#define AMDGPU_GEM_OP_GET_MAPPING_INFO		2
+
+struct drm_amdgpu_gem_vm_entry {
+	/* Start of mapping (in bytes) */
+	__u64 addr;
+
+	/* Size of mapping (in bytes) */
+	__u64 size;
+
+	/* Mapping offset */
+	__u64 offset;
+
+	/* flags needed to recreate mapping */
+	__u64 flags;
+};
 
 /* Sets or returns a value associated with a buffer. */
 struct drm_amdgpu_gem_op {
@@ -809,8 +824,12 @@ struct drm_amdgpu_gem_op {
 	__u32	handle;
 	/** AMDGPU_GEM_OP_* */
 	__u32	op;
-	/** Input or return value */
+	/** Input or return value. For MAPPING_INFO op: pointer to array of struct drm_amdgpu_gem_vm_entry */
 	__u64	value;
+	/** For MAPPING_INFO op: number of mappings (in/out) */
+	__u32	num_entries;
+
+	__u32	padding;
 };
 
 #define AMDGPU_GEM_LIST_HANDLES_FLAG_IS_IMPORT	(1 << 0)

From 85705b18ae7674347f8675f64b2b3115fb1d5629 Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Wed, 19 Feb 2025 10:01:32 -0500
Subject: [PATCH 0849/1455] drm/amdgpu: Allow kfd CRIU with no buffer objects

The kfd CRIU checkpoint ioctl would return an error if trying
to checkpoint a process with no kfd buffer objects.

This is a normal case and should not be an error.

Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: David Francis <David.Francis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 43115a374469..8535a52a62ca 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -2571,8 +2571,8 @@ static int criu_restore(struct file *filep,
 	pr_debug("CRIU restore (num_devices:%u num_bos:%u num_objects:%u priv_data_size:%llu)\n",
 		 args->num_devices, args->num_bos, args->num_objects, args->priv_data_size);
 
-	if (!args->bos || !args->devices || !args->priv_data || !args->priv_data_size ||
-	    !args->num_devices || !args->num_bos)
+	if ((args->num_bos > 0 && !args->bos) || !args->devices || !args->priv_data ||
+	    !args->priv_data_size || !args->num_devices)
 		return -EINVAL;
 
 	mutex_lock(&p->mutex);

From c661219cd7be75bb5599b525f16a455a058eb516 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:03 +0200
Subject: [PATCH 0850/1455] drm/amdgpu: Power up UVD 3 for FW validation (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unlike later versions, UVD 3 has firmware validation.
For this to work, the UVD should be powered up correctly.

When DPM is enabled and the display clock is off,
the SMU may choose a power state which doesn't power
the UVD, which can result in failure to initialize UVD.

v2:
Add code comments to explain about the UVD power state
and how UVD clock is turned on/off.

Fixes: b38f3e80ecec ("drm amdgpu: SI UVD v3_1 (v2)")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c | 29 +++++++++++++++++++++++----
 1 file changed, 25 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c
index 5dbaebb592b3..2e79a3afc774 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c
@@ -623,7 +623,22 @@ static void uvd_v3_1_enable_mgcg(struct amdgpu_device *adev,
  *
  * @ip_block: Pointer to the amdgpu_ip_block for this hw instance.
  *
- * Initialize the hardware, boot up the VCPU and do some testing
+ * Initialize the hardware, boot up the VCPU and do some testing.
+ *
+ * On SI, the UVD is meant to be used in a specific power state,
+ * or alternatively the driver can manually enable its clock.
+ * In amdgpu we use the dedicated UVD power state when DPM is enabled.
+ * Calling amdgpu_dpm_enable_uvd makes DPM select the UVD power state
+ * for the SMU and afterwards enables the UVD clock.
+ * This is automatically done by amdgpu_uvd_ring_begin_use when work
+ * is submitted to the UVD ring. Here, we have to call it manually
+ * in order to power up UVD before firmware validation.
+ *
+ * Note that we must not disable the UVD clock here, as that would
+ * cause the ring test to fail. However, UVD is powered off
+ * automatically after the ring test: amdgpu_uvd_ring_end_use calls
+ * the UVD idle work handler which will disable the UVD clock when
+ * all fences are signalled.
  */
 static int uvd_v3_1_hw_init(struct amdgpu_ip_block *ip_block)
 {
@@ -633,6 +648,15 @@ static int uvd_v3_1_hw_init(struct amdgpu_ip_block *ip_block)
 	int r;
 
 	uvd_v3_1_mc_resume(adev);
+	uvd_v3_1_enable_mgcg(adev, true);
+
+	/* Make sure UVD is powered during FW validation.
+	 * It's going to be automatically powered off after the ring test.
+	 */
+	if (adev->pm.dpm_enabled)
+		amdgpu_dpm_enable_uvd(adev, true);
+	else
+		amdgpu_asic_set_uvd_clocks(adev, 53300, 40000);
 
 	r = uvd_v3_1_fw_validate(adev);
 	if (r) {
@@ -640,9 +664,6 @@ static int uvd_v3_1_hw_init(struct amdgpu_ip_block *ip_block)
 		return r;
 	}
 
-	uvd_v3_1_enable_mgcg(adev, true);
-	amdgpu_asic_set_uvd_clocks(adev, 53300, 40000);
-
 	uvd_v3_1_start(adev);
 
 	r = amdgpu_ring_test_helper(ring);

From 3a0c3a4035f995e1f993dfaf4d63dc19e9b4bc1c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:04 +0200
Subject: [PATCH 0851/1455] drm/amd/pm: Disable ULV even if unsupported (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Always send PPSMC_MSG_DisableULV to the SMC, even if ULV mode
is unsupported, to make sure it is properly turned off.

v3:
Simplify si_disable_ulv further.
Always check the return value of amdgpu_si_send_msg_to_smc.

Fixes: 841686df9f7d ("drm/amdgpu: add SI DPM support (v4)")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index 52e732be59e3..e71070a23b91 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -5637,14 +5637,10 @@ static int si_populate_smc_t(struct amdgpu_device *adev,
 
 static int si_disable_ulv(struct amdgpu_device *adev)
 {
-	struct si_power_info *si_pi = si_get_pi(adev);
-	struct si_ulv_param *ulv = &si_pi->ulv;
+	PPSMC_Result r;
 
-	if (ulv->supported)
-		return (amdgpu_si_send_msg_to_smc(adev, PPSMC_MSG_DisableULV) == PPSMC_Result_OK) ?
-			0 : -EINVAL;
-
-	return 0;
+	r = amdgpu_si_send_msg_to_smc(adev, PPSMC_MSG_DisableULV);
+	return (r == PPSMC_Result_OK) ? 0 : -EINVAL;
 }
 
 static bool si_is_state_ulv_compatible(struct amdgpu_device *adev,

From 813d13524a3bdcc5f0253e06542440ca74c2653a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:05 +0200
Subject: [PATCH 0852/1455] drm/amd/pm: Increase SMC timeout on SI and warn
 (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The SMC can take an excessive amount of time to process some
messages under some conditions.

Background:
Sending a message to the SMC works by writing the message into
the mmSMC_MESSAGE_0 register and its optional parameter into
the mmSMC_SCRATCH0, and then polling mmSMC_RESP_0. Previously
the timeout was AMDGPU_MAX_USEC_TIMEOUT, ie. 100 ms.

Increase the timeout to 200 ms for all messages and to 1 sec for
a few messages which I've observed to be especially slow:
PPSMC_MSG_NoForcedLevel
PPSMC_MSG_SetEnabledLevels
PPSMC_MSG_SetForcedLevels
PPSMC_MSG_DisableULV
PPSMC_MSG_SwitchToSwState

This fixes the following problems on Tahiti when switching
from a lower clock power state to a higher clock state, such
as when DC turns on a display which was previously turned off.

* si_restrict_performance_levels_before_switch would fail
  (if the user previously forced high clocks using sysfs)
* si_set_sw_state would fail (always)

It turns out that both of those failures were SMC timeouts and
that the SMC actually didn't fail or hang, just needs more time
to process those.

Add a warning when there is an SMC timeout to make it easier to
identify this type of problem in the future.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c | 26 ++++++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c
index 4e65ab9e931c..281a5e377aee 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_smc.c
@@ -172,20 +172,42 @@ PPSMC_Result amdgpu_si_send_msg_to_smc(struct amdgpu_device *adev,
 {
 	u32 tmp;
 	int i;
+	int usec_timeout;
+
+	/* SMC seems to process some messages exceptionally slowly. */
+	switch (msg) {
+	case PPSMC_MSG_NoForcedLevel:
+	case PPSMC_MSG_SetEnabledLevels:
+	case PPSMC_MSG_SetForcedLevels:
+	case PPSMC_MSG_DisableULV:
+	case PPSMC_MSG_SwitchToSwState:
+		usec_timeout = 1000000; /* 1 sec */
+		break;
+	default:
+		usec_timeout = 200000; /* 200 ms */
+		break;
+	}
 
 	if (!amdgpu_si_is_smc_running(adev))
 		return PPSMC_Result_Failed;
 
 	WREG32(mmSMC_MESSAGE_0, msg);
 
-	for (i = 0; i < adev->usec_timeout; i++) {
+	for (i = 0; i < usec_timeout; i++) {
 		tmp = RREG32(mmSMC_RESP_0);
 		if (tmp != 0)
 			break;
 		udelay(1);
 	}
 
-	return (PPSMC_Result)RREG32(mmSMC_RESP_0);
+	tmp = RREG32(mmSMC_RESP_0);
+	if (tmp == 0) {
+		drm_warn(adev_to_drm(adev),
+			"%s timeout on message: %x (SMC_SCRATCH0: %x)\n",
+			__func__, msg, RREG32(mmSMC_SCRATCH0));
+	}
+
+	return (PPSMC_Result)tmp;
 }
 
 PPSMC_Result amdgpu_si_wait_for_smc_inactive(struct amdgpu_device *adev)

From a43b2cec04b02743338aa78f837ee0bdf066a6d5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:06 +0200
Subject: [PATCH 0853/1455] drm/amd/pm: Fix si_upload_smc_data (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The si_upload_smc_data function uses si_write_smc_soft_register
to set some register values in the SMC, and expects the result
to be PPSMC_Result_OK which is 1.

The PPSMC_Result_OK / PPSMC_Result_Failed values are used for
checking the result of a command sent to the SMC.
However, the si_write_smc_soft_register actually doesn't send
any commands to the SMC and returns zero on success,
so this check was incorrect.

Fix that by not checking the return value, just like other
calls to si_write_smc_soft_register.

v3:
Additionally, when no display is plugged in, there is no need
to restrict MCLK switching, so program the registers to zero.

Fixes: 841686df9f7d ("drm/amdgpu: add SI DPM support (v4)")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 43 ++++++++++++----------
 1 file changed, 24 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index e71070a23b91..6736c592dfdc 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -5813,9 +5813,9 @@ static int si_upload_smc_data(struct amdgpu_device *adev)
 {
 	struct amdgpu_crtc *amdgpu_crtc = NULL;
 	int i;
-
-	if (adev->pm.dpm.new_active_crtc_count == 0)
-		return 0;
+	u32 crtc_index = 0;
+	u32 mclk_change_block_cp_min = 0;
+	u32 mclk_change_block_cp_max = 0;
 
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
 		if (adev->pm.dpm.new_active_crtcs & (1 << i)) {
@@ -5824,26 +5824,31 @@ static int si_upload_smc_data(struct amdgpu_device *adev)
 		}
 	}
 
-	if (amdgpu_crtc == NULL)
-		return 0;
+	/* When a display is plugged in, program these so that the SMC
+	 * performs MCLK switching when it doesn't cause flickering.
+	 * When no display is plugged in, there is no need to restrict
+	 * MCLK switching, so program them to zero.
+	 */
+	if (adev->pm.dpm.new_active_crtc_count && amdgpu_crtc) {
+		crtc_index = amdgpu_crtc->crtc_id;
 
-	if (amdgpu_crtc->line_time <= 0)
-		return 0;
+		if (amdgpu_crtc->line_time) {
+			mclk_change_block_cp_min = amdgpu_crtc->wm_high / amdgpu_crtc->line_time;
+			mclk_change_block_cp_max = amdgpu_crtc->wm_low / amdgpu_crtc->line_time;
+		}
+	}
 
-	if (si_write_smc_soft_register(adev,
-				       SI_SMC_SOFT_REGISTER_crtc_index,
-				       amdgpu_crtc->crtc_id) != PPSMC_Result_OK)
-		return 0;
+	si_write_smc_soft_register(adev,
+		SI_SMC_SOFT_REGISTER_crtc_index,
+		crtc_index);
 
-	if (si_write_smc_soft_register(adev,
-				       SI_SMC_SOFT_REGISTER_mclk_change_block_cp_min,
-				       amdgpu_crtc->wm_high / amdgpu_crtc->line_time) != PPSMC_Result_OK)
-		return 0;
+	si_write_smc_soft_register(adev,
+		SI_SMC_SOFT_REGISTER_mclk_change_block_cp_min,
+		mclk_change_block_cp_min);
 
-	if (si_write_smc_soft_register(adev,
-				       SI_SMC_SOFT_REGISTER_mclk_change_block_cp_max,
-				       amdgpu_crtc->wm_low / amdgpu_crtc->line_time) != PPSMC_Result_OK)
-		return 0;
+	si_write_smc_soft_register(adev,
+		SI_SMC_SOFT_REGISTER_mclk_change_block_cp_max,
+		mclk_change_block_cp_max);
 
 	return 0;
 }

From ce025130127437dc884c84c254170e27b2ce9309 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:07 +0200
Subject: [PATCH 0854/1455] drm/amd/pm: Adjust si_upload_smc_data register
 programming (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Based on some comments in dm_pp_display_configuration
above the crtc_index and line_time fields, these values
are programmed to the SMC to work around an SMC hang
when it switches MCLK.

According to Alex, the Windows driver programs them to:
mclk_change_block_cp_min = 200 / line_time
mclk_change_block_cp_max = 100 / line_time
Let's use the same for the sake of consistency.

Previously we used the watermark values, but it seemed buggy
as the code was mixing up low/high and A/B watermarks, and
was not saving a low watermark value on DCE 6, so
mclk_change_block_cp_max would be always zero previously.

Split this change off from the previous si_upload_smc_data
to make it easier to bisect, in case it causes any issues.

Fixes: 841686df9f7d ("drm/amdgpu: add SI DPM support (v4)")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index 6736c592dfdc..fb008c5980d6 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -5833,8 +5833,8 @@ static int si_upload_smc_data(struct amdgpu_device *adev)
 		crtc_index = amdgpu_crtc->crtc_id;
 
 		if (amdgpu_crtc->line_time) {
-			mclk_change_block_cp_min = amdgpu_crtc->wm_high / amdgpu_crtc->line_time;
-			mclk_change_block_cp_max = amdgpu_crtc->wm_low / amdgpu_crtc->line_time;
+			mclk_change_block_cp_min = 200 / amdgpu_crtc->line_time;
+			mclk_change_block_cp_max = 100 / amdgpu_crtc->line_time;
 		}
 	}
 

From 9003a0746864f39a0ef72bd45f8e1ad85d930d67 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:08 +0200
Subject: [PATCH 0855/1455] drm/amd/pm: Treat zero vblank time as too short in
 si_dpm (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some parts of the code base expect that MCLK switching is turned
off when the vblank time is set to zero.

According to pp_pm_compute_clocks the non-DC code has issues
with MCLK switching with refresh rates over 120 Hz.

v3:
Add code comment to explain this better.
Add an if statement instead of changing the switch_limit.

Fixes: 841686df9f7d ("drm/amdgpu: add SI DPM support (v4)")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index fb008c5980d6..c11c4cc111df 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -3085,7 +3085,13 @@ static bool si_dpm_vblank_too_short(void *handle)
 	/* we never hit the non-gddr5 limit so disable it */
 	u32 switch_limit = adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5 ? 450 : 0;
 
-	if (vblank_time < switch_limit)
+	/* Consider zero vblank time too short and disable MCLK switching.
+	 * Note that the vblank time is set to maximum when no displays are attached,
+	 * so we'll still enable MCLK switching in that case.
+	 */
+	if (vblank_time == 0)
+		return true;
+	else if (vblank_time < switch_limit)
 		return true;
 	else
 		return false;

From ed3803533c7bf7df88bc3fc9f70bd317e1228ea8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:09 +0200
Subject: [PATCH 0856/1455] drm/amd/pm: Disable MCLK switching with non-DC at
 120 Hz+ (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

According to pp_pm_compute_clocks the non-DC display code
has "issues with mclk switching with refresh rates over 120 hz".
The workaround is to disable MCLK switching in this case.

Do the same for legacy DPM.

Fixes: 6ddbd37f1074 ("drm/amd/pm: optimize the amdgpu_pm_compute_clocks() implementations")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
index 42efe838fa85..2d2d2d5e6763 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
@@ -66,6 +66,13 @@ u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev)
 					(amdgpu_crtc->v_border * 2));
 
 				vblank_time_us = vblank_in_pixels * 1000 / amdgpu_crtc->hw_mode.clock;
+
+				/* we have issues with mclk switching with
+				 * refresh rates over 120 hz on the non-DC code.
+				 */
+				if (drm_mode_vrefresh(&amdgpu_crtc->hw_mode) > 120)
+					vblank_time_us = 0;
+
 				break;
 			}
 		}

From 7009e3af0474aca5f64262b3c72fb6e23b232f9b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:10 +0200
Subject: [PATCH 0857/1455] drm/amd/pm: Disable SCLK switching on Oland with
 high pixel clocks (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Port of commit 227545b9a08c ("drm/radeon/dpm: Disable sclk
switching on Oland when two 4K 60Hz monitors are connected")

This is an ad-hoc DPM fix, necessary because we don't have
proper bandwidth calculation for DCE 6.

We define "high pixelclock" for SI as higher than necessary
for 4K 30Hz. For example, 4K 60Hz and 1080p 144Hz fall into
this category.

When two high pixel clock displays are connected to Oland,
additionally disable shader clock switching, which results in
a higher voltage, thereby addressing some visible flickering.

v2:
Add more comments.
v3:
Split into two commits for easier review.

Fixes: 841686df9f7d ("drm/amdgpu: add SI DPM support (v4)")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 31 ++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index c11c4cc111df..4236700fc1ad 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -3449,12 +3449,14 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 {
 	struct  si_ps *ps = si_get_ps(rps);
 	struct amdgpu_clock_and_voltage_limits *max_limits;
+	struct amdgpu_connector *conn;
 	bool disable_mclk_switching = false;
 	bool disable_sclk_switching = false;
 	u32 mclk, sclk;
 	u16 vddc, vddci, min_vce_voltage = 0;
 	u32 max_sclk_vddc, max_mclk_vddci, max_mclk_vddc;
 	u32 max_sclk = 0, max_mclk = 0;
+	u32 high_pixelclock_count = 0;
 	int i;
 
 	if (adev->asic_type == CHIP_HAINAN) {
@@ -3482,6 +3484,35 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 		}
 	}
 
+	/* We define "high pixelclock" for SI as higher than necessary for 4K 30Hz.
+	 * For example, 4K 60Hz and 1080p 144Hz fall into this category.
+	 * Find number of such displays connected.
+	 */
+	for (i = 0; i < adev->mode_info.num_crtc; i++) {
+		if (!(adev->pm.dpm.new_active_crtcs & (1 << i)) ||
+			!adev->mode_info.crtcs[i]->enabled)
+			continue;
+
+		conn = to_amdgpu_connector(adev->mode_info.crtcs[i]->connector);
+
+		if (conn->pixelclock_for_modeset > 297000)
+			high_pixelclock_count++;
+	}
+
+	/* These are some ad-hoc fixes to some issues observed with SI GPUs.
+	 * They are necessary because we don't have something like dce_calcs
+	 * for these GPUs to calculate bandwidth requirements.
+	 */
+	if (high_pixelclock_count) {
+		/* On Oland, we observe some flickering when two 4K 60Hz
+		 * displays are connected, possibly because voltage is too low.
+		 * Raise the voltage by requiring a higher SCLK.
+		 * (Voltage cannot be adjusted independently without also SCLK.)
+		 */
+		if (high_pixelclock_count > 1 && adev->asic_type == CHIP_OLAND)
+			disable_sclk_switching = true;
+	}
+
 	if (rps->vce_active) {
 		rps->evclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].evclk;
 		rps->ecclk = adev->pm.dpm.vce_states[adev->pm.dpm.vce_level].ecclk;

From 6df0768c0ddbed8f4be900489a1502f0bd4bde87 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:11 +0200
Subject: [PATCH 0858/1455] drm/amd/pm: Remove wm_low and wm_high fields from
 amdgpu_crtc (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

These fields were only used by si_dpm and are not necessary
anymore. They also may have been incorrect because:
- wm_high was set to the LOW_WATERMARK field of watermark A.
- wm_low was not set on DCE 6 and was always zero.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h | 2 --
 drivers/gpu/drm/amd/amdgpu/dce_v10_0.c   | 3 +--
 drivers/gpu/drm/amd/amdgpu/dce_v11_0.c   | 3 +--
 drivers/gpu/drm/amd/amdgpu/dce_v6_0.c    | 1 -
 drivers/gpu/drm/amd/amdgpu/dce_v8_0.c    | 3 +--
 5 files changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index 6da4f946cac0..20460cfd09bc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -496,8 +496,6 @@ struct amdgpu_crtc {
 	struct drm_connector *connector;
 	/* for dpm */
 	u32 line_time;
-	u32 wm_low;
-	u32 wm_high;
 	u32 lb_vblank_lead_lines;
 	struct drm_display_mode hw_mode;
 	/* for virtual dce */
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index ba73518f5cdf..72ca6538b2e4 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -1141,8 +1141,7 @@ static void dce_v10_0_program_watermarks(struct amdgpu_device *adev,
 
 	/* save values for DPM */
 	amdgpu_crtc->line_time = line_time;
-	amdgpu_crtc->wm_high = latency_watermark_a;
-	amdgpu_crtc->wm_low = latency_watermark_b;
+
 	/* Save number of lines the linebuffer leads before the scanout */
 	amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index b01d88d078fa..e84608891300 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -1173,8 +1173,7 @@ static void dce_v11_0_program_watermarks(struct amdgpu_device *adev,
 
 	/* save values for DPM */
 	amdgpu_crtc->line_time = line_time;
-	amdgpu_crtc->wm_high = latency_watermark_a;
-	amdgpu_crtc->wm_low = latency_watermark_b;
+
 	/* Save number of lines the linebuffer leads before the scanout */
 	amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
index 81760a26f2ff..acc887a58518 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c
@@ -1034,7 +1034,6 @@ static void dce_v6_0_program_watermarks(struct amdgpu_device *adev,
 
 	/* save values for DPM */
 	amdgpu_crtc->line_time = line_time;
-	amdgpu_crtc->wm_high = latency_watermark_a;
 
 	/* Save number of lines the linebuffer leads before the scanout */
 	amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 19a265bd4d19..2ccd6aad8dd6 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -1096,8 +1096,7 @@ static void dce_v8_0_program_watermarks(struct amdgpu_device *adev,
 
 	/* save values for DPM */
 	amdgpu_crtc->line_time = line_time;
-	amdgpu_crtc->wm_high = latency_watermark_a;
-	amdgpu_crtc->wm_low = latency_watermark_b;
+
 	/* Save number of lines the linebuffer leads before the scanout */
 	amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines;
 }

From 98fb1d5596ff0499c784180b2e39a841049c4d90 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Thu, 28 Aug 2025 17:11:12 +0200
Subject: [PATCH 0859/1455] drm/amd/pm: Print VCE clocks too in si_dpm (v3)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

They are part of a power state too and should be printed
alongside the rest of the data from the power state.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index 4236700fc1ad..6595a611ce6e 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -7992,6 +7992,7 @@ static void si_dpm_print_power_state(void *handle,
 	amdgpu_dpm_dbg_print_class_info(adev, rps->class, rps->class2);
 	amdgpu_dpm_dbg_print_cap_info(adev, rps->caps);
 	drm_dbg(adev_to_drm(adev), "\tuvd    vclk: %d dclk: %d\n", rps->vclk, rps->dclk);
+	drm_dbg(adev_to_drm(adev), "\tvce    evclk: %d ecclk: %d\n", rps->evclk, rps->ecclk);
 	for (i = 0; i < ps->performance_level_count; i++) {
 		pl = &ps->performance_levels[i];
 		drm_dbg(adev_to_drm(adev), "\t\tpower level %d    sclk: %u mclk: %u vddc: %u vddci: %u pcie gen: %u\n",

From 1ee9d1a0962c13ba5ab7e47d33a80e3b8dc4b52e Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Tue, 2 Sep 2025 13:40:50 +0100
Subject: [PATCH 0860/1455] drm/amd/amdgpu: Fix missing error return on kzalloc
 failure

Currently the kzalloc failure check just sets reports the failure
and sets the variable ret to -ENOMEM, which is not checked later
for this specific error. Fix this by just returning -ENOMEM rather
than setting ret.

Fixes: 4fb930715468 ("drm/amd/amdgpu: remove redundant host to psp cmd buf allocations")
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 1d6e1d5de8fa..3696f48c233b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -448,7 +448,7 @@ static int psp_sw_init(struct amdgpu_ip_block *ip_block)
 	psp->cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
 	if (!psp->cmd) {
 		dev_err(adev->dev, "Failed to allocate memory to command buffer!\n");
-		ret = -ENOMEM;
+		return -ENOMEM;
 	}
 
 	adev->psp.xgmi_context.supports_extended_data =

From ad83b1da5b786ee2d245e41ce55cb1c71fed7c22 Mon Sep 17 00:00:00 2001
From: Zhanjun Dong <zhanjun.dong@intel.com>
Date: Fri, 29 Aug 2025 12:04:27 -0400
Subject: [PATCH 0861/1455] drm/xe/guc: Increase GuC crash dump buffer size

There are platforms already have a maximum dump size of 12KB, to avoid
data truncating, increase GuC crash dump buffer size to 16KB.

Signed-off-by: Zhanjun Dong <zhanjun.dong@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250829160427.1245732-1-zhanjun.dong@intel.com
---
 drivers/gpu/drm/xe/xe_guc_log.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_log.h b/drivers/gpu/drm/xe/xe_guc_log.h
index f1e2b0be90a9..98a47ac42b08 100644
--- a/drivers/gpu/drm/xe/xe_guc_log.h
+++ b/drivers/gpu/drm/xe/xe_guc_log.h
@@ -17,7 +17,7 @@ struct xe_device;
 #define DEBUG_BUFFER_SIZE       SZ_8M
 #define CAPTURE_BUFFER_SIZE     SZ_2M
 #else
-#define CRASH_BUFFER_SIZE	SZ_8K
+#define CRASH_BUFFER_SIZE	SZ_16K
 #define DEBUG_BUFFER_SIZE	SZ_64K
 #define CAPTURE_BUFFER_SIZE	SZ_1M
 #endif

From 1e6d36e15b8873fc3ade8b714d35d69efd5e1427 Mon Sep 17 00:00:00 2001
From: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Date: Fri, 29 Aug 2025 11:58:21 +0200
Subject: [PATCH 0862/1455] drm/amdgpu/amdkfd: Avoid a couple hundred
 -Wflex-array-member-not-at-end warnings

-Wflex-array-member-not-at-end was introduced in GCC-14, and we are
getting ready to enable it, globally.

Move the conflicting declarations to the end of the corresponding
structures. Notice that `struct dev_pagemap` is a flexible structure,
this is a structure that contains a flexible-array member.

struct dev_pagemap always has room for at least one range. amdgpu only
uses a single range. Therefore no change are needed to the allocation
of struct amdgpu_device.

Fix 283 of the following type of warnings:
    283 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h:111:28: warning: structure containing a flexible array member is not at the end of another structure [-Wflex-array-member-not-at-end]

Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h        | 8 +++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 8 +++++---
 2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 01f53700694b..17848ce65d1f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1149,9 +1149,6 @@ struct amdgpu_device {
 	/* for userq and VM fences */
 	struct amdgpu_seq64		seq64;
 
-	/* KFD */
-	struct amdgpu_kfd_dev		kfd;
-
 	/* UMC */
 	struct amdgpu_umc		umc;
 
@@ -1314,6 +1311,11 @@ struct amdgpu_device {
 	struct mutex                    userq_mutex;
 	bool                            userq_halt_for_enforce_isolation;
 	struct amdgpu_uid *uid_info;
+
+	/* KFD
+	 * Must be last --ends in a flexible-array member.
+	 */
+	struct amdgpu_kfd_dev		kfd;
 };
 
 static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 33eb4826b58b..127927b16ee2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -107,11 +107,13 @@ struct amdgpu_kfd_dev {
 	bool init_complete;
 	struct work_struct reset_work;
 
-	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping */
-	struct dev_pagemap pgmap;
-
 	/* Client for KFD BO GEM handle allocations */
 	struct drm_client_dev client;
+
+	/* HMM page migration MEMORY_DEVICE_PRIVATE mapping
+	 * Must be last --ends in a flexible-array member.
+	 */
+	struct dev_pagemap pgmap;
 };
 
 enum kgd_engine_type {

From 8d6f16f1f082881aa50ea7ae537b604dec647ed6 Mon Sep 17 00:00:00 2001
From: Tangudu Tilak Tirumalesh <tilak.tirumalesh.tangudu@intel.com>
Date: Wed, 27 Aug 2025 12:57:55 -0700
Subject: [PATCH 0863/1455] drm/xe: Extend Wa_22021007897 to Xe3 platforms

WA 22021007897 should also be applied to Graphics Versions 30.00, 30.01
and 30.03. To make it simple, simply use the range [3000, 3003] that
should be ok as there isn't a 3002 and if it's added, the WA list would
need to be revisited anyway.

Cc: Matt Atwood <matthew.s.atwood@intel.com>
Cc: Gustavo Sousa <gustavo.sousa@intel.com>
Signed-off-by: Tangudu Tilak Tirumalesh <tilak.tirumalesh.tangudu@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250827-wa-22021007897-v1-1-96922eb52af4@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_wa.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index 52c7df4c3afd..e53edf1e5100 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -879,6 +879,10 @@ static const struct xe_rtp_entry_sr lrc_was[] = {
 			     DIS_PARTIAL_AUTOSTRIP |
 			     DIS_AUTOSTRIP))
 	},
+	{ XE_RTP_NAME("22021007897"),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3003), ENGINE_CLASS(RENDER)),
+	  XE_RTP_ACTIONS(SET(COMMON_SLICE_CHICKEN4, SBE_PUSH_CONSTANT_BEHIND_FIX_ENABLE))
+	},
 };
 
 static __maybe_unused const struct xe_rtp_entry oob_was[] = {

From 56e6d568854519ed8e3f57444b8073a32d963fe6 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Fri, 29 Aug 2025 10:22:31 -0700
Subject: [PATCH 0864/1455] drm/xe: Add clearing stats to GT debugfs

It helpful to clear GT stats, run a test cases which is being profiled,
and look at the results of the stats from the individual test case. Make
stats entry writable and upon write clear the stats.

v5:
 - Drop clear_stats debugfs entry (Lucas)
v6:
 - Use xe_gt_stats_clear rather than helper (Michal)
 - Rework loop in xe_gt_stats_clear (Michal)

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250829172232.1308004-2-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_gt_debugfs.c | 20 +++++++++++++++++++-
 drivers/gpu/drm/xe/xe_gt_stats.c   | 14 ++++++++++++++
 drivers/gpu/drm/xe/xe_gt_stats.h   |  1 +
 3 files changed, 34 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index bf3a67b5951c..7411c5f5c739 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -327,7 +327,6 @@ static const struct drm_info_list vf_safe_debugfs_list[] = {
 	{"default_lrc_bcs", .show = xe_gt_debugfs_simple_show, .data = bcs_default_lrc},
 	{"default_lrc_vcs", .show = xe_gt_debugfs_simple_show, .data = vcs_default_lrc},
 	{"default_lrc_vecs", .show = xe_gt_debugfs_simple_show, .data = vecs_default_lrc},
-	{"stats", .show = xe_gt_debugfs_simple_show, .data = xe_gt_stats_print_info},
 	{"hwconfig", .show = xe_gt_debugfs_simple_show, .data = hwconfig},
 };
 
@@ -363,6 +362,24 @@ static ssize_t write_to_gt_call(const char __user *userbuf, size_t count, loff_t
 	return count;
 }
 
+static ssize_t stats_write(struct file *file, const char __user *userbuf,
+			   size_t count, loff_t *ppos)
+{
+	struct seq_file *s = file->private_data;
+	struct xe_gt *gt = s->private;
+
+	return write_to_gt_call(userbuf, count, ppos, xe_gt_stats_clear, gt);
+}
+
+static int stats_show(struct seq_file *s, void *unused)
+{
+	struct drm_printer p = drm_seq_file_printer(s);
+	struct xe_gt *gt = s->private;
+
+	return xe_gt_stats_print_info(gt, &p);
+}
+DEFINE_SHOW_STORE_ATTRIBUTE(stats);
+
 static void force_reset(struct xe_gt *gt)
 {
 	struct xe_device *xe = gt_to_xe(gt);
@@ -448,6 +465,7 @@ void xe_gt_debugfs_register(struct xe_gt *gt)
 	root->d_inode->i_private = gt;
 
 	/* VF safe */
+	debugfs_create_file("stats", 0600, root, gt, &stats_fops);
 	debugfs_create_file("force_reset", 0600, root, gt, &force_reset_fops);
 	debugfs_create_file("force_reset_sync", 0600, root, gt, &force_reset_sync_fops);
 
diff --git a/drivers/gpu/drm/xe/xe_gt_stats.c b/drivers/gpu/drm/xe/xe_gt_stats.c
index 30f942671c2b..17056d832138 100644
--- a/drivers/gpu/drm/xe/xe_gt_stats.c
+++ b/drivers/gpu/drm/xe/xe_gt_stats.c
@@ -50,3 +50,17 @@ int xe_gt_stats_print_info(struct xe_gt *gt, struct drm_printer *p)
 
 	return 0;
 }
+
+/**
+ * xe_gt_stats_clear - Clear the GT stats
+ * @gt: GT structure
+ *
+ * This clear (zeros) all the available GT stats.
+ */
+void xe_gt_stats_clear(struct xe_gt *gt)
+{
+	int id;
+
+	for (id = 0; id < ARRAY_SIZE(gt->stats.counters); ++id)
+		atomic64_set(&gt->stats.counters[id], 0);
+}
diff --git a/drivers/gpu/drm/xe/xe_gt_stats.h b/drivers/gpu/drm/xe/xe_gt_stats.h
index 38325ef53617..e8aea32bc971 100644
--- a/drivers/gpu/drm/xe/xe_gt_stats.h
+++ b/drivers/gpu/drm/xe/xe_gt_stats.h
@@ -13,6 +13,7 @@ struct drm_printer;
 
 #ifdef CONFIG_DEBUG_FS
 int xe_gt_stats_print_info(struct xe_gt *gt, struct drm_printer *p);
+void xe_gt_stats_clear(struct xe_gt *gt);
 void xe_gt_stats_incr(struct xe_gt *gt, const enum xe_gt_stats_id id, int incr);
 #else
 static inline void

From 4208fac3dce52d4328007028ec88dd5d0e78dcf2 Mon Sep 17 00:00:00 2001
From: Matthew Brost <matthew.brost@intel.com>
Date: Fri, 29 Aug 2025 10:22:32 -0700
Subject: [PATCH 0865/1455] drm/xe: Add more SVM GT stats

Add more SVM GT stats which give visibility to where time is spent in
the SVM page fault handler. Stats include number of faults at a given
size, total SVM page fault time, migration time in us, copy time in us,
copy kb, get pages time in us, and bind time in us. Will help in tuning
SVM for performance.

v2:
 - Include local changes
v3:
 - Add tlb invalidation + valid page fault + per size copy size stats
v4:
 - Ensure gt not NULL when incrementing SVM copy stats
 - Normalize stats names
 - Use magic macros to generate increment functions for ranges
v7:
 - Use DEF_STAT_STR (Michal)

Signed-off-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Francois Dugast <francois.dugast@intel.com>
Link: https://lore.kernel.org/r/20250829172232.1308004-3-matthew.brost@intel.com
---
 drivers/gpu/drm/xe/xe_gt_stats.c       |  43 +++++-
 drivers/gpu/drm/xe/xe_gt_stats_types.h |  33 +++++
 drivers/gpu/drm/xe/xe_svm.c            | 194 +++++++++++++++++++++++--
 3 files changed, 257 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_stats.c b/drivers/gpu/drm/xe/xe_gt_stats.c
index 17056d832138..5f74706bab81 100644
--- a/drivers/gpu/drm/xe/xe_gt_stats.c
+++ b/drivers/gpu/drm/xe/xe_gt_stats.c
@@ -26,11 +26,46 @@ void xe_gt_stats_incr(struct xe_gt *gt, const enum xe_gt_stats_id id, int incr)
 	atomic64_add(incr, &gt->stats.counters[id]);
 }
 
+#define DEF_STAT_STR(ID, name) [XE_GT_STATS_ID_##ID] = name
+
 static const char *const stat_description[__XE_GT_STATS_NUM_IDS] = {
-	"svm_pagefault_count",
-	"tlb_inval_count",
-	"vma_pagefault_count",
-	"vma_pagefault_kb",
+	DEF_STAT_STR(SVM_PAGEFAULT_COUNT, "svm_pagefault_count"),
+	DEF_STAT_STR(TLB_INVAL, "tlb_inval_count"),
+	DEF_STAT_STR(SVM_TLB_INVAL_COUNT, "svm_tlb_inval_count"),
+	DEF_STAT_STR(SVM_TLB_INVAL_US, "svm_tlb_inval_us"),
+	DEF_STAT_STR(VMA_PAGEFAULT_COUNT, "vma_pagefault_count"),
+	DEF_STAT_STR(VMA_PAGEFAULT_KB, "vma_pagefault_kb"),
+	DEF_STAT_STR(SVM_4K_PAGEFAULT_COUNT, "svm_4K_pagefault_count"),
+	DEF_STAT_STR(SVM_64K_PAGEFAULT_COUNT, "svm_64K_pagefault_count"),
+	DEF_STAT_STR(SVM_2M_PAGEFAULT_COUNT, "svm_2M_pagefault_count"),
+	DEF_STAT_STR(SVM_4K_VALID_PAGEFAULT_COUNT, "svm_4K_valid_pagefault_count"),
+	DEF_STAT_STR(SVM_64K_VALID_PAGEFAULT_COUNT, "svm_64K_valid_pagefault_count"),
+	DEF_STAT_STR(SVM_2M_VALID_PAGEFAULT_COUNT, "svm_2M_valid_pagefault_count"),
+	DEF_STAT_STR(SVM_4K_PAGEFAULT_US, "svm_4K_pagefault_us"),
+	DEF_STAT_STR(SVM_64K_PAGEFAULT_US, "svm_64K_pagefault_us"),
+	DEF_STAT_STR(SVM_2M_PAGEFAULT_US, "svm_2M_pagefault_us"),
+	DEF_STAT_STR(SVM_4K_MIGRATE_COUNT, "svm_4K_migrate_count"),
+	DEF_STAT_STR(SVM_64K_MIGRATE_COUNT, "svm_64K_migrate_count"),
+	DEF_STAT_STR(SVM_2M_MIGRATE_COUNT, "svm_2M_migrate_count"),
+	DEF_STAT_STR(SVM_4K_MIGRATE_US, "svm_4K_migrate_us"),
+	DEF_STAT_STR(SVM_64K_MIGRATE_US, "svm_64K_migrate_us"),
+	DEF_STAT_STR(SVM_2M_MIGRATE_US, "svm_2M_migrate_us"),
+	DEF_STAT_STR(SVM_DEVICE_COPY_US, "svm_device_copy_us"),
+	DEF_STAT_STR(SVM_4K_DEVICE_COPY_US, "svm_4K_device_copy_us"),
+	DEF_STAT_STR(SVM_64K_DEVICE_COPY_US, "svm_64K_device_copy_us"),
+	DEF_STAT_STR(SVM_2M_DEVICE_COPY_US, "svm_2M_device_copy_us"),
+	DEF_STAT_STR(SVM_CPU_COPY_US, "svm_cpu_copy_us"),
+	DEF_STAT_STR(SVM_4K_CPU_COPY_US, "svm_4K_cpu_copy_us"),
+	DEF_STAT_STR(SVM_64K_CPU_COPY_US, "svm_64K_cpu_copy_us"),
+	DEF_STAT_STR(SVM_2M_CPU_COPY_US, "svm_2M_cpu_copy_us"),
+	DEF_STAT_STR(SVM_DEVICE_COPY_KB, "svm_device_copy_kb"),
+	DEF_STAT_STR(SVM_CPU_COPY_KB, "svm_cpu_copy_kb"),
+	DEF_STAT_STR(SVM_4K_GET_PAGES_US, "svm_4K_get_pages_us"),
+	DEF_STAT_STR(SVM_64K_GET_PAGES_US, "svm_64K_get_pages_us"),
+	DEF_STAT_STR(SVM_2M_GET_PAGES_US, "svm_2M_get_pages_us"),
+	DEF_STAT_STR(SVM_4K_BIND_US, "svm_4K_bind_us"),
+	DEF_STAT_STR(SVM_64K_BIND_US, "svm_64K_bind_us"),
+	DEF_STAT_STR(SVM_2M_BIND_US, "svm_2M_bind_us"),
 };
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_gt_stats_types.h b/drivers/gpu/drm/xe/xe_gt_stats_types.h
index be3244d7133c..d8348a8de2e1 100644
--- a/drivers/gpu/drm/xe/xe_gt_stats_types.h
+++ b/drivers/gpu/drm/xe/xe_gt_stats_types.h
@@ -9,8 +9,41 @@
 enum xe_gt_stats_id {
 	XE_GT_STATS_ID_SVM_PAGEFAULT_COUNT,
 	XE_GT_STATS_ID_TLB_INVAL,
+	XE_GT_STATS_ID_SVM_TLB_INVAL_COUNT,
+	XE_GT_STATS_ID_SVM_TLB_INVAL_US,
 	XE_GT_STATS_ID_VMA_PAGEFAULT_COUNT,
 	XE_GT_STATS_ID_VMA_PAGEFAULT_KB,
+	XE_GT_STATS_ID_SVM_4K_PAGEFAULT_COUNT,
+	XE_GT_STATS_ID_SVM_64K_PAGEFAULT_COUNT,
+	XE_GT_STATS_ID_SVM_2M_PAGEFAULT_COUNT,
+	XE_GT_STATS_ID_SVM_4K_VALID_PAGEFAULT_COUNT,
+	XE_GT_STATS_ID_SVM_64K_VALID_PAGEFAULT_COUNT,
+	XE_GT_STATS_ID_SVM_2M_VALID_PAGEFAULT_COUNT,
+	XE_GT_STATS_ID_SVM_4K_PAGEFAULT_US,
+	XE_GT_STATS_ID_SVM_64K_PAGEFAULT_US,
+	XE_GT_STATS_ID_SVM_2M_PAGEFAULT_US,
+	XE_GT_STATS_ID_SVM_4K_MIGRATE_COUNT,
+	XE_GT_STATS_ID_SVM_64K_MIGRATE_COUNT,
+	XE_GT_STATS_ID_SVM_2M_MIGRATE_COUNT,
+	XE_GT_STATS_ID_SVM_4K_MIGRATE_US,
+	XE_GT_STATS_ID_SVM_64K_MIGRATE_US,
+	XE_GT_STATS_ID_SVM_2M_MIGRATE_US,
+	XE_GT_STATS_ID_SVM_DEVICE_COPY_US,
+	XE_GT_STATS_ID_SVM_4K_DEVICE_COPY_US,
+	XE_GT_STATS_ID_SVM_64K_DEVICE_COPY_US,
+	XE_GT_STATS_ID_SVM_2M_DEVICE_COPY_US,
+	XE_GT_STATS_ID_SVM_CPU_COPY_US,
+	XE_GT_STATS_ID_SVM_4K_CPU_COPY_US,
+	XE_GT_STATS_ID_SVM_64K_CPU_COPY_US,
+	XE_GT_STATS_ID_SVM_2M_CPU_COPY_US,
+	XE_GT_STATS_ID_SVM_DEVICE_COPY_KB,
+	XE_GT_STATS_ID_SVM_CPU_COPY_KB,
+	XE_GT_STATS_ID_SVM_4K_GET_PAGES_US,
+	XE_GT_STATS_ID_SVM_64K_GET_PAGES_US,
+	XE_GT_STATS_ID_SVM_2M_GET_PAGES_US,
+	XE_GT_STATS_ID_SVM_4K_BIND_US,
+	XE_GT_STATS_ID_SVM_64K_BIND_US,
+	XE_GT_STATS_ID_SVM_2M_BIND_US,
 	/* must be the last entry */
 	__XE_GT_STATS_NUM_IDS,
 };
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 76c6d74c1208..4834282e4b64 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -6,6 +6,7 @@
 #include <drm/drm_drv.h>
 
 #include "xe_bo.h"
+#include "xe_exec_queue_types.h"
 #include "xe_gt_stats.h"
 #include "xe_migrate.h"
 #include "xe_module.h"
@@ -112,6 +113,11 @@ xe_svm_garbage_collector_add_range(struct xe_vm *vm, struct xe_svm_range *range,
 		   &vm->svm.garbage_collector.work);
 }
 
+static void xe_svm_tlb_inval_count_stats_incr(struct xe_gt *gt)
+{
+	xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_TLB_INVAL_COUNT, 1);
+}
+
 static u8
 xe_svm_range_notifier_event_begin(struct xe_vm *vm, struct drm_gpusvm_range *r,
 				  const struct mmu_notifier_range *mmu_range,
@@ -144,13 +150,19 @@ xe_svm_range_notifier_event_begin(struct xe_vm *vm, struct drm_gpusvm_range *r,
 	 */
 	for_each_tile(tile, xe, id)
 		if (xe_pt_zap_ptes_range(tile, vm, range)) {
-			tile_mask |= BIT(id);
 			/*
 			 * WRITE_ONCE pairs with READ_ONCE in
 			 * xe_vm_has_valid_gpu_mapping()
 			 */
 			WRITE_ONCE(range->tile_invalidated,
 				   range->tile_invalidated | BIT(id));
+
+			if (!(tile_mask & BIT(id))) {
+				xe_svm_tlb_inval_count_stats_incr(tile->primary_gt);
+				if (tile->media_gt)
+					xe_svm_tlb_inval_count_stats_incr(tile->media_gt);
+				tile_mask |= BIT(id);
+			}
 		}
 
 	return tile_mask;
@@ -170,6 +182,24 @@ xe_svm_range_notifier_event_end(struct xe_vm *vm, struct drm_gpusvm_range *r,
 						   mmu_range);
 }
 
+static s64 xe_svm_stats_ktime_us_delta(ktime_t start)
+{
+	return IS_ENABLED(CONFIG_DEBUG_FS) ?
+		ktime_us_delta(ktime_get(), start) : 0;
+}
+
+static void xe_svm_tlb_inval_us_stats_incr(struct xe_gt *gt, ktime_t start)
+{
+	s64 us_delta = xe_svm_stats_ktime_us_delta(start);
+
+	xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_TLB_INVAL_US, us_delta);
+}
+
+static ktime_t xe_svm_stats_ktime_get(void)
+{
+	return IS_ENABLED(CONFIG_DEBUG_FS) ? ktime_get() : 0;
+}
+
 static void xe_svm_invalidate(struct drm_gpusvm *gpusvm,
 			      struct drm_gpusvm_notifier *notifier,
 			      const struct mmu_notifier_range *mmu_range)
@@ -177,8 +207,10 @@ static void xe_svm_invalidate(struct drm_gpusvm *gpusvm,
 	struct xe_vm *vm = gpusvm_to_vm(gpusvm);
 	struct xe_device *xe = vm->xe;
 	struct drm_gpusvm_range *r, *first;
+	struct xe_tile *tile;
+	ktime_t start = xe_svm_stats_ktime_get();
 	u64 adj_start = mmu_range->start, adj_end = mmu_range->end;
-	u8 tile_mask = 0;
+	u8 tile_mask = 0, id;
 	long err;
 
 	xe_svm_assert_in_notifier(vm);
@@ -231,6 +263,13 @@ range_notifier_event_end:
 	r = first;
 	drm_gpusvm_for_each_range(r, notifier, adj_start, adj_end)
 		xe_svm_range_notifier_event_end(vm, r, mmu_range);
+	for_each_tile(tile, xe, id) {
+		if (tile_mask & BIT(id)) {
+			xe_svm_tlb_inval_us_stats_incr(tile->primary_gt, start);
+			if (tile->media_gt)
+				xe_svm_tlb_inval_us_stats_incr(tile->media_gt, start);
+		}
+	}
 }
 
 static int __xe_svm_garbage_collector(struct xe_vm *vm,
@@ -384,11 +423,66 @@ enum xe_svm_copy_dir {
 	XE_SVM_COPY_TO_SRAM,
 };
 
+static void xe_svm_copy_kb_stats_incr(struct xe_gt *gt,
+				      const enum xe_svm_copy_dir dir,
+				      int kb)
+{
+	if (dir == XE_SVM_COPY_TO_VRAM)
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_DEVICE_COPY_KB, kb);
+	else
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_CPU_COPY_KB, kb);
+}
+
+static void xe_svm_copy_us_stats_incr(struct xe_gt *gt,
+				      const enum xe_svm_copy_dir dir,
+				      unsigned long npages,
+				      ktime_t start)
+{
+	s64 us_delta = xe_svm_stats_ktime_us_delta(start);
+
+	if (dir == XE_SVM_COPY_TO_VRAM) {
+		switch (npages) {
+		case 1:
+			xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_4K_DEVICE_COPY_US,
+					 us_delta);
+			break;
+		case 16:
+			xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_64K_DEVICE_COPY_US,
+					 us_delta);
+			break;
+		case 512:
+			xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_2M_DEVICE_COPY_US,
+					 us_delta);
+			break;
+		}
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_DEVICE_COPY_US,
+				 us_delta);
+	} else {
+		switch (npages) {
+		case 1:
+			xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_4K_CPU_COPY_US,
+					 us_delta);
+			break;
+		case 16:
+			xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_64K_CPU_COPY_US,
+					 us_delta);
+			break;
+		case 512:
+			xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_2M_CPU_COPY_US,
+					 us_delta);
+			break;
+		}
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_CPU_COPY_US,
+				 us_delta);
+	}
+}
+
 static int xe_svm_copy(struct page **pages,
 		       struct drm_pagemap_addr *pagemap_addr,
 		       unsigned long npages, const enum xe_svm_copy_dir dir)
 {
 	struct xe_vram_region *vr = NULL;
+	struct xe_gt *gt = NULL;
 	struct xe_device *xe;
 	struct dma_fence *fence = NULL;
 	unsigned long i;
@@ -396,6 +490,7 @@ static int xe_svm_copy(struct page **pages,
 	u64 vram_addr = XE_VRAM_ADDR_INVALID;
 	int err = 0, pos = 0;
 	bool sram = dir == XE_SVM_COPY_TO_SRAM;
+	ktime_t start = xe_svm_stats_ktime_get();
 
 	/*
 	 * This flow is complex: it locates physically contiguous device pages,
@@ -422,6 +517,7 @@ static int xe_svm_copy(struct page **pages,
 
 		if (!vr && spage) {
 			vr = page_to_vr(spage);
+			gt = xe_migrate_exec_queue(vr->migrate)->gt;
 			xe = vr->xe;
 		}
 		XE_WARN_ON(spage && page_to_vr(spage) != vr);
@@ -461,6 +557,9 @@ static int xe_svm_copy(struct page **pages,
 			int incr = (match && last) ? 1 : 0;
 
 			if (vram_addr != XE_VRAM_ADDR_INVALID) {
+				xe_svm_copy_kb_stats_incr(gt, dir,
+							  (i - pos + incr) *
+							  (PAGE_SIZE / SZ_1K));
 				if (sram) {
 					vm_dbg(&xe->drm,
 					       "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%ld",
@@ -499,6 +598,8 @@ static int xe_svm_copy(struct page **pages,
 
 			/* Extra mismatched device page, copy it */
 			if (!match && last && vram_addr != XE_VRAM_ADDR_INVALID) {
+				xe_svm_copy_kb_stats_incr(gt, dir,
+							  (PAGE_SIZE / SZ_1K));
 				if (sram) {
 					vm_dbg(&xe->drm,
 					       "COPY TO SRAM - 0x%016llx -> 0x%016llx, NPAGES=%d",
@@ -532,6 +633,14 @@ err_out:
 		dma_fence_put(fence);
 	}
 
+	/*
+	 * XXX: We can't derive the GT here (or anywhere in this functions, but
+	 * compute always uses the primary GT so accumlate stats on the likely
+	 * GT of the fault.
+	 */
+	if (gt)
+		xe_svm_copy_us_stats_incr(gt, dir, npages, start);
+
 	return err;
 #undef XE_MIGRATE_CHUNK_SIZE
 #undef XE_VRAM_ADDR_INVALID
@@ -845,6 +954,55 @@ bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range, struct xe_vm
 	return true;
 }
 
+#define DECL_SVM_RANGE_COUNT_STATS(elem, stat) \
+static void xe_svm_range_##elem##_count_stats_incr(struct xe_gt *gt, \
+						   struct xe_svm_range *range) \
+{ \
+	switch (xe_svm_range_size(range)) { \
+	case SZ_4K: \
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_4K_##stat##_COUNT, 1); \
+		break; \
+	case SZ_64K: \
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_64K_##stat##_COUNT, 1); \
+		break; \
+	case SZ_2M: \
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_2M_##stat##_COUNT, 1); \
+		break; \
+	} \
+} \
+
+DECL_SVM_RANGE_COUNT_STATS(fault, PAGEFAULT)
+DECL_SVM_RANGE_COUNT_STATS(valid_fault, VALID_PAGEFAULT)
+DECL_SVM_RANGE_COUNT_STATS(migrate, MIGRATE)
+
+#define DECL_SVM_RANGE_US_STATS(elem, stat) \
+static void xe_svm_range_##elem##_us_stats_incr(struct xe_gt *gt, \
+						struct xe_svm_range *range, \
+						ktime_t start) \
+{ \
+	s64 us_delta = xe_svm_stats_ktime_us_delta(start); \
+\
+	switch (xe_svm_range_size(range)) { \
+	case SZ_4K: \
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_4K_##stat##_US, \
+				 us_delta); \
+		break; \
+	case SZ_64K: \
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_64K_##stat##_US, \
+				 us_delta); \
+		break; \
+	case SZ_2M: \
+		xe_gt_stats_incr(gt, XE_GT_STATS_ID_SVM_2M_##stat##_US, \
+				 us_delta); \
+		break; \
+	} \
+} \
+
+DECL_SVM_RANGE_US_STATS(migrate, MIGRATE)
+DECL_SVM_RANGE_US_STATS(get_pages, GET_PAGES)
+DECL_SVM_RANGE_US_STATS(bind, BIND)
+DECL_SVM_RANGE_US_STATS(fault, PAGEFAULT)
+
 static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 				     struct xe_gt *gt, u64 fault_addr,
 				     bool need_vram)
@@ -866,6 +1024,7 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 	struct xe_tile *tile = gt_to_tile(gt);
 	int migrate_try_count = ctx.devmem_only ? 3 : 1;
 	ktime_t end = 0;
+	ktime_t start = xe_svm_stats_ktime_get(), bind_start, get_pages_start;
 	int err;
 
 	lockdep_assert_held_write(&vm->lock);
@@ -884,23 +1043,34 @@ retry:
 	if (IS_ERR(range))
 		return PTR_ERR(range);
 
-	if (ctx.devmem_only && !range->base.flags.migrate_devmem)
-		return -EACCES;
+	xe_svm_range_fault_count_stats_incr(gt, range);
 
-	if (xe_svm_range_is_valid(range, tile, ctx.devmem_only))
-		return 0;
+	if (ctx.devmem_only && !range->base.flags.migrate_devmem) {
+		err = -EACCES;
+		goto out;
+	}
+
+	if (xe_svm_range_is_valid(range, tile, ctx.devmem_only)) {
+		xe_svm_range_valid_fault_count_stats_incr(gt, range);
+		range_debug(range, "PAGE FAULT - VALID");
+		goto out;
+	}
 
 	range_debug(range, "PAGE FAULT");
 
 	dpagemap = xe_vma_resolve_pagemap(vma, tile);
 	if (--migrate_try_count >= 0 &&
 	    xe_svm_range_needs_migrate_to_vram(range, vma, !!dpagemap || ctx.devmem_only)) {
+		ktime_t migrate_start = xe_svm_stats_ktime_get();
+
 		/* TODO : For multi-device dpagemap will be used to find the
 		 * remote tile and remote device. Will need to modify
 		 * xe_svm_alloc_vram to use dpagemap for future multi-device
 		 * support.
 		 */
+		xe_svm_range_migrate_count_stats_incr(gt, range);
 		err = xe_svm_alloc_vram(tile, range, &ctx);
+		xe_svm_range_migrate_us_stats_incr(gt, range, migrate_start);
 		ctx.timeslice_ms <<= 1;	/* Double timeslice if we have to retry */
 		if (err) {
 			if (migrate_try_count || !ctx.devmem_only) {
@@ -917,6 +1087,8 @@ retry:
 		}
 	}
 
+	get_pages_start = xe_svm_stats_ktime_get();
+
 	range_debug(range, "GET PAGES");
 	err = xe_svm_range_get_pages(vm, range, &ctx);
 	/* Corner where CPU mappings have changed */
@@ -936,11 +1108,13 @@ retry:
 	}
 	if (err) {
 		range_debug(range, "PAGE FAULT - FAIL PAGE COLLECT");
-		goto err_out;
+		goto out;
 	}
 
+	xe_svm_range_get_pages_us_stats_incr(gt, range, get_pages_start);
 	range_debug(range, "PAGE FAULT - BIND");
 
+	bind_start = xe_svm_stats_ktime_get();
 retry_bind:
 	xe_vm_lock(vm, false);
 	fence = xe_vm_range_rebind(vm, vma, range, BIT(tile->id));
@@ -954,14 +1128,16 @@ retry_bind:
 		}
 		if (xe_vm_validate_should_retry(NULL, err, &end))
 			goto retry_bind;
-		goto err_out;
+		goto out;
 	}
 	xe_vm_unlock(vm);
 
 	dma_fence_wait(fence, false);
 	dma_fence_put(fence);
+	xe_svm_range_bind_us_stats_incr(gt, range, bind_start);
 
-err_out:
+out:
+	xe_svm_range_fault_us_stats_incr(gt, range, start);
 
 	return err;
 }

From 765531faa42ed76d06f959083dae0da3e8c1a529 Mon Sep 17 00:00:00 2001
From: Mika Kahola <mika.kahola@intel.com>
Date: Fri, 29 Aug 2025 13:12:26 +0300
Subject: [PATCH 0866/1455] drm/i915/display: Fix possible overflow on tc power
 domain selection

There is a possibility that intel_encoder_to_tc() function
returns negative i.e. TC_PORT_NONE (-1) value which may cause
tc_port_power_domain() function to overflow. To fix this,
let's add additional check that returns invalid power domain
i.e. POWER_DOMAIN_INVALID in case tc port equals TC_PORT_NONE.

Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250829101226.4085757-1-mika.kahola@intel.com
---
 drivers/gpu/drm/i915/display/intel_tc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 583fed5a386d..23745fc99d35 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -249,6 +249,9 @@ tc_port_power_domain(struct intel_tc_port *tc)
 {
 	enum tc_port tc_port = intel_encoder_to_tc(&tc->dig_port->base);
 
+	if (tc_port == TC_PORT_NONE)
+		return POWER_DOMAIN_INVALID;
+
 	return POWER_DOMAIN_PORT_DDI_LANES_TC1 + tc_port - TC_PORT_1;
 }
 

From 82b13dd5d93d9d170c096f042eb470add22c338c Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@linux.dev>
Date: Tue, 2 Sep 2025 10:10:42 +0200
Subject: [PATCH 0867/1455] drm/i915: Replace kmalloc() + copy_from_user() with
 memdup_user()

Replace kmalloc() followed by copy_from_user() with memdup_user() to
improve and simplify set_context_image(), and to silence the following
Coccinelle/coccicheck warning reported by memdup_user.cocci:

  WARNING opportunity for memdup_user

No functional changes intended.

Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Tvrtko Ursulin <tursulin@ursulin.net>
Link: https://lore.kernel.org/r/20250902081046.35463-2-thorsten.blum@linux.dev
---
 drivers/gpu/drm/i915/gem/i915_gem_context.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c
index 15835952352e..ed6599694835 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c
@@ -2158,18 +2158,12 @@ static int set_context_image(struct i915_gem_context *ctx,
 		goto out_ce;
 	}
 
-	state = kmalloc(ce->engine->context_size, GFP_KERNEL);
-	if (!state) {
-		ret = -ENOMEM;
+	state = memdup_user(u64_to_user_ptr(user.image), ce->engine->context_size);
+	if (IS_ERR(state)) {
+		ret = PTR_ERR(state);
 		goto out_ce;
 	}
 
-	if (copy_from_user(state, u64_to_user_ptr(user.image),
-			   ce->engine->context_size)) {
-		ret = -EFAULT;
-		goto out_state;
-	}
-
 	shmem_state = shmem_create_from_data(ce->engine->name,
 					     state, ce->engine->context_size);
 	if (IS_ERR(shmem_state)) {

From 38f9262b05654854954763e4fac13b0ba7fd6a33 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 29 Aug 2025 14:33:11 +0300
Subject: [PATCH 0868/1455] drm/i915/dp: convert open-coded timeout to
 poll_timeout_us()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use poll_timeout_us() instead of open-coding the timeout loop.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/20250829113311.1930618-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index bed7cba1ca68..958e48859beb 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -27,6 +27,7 @@
 
 #include <linux/export.h>
 #include <linux/i2c.h>
+#include <linux/iopoll.h>
 #include <linux/log2.h>
 #include <linux/math.h>
 #include <linux/notifier.h>
@@ -5662,14 +5663,9 @@ bool intel_digital_port_connected_locked(struct intel_encoder *encoder)
 	intel_wakeref_t wakeref;
 
 	with_intel_display_power(display, POWER_DOMAIN_DISPLAY_CORE, wakeref) {
-		unsigned long wait_expires = jiffies + msecs_to_jiffies_timeout(4);
-
-		do {
-			is_connected = dig_port->connected(encoder);
-			if (is_connected || is_glitch_free)
-				break;
-			usleep_range(10, 30);
-		} while (time_before(jiffies, wait_expires));
+		poll_timeout_us(is_connected = dig_port->connected(encoder),
+				is_connected || is_glitch_free,
+				30, 4000, false);
 	}
 
 	return is_connected;

From 33a7776f9bf748e4b6387d1d579ac0c5eab26785 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Tue, 2 Sep 2025 14:15:50 -0700
Subject: [PATCH 0869/1455] drm/tidss: dispc: Explicitly include bitfield.h

After a recent series to use FIELD_PREP and FIELD_MODIFY in
tidss_dispc.c, there are many errors when bitfield.h is not implicitly
included, such as when building allmodconfig for ARCH=hexagon:

  drivers/gpu/drm/tidss/tidss_dispc.c:1116:2: error: call to undeclared function 'FIELD_MODIFY'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
   1116 |         VP_REG_FLD_MOD(dispc, hw_videoport, DISPC_VP_CONTROL, v,
        |         ^
  drivers/gpu/drm/tidss/tidss_dispc.c:631:3: note: expanded from macro 'VP_REG_FLD_MOD'
    631 |                 FIELD_MODIFY((mask), &_reg, (val));                     \
        |                 ^
  drivers/gpu/drm/tidss/tidss_dispc.c:1140:2: error: call to undeclared function 'FIELD_MODIFY'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
   1140 |         FIELD_MODIFY(DISPC_VP_DSS_OLDI_CFG_MAP_MASK, &oldi_cfg,
        |         ^
  drivers/gpu/drm/tidss/tidss_dispc.c:1203:10: error: call to undeclared function 'FIELD_PREP'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
   1203 |                        FIELD_PREP(DISPC_VP_TIMING_H_SYNC_PULSE_MASK, hsw - 1) |
        |                        ^
  ...

Explicitly include bitfield.h to resolve the errors.

Fixes: 9accc8b10de8 ("drm/tidss: dispc: Get rid of FLD_VAL")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250902-drm-tidss-fix-missing-bitfield-h-v1-1-aaad4a285f98@kernel.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/tidss/tidss_dispc.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/tidss/tidss_dispc.c b/drivers/gpu/drm/tidss/tidss_dispc.c
index 97e2cd7b1e20..7c8c15a5c39b 100644
--- a/drivers/gpu/drm/tidss/tidss_dispc.c
+++ b/drivers/gpu/drm/tidss/tidss_dispc.c
@@ -4,6 +4,7 @@
  * Author: Jyri Sarha <jsarha@ti.com>
  */
 
+#include <linux/bitfield.h>
 #include <linux/clk.h>
 #include <linux/delay.h>
 #include <linux/dma-mapping.h>

From 47a5e9eb3ca0bb67c5d258229823b38cd113ecda Mon Sep 17 00:00:00 2001
From: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Date: Wed, 14 May 2025 16:52:29 -0700
Subject: [PATCH 0870/1455] drm/msm/dpu: Drop maxwidth from dpu_lm_sub_blks
 struct

Drop the unused maxwidth field from the dpu_lm_sub_blks struct

Signed-off-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/653759/
Link: https://lore.kernel.org/r/20250514-max-mixer-width-v1-1-c8ba0d9bb858@oss.qualcomm.com
[DB: also handled SM8750 block]
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c | 5 -----
 drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h | 1 -
 2 files changed, 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
index e824cd64fd3f..6641455c4ec6 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.c
@@ -338,7 +338,6 @@ static const struct dpu_sspp_sub_blks dpu_dma_sblk = _DMA_SBLK();
  *************************************************************/
 
 static const struct dpu_lm_sub_blks msm8998_lm_sblk = {
-	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 7, /* excluding base layer */
 	.blendstage_base = { /* offsets relative to mixer base */
 		0x20, 0x50, 0x80, 0xb0, 0x230,
@@ -347,7 +346,6 @@ static const struct dpu_lm_sub_blks msm8998_lm_sblk = {
 };
 
 static const struct dpu_lm_sub_blks sdm845_lm_sblk = {
-	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 11, /* excluding base layer */
 	.blendstage_base = { /* offsets relative to mixer base */
 		0x20, 0x38, 0x50, 0x68, 0x80, 0x98,
@@ -356,7 +354,6 @@ static const struct dpu_lm_sub_blks sdm845_lm_sblk = {
 };
 
 static const struct dpu_lm_sub_blks sc7180_lm_sblk = {
-	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 7, /* excluding base layer */
 	.blendstage_base = { /* offsets relative to mixer base */
 		0x20, 0x38, 0x50, 0x68, 0x80, 0x98, 0xb0
@@ -364,7 +361,6 @@ static const struct dpu_lm_sub_blks sc7180_lm_sblk = {
 };
 
 static const struct dpu_lm_sub_blks sm8750_lm_sblk = {
-	.maxwidth = DEFAULT_DPU_OUTPUT_LINE_WIDTH,
 	.maxblendstages = 11, /* excluding base layer */
 	.blendstage_base = { /* offsets relative to mixer base */
 		/* 0x40 + n*0x30 */
@@ -374,7 +370,6 @@ static const struct dpu_lm_sub_blks sm8750_lm_sblk = {
 };
 
 static const struct dpu_lm_sub_blks qcm2290_lm_sblk = {
-	.maxwidth = DEFAULT_DPU_LINE_WIDTH,
 	.maxblendstages = 4, /* excluding base layer */
 	.blendstage_base = { /* offsets relative to mixer base */
 		0x20, 0x38, 0x50, 0x68
diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
index a78bb2c334e3..f0768f54e9b3 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h
@@ -307,7 +307,6 @@ struct dpu_sspp_sub_blks {
  * @blendstage_base:        Blend-stage register base offset
  */
 struct dpu_lm_sub_blks {
-	u32 maxwidth;
 	u32 maxblendstages;
 	u32 blendstage_base[MAX_BLOCKS];
 };

From 6fc616723bb5fd4289d7422fa013da062b44ae55 Mon Sep 17 00:00:00 2001
From: Jun Nie <jun.nie@linaro.org>
Date: Tue, 19 Aug 2025 09:30:55 +0800
Subject: [PATCH 0871/1455] drm/msm: Do not validate SSPP when it is not ready

Current code will validate current plane and previous plane to
confirm they can share a SSPP with multi-rect mode. The SSPP
is already allocated for previous plane, while current plane
is not associated with any SSPP yet. Null pointer is referenced
when validating the SSPP of current plane. Skip SSPP validation
for current plane.

Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020
Mem abort info:
  ESR = 0x0000000096000004
  EC = 0x25: DABT (current EL), IL = 32 bits
  SET = 0, FnV = 0
  EA = 0, S1PTW = 0
  FSC = 0x04: level 0 translation fault
Data abort info:
  ISV = 0, ISS = 0x00000004, ISS2 = 0x00000000
  CM = 0, WnR = 0, TnD = 0, TagAccess = 0
  GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
user pgtable: 4k pages, 48-bit VAs, pgdp=0000000888ac3000
[0000000000000020] pgd=0000000000000000, p4d=0000000000000000
Internal error: Oops: 0000000096000004 [#1]  SMP
Modules linked in:
CPU: 4 UID: 0 PID: 1891 Comm: modetest Tainted: G S                  6.15.0-rc2-g3ee3f6e1202e #335 PREEMPT
Tainted: [S]=CPU_OUT_OF_SPEC
Hardware name: SM8650 EV1 rev1 4slam 2et (DT)
pstate: 63400009 (nZCv daif +PAN -UAO +TCO +DIT -SSBS BTYPE=--)
pc : dpu_plane_is_multirect_capable+0x68/0x90
lr : dpu_assign_plane_resources+0x288/0x410
sp : ffff800093dcb770
x29: ffff800093dcb770 x28: 0000000000002000 x27: ffff000817c6c000
x26: ffff000806b46368 x25: ffff0008013f6080 x24: ffff00080cbf4800
x23: ffff000810842680 x22: ffff0008013f1080 x21: ffff00080cc86080
x20: ffff000806b463b0 x19: ffff00080cbf5a00 x18: 00000000ffffffff
x17: 707a5f657a696c61 x16: 0000000000000003 x15: 0000000000002200
x14: 00000000ffffffff x13: 00aaaaaa00aaaaaa x12: 0000000000000000
x11: ffff000817c6e2b8 x10: 0000000000000000 x9 : ffff80008106a950
x8 : ffff00080cbf48f4 x7 : 0000000000000000 x6 : 0000000000000000
x5 : 0000000000000000 x4 : 0000000000000438 x3 : 0000000000000438
x2 : ffff800082e245e0 x1 : 0000000000000008 x0 : 0000000000000000
Call trace:
 dpu_plane_is_multirect_capable+0x68/0x90 (P)
 dpu_crtc_atomic_check+0x5bc/0x650
 drm_atomic_helper_check_planes+0x13c/0x220
 drm_atomic_helper_check+0x58/0xb8
 msm_atomic_check+0xd8/0xf0
 drm_atomic_check_only+0x4a8/0x968
 drm_atomic_commit+0x50/0xd8
 drm_atomic_helper_update_plane+0x140/0x188
 __setplane_atomic+0xfc/0x148
 drm_mode_setplane+0x164/0x378
 drm_ioctl_kernel+0xc0/0x140
 drm_ioctl+0x20c/0x500
 __arm64_sys_ioctl+0xbc/0xf8
 invoke_syscall+0x50/0x120
 el0_svc_common.constprop.0+0x48/0xf8
 do_el0_svc+0x28/0x40
 el0_svc+0x30/0xd0
 el0t_64_sync_handler+0x144/0x168
 el0t_64_sync+0x198/0x1a0
Code: b9402021 370fffc1 f9401441 3707ff81 (f94010a1)
---[ end trace 0000000000000000 ]---

Fixes: 3ed12a3664b36 ("drm/msm/dpu: allow sharing SSPP between planes")
Signed-off-by: Jun Nie <jun.nie@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/669224/
Link: https://lore.kernel.org/r/20250819-v6-16-rc2-quad-pipe-upstream-v15-1-2c7a85089db8@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
index 6859e8ef6b05..f54cf0faa1c7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c
@@ -922,6 +922,9 @@ static int dpu_plane_is_multirect_capable(struct dpu_hw_sspp *sspp,
 	if (MSM_FORMAT_IS_YUV(fmt))
 		return false;
 
+	if (!sspp)
+		return true;
+
 	if (!test_bit(DPU_SSPP_SMART_DMA_V1, &sspp->cap->features) &&
 	    !test_bit(DPU_SSPP_SMART_DMA_V2, &sspp->cap->features))
 		return false;
@@ -1028,6 +1031,7 @@ static int dpu_plane_try_multirect_shared(struct dpu_plane_state *pstate,
 	    prev_pipe->multirect_mode != DPU_SSPP_MULTIRECT_NONE)
 		return false;
 
+	/* Do not validate SSPP of current plane when it is not ready */
 	if (!dpu_plane_is_multirect_capable(pipe->sspp, pipe_cfg, fmt) ||
 	    !dpu_plane_is_multirect_capable(prev_pipe->sspp, prev_pipe_cfg, prev_fmt))
 		return false;

From 4b09db0c13eeb12102827d7d16128c1713319576 Mon Sep 17 00:00:00 2001
From: Jun Nie <jun.nie@linaro.org>
Date: Tue, 19 Aug 2025 09:30:56 +0800
Subject: [PATCH 0872/1455] drm/msm/dpu: polish log for resource allocation

It is more likely that resource allocation may fail in complex usage
case, such as quad-pipe case, than existing usage cases.
A resource type ID is printed on failure in the current implementation,
but the raw ID number is not explicit enough to help easily understand
which resource caused the failure, so add a table to match the type ID
to an human readable resource name and use it in the error print.

Signed-off-by: Jun Nie <jun.nie@linaro.org>
Reviewed-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/669225/
Link: https://lore.kernel.org/r/20250819-v6-16-rc2-quad-pipe-upstream-v15-2-2c7a85089db8@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c | 23 +++++++++++++++++++----
 1 file changed, 19 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
index 25382120cb1a..2c77c74fac0f 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c
@@ -865,6 +865,21 @@ void dpu_rm_release_all_sspp(struct dpu_global_state *global_state,
 		ARRAY_SIZE(global_state->sspp_to_crtc_id), crtc_id);
 }
 
+static char *dpu_hw_blk_type_name[] = {
+	[DPU_HW_BLK_TOP] = "TOP",
+	[DPU_HW_BLK_SSPP] = "SSPP",
+	[DPU_HW_BLK_LM] = "LM",
+	[DPU_HW_BLK_CTL] = "CTL",
+	[DPU_HW_BLK_PINGPONG] = "pingpong",
+	[DPU_HW_BLK_INTF] = "INTF",
+	[DPU_HW_BLK_WB] = "WB",
+	[DPU_HW_BLK_DSPP] = "DSPP",
+	[DPU_HW_BLK_MERGE_3D] = "merge_3d",
+	[DPU_HW_BLK_DSC] = "DSC",
+	[DPU_HW_BLK_CDM] = "CDM",
+	[DPU_HW_BLK_MAX] = "unknown",
+};
+
 /**
  * dpu_rm_get_assigned_resources - Get hw resources of the given type that are
  *     assigned to this encoder
@@ -946,13 +961,13 @@ int dpu_rm_get_assigned_resources(struct dpu_rm *rm,
 		}
 
 		if (num_blks == blks_size) {
-			DPU_ERROR("More than %d resources assigned to crtc %d\n",
-				  blks_size, crtc_id);
+			DPU_ERROR("More than %d %s assigned to crtc %d\n",
+				  blks_size, dpu_hw_blk_type_name[type], crtc_id);
 			break;
 		}
 		if (!hw_blks[i]) {
-			DPU_ERROR("Allocated resource %d unavailable to assign to crtc %d\n",
-				  type, crtc_id);
+			DPU_ERROR("%s unavailable to assign to crtc %d\n",
+				  dpu_hw_blk_type_name[type], crtc_id);
 			break;
 		}
 		blks[num_blks++] = hw_blks[i];

From 51e085ee4fdc61d88621e75a1c6f5c4fafb3fe35 Mon Sep 17 00:00:00 2001
From: Jun Nie <jun.nie@linaro.org>
Date: Tue, 19 Aug 2025 09:30:57 +0800
Subject: [PATCH 0873/1455] drm/msm/dpu: decide right side per last bit

Currently, only one pair of mixers is supported, so a non-zero counter
value is sufficient to identify the correct mixer within that pair.
However, future implementations may involve multiple mixer pairs. With
the current implementation, all mixers within the second pair would be
incorrectly selected as right mixer. To correctly select the mixer
within a pair, test the least significant bit of the counter. If the
least significant bit is not set, select the mixer as left one;
otherwise, select the mixer as right one for all pairs.

Signed-off-by: Jun Nie <jun.nie@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reviewed-by: Jessica Zhang <quic_jesszhan@quicinc.com>
Patchwork: https://patchwork.freedesktop.org/patch/669226/
Link: https://lore.kernel.org/r/20250819-v6-16-rc2-quad-pipe-upstream-v15-3-2c7a85089db8@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
index d59512e45af0..4b970a59deaf 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c
@@ -377,11 +377,10 @@ static void _dpu_crtc_setup_blend_cfg(struct dpu_crtc_mixer *mixer,
 static void _dpu_crtc_program_lm_output_roi(struct drm_crtc *crtc)
 {
 	struct dpu_crtc_state *crtc_state;
-	int lm_idx, lm_horiz_position;
+	int lm_idx;
 
 	crtc_state = to_dpu_crtc_state(crtc->state);
 
-	lm_horiz_position = 0;
 	for (lm_idx = 0; lm_idx < crtc_state->num_mixers; lm_idx++) {
 		const struct drm_rect *lm_roi = &crtc_state->lm_bounds[lm_idx];
 		struct dpu_hw_mixer *hw_lm = crtc_state->mixers[lm_idx].hw_lm;
@@ -392,7 +391,7 @@ static void _dpu_crtc_program_lm_output_roi(struct drm_crtc *crtc)
 
 		cfg.out_width = drm_rect_width(lm_roi);
 		cfg.out_height = drm_rect_height(lm_roi);
-		cfg.right_mixer = lm_horiz_position++;
+		cfg.right_mixer = lm_idx & 0x1;
 		cfg.flags = 0;
 		hw_lm->ops.setup_mixer_out(hw_lm, &cfg);
 	}

From 98463a8d9c463128d75086516f7f50c3f2a82afb Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 17:49:29 +0300
Subject: [PATCH 0874/1455] drm/i915/power: drop a couple of &i915->drm usages

Switch from &i915->drm to display->drm.

Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/20250902144929.3026700-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_power_well.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 31c2a07bb188..0c48d9cf3528 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -500,7 +500,6 @@ static void icl_tc_port_assert_ref_held(struct intel_display *display,
 
 static void icl_tc_cold_exit(struct intel_display *display)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	int ret, tries = 0;
 
 	while (1) {
@@ -515,7 +514,7 @@ static void icl_tc_cold_exit(struct intel_display *display)
 		msleep(1);
 
 	/* TODO: turn failure into a error as soon i915 CI updates ICL IFWI */
-	drm_dbg_kms(&i915->drm, "TC cold block %s\n", ret ? "failed" :
+	drm_dbg_kms(display->drm, "TC cold block %s\n", ret ? "failed" :
 		    "succeeded");
 }
 
@@ -1766,7 +1765,6 @@ static void chv_pipe_power_well_disable(struct intel_display *display,
 static void
 tgl_tc_cold_request(struct intel_display *display, bool block)
 {
-	struct drm_i915_private *i915 = to_i915(display->drm);
 	u8 tries = 0;
 	int ret;
 
@@ -1799,10 +1797,9 @@ tgl_tc_cold_request(struct intel_display *display, bool block)
 	}
 
 	if (ret)
-		drm_err(&i915->drm, "TC cold %sblock failed\n",
-			block ? "" : "un");
+		drm_err(display->drm, "TC cold %sblock failed\n", block ? "" : "un");
 	else
-		drm_dbg_kms(&i915->drm, "TC cold %sblock succeeded\n",
+		drm_dbg_kms(display->drm, "TC cold %sblock succeeded\n",
 			    block ? "" : "un");
 }
 

From e9a1d54ed9233b73737052c42115fe593f82ef84 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Mon, 1 Sep 2025 13:10:33 +0300
Subject: [PATCH 0875/1455] drm/i915/bios: Remove unnecessary checks of PSR
 idle frames in VBT binary
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

PSR idle frames in VBT binary is a 4 bits wide bitfield. Checking if it's
below 0 or over 15 doesn't make sense. Remove these checks.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Link: https://lore.kernel.org/r/20250901101033.4176277-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_bios.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_bios.c b/drivers/gpu/drm/i915/display/intel_bios.c
index 7d7193a05d88..3596dce84c28 100644
--- a/drivers/gpu/drm/i915/display/intel_bios.c
+++ b/drivers/gpu/drm/i915/display/intel_bios.c
@@ -1567,10 +1567,7 @@ parse_psr(struct intel_display *display,
 
 	panel->vbt.psr.full_link = psr_table->full_link;
 	panel->vbt.psr.require_aux_wakeup = psr_table->require_aux_to_wakeup;
-
-	/* Allowed VBT values goes from 0 to 15 */
-	panel->vbt.psr.idle_frames = psr_table->idle_frames < 0 ? 0 :
-		psr_table->idle_frames > 15 ? 15 : psr_table->idle_frames;
+	panel->vbt.psr.idle_frames = psr_table->idle_frames;
 
 	/*
 	 * New psr options 0=500us, 1=100us, 2=2500us, 3=0us

From 5afa9d2a9bb1410f816e0123846047288b16e4b9 Mon Sep 17 00:00:00 2001
From: Chia-I Wu <olvaffe@gmail.com>
Date: Thu, 28 Aug 2025 13:01:16 -0700
Subject: [PATCH 0876/1455] drm/panthor: check bo offset alignment in vm bind

Fail early from panthor_vm_bind_prepare_op_ctx instead of late from
ops->map_pages.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250828200116.3532255-1-olvaffe@gmail.com
---
 drivers/gpu/drm/panthor/panthor_mmu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/panthor/panthor_mmu.c b/drivers/gpu/drm/panthor/panthor_mmu.c
index bad720f23eb7..6dec4354e378 100644
--- a/drivers/gpu/drm/panthor/panthor_mmu.c
+++ b/drivers/gpu/drm/panthor/panthor_mmu.c
@@ -1222,7 +1222,7 @@ static int panthor_vm_prepare_map_op_ctx(struct panthor_vm_op_ctx *op_ctx,
 	    (flags & DRM_PANTHOR_VM_BIND_OP_TYPE_MASK) != DRM_PANTHOR_VM_BIND_OP_TYPE_MAP)
 		return -EINVAL;
 
-	/* Make sure the VA and size are aligned and in-bounds. */
+	/* Make sure the VA and size are in-bounds. */
 	if (size > bo->base.base.size || offset > bo->base.base.size - size)
 		return -EINVAL;
 
@@ -2447,7 +2447,7 @@ panthor_vm_bind_prepare_op_ctx(struct drm_file *file,
 	int ret;
 
 	/* Aligned on page size. */
-	if (!IS_ALIGNED(op->va | op->size, vm_pgsz))
+	if (!IS_ALIGNED(op->va | op->size | op->bo_offset, vm_pgsz))
 		return -EINVAL;
 
 	switch (op->flags & DRM_PANTHOR_VM_BIND_OP_TYPE_MASK) {

From 9931e4be11f2129a20ffd908bc364598a63016f8 Mon Sep 17 00:00:00 2001
From: Antheas Kapenekakis <lkml@antheas.dev>
Date: Fri, 29 Aug 2025 16:55:36 +0200
Subject: [PATCH 0877/1455] drm: panel-backlight-quirks: Make EDID match
 optional
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, having a valid panel_id match is required to use the quirk
system. For certain devices, we know that all SKUs need a certain quirk.
Therefore, allow not specifying ident by only checking for a match
if panel_id is non-zero.

Tested-by: Philip Müller <philm@manjaro.org>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Antheas Kapenekakis <lkml@antheas.dev>
Link: https://lore.kernel.org/r/20250829145541.512671-2-lkml@antheas.dev
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 drivers/gpu/drm/drm_panel_backlight_quirks.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_panel_backlight_quirks.c b/drivers/gpu/drm/drm_panel_backlight_quirks.c
index 598f812b7cb3..b38b33e26ea5 100644
--- a/drivers/gpu/drm/drm_panel_backlight_quirks.c
+++ b/drivers/gpu/drm/drm_panel_backlight_quirks.c
@@ -50,7 +50,7 @@ static bool drm_panel_min_backlight_quirk_matches(const struct drm_panel_min_bac
 	if (!dmi_match(quirk->dmi_match.field, quirk->dmi_match.value))
 		return false;
 
-	if (!drm_edid_match(edid, &quirk->ident))
+	if (quirk->ident.panel_id && !drm_edid_match(edid, &quirk->ident))
 		return false;
 
 	return true;

From 6eee1ef9e59853a49e926d116a004c53a9819dfd Mon Sep 17 00:00:00 2001
From: Antheas Kapenekakis <lkml@antheas.dev>
Date: Fri, 29 Aug 2025 16:55:37 +0200
Subject: [PATCH 0878/1455] drm: panel-backlight-quirks: Convert brightness
 quirk to generic structure
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently, the brightness quirk is limited to minimum brightness only.
Refactor it to a structure, so that more quirks can be added in the
future. Reserve 0 value for "no quirk", and use u16 to allow minimum
brightness up to 255.

Tested-by: Philip Müller <philm@manjaro.org>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Antheas Kapenekakis <lkml@antheas.dev>
Link: https://lore.kernel.org/r/20250829145541.512671-3-lkml@antheas.dev
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 12 ++++--
 drivers/gpu/drm/drm_panel_backlight_quirks.c  | 41 ++++++++++---------
 include/drm/drm_utils.h                       |  7 +++-
 3 files changed, 35 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index cd0e2976e268..4ad80ae615a2 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3612,11 +3612,11 @@ static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = {
 
 static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 {
+	const struct drm_panel_backlight_quirk *panel_backlight_quirk;
 	struct amdgpu_dm_backlight_caps *caps;
 	struct drm_connector *conn_base;
 	struct amdgpu_device *adev;
 	struct drm_luminance_range_info *luminance_range;
-	int min_input_signal_override;
 
 	if (aconnector->bl_idx == -1 ||
 	    aconnector->dc_link->connector_signal != SIGNAL_TYPE_EDP)
@@ -3656,9 +3656,13 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 	else
 		caps->aux_min_input_signal = 1;
 
-	min_input_signal_override = drm_get_panel_min_brightness_quirk(aconnector->drm_edid);
-	if (min_input_signal_override >= 0)
-		caps->min_input_signal = min_input_signal_override;
+	panel_backlight_quirk =
+		drm_get_panel_backlight_quirk(aconnector->drm_edid);
+	if (!IS_ERR_OR_NULL(panel_backlight_quirk)) {
+		if (panel_backlight_quirk->min_brightness)
+			caps->min_input_signal =
+				panel_backlight_quirk->min_brightness - 1;
+	}
 }
 
 DEFINE_FREE(sink_release, struct dc_sink *, if (_T) dc_sink_release(_T))
diff --git a/drivers/gpu/drm/drm_panel_backlight_quirks.c b/drivers/gpu/drm/drm_panel_backlight_quirks.c
index b38b33e26ea5..702726c20ccc 100644
--- a/drivers/gpu/drm/drm_panel_backlight_quirks.c
+++ b/drivers/gpu/drm/drm_panel_backlight_quirks.c
@@ -8,23 +8,23 @@
 #include <drm/drm_edid.h>
 #include <drm/drm_utils.h>
 
-struct drm_panel_min_backlight_quirk {
+struct drm_get_panel_backlight_quirk {
 	struct {
 		enum dmi_field field;
 		const char * const value;
 	} dmi_match;
 	struct drm_edid_ident ident;
-	u8 min_brightness;
+	struct drm_panel_backlight_quirk quirk;
 };
 
-static const struct drm_panel_min_backlight_quirk drm_panel_min_backlight_quirks[] = {
+static const struct drm_get_panel_backlight_quirk drm_panel_min_backlight_quirks[] = {
 	/* 13 inch matte panel */
 	{
 		.dmi_match.field = DMI_BOARD_VENDOR,
 		.dmi_match.value = "Framework",
 		.ident.panel_id = drm_edid_encode_panel_id('B', 'O', 'E', 0x0bca),
 		.ident.name = "NE135FBM-N41",
-		.min_brightness = 0,
+		.quirk = { .min_brightness = 1, },
 	},
 	/* 13 inch glossy panel */
 	{
@@ -32,7 +32,7 @@ static const struct drm_panel_min_backlight_quirk drm_panel_min_backlight_quirks
 		.dmi_match.value = "Framework",
 		.ident.panel_id = drm_edid_encode_panel_id('B', 'O', 'E', 0x095f),
 		.ident.name = "NE135FBM-N41",
-		.min_brightness = 0,
+		.quirk = { .min_brightness = 1, },
 	},
 	/* 13 inch 2.8k panel */
 	{
@@ -40,12 +40,13 @@ static const struct drm_panel_min_backlight_quirk drm_panel_min_backlight_quirks
 		.dmi_match.value = "Framework",
 		.ident.panel_id = drm_edid_encode_panel_id('B', 'O', 'E', 0x0cb4),
 		.ident.name = "NE135A1M-NY1",
-		.min_brightness = 0,
+		.quirk = { .min_brightness = 1, },
 	},
 };
 
-static bool drm_panel_min_backlight_quirk_matches(const struct drm_panel_min_backlight_quirk *quirk,
-						  const struct drm_edid *edid)
+static bool drm_panel_min_backlight_quirk_matches(
+	const struct drm_get_panel_backlight_quirk *quirk,
+	const struct drm_edid *edid)
 {
 	if (!dmi_match(quirk->dmi_match.field, quirk->dmi_match.value))
 		return false;
@@ -57,39 +58,39 @@ static bool drm_panel_min_backlight_quirk_matches(const struct drm_panel_min_bac
 }
 
 /**
- * drm_get_panel_min_brightness_quirk - Get minimum supported brightness level for a panel.
+ * drm_get_panel_backlight_quirk - Get backlight quirks for a panel
  * @edid: EDID of the panel to check
  *
  * This function checks for platform specific (e.g. DMI based) quirks
  * providing info on the minimum backlight brightness for systems where this
- * cannot be probed correctly from the hard-/firm-ware.
+ * cannot be probed correctly from the hard-/firm-ware and other sources.
  *
  * Returns:
- * A negative error value or
- * an override value in the range [0, 255] representing 0-100% to be scaled to
- * the drivers target range.
+ * a drm_panel_backlight_quirk struct if a quirk was found, otherwise an
+ * error pointer.
  */
-int drm_get_panel_min_brightness_quirk(const struct drm_edid *edid)
+const struct drm_panel_backlight_quirk *
+drm_get_panel_backlight_quirk(const struct drm_edid *edid)
 {
-	const struct drm_panel_min_backlight_quirk *quirk;
+	const struct drm_get_panel_backlight_quirk *quirk;
 	size_t i;
 
 	if (!IS_ENABLED(CONFIG_DMI))
-		return -ENODATA;
+		return ERR_PTR(-ENODATA);
 
 	if (!edid)
-		return -EINVAL;
+		return ERR_PTR(-EINVAL);
 
 	for (i = 0; i < ARRAY_SIZE(drm_panel_min_backlight_quirks); i++) {
 		quirk = &drm_panel_min_backlight_quirks[i];
 
 		if (drm_panel_min_backlight_quirk_matches(quirk, edid))
-			return quirk->min_brightness;
+			return &quirk->quirk;
 	}
 
-	return -ENODATA;
+	return ERR_PTR(-ENODATA);
 }
-EXPORT_SYMBOL(drm_get_panel_min_brightness_quirk);
+EXPORT_SYMBOL(drm_get_panel_backlight_quirk);
 
 MODULE_DESCRIPTION("Quirks for panel backlight overrides");
 MODULE_LICENSE("GPL");
diff --git a/include/drm/drm_utils.h b/include/drm/drm_utils.h
index 15fa9b6865f4..82eeee4a58ab 100644
--- a/include/drm/drm_utils.h
+++ b/include/drm/drm_utils.h
@@ -16,7 +16,12 @@ struct drm_edid;
 
 int drm_get_panel_orientation_quirk(int width, int height);
 
-int drm_get_panel_min_brightness_quirk(const struct drm_edid *edid);
+struct drm_panel_backlight_quirk {
+	u16 min_brightness;
+};
+
+const struct drm_panel_backlight_quirk *
+drm_get_panel_backlight_quirk(const struct drm_edid *edid);
 
 signed long drm_timeout_abs_to_jiffies(int64_t timeout_nsec);
 

From f7033fab81d82e9a56092bb412886df1d2591ae6 Mon Sep 17 00:00:00 2001
From: Antheas Kapenekakis <lkml@antheas.dev>
Date: Fri, 29 Aug 2025 16:55:38 +0200
Subject: [PATCH 0879/1455] drm: panel-backlight-quirks: Add secondary DMI
 match
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Using a single DMI match only allows matching per manufacturer.
Introduce a second optional match to allow matching make/model.
In addition, make DMI optional to allow matching only by EDID.

Tested-by: Philip Müller <philm@manjaro.org>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Antheas Kapenekakis <lkml@antheas.dev>
Link: https://lore.kernel.org/r/20250829145541.512671-4-lkml@antheas.dev
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 drivers/gpu/drm/drm_panel_backlight_quirks.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/drm_panel_backlight_quirks.c b/drivers/gpu/drm/drm_panel_backlight_quirks.c
index 702726c20ccc..3d386a96e50e 100644
--- a/drivers/gpu/drm/drm_panel_backlight_quirks.c
+++ b/drivers/gpu/drm/drm_panel_backlight_quirks.c
@@ -8,11 +8,14 @@
 #include <drm/drm_edid.h>
 #include <drm/drm_utils.h>
 
+struct drm_panel_match {
+	enum dmi_field field;
+	const char * const value;
+};
+
 struct drm_get_panel_backlight_quirk {
-	struct {
-		enum dmi_field field;
-		const char * const value;
-	} dmi_match;
+	struct drm_panel_match dmi_match;
+	struct drm_panel_match dmi_match_other;
 	struct drm_edid_ident ident;
 	struct drm_panel_backlight_quirk quirk;
 };
@@ -48,7 +51,13 @@ static bool drm_panel_min_backlight_quirk_matches(
 	const struct drm_get_panel_backlight_quirk *quirk,
 	const struct drm_edid *edid)
 {
-	if (!dmi_match(quirk->dmi_match.field, quirk->dmi_match.value))
+	if (quirk->dmi_match.field &&
+	    !dmi_match(quirk->dmi_match.field, quirk->dmi_match.value))
+		return false;
+
+	if (quirk->dmi_match_other.field &&
+	    !dmi_match(quirk->dmi_match_other.field,
+		       quirk->dmi_match_other.value))
 		return false;
 
 	if (quirk->ident.panel_id && !drm_edid_match(edid, &quirk->ident))

From aef10b1138e995ba9aa4357ed78cd05686cabbe1 Mon Sep 17 00:00:00 2001
From: Antheas Kapenekakis <lkml@antheas.dev>
Date: Fri, 29 Aug 2025 16:55:39 +0200
Subject: [PATCH 0880/1455] drm: panel-backlight-quirks: Add brightness mask
 quirk
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Certain OLED devices malfunction on specific brightness levels.
Specifically, when DP_SOURCE_BACKLIGHT_LEVEL is written to with
the first byte being 0x00 and sometimes 0x01, the panel forcibly
turns off until the device sleeps again.

Below are some examples. This was found by iterating over brighness
ranges while printing DP_SOURCE_BACKLIGHT_LEVEL. It was found that
the screen would malfunction on specific values, and some of them
were collected.

Therefore, introduce a quirk where the minor byte of brightness is
OR'd with 0x03 to avoid the range of invalid values.

This quirk was tested by removing the workarounds and iterating
from 0 to 50_000 value ranges with a cadence of 0.2s/it. The
range of the panel is 1000...400_000, so the values were slightly
interpolated during testing. The custom brightness curve added on
6.15 was disabled.

 86016:  10101000000000000
 86272:  10101000100000000
 87808:  10101011100000000
251648: 111101011100000000
251649: 111101011100000001

 86144:  10101000010000000
 87809:  10101011100000001
251650: 111101011100000010

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3803
Tested-by: Philip Müller <philm@manjaro.org>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Antheas Kapenekakis <lkml@antheas.dev>
Link: https://lore.kernel.org/r/20250829145541.512671-5-lkml@antheas.dev
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |  7 ++++
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  5 +++
 drivers/gpu/drm/drm_panel_backlight_quirks.c  | 36 +++++++++++++++++++
 include/drm/drm_utils.h                       |  1 +
 4 files changed, 49 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 4ad80ae615a2..156f2aae6828 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3662,6 +3662,9 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 		if (panel_backlight_quirk->min_brightness)
 			caps->min_input_signal =
 				panel_backlight_quirk->min_brightness - 1;
+		if (panel_backlight_quirk->brightness_mask)
+			caps->brightness_mask =
+				panel_backlight_quirk->brightness_mask;
 	}
 }
 
@@ -4862,6 +4865,10 @@ static void amdgpu_dm_backlight_set_level(struct amdgpu_display_manager *dm,
 	brightness = convert_brightness_from_user(caps, dm->brightness[bl_idx]);
 	link = (struct dc_link *)dm->backlight_link[bl_idx];
 
+	/* Apply brightness quirk */
+	if (caps->brightness_mask)
+		brightness |= caps->brightness_mask;
+
 	/* Change brightness based on AUX property */
 	mutex_lock(&dm->dc_lock);
 	if (dm->dc->caps.ips_support && dm->dc->ctx->dmub_srv->idle_allowed) {
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index b937da0a4e4a..60ce2ceb653a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -200,6 +200,11 @@ struct amdgpu_dm_backlight_caps {
 	 * @aux_support: Describes if the display supports AUX backlight.
 	 */
 	bool aux_support;
+	/**
+	 * @brightness_mask: After deriving brightness, OR it with this mask.
+	 * Workaround for panels with issues with certain brightness values.
+	 */
+	u32 brightness_mask;
 	/**
 	 * @ac_level: the default brightness if booted on AC
 	 */
diff --git a/drivers/gpu/drm/drm_panel_backlight_quirks.c b/drivers/gpu/drm/drm_panel_backlight_quirks.c
index 3d386a96e50e..2bdbd5583d32 100644
--- a/drivers/gpu/drm/drm_panel_backlight_quirks.c
+++ b/drivers/gpu/drm/drm_panel_backlight_quirks.c
@@ -45,6 +45,42 @@ static const struct drm_get_panel_backlight_quirk drm_panel_min_backlight_quirks
 		.ident.name = "NE135A1M-NY1",
 		.quirk = { .min_brightness = 1, },
 	},
+	/* Have OLED Panels with brightness issue when last byte is 0/1 */
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "AYANEO",
+		.dmi_match_other.field = DMI_PRODUCT_NAME,
+		.dmi_match_other.value = "AYANEO 3",
+		.quirk = { .brightness_mask = 3, },
+	},
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "ZOTAC",
+		.dmi_match_other.field = DMI_BOARD_NAME,
+		.dmi_match_other.value = "G0A1W",
+		.quirk = { .brightness_mask = 3, },
+	},
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "ZOTAC",
+		.dmi_match_other.field = DMI_BOARD_NAME,
+		.dmi_match_other.value = "G1A1W",
+		.quirk = { .brightness_mask = 3, },
+	},
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "ONE-NETBOOK",
+		.dmi_match_other.field = DMI_PRODUCT_NAME,
+		.dmi_match_other.value = "ONEXPLAYER F1Pro",
+		.quirk = { .brightness_mask = 3, },
+	},
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "ONE-NETBOOK",
+		.dmi_match_other.field = DMI_PRODUCT_NAME,
+		.dmi_match_other.value = "ONEXPLAYER F1 EVA-02",
+		.quirk = { .brightness_mask = 3, },
+	},
 };
 
 static bool drm_panel_min_backlight_quirk_matches(
diff --git a/include/drm/drm_utils.h b/include/drm/drm_utils.h
index 82eeee4a58ab..6a46f755daba 100644
--- a/include/drm/drm_utils.h
+++ b/include/drm/drm_utils.h
@@ -18,6 +18,7 @@ int drm_get_panel_orientation_quirk(int width, int height);
 
 struct drm_panel_backlight_quirk {
 	u16 min_brightness;
+	u32 brightness_mask;
 };
 
 const struct drm_panel_backlight_quirk *

From bf0365b005d9a486d2ddf0c8ec0a578d0f3aabed Mon Sep 17 00:00:00 2001
From: Antheas Kapenekakis <lkml@antheas.dev>
Date: Fri, 29 Aug 2025 16:55:40 +0200
Subject: [PATCH 0881/1455] drm: panel-backlight-quirks: Add Steam Deck
 brightness quirk

On the SteamOS kernel, Valve universally makes minimum brightness 0
for all devices. SteamOS is (was?) meant for the Steam Deck, so
enabling it universally is reasonable. However, it causes issues in
certain devices. Therefore, introduce it just for the Steam Deck here.

SteamOS kernel does not have a public mirror, but this replaces commit
806dd74bb225 ("amd/drm: override backlight min value from 12 -> 0")
in the latest, as of this writing, SteamOS kernel (6.11.11-valve24).
See unofficial mirror reconstructed from sources below.

Link: https://gitlab.com/evlaV/linux-integration/-/commit/806dd74bb225
Reviewed-by: Robert Beckett <bob.beckett@collabora.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Antheas Kapenekakis <lkml@antheas.dev>
Link: https://lore.kernel.org/r/20250829145541.512671-6-lkml@antheas.dev
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 drivers/gpu/drm/drm_panel_backlight_quirks.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/drm_panel_backlight_quirks.c b/drivers/gpu/drm/drm_panel_backlight_quirks.c
index 2bdbd5583d32..537dc6dd0534 100644
--- a/drivers/gpu/drm/drm_panel_backlight_quirks.c
+++ b/drivers/gpu/drm/drm_panel_backlight_quirks.c
@@ -45,6 +45,21 @@ static const struct drm_get_panel_backlight_quirk drm_panel_min_backlight_quirks
 		.ident.name = "NE135A1M-NY1",
 		.quirk = { .min_brightness = 1, },
 	},
+	/* Steam Deck models */
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "Valve",
+		.dmi_match_other.field = DMI_PRODUCT_NAME,
+		.dmi_match_other.value = "Jupiter",
+		.quirk = { .min_brightness = 1, },
+	},
+	{
+		.dmi_match.field = DMI_SYS_VENDOR,
+		.dmi_match.value = "Valve",
+		.dmi_match_other.field = DMI_PRODUCT_NAME,
+		.dmi_match_other.value = "Galileo",
+		.quirk = { .min_brightness = 1, },
+	},
 	/* Have OLED Panels with brightness issue when last byte is 0/1 */
 	{
 		.dmi_match.field = DMI_SYS_VENDOR,

From a3ae3384be7704fcf41279f13190bd8a11204bea Mon Sep 17 00:00:00 2001
From: Antheas Kapenekakis <lkml@antheas.dev>
Date: Fri, 29 Aug 2025 16:55:41 +0200
Subject: [PATCH 0882/1455] drm: panel-backlight-quirks: Log applied panel
 brightness quirks

Currently, when a panel brightness quirk is applied, there is no log
indicating that a quirk was applied. Unwrap the drm device on its own
and use drm_info() to log when a quirk is applied.

Suggested-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Antheas Kapenekakis <lkml@antheas.dev>
Link: https://lore.kernel.org/r/20250829145541.512671-7-lkml@antheas.dev
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
(Correct a missing -1 in the message math)
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c    | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 156f2aae6828..18f0e0912ea6 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3617,13 +3617,15 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 	struct drm_connector *conn_base;
 	struct amdgpu_device *adev;
 	struct drm_luminance_range_info *luminance_range;
+	struct drm_device *drm;
 
 	if (aconnector->bl_idx == -1 ||
 	    aconnector->dc_link->connector_signal != SIGNAL_TYPE_EDP)
 		return;
 
 	conn_base = &aconnector->base;
-	adev = drm_to_adev(conn_base->dev);
+	drm = conn_base->dev;
+	adev = drm_to_adev(drm);
 
 	caps = &adev->dm.backlight_caps[aconnector->bl_idx];
 	caps->ext_caps = &aconnector->dc_link->dpcd_sink_ext_caps;
@@ -3659,12 +3661,20 @@ static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
 	panel_backlight_quirk =
 		drm_get_panel_backlight_quirk(aconnector->drm_edid);
 	if (!IS_ERR_OR_NULL(panel_backlight_quirk)) {
-		if (panel_backlight_quirk->min_brightness)
+		if (panel_backlight_quirk->min_brightness) {
 			caps->min_input_signal =
 				panel_backlight_quirk->min_brightness - 1;
-		if (panel_backlight_quirk->brightness_mask)
+			drm_info(drm,
+				 "Applying panel backlight quirk, min_brightness: %d\n",
+				 caps->min_input_signal);
+		}
+		if (panel_backlight_quirk->brightness_mask) {
+			drm_info(drm,
+				 "Applying panel backlight quirk, brightness_mask: 0x%X\n",
+				 panel_backlight_quirk->brightness_mask);
 			caps->brightness_mask =
 				panel_backlight_quirk->brightness_mask;
+		}
 	}
 }
 

From ac77c8f12edd894e8d3fe732d2b1c6c3de09c1aa Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 14:58:12 +0300
Subject: [PATCH 0883/1455] dt-bindings: display/msm: dp-controller: allow eDP
 for SA8775P

On Qualcomm SA8775P the DP controller might be driving either a
DisplayPort or a eDP sink (depending on the PHY that is tied to the
controller). Reflect that in the schema.

Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672569/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-1-7526f0311eaa@oss.qualcomm.com
---
 .../bindings/display/msm/dp-controller.yaml   | 25 ++++++++++++++-----
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index 9923b065323b..aed3bafa67e3 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -174,12 +174,25 @@ allOf:
       properties:
         "#sound-dai-cells": false
     else:
-      properties:
-        aux-bus: false
-        reg:
-          minItems: 5
-      required:
-        - "#sound-dai-cells"
+      if:
+        properties:
+          compatible:
+            contains:
+              enum:
+                - qcom,sa8775p-dp
+      then:
+        oneOf:
+          - required:
+              - aux-bus
+          - required:
+              - "#sound-dai-cells"
+      else:
+        properties:
+          aux-bus: false
+          reg:
+            minItems: 5
+        required:
+          - "#sound-dai-cells"
 
 additionalProperties: false
 

From 5484265fd685d752bb75dd8b34f617dd9991ba5c Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 14:58:13 +0300
Subject: [PATCH 0884/1455] dt-bindings: display/msm: dp-controller: fix
 fallback for SM6350

Currently SM6350 uses qcom,sm8350-dp as a fallback compatible entry.
This works, but adding DP MST support will reveal that this SoC is
not fully compatible with SM8350 platform: the former one doesn't
provide MST support, while the latter one will get it. DT schema for
SM8350 is going to demand MST-related clocks which SM6350 doesn't
provide.

Add new entry for SM6350 with fallback to SC7180 (which belongs to the
same generation and also doesn't have MST support). SC7180 has been
supported by the Linux kernel long ago (and long before SM8350 support
was added).

Fixes: 39086151593a ("dt-bindings: display: msm: dp-controller: document SM6350 compatible")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/672571/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-2-7526f0311eaa@oss.qualcomm.com
---
 .../bindings/display/msm/dp-controller.yaml        | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index aed3bafa67e3..0f814aa6f514 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -31,13 +31,25 @@ properties:
           - qcom,sm8650-dp
       - items:
           - enum:
-              - qcom,sar2130p-dp
               - qcom,sm6350-dp
+          - const: qcom,sc7180-dp
+
+      # deprecated entry for compatibility with old DT
+      - items:
+          - enum:
+              - qcom,sm6350-dp
+          - const: qcom,sm8350-dp
+        deprecated: true
+
+      - items:
+          - enum:
+              - qcom,sar2130p-dp
               - qcom,sm8150-dp
               - qcom,sm8250-dp
               - qcom,sm8450-dp
               - qcom,sm8550-dp
           - const: qcom,sm8350-dp
+
       - items:
           - enum:
               - qcom,sm8750-dp

From afc9e583c0b28f885ef6aa5ef6ea098204c263ed Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 14:58:14 +0300
Subject: [PATCH 0885/1455] dt-bindings: display/msm: dp-controller: document
 DP on SM7150

The qcom,sm7150-dp compatible is documented in schema. Mark DisplayPort
controller as compatible with SM8350.

Fixes: 726eded12dd7 ("dt-bindings: display/msm: Add SM7150 MDSS")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/672572/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-3-7526f0311eaa@oss.qualcomm.com
---
 .../devicetree/bindings/display/msm/dp-controller.yaml      | 1 +
 .../devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml   | 6 ++++--
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index 0f814aa6f514..a18183f7ec21 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -44,6 +44,7 @@ properties:
       - items:
           - enum:
               - qcom,sar2130p-dp
+              - qcom,sm7150-dp
               - qcom,sm8150-dp
               - qcom,sm8250-dp
               - qcom,sm8450-dp
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml
index 13c5d5ffabde..c5d209019124 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml
@@ -61,7 +61,8 @@ patternProperties:
     additionalProperties: true
     properties:
       compatible:
-        const: qcom,sm7150-dp
+        contains:
+          const: qcom,sm7150-dp
 
   "^dsi@[0-9a-f]+$":
     type: object
@@ -378,7 +379,8 @@ examples:
         };
 
         displayport-controller@ae90000 {
-            compatible = "qcom,sm7150-dp";
+            compatible = "qcom,sm7150-dp",
+                         "qcom,sm8350-dp";
             reg = <0xae90000 0x200>,
                   <0xae90200 0x200>,
                   <0xae90400 0xc00>,

From 2c97ee44d29e90e0e1687c15bca4f1aa8eff3d78 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Wed, 3 Sep 2025 14:58:15 +0300
Subject: [PATCH 0886/1455] dt-bindings: display/msm: qcom,x1e80100-mdss:
 correct DP addresses

Fix c&p error and correct example to use 32-bit addressing (as the rest
of the example DT does) instead of 64-bit (as the platform does).  It
got unnoticed before since DP controller node wasn't validated against
DT schema because of the missing compatible.

Fixes: 81de267367d4 ("dt-bindings: display/msm: Document MDSS on X1E80100")
Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672574/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-4-7526f0311eaa@oss.qualcomm.com
---
 .../bindings/display/msm/qcom,x1e80100-mdss.yaml       | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml
index 3b01a0e47333..e35230a86437 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml
@@ -170,11 +170,11 @@ examples:
 
         displayport-controller@ae90000 {
             compatible = "qcom,x1e80100-dp";
-            reg = <0 0xae90000 0 0x200>,
-                  <0 0xae90200 0 0x200>,
-                  <0 0xae90400 0 0x600>,
-                  <0 0xae91000 0 0x400>,
-                  <0 0xae91400 0 0x400>;
+            reg = <0xae90000 0x200>,
+                  <0xae90200 0x200>,
+                  <0xae90400 0x600>,
+                  <0xae91000 0x400>,
+                  <0xae91400 0x400>;
 
             interrupt-parent = <&mdss>;
             interrupts = <12>;

From 3259ffe5b14d977c52e3e56f31cf07ca21fb314d Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Wed, 3 Sep 2025 14:58:16 +0300
Subject: [PATCH 0887/1455] dt-bindings: display/msm: dp-controller: add
 X1E80100

Add X1E80100 to the dp-controller bindings, it has DisplayPort
controller similar to other platforms, but it uses its own compatible
string.

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672578/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-5-7526f0311eaa@oss.qualcomm.com
---
 .../devicetree/bindings/display/msm/dp-controller.yaml         | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index a18183f7ec21..6316c929fd66 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -29,6 +29,8 @@ properties:
           - qcom,sdm845-dp
           - qcom,sm8350-dp
           - qcom,sm8650-dp
+          - qcom,x1e80100-dp
+
       - items:
           - enum:
               - qcom,sm6350-dp
@@ -193,6 +195,7 @@ allOf:
             contains:
               enum:
                 - qcom,sa8775p-dp
+                - qcom,x1e80100-dp
       then:
         oneOf:
           - required:

From 0253f5ef8da491e8e694763c4aa2ac2f48ede932 Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Wed, 3 Sep 2025 14:58:17 +0300
Subject: [PATCH 0888/1455] dt-bindings: display/msm: drop
 assigned-clock-parents for dp controller

DP controller schema documents assigned-clocks and
assigned-clock-parents. However these assignments should not be a part
of the ABI: there are no actual requirements on the order of the
assignments, MST cases require different number of clocks to be
assigned, etc.

Instead of fixing up the documentation, drop the assigned-clock-parents
and assigned-clocks from the bindings. The generic clock/clock.yaml
already covers these properties.

Suggested-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672586/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-6-7526f0311eaa@oss.qualcomm.com
---
 .../devicetree/bindings/display/msm/dp-controller.yaml | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index 6316c929fd66..afe01332d66c 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -86,16 +86,6 @@ properties:
       - const: ctrl_link_iface
       - const: stream_pixel
 
-  assigned-clocks:
-    items:
-      - description: link clock source
-      - description: pixel clock source
-
-  assigned-clock-parents:
-    items:
-      - description: phy 0 parent
-      - description: phy 1 parent
-
   phys:
     maxItems: 1
 

From 9be5c47908e669db4934ef9c129b28bbc879a8be Mon Sep 17 00:00:00 2001
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
Date: Wed, 3 Sep 2025 14:58:18 +0300
Subject: [PATCH 0889/1455] dt-bindings: display/msm: expand to support MST

On a vast majority of Qualcomm chipsets DisplayPort controller can
support several MST streams (up to 4x). To support MST these chipsets
use up to 4 stream pixel clocks for the DisplayPort controller and
several extra register regions. Expand corresponding region and clock
bindings for these platforms and fix example schema files to follow
updated bindings.

Note: On chipsets that support MST, the number of streams supported
can vary between controllers. For example, SA8775P supports 4 MST
streams on mdss_dp0 but only 2 streams on mdss_dp1.

Signed-off-by: Abhinav Kumar <quic_abhinavk@quicinc.com>
Signed-off-by: Jessica Zhang <jessica.zhang@oss.qualcomm.com>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Patchwork: https://patchwork.freedesktop.org/patch/672585/
Link: https://lore.kernel.org/r/20250903-dp_mst_bindings-v8-7-7526f0311eaa@oss.qualcomm.com
---
 .../bindings/display/msm/dp-controller.yaml   | 103 +++++++++++++++++-
 .../display/msm/qcom,sa8775p-mdss.yaml        |  26 ++++-
 .../display/msm/qcom,sar2130p-mdss.yaml       |  10 +-
 .../display/msm/qcom,sc7280-mdss.yaml         |   3 +-
 .../display/msm/qcom,sm7150-mdss.yaml         |  10 +-
 .../display/msm/qcom,sm8750-mdss.yaml         |  10 +-
 .../display/msm/qcom,x1e80100-mdss.yaml       |  10 +-
 7 files changed, 150 insertions(+), 22 deletions(-)

diff --git a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
index afe01332d66c..aeb4e4f36044 100644
--- a/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
+++ b/Documentation/devicetree/bindings/display/msm/dp-controller.yaml
@@ -66,25 +66,37 @@ properties:
       - description: link register block
       - description: p0 register block
       - description: p1 register block
+      - description: p2 register block
+      - description: p3 register block
+      - description: mst2link register block
+      - description: mst3link register block
 
   interrupts:
     maxItems: 1
 
   clocks:
+    minItems: 5
     items:
       - description: AHB clock to enable register access
       - description: Display Port AUX clock
       - description: Display Port Link clock
       - description: Link interface clock between DP and PHY
-      - description: Display Port Pixel clock
+      - description: Display Port stream 0 Pixel clock
+      - description: Display Port stream 1 Pixel clock
+      - description: Display Port stream 2 Pixel clock
+      - description: Display Port stream 3 Pixel clock
 
   clock-names:
+    minItems: 5
     items:
       - const: core_iface
       - const: core_aux
       - const: ctrl_link
       - const: ctrl_link_iface
       - const: stream_pixel
+      - const: stream_1_pixel
+      - const: stream_2_pixel
+      - const: stream_3_pixel
 
   phys:
     maxItems: 1
@@ -166,7 +178,6 @@ required:
 allOf:
   # AUX BUS does not exist on DP controllers
   # Audio output also is present only on DP output
-  # p1 regions is present on DP, but not on eDP
   - if:
       properties:
         compatible:
@@ -195,11 +206,95 @@ allOf:
       else:
         properties:
           aux-bus: false
-          reg:
-            minItems: 5
         required:
           - "#sound-dai-cells"
 
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              # these platforms support SST only
+              - qcom,sc7180-dp
+              - qcom,sc7280-dp
+              - qcom,sc7280-edp
+              - qcom,sc8180x-edp
+              - qcom,sc8280xp-edp
+    then:
+      properties:
+        reg:
+          minItems: 5
+          maxItems: 5
+        clocks:
+          minItems: 5
+          maxItems: 5
+        clocks-names:
+          minItems: 5
+          maxItems: 5
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              # these platforms support 2 streams MST on some interfaces,
+              # others are SST only
+              - qcom,sc8280xp-dp
+              - qcom,x1e80100-dp
+    then:
+      properties:
+        reg:
+          minItems: 5
+          maxItems: 5
+        clocks:
+          minItems: 5
+          maxItems: 6
+        clocks-names:
+          minItems: 5
+          maxItems: 6
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            # 2 streams MST
+            enum:
+              - qcom,sc8180x-dp
+              - qcom,sdm845-dp
+              - qcom,sm8350-dp
+              - qcom,sm8650-dp
+    then:
+      properties:
+        reg:
+          minItems: 5
+          maxItems: 5
+        clocks:
+          minItems: 6
+          maxItems: 6
+        clocks-names:
+          minItems: 6
+          maxItems: 6
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
+              # these platforms support 4 stream MST on first DP,
+              # 2 streams MST on the second one.
+              - qcom,sa8775p-dp
+    then:
+      properties:
+        reg:
+          minItems: 9
+          maxItems: 9
+        clocks:
+          minItems: 6
+          maxItems: 8
+        clocks-names:
+          minItems: 6
+          maxItems: 8
+
 additionalProperties: false
 
 examples:
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml
index 1053b3bc4908..e2730a2f25cf 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sa8775p-mdss.yaml
@@ -375,7 +375,11 @@ examples:
                   <0xaf54200 0x0c0>,
                   <0xaf55000 0x770>,
                   <0xaf56000 0x09c>,
-                  <0xaf57000 0x09c>;
+                  <0xaf57000 0x09c>,
+                  <0xaf58000 0x09c>,
+                  <0xaf59000 0x09c>,
+                  <0xaf5a000 0x23c>,
+                  <0xaf5b000 0x23c>;
 
             interrupt-parent = <&mdss0>;
             interrupts = <12>;
@@ -384,16 +388,28 @@ examples:
                      <&dispcc_dptx0_aux_clk>,
                      <&dispcc_dptx0_link_clk>,
                      <&dispcc_dptx0_link_intf_clk>,
-                     <&dispcc_dptx0_pixel0_clk>;
+                     <&dispcc_dptx0_pixel0_clk>,
+                     <&dispcc_dptx0_pixel1_clk>,
+                     <&dispcc_dptx0_pixel2_clk>,
+                     <&dispcc_dptx0_pixel3_clk>;
             clock-names = "core_iface",
                           "core_aux",
                           "ctrl_link",
                           "ctrl_link_iface",
-                          "stream_pixel";
+                          "stream_pixel",
+                          "stream_1_pixel",
+                          "stream_2_pixel",
+                          "stream_3_pixel";
 
             assigned-clocks = <&dispcc_mdss_dptx0_link_clk_src>,
-                              <&dispcc_mdss_dptx0_pixel0_clk_src>;
-            assigned-clock-parents = <&mdss0_dp0_phy 0>, <&mdss0_dp0_phy 1>;
+                              <&dispcc_mdss_dptx0_pixel0_clk_src>,
+                              <&dispcc_mdss_dptx0_pixel1_clk_src>,
+                              <&dispcc_mdss_dptx0_pixel2_clk_src>,
+                              <&dispcc_mdss_dptx0_pixel3_clk_src>;
+            assigned-clock-parents = <&mdss0_dp0_phy 0>,
+                                     <&mdss0_dp0_phy 1>,
+                                     <&mdss0_dp0_phy 1>,
+                                     <&mdss0_dp0_phy 1>;
 
             phys = <&mdss0_dp0_phy>;
             phy-names = "dp";
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml
index 870144b53cec..44c1bb9e4109 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sar2130p-mdss.yaml
@@ -207,16 +207,20 @@ examples:
                      <&dispcc_disp_cc_mdss_dptx0_aux_clk>,
                      <&dispcc_disp_cc_mdss_dptx0_link_clk>,
                      <&dispcc_disp_cc_mdss_dptx0_link_intf_clk>,
-                     <&dispcc_disp_cc_mdss_dptx0_pixel0_clk>;
+                     <&dispcc_disp_cc_mdss_dptx0_pixel0_clk>,
+                     <&dispcc_disp_cc_mdss_dptx0_pixel1_clk>;
             clock-names = "core_iface",
                           "core_aux",
                           "ctrl_link",
                           "ctrl_link_iface",
-                          "stream_pixel";
+                          "stream_pixel",
+                          "stream_1_pixel";
 
             assigned-clocks = <&dispcc_disp_cc_mdss_dptx0_link_clk_src>,
-                              <&dispcc_disp_cc_mdss_dptx0_pixel0_clk_src>;
+                              <&dispcc_disp_cc_mdss_dptx0_pixel0_clk_src>,
+                              <&dispcc_disp_cc_mdss_dptx0_pixel1_clk_src>;
             assigned-clock-parents = <&usb_dp_qmpphy_QMP_USB43DP_DP_LINK_CLK>,
+                                     <&usb_dp_qmpphy_QMP_USB43DP_DP_VCO_DIV_CLK>,
                                      <&usb_dp_qmpphy_QMP_USB43DP_DP_VCO_DIV_CLK>;
 
             phys = <&usb_dp_qmpphy QMP_USB43DP_DP_PHY>;
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sc7280-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sc7280-mdss.yaml
index 2947f27e0585..b643d3adf669 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sc7280-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sc7280-mdss.yaml
@@ -281,7 +281,8 @@ examples:
             reg = <0xaea0000 0x200>,
                   <0xaea0200 0x200>,
                   <0xaea0400 0xc00>,
-                  <0xaea1000 0x400>;
+                  <0xaea1000 0x400>,
+                  <0xaea1400 0x400>;
 
             interrupt-parent = <&mdss>;
             interrupts = <14>;
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml
index c5d209019124..9b0621d88d50 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm7150-mdss.yaml
@@ -394,16 +394,20 @@ examples:
                      <&dispcc_mdss_dp_aux_clk>,
                      <&dispcc_mdss_dp_link_clk>,
                      <&dispcc_mdss_dp_link_intf_clk>,
-                     <&dispcc_mdss_dp_pixel_clk>;
+                     <&dispcc_mdss_dp_pixel_clk>,
+                     <&dispcc_mdss_dp_pixel1_clk>;
             clock-names = "core_iface",
                           "core_aux",
                           "ctrl_link",
                           "ctrl_link_iface",
-                          "stream_pixel";
+                          "stream_pixel",
+                          "stream_1_pixel";
 
             assigned-clocks = <&dispcc_mdss_dp_link_clk_src>,
-                              <&dispcc_mdss_dp_pixel_clk_src>;
+                              <&dispcc_mdss_dp_pixel_clk_src>,
+                              <&dispcc_mdss_dp_pixel1_clk_src>;
             assigned-clock-parents = <&dp_phy 0>,
+                                     <&dp_phy 1>,
                                      <&dp_phy 1>;
 
             operating-points-v2 = <&dp_opp_table>;
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml
index 72c70edc1fb0..4151f475f3bc 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,sm8750-mdss.yaml
@@ -401,16 +401,20 @@ examples:
                          <&disp_cc_mdss_dptx0_aux_clk>,
                          <&disp_cc_mdss_dptx0_link_clk>,
                          <&disp_cc_mdss_dptx0_link_intf_clk>,
-                         <&disp_cc_mdss_dptx0_pixel0_clk>;
+                         <&disp_cc_mdss_dptx0_pixel0_clk>,
+                         <&disp_cc_mdss_dptx0_pixel1_clk>;
                 clock-names = "core_iface",
                               "core_aux",
                               "ctrl_link",
                               "ctrl_link_iface",
-                              "stream_pixel";
+                              "stream_pixel",
+                              "stream_1_pixel";
 
                 assigned-clocks = <&disp_cc_mdss_dptx0_link_clk_src>,
-                                  <&disp_cc_mdss_dptx0_pixel0_clk_src>;
+                                  <&disp_cc_mdss_dptx0_pixel0_clk_src>,
+                                  <&disp_cc_mdss_dptx0_pixel1_clk_src>;
                 assigned-clock-parents = <&usb_dp_qmpphy QMP_USB43DP_DP_LINK_CLK>,
+                                         <&usb_dp_qmpphy QMP_USB43DP_DP_VCO_DIV_CLK>,
                                          <&usb_dp_qmpphy QMP_USB43DP_DP_VCO_DIV_CLK>;
 
                 operating-points-v2 = <&dp_opp_table>;
diff --git a/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml b/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml
index e35230a86437..8d698a2e055a 100644
--- a/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml
+++ b/Documentation/devicetree/bindings/display/msm/qcom,x1e80100-mdss.yaml
@@ -183,15 +183,19 @@ examples:
                <&dispcc_dptx0_aux_clk>,
                <&dispcc_dptx0_link_clk>,
                <&dispcc_dptx0_link_intf_clk>,
-               <&dispcc_dptx0_pixel0_clk>;
+               <&dispcc_dptx0_pixel0_clk>,
+               <&dispcc_dptx0_pixel1_clk>;
             clock-names = "core_iface", "core_aux",
                     "ctrl_link",
                     "ctrl_link_iface",
-                    "stream_pixel";
+                    "stream_pixel",
+                    "stream_1_pixel";
 
             assigned-clocks = <&dispcc_mdss_dptx0_link_clk_src>,
-                  <&dispcc_mdss_dptx0_pixel0_clk_src>;
+                  <&dispcc_mdss_dptx0_pixel0_clk_src>,
+                  <&dispcc_mdss_dptx0_pixel1_clk_src>;
             assigned-clock-parents = <&usb_1_ss0_qmpphy QMP_USB43DP_DP_LINK_CLK>,
+                  <&usb_1_ss0_qmpphy QMP_USB43DP_DP_VCO_DIV_CLK>,
                   <&usb_1_ss0_qmpphy QMP_USB43DP_DP_VCO_DIV_CLK>;
 
             operating-points-v2 = <&mdss_dp0_opp_table>;

From e339a73737d365dc88e1994d561112ef2c21ad88 Mon Sep 17 00:00:00 2001
From: Eric Biggers <ebiggers@kernel.org>
Date: Thu, 21 Aug 2025 13:56:13 -0400
Subject: [PATCH 0890/1455] drm/bridge: it6505: Use SHA-1 library instead of
 crypto_shash

Instead of using the "sha1" crypto_shash, simply call the sha1() library
function.  This is simpler and faster.

Signed-off-by: Eric Biggers <ebiggers@kernel.org>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://patch.msgid.link/20250821175613.14717-1-ebiggers@kernel.org
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
---
 drivers/gpu/drm/bridge/Kconfig      |  3 +--
 drivers/gpu/drm/bridge/ite-it6505.c | 33 ++---------------------------
 2 files changed, 3 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig
index 6945029b3592..a250afd8d662 100644
--- a/drivers/gpu/drm/bridge/Kconfig
+++ b/drivers/gpu/drm/bridge/Kconfig
@@ -120,8 +120,7 @@ config DRM_ITE_IT6505
 	select DRM_DISPLAY_DP_AUX_BUS
 	select DRM_KMS_HELPER
 	select EXTCON
-	select CRYPTO
-	select CRYPTO_HASH
+	select CRYPTO_LIB_SHA1
 	select REGMAP_I2C
 	help
 	  ITE IT6505 DisplayPort bridge chip driver.
diff --git a/drivers/gpu/drm/bridge/ite-it6505.c b/drivers/gpu/drm/bridge/ite-it6505.c
index 89649c17ffad..a094803ba7aa 100644
--- a/drivers/gpu/drm/bridge/ite-it6505.c
+++ b/drivers/gpu/drm/bridge/ite-it6505.c
@@ -21,7 +21,7 @@
 #include <linux/wait.h>
 #include <linux/bitfield.h>
 
-#include <crypto/hash.h>
+#include <crypto/sha1.h>
 
 #include <drm/display/drm_dp_helper.h>
 #include <drm/display/drm_hdcp_helper.h>
@@ -2107,35 +2107,6 @@ static void it6505_hdcp_part1_auth(struct it6505 *it6505)
 	it6505->hdcp_status = HDCP_AUTH_GOING;
 }
 
-static int it6505_sha1_digest(struct it6505 *it6505, u8 *sha1_input,
-			      unsigned int size, u8 *output_av)
-{
-	struct shash_desc *desc;
-	struct crypto_shash *tfm;
-	int err;
-	struct device *dev = it6505->dev;
-
-	tfm = crypto_alloc_shash("sha1", 0, 0);
-	if (IS_ERR(tfm)) {
-		dev_err(dev, "crypto_alloc_shash sha1 failed");
-		return PTR_ERR(tfm);
-	}
-	desc = kzalloc(sizeof(*desc) + crypto_shash_descsize(tfm), GFP_KERNEL);
-	if (!desc) {
-		crypto_free_shash(tfm);
-		return -ENOMEM;
-	}
-
-	desc->tfm = tfm;
-	err = crypto_shash_digest(desc, sha1_input, size, output_av);
-	if (err)
-		dev_err(dev, "crypto_shash_digest sha1 failed");
-
-	crypto_free_shash(tfm);
-	kfree(desc);
-	return err;
-}
-
 static int it6505_setup_sha1_input(struct it6505 *it6505, u8 *sha1_input)
 {
 	struct device *dev = it6505->dev;
@@ -2205,7 +2176,7 @@ static bool it6505_hdcp_part2_ksvlist_check(struct it6505 *it6505)
 		return false;
 	}
 
-	it6505_sha1_digest(it6505, it6505->sha1_input, i, (u8 *)av);
+	sha1(it6505->sha1_input, i, (u8 *)av);
 	/*1B-05 V' must retry 3 times */
 	for (retry = 0; retry < 3; retry++) {
 		err = it6505_get_dpcd(it6505, DP_AUX_HDCP_V_PRIME(0), (u8 *)bv,

From 294912f93d0fd4d537b7605d2e5fa6f735794c75 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 28 Aug 2025 16:19:32 +0530
Subject: [PATCH 0891/1455] drm/xe: Fix indentation in
 xe_zap_ptes_in_madvise_range

Fix misleading indentation around WRITE_ONCE in pte zap loop.
No functional change intended.

Fixes: ada7486c5668 ("drm/xe: Implement madvise ioctl for xe")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://lore.kernel.org/r/20250828104933.3839825-2-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm_madvise.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 09c5783ee523..732577dfe519 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -201,12 +201,12 @@ static u8 xe_zap_ptes_in_madvise_range(struct xe_vm *vm, u64 start, u64 end)
 				if (xe_pt_zap_ptes(tile, vma)) {
 					tile_mask |= BIT(id);
 
-				/*
-				 * WRITE_ONCE pairs with READ_ONCE
-				 * in xe_vm_has_valid_gpu_mapping()
-				 */
-				WRITE_ONCE(vma->tile_invalidated,
-					   vma->tile_invalidated | BIT(id));
+					/*
+					 * WRITE_ONCE pairs with READ_ONCE
+					 * in xe_vm_has_valid_gpu_mapping()
+					 */
+					WRITE_ONCE(vma->tile_invalidated,
+						   vma->tile_invalidated | BIT(id));
 				}
 			}
 		}

From fece859855497d444b5efef23d022ab788f14303 Mon Sep 17 00:00:00 2001
From: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Date: Thu, 28 Aug 2025 16:19:33 +0530
Subject: [PATCH 0892/1455] drm/xe/vm: Fix error handling in
 xe_vm_query_vmas_attrs_ioctl()

copy_to_user() returns the number of bytes not copied on failure, not a
negative error code. Update the logic to return -EFAULT instead of the
number of bytes to correctly signal the error.

Fixes: 418807860e94 ("drm/xe/uapi: Add UAPI for querying VMA count and memory attributes")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://lore.kernel.org/r/20250828104933.3839825-3-himal.prasad.ghimiray@intel.com
Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 3ff3c67aa79d..f55f96bb240a 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2288,6 +2288,8 @@ int xe_vm_query_vmas_attrs_ioctl(struct drm_device *dev, void *data, struct drm_
 
 	err = copy_to_user(attrs_user, mem_attrs,
 			   args->sizeof_mem_range_attr * args->num_mem_ranges);
+	if (err)
+		err = -EFAULT;
 
 free_mem_attrs:
 	kvfree(mem_attrs);

From 2a1eea8fd601db4c52f0d14f8871663b7b052c91 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Wed, 3 Sep 2025 09:31:06 +0100
Subject: [PATCH 0893/1455] drm/sysfb: Remove double assignment to pointer
 crtc_state

The declaration of pointer crtc_state includes an assignment to
crtc_state. The double assignment of crtc_state is redundant and
can be removed.

Fixes: 061963cd9e5b ("drm/sysfb: Blit to CRTC destination format")
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250903083106.2703580-1-colin.i.king@gmail.com
---
 drivers/gpu/drm/sysfb/drm_sysfb_modeset.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
index 963c380fea64..ddb4a7523ee6 100644
--- a/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
+++ b/drivers/gpu/drm/sysfb/drm_sysfb_modeset.c
@@ -238,8 +238,7 @@ void drm_sysfb_plane_helper_atomic_update(struct drm_plane *plane, struct drm_at
 	struct drm_shadow_plane_state *shadow_plane_state = to_drm_shadow_plane_state(plane_state);
 	struct drm_framebuffer *fb = plane_state->fb;
 	unsigned int dst_pitch = sysfb->fb_pitch;
-	struct drm_crtc_state *crtc_state = crtc_state =
-		drm_atomic_get_new_crtc_state(state, plane_state->crtc);
+	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state, plane_state->crtc);
 	struct drm_sysfb_crtc_state *sysfb_crtc_state = to_drm_sysfb_crtc_state(crtc_state);
 	const struct drm_format_info *dst_format = sysfb_crtc_state->format;
 	struct drm_atomic_helper_damage_iter iter;

From a5db45bfe9469e85e416035367ead7e0b94c642c Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 08:49:20 +0200
Subject: [PATCH 0894/1455] drm/ast: Do not print DRAM info

Most of the information in the DRAM status output is irrelevant; some
is even wrong. Only the DRAM type is used on some older models. Drop
the output entirely.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/20250826065032.344412-2-tzimmermann@suse.de
---
 drivers/gpu/drm/ast/ast_main.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 44b9b5f659fc..830f1c5fe893 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -355,8 +355,6 @@ struct drm_device *ast_device_create(struct pci_dev *pdev,
 	ret = ast_get_dram_info(ast);
 	if (ret)
 		return ERR_PTR(ret);
-	drm_info(dev, "dram MCLK=%u Mhz type=%d bus_width=%d\n",
-		 ast->mclk, ast->dram_type, ast->dram_bus_width);
 
 	ast_detect_tx_chip(ast, need_post);
 	switch (ast->tx_chip) {

From dc2a40f44e5fb815c3b4506debfe73ada4ea1640 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 08:49:21 +0200
Subject: [PATCH 0895/1455] drm/ast: Remove unused dram_bus_width field

The DRAM bus width is not necessary for the driver. Remove it.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/20250826065032.344412-3-tzimmermann@suse.de
---
 drivers/gpu/drm/ast/ast_drv.h  | 1 -
 drivers/gpu/drm/ast/ast_main.c | 6 ------
 2 files changed, 7 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index e37a55295ed7..403b86f00a54 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -172,7 +172,6 @@ struct ast_device {
 	enum ast_config_mode config_mode;
 	enum ast_chip chip;
 
-	uint32_t dram_bus_width;
 	uint32_t dram_type;
 	uint32_t mclk;
 
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 830f1c5fe893..0b3fb6856db7 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -242,7 +242,6 @@ static int ast_get_dram_info(struct ast_device *ast)
 		break;
 	case ast_use_defaults:
 	default:
-		ast->dram_bus_width = 16;
 		ast->dram_type = AST_DRAM_1Gx16;
 		if (IS_AST_GEN6(ast))
 			ast->mclk = 800;
@@ -251,11 +250,6 @@ static int ast_get_dram_info(struct ast_device *ast)
 		return 0;
 	}
 
-	if (mcr_cfg & 0x40)
-		ast->dram_bus_width = 16;
-	else
-		ast->dram_bus_width = 32;
-
 	if (IS_AST_GEN6(ast)) {
 		switch (mcr_cfg & 0x03) {
 		case 0:

From e91153028ddba13ec545afcf9b6b350191ee4bf7 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 08:49:22 +0200
Subject: [PATCH 0896/1455] drm/ast: Remove unused mclk field

The memory clock is not necessary for the driver. In default for
AST2600 is event incorrect; should be 800 MHz. Remove it.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/20250826065032.344412-4-tzimmermann@suse.de
---
 drivers/gpu/drm/ast/ast_drv.h  |  1 -
 drivers/gpu/drm/ast/ast_main.c | 26 --------------------------
 2 files changed, 27 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 403b86f00a54..c9c933b5a70d 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -173,7 +173,6 @@ struct ast_device {
 	enum ast_chip chip;
 
 	uint32_t dram_type;
-	uint32_t mclk;
 
 	void __iomem	*vram;
 	unsigned long	vram_base;
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 0b3fb6856db7..492cc8461eb0 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -215,7 +215,6 @@ static int ast_get_dram_info(struct ast_device *ast)
 	struct drm_device *dev = &ast->base;
 	struct device_node *np = dev->dev->of_node;
 	uint32_t mcr_cfg, mcr_scu_mpll, mcr_scu_strap;
-	uint32_t denum, num, div, ref_pll, dsel;
 
 	switch (ast->config_mode) {
 	case ast_use_dt:
@@ -243,10 +242,6 @@ static int ast_get_dram_info(struct ast_device *ast)
 	case ast_use_defaults:
 	default:
 		ast->dram_type = AST_DRAM_1Gx16;
-		if (IS_AST_GEN6(ast))
-			ast->mclk = 800;
-		else
-			ast->mclk = 396;
 		return 0;
 	}
 
@@ -300,27 +295,6 @@ static int ast_get_dram_info(struct ast_device *ast)
 		}
 	}
 
-	if (mcr_scu_strap & 0x2000)
-		ref_pll = 14318;
-	else
-		ref_pll = 12000;
-
-	denum = mcr_scu_mpll & 0x1f;
-	num = (mcr_scu_mpll & 0x3fe0) >> 5;
-	dsel = (mcr_scu_mpll & 0xc000) >> 14;
-	switch (dsel) {
-	case 3:
-		div = 0x4;
-		break;
-	case 2:
-	case 1:
-		div = 0x2;
-		break;
-	default:
-		div = 0x1;
-		break;
-	}
-	ast->mclk = ref_pll * (num + 2) / ((denum + 2) * (div * 1000));
 	return 0;
 }
 

From db216056a63b9ca415b72963605defad8107d6fc Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 08:49:23 +0200
Subject: [PATCH 0897/1455] drm/ast: Remove unused SCU-MPLL and SCU-STRAP
 values

The ast driver used SCU-MPLL and SCU-STRAP to compute the memory
clock. Remove the now unused values.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/20250826065032.344412-5-tzimmermann@suse.de
---
 drivers/gpu/drm/ast/ast_main.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 492cc8461eb0..41ff880cfdec 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -214,7 +214,7 @@ static int ast_get_dram_info(struct ast_device *ast)
 {
 	struct drm_device *dev = &ast->base;
 	struct device_node *np = dev->dev->of_node;
-	uint32_t mcr_cfg, mcr_scu_mpll, mcr_scu_strap;
+	uint32_t mcr_cfg;
 
 	switch (ast->config_mode) {
 	case ast_use_dt:
@@ -222,22 +222,13 @@ static int ast_get_dram_info(struct ast_device *ast)
 		 * If some properties are missing, use reasonable
 		 * defaults for GEN5
 		 */
-		if (of_property_read_u32(np, "aspeed,mcr-configuration",
-					 &mcr_cfg))
+		if (of_property_read_u32(np, "aspeed,mcr-configuration", &mcr_cfg))
 			mcr_cfg = 0x00000577;
-		if (of_property_read_u32(np, "aspeed,mcr-scu-mpll",
-					 &mcr_scu_mpll))
-			mcr_scu_mpll = 0x000050C0;
-		if (of_property_read_u32(np, "aspeed,mcr-scu-strap",
-					 &mcr_scu_strap))
-			mcr_scu_strap = 0;
 		break;
 	case ast_use_p2a:
 		ast_write32(ast, 0xf004, 0x1e6e0000);
 		ast_write32(ast, 0xf000, 0x1);
 		mcr_cfg = ast_read32(ast, 0x10004);
-		mcr_scu_mpll = ast_read32(ast, 0x10120);
-		mcr_scu_strap = ast_read32(ast, 0x10170);
 		break;
 	case ast_use_defaults:
 	default:

From 6d580cf8093fc640e6e4cef3932ef176f1ea1c87 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 08:49:24 +0200
Subject: [PATCH 0898/1455] drm/ast: Move DRAM info next to its only user

The only place in the ast driver that uses the DRAM type is the
P2A DRAM initialization for Gen2 and Gen3 of the chip. Condense
the code in ast_get_dram_info() to exactly this use case and move
it into the Gen's custom source file. Remove the field dram_type
from struct ast_device.

The AST_DRAM_ constants are also used in Gen4 POST helpers, but
independently from the dram_type field. No changes there.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/20250826065032.344412-6-tzimmermann@suse.de
---
 drivers/gpu/drm/ast/ast_2100.c | 47 +++++++++++++++++--
 drivers/gpu/drm/ast/ast_drv.h  |  2 -
 drivers/gpu/drm/ast/ast_main.c | 83 ----------------------------------
 3 files changed, 44 insertions(+), 88 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_2100.c b/drivers/gpu/drm/ast/ast_2100.c
index 477ee15eff5d..44c33dd050eb 100644
--- a/drivers/gpu/drm/ast/ast_2100.c
+++ b/drivers/gpu/drm/ast/ast_2100.c
@@ -31,6 +31,38 @@
 #include "ast_drv.h"
 #include "ast_post.h"
 
+/*
+ * DRAM type
+ */
+
+static int ast_2100_get_dram_type_p2a(struct ast_device *ast)
+{
+	u32 mcr_cfg;
+	int dram_type;
+
+	ast_write32(ast, 0xf004, 0x1e6e0000);
+	ast_write32(ast, 0xf000, 0x1);
+	mcr_cfg = ast_read32(ast, 0x10004);
+
+	switch (mcr_cfg & 0x0c) {
+	case 0:
+	case 4:
+		dram_type = AST_DRAM_512Mx16;
+		break;
+	case 8:
+		if (mcr_cfg & 0x40)
+			dram_type = AST_DRAM_1Gx16;
+		else
+			dram_type = AST_DRAM_512Mx32;
+		break;
+	case 0xc:
+		dram_type = AST_DRAM_1Gx32;
+		break;
+	}
+
+	return dram_type;
+}
+
 /*
  * POST
  */
@@ -266,6 +298,9 @@ static void ast_post_chip_2100(struct ast_device *ast)
 	u8 j;
 	u32 data, temp, i;
 	const struct ast_dramstruct *dram_reg_info;
+	int dram_type;
+
+	dram_type = ast_2100_get_dram_type_p2a(ast);
 
 	j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff);
 
@@ -292,11 +327,17 @@ static void ast_post_chip_2100(struct ast_device *ast)
 				for (i = 0; i < 15; i++)
 					udelay(dram_reg_info->data);
 			} else if (AST_DRAMSTRUCT_IS(dram_reg_info, DRAM_TYPE)) {
-				data = dram_reg_info->data;
-				if (ast->dram_type == AST_DRAM_1Gx16)
+				switch (dram_type) {
+				case AST_DRAM_1Gx16:
 					data = 0x00000d89;
-				else if (ast->dram_type == AST_DRAM_1Gx32)
+					break;
+				case AST_DRAM_1Gx32:
 					data = 0x00000c8d;
+					break;
+				default:
+					data = dram_reg_info->data;
+					break;
+				};
 
 				temp = ast_read32(ast, 0x12070);
 				temp &= 0xc;
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index c9c933b5a70d..4c29ae9fb511 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -172,8 +172,6 @@ struct ast_device {
 	enum ast_config_mode config_mode;
 	enum ast_chip chip;
 
-	uint32_t dram_type;
-
 	void __iomem	*vram;
 	unsigned long	vram_base;
 	unsigned long	vram_size;
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 41ff880cfdec..3eea6a6cdacd 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -210,85 +210,6 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post)
 	drm_info(dev, "Using %s\n", info_str[ast->tx_chip]);
 }
 
-static int ast_get_dram_info(struct ast_device *ast)
-{
-	struct drm_device *dev = &ast->base;
-	struct device_node *np = dev->dev->of_node;
-	uint32_t mcr_cfg;
-
-	switch (ast->config_mode) {
-	case ast_use_dt:
-		/*
-		 * If some properties are missing, use reasonable
-		 * defaults for GEN5
-		 */
-		if (of_property_read_u32(np, "aspeed,mcr-configuration", &mcr_cfg))
-			mcr_cfg = 0x00000577;
-		break;
-	case ast_use_p2a:
-		ast_write32(ast, 0xf004, 0x1e6e0000);
-		ast_write32(ast, 0xf000, 0x1);
-		mcr_cfg = ast_read32(ast, 0x10004);
-		break;
-	case ast_use_defaults:
-	default:
-		ast->dram_type = AST_DRAM_1Gx16;
-		return 0;
-	}
-
-	if (IS_AST_GEN6(ast)) {
-		switch (mcr_cfg & 0x03) {
-		case 0:
-			ast->dram_type = AST_DRAM_1Gx16;
-			break;
-		default:
-		case 1:
-			ast->dram_type = AST_DRAM_2Gx16;
-			break;
-		case 2:
-			ast->dram_type = AST_DRAM_4Gx16;
-			break;
-		case 3:
-			ast->dram_type = AST_DRAM_8Gx16;
-			break;
-		}
-	} else if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast)) {
-		switch (mcr_cfg & 0x03) {
-		case 0:
-			ast->dram_type = AST_DRAM_512Mx16;
-			break;
-		default:
-		case 1:
-			ast->dram_type = AST_DRAM_1Gx16;
-			break;
-		case 2:
-			ast->dram_type = AST_DRAM_2Gx16;
-			break;
-		case 3:
-			ast->dram_type = AST_DRAM_4Gx16;
-			break;
-		}
-	} else {
-		switch (mcr_cfg & 0x0c) {
-		case 0:
-		case 4:
-			ast->dram_type = AST_DRAM_512Mx16;
-			break;
-		case 8:
-			if (mcr_cfg & 0x40)
-				ast->dram_type = AST_DRAM_1Gx16;
-			else
-				ast->dram_type = AST_DRAM_512Mx32;
-			break;
-		case 0xc:
-			ast->dram_type = AST_DRAM_1Gx32;
-			break;
-		}
-	}
-
-	return 0;
-}
-
 struct drm_device *ast_device_create(struct pci_dev *pdev,
 				     const struct drm_driver *drv,
 				     enum ast_chip chip,
@@ -311,10 +232,6 @@ struct drm_device *ast_device_create(struct pci_dev *pdev,
 	ast->regs = regs;
 	ast->ioregs = ioregs;
 
-	ret = ast_get_dram_info(ast);
-	if (ret)
-		return ERR_PTR(ret);
-
 	ast_detect_tx_chip(ast, need_post);
 	switch (ast->tx_chip) {
 	case AST_TX_ASTDP:

From 03e7ae93c6e32206797c13118659a966ae84a3bb Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Tue, 26 Aug 2025 08:49:25 +0200
Subject: [PATCH 0899/1455] drm/ast: Put AST_DRAM_ constants into enum
 ast_dram_layout

The AST_DRAM_ constants belong together, so put them in an enum
type. Rename type and variables to 'drm_layout', as there's already
another DRAM type in the ast driver (AST_DDR2, AST_DDR3).

v2:
- avoid compiler warning with switch default (Dan)

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/20250826065032.344412-7-tzimmermann@suse.de
---
 drivers/gpu/drm/ast/ast_2100.c | 21 ++++++++++-----------
 drivers/gpu/drm/ast/ast_drv.h  | 16 +++++++++-------
 2 files changed, 19 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/ast/ast_2100.c b/drivers/gpu/drm/ast/ast_2100.c
index 44c33dd050eb..91541c8eaff7 100644
--- a/drivers/gpu/drm/ast/ast_2100.c
+++ b/drivers/gpu/drm/ast/ast_2100.c
@@ -35,10 +35,10 @@
  * DRAM type
  */
 
-static int ast_2100_get_dram_type_p2a(struct ast_device *ast)
+static enum ast_dram_layout ast_2100_get_dram_layout_p2a(struct ast_device *ast)
 {
 	u32 mcr_cfg;
-	int dram_type;
+	enum ast_dram_layout dram_layout;
 
 	ast_write32(ast, 0xf004, 0x1e6e0000);
 	ast_write32(ast, 0xf000, 0x1);
@@ -47,20 +47,21 @@ static int ast_2100_get_dram_type_p2a(struct ast_device *ast)
 	switch (mcr_cfg & 0x0c) {
 	case 0:
 	case 4:
-		dram_type = AST_DRAM_512Mx16;
+	default:
+		dram_layout = AST_DRAM_512Mx16;
 		break;
 	case 8:
 		if (mcr_cfg & 0x40)
-			dram_type = AST_DRAM_1Gx16;
+			dram_layout = AST_DRAM_1Gx16;
 		else
-			dram_type = AST_DRAM_512Mx32;
+			dram_layout = AST_DRAM_512Mx32;
 		break;
 	case 0xc:
-		dram_type = AST_DRAM_1Gx32;
+		dram_layout = AST_DRAM_1Gx32;
 		break;
 	}
 
-	return dram_type;
+	return dram_layout;
 }
 
 /*
@@ -298,9 +299,7 @@ static void ast_post_chip_2100(struct ast_device *ast)
 	u8 j;
 	u32 data, temp, i;
 	const struct ast_dramstruct *dram_reg_info;
-	int dram_type;
-
-	dram_type = ast_2100_get_dram_type_p2a(ast);
+	enum ast_dram_layout dram_layout  = ast_2100_get_dram_layout_p2a(ast);
 
 	j = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd0, 0xff);
 
@@ -327,7 +326,7 @@ static void ast_post_chip_2100(struct ast_device *ast)
 				for (i = 0; i < 15; i++)
 					udelay(dram_reg_info->data);
 			} else if (AST_DRAMSTRUCT_IS(dram_reg_info, DRAM_TYPE)) {
-				switch (dram_type) {
+				switch (dram_layout) {
 				case AST_DRAM_1Gx16:
 					data = 0x00000d89;
 					break;
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index 4c29ae9fb511..c15aef014f69 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -98,13 +98,15 @@ enum ast_config_mode {
 	ast_use_defaults
 };
 
-#define AST_DRAM_512Mx16 0
-#define AST_DRAM_1Gx16   1
-#define AST_DRAM_512Mx32 2
-#define AST_DRAM_1Gx32   3
-#define AST_DRAM_2Gx16   6
-#define AST_DRAM_4Gx16   7
-#define AST_DRAM_8Gx16   8
+enum ast_dram_layout {
+	AST_DRAM_512Mx16 = 0,
+	AST_DRAM_1Gx16 = 1,
+	AST_DRAM_512Mx32 = 2,
+	AST_DRAM_1Gx32 = 3,
+	AST_DRAM_2Gx16 = 6,
+	AST_DRAM_4Gx16 = 7,
+	AST_DRAM_8Gx16 = 8,
+};
 
 /*
  * Hardware cursor

From a304609be24ec94ddb9b18a5cf6357e58466ac5e Mon Sep 17 00:00:00 2001
From: Luca Coelho <luciano.coelho@intel.com>
Date: Wed, 27 Aug 2025 13:12:06 +0300
Subject: [PATCH 0900/1455] drm/i915: use REG_BIT on FW_BLC_SELF_* macros

Use REG_BIT() instead of open coding the shift in the FW_BLC_SELF_*
macro definitions to avoid potentially typing them as 'int'.

For example, this happens when we pass them to _MASKED_BIT_ENABLE(),
because of the typeof() construct there.  When we pass 1 << 15 (the
FW_BLC_SELF_EN macro), we get typeof(1 << 15), which is 'int'.  Then
the value becomes negative (-2147450880) and we try to assign it to a
'u32'.

In practice this is not a problem though, because when we try to
assign -2147450880 to the u32, that becomes 0x80008000, which was the
intended result.

Link: https://lore.kernel.org/intel-gfx/20250827111109.401604-1-luciano.coelho@intel.com/
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
---
 drivers/gpu/drm/i915/i915_reg.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index b283b25d8368..354ef75ef6a5 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -412,9 +412,9 @@
 #define FW_BLC		_MMIO(0x20d8)
 #define FW_BLC2		_MMIO(0x20dc)
 #define FW_BLC_SELF	_MMIO(0x20e0) /* 915+ only */
-#define   FW_BLC_SELF_EN_MASK      (1 << 31)
-#define   FW_BLC_SELF_FIFO_MASK    (1 << 16) /* 945 only */
-#define   FW_BLC_SELF_EN           (1 << 15) /* 945 only */
+#define   FW_BLC_SELF_EN_MASK      REG_BIT(31)
+#define   FW_BLC_SELF_FIFO_MASK    REG_BIT(16) /* 945 only */
+#define   FW_BLC_SELF_EN           REG_BIT(15) /* 945 only */
 #define MM_BURST_LENGTH     0x00700000
 #define MM_FIFO_WATERMARK   0x0001F000
 #define LM_BURST_LENGTH     0x00000700

From a85ead6d7f74438bc779927fe7b78b0c86addb6b Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 22:11:03 +0200
Subject: [PATCH 0901/1455] drm/xe/debugfs: Move sa_info from gt to tile
 directory

Our drm-based suballocator is implemented per-tile so it is better
to show its debug information also per-tile debugfs directory, not
under per-gt directory as it is done today.

To allow adding more per-tile attributes, prepare necessary helper
functions, like we already did for per-gt or per-uc attributes.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250829201106.1263-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/Makefile          |   1 +
 drivers/gpu/drm/xe/xe_debugfs.c      |  20 +---
 drivers/gpu/drm/xe/xe_gt_debugfs.c   |  13 ---
 drivers/gpu/drm/xe/xe_tile_debugfs.c | 135 +++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_tile_debugfs.h |  13 +++
 5 files changed, 151 insertions(+), 31 deletions(-)
 create mode 100644 drivers/gpu/drm/xe/xe_tile_debugfs.c
 create mode 100644 drivers/gpu/drm/xe/xe_tile_debugfs.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 987e4fe10538..95c0f0a9b403 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -325,6 +325,7 @@ ifeq ($(CONFIG_DEBUG_FS),y)
 		xe_gt_stats.o \
 		xe_guc_debugfs.o \
 		xe_huc_debugfs.o \
+		xe_tile_debugfs.o \
 		xe_uc_debugfs.o
 
 	xe-$(CONFIG_PCI_IOV) += xe_gt_sriov_pf_debugfs.o
diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 8d6df6bd885e..4b71570529a6 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -25,6 +25,7 @@
 #include "xe_sriov.h"
 #include "xe_sriov_pf.h"
 #include "xe_step.h"
+#include "xe_tile_debugfs.h"
 #include "xe_wa.h"
 #include "xe_vsec.h"
 
@@ -329,23 +330,6 @@ static const struct file_operations atomic_svm_timeslice_ms_fops = {
 	.write = atomic_svm_timeslice_ms_set,
 };
 
-static void create_tile_debugfs(struct xe_tile *tile, struct dentry *root)
-{
-	char name[8];
-
-	snprintf(name, sizeof(name), "tile%u", tile->id);
-	tile->debugfs = debugfs_create_dir(name, root);
-	if (IS_ERR(tile->debugfs))
-		return;
-
-	/*
-	 * Store the xe_tile pointer as private data of the tile/ directory
-	 * node so other tile specific attributes under that directory may
-	 * refer to it by looking at its parent node private data.
-	 */
-	tile->debugfs->d_inode->i_private = tile;
-}
-
 void xe_debugfs_register(struct xe_device *xe)
 {
 	struct ttm_device *bdev = &xe->ttm;
@@ -398,7 +382,7 @@ void xe_debugfs_register(struct xe_device *xe)
 		ttm_resource_manager_create_debugfs(man, root, "stolen_mm");
 
 	for_each_tile(tile, xe, tile_id)
-		create_tile_debugfs(tile, root);
+		xe_tile_debugfs_register(tile);
 
 	for_each_gt(gt, xe, id)
 		xe_gt_debugfs_register(gt);
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index 7411c5f5c739..5aa1eded278d 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -123,18 +123,6 @@ static int powergate_info(struct xe_gt *gt, struct drm_printer *p)
 	return ret;
 }
 
-static int sa_info(struct xe_gt *gt, struct drm_printer *p)
-{
-	struct xe_tile *tile = gt_to_tile(gt);
-
-	xe_pm_runtime_get(gt_to_xe(gt));
-	drm_suballoc_dump_debug_info(&tile->mem.kernel_bb_pool->base, p,
-				     xe_sa_manager_gpu_addr(tile->mem.kernel_bb_pool));
-	xe_pm_runtime_put(gt_to_xe(gt));
-
-	return 0;
-}
-
 static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p)
 {
 	struct xe_tile *tile = gt_to_tile(gt);
@@ -316,7 +304,6 @@ static int hwconfig(struct xe_gt *gt, struct drm_printer *p)
  * - without access to the PF specific data
  */
 static const struct drm_info_list vf_safe_debugfs_list[] = {
-	{"sa_info", .show = xe_gt_debugfs_simple_show, .data = sa_info},
 	{"topology", .show = xe_gt_debugfs_simple_show, .data = topology},
 	{"ggtt", .show = xe_gt_debugfs_simple_show, .data = ggtt},
 	{"register-save-restore", .show = xe_gt_debugfs_simple_show, .data = register_save_restore},
diff --git a/drivers/gpu/drm/xe/xe_tile_debugfs.c b/drivers/gpu/drm/xe/xe_tile_debugfs.c
new file mode 100644
index 000000000000..5523874cba7b
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_tile_debugfs.c
@@ -0,0 +1,135 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/debugfs.h>
+#include <drm/drm_debugfs.h>
+
+#include "xe_pm.h"
+#include "xe_sa.h"
+#include "xe_tile_debugfs.h"
+
+static struct xe_tile *node_to_tile(struct drm_info_node *node)
+{
+	return node->dent->d_parent->d_inode->i_private;
+}
+
+/**
+ * tile_debugfs_simple_show - A show callback for struct drm_info_list
+ * @m: the &seq_file
+ * @data: data used by the drm debugfs helpers
+ *
+ * This callback can be used in struct drm_info_list to describe debugfs
+ * files that are &xe_tile specific.
+ *
+ * It is assumed that those debugfs files will be created on directory entry
+ * which struct dentry d_inode->i_private points to &xe_tile.
+ *
+ *      /sys/kernel/debug/dri/0/
+ *      ├── tile0/		# tile = dentry->d_inode->i_private
+ *      │   │   ├── id		# tile = dentry->d_parent->d_inode->i_private
+ *
+ * This function assumes that &m->private will be set to the &struct
+ * drm_info_node corresponding to the instance of the info on a given &struct
+ * drm_minor (see struct drm_info_list.show for details).
+ *
+ * This function also assumes that struct drm_info_list.data will point to the
+ * function code that will actually print a file content::
+ *
+ *   int (*print)(struct xe_tile *, struct drm_printer *)
+ *
+ * Example::
+ *
+ *    int tile_id(struct xe_tile *tile, struct drm_printer *p)
+ *    {
+ *        drm_printf(p, "%u\n", tile->id);
+ *        return 0;
+ *    }
+ *
+ *    static const struct drm_info_list info[] = {
+ *        { name = "id", .show = tile_debugfs_simple_show, .data = tile_id },
+ *    };
+ *
+ *    dir = debugfs_create_dir("tile0", parent);
+ *    dir->d_inode->i_private = tile;
+ *    drm_debugfs_create_files(info, ARRAY_SIZE(info), dir, minor);
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+static int tile_debugfs_simple_show(struct seq_file *m, void *data)
+{
+	struct drm_printer p = drm_seq_file_printer(m);
+	struct drm_info_node *node = m->private;
+	struct xe_tile *tile = node_to_tile(node);
+	int (*print)(struct xe_tile *, struct drm_printer *) = node->info_ent->data;
+
+	return print(tile, &p);
+}
+
+/**
+ * tile_debugfs_show_with_rpm - A show callback for struct drm_info_list
+ * @m: the &seq_file
+ * @data: data used by the drm debugfs helpers
+ *
+ * Similar to tile_debugfs_simple_show() but implicitly takes a RPM ref.
+ *
+ * Return: 0 on success or a negative error code on failure.
+ */
+static int tile_debugfs_show_with_rpm(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct xe_tile *tile = node_to_tile(node);
+	struct xe_device *xe = tile_to_xe(tile);
+	int ret;
+
+	xe_pm_runtime_get(xe);
+	ret = tile_debugfs_simple_show(m, data);
+	xe_pm_runtime_put(xe);
+
+	return ret;
+}
+
+static int sa_info(struct xe_tile *tile, struct drm_printer *p)
+{
+	drm_suballoc_dump_debug_info(&tile->mem.kernel_bb_pool->base, p,
+				     xe_sa_manager_gpu_addr(tile->mem.kernel_bb_pool));
+
+	return 0;
+}
+
+/* only for debugfs files which can be safely used on the VF */
+static const struct drm_info_list vf_safe_debugfs_list[] = {
+	{ "sa_info", .show = tile_debugfs_show_with_rpm, .data = sa_info },
+};
+
+/**
+ * xe_tile_debugfs_register - Register tile's debugfs attributes
+ * @tile: the &xe_tile to register
+ *
+ * Create debugfs sub-directory with a name that includes a tile ID and
+ * then creates set of debugfs files (attributes) specific to this tile.
+ */
+void xe_tile_debugfs_register(struct xe_tile *tile)
+{
+	struct xe_device *xe = tile_to_xe(tile);
+	struct drm_minor *minor = xe->drm.primary;
+	struct dentry *root = minor->debugfs_root;
+	char name[8];
+
+	snprintf(name, sizeof(name), "tile%u", tile->id);
+	tile->debugfs = debugfs_create_dir(name, root);
+	if (IS_ERR(tile->debugfs))
+		return;
+
+	/*
+	 * Store the xe_tile pointer as private data of the tile/ directory
+	 * node so other tile specific attributes under that directory may
+	 * refer to it by looking at its parent node private data.
+	 */
+	tile->debugfs->d_inode->i_private = tile;
+
+	drm_debugfs_create_files(vf_safe_debugfs_list,
+				 ARRAY_SIZE(vf_safe_debugfs_list),
+				 tile->debugfs, minor);
+}
diff --git a/drivers/gpu/drm/xe/xe_tile_debugfs.h b/drivers/gpu/drm/xe/xe_tile_debugfs.h
new file mode 100644
index 000000000000..0e5f724de37f
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_tile_debugfs.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_TILE_DEBUGFS_H_
+#define _XE_TILE_DEBUGFS_H_
+
+struct xe_tile;
+
+void xe_tile_debugfs_register(struct xe_tile *tile);
+
+#endif

From 00cbd55557a98d2ed111967dc5a8e5e8ac5c8dc0 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:43 +0300
Subject: [PATCH 0902/1455] drm/i915/hdmi: use generic poll_timeout_us()
 instead of __wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
__wait_for(), which used to be 1, 2, 4, and 8 ms in this particular
case.

Use an arbitrary constant 4 ms sleep instead. The timeout remains,
varying between 20 ms and 3000 ms.

Cc: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/fc3a67f9de0049f415a276bba1c11a4df97e01d6.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_hdmi.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdmi.c b/drivers/gpu/drm/i915/display/intel_hdmi.c
index 0a8af7bb80ef..4ab7e2e3bfd4 100644
--- a/drivers/gpu/drm/i915/display/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/display/intel_hdmi.c
@@ -29,6 +29,7 @@
 #include <linux/delay.h>
 #include <linux/hdmi.h>
 #include <linux/i2c.h>
+#include <linux/iopoll.h>
 #include <linux/slab.h>
 #include <linux/string_helpers.h>
 
@@ -1690,11 +1691,10 @@ intel_hdmi_hdcp2_wait_for_msg(struct intel_digital_port *dig_port,
 	if (timeout < 0)
 		return timeout;
 
-	ret = __wait_for(ret = hdcp2_detect_msg_availability(dig_port,
-							     msg_id, &msg_ready,
-							     &msg_sz),
-			 !ret && msg_ready && msg_sz, timeout * 1000,
-			 1000, 5 * 1000);
+	ret = poll_timeout_us(ret = hdcp2_detect_msg_availability(dig_port, msg_id,
+								  &msg_ready, &msg_sz),
+			      !ret && msg_ready && msg_sz,
+			      4000, timeout * 1000, false);
 	if (ret)
 		drm_dbg_kms(display->drm,
 			    "msg_id: %d, ret: %d, timeout: %d\n",

From 6f1759df0f83e376b0b40017fa06f081963f5704 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:44 +0300
Subject: [PATCH 0903/1455] drm/i915/hdcp: use generic poll_timeout_us()
 instead of __wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
__wait_for(), which used to be 1, 2, 4, ... 64, and 128 ms in this
particular case.

Use an arbitrary 100 ms sleep instead. The timeout remains at 5000 ms.

Cc: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/bfc9f941ec1628830644f1419d606e3d085aaba0.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index e11cf26ed0bf..7af3dffa1eb0 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -11,6 +11,7 @@
 #include <linux/component.h>
 #include <linux/debugfs.h>
 #include <linux/i2c.h>
+#include <linux/iopoll.h>
 #include <linux/random.h>
 
 #include <drm/display/drm_hdcp_helper.h>
@@ -326,10 +327,9 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port,
 	bool ksv_ready;
 
 	/* Poll for ksv list ready (spec says max time allowed is 5s) */
-	ret = __wait_for(read_ret = shim->read_ksv_ready(dig_port,
-							 &ksv_ready),
-			 read_ret || ksv_ready, 5 * 1000 * 1000, 1000,
-			 100 * 1000);
+	ret = poll_timeout_us(read_ret = shim->read_ksv_ready(dig_port, &ksv_ready),
+			      read_ret || ksv_ready,
+			      100 * 1000, 5 * 1000 * 1000, false);
 	if (ret)
 		return ret;
 	if (read_ret)

From e54d34e4b309787fc0d4202a638143490c2fe319 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:45 +0300
Subject: [PATCH 0904/1455] drm/i915/hdcp: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 100 us sleep instead. The timeout remains at 1
ms.

While at it, use the last failing value for debug logging instead of
reading it again.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/2871a07337401c25ef3df44073c5e78fedc45e8e.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 7af3dffa1eb0..910e34bd73a9 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -817,6 +817,7 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 	enum port port = dig_port->base.port;
 	unsigned long r0_prime_gen_start;
 	int ret, i, tries = 2;
+	u32 val;
 	union {
 		u32 reg[2];
 		u8 shim[DRM_HDCP_AN_LEN];
@@ -905,8 +906,10 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 		       HDCP_CONF_AUTH_AND_ENC);
 
 	/* Wait for R0 ready */
-	if (wait_for(intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) &
-		     (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC), 1)) {
+	ret = poll_timeout_us(val = intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)),
+			      val & (HDCP_STATUS_R0_READY | HDCP_STATUS_ENC),
+			      100, 1000, false);
+	if (ret) {
 		drm_err(display->drm, "Timed out waiting for R0 ready\n");
 		return -ETIMEDOUT;
 	}
@@ -938,16 +941,16 @@ static int intel_hdcp_auth(struct intel_connector *connector)
 			       ri.reg);
 
 		/* Wait for Ri prime match */
-		if (!wait_for(intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)) &
-			      (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC), 1))
+		ret = poll_timeout_us(val = intel_de_read(display, HDCP_STATUS(display, cpu_transcoder, port)),
+				      val & (HDCP_STATUS_RI_MATCH | HDCP_STATUS_ENC),
+				      100, 1000, false);
+		if (!ret)
 			break;
 	}
 
 	if (i == tries) {
 		drm_dbg_kms(display->drm,
-			    "Timed out waiting for Ri prime match (%x)\n",
-			    intel_de_read(display,
-					  HDCP_STATUS(display, cpu_transcoder, port)));
+			    "Timed out waiting for Ri prime match (%x)\n", val);
 		return -ETIMEDOUT;
 	}
 

From 17d56494c1b8f3e61e4ee9a8aedf1e3733a83eaa Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:46 +0300
Subject: [PATCH 0905/1455] drm/i915/dsi: use generic poll_timeout_us() instead
 of wait_for_us()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The sleep and timeout remain the same as for wait_for_us().

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/02ebcd2864819b7eaf9cf455aa2b968980a2f671.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/icl_dsi.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/icl_dsi.c b/drivers/gpu/drm/i915/display/icl_dsi.c
index 44310984bb57..37faa8f19f6e 100644
--- a/drivers/gpu/drm/i915/display/icl_dsi.c
+++ b/drivers/gpu/drm/i915/display/icl_dsi.c
@@ -25,6 +25,8 @@
  *   Jani Nikula <jani.nikula@intel.com>
  */
 
+#include <linux/iopoll.h>
+
 #include <drm/display/drm_dsc_helper.h>
 #include <drm/drm_atomic_helper.h>
 #include <drm/drm_fixed.h>
@@ -72,8 +74,12 @@ static int payload_credits_available(struct intel_display *display,
 static bool wait_for_header_credits(struct intel_display *display,
 				    enum transcoder dsi_trans, int hdr_credit)
 {
-	if (wait_for_us(header_credits_available(display, dsi_trans) >=
-			hdr_credit, 100)) {
+	int ret, available;
+
+	ret = poll_timeout_us(available = header_credits_available(display, dsi_trans),
+			      available >= hdr_credit,
+			      10, 100, false);
+	if (ret) {
 		drm_err(display->drm, "DSI header credits not released\n");
 		return false;
 	}
@@ -84,8 +90,12 @@ static bool wait_for_header_credits(struct intel_display *display,
 static bool wait_for_payload_credits(struct intel_display *display,
 				     enum transcoder dsi_trans, int payld_credit)
 {
-	if (wait_for_us(payload_credits_available(display, dsi_trans) >=
-			payld_credit, 100)) {
+	int ret, available;
+
+	ret = poll_timeout_us(available = payload_credits_available(display, dsi_trans),
+			      available >= payld_credit,
+			      10, 100, false);
+	if (ret) {
 		drm_err(display->drm, "DSI payload credits not released\n");
 		return false;
 	}

From 0399a3368503c75043afbb680071dabc723aa9d7 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:47 +0300
Subject: [PATCH 0906/1455] drm/i915/dsi-pll: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 500 us sleep instead. The timeout remains at
20 ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/476fcc5aad9e2ddbf6d8c14bd5ff5cbf071c5dca.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/vlv_dsi_pll.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
index 83afe1315e96..f078b9cda96c 100644
--- a/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
+++ b/drivers/gpu/drm/i915/display/vlv_dsi_pll.c
@@ -25,12 +25,12 @@
  *	Yogesh Mohan Marimuthu <yogesh.mohan.marimuthu@intel.com>
  */
 
+#include <linux/iopoll.h>
 #include <linux/kernel.h>
 #include <linux/string_helpers.h>
 
 #include <drm/drm_print.h>
 
-#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_types.h"
 #include "intel_dsi.h"
@@ -214,6 +214,8 @@ void vlv_dsi_pll_enable(struct intel_encoder *encoder,
 			const struct intel_crtc_state *config)
 {
 	struct intel_display *display = to_intel_display(encoder);
+	u32 val;
+	int ret;
 
 	drm_dbg_kms(display->drm, "\n");
 
@@ -231,9 +233,10 @@ void vlv_dsi_pll_enable(struct intel_encoder *encoder,
 
 	vlv_cck_write(display->drm, CCK_REG_DSI_PLL_CONTROL, config->dsi_pll.ctrl);
 
-	if (wait_for(vlv_cck_read(display->drm, CCK_REG_DSI_PLL_CONTROL) &
-						DSI_PLL_LOCK, 20)) {
-
+	ret = poll_timeout_us(val = vlv_cck_read(display->drm, CCK_REG_DSI_PLL_CONTROL),
+			      val & DSI_PLL_LOCK,
+			      500, 20 * 1000, false);
+	if (ret) {
 		vlv_cck_put(display->drm);
 		drm_err(display->drm, "DSI PLL lock failed\n");
 		return;

From 11080a5ea877d7da163a93975559b9607f629d49 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:48 +0300
Subject: [PATCH 0907/1455] drm/i915/gmbus: use generic poll_timeout*() instead
 of wait_for*()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The "two tier" wait_for_us() + wait_for() combination appeared without
much explanation in commit 4e6c2d58ba86 ("drm/i915: Take forcewake once
for the entire GMBUS transaction"). Try to mimic roughly the same with
the generic helpers.

wait_for_us() with 10 us or shorter timeouts ends up in
_wait_for_atomic(). Thus use poll_timeout_us_atomic() for the first try,
with the same 2 us timeout and no sleep.

For the fallback, the functional change is losing the exponentially
growing sleep of wait_for(), which used to be 10, 20, 40, ..., 640, and
1280 us. Use an arbitrary constant 500 us sleep instead. The timeout
remains at 50 ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/564b397352c53a1116519fb2d53050c0426bc0dc.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_gmbus.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_gmbus.c b/drivers/gpu/drm/i915/display/intel_gmbus.c
index 063335053d13..358210adb8f8 100644
--- a/drivers/gpu/drm/i915/display/intel_gmbus.c
+++ b/drivers/gpu/drm/i915/display/intel_gmbus.c
@@ -30,13 +30,13 @@
 #include <linux/export.h>
 #include <linux/i2c-algo-bit.h>
 #include <linux/i2c.h>
+#include <linux/iopoll.h>
 
 #include <drm/display/drm_hdcp_helper.h>
 
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
-#include "i915_utils.h"
 #include "intel_de.h"
 #include "intel_display_regs.h"
 #include "intel_display_types.h"
@@ -415,11 +415,14 @@ static int gmbus_wait(struct intel_display *display, u32 status, u32 irq_en)
 	intel_de_write_fw(display, GMBUS4(display), irq_en);
 
 	status |= GMBUS_SATOER;
-	ret = wait_for_us((gmbus2 = intel_de_read_fw(display, GMBUS2(display))) & status,
-			  2);
+
+	ret = poll_timeout_us_atomic(gmbus2 = intel_de_read_fw(display, GMBUS2(display)),
+				     gmbus2 & status,
+				     0, 2, false);
 	if (ret)
-		ret = wait_for((gmbus2 = intel_de_read_fw(display, GMBUS2(display))) & status,
-			       50);
+		ret = poll_timeout_us(gmbus2 = intel_de_read_fw(display, GMBUS2(display)),
+				      gmbus2 & status,
+				      500, 50 * 1000, false);
 
 	intel_de_write_fw(display, GMBUS4(display), 0);
 	remove_wait_queue(&display->gmbus.wait_queue, &wait);

From 7f38db435cf703cc432810362d84ed8be59202ea Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:49 +0300
Subject: [PATCH 0908/1455] drm/i915/wm: use generic poll_timeout_us() instead
 of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 500 us sleep instead. The timeout remains at 3
ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/52c80860ea7b98e84f2386ed6cdd761f03190b1e.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/i9xx_wm.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c
index 591acce2a4b1..060aff765994 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.c
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.c
@@ -3,6 +3,8 @@
  * Copyright © 2023 Intel Corporation
  */
 
+#include <linux/iopoll.h>
+
 #include "soc/intel_dram.h"
 
 #include "i915_drv.h"
@@ -112,6 +114,7 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *dis
 static void chv_set_memory_dvfs(struct intel_display *display, bool enable)
 {
 	u32 val;
+	int ret;
 
 	vlv_punit_get(display->drm);
 
@@ -124,8 +127,10 @@ static void chv_set_memory_dvfs(struct intel_display *display, bool enable)
 	val |= FORCE_DDR_FREQ_REQ_ACK;
 	vlv_punit_write(display->drm, PUNIT_REG_DDR_SETUP2, val);
 
-	if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2) &
-		      FORCE_DDR_FREQ_REQ_ACK) == 0, 3))
+	ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2),
+			      (val & FORCE_DDR_FREQ_REQ_ACK) == 0,
+			      500, 3000, false);
+	if (ret)
 		drm_err(display->drm,
 			"timed out waiting for Punit DDR DVFS request\n");
 
@@ -3905,6 +3910,7 @@ static void vlv_wm_get_hw_state(struct intel_display *display)
 	struct vlv_wm_values *wm = &display->wm.vlv;
 	struct intel_crtc *crtc;
 	u32 val;
+	int ret;
 
 	vlv_read_wm_values(display, wm);
 
@@ -3931,8 +3937,10 @@ static void vlv_wm_get_hw_state(struct intel_display *display)
 		val |= FORCE_DDR_FREQ_REQ_ACK;
 		vlv_punit_write(display->drm, PUNIT_REG_DDR_SETUP2, val);
 
-		if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2) &
-			      FORCE_DDR_FREQ_REQ_ACK) == 0, 3)) {
+		ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DDR_SETUP2),
+				      (val & FORCE_DDR_FREQ_REQ_ACK) == 0,
+				      500, 3000, false);
+		if (ret) {
 			drm_dbg_kms(display->drm,
 				    "Punit not acking DDR DVFS request, "
 				    "assuming DDR DVFS is disabled\n");

From 032a3bd4d673a3efb2575914bd96ea60c9124993 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:50 +0300
Subject: [PATCH 0909/1455] drm/i915/cdclk: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 500 us sleep instead. The timeout remains at
50 ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/6d50031411d5517508867d4b595ce90a2b44073b.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_cdclk.c | 37 ++++++++++++----------
 1 file changed, 20 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_cdclk.c b/drivers/gpu/drm/i915/display/intel_cdclk.c
index d7ba3970e1e9..9725eebe5706 100644
--- a/drivers/gpu/drm/i915/display/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/display/intel_cdclk.c
@@ -22,6 +22,7 @@
  */
 
 #include <linux/debugfs.h>
+#include <linux/iopoll.h>
 #include <linux/time.h>
 
 #include <drm/drm_fixed.h>
@@ -673,6 +674,7 @@ static void vlv_set_cdclk(struct intel_display *display,
 	int cdclk = cdclk_config->cdclk;
 	u32 val, cmd = cdclk_config->voltage_level;
 	intel_wakeref_t wakeref;
+	int ret;
 
 	switch (cdclk) {
 	case 400000:
@@ -703,12 +705,12 @@ static void vlv_set_cdclk(struct intel_display *display,
 	val &= ~DSPFREQGUAR_MASK;
 	val |= (cmd << DSPFREQGUAR_SHIFT);
 	vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val);
-	if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) &
-		      DSPFREQSTAT_MASK) == (cmd << DSPFREQSTAT_SHIFT),
-		     50)) {
-		drm_err(display->drm,
-			"timed out waiting for CDclk change\n");
-	}
+
+	ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM),
+			      (val & DSPFREQSTAT_MASK) == (cmd << DSPFREQSTAT_SHIFT),
+			      500, 50 * 1000, false);
+	if (ret)
+		drm_err(display->drm, "timed out waiting for CDCLK change\n");
 
 	if (cdclk == 400000) {
 		u32 divider;
@@ -722,11 +724,11 @@ static void vlv_set_cdclk(struct intel_display *display,
 		val |= divider;
 		vlv_cck_write(display->drm, CCK_DISPLAY_CLOCK_CONTROL, val);
 
-		if (wait_for((vlv_cck_read(display->drm, CCK_DISPLAY_CLOCK_CONTROL) &
-			      CCK_FREQUENCY_STATUS) == (divider << CCK_FREQUENCY_STATUS_SHIFT),
-			     50))
-			drm_err(display->drm,
-				"timed out waiting for CDclk change\n");
+		ret = poll_timeout_us(val = vlv_cck_read(display->drm, CCK_DISPLAY_CLOCK_CONTROL),
+				      (val & CCK_FREQUENCY_STATUS) == (divider << CCK_FREQUENCY_STATUS_SHIFT),
+				      500, 50 * 1000, false);
+		if (ret)
+			drm_err(display->drm, "timed out waiting for CDCLK change\n");
 	}
 
 	/* adjust self-refresh exit latency value */
@@ -762,6 +764,7 @@ static void chv_set_cdclk(struct intel_display *display,
 	int cdclk = cdclk_config->cdclk;
 	u32 val, cmd = cdclk_config->voltage_level;
 	intel_wakeref_t wakeref;
+	int ret;
 
 	switch (cdclk) {
 	case 333333:
@@ -787,12 +790,12 @@ static void chv_set_cdclk(struct intel_display *display,
 	val &= ~DSPFREQGUAR_MASK_CHV;
 	val |= (cmd << DSPFREQGUAR_SHIFT_CHV);
 	vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, val);
-	if (wait_for((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) &
-		      DSPFREQSTAT_MASK_CHV) == (cmd << DSPFREQSTAT_SHIFT_CHV),
-		     50)) {
-		drm_err(display->drm,
-			"timed out waiting for CDclk change\n");
-	}
+
+	ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM),
+			      (val & DSPFREQSTAT_MASK_CHV) == (cmd << DSPFREQSTAT_SHIFT_CHV),
+			      500, 50 * 1000, false);
+	if (ret)
+		drm_err(display->drm, "timed out waiting for CDCLK change\n");
 
 	vlv_punit_put(display->drm);
 

From c698a48e9f8019c01eab83a962289ad4e81db938 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:51 +0300
Subject: [PATCH 0910/1455] drm/i915/power: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 100 us sleep instead. The timeout remains at 1
ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/900680516b047ae32e3298b5cdbcede0393e0466.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_power.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index d1d3b6f89e2a..7340d5a71673 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -3,6 +3,7 @@
  * Copyright © 2019 Intel Corporation
  */
 
+#include <linux/iopoll.h>
 #include <linux/string_helpers.h>
 
 #include "soc/intel_dram.h"
@@ -1310,8 +1311,10 @@ static void hsw_disable_lcpll(struct intel_display *display,
 	hsw_write_dcomp(display, val);
 	ndelay(100);
 
-	if (wait_for((hsw_read_dcomp(display) &
-		      D_COMP_RCOMP_IN_PROGRESS) == 0, 1))
+	ret = poll_timeout_us(val = hsw_read_dcomp(display),
+			      (val & D_COMP_RCOMP_IN_PROGRESS) == 0,
+			      100, 1000, false);
+	if (ret)
 		drm_err(display->drm, "D_COMP RCOMP still in progress\n");
 
 	if (allow_power_down) {

From a8eb4cdcedd836b22459b9f6e68b60001bd04bac Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:52 +0300
Subject: [PATCH 0911/1455] drm/i915/power-well: use generic poll_timeout_us()
 instead of wait_for() for DKL PHY
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 100 us sleep instead. The timeout remains at 1
ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/663c9edf4a98b09121d7200f8d734ebc829da85b.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../gpu/drm/i915/display/intel_display_power_well.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index 0c48d9cf3528..c7a1fcb9d6ad 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -3,6 +3,8 @@
  * Copyright © 2022 Intel Corporation
  */
 
+#include <linux/iopoll.h>
+
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
@@ -527,6 +529,8 @@ icl_tc_phy_aux_power_well_enable(struct intel_display *display,
 	const struct i915_power_well_regs *regs = power_well->desc->ops->regs;
 	bool is_tbt = power_well->desc->is_tc_tbt;
 	bool timeout_expected;
+	u32 val;
+	int ret;
 
 	icl_tc_port_assert_ref_held(display, power_well, dig_port);
 
@@ -553,10 +557,11 @@ icl_tc_phy_aux_power_well_enable(struct intel_display *display,
 
 		tc_port = TGL_AUX_PW_TO_TC_PORT(i915_power_well_instance(power_well)->hsw.idx);
 
-		if (wait_for(intel_dkl_phy_read(display, DKL_CMN_UC_DW_27(tc_port)) &
-			     DKL_CMN_UC_DW27_UC_HEALTH, 1))
-			drm_warn(display->drm,
-				 "Timeout waiting TC uC health\n");
+		ret = poll_timeout_us(val = intel_dkl_phy_read(display, DKL_CMN_UC_DW_27(tc_port)),
+				      val & DKL_CMN_UC_DW27_UC_HEALTH,
+				      100, 1000, false);
+		if (ret)
+			drm_warn(display->drm, "Timeout waiting TC uC health\n");
 	}
 }
 

From e38e3c50a14c76bf130a688bffd130797b09dfe6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:53 +0300
Subject: [PATCH 0912/1455] drm/i915/power-well: use generic poll_timeout_us()
 instead of wait_for() for VLV/CHV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 500 us sleep instead. The timeout remains at
100 ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/c644b7b5611a3c047ea5d3d52acd91830b2fa6b4.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../i915/display/intel_display_power_well.c   | 29 ++++++++++---------
 1 file changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c
index c7a1fcb9d6ad..5e88b930f5aa 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power_well.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c
@@ -8,7 +8,6 @@
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
-#include "i915_utils.h"
 #include "intel_backlight_regs.h"
 #include "intel_combo_phy.h"
 #include "intel_combo_phy_regs.h"
@@ -1127,6 +1126,8 @@ static void vlv_set_power_well(struct intel_display *display,
 	u32 mask;
 	u32 state;
 	u32 ctrl;
+	u32 val;
+	int ret;
 
 	mask = PUNIT_PWRGT_MASK(pw_idx);
 	state = enable ? PUNIT_PWRGT_PWR_ON(pw_idx) :
@@ -1134,10 +1135,8 @@ static void vlv_set_power_well(struct intel_display *display,
 
 	vlv_punit_get(display->drm);
 
-#define COND \
-	((vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS) & mask) == state)
-
-	if (COND)
+	val = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS);
+	if ((val & mask) == state)
 		goto out;
 
 	ctrl = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL);
@@ -1145,14 +1144,15 @@ static void vlv_set_power_well(struct intel_display *display,
 	ctrl |= state;
 	vlv_punit_write(display->drm, PUNIT_REG_PWRGT_CTRL, ctrl);
 
-	if (wait_for(COND, 100))
+	ret = poll_timeout_us(val = vlv_punit_read(display->drm, PUNIT_REG_PWRGT_STATUS),
+			      (val & mask) == state,
+			      500, 100 * 1000, false);
+	if (ret)
 		drm_err(display->drm,
 			"timeout setting power well state %08x (%08x)\n",
 			state,
 			vlv_punit_read(display->drm, PUNIT_REG_PWRGT_CTRL));
 
-#undef COND
-
 out:
 	vlv_punit_put(display->drm);
 }
@@ -1716,23 +1716,24 @@ static void chv_set_pipe_power_well(struct intel_display *display,
 	enum pipe pipe = PIPE_A;
 	u32 state;
 	u32 ctrl;
+	int ret;
 
 	state = enable ? DP_SSS_PWR_ON(pipe) : DP_SSS_PWR_GATE(pipe);
 
 	vlv_punit_get(display->drm);
 
-#define COND \
-	((vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM) & DP_SSS_MASK(pipe)) == state)
-
-	if (COND)
+	ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM);
+	if ((ctrl & DP_SSS_MASK(pipe)) == state)
 		goto out;
 
-	ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM);
 	ctrl &= ~DP_SSC_MASK(pipe);
 	ctrl |= enable ? DP_SSC_PWR_ON(pipe) : DP_SSC_PWR_GATE(pipe);
 	vlv_punit_write(display->drm, PUNIT_REG_DSPSSPM, ctrl);
 
-	if (wait_for(COND, 100))
+	ret = poll_timeout_us(ctrl = vlv_punit_read(display->drm, PUNIT_REG_DSPSSPM),
+			      (ctrl & DP_SSS_MASK(pipe)) == state,
+			      500, 100 * 1000, false);
+	if (ret)
 		drm_err(display->drm,
 			"timeout setting power well state %08x (%08x)\n",
 			state,

From 39b555b3eab503a47396d5ef1dcc7e22de2c2e94 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:54 +0300
Subject: [PATCH 0913/1455] drm/i915/dp: use generic poll_timeout_us() instead
 of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 1 ms sleep instead. The timeouts remain, being
500 ms or 1000 ms depending on the case.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/83d3417d4e5af1db13eb4c6eaa48b5f9c12caeb4.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 958e48859beb..2eab591a8ef5 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -3879,10 +3879,11 @@ static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp)
 	if (ret < 0)
 		return ret;
 	/* Wait for PCON to be FRL Ready */
-	wait_for(is_active = drm_dp_pcon_is_frl_ready(&intel_dp->aux) == true, TIMEOUT_FRL_READY_MS);
-
-	if (!is_active)
-		return -ETIMEDOUT;
+	ret = poll_timeout_us(is_active = drm_dp_pcon_is_frl_ready(&intel_dp->aux),
+			      is_active,
+			      1000, TIMEOUT_FRL_READY_MS * 1000, false);
+	if (ret)
+		return ret;
 
 	ret = drm_dp_pcon_frl_configure_1(&intel_dp->aux, max_frl_bw,
 					  DP_PCON_ENABLE_SEQUENTIAL_LINK);
@@ -3899,12 +3900,11 @@ static int intel_dp_pcon_start_frl_training(struct intel_dp *intel_dp)
 	 * Wait for FRL to be completed
 	 * Check if the HDMI Link is up and active.
 	 */
-	wait_for(is_active =
-		 intel_dp_pcon_is_frl_trained(intel_dp, max_frl_bw_mask, &frl_trained_mask),
-		 TIMEOUT_HDMI_LINK_ACTIVE_MS);
-
-	if (!is_active)
-		return -ETIMEDOUT;
+	ret = poll_timeout_us(is_active = intel_dp_pcon_is_frl_trained(intel_dp, max_frl_bw_mask, &frl_trained_mask),
+			      is_active,
+			      1000, TIMEOUT_HDMI_LINK_ACTIVE_MS * 1000, false);
+	if (ret)
+		return ret;
 
 frl_trained:
 	drm_dbg(display->drm, "FRL_TRAINED_MASK = %u\n", frl_trained_mask);

From 476721f28ee8f97abc89103c89279766167ef95a Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:55 +0300
Subject: [PATCH 0914/1455] drm/i915/dp: use generic poll_timeout_us() instead
 of wait_for() in link training
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 500 us sleep instead. The timeout remains at
500 ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/29ab4738758fe844dc1323c4a59d5d6bdcf87308.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../drm/i915/display/intel_dp_link_training.c   | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_link_training.c b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
index 9d14b4e59545..27f3716bdc1f 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_link_training.c
@@ -22,6 +22,7 @@
  */
 
 #include <linux/debugfs.h>
+#include <linux/iopoll.h>
 
 #include <drm/display/drm_dp_helper.h>
 #include <drm/drm_print.h>
@@ -1135,15 +1136,19 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp,
 {
 	struct intel_display *display = to_intel_display(intel_dp);
 	struct intel_encoder *encoder = &dp_to_dig_port(intel_dp)->base;
+	int ret;
 
 	intel_dp->link.active = true;
 
 	intel_dp_program_link_training_pattern(intel_dp, crtc_state, DP_PHY_DPRX,
 					       DP_TRAINING_PATTERN_DISABLE);
 
-	if (intel_dp_is_uhbr(crtc_state) &&
-	    wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) {
-		lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n");
+	if (intel_dp_is_uhbr(crtc_state)) {
+		ret = poll_timeout_us(ret = intel_dp_128b132b_intra_hop(intel_dp, crtc_state),
+				      ret == 0,
+				      500, 500 * 1000, false);
+		if (ret)
+			lt_dbg(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clearing\n");
 	}
 
 	intel_hpd_unblock(encoder);
@@ -1581,8 +1586,12 @@ intel_dp_128b132b_link_train(struct intel_dp *intel_dp,
 			     int lttpr_count)
 {
 	bool passed = false;
+	int ret;
 
-	if (wait_for(intel_dp_128b132b_intra_hop(intel_dp, crtc_state) == 0, 500)) {
+	ret = poll_timeout_us(ret = intel_dp_128b132b_intra_hop(intel_dp, crtc_state),
+			      ret == 0,
+			      500, 500 * 1000, false);
+	if (ret) {
 		lt_err(intel_dp, DP_PHY_DPRX, "128b/132b intra-hop not clear\n");
 		goto out;
 	}

From 46013820f40220ca4ddcf4892401affe41d69747 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:56 +0300
Subject: [PATCH 0915/1455] drm/i915/vblank: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 500 us sleep instead. The timeout remains at
100 ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/162dff5862d3213304491a6d2eb31a57346b523e.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_vblank.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_vblank.c b/drivers/gpu/drm/i915/display/intel_vblank.c
index 46d6db5fed11..c15234c1d96e 100644
--- a/drivers/gpu/drm/i915/display/intel_vblank.c
+++ b/drivers/gpu/drm/i915/display/intel_vblank.c
@@ -3,6 +3,8 @@
  * Copyright © 2022-2023 Intel Corporation
  */
 
+#include <linux/iopoll.h>
+
 #include <drm/drm_vblank.h>
 
 #include "i915_drv.h"
@@ -493,9 +495,14 @@ static void wait_for_pipe_scanline_moving(struct intel_crtc *crtc, bool state)
 {
 	struct intel_display *display = to_intel_display(crtc);
 	enum pipe pipe = crtc->pipe;
+	bool is_moving;
+	int ret;
 
 	/* Wait for the display line to settle/start moving */
-	if (wait_for(pipe_scanline_is_moving(display, pipe) == state, 100))
+	ret = poll_timeout_us(is_moving = pipe_scanline_is_moving(display, pipe),
+			      is_moving == state,
+			      500, 100 * 1000, false);
+	if (ret)
 		drm_err(display->drm,
 			"pipe %c scanline %s wait timed out\n",
 			pipe_name(pipe), str_on_off(state));

From 7987b7643312f1b63f70da433b23b32e2a85160c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:57 +0300
Subject: [PATCH 0916/1455] drm/i915/tc: use generic poll_timeout_us() instead
 of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 200 us sleep for the 5 ms timeout, and 1000 us
sleep for the 500 ms timeout. The timeouts remain the same.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/50cd06b61210f541d5bb52a36af2d8bf059dd3a1.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_tc.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_tc.c b/drivers/gpu/drm/i915/display/intel_tc.c
index 23745fc99d35..c4a5601c5107 100644
--- a/drivers/gpu/drm/i915/display/intel_tc.c
+++ b/drivers/gpu/drm/i915/display/intel_tc.c
@@ -3,6 +3,8 @@
  * Copyright © 2019 Intel Corporation
  */
 
+#include <linux/iopoll.h>
+
 #include <drm/drm_print.h>
 
 #include "i915_reg.h"
@@ -1050,8 +1052,13 @@ static bool
 xelpdp_tc_phy_wait_for_tcss_power(struct intel_tc_port *tc, bool enabled)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
+	bool is_enabled;
+	int ret;
 
-	if (wait_for(xelpdp_tc_phy_tcss_power_is_enabled(tc) == enabled, 5)) {
+	ret = poll_timeout_us(is_enabled = xelpdp_tc_phy_tcss_power_is_enabled(tc),
+			      is_enabled == enabled,
+			      200, 5000, false);
+	if (ret) {
 		drm_dbg_kms(display->drm,
 			    "Port %s: timeout waiting for TCSS power to get %s\n",
 			    str_enabled_disabled(enabled),
@@ -1332,8 +1339,13 @@ static bool tc_phy_is_connected(struct intel_tc_port *tc,
 static bool tc_phy_wait_for_ready(struct intel_tc_port *tc)
 {
 	struct intel_display *display = to_intel_display(tc->dig_port);
+	bool is_ready;
+	int ret;
 
-	if (wait_for(tc_phy_is_ready(tc), 500)) {
+	ret = poll_timeout_us(is_ready = tc_phy_is_ready(tc),
+			      is_ready,
+			      1000, 500 * 1000, false);
+	if (ret) {
 		drm_err(display->drm, "Port %s: timeout waiting for PHY ready\n",
 			tc->port_name);
 

From bcd23d8ee01c2abcea782aaf40c74d54a3fdab82 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:58 +0300
Subject: [PATCH 0917/1455] drm/i915/dsb: use generic poll_timeout_us() instead
 of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 100 us sleep instead. The timeout remains at 1
ms.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/979eae02af1184b3756746ace61379dd1947a79b.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dsb.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dsb.c b/drivers/gpu/drm/i915/display/intel_dsb.c
index 53d8ae3a70e9..dee44d45b668 100644
--- a/drivers/gpu/drm/i915/display/intel_dsb.c
+++ b/drivers/gpu/drm/i915/display/intel_dsb.c
@@ -4,10 +4,11 @@
  *
  */
 
+#include <linux/iopoll.h>
+
 #include <drm/drm_print.h>
 #include <drm/drm_vblank.h>
 
-#include "i915_utils.h"
 #include "intel_crtc.h"
 #include "intel_de.h"
 #include "intel_display_regs.h"
@@ -871,8 +872,13 @@ void intel_dsb_wait(struct intel_dsb *dsb)
 	struct intel_crtc *crtc = dsb->crtc;
 	struct intel_display *display = to_intel_display(crtc->base.dev);
 	enum pipe pipe = crtc->pipe;
+	bool is_busy;
+	int ret;
 
-	if (wait_for(!is_dsb_busy(display, pipe, dsb->id), 1)) {
+	ret = poll_timeout_us(is_busy = is_dsb_busy(display, pipe, dsb->id),
+			      !is_busy,
+			      100, 1000, false);
+	if (ret) {
 		u32 offset = intel_dsb_buffer_ggtt_offset(&dsb->dsb_buf);
 
 		intel_de_write_fw(display, DSB_CTRL(pipe, dsb->id),

From 19b31edb0a8da0d0e9406a7ced6bda3b68577d9e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:20:59 +0300
Subject: [PATCH 0918/1455] drm/i915/lspcon: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 5 ms sleep instead. The timeouts remain, being
400 ms or 800 ms, depending on the case.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/4065fa96c0ef6afd51a384f365761d2ca802256b.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_lspcon.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_lspcon.c b/drivers/gpu/drm/i915/display/intel_lspcon.c
index abc4b562083d..d56026c4efdd 100644
--- a/drivers/gpu/drm/i915/display/intel_lspcon.c
+++ b/drivers/gpu/drm/i915/display/intel_lspcon.c
@@ -23,6 +23,8 @@
  *
  */
 
+#include <linux/iopoll.h>
+
 #include <drm/display/drm_dp_dual_mode_helper.h>
 #include <drm/display/drm_hdmi_helper.h>
 #include <drm/drm_atomic_helper.h>
@@ -181,6 +183,8 @@ static enum drm_lspcon_mode lspcon_wait_mode(struct intel_lspcon *lspcon,
 	struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon);
 	struct intel_display *display = to_intel_display(intel_dp);
 	enum drm_lspcon_mode current_mode;
+	int timeout_us;
+	int ret;
 
 	current_mode = lspcon_get_current_mode(lspcon);
 	if (current_mode == mode)
@@ -189,9 +193,12 @@ static enum drm_lspcon_mode lspcon_wait_mode(struct intel_lspcon *lspcon,
 	drm_dbg_kms(display->drm, "Waiting for LSPCON mode %s to settle\n",
 		    lspcon_mode_name(mode));
 
-	wait_for((current_mode = lspcon_get_current_mode(lspcon)) == mode,
-		 lspcon_get_mode_settle_timeout(lspcon));
-	if (current_mode != mode)
+	timeout_us = lspcon_get_mode_settle_timeout(lspcon) * 1000;
+
+	ret = poll_timeout_us(current_mode = lspcon_get_current_mode(lspcon),
+			      current_mode == mode,
+			      5000, timeout_us, false);
+	if (ret)
 		drm_err(display->drm, "LSPCON mode hasn't settled\n");
 
 out:

From f1415900089f4cfbaa8eb992fb9186e5840f8c34 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:21:00 +0300
Subject: [PATCH 0919/1455] drm/i915/opregion: use generic poll_timeout_us()
 instead of wait_for()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Prefer generic poll helpers over i915 custom helpers.

The functional change is losing the exponentially growing sleep of
wait_for(), which used to be 10, 20, 40, ..., 640, and 1280 us.

Use an arbitrary constant 1 ms sleep instead. The timeout remains, being
opregion defined, 50 ms by default, and 1500 ms at most.

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/63db3a1e1db9e55a18ed322c55f2dffe511a10bb.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_opregion.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_opregion.c b/drivers/gpu/drm/i915/display/intel_opregion.c
index 380314a3b4d9..cbc220310813 100644
--- a/drivers/gpu/drm/i915/display/intel_opregion.c
+++ b/drivers/gpu/drm/i915/display/intel_opregion.c
@@ -28,13 +28,13 @@
 #include <linux/acpi.h>
 #include <linux/debugfs.h>
 #include <linux/dmi.h>
+#include <linux/iopoll.h>
 #include <acpi/video.h>
 
 #include <drm/drm_edid.h>
 #include <drm/drm_file.h>
 #include <drm/drm_print.h>
 
-#include "i915_utils.h"
 #include "intel_acpi.h"
 #include "intel_backlight.h"
 #include "intel_display_core.h"
@@ -357,10 +357,12 @@ static int swsci(struct intel_display *display,
 	pci_write_config_word(pdev, SWSCI, swsci_val);
 
 	/* Poll for the result. */
-#define C (((scic = swsci->scic) & SWSCI_SCIC_INDICATOR) == 0)
-	if (wait_for(C, dslp)) {
+	ret = poll_timeout_us(scic = swsci->scic,
+			      (scic & SWSCI_SCIC_INDICATOR) == 0,
+			      1000, dslp * 1000, false);
+	if (ret) {
 		drm_dbg(display->drm, "SWSCI request timed out\n");
-		return -ETIMEDOUT;
+		return ret;
 	}
 
 	scic = (scic & SWSCI_SCIC_EXIT_STATUS_MASK) >>

From 54846c2ea78e8a1ce61d2ef604f56f25783d855d Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:21:01 +0300
Subject: [PATCH 0920/1455] drm/i915/ddi: prefer poll_timeout_us() over
 readx_poll_timeout()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unify on using poll_timeout_us() throughout instead of mixing with
readx_poll_timeout(). While the latter can be ever so slightly simpler,
they are both complicated enough that it's better to unify on one
approach only.

While at it, better separate the handling of error returns from
drm_dp_dpcd_readb() and the actual status byte. This is best achieved by
inlining the read_fec_detected_status() function, and switching to
drm_dp_dpcd_read_byte().

v2: Use drm_dp_dpcd_read_byte() (Imre)

Cc: Imre Deak <imre.deak@intel.com>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/63b10a36c7ab545c640b24bc8fc007ce2ea74623.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_ddi.c | 32 ++++++++----------------
 1 file changed, 11 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 4e4ea3a0ff83..46017091bb0b 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -2340,34 +2340,24 @@ static void intel_dp_sink_set_fec_ready(struct intel_dp *intel_dp,
 		drm_dbg_kms(display->drm, "Failed to clear FEC detected flags\n");
 }
 
-static int read_fec_detected_status(struct drm_dp_aux *aux)
-{
-	int ret;
-	u8 status;
-
-	ret = drm_dp_dpcd_readb(aux, DP_FEC_STATUS, &status);
-	if (ret < 0)
-		return ret;
-
-	return status;
-}
-
 static int wait_for_fec_detected(struct drm_dp_aux *aux, bool enabled)
 {
 	struct intel_display *display = to_intel_display(aux->drm_dev);
 	int mask = enabled ? DP_FEC_DECODE_EN_DETECTED : DP_FEC_DECODE_DIS_DETECTED;
-	int status;
-	int err;
+	u8 status = 0;
+	int ret, err;
 
-	err = readx_poll_timeout(read_fec_detected_status, aux, status,
-				 status & mask || status < 0,
-				 10000, 200000);
+	ret = poll_timeout_us(err = drm_dp_dpcd_read_byte(aux, DP_FEC_STATUS, &status),
+			      err || (status & mask),
+			      10 * 1000, 200 * 1000, false);
 
-	if (err || status < 0) {
+	/* Either can be non-zero, but not both */
+	ret = ret ?: err;
+	if (ret) {
 		drm_dbg_kms(display->drm,
-			    "Failed waiting for FEC %s to get detected: %d (status %d)\n",
-			    str_enabled_disabled(enabled), err, status);
-		return err ? err : status;
+			    "Failed waiting for FEC %s to get detected: %d (status 0x%02x)\n",
+			    str_enabled_disabled(enabled), ret, status);
+		return ret;
 	}
 
 	return 0;

From f7e60f1570b4de044055e4269c30be826ef16a9b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 28 Aug 2025 15:21:02 +0300
Subject: [PATCH 0921/1455] drm/i915/pps: prefer poll_timeout_us() over
 read_poll_timeout()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Unify on using poll_timeout_us() throughout instead of mixing with
readx_poll_timeout().

Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/260fd455df743453f123d96fc01e7ca96a36f0fa.1756383233.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_pps.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_pps.c b/drivers/gpu/drm/i915/display/intel_pps.c
index b84eb43bd2d0..327e0de86f1e 100644
--- a/drivers/gpu/drm/i915/display/intel_pps.c
+++ b/drivers/gpu/drm/i915/display/intel_pps.c
@@ -627,11 +627,9 @@ static void wait_panel_status(struct intel_dp *intel_dp,
 		    intel_de_read(display, pp_stat_reg),
 		    intel_de_read(display, pp_ctrl_reg));
 
-	ret = read_poll_timeout(intel_de_read, val,
-				(val & mask) == value,
-				10 * 1000, 5000 * 1000, true,
-				display, pp_stat_reg);
-
+	ret = poll_timeout_us(val = intel_de_read(display, pp_stat_reg),
+			      (val & mask) == value,
+			      10 * 1000, 5000 * 1000, true);
 	if (ret) {
 		drm_err(display->drm,
 			"[ENCODER:%d:%s] %s panel status timeout: PP_STATUS: 0x%08x PP_CONTROL: 0x%08x\n",

From 2f509fe6a42cda845890273fe759fb7ba9edad97 Mon Sep 17 00:00:00 2001
From: Lizhi Hou <lizhi.hou@amd.com>
Date: Tue, 2 Sep 2025 22:34:02 -0700
Subject: [PATCH 0922/1455] accel/amdxdna: Add ioctl
 DRM_IOCTL_AMDXDNA_GET_ARRAY

Add interface for applications to get information array. The application
provides a buffer pointer along with information type, maximum number of
entries and maximum size of each entry. The buffer may also contain match
conditions based on the information type. After the ioctl completes, the
actual number of entries and entry size are returned. (see [1], used by
driver runtime library)

[1] https://github.com/amd/xdna-driver/blob/main/src/shim/host/platform_host.cpp#L337

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250903053402.2103196-1-lizhi.hou@amd.com
---
 drivers/accel/amdxdna/aie2_pci.c        | 112 +++++++++++++++++++-----
 drivers/accel/amdxdna/amdxdna_pci_drv.c |  27 ++++++
 drivers/accel/amdxdna/amdxdna_pci_drv.h |   1 +
 include/uapi/drm/amdxdna_accel.h        | 111 +++++++++++++++++++++++
 4 files changed, 227 insertions(+), 24 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_pci.c b/drivers/accel/amdxdna/aie2_pci.c
index 7a3449541107..87c425e3d2b9 100644
--- a/drivers/accel/amdxdna/aie2_pci.c
+++ b/drivers/accel/amdxdna/aie2_pci.c
@@ -785,11 +785,12 @@ static int aie2_get_clock_metadata(struct amdxdna_client *client,
 
 static int aie2_hwctx_status_cb(struct amdxdna_hwctx *hwctx, void *arg)
 {
-	struct amdxdna_drm_query_hwctx *tmp __free(kfree) = NULL;
-	struct amdxdna_drm_get_info *get_info_args = arg;
-	struct amdxdna_drm_query_hwctx __user *buf;
+	struct amdxdna_drm_hwctx_entry *tmp __free(kfree) = NULL;
+	struct amdxdna_drm_get_array *array_args = arg;
+	struct amdxdna_drm_hwctx_entry __user *buf;
+	u32 size;
 
-	if (get_info_args->buffer_size < sizeof(*tmp))
+	if (!array_args->num_element)
 		return -EINVAL;
 
 	tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
@@ -802,14 +803,23 @@ static int aie2_hwctx_status_cb(struct amdxdna_hwctx *hwctx, void *arg)
 	tmp->num_col = hwctx->num_col;
 	tmp->command_submissions = hwctx->priv->seq;
 	tmp->command_completions = hwctx->priv->completed;
+	tmp->pasid = hwctx->client->pasid;
+	tmp->priority = hwctx->qos.priority;
+	tmp->gops = hwctx->qos.gops;
+	tmp->fps = hwctx->qos.fps;
+	tmp->dma_bandwidth = hwctx->qos.dma_bandwidth;
+	tmp->latency = hwctx->qos.latency;
+	tmp->frame_exec_time = hwctx->qos.frame_exec_time;
+	tmp->state = AMDXDNA_HWCTX_STATE_ACTIVE;
 
-	buf = u64_to_user_ptr(get_info_args->buffer);
+	buf = u64_to_user_ptr(array_args->buffer);
+	size = min(sizeof(*tmp), array_args->element_size);
 
-	if (copy_to_user(buf, tmp, sizeof(*tmp)))
+	if (copy_to_user(buf, tmp, size))
 		return -EFAULT;
 
-	get_info_args->buffer += sizeof(*tmp);
-	get_info_args->buffer_size -= sizeof(*tmp);
+	array_args->buffer += size;
+	array_args->num_element--;
 
 	return 0;
 }
@@ -817,23 +827,24 @@ static int aie2_hwctx_status_cb(struct amdxdna_hwctx *hwctx, void *arg)
 static int aie2_get_hwctx_status(struct amdxdna_client *client,
 				 struct amdxdna_drm_get_info *args)
 {
+	struct amdxdna_drm_get_array array_args;
 	struct amdxdna_dev *xdna = client->xdna;
-	struct amdxdna_drm_get_info info_args;
 	struct amdxdna_client *tmp_client;
 	int ret;
 
 	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
 
-	info_args.buffer = args->buffer;
-	info_args.buffer_size = args->buffer_size;
-
+	array_args.element_size = sizeof(struct amdxdna_drm_query_hwctx);
+	array_args.buffer = args->buffer;
+	array_args.num_element = args->buffer_size / array_args.element_size;
 	list_for_each_entry(tmp_client, &xdna->client_list, node) {
-		ret = amdxdna_hwctx_walk(tmp_client, &info_args, aie2_hwctx_status_cb);
+		ret = amdxdna_hwctx_walk(tmp_client, &array_args,
+					 aie2_hwctx_status_cb);
 		if (ret)
 			break;
 	}
 
-	args->buffer_size = (u32)(info_args.buffer - args->buffer);
+	args->buffer_size -= (u32)(array_args.buffer - args->buffer);
 	return ret;
 }
 
@@ -877,6 +888,58 @@ static int aie2_get_info(struct amdxdna_client *client, struct amdxdna_drm_get_i
 	return ret;
 }
 
+static int aie2_query_ctx_status_array(struct amdxdna_client *client,
+				       struct amdxdna_drm_get_array *args)
+{
+	struct amdxdna_drm_get_array array_args;
+	struct amdxdna_dev *xdna = client->xdna;
+	struct amdxdna_client *tmp_client;
+	int ret;
+
+	drm_WARN_ON(&xdna->ddev, !mutex_is_locked(&xdna->dev_lock));
+
+	array_args.element_size = min(args->element_size,
+				      sizeof(struct amdxdna_drm_hwctx_entry));
+	array_args.buffer = args->buffer;
+	array_args.num_element = args->num_element * args->element_size /
+				array_args.element_size;
+	list_for_each_entry(tmp_client, &xdna->client_list, node) {
+		ret = amdxdna_hwctx_walk(tmp_client, &array_args,
+					 aie2_hwctx_status_cb);
+		if (ret)
+			break;
+	}
+
+	args->element_size = array_args.element_size;
+	args->num_element = (u32)((array_args.buffer - args->buffer) /
+				  args->element_size);
+
+	return ret;
+}
+
+static int aie2_get_array(struct amdxdna_client *client,
+			  struct amdxdna_drm_get_array *args)
+{
+	struct amdxdna_dev *xdna = client->xdna;
+	int ret, idx;
+
+	if (!drm_dev_enter(&xdna->ddev, &idx))
+		return -ENODEV;
+
+	switch (args->param) {
+	case DRM_AMDXDNA_HW_CONTEXT_ALL:
+		ret = aie2_query_ctx_status_array(client, args);
+		break;
+	default:
+		XDNA_ERR(xdna, "Not supported request parameter %u", args->param);
+		ret = -EOPNOTSUPP;
+	}
+	XDNA_DBG(xdna, "Got param %d", args->param);
+
+	drm_dev_exit(idx);
+	return ret;
+}
+
 static int aie2_set_power_mode(struct amdxdna_client *client,
 			       struct amdxdna_drm_set_state *args)
 {
@@ -926,15 +989,16 @@ static int aie2_set_state(struct amdxdna_client *client,
 }
 
 const struct amdxdna_dev_ops aie2_ops = {
-	.init           = aie2_init,
-	.fini           = aie2_fini,
-	.resume         = aie2_hw_resume,
-	.suspend        = aie2_hw_suspend,
-	.get_aie_info   = aie2_get_info,
-	.set_aie_state	= aie2_set_state,
-	.hwctx_init     = aie2_hwctx_init,
-	.hwctx_fini     = aie2_hwctx_fini,
-	.hwctx_config   = aie2_hwctx_config,
-	.cmd_submit     = aie2_cmd_submit,
+	.init = aie2_init,
+	.fini = aie2_fini,
+	.resume = aie2_hw_resume,
+	.suspend = aie2_hw_suspend,
+	.get_aie_info = aie2_get_info,
+	.set_aie_state = aie2_set_state,
+	.hwctx_init = aie2_hwctx_init,
+	.hwctx_fini = aie2_hwctx_fini,
+	.hwctx_config = aie2_hwctx_config,
+	.cmd_submit = aie2_cmd_submit,
 	.hmm_invalidate = aie2_hmm_invalidate,
+	.get_array = aie2_get_array,
 };
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.c b/drivers/accel/amdxdna/amdxdna_pci_drv.c
index 8ef5e4f27f5e..569cd703729d 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.c
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.c
@@ -26,6 +26,13 @@ MODULE_FIRMWARE("amdnpu/17f0_10/npu.sbin");
 MODULE_FIRMWARE("amdnpu/17f0_11/npu.sbin");
 MODULE_FIRMWARE("amdnpu/17f0_20/npu.sbin");
 
+/*
+ * 0.0: Initial version
+ * 0.1: Support getting all hardware contexts by DRM_IOCTL_AMDXDNA_GET_ARRAY
+ */
+#define AMDXDNA_DRIVER_MAJOR		0
+#define AMDXDNA_DRIVER_MINOR		1
+
 /*
  * Bind the driver base on (vendor_id, device_id) pair and later use the
  * (device_id, rev_id) pair as a key to select the devices. The devices with
@@ -164,6 +171,23 @@ static int amdxdna_drm_get_info_ioctl(struct drm_device *dev, void *data, struct
 	return ret;
 }
 
+static int amdxdna_drm_get_array_ioctl(struct drm_device *dev, void *data,
+				       struct drm_file *filp)
+{
+	struct amdxdna_client *client = filp->driver_priv;
+	struct amdxdna_dev *xdna = to_xdna_dev(dev);
+	struct amdxdna_drm_get_array *args = data;
+
+	if (!xdna->dev_info->ops->get_array)
+		return -EOPNOTSUPP;
+
+	if (args->pad || !args->num_element || !args->element_size)
+		return -EINVAL;
+
+	guard(mutex)(&xdna->dev_lock);
+	return xdna->dev_info->ops->get_array(client, args);
+}
+
 static int amdxdna_drm_set_state_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 {
 	struct amdxdna_client *client = filp->driver_priv;
@@ -195,6 +219,7 @@ static const struct drm_ioctl_desc amdxdna_drm_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(AMDXDNA_EXEC_CMD, amdxdna_drm_submit_cmd_ioctl, 0),
 	/* AIE hardware */
 	DRM_IOCTL_DEF_DRV(AMDXDNA_GET_INFO, amdxdna_drm_get_info_ioctl, 0),
+	DRM_IOCTL_DEF_DRV(AMDXDNA_GET_ARRAY, amdxdna_drm_get_array_ioctl, 0),
 	DRM_IOCTL_DEF_DRV(AMDXDNA_SET_STATE, amdxdna_drm_set_state_ioctl, DRM_ROOT_ONLY),
 };
 
@@ -218,6 +243,8 @@ const struct drm_driver amdxdna_drm_drv = {
 	.fops = &amdxdna_fops,
 	.name = "amdxdna_accel_driver",
 	.desc = "AMD XDNA DRM implementation",
+	.major = AMDXDNA_DRIVER_MAJOR,
+	.minor = AMDXDNA_DRIVER_MINOR,
 	.open = amdxdna_drm_open,
 	.postclose = amdxdna_drm_close,
 	.ioctls = amdxdna_drm_ioctls,
diff --git a/drivers/accel/amdxdna/amdxdna_pci_drv.h b/drivers/accel/amdxdna/amdxdna_pci_drv.h
index b6b3b424d1d5..72d6696d49da 100644
--- a/drivers/accel/amdxdna/amdxdna_pci_drv.h
+++ b/drivers/accel/amdxdna/amdxdna_pci_drv.h
@@ -58,6 +58,7 @@ struct amdxdna_dev_ops {
 	int (*cmd_submit)(struct amdxdna_hwctx *hwctx, struct amdxdna_sched_job *job, u64 *seq);
 	int (*get_aie_info)(struct amdxdna_client *client, struct amdxdna_drm_get_info *args);
 	int (*set_aie_state)(struct amdxdna_client *client, struct amdxdna_drm_set_state *args);
+	int (*get_array)(struct amdxdna_client *client, struct amdxdna_drm_get_array *args);
 };
 
 /*
diff --git a/include/uapi/drm/amdxdna_accel.h b/include/uapi/drm/amdxdna_accel.h
index ce523e9ccc52..a1fb9785db77 100644
--- a/include/uapi/drm/amdxdna_accel.h
+++ b/include/uapi/drm/amdxdna_accel.h
@@ -34,6 +34,7 @@ enum amdxdna_drm_ioctl_id {
 	DRM_AMDXDNA_EXEC_CMD,
 	DRM_AMDXDNA_GET_INFO,
 	DRM_AMDXDNA_SET_STATE,
+	DRM_AMDXDNA_GET_ARRAY = 10,
 };
 
 /**
@@ -455,6 +456,112 @@ struct amdxdna_drm_get_info {
 	__u64 buffer; /* in/out */
 };
 
+#define AMDXDNA_HWCTX_STATE_IDLE	0
+#define AMDXDNA_HWCTX_STATE_ACTIVE	1
+
+/**
+ * struct amdxdna_drm_hwctx_entry - The hardware context array entry
+ */
+struct amdxdna_drm_hwctx_entry {
+	/** @context_id: Context ID. */
+	__u32 context_id;
+	/** @start_col: Start AIE array column assigned to context. */
+	__u32 start_col;
+	/** @num_col: Number of AIE array columns assigned to context. */
+	__u32 num_col;
+	/** @hwctx_id: The real hardware context id. */
+	__u32 hwctx_id;
+	/** @pid: ID of process which created this context. */
+	__s64 pid;
+	/** @command_submissions: Number of commands submitted. */
+	__u64 command_submissions;
+	/** @command_completions: Number of commands completed. */
+	__u64 command_completions;
+	/** @migrations: Number of times been migrated. */
+	__u64 migrations;
+	/** @preemptions: Number of times been preempted. */
+	__u64 preemptions;
+	/** @errors: Number of errors happened. */
+	__u64 errors;
+	/** @priority: Context priority. */
+	__u64 priority;
+	/** @heap_usage: Usage of device heap buffer. */
+	__u64 heap_usage;
+	/** @suspensions: Number of times been suspended. */
+	__u64 suspensions;
+	/**
+	 * @state: Context state.
+	 * %AMDXDNA_HWCTX_STATE_IDLE
+	 * %AMDXDNA_HWCTX_STATE_ACTIVE
+	 */
+	__u32 state;
+	/** @pasid: PASID been bound. */
+	__u32 pasid;
+	/** @gops: Giga operations per second. */
+	__u32 gops;
+	/** @fps: Frames per second. */
+	__u32 fps;
+	/** @dma_bandwidth: DMA bandwidth. */
+	__u32 dma_bandwidth;
+	/** @latency: Frame response latency. */
+	__u32 latency;
+	/** @frame_exec_time: Frame execution time. */
+	__u32 frame_exec_time;
+	/** @txn_op_idx: Index of last control code executed. */
+	__u32 txn_op_idx;
+	/** @ctx_pc: Program counter. */
+	__u32 ctx_pc;
+	/** @fatal_error_type: Fatal error type if context crashes. */
+	__u32 fatal_error_type;
+	/** @fatal_error_exception_type: Firmware exception type. */
+	__u32 fatal_error_exception_type;
+	/** @fatal_error_exception_pc: Firmware exception program counter. */
+	__u32 fatal_error_exception_pc;
+	/** @fatal_error_app_module: Exception module name. */
+	__u32 fatal_error_app_module;
+	/** @pad: Structure pad. */
+	__u32 pad;
+};
+
+#define DRM_AMDXDNA_HW_CONTEXT_ALL	0
+
+/**
+ * struct amdxdna_drm_get_array - Get information array.
+ */
+struct amdxdna_drm_get_array {
+	/**
+	 * @param:
+	 *
+	 * Supported params:
+	 *
+	 * %DRM_AMDXDNA_HW_CONTEXT_ALL:
+	 * Returns all created hardware contexts.
+	 */
+	__u32 param;
+	/**
+	 * @element_size:
+	 *
+	 * Specifies maximum element size and returns the actual element size.
+	 */
+	__u32 element_size;
+	/**
+	 * @num_element:
+	 *
+	 * Specifies maximum number of elements and returns the actual number
+	 * of elements.
+	 */
+	__u32 num_element; /* in/out */
+	/** @pad: MBZ */
+	__u32 pad;
+	/**
+	 * @buffer:
+	 *
+	 * Specifies the match conditions and returns the matched information
+	 * array.
+	 */
+	__u64 buffer;
+};
+
 enum amdxdna_drm_set_param {
 	DRM_AMDXDNA_SET_POWER_MODE,
 	DRM_AMDXDNA_WRITE_AIE_MEM,
@@ -519,6 +626,10 @@ struct amdxdna_drm_set_power_mode {
 	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDXDNA_SET_STATE, \
 		 struct amdxdna_drm_set_state)
 
+#define DRM_IOCTL_AMDXDNA_GET_ARRAY \
+	DRM_IOWR(DRM_COMMAND_BASE + DRM_AMDXDNA_GET_ARRAY, \
+		 struct amdxdna_drm_get_array)
+
 #if defined(__cplusplus)
 } /* extern c end */
 #endif

From 467971a908761540cc60b4da22639c440dc05462 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:37 +0200
Subject: [PATCH 0923/1455] rust: page: implement BorrowedPage

Currently, a Page always owns the underlying struct page.

However, sometimes a struct page may be owned by some other entity, e.g.
a vmalloc allocation.

Hence, introduce BorrowedPage to support such cases, until the Ownable
solution [1] lands.

This is required by the scatterlist abstractions.

Acked-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Link: https://lore.kernel.org/rust-for-linux/ZnCzLIly3DRK2eab@boqun-archlinux/ [1]
Link: https://lore.kernel.org/r/20250820145434.94745-2-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/bindings/bindings_helper.h |  1 +
 rust/kernel/page.rs             | 75 ++++++++++++++++++++++++++++++++-
 2 files changed, 75 insertions(+), 1 deletion(-)

diff --git a/rust/bindings/bindings_helper.h b/rust/bindings/bindings_helper.h
index 84d60635e8a9..0e140e07758b 100644
--- a/rust/bindings/bindings_helper.h
+++ b/rust/bindings/bindings_helper.h
@@ -57,6 +57,7 @@
 #include <linux/jiffies.h>
 #include <linux/jump_label.h>
 #include <linux/mdio.h>
+#include <linux/mm.h>
 #include <linux/miscdevice.h>
 #include <linux/of_device.h>
 #include <linux/pci.h>
diff --git a/rust/kernel/page.rs b/rust/kernel/page.rs
index 7c1b17246ed5..631718a6ad7d 100644
--- a/rust/kernel/page.rs
+++ b/rust/kernel/page.rs
@@ -9,7 +9,12 @@ use crate::{
     error::Result,
     uaccess::UserSliceReader,
 };
-use core::ptr::{self, NonNull};
+use core::{
+    marker::PhantomData,
+    mem::ManuallyDrop,
+    ops::Deref,
+    ptr::{self, NonNull},
+};
 
 /// A bitwise shift for the page size.
 pub const PAGE_SHIFT: usize = bindings::PAGE_SHIFT as usize;
@@ -30,6 +35,74 @@ pub const fn page_align(addr: usize) -> usize {
     (addr + (PAGE_SIZE - 1)) & PAGE_MASK
 }
 
+/// Representation of a non-owning reference to a [`Page`].
+///
+/// This type provides a borrowed version of a [`Page`] that is owned by some other entity, e.g. a
+/// [`Vmalloc`] allocation such as [`VBox`].
+///
+/// # Example
+///
+/// ```
+/// # use kernel::{bindings, prelude::*};
+/// use kernel::page::{BorrowedPage, Page, PAGE_SIZE};
+/// # use core::{mem::MaybeUninit, ptr, ptr::NonNull };
+///
+/// fn borrow_page<'a>(vbox: &'a mut VBox<MaybeUninit<[u8; PAGE_SIZE]>>) -> BorrowedPage<'a> {
+///     let ptr = ptr::from_ref(&**vbox);
+///
+///     // SAFETY: `ptr` is a valid pointer to `Vmalloc` memory.
+///     let page = unsafe { bindings::vmalloc_to_page(ptr.cast()) };
+///
+///     // SAFETY: `vmalloc_to_page` returns a valid pointer to a `struct page` for a valid
+///     // pointer to `Vmalloc` memory.
+///     let page = unsafe { NonNull::new_unchecked(page) };
+///
+///     // SAFETY:
+///     // - `self.0` is a valid pointer to a `struct page`.
+///     // - `self.0` is valid for the entire lifetime of `self`.
+///     unsafe { BorrowedPage::from_raw(page) }
+/// }
+///
+/// let mut vbox = VBox::<[u8; PAGE_SIZE]>::new_uninit(GFP_KERNEL)?;
+/// let page = borrow_page(&mut vbox);
+///
+/// // SAFETY: There is no concurrent read or write to this page.
+/// unsafe { page.fill_zero_raw(0, PAGE_SIZE)? };
+/// # Ok::<(), Error>(())
+/// ```
+///
+/// # Invariants
+///
+/// The borrowed underlying pointer to a `struct page` is valid for the entire lifetime `'a`.
+///
+/// [`VBox`]: kernel::alloc::VBox
+/// [`Vmalloc`]: kernel::alloc::allocator::Vmalloc
+pub struct BorrowedPage<'a>(ManuallyDrop<Page>, PhantomData<&'a Page>);
+
+impl<'a> BorrowedPage<'a> {
+    /// Constructs a [`BorrowedPage`] from a raw pointer to a `struct page`.
+    ///
+    /// # Safety
+    ///
+    /// - `ptr` must point to a valid `bindings::page`.
+    /// - `ptr` must remain valid for the entire lifetime `'a`.
+    pub unsafe fn from_raw(ptr: NonNull<bindings::page>) -> Self {
+        let page = Page { page: ptr };
+
+        // INVARIANT: The safety requirements guarantee that `ptr` is valid for the entire lifetime
+        // `'a`.
+        Self(ManuallyDrop::new(page), PhantomData)
+    }
+}
+
+impl<'a> Deref for BorrowedPage<'a> {
+    type Target = Page;
+
+    fn deref(&self) -> &Self::Target {
+        &self.0
+    }
+}
+
 /// A pointer to a page that owns the page allocation.
 ///
 /// # Invariants

From 8e92c9902ff11a1c2aab229a3d7d4c1d7e5b698f Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:38 +0200
Subject: [PATCH 0924/1455] rust: alloc: vmalloc: implement Vmalloc::to_page()

Implement an abstraction of vmalloc_to_page() for subsequent use in the
AsPageIter implementation of VBox and VVec.

Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Link: https://lore.kernel.org/r/20250820145434.94745-3-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/alloc/allocator.rs | 49 ++++++++++++++++++++++++++++++++++
 1 file changed, 49 insertions(+)

diff --git a/rust/kernel/alloc/allocator.rs b/rust/kernel/alloc/allocator.rs
index aa2dfa9dca4c..2315f5063011 100644
--- a/rust/kernel/alloc/allocator.rs
+++ b/rust/kernel/alloc/allocator.rs
@@ -15,6 +15,7 @@ use core::ptr::NonNull;
 
 use crate::alloc::{AllocError, Allocator};
 use crate::bindings;
+use crate::page;
 use crate::pr_warn;
 
 /// The contiguous kernel allocator.
@@ -140,6 +141,54 @@ unsafe impl Allocator for Kmalloc {
     }
 }
 
+impl Vmalloc {
+    /// Convert a pointer to a [`Vmalloc`] allocation to a [`page::BorrowedPage`].
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// # use core::ptr::{NonNull, from_mut};
+    /// # use kernel::{page, prelude::*};
+    /// use kernel::alloc::allocator::Vmalloc;
+    ///
+    /// let mut vbox = VBox::<[u8; page::PAGE_SIZE]>::new_uninit(GFP_KERNEL)?;
+    ///
+    /// {
+    ///     // SAFETY: By the type invariant of `Box` the inner pointer of `vbox` is non-null.
+    ///     let ptr = unsafe { NonNull::new_unchecked(from_mut(&mut *vbox)) };
+    ///
+    ///     // SAFETY:
+    ///     // `ptr` is a valid pointer to a `Vmalloc` allocation.
+    ///     // `ptr` is valid for the entire lifetime of `page`.
+    ///     let page = unsafe { Vmalloc::to_page(ptr.cast()) };
+    ///
+    ///     // SAFETY: There is no concurrent read or write to the same page.
+    ///     unsafe { page.fill_zero_raw(0, page::PAGE_SIZE)? };
+    /// }
+    /// # Ok::<(), Error>(())
+    /// ```
+    ///
+    /// # Safety
+    ///
+    /// - `ptr` must be a valid pointer to a [`Vmalloc`] allocation.
+    /// - `ptr` must remain valid for the entire duration of `'a`.
+    pub unsafe fn to_page<'a>(ptr: NonNull<u8>) -> page::BorrowedPage<'a> {
+        // SAFETY: `ptr` is a valid pointer to `Vmalloc` memory.
+        let page = unsafe { bindings::vmalloc_to_page(ptr.as_ptr().cast()) };
+
+        // SAFETY: `vmalloc_to_page` returns a valid pointer to a `struct page` for a valid pointer
+        // to `Vmalloc` memory.
+        let page = unsafe { NonNull::new_unchecked(page) };
+
+        // SAFETY:
+        // - `page` is a valid pointer to a `struct page`, given that by the safety requirements of
+        //   this function `ptr` is a valid pointer to a `Vmalloc` allocation.
+        // - By the safety requirements of this function `ptr` is valid for the entire lifetime of
+        //   `'a`.
+        unsafe { page::BorrowedPage::from_raw(page) }
+    }
+}
+
 // SAFETY: `realloc` delegates to `ReallocFunc::call`, which guarantees that
 // - memory remains valid until it is explicitly freed,
 // - passing a pointer to a valid memory allocation is OK,

From 2506af5f8109a387a5e8e9e3d7c498480b8033db Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 4 Sep 2025 00:33:30 +0200
Subject: [PATCH 0925/1455] drm/xe/guc: Set upper limit of H2G retries over CTB

The GuC communication protocol allows GuC to send NO_RESPONSE_RETRY
reply message to indicate that due to some interim condition it can
not handle incoming H2G request and the host shall resend it.

But in some cases, due to errors, this unsatisfied condition might
be final and this could lead to endless retries as it was recently
seen on the CI:

 [drm] GT0: PF: VF1 FLR didn't finish in 5000 ms (-ETIMEDOUT)
 [drm] GT0: PF: VF1 resource sanitizing failed (-ETIMEDOUT)
 [drm] GT0: PF: VF1 FLR failed!
 [drm:guc_ct_send_recv [xe]] GT0: H2G action 0x5503 retrying: reason 0x0
 [drm:guc_ct_send_recv [xe]] GT0: H2G action 0x5503 retrying: reason 0x0
 [drm:guc_ct_send_recv [xe]] GT0: H2G action 0x5503 retrying: reason 0x0
 [drm:guc_ct_send_recv [xe]] GT0: H2G action 0x5503 retrying: reason 0x0

To avoid such dangerous loops allow only limited number of retries
(for now 50) and add some delays (n * 5ms) to slow down the rate of
resending this repeated request.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Julia Filipchuk <julia.filipchuk@intel.com>
Link: https://lore.kernel.org/r/20250903223330.6408-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_guc_ct.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index e431ff73227c..f40543b040d6 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -1079,11 +1079,15 @@ static bool retry_failure(struct xe_guc_ct *ct, int ret)
 	return true;
 }
 
+#define GUC_SEND_RETRY_LIMIT	50
+#define GUC_SEND_RETRY_MSLEEP	5
+
 static int guc_ct_send_recv(struct xe_guc_ct *ct, const u32 *action, u32 len,
 			    u32 *response_buffer, bool no_fail)
 {
 	struct xe_gt *gt = ct_to_gt(ct);
 	struct g2h_fence g2h_fence;
+	unsigned int retries = 0;
 	int ret = 0;
 
 	/*
@@ -1148,6 +1152,12 @@ retry_same_fence:
 		xe_gt_dbg(gt, "H2G action %#x retrying: reason %#x\n",
 			  action[0], g2h_fence.reason);
 		mutex_unlock(&ct->lock);
+		if (++retries > GUC_SEND_RETRY_LIMIT) {
+			xe_gt_err(gt, "H2G action %#x reached retry limit=%u, aborting\n",
+				  action[0], GUC_SEND_RETRY_LIMIT);
+			return -ELOOP;
+		}
+		msleep(GUC_SEND_RETRY_MSLEEP * retries);
 		goto retry;
 	}
 	if (g2h_fence.fail) {

From 28fa7f5243c02a812af50707c9805c2b57deceeb Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Thu, 21 Aug 2025 13:52:12 -0700
Subject: [PATCH 0926/1455] drm/bridge: cdns-dsi: Select VIDEOMODE_HELPERS

When no other driver selects CONFIG_VIDEOMODE_HELPERS but
CONFIG_DRM_CDNS_DSI is enabled, there is a linker or modpost error:

  ERROR: modpost: "drm_display_mode_to_videomode" [drivers/gpu/drm/bridge/cadence/cdns-dsi.ko] undefined!

Select VIDEOMODE_HELPERS to ensure that this helper function is
available to the driver.

Fixes: ce4bc5ca7c1d ("drm/bridge: cdns-dsi: Use video mode and clean up cdns_dsi_mode2cfg()")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250821-cdns-videohelpers-v1-1-853e021908cf@kernel.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/cadence/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/bridge/cadence/Kconfig b/drivers/gpu/drm/bridge/cadence/Kconfig
index cced81633ddc..f1d8a8a151d8 100644
--- a/drivers/gpu/drm/bridge/cadence/Kconfig
+++ b/drivers/gpu/drm/bridge/cadence/Kconfig
@@ -6,6 +6,7 @@ config DRM_CDNS_DSI
 	select DRM_PANEL_BRIDGE
 	select GENERIC_PHY
 	select GENERIC_PHY_MIPI_DPHY
+	select VIDEOMODE_HELPERS
 	depends on OF
 	help
 	  Support Cadence DPI to DSI bridge. This is an internal

From 079a5c83dbd23db7a6eed8f558cf75e264d8a17b Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 4 Sep 2025 12:35:21 +0200
Subject: [PATCH 0927/1455] drm/xe/configfs: Don't touch survivability_mode on
 fini

This is a user controlled configfs attribute, we should not
modify that outside the configfs attr.store() implementation.

Fixes: bc417e54e24b ("drm/xe: Enable configfs support for survivability mode")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250904103521.7130-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_survivability_mode.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index 53c5af4b810c..7999cc5262a5 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -44,6 +44,8 @@
  *
  *	# echo 1 > /sys/kernel/config/xe/0000:03:00.0/survivability_mode
  *
+ * It is the responsibility of the user to clear the mode once firmware flash is complete.
+ *
  * Refer :ref:`xe_configfs` for more details on how to use configfs
  *
  * Survivability mode is indicated by the below admin-only readable sysfs which provides additional
@@ -180,7 +182,6 @@ static void xe_survivability_mode_fini(void *arg)
 	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
 	struct device *dev = &pdev->dev;
 
-	xe_configfs_clear_survivability_mode(pdev);
 	sysfs_remove_file(&dev->kobj, &dev_attr_survivability_mode.attr);
 }
 

From b076d321771204d4b711df99e904dc2efdc78856 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 2 Sep 2025 15:17:43 +0200
Subject: [PATCH 0928/1455] drm/xe/configfs: Prepare to filter-out configfs
 attributes

Implement empty ops.is_visible hook to allow filtering-out any not
supported attributes, as not all of them are applicable on all xe
platforms.

Since during creation of each new configfs directory we are looking
for xe device descriptor to validate that xe driver supports given
PCI device, store reference to that descriptor to allow later use
while doing attribute filtering.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250902131744.5076-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_configfs.c | 20 ++++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 1025d3979b06..43f000260752 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -134,6 +134,8 @@ struct xe_config_group_device {
 
 	/* protects attributes */
 	struct mutex lock;
+	/* matching descriptor */
+	const struct xe_device_desc *desc;
 };
 
 static const struct xe_config_device device_defaults = {
@@ -362,8 +364,21 @@ static struct configfs_item_operations xe_config_device_ops = {
 	.release	= xe_config_device_release,
 };
 
+static bool xe_config_device_is_visible(struct config_item *item,
+					struct configfs_attribute *attr, int n)
+{
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+
+	return dev->desc; /* shall be always true */
+}
+
+static struct configfs_group_operations xe_config_device_group_ops = {
+	.is_visible	= xe_config_device_is_visible,
+};
+
 static const struct config_item_type xe_config_device_type = {
 	.ct_item_ops	= &xe_config_device_ops,
+	.ct_group_ops	= &xe_config_device_group_ops,
 	.ct_attrs	= xe_config_device_attrs,
 	.ct_owner	= THIS_MODULE,
 };
@@ -442,6 +457,7 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
+	dev->desc = match;
 	set_device_defaults(&dev->config);
 
 	config_group_init_type_name(&dev->group, name, &xe_config_device_type);
@@ -451,12 +467,12 @@ static struct config_group *xe_config_make_device_group(struct config_group *gro
 	return &dev->group;
 }
 
-static struct configfs_group_operations xe_config_device_group_ops = {
+static struct configfs_group_operations xe_config_group_ops = {
 	.make_group	= xe_config_make_device_group,
 };
 
 static const struct config_item_type xe_configfs_type = {
-	.ct_group_ops	= &xe_config_device_group_ops,
+	.ct_group_ops	= &xe_config_group_ops,
 	.ct_owner	= THIS_MODULE,
 };
 

From 3088f485dea2e98c8acb07495759363c5ae546bd Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 2 Sep 2025 15:17:44 +0200
Subject: [PATCH 0929/1455] drm/xe/configfs: Don't expose survivability_mode if
 not applicable

The survivability_mode attribute is applicable only for DGFX and
platforms newer than BATTLEMAGE. Use .is_visible() hook to hide
this attribute when above conditions are not met. Remove code that
was trying to fix such configuration during the runtime.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250902131744.5076-4-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_configfs.c           | 24 ++++++----------------
 drivers/gpu/drm/xe/xe_configfs.h           |  2 --
 drivers/gpu/drm/xe/xe_survivability_mode.c | 11 +---------
 3 files changed, 7 insertions(+), 30 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 43f000260752..0337811864cd 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -369,7 +369,12 @@ static bool xe_config_device_is_visible(struct config_item *item,
 {
 	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 
-	return dev->desc; /* shall be always true */
+	if (attr == &attr_survivability_mode) {
+		if (!dev->desc->is_dgfx || dev->desc->platform < XE_BATTLEMAGE)
+			return false;
+	}
+
+	return true;
 }
 
 static struct configfs_group_operations xe_config_device_group_ops = {
@@ -558,23 +563,6 @@ bool xe_configfs_get_survivability_mode(struct pci_dev *pdev)
 	return mode;
 }
 
-/**
- * xe_configfs_clear_survivability_mode - clear configfs survivability mode
- * @pdev: pci device
- */
-void xe_configfs_clear_survivability_mode(struct pci_dev *pdev)
-{
-	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
-
-	if (!dev)
-		return;
-
-	guard(mutex)(&dev->lock);
-	dev->config.survivability_mode = 0;
-
-	config_group_put(&dev->group);
-}
-
 /**
  * xe_configfs_get_engines_allowed - get engine allowed mask from configfs
  * @pdev: pci device
diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
index 58c8c3164000..1402e863b71c 100644
--- a/drivers/gpu/drm/xe/xe_configfs.h
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -15,7 +15,6 @@ int xe_configfs_init(void);
 void xe_configfs_exit(void);
 void xe_configfs_check_device(struct pci_dev *pdev);
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
-void xe_configfs_clear_survivability_mode(struct pci_dev *pdev);
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
 bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev);
 #else
@@ -23,7 +22,6 @@ static inline int xe_configfs_init(void) { return 0; }
 static inline void xe_configfs_exit(void) { }
 static inline void xe_configfs_check_device(struct pci_dev *pdev) { }
 static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; }
-static inline void xe_configfs_clear_survivability_mode(struct pci_dev *pdev) { }
 static inline u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { return U64_MAX; }
 static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
 #endif
diff --git a/drivers/gpu/drm/xe/xe_survivability_mode.c b/drivers/gpu/drm/xe/xe_survivability_mode.c
index 7999cc5262a5..1662bfddd4bc 100644
--- a/drivers/gpu/drm/xe/xe_survivability_mode.c
+++ b/drivers/gpu/drm/xe/xe_survivability_mode.c
@@ -289,19 +289,10 @@ bool xe_survivability_mode_is_requested(struct xe_device *xe)
 	u32 data;
 	bool survivability_mode;
 
-	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
+	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe) || xe->info.platform < XE_BATTLEMAGE)
 		return false;
 
 	survivability_mode = xe_configfs_get_survivability_mode(pdev);
-
-	if (xe->info.platform < XE_BATTLEMAGE) {
-		if (survivability_mode) {
-			dev_err(&pdev->dev, "Survivability Mode is not supported on this card\n");
-			xe_configfs_clear_survivability_mode(pdev);
-		}
-		return false;
-	}
-
 	/* Enable survivability mode if set via configfs */
 	if (survivability_mode)
 		return true;

From 7937dca770393d34d1ad217580c5446d2e45417f Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:39 +0200
Subject: [PATCH 0930/1455] rust: alloc: implement VmallocPageIter

Introduce the VmallocPageIter type; an instance of VmallocPageIter may
be exposed by owners of vmalloc allocations to provide borrowed access
to its backing pages.

For instance, this is useful to access and borrow the backing pages of
allocation primitives, such as Box and Vec, backing a scatterlist.

Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Suggested-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250820145434.94745-4-dakr@kernel.org
[ Drop VmallocPageIter::base_address(), move to allocator/iter.rs and
  stub VmallocPageIter for allocator_test.rs. - Danilo ]
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/alloc/allocator.rs      |   3 +
 rust/kernel/alloc/allocator/iter.rs | 102 ++++++++++++++++++++++++++++
 rust/kernel/alloc/allocator_test.rs |  29 ++++++++
 3 files changed, 134 insertions(+)
 create mode 100644 rust/kernel/alloc/allocator/iter.rs

diff --git a/rust/kernel/alloc/allocator.rs b/rust/kernel/alloc/allocator.rs
index 2315f5063011..90e03ad15760 100644
--- a/rust/kernel/alloc/allocator.rs
+++ b/rust/kernel/alloc/allocator.rs
@@ -18,6 +18,9 @@ use crate::bindings;
 use crate::page;
 use crate::pr_warn;
 
+mod iter;
+pub use self::iter::VmallocPageIter;
+
 /// The contiguous kernel allocator.
 ///
 /// `Kmalloc` is typically used for physically contiguous allocations up to page size, but also
diff --git a/rust/kernel/alloc/allocator/iter.rs b/rust/kernel/alloc/allocator/iter.rs
new file mode 100644
index 000000000000..5759f86029b7
--- /dev/null
+++ b/rust/kernel/alloc/allocator/iter.rs
@@ -0,0 +1,102 @@
+// SPDX-License-Identifier: GPL-2.0
+
+use super::Vmalloc;
+use crate::page;
+use core::marker::PhantomData;
+use core::ptr::NonNull;
+
+/// An [`Iterator`] of [`page::BorrowedPage`] items owned by a [`Vmalloc`] allocation.
+///
+/// # Guarantees
+///
+/// The pages iterated by the [`Iterator`] appear in the order as they are mapped in the CPU's
+/// virtual address space ascendingly.
+///
+/// # Invariants
+///
+/// - `buf` is a valid and [`page::PAGE_SIZE`] aligned pointer into a [`Vmalloc`] allocation.
+/// - `size` is the number of bytes from `buf` until the end of the [`Vmalloc`] allocation `buf`
+///   points to.
+pub struct VmallocPageIter<'a> {
+    /// The base address of the [`Vmalloc`] buffer.
+    buf: NonNull<u8>,
+    /// The size of the buffer pointed to by `buf` in bytes.
+    size: usize,
+    /// The current page index of the [`Iterator`].
+    index: usize,
+    _p: PhantomData<page::BorrowedPage<'a>>,
+}
+
+impl<'a> Iterator for VmallocPageIter<'a> {
+    type Item = page::BorrowedPage<'a>;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        let offset = self.index.checked_mul(page::PAGE_SIZE)?;
+
+        // Even though `self.size()` may be smaller than `Self::page_count() * page::PAGE_SIZE`, it
+        // is always a number between `(Self::page_count() - 1) * page::PAGE_SIZE` and
+        // `Self::page_count() * page::PAGE_SIZE`, hence the check below is sufficient.
+        if offset < self.size() {
+            self.index += 1;
+        } else {
+            return None;
+        }
+
+        // TODO: Use `NonNull::add()` instead, once the minimum supported compiler version is
+        // bumped to 1.80 or later.
+        //
+        // SAFETY: `offset` is in the interval `[0, (self.page_count() - 1) * page::PAGE_SIZE]`,
+        // hence the resulting pointer is guaranteed to be within the same allocation.
+        let ptr = unsafe { self.buf.as_ptr().add(offset) };
+
+        // SAFETY: `ptr` is guaranteed to be non-null given that it is derived from `self.buf`.
+        let ptr = unsafe { NonNull::new_unchecked(ptr) };
+
+        // SAFETY:
+        // - `ptr` is a valid pointer to a `Vmalloc` allocation.
+        // - `ptr` is valid for the duration of `'a`.
+        Some(unsafe { Vmalloc::to_page(ptr) })
+    }
+
+    fn size_hint(&self) -> (usize, Option<usize>) {
+        let remaining = self.page_count().saturating_sub(self.index);
+
+        (remaining, Some(remaining))
+    }
+}
+
+impl<'a> VmallocPageIter<'a> {
+    /// Creates a new [`VmallocPageIter`] instance.
+    ///
+    /// # Safety
+    ///
+    /// - `buf` must be a [`page::PAGE_SIZE`] aligned pointer into a [`Vmalloc`] allocation.
+    /// - `buf` must be valid for at least the lifetime of `'a`.
+    /// - `size` must be the number of bytes from `buf` until the end of the [`Vmalloc`] allocation
+    ///   `buf` points to.
+    pub unsafe fn new(buf: NonNull<u8>, size: usize) -> Self {
+        // INVARIANT: By the safety requirements, `buf` is a valid and `page::PAGE_SIZE` aligned
+        // pointer into a [`Vmalloc`] allocation.
+        Self {
+            buf,
+            size,
+            index: 0,
+            _p: PhantomData,
+        }
+    }
+
+    /// Returns the size of the backing [`Vmalloc`] allocation in bytes.
+    ///
+    /// Note that this is the size the [`Vmalloc`] allocation has been allocated with. Hence, this
+    /// number may be smaller than `[`Self::page_count`] * [`page::PAGE_SIZE`]`.
+    #[inline]
+    pub fn size(&self) -> usize {
+        self.size
+    }
+
+    /// Returns the number of pages owned by the backing [`Vmalloc`] allocation.
+    #[inline]
+    pub fn page_count(&self) -> usize {
+        self.size().div_ceil(page::PAGE_SIZE)
+    }
+}
diff --git a/rust/kernel/alloc/allocator_test.rs b/rust/kernel/alloc/allocator_test.rs
index a3074480bd8d..f46b4b671389 100644
--- a/rust/kernel/alloc/allocator_test.rs
+++ b/rust/kernel/alloc/allocator_test.rs
@@ -12,8 +12,10 @@
 use super::{flags::*, AllocError, Allocator, Flags};
 use core::alloc::Layout;
 use core::cmp;
+use core::marker::PhantomData;
 use core::ptr;
 use core::ptr::NonNull;
+use kernel::page;
 
 /// The userspace allocator based on libc.
 pub struct Cmalloc;
@@ -22,6 +24,33 @@ pub type Kmalloc = Cmalloc;
 pub type Vmalloc = Kmalloc;
 pub type KVmalloc = Kmalloc;
 
+pub struct VmallocPageIter<'a> {
+    _p: PhantomData<page::BorrowedPage<'a>>,
+}
+
+impl<'a> Iterator for VmallocPageIter<'a> {
+    type Item = page::BorrowedPage<'a>;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        None
+    }
+}
+
+impl<'a> VmallocPageIter<'a> {
+    #[allow(clippy::missing_safety_doc)]
+    pub unsafe fn new(_buf: NonNull<u8>, _size: usize) -> Self {
+        Self { _p: PhantomData }
+    }
+
+    pub fn size(&self) -> usize {
+        0
+    }
+
+    pub fn page_count(&self) -> usize {
+        0
+    }
+}
+
 extern "C" {
     #[link_name = "aligned_alloc"]
     fn libc_aligned_alloc(align: usize, size: usize) -> *mut crate::ffi::c_void;

From 866ec3bab19c33a66ad384bc69dff1cd83d7ed12 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:40 +0200
Subject: [PATCH 0931/1455] rust: page: define trait AsPageIter

The AsPageIter trait provides a common interface for types that
provide a page iterator, such as VmallocPageIter.

Subsequent patches will leverage this to let VBox and VVec provide a
VmallocPageIter though this trait.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Link: https://lore.kernel.org/r/20250820145434.94745-5-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/page.rs | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/rust/kernel/page.rs b/rust/kernel/page.rs
index 631718a6ad7d..75ef096075cb 100644
--- a/rust/kernel/page.rs
+++ b/rust/kernel/page.rs
@@ -103,6 +103,18 @@ impl<'a> Deref for BorrowedPage<'a> {
     }
 }
 
+/// Trait to be implemented by types which provide an [`Iterator`] implementation of
+/// [`BorrowedPage`] items, such as [`VmallocPageIter`](kernel::alloc::allocator::VmallocPageIter).
+pub trait AsPageIter {
+    /// The [`Iterator`] type, e.g. [`VmallocPageIter`](kernel::alloc::allocator::VmallocPageIter).
+    type Iter<'a>: Iterator<Item = BorrowedPage<'a>>
+    where
+        Self: 'a;
+
+    /// Returns an [`Iterator`] of [`BorrowedPage`] items over all pages owned by `self`.
+    fn page_iter(&mut self) -> Self::Iter<'_>;
+}
+
 /// A pointer to a page that owns the page allocation.
 ///
 /// # Invariants

From 671618432f46c553d9b82ad701ac054e16c8d55a Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:41 +0200
Subject: [PATCH 0932/1455] rust: alloc: kbox: implement AsPageIter for VBox

Implement AsPageIter for VBox; this allows to iterate and borrow the
backing pages of a VBox. This, for instance, is useful in combination
with VBox backing a scatterlist.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Link: https://lore.kernel.org/r/20250820145434.94745-6-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/alloc/kbox.rs | 40 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 39 insertions(+), 1 deletion(-)

diff --git a/rust/kernel/alloc/kbox.rs b/rust/kernel/alloc/kbox.rs
index 856d05aa60f1..b69ff4a1d748 100644
--- a/rust/kernel/alloc/kbox.rs
+++ b/rust/kernel/alloc/kbox.rs
@@ -3,7 +3,7 @@
 //! Implementation of [`Box`].
 
 #[allow(unused_imports)] // Used in doc comments.
-use super::allocator::{KVmalloc, Kmalloc, Vmalloc};
+use super::allocator::{KVmalloc, Kmalloc, Vmalloc, VmallocPageIter};
 use super::{AllocError, Allocator, Flags};
 use core::alloc::Layout;
 use core::borrow::{Borrow, BorrowMut};
@@ -18,6 +18,7 @@ use core::result::Result;
 
 use crate::ffi::c_void;
 use crate::init::InPlaceInit;
+use crate::page::AsPageIter;
 use crate::types::ForeignOwnable;
 use pin_init::{InPlaceWrite, Init, PinInit, ZeroableOption};
 
@@ -598,3 +599,40 @@ where
         unsafe { A::free(self.0.cast(), layout) };
     }
 }
+
+/// # Examples
+///
+/// ```
+/// # use kernel::prelude::*;
+/// use kernel::alloc::allocator::VmallocPageIter;
+/// use kernel::page::{AsPageIter, PAGE_SIZE};
+///
+/// let mut vbox = VBox::new((), GFP_KERNEL)?;
+///
+/// assert!(vbox.page_iter().next().is_none());
+///
+/// let mut vbox = VBox::<[u8; PAGE_SIZE]>::new_uninit(GFP_KERNEL)?;
+///
+/// let page = vbox.page_iter().next().expect("At least one page should be available.\n");
+///
+/// // SAFETY: There is no concurrent read or write to the same page.
+/// unsafe { page.fill_zero_raw(0, PAGE_SIZE)? };
+/// # Ok::<(), Error>(())
+/// ```
+impl<T> AsPageIter for VBox<T> {
+    type Iter<'a>
+        = VmallocPageIter<'a>
+    where
+        T: 'a;
+
+    fn page_iter(&mut self) -> Self::Iter<'_> {
+        let ptr = self.0.cast();
+        let size = core::mem::size_of::<T>();
+
+        // SAFETY:
+        // - `ptr` is a valid pointer to the beginning of a `Vmalloc` allocation.
+        // - `ptr` is guaranteed to be valid for the lifetime of `'a`.
+        // - `size` is the size of the `Vmalloc` allocation `ptr` points to.
+        unsafe { VmallocPageIter::new(ptr, size) }
+    }
+}

From 9acb4e630c3f3fc070b9962683fdde5ba1c4c70c Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:42 +0200
Subject: [PATCH 0933/1455] rust: alloc: layout: implement ArrayLayout::size()

Provide a convenience method for ArrayLayout to calculate the size of
the ArrayLayout in bytes.

Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Abdiel Janulgue <abdiel.janulgue@gmail.com>
Link: https://lore.kernel.org/r/20250820145434.94745-7-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/alloc/layout.rs | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/rust/kernel/alloc/layout.rs b/rust/kernel/alloc/layout.rs
index 93ed514f7cc7..666accb7859c 100644
--- a/rust/kernel/alloc/layout.rs
+++ b/rust/kernel/alloc/layout.rs
@@ -98,6 +98,11 @@ impl<T> ArrayLayout<T> {
     pub const fn is_empty(&self) -> bool {
         self.len == 0
     }
+
+    /// Returns the size of the [`ArrayLayout`] in bytes.
+    pub const fn size(&self) -> usize {
+        self.len() * core::mem::size_of::<T>()
+    }
 }
 
 impl<T> From<ArrayLayout<T>> for Layout {

From 779db37373a38f23e2534c7f4ffe8188f5237988 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Wed, 20 Aug 2025 16:53:43 +0200
Subject: [PATCH 0934/1455] rust: alloc: kvec: implement AsPageIter for VVec

Implement AsPageIter for VVec; this allows to iterate and borrow the
backing pages of a VVec. This, for instance, is useful in combination
with VVec backing a scatterlist.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Link: https://lore.kernel.org/r/20250820145434.94745-8-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/kernel/alloc/kvec.rs | 40 ++++++++++++++++++++++++++++++++++++++-
 1 file changed, 39 insertions(+), 1 deletion(-)

diff --git a/rust/kernel/alloc/kvec.rs b/rust/kernel/alloc/kvec.rs
index 3c72e0bdddb8..ac438e70a1ed 100644
--- a/rust/kernel/alloc/kvec.rs
+++ b/rust/kernel/alloc/kvec.rs
@@ -3,10 +3,11 @@
 //! Implementation of [`Vec`].
 
 use super::{
-    allocator::{KVmalloc, Kmalloc, Vmalloc},
+    allocator::{KVmalloc, Kmalloc, Vmalloc, VmallocPageIter},
     layout::ArrayLayout,
     AllocError, Allocator, Box, Flags,
 };
+use crate::page::AsPageIter;
 use core::{
     borrow::{Borrow, BorrowMut},
     fmt,
@@ -1017,6 +1018,43 @@ where
     }
 }
 
+/// # Examples
+///
+/// ```
+/// # use kernel::prelude::*;
+/// use kernel::alloc::allocator::VmallocPageIter;
+/// use kernel::page::{AsPageIter, PAGE_SIZE};
+///
+/// let mut vec = VVec::<u8>::new();
+///
+/// assert!(vec.page_iter().next().is_none());
+///
+/// vec.reserve(PAGE_SIZE, GFP_KERNEL)?;
+///
+/// let page = vec.page_iter().next().expect("At least one page should be available.\n");
+///
+/// // SAFETY: There is no concurrent read or write to the same page.
+/// unsafe { page.fill_zero_raw(0, PAGE_SIZE)? };
+/// # Ok::<(), Error>(())
+/// ```
+impl<T> AsPageIter for VVec<T> {
+    type Iter<'a>
+        = VmallocPageIter<'a>
+    where
+        T: 'a;
+
+    fn page_iter(&mut self) -> Self::Iter<'_> {
+        let ptr = self.ptr.cast();
+        let size = self.layout.size();
+
+        // SAFETY:
+        // - `ptr` is a valid pointer to the beginning of a `Vmalloc` allocation.
+        // - `ptr` is guaranteed to be valid for the lifetime of `'a`.
+        // - `size` is the size of the `Vmalloc` allocation `ptr` points to.
+        unsafe { VmallocPageIter::new(ptr, size) }
+    }
+}
+
 /// An [`Iterator`] implementation for [`Vec`] that moves elements out of a vector.
 ///
 /// This structure is created by the [`Vec::into_iter`] method on [`Vec`] (provided by the

From c2437c43cfb1f7f461669ba3cd74936fe345640b Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Thu, 28 Aug 2025 15:32:14 +0200
Subject: [PATCH 0935/1455] rust: dma: implement DataDirection

Add the `DataDirection` struct, a newtype wrapper around the C
`enum dma_data_direction`.

This provides a type-safe Rust interface for specifying the direction of
DMA transfers.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Link: https://lore.kernel.org/r/20250828133323.53311-2-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/bindings/bindings_helper.h |  1 +
 rust/kernel/dma.rs              | 68 +++++++++++++++++++++++++++++++++
 2 files changed, 69 insertions(+)

diff --git a/rust/bindings/bindings_helper.h b/rust/bindings/bindings_helper.h
index 0e140e07758b..c2cc52ee9945 100644
--- a/rust/bindings/bindings_helper.h
+++ b/rust/bindings/bindings_helper.h
@@ -47,6 +47,7 @@
 #include <linux/cpumask.h>
 #include <linux/cred.h>
 #include <linux/device/faux.h>
+#include <linux/dma-direction.h>
 #include <linux/dma-mapping.h>
 #include <linux/errname.h>
 #include <linux/ethtool.h>
diff --git a/rust/kernel/dma.rs b/rust/kernel/dma.rs
index 2bc8ab51ec28..27b25f041f32 100644
--- a/rust/kernel/dma.rs
+++ b/rust/kernel/dma.rs
@@ -244,6 +244,74 @@ pub mod attrs {
     pub const DMA_ATTR_PRIVILEGED: Attrs = Attrs(bindings::DMA_ATTR_PRIVILEGED);
 }
 
+/// DMA data direction.
+///
+/// Corresponds to the C [`enum dma_data_direction`].
+///
+/// [`enum dma_data_direction`]: srctree/include/linux/dma-direction.h
+#[derive(Copy, Clone, PartialEq, Eq, Debug)]
+#[repr(u32)]
+pub enum DataDirection {
+    /// The DMA mapping is for bidirectional data transfer.
+    ///
+    /// This is used when the buffer can be both read from and written to by the device.
+    /// The cache for the corresponding memory region is both flushed and invalidated.
+    Bidirectional = Self::const_cast(bindings::dma_data_direction_DMA_BIDIRECTIONAL),
+
+    /// The DMA mapping is for data transfer from memory to the device (write).
+    ///
+    /// The CPU has prepared data in the buffer, and the device will read it.
+    /// The cache for the corresponding memory region is flushed before device access.
+    ToDevice = Self::const_cast(bindings::dma_data_direction_DMA_TO_DEVICE),
+
+    /// The DMA mapping is for data transfer from the device to memory (read).
+    ///
+    /// The device will write data into the buffer for the CPU to read.
+    /// The cache for the corresponding memory region is invalidated before CPU access.
+    FromDevice = Self::const_cast(bindings::dma_data_direction_DMA_FROM_DEVICE),
+
+    /// The DMA mapping is not for data transfer.
+    ///
+    /// This is primarily for debugging purposes. With this direction, the DMA mapping API
+    /// will not perform any cache coherency operations.
+    None = Self::const_cast(bindings::dma_data_direction_DMA_NONE),
+}
+
+impl DataDirection {
+    /// Casts the bindgen-generated enum type to a `u32` at compile time.
+    ///
+    /// This function will cause a compile-time error if the underlying value of the
+    /// C enum is out of bounds for `u32`.
+    const fn const_cast(val: bindings::dma_data_direction) -> u32 {
+        // CAST: The C standard allows compilers to choose different integer types for enums.
+        // To safely check the value, we cast it to a wide signed integer type (`i128`)
+        // which can hold any standard C integer enum type without truncation.
+        let wide_val = val as i128;
+
+        // Check if the value is outside the valid range for the target type `u32`.
+        // CAST: `u32::MAX` is cast to `i128` to match the type of `wide_val` for the comparison.
+        if wide_val < 0 || wide_val > u32::MAX as i128 {
+            // Trigger a compile-time error in a const context.
+            build_error!("C enum value is out of bounds for the target type `u32`.");
+        }
+
+        // CAST: This cast is valid because the check above guarantees that `wide_val`
+        // is within the representable range of `u32`.
+        wide_val as u32
+    }
+}
+
+impl From<DataDirection> for bindings::dma_data_direction {
+    /// Returns the raw representation of [`enum dma_data_direction`].
+    fn from(direction: DataDirection) -> Self {
+        // CAST: `direction as u32` gets the underlying representation of our `#[repr(u32)]` enum.
+        // The subsequent cast to `Self` (the bindgen type) assumes the C enum is compatible
+        // with the enum variants of `DataDirection`, which is a valid assumption given our
+        // compile-time checks.
+        direction as u32 as Self
+    }
+}
+
 /// An abstraction of the `dma_alloc_coherent` API.
 ///
 /// This is an abstraction around the `dma_alloc_coherent` API which is used to allocate and map

From c7081ec661bd2e9bfabe665b72c91698c396792c Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Thu, 28 Aug 2025 15:32:15 +0200
Subject: [PATCH 0936/1455] rust: dma: add type alias for bindings::dma_addr_t

Add a type alias for bindings::dma_addr_t (DmaAddress), such that we do
not have to access bindings directly.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Suggested-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250828133323.53311-3-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 drivers/gpu/nova-core/falcon.rs |  4 ++--
 rust/kernel/dma.rs              | 18 ++++++++++++++----
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index be91aac6976a..2dbcdf26697b 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -4,8 +4,8 @@
 
 use core::ops::Deref;
 use hal::FalconHal;
-use kernel::bindings;
 use kernel::device;
+use kernel::dma::DmaAddress;
 use kernel::prelude::*;
 use kernel::sync::aref::ARef;
 use kernel::time::Delta;
@@ -455,7 +455,7 @@ impl<E: FalconEngine + 'static> Falcon<E> {
                 fw.dma_handle_with_offset(load_offsets.src_start as usize)?,
             ),
         };
-        if dma_start % bindings::dma_addr_t::from(DMA_LEN) > 0 {
+        if dma_start % DmaAddress::from(DMA_LEN) > 0 {
             dev_err!(
                 self.dev,
                 "DMA transfer start addresses must be a multiple of {}",
diff --git a/rust/kernel/dma.rs b/rust/kernel/dma.rs
index 27b25f041f32..b2a6282876da 100644
--- a/rust/kernel/dma.rs
+++ b/rust/kernel/dma.rs
@@ -13,6 +13,16 @@ use crate::{
     types::ARef,
 };
 
+/// DMA address type.
+///
+/// Represents a bus address used for Direct Memory Access (DMA) operations.
+///
+/// This is an alias of the kernel's `dma_addr_t`, which may be `u32` or `u64` depending on
+/// `CONFIG_ARCH_DMA_ADDR_T_64BIT`.
+///
+/// Note that this may be `u64` even on 32-bit architectures.
+pub type DmaAddress = bindings::dma_addr_t;
+
 /// Trait to be implemented by DMA capable bus devices.
 ///
 /// The [`dma::Device`](Device) trait should be implemented by bus specific device representations,
@@ -343,7 +353,7 @@ impl From<DataDirection> for bindings::dma_data_direction {
 // entire `CoherentAllocation` including the allocated memory itself.
 pub struct CoherentAllocation<T: AsBytes + FromBytes> {
     dev: ARef<device::Device>,
-    dma_handle: bindings::dma_addr_t,
+    dma_handle: DmaAddress,
     count: usize,
     cpu_addr: *mut T,
     dma_attrs: Attrs,
@@ -444,7 +454,7 @@ impl<T: AsBytes + FromBytes> CoherentAllocation<T> {
 
     /// Returns a DMA handle which may be given to the device as the DMA address base of
     /// the region.
-    pub fn dma_handle(&self) -> bindings::dma_addr_t {
+    pub fn dma_handle(&self) -> DmaAddress {
         self.dma_handle
     }
 
@@ -452,13 +462,13 @@ impl<T: AsBytes + FromBytes> CoherentAllocation<T> {
     /// device as the DMA address base of the region.
     ///
     /// Returns `EINVAL` if `offset` is not within the bounds of the allocation.
-    pub fn dma_handle_with_offset(&self, offset: usize) -> Result<bindings::dma_addr_t> {
+    pub fn dma_handle_with_offset(&self, offset: usize) -> Result<DmaAddress> {
         if offset >= self.count {
             Err(EINVAL)
         } else {
             // INVARIANT: The type invariant of `Self` guarantees that `size_of::<T> * count` fits
             // into a `usize`, and `offset` is inferior to `count`.
-            Ok(self.dma_handle + (offset * core::mem::size_of::<T>()) as bindings::dma_addr_t)
+            Ok(self.dma_handle + (offset * core::mem::size_of::<T>()) as DmaAddress)
         }
     }
 

From 05aa6fb1c21d7fb9df735da24096d793223789d5 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Thu, 28 Aug 2025 15:32:16 +0200
Subject: [PATCH 0937/1455] rust: scatterlist: Add abstraction for sg_table

Add a safe Rust abstraction for the kernel's scatter-gather list
facilities (`struct scatterlist` and `struct sg_table`).

This commit introduces `SGTable<T>`, a wrapper that uses a generic
parameter to provide compile-time guarantees about ownership and lifetime.

The abstraction provides two primary states:
- `SGTable<Owned<P>>`: Represents a table whose resources are fully
  managed by Rust. It takes ownership of a page provider `P`, allocates
  the underlying `struct sg_table`, maps it for DMA, and handles all
  cleanup automatically upon drop. The DMA mapping's lifetime is tied to
  the associated device using `Devres`, ensuring it is correctly unmapped
  before the device is unbound.
- `SGTable<Borrowed>` (or just `SGTable`): A zero-cost representation of
  an externally managed `struct sg_table`. It is created from a raw
  pointer using `SGTable::from_raw()` and provides a lifetime-bound
  reference (`&'a SGTable`) for operations like iteration.

The API exposes a safe iterator that yields `&SGEntry` references,
allowing drivers to easily access the DMA address and length of each
segment in the list.

Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Co-developed-by: Abdiel Janulgue <abdiel.janulgue@gmail.com>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@gmail.com>
Link: https://lore.kernel.org/r/20250828133323.53311-4-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 rust/helpers/helpers.c     |   1 +
 rust/helpers/scatterlist.c |  24 ++
 rust/kernel/lib.rs         |   1 +
 rust/kernel/scatterlist.rs | 491 +++++++++++++++++++++++++++++++++++++
 4 files changed, 517 insertions(+)
 create mode 100644 rust/helpers/scatterlist.c
 create mode 100644 rust/kernel/scatterlist.rs

diff --git a/rust/helpers/helpers.c b/rust/helpers/helpers.c
index 7cf7fe95e41d..e94542bf6ea7 100644
--- a/rust/helpers/helpers.c
+++ b/rust/helpers/helpers.c
@@ -39,6 +39,7 @@
 #include "rcu.c"
 #include "refcount.c"
 #include "regulator.c"
+#include "scatterlist.c"
 #include "security.c"
 #include "signal.c"
 #include "slab.c"
diff --git a/rust/helpers/scatterlist.c b/rust/helpers/scatterlist.c
new file mode 100644
index 000000000000..80c956ee09ab
--- /dev/null
+++ b/rust/helpers/scatterlist.c
@@ -0,0 +1,24 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/dma-direction.h>
+
+dma_addr_t rust_helper_sg_dma_address(struct scatterlist *sg)
+{
+	return sg_dma_address(sg);
+}
+
+unsigned int rust_helper_sg_dma_len(struct scatterlist *sg)
+{
+	return sg_dma_len(sg);
+}
+
+struct scatterlist *rust_helper_sg_next(struct scatterlist *sg)
+{
+	return sg_next(sg);
+}
+
+void rust_helper_dma_unmap_sgtable(struct device *dev, struct sg_table *sgt,
+				   enum dma_data_direction dir, unsigned long attrs)
+{
+	return dma_unmap_sgtable(dev, sgt, dir, attrs);
+}
diff --git a/rust/kernel/lib.rs b/rust/kernel/lib.rs
index c859a8984bae..fcffc3988a90 100644
--- a/rust/kernel/lib.rs
+++ b/rust/kernel/lib.rs
@@ -114,6 +114,7 @@ pub mod print;
 pub mod rbtree;
 pub mod regulator;
 pub mod revocable;
+pub mod scatterlist;
 pub mod security;
 pub mod seq_file;
 pub mod sizes;
diff --git a/rust/kernel/scatterlist.rs b/rust/kernel/scatterlist.rs
new file mode 100644
index 000000000000..9709dff60b5a
--- /dev/null
+++ b/rust/kernel/scatterlist.rs
@@ -0,0 +1,491 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! Abstractions for scatter-gather lists.
+//!
+//! C header: [`include/linux/scatterlist.h`](srctree/include/linux/scatterlist.h)
+//!
+//! Scatter-gather (SG) I/O is a memory access technique that allows devices to perform DMA
+//! operations on data buffers that are not physically contiguous in memory. It works by creating a
+//! "scatter-gather list", an array where each entry specifies the address and length of a
+//! physically contiguous memory segment.
+//!
+//! The device's DMA controller can then read this list and process the segments sequentially as
+//! part of one logical I/O request. This avoids the need for a single, large, physically contiguous
+//! memory buffer, which can be difficult or impossible to allocate.
+//!
+//! This module provides safe Rust abstractions over the kernel's `struct scatterlist` and
+//! `struct sg_table` types.
+//!
+//! The main entry point is the [`SGTable`] type, which represents a complete scatter-gather table.
+//! It can be either:
+//!
+//! - An owned table ([`SGTable<Owned<P>>`]), created from a Rust memory buffer (e.g., [`VVec`]).
+//!   This type manages the allocation of the `struct sg_table`, the DMA mapping of the buffer, and
+//!   the automatic cleanup of all resources.
+//! - A borrowed reference (&[`SGTable`]), which provides safe, read-only access to a table that was
+//!   allocated by other (e.g., C) code.
+//!
+//! Individual entries in the table are represented by [`SGEntry`], which can be accessed by
+//! iterating over an [`SGTable`].
+
+use crate::{
+    alloc,
+    alloc::allocator::VmallocPageIter,
+    bindings,
+    device::{Bound, Device},
+    devres::Devres,
+    dma, error,
+    io::resource::ResourceSize,
+    page,
+    prelude::*,
+    types::{ARef, Opaque},
+};
+use core::{ops::Deref, ptr::NonNull};
+
+/// A single entry in a scatter-gather list.
+///
+/// An `SGEntry` represents a single, physically contiguous segment of memory that has been mapped
+/// for DMA.
+///
+/// Instances of this struct are obtained by iterating over an [`SGTable`]. Drivers do not create
+/// or own [`SGEntry`] objects directly.
+#[repr(transparent)]
+pub struct SGEntry(Opaque<bindings::scatterlist>);
+
+// SAFETY: `SGEntry` can be sent to any task.
+unsafe impl Send for SGEntry {}
+
+// SAFETY: `SGEntry` has no interior mutability and can be accessed concurrently.
+unsafe impl Sync for SGEntry {}
+
+impl SGEntry {
+    /// Convert a raw `struct scatterlist *` to a `&'a SGEntry`.
+    ///
+    /// # Safety
+    ///
+    /// Callers must ensure that the `struct scatterlist` pointed to by `ptr` is valid for the
+    /// lifetime `'a`.
+    #[inline]
+    unsafe fn from_raw<'a>(ptr: *mut bindings::scatterlist) -> &'a Self {
+        // SAFETY: The safety requirements of this function guarantee that `ptr` is a valid pointer
+        // to a `struct scatterlist` for the duration of `'a`.
+        unsafe { &*ptr.cast() }
+    }
+
+    /// Obtain the raw `struct scatterlist *`.
+    #[inline]
+    fn as_raw(&self) -> *mut bindings::scatterlist {
+        self.0.get()
+    }
+
+    /// Returns the DMA address of this SG entry.
+    ///
+    /// This is the address that the device should use to access the memory segment.
+    #[inline]
+    pub fn dma_address(&self) -> dma::DmaAddress {
+        // SAFETY: `self.as_raw()` is a valid pointer to a `struct scatterlist`.
+        unsafe { bindings::sg_dma_address(self.as_raw()) }
+    }
+
+    /// Returns the length of this SG entry in bytes.
+    #[inline]
+    pub fn dma_len(&self) -> ResourceSize {
+        #[allow(clippy::useless_conversion)]
+        // SAFETY: `self.as_raw()` is a valid pointer to a `struct scatterlist`.
+        unsafe { bindings::sg_dma_len(self.as_raw()) }.into()
+    }
+}
+
+/// The borrowed generic type of an [`SGTable`], representing a borrowed or externally managed
+/// table.
+#[repr(transparent)]
+pub struct Borrowed(Opaque<bindings::sg_table>);
+
+// SAFETY: `Borrowed` can be sent to any task.
+unsafe impl Send for Borrowed {}
+
+// SAFETY: `Borrowed` has no interior mutability and can be accessed concurrently.
+unsafe impl Sync for Borrowed {}
+
+/// A scatter-gather table.
+///
+/// This struct is a wrapper around the kernel's `struct sg_table`. It manages a list of DMA-mapped
+/// memory segments that can be passed to a device for I/O operations.
+///
+/// The generic parameter `T` is used as a generic type to distinguish between owned and borrowed
+/// tables.
+///
+///  - [`SGTable<Owned>`]: An owned table created and managed entirely by Rust code. It handles
+///    allocation, DMA mapping, and cleanup of all associated resources. See [`SGTable::new`].
+///  - [`SGTable<Borrowed>`} (or simply [`SGTable`]): Represents a table whose lifetime is managed
+///    externally. It can be used safely via a borrowed reference `&'a SGTable`, where `'a` is the
+///    external lifetime.
+///
+/// All [`SGTable`] variants can be iterated over the individual [`SGEntry`]s.
+#[repr(transparent)]
+#[pin_data]
+pub struct SGTable<T: private::Sealed = Borrowed> {
+    #[pin]
+    inner: T,
+}
+
+impl SGTable {
+    /// Creates a borrowed `&'a SGTable` from a raw `struct sg_table` pointer.
+    ///
+    /// This allows safe access to an `sg_table` that is managed elsewhere (for example, in C code).
+    ///
+    /// # Safety
+    ///
+    /// Callers must ensure that:
+    ///
+    /// - the `struct sg_table` pointed to by `ptr` is valid for the entire lifetime of `'a`,
+    /// - the data behind `ptr` is not modified concurrently for the duration of `'a`.
+    #[inline]
+    pub unsafe fn from_raw<'a>(ptr: *mut bindings::sg_table) -> &'a Self {
+        // SAFETY: The safety requirements of this function guarantee that `ptr` is a valid pointer
+        // to a `struct sg_table` for the duration of `'a`.
+        unsafe { &*ptr.cast() }
+    }
+
+    #[inline]
+    fn as_raw(&self) -> *mut bindings::sg_table {
+        self.inner.0.get()
+    }
+
+    /// Returns an [`SGTableIter`] bound to the lifetime of `self`.
+    pub fn iter(&self) -> SGTableIter<'_> {
+        // SAFETY: `self.as_raw()` is a valid pointer to a `struct sg_table`.
+        let nents = unsafe { (*self.as_raw()).nents };
+
+        let pos = if nents > 0 {
+            // SAFETY: `self.as_raw()` is a valid pointer to a `struct sg_table`.
+            let ptr = unsafe { (*self.as_raw()).sgl };
+
+            // SAFETY: `ptr` is guaranteed to be a valid pointer to a `struct scatterlist`.
+            Some(unsafe { SGEntry::from_raw(ptr) })
+        } else {
+            None
+        };
+
+        SGTableIter { pos, nents }
+    }
+}
+
+/// Represents the DMA mapping state of a `struct sg_table`.
+///
+/// This is used as an inner type of [`Owned`] to manage the DMA mapping lifecycle.
+///
+/// # Invariants
+///
+/// - `sgt` is a valid pointer to a `struct sg_table` for the entire lifetime of the
+///   [`DmaMappedSgt`].
+/// - `sgt` is always DMA mapped.
+struct DmaMappedSgt {
+    sgt: NonNull<bindings::sg_table>,
+    dev: ARef<Device>,
+    dir: dma::DataDirection,
+}
+
+// SAFETY: `DmaMappedSgt` can be sent to any task.
+unsafe impl Send for DmaMappedSgt {}
+
+// SAFETY: `DmaMappedSgt` has no interior mutability and can be accessed concurrently.
+unsafe impl Sync for DmaMappedSgt {}
+
+impl DmaMappedSgt {
+    /// # Safety
+    ///
+    /// - `sgt` must be a valid pointer to a `struct sg_table` for the entire lifetime of the
+    ///   returned [`DmaMappedSgt`].
+    /// - The caller must guarantee that `sgt` remains DMA mapped for the entire lifetime of
+    ///   [`DmaMappedSgt`].
+    unsafe fn new(
+        sgt: NonNull<bindings::sg_table>,
+        dev: &Device<Bound>,
+        dir: dma::DataDirection,
+    ) -> Result<Self> {
+        // SAFETY:
+        // - `dev.as_raw()` is a valid pointer to a `struct device`, which is guaranteed to be
+        //   bound to a driver for the duration of this call.
+        // - `sgt` is a valid pointer to a `struct sg_table`.
+        error::to_result(unsafe {
+            bindings::dma_map_sgtable(dev.as_raw(), sgt.as_ptr(), dir.into(), 0)
+        })?;
+
+        // INVARIANT: By the safety requirements of this function it is guaranteed that `sgt` is
+        // valid for the entire lifetime of this object instance.
+        Ok(Self {
+            sgt,
+            dev: dev.into(),
+            dir,
+        })
+    }
+}
+
+impl Drop for DmaMappedSgt {
+    #[inline]
+    fn drop(&mut self) {
+        // SAFETY:
+        // - `self.dev.as_raw()` is a pointer to a valid `struct device`.
+        // - `self.dev` is the same device the mapping has been created for in `Self::new()`.
+        // - `self.sgt.as_ptr()` is a valid pointer to a `struct sg_table` by the type invariants
+        //   of `Self`.
+        // - `self.dir` is the same `dma::DataDirection` the mapping has been created with in
+        //   `Self::new()`.
+        unsafe {
+            bindings::dma_unmap_sgtable(self.dev.as_raw(), self.sgt.as_ptr(), self.dir.into(), 0)
+        };
+    }
+}
+
+/// A transparent wrapper around a `struct sg_table`.
+///
+/// While we could also create the `struct sg_table` in the constructor of [`Owned`], we can't tear
+/// down the `struct sg_table` in [`Owned::drop`]; the drop order in [`Owned`] matters.
+#[repr(transparent)]
+struct RawSGTable(Opaque<bindings::sg_table>);
+
+// SAFETY: `RawSGTable` can be sent to any task.
+unsafe impl Send for RawSGTable {}
+
+// SAFETY: `RawSGTable` has no interior mutability and can be accessed concurrently.
+unsafe impl Sync for RawSGTable {}
+
+impl RawSGTable {
+    /// # Safety
+    ///
+    /// - `pages` must be a slice of valid `struct page *`.
+    /// - The pages pointed to by `pages` must remain valid for the entire lifetime of the returned
+    ///   [`RawSGTable`].
+    unsafe fn new(
+        pages: &mut [*mut bindings::page],
+        size: usize,
+        max_segment: u32,
+        flags: alloc::Flags,
+    ) -> Result<Self> {
+        // `sg_alloc_table_from_pages_segment()` expects at least one page, otherwise it
+        // produces a NPE.
+        if pages.is_empty() {
+            return Err(EINVAL);
+        }
+
+        let sgt = Opaque::zeroed();
+        // SAFETY:
+        // - `sgt.get()` is a valid pointer to uninitialized memory.
+        // - As by the check above, `pages` is not empty.
+        error::to_result(unsafe {
+            bindings::sg_alloc_table_from_pages_segment(
+                sgt.get(),
+                pages.as_mut_ptr(),
+                pages.len().try_into()?,
+                0,
+                size,
+                max_segment,
+                flags.as_raw(),
+            )
+        })?;
+
+        Ok(Self(sgt))
+    }
+
+    #[inline]
+    fn as_raw(&self) -> *mut bindings::sg_table {
+        self.0.get()
+    }
+}
+
+impl Drop for RawSGTable {
+    #[inline]
+    fn drop(&mut self) {
+        // SAFETY: `sgt` is a valid and initialized `struct sg_table`.
+        unsafe { bindings::sg_free_table(self.0.get()) };
+    }
+}
+
+/// The [`Owned`] generic type of an [`SGTable`].
+///
+/// A [`SGTable<Owned>`] signifies that the [`SGTable`] owns all associated resources:
+///
+/// - The backing memory pages.
+/// - The `struct sg_table` allocation (`sgt`).
+/// - The DMA mapping, managed through a [`Devres`]-managed `DmaMappedSgt`.
+///
+/// Users interact with this type through the [`SGTable`] handle and do not need to manage
+/// [`Owned`] directly.
+#[pin_data]
+pub struct Owned<P> {
+    // Note: The drop order is relevant; we first have to unmap the `struct sg_table`, then free the
+    // `struct sg_table` and finally free the backing pages.
+    #[pin]
+    dma: Devres<DmaMappedSgt>,
+    sgt: RawSGTable,
+    _pages: P,
+}
+
+// SAFETY: `Owned` can be sent to any task if `P` can be send to any task.
+unsafe impl<P: Send> Send for Owned<P> {}
+
+// SAFETY: `Owned` has no interior mutability and can be accessed concurrently if `P` can be
+// accessed concurrently.
+unsafe impl<P: Sync> Sync for Owned<P> {}
+
+impl<P> Owned<P>
+where
+    for<'a> P: page::AsPageIter<Iter<'a> = VmallocPageIter<'a>> + 'static,
+{
+    fn new(
+        dev: &Device<Bound>,
+        mut pages: P,
+        dir: dma::DataDirection,
+        flags: alloc::Flags,
+    ) -> Result<impl PinInit<Self, Error> + '_> {
+        let page_iter = pages.page_iter();
+        let size = page_iter.size();
+
+        let mut page_vec: KVec<*mut bindings::page> =
+            KVec::with_capacity(page_iter.page_count(), flags)?;
+
+        for page in page_iter {
+            page_vec.push(page.as_ptr(), flags)?;
+        }
+
+        // `dma_max_mapping_size` returns `size_t`, but `sg_alloc_table_from_pages_segment()` takes
+        // an `unsigned int`.
+        //
+        // SAFETY: `dev.as_raw()` is a valid pointer to a `struct device`.
+        let max_segment = match unsafe { bindings::dma_max_mapping_size(dev.as_raw()) } {
+            0 => u32::MAX,
+            max_segment => u32::try_from(max_segment).unwrap_or(u32::MAX),
+        };
+
+        Ok(try_pin_init!(&this in Self {
+            // SAFETY:
+            // - `page_vec` is a `KVec` of valid `struct page *` obtained from `pages`.
+            // - The pages contained in `pages` remain valid for the entire lifetime of the
+            //   `RawSGTable`.
+            sgt: unsafe { RawSGTable::new(&mut page_vec, size, max_segment, flags) }?,
+            dma <- {
+                // SAFETY: `this` is a valid pointer to uninitialized memory.
+                let sgt = unsafe { &raw mut (*this.as_ptr()).sgt }.cast();
+
+                // SAFETY: `sgt` is guaranteed to be non-null.
+                let sgt = unsafe { NonNull::new_unchecked(sgt) };
+
+                // SAFETY:
+                // - It is guaranteed that the object returned by `DmaMappedSgt::new` won't out-live
+                //   `sgt`.
+                // - `sgt` is never DMA unmapped manually.
+                Devres::new(dev, unsafe { DmaMappedSgt::new(sgt, dev, dir) })
+            },
+            _pages: pages,
+        }))
+    }
+}
+
+impl<P> SGTable<Owned<P>>
+where
+    for<'a> P: page::AsPageIter<Iter<'a> = VmallocPageIter<'a>> + 'static,
+{
+    /// Allocates a new scatter-gather table from the given pages and maps it for DMA.
+    ///
+    /// This constructor creates a new [`SGTable<Owned>`] that takes ownership of `P`.
+    /// It allocates a `struct sg_table`, populates it with entries corresponding to the physical
+    /// pages of `P`, and maps the table for DMA with the specified [`Device`] and
+    /// [`dma::DataDirection`].
+    ///
+    /// The DMA mapping is managed through [`Devres`], ensuring that the DMA mapping is unmapped
+    /// once the associated [`Device`] is unbound, or when the [`SGTable<Owned>`] is dropped.
+    ///
+    /// # Parameters
+    ///
+    /// * `dev`: The [`Device`] that will be performing the DMA.
+    /// * `pages`: The entity providing the backing pages. It must implement [`page::AsPageIter`].
+    ///   The ownership of this entity is moved into the new [`SGTable<Owned>`].
+    /// * `dir`: The [`dma::DataDirection`] of the DMA transfer.
+    /// * `flags`: Allocation flags for internal allocations (e.g., [`GFP_KERNEL`]).
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// use kernel::{
+    ///     device::{Bound, Device},
+    ///     dma, page,
+    ///     prelude::*,
+    ///     scatterlist::{SGTable, Owned},
+    /// };
+    ///
+    /// fn test(dev: &Device<Bound>) -> Result {
+    ///     let size = 4 * page::PAGE_SIZE;
+    ///     let pages = VVec::<u8>::with_capacity(size, GFP_KERNEL)?;
+    ///
+    ///     let sgt = KBox::pin_init(SGTable::new(
+    ///         dev,
+    ///         pages,
+    ///         dma::DataDirection::ToDevice,
+    ///         GFP_KERNEL,
+    ///     ), GFP_KERNEL)?;
+    ///
+    ///     Ok(())
+    /// }
+    /// ```
+    pub fn new(
+        dev: &Device<Bound>,
+        pages: P,
+        dir: dma::DataDirection,
+        flags: alloc::Flags,
+    ) -> impl PinInit<Self, Error> + '_ {
+        try_pin_init!(Self {
+            inner <- Owned::new(dev, pages, dir, flags)?
+        })
+    }
+}
+
+impl<P> Deref for SGTable<Owned<P>> {
+    type Target = SGTable;
+
+    #[inline]
+    fn deref(&self) -> &Self::Target {
+        // SAFETY:
+        // - `self.inner.sgt.as_raw()` is a valid pointer to a `struct sg_table` for the entire
+        //   lifetime of `self`.
+        // - The backing `struct sg_table` is not modified for the entire lifetime of `self`.
+        unsafe { SGTable::from_raw(self.inner.sgt.as_raw()) }
+    }
+}
+
+mod private {
+    pub trait Sealed {}
+
+    impl Sealed for super::Borrowed {}
+    impl<P> Sealed for super::Owned<P> {}
+}
+
+/// An [`Iterator`] over the DMA mapped [`SGEntry`] items of an [`SGTable`].
+///
+/// Note that the existence of an [`SGTableIter`] does not guarantee that the [`SGEntry`] items
+/// actually remain DMA mapped; they are prone to be unmapped on device unbind.
+pub struct SGTableIter<'a> {
+    pos: Option<&'a SGEntry>,
+    /// The number of DMA mapped entries in a `struct sg_table`.
+    nents: c_uint,
+}
+
+impl<'a> Iterator for SGTableIter<'a> {
+    type Item = &'a SGEntry;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        let entry = self.pos?;
+        self.nents = self.nents.saturating_sub(1);
+
+        // SAFETY: `entry.as_raw()` is a valid pointer to a `struct scatterlist`.
+        let next = unsafe { bindings::sg_next(entry.as_raw()) };
+
+        self.pos = (!next.is_null() && self.nents > 0).then(|| {
+            // SAFETY: If `next` is not NULL, `sg_next()` guarantees to return a valid pointer to
+            // the next `struct scatterlist`.
+            unsafe { SGEntry::from_raw(next) }
+        });
+
+        Some(entry)
+    }
+}

From 5444799d701cbf3c4b18c88fef8034b35d159615 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Thu, 28 Aug 2025 15:32:17 +0200
Subject: [PATCH 0938/1455] samples: rust: dma: add sample code for SGTable

Add sample code for allocating and mapping a scatter-gather table
(`SGTable`).

Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Lyude Paul <lyude@redhat.com>
Co-developed-by: Abdiel Janulgue <abdiel.janulgue@gmail.com>
Signed-off-by: Abdiel Janulgue <abdiel.janulgue@gmail.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250828133323.53311-5-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 samples/rust/rust_dma.rs | 35 ++++++++++++++++++++++++++---------
 1 file changed, 26 insertions(+), 9 deletions(-)

diff --git a/samples/rust/rust_dma.rs b/samples/rust/rust_dma.rs
index c5e7cce68654..04007e29fd85 100644
--- a/samples/rust/rust_dma.rs
+++ b/samples/rust/rust_dma.rs
@@ -7,15 +7,19 @@
 use kernel::{
     bindings,
     device::Core,
-    dma::{CoherentAllocation, Device, DmaMask},
-    pci,
+    dma::{CoherentAllocation, DataDirection, Device, DmaMask},
+    page, pci,
     prelude::*,
+    scatterlist::{Owned, SGTable},
     types::ARef,
 };
 
+#[pin_data(PinnedDrop)]
 struct DmaSampleDriver {
     pdev: ARef<pci::Device>,
     ca: CoherentAllocation<MyStruct>,
+    #[pin]
+    sgt: SGTable<Owned<VVec<u8>>>,
 }
 
 const TEST_VALUES: [(u32, u32); 5] = [
@@ -70,21 +74,30 @@ impl pci::Driver for DmaSampleDriver {
             kernel::dma_write!(ca[i] = MyStruct::new(value.0, value.1))?;
         }
 
-        let drvdata = KBox::new(
-            Self {
+        let size = 4 * page::PAGE_SIZE;
+        let pages = VVec::with_capacity(size, GFP_KERNEL)?;
+
+        let sgt = SGTable::new(pdev.as_ref(), pages, DataDirection::ToDevice, GFP_KERNEL);
+
+        let drvdata = KBox::pin_init(
+            try_pin_init!(Self {
                 pdev: pdev.into(),
                 ca,
-            },
+                sgt <- sgt,
+            }),
             GFP_KERNEL,
         )?;
 
-        Ok(drvdata.into())
+        Ok(drvdata)
     }
 }
 
-impl Drop for DmaSampleDriver {
-    fn drop(&mut self) {
-        dev_info!(self.pdev.as_ref(), "Unload DMA test driver.\n");
+#[pinned_drop]
+impl PinnedDrop for DmaSampleDriver {
+    fn drop(self: Pin<&mut Self>) {
+        let dev = self.pdev.as_ref();
+
+        dev_info!(dev, "Unload DMA test driver.\n");
 
         for (i, value) in TEST_VALUES.into_iter().enumerate() {
             let val0 = kernel::dma_read!(self.ca[i].h);
@@ -99,6 +112,10 @@ impl Drop for DmaSampleDriver {
                 assert_eq!(val1, value.1);
             }
         }
+
+        for (i, entry) in self.sgt.iter().enumerate() {
+            dev_info!(dev, "Entry[{}]: DMA address: {:#x}", i, entry.dma_address());
+        }
     }
 }
 

From c58466b85b16f139a4afc6163aa57d7445dae806 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Thu, 28 Aug 2025 15:32:18 +0200
Subject: [PATCH 0939/1455] MAINTAINERS: rust: dma: add scatterlist files

Rename the "DMA MAPPING HELPERS DEVICE DRIVER API [RUST]" maintainers
entry to "DMA MAPPING & SCATTERLIST API [RUST]" and add the
corresponding scatterlist files.

Reviewed-by: Lyude Paul <lyude@redhat.com>
Acked-by: Miguel Ojeda <ojeda@kernel.org>
Link: https://lore.kernel.org/r/20250828133323.53311-6-dakr@kernel.org
Signed-off-by: Danilo Krummrich <dakr@kernel.org>
---
 MAINTAINERS | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 936650c7eff9..b97760467f09 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7238,7 +7238,7 @@ F:	include/linux/dma-mapping.h
 F:	include/linux/swiotlb.h
 F:	kernel/dma/
 
-DMA MAPPING HELPERS DEVICE DRIVER API [RUST]
+DMA MAPPING & SCATTERLIST API [RUST]
 M:	Abdiel Janulgue <abdiel.janulgue@gmail.com>
 M:	Danilo Krummrich <dakr@kernel.org>
 R:	Daniel Almeida <daniel.almeida@collabora.com>
@@ -7249,7 +7249,9 @@ S:	Supported
 W:	https://rust-for-linux.com
 T:	git https://github.com/Rust-for-Linux/linux.git alloc-next
 F:	rust/helpers/dma.c
+F:	rust/helpers/scatterlist.c
 F:	rust/kernel/dma.rs
+F:	rust/kernel/scatterlist.rs
 F:	samples/rust/rust_dma.rs
 
 DMA-BUF HEAPS FRAMEWORK

From 4c67b73907214994f87cad795195c46fe63c1e1c Mon Sep 17 00:00:00 2001
From: Min Ma <mamin506@gmail.com>
Date: Sat, 30 Aug 2025 17:12:28 -0700
Subject: [PATCH 0940/1455] MAINTAINERS: Update Min Ma's email for AMD XDNA
 driver

I recently left AMD and would like to continue participating in
the review and maintenance of the XDNA driver using my personal
email address.

Signed-off-by: Min Ma <mamin506@gmail.com>
Reviewed-by: Lizhi Hou <lizhi.hou@amd.com>
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Link: https://lore.kernel.org/r/20250831001228.592-1-mamin506@gmail.com
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 88851907b672..c6f78b3702f6 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1243,7 +1243,7 @@ F:	drivers/spi/spi-amd.c
 F:	drivers/spi/spi-amd.h
 
 AMD XDNA DRIVER
-M:	Min Ma <min.ma@amd.com>
+M:	Min Ma <mamin506@gmail.com>
 M:	Lizhi Hou <lizhi.hou@amd.com>
 L:	dri-devel@lists.freedesktop.org
 S:	Supported

From 3b5b2567f8d2e16f7200ea36f62e00bf79248bfb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 29 Aug 2025 08:39:25 +0300
Subject: [PATCH 0941/1455] drm/i915/alpm: Calculate silence period
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Calculate silence period instead of hardcoding it in switch case.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Animesh Manna <animesh.manna@intel.com>
Link: https://lore.kernel.org/r/20250829053929.3585636-2-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_alpm.c | 37 +++++++++++------------
 1 file changed, 17 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c
index dfdde8e4eabe..e48a4218c163 100644
--- a/drivers/gpu/drm/i915/display/intel_alpm.c
+++ b/drivers/gpu/drm/i915/display/intel_alpm.c
@@ -16,6 +16,12 @@
 #include "intel_psr.h"
 #include "intel_psr_regs.h"
 
+#define SILENCE_PERIOD_MIN_TIME	80
+#define SILENCE_PERIOD_MAX_TIME	180
+#define SILENCE_PERIOD_TIME	(SILENCE_PERIOD_MIN_TIME +	\
+				(SILENCE_PERIOD_MAX_TIME -	\
+				 SILENCE_PERIOD_MIN_TIME) / 2)
+
 bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp)
 {
 	return intel_dp->alpm_dpcd & DP_ALPM_CAP;
@@ -44,11 +50,15 @@ void intel_alpm_init(struct intel_dp *intel_dp)
 	mutex_init(&intel_dp->alpm_parameters.lock);
 }
 
+static int get_silence_period_symbols(const struct intel_crtc_state *crtc_state)
+{
+	return SILENCE_PERIOD_TIME * intel_dp_link_symbol_clock(crtc_state->port_clock) /
+		1000 / 1000;
+}
+
 /*
  * See Bspec: 71632 for the table
  *
- * Silence_period = tSilence,Min + ((tSilence,Max - tSilence,Min) / 2)
- *
  * Half cycle duration:
  *
  * Link rates 1.62 - 4.32 and tLFPS_Cycle = 70 ns
@@ -60,53 +70,41 @@ void intel_alpm_init(struct intel_dp *intel_dp)
  * FLOOR( LFPS Period in Symbol clocks /
  * (2 * PORT_ALPM_LFPS_CTL[ LFPS Cycle Count ]) )
  */
-static bool _lnl_get_silence_period_and_lfps_half_cycle(int link_rate,
-							int *silence_period,
-							int *lfps_half_cycle)
+static bool _lnl_get_lfps_half_cycle(int link_rate, int *lfps_half_cycle)
 {
 	switch (link_rate) {
 	case 162000:
-		*silence_period = 20;
 		*lfps_half_cycle = 5;
 		break;
 	case 216000:
-		*silence_period = 27;
 		*lfps_half_cycle = 7;
 		break;
 	case 243000:
-		*silence_period = 31;
 		*lfps_half_cycle = 8;
 		break;
 	case 270000:
-		*silence_period = 34;
 		*lfps_half_cycle = 9;
 		break;
 	case 324000:
-		*silence_period = 41;
 		*lfps_half_cycle = 11;
 		break;
 	case 432000:
-		*silence_period = 56;
 		*lfps_half_cycle = 15;
 		break;
 	case 540000:
-		*silence_period = 69;
 		*lfps_half_cycle = 12;
 		break;
 	case 648000:
-		*silence_period = 84;
 		*lfps_half_cycle = 15;
 		break;
 	case 675000:
-		*silence_period = 87;
 		*lfps_half_cycle = 15;
 		break;
 	case 810000:
-		*silence_period = 104;
 		*lfps_half_cycle = 19;
 		break;
 	default:
-		*silence_period = *lfps_half_cycle = -1;
+		*lfps_half_cycle = -1;
 		return false;
 	}
 	return true;
@@ -160,10 +158,9 @@ _lnl_compute_aux_less_alpm_params(struct intel_dp *intel_dp,
 		_lnl_compute_aux_less_wake_time(crtc_state->port_clock);
 	aux_less_wake_lines = intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode,
 						       aux_less_wake_time);
-
-	if (!_lnl_get_silence_period_and_lfps_half_cycle(crtc_state->port_clock,
-							 &silence_period,
-							 &lfps_half_cycle))
+	silence_period = get_silence_period_symbols(crtc_state);
+	if (!_lnl_get_lfps_half_cycle(crtc_state->port_clock,
+				      &lfps_half_cycle))
 		return false;
 
 	if (aux_less_wake_lines > ALPM_CTL_AUX_LESS_WAKE_TIME_MASK ||

From e9c62c8654875bb583fa07701e3921540550a379 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 29 Aug 2025 08:39:26 +0300
Subject: [PATCH 0942/1455] drm/i915/alpm: Add own define for LFPS count
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add own define for LFPS count and use it for the configuration. This new
define will be used for calculating ALPM parameters as well.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Animesh Manna <animesh.manna@intel.com>
Link: https://lore.kernel.org/r/20250829053929.3585636-3-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_alpm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c
index e48a4218c163..ae556a885c2a 100644
--- a/drivers/gpu/drm/i915/display/intel_alpm.c
+++ b/drivers/gpu/drm/i915/display/intel_alpm.c
@@ -22,6 +22,8 @@
 				(SILENCE_PERIOD_MAX_TIME -	\
 				 SILENCE_PERIOD_MIN_TIME) / 2)
 
+#define LFPS_CYCLE_COUNT 10
+
 bool intel_alpm_aux_wake_supported(struct intel_dp *intel_dp)
 {
 	return intel_dp->alpm_dpcd & DP_ALPM_CAP;
@@ -403,7 +405,7 @@ void intel_alpm_port_configure(struct intel_dp *intel_dp,
 			PORT_ALPM_CTL_MAX_PHY_SWING_HOLD(0) |
 			PORT_ALPM_CTL_SILENCE_PERIOD(
 				intel_dp->alpm_parameters.silence_period_sym_clocks);
-		lfps_ctl_val = PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT(10) |
+		lfps_ctl_val = PORT_ALPM_LFPS_CTL_LFPS_CYCLE_COUNT(LFPS_CYCLE_COUNT) |
 			PORT_ALPM_LFPS_CTL_LFPS_HALF_CYCLE_DURATION(
 				intel_dp->alpm_parameters.lfps_half_cycle_num_of_syms) |
 			PORT_ALPM_LFPS_CTL_FIRST_LFPS_HALF_CYCLE_DURATION(

From cc2189a97848a41d107c03da6ddd93d94ad27e77 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 29 Aug 2025 08:39:27 +0300
Subject: [PATCH 0943/1455] drm/i915/alpm: Replace hardcoded LFPS cycle with
 proper calculation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently LFPS is hadcoded for different port clocks. Replace this with
proper calculation.

v2: replace hardcoded 20 with 2 * LFPS_CYCLE_COUNT

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Animesh Manna <animesh.manna@intel.com>
Link: https://lore.kernel.org/r/20250829053929.3585636-4-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_alpm.c | 90 ++++++++++-------------
 1 file changed, 38 insertions(+), 52 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c
index ae556a885c2a..b061f7165bba 100644
--- a/drivers/gpu/drm/i915/display/intel_alpm.c
+++ b/drivers/gpu/drm/i915/display/intel_alpm.c
@@ -58,58 +58,43 @@ static int get_silence_period_symbols(const struct intel_crtc_state *crtc_state)
 		1000 / 1000;
 }
 
-/*
- * See Bspec: 71632 for the table
- *
- * Half cycle duration:
- *
- * Link rates 1.62 - 4.32 and tLFPS_Cycle = 70 ns
- * FLOOR( (Link Rate * tLFPS_Cycle) / (2 * 10) )
- *
- * Link rates 5.4 - 8.1
- * PORT_ALPM_LFPS_CTL[ LFPS Cycle Count ] = 10
- * LFPS Period chosen is the mid-point of the min:max values from the table
- * FLOOR( LFPS Period in Symbol clocks /
- * (2 * PORT_ALPM_LFPS_CTL[ LFPS Cycle Count ]) )
- */
-static bool _lnl_get_lfps_half_cycle(int link_rate, int *lfps_half_cycle)
+static int get_lfps_cycle_min_max_time(const struct intel_crtc_state *crtc_state,
+				       int *min, int *max)
 {
-	switch (link_rate) {
-	case 162000:
-		*lfps_half_cycle = 5;
-		break;
-	case 216000:
-		*lfps_half_cycle = 7;
-		break;
-	case 243000:
-		*lfps_half_cycle = 8;
-		break;
-	case 270000:
-		*lfps_half_cycle = 9;
-		break;
-	case 324000:
-		*lfps_half_cycle = 11;
-		break;
-	case 432000:
-		*lfps_half_cycle = 15;
-		break;
-	case 540000:
-		*lfps_half_cycle = 12;
-		break;
-	case 648000:
-		*lfps_half_cycle = 15;
-		break;
-	case 675000:
-		*lfps_half_cycle = 15;
-		break;
-	case 810000:
-		*lfps_half_cycle = 19;
-		break;
-	default:
-		*lfps_half_cycle = -1;
-		return false;
+	if (crtc_state->port_clock < 540000) {
+		*min = 65 * LFPS_CYCLE_COUNT;
+		*max = 75 * LFPS_CYCLE_COUNT;
+	} else if (crtc_state->port_clock <= 810000) {
+		*min = 140;
+		*max = 800;
+	} else {
+		*min = *max = -1;
+		return -1;
 	}
-	return true;
+
+	return 0;
+}
+
+static int get_lfps_cycle_time(const struct intel_crtc_state *crtc_state)
+{
+	int tlfps_cycle_min, tlfps_cycle_max, ret;
+
+	ret = get_lfps_cycle_min_max_time(crtc_state, &tlfps_cycle_min,
+					  &tlfps_cycle_max);
+	if (ret)
+		return ret;
+
+	return tlfps_cycle_min +  (tlfps_cycle_max - tlfps_cycle_min) / 2;
+}
+
+static int get_lfps_half_cycle_clocks(const struct intel_crtc_state *crtc_state)
+{
+	int lfps_cycle_time = get_lfps_cycle_time(crtc_state);
+
+	if (lfps_cycle_time < 0)
+		return -1;
+
+	return lfps_cycle_time * crtc_state->port_clock / 1000 / 1000 / (2 * LFPS_CYCLE_COUNT);
 }
 
 /*
@@ -161,8 +146,9 @@ _lnl_compute_aux_less_alpm_params(struct intel_dp *intel_dp,
 	aux_less_wake_lines = intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode,
 						       aux_less_wake_time);
 	silence_period = get_silence_period_symbols(crtc_state);
-	if (!_lnl_get_lfps_half_cycle(crtc_state->port_clock,
-				      &lfps_half_cycle))
+
+	lfps_half_cycle = get_lfps_half_cycle_clocks(crtc_state);
+	if (lfps_half_cycle < 0)
 		return false;
 
 	if (aux_less_wake_lines > ALPM_CTL_AUX_LESS_WAKE_TIME_MASK ||

From d074a40b8828fe29c2bab59d0c4a26c940af7e88 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 29 Aug 2025 08:39:28 +0300
Subject: [PATCH 0944/1455] drm/i915/alpm: Use actual lfps cycle and silence
 periods in wake time
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we are using maximum lfps cycle and silence period times when
calculating AUXLess wake time. Use actual values instead.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Animesh Manna <animesh.manna@intel.com>
Link: https://lore.kernel.org/r/20250829053929.3585636-5-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_alpm.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_alpm.c b/drivers/gpu/drm/i915/display/intel_alpm.c
index b061f7165bba..ed7a7ed486b5 100644
--- a/drivers/gpu/drm/i915/display/intel_alpm.c
+++ b/drivers/gpu/drm/i915/display/intel_alpm.c
@@ -116,21 +116,19 @@ static int get_lfps_half_cycle_clocks(const struct intel_crtc_state *crtc_state)
  * tML_PHY_LOCK = TPS4 Length * ( 10 / (Link Rate in MHz) )
  * TPS4 Length = 252 Symbols
  */
-static int _lnl_compute_aux_less_wake_time(int port_clock)
+static int _lnl_compute_aux_less_wake_time(const struct intel_crtc_state *crtc_state)
 {
 	int tphy2_p2_to_p0 = 12 * 1000;
-	int tlfps_period_max = 800;
-	int tsilence_max = 180;
 	int t1 = 50 * 1000;
 	int tps4 = 252;
 	/* port_clock is link rate in 10kbit/s units */
-	int tml_phy_lock = 1000 * 1000 * tps4 / port_clock;
+	int tml_phy_lock = 1000 * 1000 * tps4 / crtc_state->port_clock;
 	int num_ml_phy_lock = 7 + DIV_ROUND_UP(6500, tml_phy_lock) + 1;
 	int t2 = num_ml_phy_lock * tml_phy_lock;
 	int tcds = 1 * t2;
 
-	return DIV_ROUND_UP(tphy2_p2_to_p0 + tlfps_period_max + tsilence_max +
-			    t1 + tcds, 1000);
+	return DIV_ROUND_UP(tphy2_p2_to_p0 + get_lfps_cycle_time(crtc_state) +
+			    SILENCE_PERIOD_TIME + t1 + tcds, 1000);
 }
 
 static int
@@ -142,7 +140,7 @@ _lnl_compute_aux_less_alpm_params(struct intel_dp *intel_dp,
 		lfps_half_cycle;
 
 	aux_less_wake_time =
-		_lnl_compute_aux_less_wake_time(crtc_state->port_clock);
+		_lnl_compute_aux_less_wake_time(crtc_state);
 	aux_less_wake_lines = intel_usecs_to_scanlines(&crtc_state->hw.adjusted_mode,
 						       aux_less_wake_time);
 	silence_period = get_silence_period_symbols(crtc_state);

From 2c907d852c57551ca5481409da33b67fd3800af3 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Fri, 29 Aug 2025 15:53:14 +0800
Subject: [PATCH 0945/1455] drm/panel: lvds: Remove unused members from main
 structure

Since commit 03fa454bb666 ("drm/panel: lvds: Simplify mode parsing"),
the width and height members of struct panel_lvds are no longer used.
Remove them.  No functional change.

Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Link: https://lore.kernel.org/r/20250829-panel-lvds-remove-width-height-v1-1-acecf0c84dc4@nxp.com
---
 drivers/gpu/drm/panel/panel-lvds.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/panel/panel-lvds.c b/drivers/gpu/drm/panel/panel-lvds.c
index 23fd535d8f47..46b07f38559f 100644
--- a/drivers/gpu/drm/panel/panel-lvds.c
+++ b/drivers/gpu/drm/panel/panel-lvds.c
@@ -28,8 +28,6 @@ struct panel_lvds {
 	struct device *dev;
 
 	const char *label;
-	unsigned int width;
-	unsigned int height;
 	struct drm_display_mode dmode;
 	u32 bus_flags;
 	unsigned int bus_format;

From 6ee8adf124102aa679d13b8c3cbe391f544ba90a Mon Sep 17 00:00:00 2001
From: Harish Chegondi <harish.chegondi@intel.com>
Date: Tue, 2 Sep 2025 16:41:59 -0700
Subject: [PATCH 0946/1455] drm/xe/mcr: Make xe_gt_mcr_get_dss_steering() input
 gt a const

Make gt, input parameter to xe_gt_mcr_get_dss_steering(), a
constant. This would allow xe_gt_mcr_get_dss_steering() to
be called from functions that have gt as const to struct xe_gt.

Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Link: https://lore.kernel.org/r/9dc621a90880f62ac8e2951afea7952277f7eb0e.1756855529.git.harish.chegondi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_mcr.c | 2 +-
 drivers/gpu/drm/xe/xe_gt_mcr.h | 3 ++-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.c b/drivers/gpu/drm/xe/xe_gt_mcr.c
index 683ac021a06d..8fb1cae91724 100644
--- a/drivers/gpu/drm/xe/xe_gt_mcr.c
+++ b/drivers/gpu/drm/xe/xe_gt_mcr.c
@@ -362,7 +362,7 @@ fallback:
  * @group: pointer to storage for steering group ID
  * @instance: pointer to storage for steering instance ID
  */
-void xe_gt_mcr_get_dss_steering(struct xe_gt *gt, unsigned int dss, u16 *group, u16 *instance)
+void xe_gt_mcr_get_dss_steering(const struct xe_gt *gt, unsigned int dss, u16 *group, u16 *instance)
 {
 	xe_gt_assert(gt, dss < XE_MAX_DSS_FUSE_BITS);
 
diff --git a/drivers/gpu/drm/xe/xe_gt_mcr.h b/drivers/gpu/drm/xe/xe_gt_mcr.h
index bc06520befab..283a1c9770e2 100644
--- a/drivers/gpu/drm/xe/xe_gt_mcr.h
+++ b/drivers/gpu/drm/xe/xe_gt_mcr.h
@@ -31,7 +31,8 @@ bool xe_gt_mcr_get_nonterminated_steering(struct xe_gt *gt,
 					  u8 *group, u8 *instance);
 
 void xe_gt_mcr_steering_dump(struct xe_gt *gt, struct drm_printer *p);
-void xe_gt_mcr_get_dss_steering(struct xe_gt *gt, unsigned int dss, u16 *group, u16 *instance);
+void xe_gt_mcr_get_dss_steering(const struct xe_gt *gt,
+				unsigned int dss, u16 *group, u16 *instance);
 u32 xe_gt_mcr_steering_info_to_dss_id(struct xe_gt *gt, u16 group, u16 instance);
 
 /*

From 5952d80514054d12d2ec72aa111c2e09eb70e881 Mon Sep 17 00:00:00 2001
From: Harish Chegondi <harish.chegondi@intel.com>
Date: Tue, 2 Sep 2025 16:42:00 -0700
Subject: [PATCH 0947/1455] drm/xe/xe2hpg: Add Wa_18041344222 for Xe2_HPG

Add Wa_18041344222 for Xe2_HPG that requires disabling
the perf mode for subslice count for eustall sampling
when the enabled slices are discontiguous.

Bspec: 79483, 56024
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/b6a631a13a9fb7360e89d679e0797fae42d5a09e.1756855529.git.harish.chegondi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_gt_regs.h |  1 +
 drivers/gpu/drm/xe/xe_gt_topology.c  | 17 +++++++++++++++++
 drivers/gpu/drm/xe/xe_gt_topology.h  |  2 ++
 drivers/gpu/drm/xe/xe_rtp.c          |  6 ++++++
 drivers/gpu/drm/xe/xe_rtp.h          |  3 +++
 drivers/gpu/drm/xe/xe_wa.c           |  7 +++++++
 6 files changed, 36 insertions(+)

diff --git a/drivers/gpu/drm/xe/regs/xe_gt_regs.h b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
index f96b2e2b3064..06cb6b02ec64 100644
--- a/drivers/gpu/drm/xe/regs/xe_gt_regs.h
+++ b/drivers/gpu/drm/xe/regs/xe_gt_regs.h
@@ -522,6 +522,7 @@
 
 #define TDL_CHICKEN				XE_REG_MCR(0xe5f4, XE_REG_OPTION_MASKED)
 #define   QID_WAIT_FOR_THREAD_NOT_RUN_DISABLE	REG_BIT(12)
+#define   EUSTALL_PERF_SAMPLING_DISABLE		REG_BIT(5)
 
 #define LSC_CHICKEN_BIT_0			XE_REG_MCR(0xe7c8)
 #define   DISABLE_D8_D16_COASLESCE		REG_BIT(30)
diff --git a/drivers/gpu/drm/xe/xe_gt_topology.c b/drivers/gpu/drm/xe/xe_gt_topology.c
index a0baa560dd71..0ed7dc9044a5 100644
--- a/drivers/gpu/drm/xe/xe_gt_topology.c
+++ b/drivers/gpu/drm/xe/xe_gt_topology.c
@@ -12,6 +12,7 @@
 #include "regs/xe_gt_regs.h"
 #include "xe_assert.h"
 #include "xe_gt.h"
+#include "xe_gt_mcr.h"
 #include "xe_gt_printk.h"
 #include "xe_mmio.h"
 #include "xe_wa.h"
@@ -328,3 +329,19 @@ bool xe_gt_has_compute_dss(struct xe_gt *gt, unsigned int dss)
 {
 	return test_bit(dss, gt->fuse_topo.c_dss_mask);
 }
+
+bool xe_gt_has_discontiguous_dss_groups(const struct xe_gt *gt)
+{
+	unsigned int xecore;
+	int last_group = -1;
+	u16 group, instance;
+
+	for_each_dss_steering(xecore, gt, group, instance) {
+		if (last_group != group) {
+			if (group - last_group > 1)
+				return true;
+			last_group = group;
+		}
+	}
+	return false;
+}
diff --git a/drivers/gpu/drm/xe/xe_gt_topology.h b/drivers/gpu/drm/xe/xe_gt_topology.h
index c8140704ad4c..d95cdd6e45be 100644
--- a/drivers/gpu/drm/xe/xe_gt_topology.h
+++ b/drivers/gpu/drm/xe/xe_gt_topology.h
@@ -47,4 +47,6 @@ xe_gt_topology_has_dss_in_quadrant(struct xe_gt *gt, int quad);
 bool xe_gt_has_geometry_dss(struct xe_gt *gt, unsigned int dss);
 bool xe_gt_has_compute_dss(struct xe_gt *gt, unsigned int dss);
 
+bool xe_gt_has_discontiguous_dss_groups(const struct xe_gt *gt);
+
 #endif /* _XE_GT_TOPOLOGY_H_ */
diff --git a/drivers/gpu/drm/xe/xe_rtp.c b/drivers/gpu/drm/xe/xe_rtp.c
index 47ea1521dc80..b5f430d59f80 100644
--- a/drivers/gpu/drm/xe/xe_rtp.c
+++ b/drivers/gpu/drm/xe/xe_rtp.c
@@ -370,3 +370,9 @@ bool xe_rtp_match_psmi_enabled(const struct xe_gt *gt,
 {
 	return xe_configfs_get_psmi_enabled(to_pci_dev(gt_to_xe(gt)->drm.dev));
 }
+
+bool xe_rtp_match_gt_has_discontiguous_dss_groups(const struct xe_gt *gt,
+						  const struct xe_hw_engine *hwe)
+{
+	return xe_gt_has_discontiguous_dss_groups(gt);
+}
diff --git a/drivers/gpu/drm/xe/xe_rtp.h b/drivers/gpu/drm/xe/xe_rtp.h
index 7951fefdbe04..ac12ddf6cde6 100644
--- a/drivers/gpu/drm/xe/xe_rtp.h
+++ b/drivers/gpu/drm/xe/xe_rtp.h
@@ -480,4 +480,7 @@ bool xe_rtp_match_not_sriov_vf(const struct xe_gt *gt,
 bool xe_rtp_match_psmi_enabled(const struct xe_gt *gt,
 			       const struct xe_hw_engine *hwe);
 
+bool xe_rtp_match_gt_has_discontiguous_dss_groups(const struct xe_gt *gt,
+						  const struct xe_hw_engine *hwe);
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index e53edf1e5100..c1fec526bed0 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -612,6 +612,13 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 		       FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(TDL_TSL_CHICKEN, RES_CHK_SPR_DIS))
 	},
+	{ XE_RTP_NAME("18041344222"),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(2001, 2002),
+		       FUNC(xe_rtp_match_first_render_or_compute),
+		       FUNC(xe_rtp_match_not_sriov_vf),
+		       FUNC(xe_rtp_match_gt_has_discontiguous_dss_groups)),
+	  XE_RTP_ACTIONS(SET(TDL_CHICKEN, EUSTALL_PERF_SAMPLING_DISABLE))
+	},
 
 	/* Xe2_LPM */
 

From c50729c68aaf93611c855752b00e49ce1fdd1558 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:32 +0100
Subject: [PATCH 0948/1455] drm/gpusvm: fix hmm_pfn_to_map_order() usage
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Handle the case where the hmm range partially covers a huge page (like
2M), otherwise we can potentially end up doing something nasty like
mapping memory which is outside the range, and maybe not even mapped by
the mm. Fix is based on the xe userptr code, which in a future patch
will directly use gpusvm, so needs alignment here.

v2:
  - Add kernel-doc (Matt B)
  - s/fls/ilog2/ (Thomas)

Reported-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-11-matthew.auld@intel.com
---
 drivers/gpu/drm/drm_gpusvm.c | 33 +++++++++++++++++++++++++++++++--
 1 file changed, 31 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index e2a9a6ae1d54..49f64d26625d 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -635,6 +635,35 @@ drm_gpusvm_range_alloc(struct drm_gpusvm *gpusvm,
 	return range;
 }
 
+/**
+ * drm_gpusvm_hmm_pfn_to_order() - Get the largest CPU mapping order.
+ * @hmm_pfn: The current hmm_pfn.
+ * @hmm_pfn_index: Index of the @hmm_pfn within the pfn array.
+ * @npages: Number of pages within the pfn array i.e the hmm range size.
+ *
+ * To allow skipping PFNs with the same flags (like when they belong to
+ * the same huge PTE) when looping over the pfn array, take a given a hmm_pfn,
+ * and return the largest order that will fit inside the CPU PTE, but also
+ * crucially accounting for the original hmm range boundaries.
+ *
+ * Return: The largest order that will safely fit within the size of the hmm_pfn
+ * CPU PTE.
+ */
+static unsigned int drm_gpusvm_hmm_pfn_to_order(unsigned long hmm_pfn,
+						unsigned long hmm_pfn_index,
+						unsigned long npages)
+{
+	unsigned long size;
+
+	size = 1UL << hmm_pfn_to_map_order(hmm_pfn);
+	size -= (hmm_pfn & ~HMM_PFN_FLAGS) & (size - 1);
+	hmm_pfn_index += size;
+	if (hmm_pfn_index > npages)
+		size -= (hmm_pfn_index - npages);
+
+	return ilog2(size);
+}
+
 /**
  * drm_gpusvm_check_pages() - Check pages
  * @gpusvm: Pointer to the GPU SVM structure
@@ -693,7 +722,7 @@ static bool drm_gpusvm_check_pages(struct drm_gpusvm *gpusvm,
 			err = -EFAULT;
 			goto err_free;
 		}
-		i += 0x1 << hmm_pfn_to_map_order(pfns[i]);
+		i += 0x1 << drm_gpusvm_hmm_pfn_to_order(pfns[i], i, npages);
 	}
 
 err_free:
@@ -1270,7 +1299,7 @@ map_pages:
 	for (i = 0, j = 0; i < npages; ++j) {
 		struct page *page = hmm_pfn_to_page(pfns[i]);
 
-		order = hmm_pfn_to_map_order(pfns[i]);
+		order = drm_gpusvm_hmm_pfn_to_order(pfns[i], i, npages);
 		if (is_device_private_page(page) ||
 		    is_device_coherent_page(page)) {
 			if (zdd != page->zone_device_data && i > 0) {

From ad70e289ed4b1f87d4f158fd1ad5a3225256929d Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:33 +0100
Subject: [PATCH 0949/1455] drm/gpusvm: use more selective dma dir in
 get_pages()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If we are only reading the memory then from the device pov the direction
can be DMA_TO_DEVICE. This aligns with the xe-userptr code. Using the
most restrictive data direction to represent the access is normally a
good idea.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-12-matthew.auld@intel.com
---
 drivers/gpu/drm/drm_gpusvm.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index 49f64d26625d..b2920b8551f5 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -1224,6 +1224,8 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
 	struct dev_pagemap *pagemap;
 	struct drm_pagemap *dpagemap;
 	struct drm_gpusvm_range_flags flags;
+	enum dma_data_direction dma_dir = ctx->read_only ? DMA_TO_DEVICE :
+							   DMA_BIDIRECTIONAL;
 
 retry:
 	hmm_range.notifier_seq = mmu_interval_read_begin(notifier);
@@ -1329,7 +1331,7 @@ map_pages:
 				dpagemap->ops->device_map(dpagemap,
 							  gpusvm->drm->dev,
 							  page, order,
-							  DMA_BIDIRECTIONAL);
+							  dma_dir);
 			if (dma_mapping_error(gpusvm->drm->dev,
 					      range->dma_addr[j].addr)) {
 				err = -EFAULT;
@@ -1351,7 +1353,7 @@ map_pages:
 			addr = dma_map_page(gpusvm->drm->dev,
 					    page, 0,
 					    PAGE_SIZE << order,
-					    DMA_BIDIRECTIONAL);
+					    dma_dir);
 			if (dma_mapping_error(gpusvm->drm->dev, addr)) {
 				err = -EFAULT;
 				goto err_unmap;
@@ -1359,7 +1361,7 @@ map_pages:
 
 			range->dma_addr[j] = drm_pagemap_addr_encode
 				(addr, DRM_INTERCONNECT_SYSTEM, order,
-				 DMA_BIDIRECTIONAL);
+				 dma_dir);
 		}
 		i += 1 << order;
 		num_dma_mapped = i;

From f70da6f99d4f40c5f481c92e3b65d5e36eaa6dc9 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:34 +0100
Subject: [PATCH 0950/1455] drm/gpusvm: pull out drm_gpusvm_pages substructure
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pull the pages stuff from the svm range into its own substructure, with
the idea of having the main pages related routines, like get_pages(),
unmap_pages() and free_pages() all operating on some lower level
structures, which can then be re-used for stuff like userptr.

v2:
  - Move seq into pages struct (Matt B)
v3:
  - Small kernel-doc fixes

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-13-matthew.auld@intel.com
---
 drivers/gpu/drm/drm_gpusvm.c | 67 +++++++++++++++++++-----------------
 drivers/gpu/drm/xe/xe_pt.c   |  2 +-
 drivers/gpu/drm/xe/xe_svm.c  | 16 ++++-----
 drivers/gpu/drm/xe/xe_svm.h  |  6 ++--
 include/drm/drm_gpusvm.h     | 48 +++++++++++++++++---------
 5 files changed, 80 insertions(+), 59 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index b2920b8551f5..fa193dd265b0 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -629,8 +629,8 @@ drm_gpusvm_range_alloc(struct drm_gpusvm *gpusvm,
 	range->itree.start = ALIGN_DOWN(fault_addr, chunk_size);
 	range->itree.last = ALIGN(fault_addr + 1, chunk_size) - 1;
 	INIT_LIST_HEAD(&range->entry);
-	range->notifier_seq = LONG_MAX;
-	range->flags.migrate_devmem = migrate_devmem ? 1 : 0;
+	range->pages.notifier_seq = LONG_MAX;
+	range->pages.flags.migrate_devmem = migrate_devmem ? 1 : 0;
 
 	return range;
 }
@@ -992,19 +992,20 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 					   struct drm_gpusvm_range *range,
 					   unsigned long npages)
 {
-	unsigned long i, j;
-	struct drm_pagemap *dpagemap = range->dpagemap;
+	struct drm_gpusvm_pages *svm_pages = &range->pages;
+	struct drm_pagemap *dpagemap = svm_pages->dpagemap;
 	struct device *dev = gpusvm->drm->dev;
+	unsigned long i, j;
 
 	lockdep_assert_held(&gpusvm->notifier_lock);
 
-	if (range->flags.has_dma_mapping) {
-		struct drm_gpusvm_range_flags flags = {
-			.__flags = range->flags.__flags,
+	if (svm_pages->flags.has_dma_mapping) {
+		struct drm_gpusvm_pages_flags flags = {
+			.__flags = svm_pages->flags.__flags,
 		};
 
 		for (i = 0, j = 0; i < npages; j++) {
-			struct drm_pagemap_addr *addr = &range->dma_addr[j];
+			struct drm_pagemap_addr *addr = &svm_pages->dma_addr[j];
 
 			if (addr->proto == DRM_INTERCONNECT_SYSTEM)
 				dma_unmap_page(dev,
@@ -1020,9 +1021,9 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 		/* WRITE_ONCE pairs with READ_ONCE for opportunistic checks */
 		flags.has_devmem_pages = false;
 		flags.has_dma_mapping = false;
-		WRITE_ONCE(range->flags.__flags, flags.__flags);
+		WRITE_ONCE(svm_pages->flags.__flags, flags.__flags);
 
-		range->dpagemap = NULL;
+		svm_pages->dpagemap = NULL;
 	}
 }
 
@@ -1036,11 +1037,13 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 static void drm_gpusvm_range_free_pages(struct drm_gpusvm *gpusvm,
 					struct drm_gpusvm_range *range)
 {
+	struct drm_gpusvm_pages *svm_pages = &range->pages;
+
 	lockdep_assert_held(&gpusvm->notifier_lock);
 
-	if (range->dma_addr) {
-		kvfree(range->dma_addr);
-		range->dma_addr = NULL;
+	if (svm_pages->dma_addr) {
+		kvfree(svm_pages->dma_addr);
+		svm_pages->dma_addr = NULL;
 	}
 }
 
@@ -1152,9 +1155,11 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_put);
 bool drm_gpusvm_range_pages_valid(struct drm_gpusvm *gpusvm,
 				  struct drm_gpusvm_range *range)
 {
+	struct drm_gpusvm_pages *svm_pages = &range->pages;
+
 	lockdep_assert_held(&gpusvm->notifier_lock);
 
-	return range->flags.has_devmem_pages || range->flags.has_dma_mapping;
+	return svm_pages->flags.has_devmem_pages || svm_pages->flags.has_dma_mapping;
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_pages_valid);
 
@@ -1172,9 +1177,10 @@ static bool
 drm_gpusvm_range_pages_valid_unlocked(struct drm_gpusvm *gpusvm,
 				      struct drm_gpusvm_range *range)
 {
+	struct drm_gpusvm_pages *svm_pages = &range->pages;
 	bool pages_valid;
 
-	if (!range->dma_addr)
+	if (!svm_pages->dma_addr)
 		return false;
 
 	drm_gpusvm_notifier_lock(gpusvm);
@@ -1201,6 +1207,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
 			       struct drm_gpusvm_range *range,
 			       const struct drm_gpusvm_ctx *ctx)
 {
+	struct drm_gpusvm_pages *svm_pages = &range->pages;
 	struct mmu_interval_notifier *notifier = &range->notifier->notifier;
 	struct hmm_range hmm_range = {
 		.default_flags = HMM_PFN_REQ_FAULT | (ctx->read_only ? 0 :
@@ -1223,7 +1230,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
 	int err = 0;
 	struct dev_pagemap *pagemap;
 	struct drm_pagemap *dpagemap;
-	struct drm_gpusvm_range_flags flags;
+	struct drm_gpusvm_pages_flags flags;
 	enum dma_data_direction dma_dir = ctx->read_only ? DMA_TO_DEVICE :
 							   DMA_BIDIRECTIONAL;
 
@@ -1269,7 +1276,7 @@ map_pages:
 	 */
 	drm_gpusvm_notifier_lock(gpusvm);
 
-	flags.__flags = range->flags.__flags;
+	flags.__flags = svm_pages->flags.__flags;
 	if (flags.unmapped) {
 		drm_gpusvm_notifier_unlock(gpusvm);
 		err = -EFAULT;
@@ -1282,13 +1289,12 @@ map_pages:
 		goto retry;
 	}
 
-	if (!range->dma_addr) {
+	if (!svm_pages->dma_addr) {
 		/* Unlock and restart mapping to allocate memory. */
 		drm_gpusvm_notifier_unlock(gpusvm);
-		range->dma_addr = kvmalloc_array(npages,
-						 sizeof(*range->dma_addr),
-						 GFP_KERNEL);
-		if (!range->dma_addr) {
+		svm_pages->dma_addr =
+			kvmalloc_array(npages, sizeof(*svm_pages->dma_addr), GFP_KERNEL);
+		if (!svm_pages->dma_addr) {
 			err = -ENOMEM;
 			goto err_free;
 		}
@@ -1327,13 +1333,13 @@ map_pages:
 					goto err_unmap;
 				}
 			}
-			range->dma_addr[j] =
+			svm_pages->dma_addr[j] =
 				dpagemap->ops->device_map(dpagemap,
 							  gpusvm->drm->dev,
 							  page, order,
 							  dma_dir);
 			if (dma_mapping_error(gpusvm->drm->dev,
-					      range->dma_addr[j].addr)) {
+					      svm_pages->dma_addr[j].addr)) {
 				err = -EFAULT;
 				goto err_unmap;
 			}
@@ -1359,7 +1365,7 @@ map_pages:
 				goto err_unmap;
 			}
 
-			range->dma_addr[j] = drm_pagemap_addr_encode
+			svm_pages->dma_addr[j] = drm_pagemap_addr_encode
 				(addr, DRM_INTERCONNECT_SYSTEM, order,
 				 dma_dir);
 		}
@@ -1370,16 +1376,16 @@ map_pages:
 
 	if (pagemap) {
 		flags.has_devmem_pages = true;
-		range->dpagemap = dpagemap;
+		svm_pages->dpagemap = dpagemap;
 	}
 
 	/* WRITE_ONCE pairs with READ_ONCE for opportunistic checks */
-	WRITE_ONCE(range->flags.__flags, flags.__flags);
+	WRITE_ONCE(svm_pages->flags.__flags, flags.__flags);
 
 	drm_gpusvm_notifier_unlock(gpusvm);
 	kvfree(pfns);
 set_seqno:
-	range->notifier_seq = hmm_range.notifier_seq;
+	svm_pages->notifier_seq = hmm_range.notifier_seq;
 
 	return 0;
 
@@ -1396,7 +1402,6 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_get_pages);
 
 /**
  * drm_gpusvm_range_unmap_pages() - Unmap pages associated with a GPU SVM range
- * drm_gpusvm_range_evict() - Evict GPU SVM range
  * @gpusvm: Pointer to the GPU SVM structure
  * @range: Pointer to the GPU SVM range structure
  * @ctx: GPU SVM context
@@ -1520,10 +1525,10 @@ void drm_gpusvm_range_set_unmapped(struct drm_gpusvm_range *range,
 {
 	lockdep_assert_held_write(&range->gpusvm->notifier_lock);
 
-	range->flags.unmapped = true;
+	range->pages.flags.unmapped = true;
 	if (drm_gpusvm_range_start(range) < mmu_range->start ||
 	    drm_gpusvm_range_end(range) > mmu_range->end)
-		range->flags.partial_unmap = true;
+		range->pages.flags.partial_unmap = true;
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_set_unmapped);
 
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index c129048a9a09..029480c9cebd 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -729,7 +729,7 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 			return -EAGAIN;
 		}
 		if (xe_svm_range_has_dma_mapping(range)) {
-			xe_res_first_dma(range->base.dma_addr, 0,
+			xe_res_first_dma(range->base.pages.dma_addr, 0,
 					 range->base.itree.last + 1 - range->base.itree.start,
 					 &curs);
 			xe_svm_range_debug(range, "BIND PREPARE - MIXED");
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 4834282e4b64..f30c89316812 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -26,9 +26,9 @@ static bool xe_svm_range_in_vram(struct xe_svm_range *range)
 	 * memory.
 	 */
 
-	struct drm_gpusvm_range_flags flags = {
+	struct drm_gpusvm_pages_flags flags = {
 		/* Pairs with WRITE_ONCE in drm_gpusvm.c */
-		.__flags = READ_ONCE(range->base.flags.__flags),
+		.__flags = READ_ONCE(range->base.pages.flags.__flags),
 	};
 
 	return flags.has_devmem_pages;
@@ -58,7 +58,7 @@ static struct xe_vm *range_to_vm(struct drm_gpusvm_range *r)
 	       (r__)->base.gpusvm,					\
 	       xe_svm_range_in_vram((r__)) ? 1 : 0,			\
 	       xe_svm_range_has_vram_binding((r__)) ? 1 : 0,		\
-	       (r__)->base.notifier_seq,				\
+	       (r__)->base.pages.notifier_seq,				\
 	       xe_svm_range_start((r__)), xe_svm_range_end((r__)),	\
 	       xe_svm_range_size((r__)))
 
@@ -134,7 +134,7 @@ xe_svm_range_notifier_event_begin(struct xe_vm *vm, struct drm_gpusvm_range *r,
 	range_debug(range, "NOTIFIER");
 
 	/* Skip if already unmapped or if no binding exist */
-	if (range->base.flags.unmapped || !range->tile_present)
+	if (range->base.pages.flags.unmapped || !range->tile_present)
 		return 0;
 
 	range_debug(range, "NOTIFIER - EXECUTE");
@@ -825,7 +825,7 @@ bool xe_svm_range_validate(struct xe_vm *vm,
 	xe_svm_notifier_lock(vm);
 
 	ret = (range->tile_present & ~range->tile_invalidated & tile_mask) == tile_mask &&
-	       (devmem_preferred == range->base.flags.has_devmem_pages);
+	       (devmem_preferred == range->base.pages.flags.has_devmem_pages);
 
 	xe_svm_notifier_unlock(vm);
 
@@ -936,7 +936,7 @@ bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range, struct xe_vm
 	struct xe_vm *vm = range_to_vm(&range->base);
 	u64 range_size = xe_svm_range_size(range);
 
-	if (!range->base.flags.migrate_devmem || !preferred_region_is_vram)
+	if (!range->base.pages.flags.migrate_devmem || !preferred_region_is_vram)
 		return false;
 
 	xe_assert(vm->xe, IS_DGFX(vm->xe));
@@ -1045,7 +1045,7 @@ retry:
 
 	xe_svm_range_fault_count_stats_incr(gt, range);
 
-	if (ctx.devmem_only && !range->base.flags.migrate_devmem) {
+	if (ctx.devmem_only && !range->base.pages.flags.migrate_devmem) {
 		err = -EACCES;
 		goto out;
 	}
@@ -1397,7 +1397,7 @@ int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range *range,
 {
 	struct drm_pagemap *dpagemap;
 
-	xe_assert(tile_to_xe(tile), range->base.flags.migrate_devmem);
+	xe_assert(tile_to_xe(tile), range->base.pages.flags.migrate_devmem);
 	range_debug(range, "ALLOCATE VRAM");
 
 	dpagemap = tile_local_pagemap(tile);
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index 9d6a8840a8b7..f0e0e009c7fc 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -105,7 +105,7 @@ struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *t
 static inline bool xe_svm_range_has_dma_mapping(struct xe_svm_range *range)
 {
 	lockdep_assert_held(&range->base.gpusvm->notifier_lock);
-	return range->base.flags.has_dma_mapping;
+	return range->base.pages.flags.has_dma_mapping;
 }
 
 /**
@@ -184,7 +184,9 @@ struct xe_vram_region;
 struct xe_svm_range {
 	struct {
 		struct interval_tree_node itree;
-		const struct drm_pagemap_addr *dma_addr;
+		struct {
+			const struct drm_pagemap_addr *dma_addr;
+		} pages;
 	} base;
 	u32 tile_present;
 	u32 tile_invalidated;
diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h
index 0e336148309d..1ee4188c3067 100644
--- a/include/drm/drm_gpusvm.h
+++ b/include/drm/drm_gpusvm.h
@@ -106,16 +106,16 @@ struct drm_gpusvm_notifier {
 };
 
 /**
- * struct drm_gpusvm_range_flags - Structure representing a GPU SVM range flags
+ * struct drm_gpusvm_pages_flags - Structure representing a GPU SVM pages flags
  *
- * @migrate_devmem: Flag indicating whether the range can be migrated to device memory
- * @unmapped: Flag indicating if the range has been unmapped
- * @partial_unmap: Flag indicating if the range has been partially unmapped
- * @has_devmem_pages: Flag indicating if the range has devmem pages
- * @has_dma_mapping: Flag indicating if the range has a DMA mapping
- * @__flags: Flags for range in u16 form (used for READ_ONCE)
+ * @migrate_devmem: Flag indicating whether the pages can be migrated to device memory
+ * @unmapped: Flag indicating if the pages has been unmapped
+ * @partial_unmap: Flag indicating if the pages has been partially unmapped
+ * @has_devmem_pages: Flag indicating if the pages has devmem pages
+ * @has_dma_mapping: Flag indicating if the pages has a DMA mapping
+ * @__flags: Flags for pages in u16 form (used for READ_ONCE)
  */
-struct drm_gpusvm_range_flags {
+struct drm_gpusvm_pages_flags {
 	union {
 		struct {
 			/* All flags below must be set upon creation */
@@ -130,6 +130,27 @@ struct drm_gpusvm_range_flags {
 	};
 };
 
+/**
+ * struct drm_gpusvm_pages - Structure representing a GPU SVM mapped pages
+ *
+ * @dma_addr: Device address array
+ * @dpagemap: The struct drm_pagemap of the device pages we're dma-mapping.
+ *            Note this is assuming only one drm_pagemap per range is allowed.
+ * @notifier_seq: Notifier sequence number of the range's pages
+ * @flags: Flags for range
+ * @flags.migrate_devmem: Flag indicating whether the range can be migrated to device memory
+ * @flags.unmapped: Flag indicating if the range has been unmapped
+ * @flags.partial_unmap: Flag indicating if the range has been partially unmapped
+ * @flags.has_devmem_pages: Flag indicating if the range has devmem pages
+ * @flags.has_dma_mapping: Flag indicating if the range has a DMA mapping
+ */
+struct drm_gpusvm_pages {
+	struct drm_pagemap_addr *dma_addr;
+	struct drm_pagemap *dpagemap;
+	unsigned long notifier_seq;
+	struct drm_gpusvm_pages_flags flags;
+};
+
 /**
  * struct drm_gpusvm_range - Structure representing a GPU SVM range
  *
@@ -138,11 +159,7 @@ struct drm_gpusvm_range_flags {
  * @refcount: Reference count for the range
  * @itree: Interval tree node for the range (inserted in GPU SVM notifier)
  * @entry: List entry to fast interval tree traversal
- * @notifier_seq: Notifier sequence number of the range's pages
- * @dma_addr: Device address array
- * @dpagemap: The struct drm_pagemap of the device pages we're dma-mapping.
- *            Note this is assuming only one drm_pagemap per range is allowed.
- * @flags: Flags for range
+ * @pages: The pages for this range.
  *
  * This structure represents a GPU SVM range used for tracking memory ranges
  * mapped in a DRM device.
@@ -153,10 +170,7 @@ struct drm_gpusvm_range {
 	struct kref refcount;
 	struct interval_tree_node itree;
 	struct list_head entry;
-	unsigned long notifier_seq;
-	struct drm_pagemap_addr *dma_addr;
-	struct drm_pagemap *dpagemap;
-	struct drm_gpusvm_range_flags flags;
+	struct drm_gpusvm_pages pages;
 };
 
 /**

From 6364afd532bcabf4a12bea44c3dd4c291e31a83e Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:35 +0100
Subject: [PATCH 0951/1455] drm/gpusvm: refactor core API to use pages struct
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Refactor the core API of get/unmap/free pages to all operate on
drm_gpusvm_pages. In the next patch we want to export a simplified core
API without needing fully blown svm range etc.

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-14-matthew.auld@intel.com
---
 drivers/gpu/drm/drm_gpusvm.c | 161 ++++++++++++++++++++++++-----------
 1 file changed, 110 insertions(+), 51 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index fa193dd265b0..eb50ed0d0f55 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -980,19 +980,18 @@ err_mmunlock:
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_find_or_insert);
 
 /**
- * __drm_gpusvm_range_unmap_pages() - Unmap pages associated with a GPU SVM range (internal)
+ * __drm_gpusvm_unmap_pages() - Unmap pages associated with GPU SVM pages (internal)
  * @gpusvm: Pointer to the GPU SVM structure
- * @range: Pointer to the GPU SVM range structure
+ * @svm_pages: Pointer to the GPU SVM pages structure
  * @npages: Number of pages to unmap
  *
- * This function unmap pages associated with a GPU SVM range. Assumes and
+ * This function unmap pages associated with a GPU SVM pages struct. Assumes and
  * asserts correct locking is in place when called.
  */
-static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
-					   struct drm_gpusvm_range *range,
-					   unsigned long npages)
+static void __drm_gpusvm_unmap_pages(struct drm_gpusvm *gpusvm,
+				     struct drm_gpusvm_pages *svm_pages,
+				     unsigned long npages)
 {
-	struct drm_gpusvm_pages *svm_pages = &range->pages;
 	struct drm_pagemap *dpagemap = svm_pages->dpagemap;
 	struct device *dev = gpusvm->drm->dev;
 	unsigned long i, j;
@@ -1028,17 +1027,15 @@ static void __drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 }
 
 /**
- * drm_gpusvm_range_free_pages() - Free pages associated with a GPU SVM range
+ * __drm_gpusvm_free_pages() - Free dma array associated with GPU SVM pages
  * @gpusvm: Pointer to the GPU SVM structure
- * @range: Pointer to the GPU SVM range structure
+ * @svm_pages: Pointer to the GPU SVM pages structure
  *
  * This function frees the dma address array associated with a GPU SVM range.
  */
-static void drm_gpusvm_range_free_pages(struct drm_gpusvm *gpusvm,
-					struct drm_gpusvm_range *range)
+static void __drm_gpusvm_free_pages(struct drm_gpusvm *gpusvm,
+				    struct drm_gpusvm_pages *svm_pages)
 {
-	struct drm_gpusvm_pages *svm_pages = &range->pages;
-
 	lockdep_assert_held(&gpusvm->notifier_lock);
 
 	if (svm_pages->dma_addr) {
@@ -1072,8 +1069,8 @@ void drm_gpusvm_range_remove(struct drm_gpusvm *gpusvm,
 		return;
 
 	drm_gpusvm_notifier_lock(gpusvm);
-	__drm_gpusvm_range_unmap_pages(gpusvm, range, npages);
-	drm_gpusvm_range_free_pages(gpusvm, range);
+	__drm_gpusvm_unmap_pages(gpusvm, &range->pages, npages);
+	__drm_gpusvm_free_pages(gpusvm, &range->pages);
 	__drm_gpusvm_range_remove(notifier, range);
 	drm_gpusvm_notifier_unlock(gpusvm);
 
@@ -1138,6 +1135,28 @@ void drm_gpusvm_range_put(struct drm_gpusvm_range *range)
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_put);
 
+/**
+ * drm_gpusvm_pages_valid() - GPU SVM range pages valid
+ * @gpusvm: Pointer to the GPU SVM structure
+ * @svm_pages: Pointer to the GPU SVM pages structure
+ *
+ * This function determines if a GPU SVM range pages are valid. Expected be
+ * called holding gpusvm->notifier_lock and as the last step before committing a
+ * GPU binding. This is akin to a notifier seqno check in the HMM documentation
+ * but due to wider notifiers (i.e., notifiers which span multiple ranges) this
+ * function is required for finer grained checking (i.e., per range) if pages
+ * are valid.
+ *
+ * Return: True if GPU SVM range has valid pages, False otherwise
+ */
+static bool drm_gpusvm_pages_valid(struct drm_gpusvm *gpusvm,
+				   struct drm_gpusvm_pages *svm_pages)
+{
+	lockdep_assert_held(&gpusvm->notifier_lock);
+
+	return svm_pages->flags.has_devmem_pages || svm_pages->flags.has_dma_mapping;
+}
+
 /**
  * drm_gpusvm_range_pages_valid() - GPU SVM range pages valid
  * @gpusvm: Pointer to the GPU SVM structure
@@ -1155,11 +1174,7 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_put);
 bool drm_gpusvm_range_pages_valid(struct drm_gpusvm *gpusvm,
 				  struct drm_gpusvm_range *range)
 {
-	struct drm_gpusvm_pages *svm_pages = &range->pages;
-
-	lockdep_assert_held(&gpusvm->notifier_lock);
-
-	return svm_pages->flags.has_devmem_pages || svm_pages->flags.has_dma_mapping;
+	return drm_gpusvm_pages_valid(gpusvm, &range->pages);
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_pages_valid);
 
@@ -1173,57 +1188,59 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_pages_valid);
  *
  * Return: True if GPU SVM range has valid pages, False otherwise
  */
-static bool
-drm_gpusvm_range_pages_valid_unlocked(struct drm_gpusvm *gpusvm,
-				      struct drm_gpusvm_range *range)
+static bool drm_gpusvm_pages_valid_unlocked(struct drm_gpusvm *gpusvm,
+					    struct drm_gpusvm_pages *svm_pages)
 {
-	struct drm_gpusvm_pages *svm_pages = &range->pages;
 	bool pages_valid;
 
 	if (!svm_pages->dma_addr)
 		return false;
 
 	drm_gpusvm_notifier_lock(gpusvm);
-	pages_valid = drm_gpusvm_range_pages_valid(gpusvm, range);
+	pages_valid = drm_gpusvm_pages_valid(gpusvm, svm_pages);
 	if (!pages_valid)
-		drm_gpusvm_range_free_pages(gpusvm, range);
+		__drm_gpusvm_free_pages(gpusvm, svm_pages);
 	drm_gpusvm_notifier_unlock(gpusvm);
 
 	return pages_valid;
 }
 
 /**
- * drm_gpusvm_range_get_pages() - Get pages for a GPU SVM range
+ * drm_gpusvm_get_pages() - Get pages and populate GPU SVM pages struct
  * @gpusvm: Pointer to the GPU SVM structure
- * @range: Pointer to the GPU SVM range structure
+ * @svm_pages: The SVM pages to populate. This will contain the dma-addresses
+ * @mm: The mm corresponding to the CPU range
+ * @notifier: The corresponding notifier for the given CPU range
+ * @pages_start: Start CPU address for the pages
+ * @pages_end: End CPU address for the pages (exclusive)
  * @ctx: GPU SVM context
  *
- * This function gets pages for a GPU SVM range and ensures they are mapped for
- * DMA access.
+ * This function gets and maps pages for CPU range and ensures they are
+ * mapped for DMA access.
  *
  * Return: 0 on success, negative error code on failure.
  */
-int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
-			       struct drm_gpusvm_range *range,
-			       const struct drm_gpusvm_ctx *ctx)
+static int drm_gpusvm_get_pages(struct drm_gpusvm *gpusvm,
+				struct drm_gpusvm_pages *svm_pages,
+				struct mm_struct *mm,
+				struct mmu_interval_notifier *notifier,
+				unsigned long pages_start,
+				unsigned long pages_end,
+				const struct drm_gpusvm_ctx *ctx)
 {
-	struct drm_gpusvm_pages *svm_pages = &range->pages;
-	struct mmu_interval_notifier *notifier = &range->notifier->notifier;
 	struct hmm_range hmm_range = {
 		.default_flags = HMM_PFN_REQ_FAULT | (ctx->read_only ? 0 :
 			HMM_PFN_REQ_WRITE),
 		.notifier = notifier,
-		.start = drm_gpusvm_range_start(range),
-		.end = drm_gpusvm_range_end(range),
+		.start = pages_start,
+		.end = pages_end,
 		.dev_private_owner = gpusvm->device_private_page_owner,
 	};
-	struct mm_struct *mm = gpusvm->mm;
 	void *zdd;
 	unsigned long timeout =
 		jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
 	unsigned long i, j;
-	unsigned long npages = npages_in_range(drm_gpusvm_range_start(range),
-					       drm_gpusvm_range_end(range));
+	unsigned long npages = npages_in_range(pages_start, pages_end);
 	unsigned long num_dma_mapped;
 	unsigned int order = 0;
 	unsigned long *pfns;
@@ -1236,7 +1253,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
 
 retry:
 	hmm_range.notifier_seq = mmu_interval_read_begin(notifier);
-	if (drm_gpusvm_range_pages_valid_unlocked(gpusvm, range))
+	if (drm_gpusvm_pages_valid_unlocked(gpusvm, svm_pages))
 		goto set_seqno;
 
 	pfns = kvmalloc_array(npages, sizeof(*pfns), GFP_KERNEL);
@@ -1390,7 +1407,7 @@ set_seqno:
 	return 0;
 
 err_unmap:
-	__drm_gpusvm_range_unmap_pages(gpusvm, range, num_dma_mapped);
+	__drm_gpusvm_unmap_pages(gpusvm, svm_pages, num_dma_mapped);
 	drm_gpusvm_notifier_unlock(gpusvm);
 err_free:
 	kvfree(pfns);
@@ -1398,8 +1415,58 @@ err_free:
 		goto retry;
 	return err;
 }
+
+/**
+ * drm_gpusvm_range_get_pages() - Get pages for a GPU SVM range
+ * @gpusvm: Pointer to the GPU SVM structure
+ * @range: Pointer to the GPU SVM range structure
+ * @ctx: GPU SVM context
+ *
+ * This function gets pages for a GPU SVM range and ensures they are mapped for
+ * DMA access.
+ *
+ * Return: 0 on success, negative error code on failure.
+ */
+int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm,
+			       struct drm_gpusvm_range *range,
+			       const struct drm_gpusvm_ctx *ctx)
+{
+	return drm_gpusvm_get_pages(gpusvm, &range->pages, gpusvm->mm,
+				    &range->notifier->notifier,
+				    drm_gpusvm_range_start(range),
+				    drm_gpusvm_range_end(range), ctx);
+}
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_get_pages);
 
+/**
+ * drm_gpusvm_unmap_pages() - Unmap GPU svm pages
+ * @gpusvm: Pointer to the GPU SVM structure
+ * @svm_pages: Pointer to the GPU SVM pages structure
+ * @npages: Number of pages in @svm_pages.
+ * @ctx: GPU SVM context
+ *
+ * This function unmaps pages associated with a GPU SVM pages struct. If
+ * @in_notifier is set, it is assumed that gpusvm->notifier_lock is held in
+ * write mode; if it is clear, it acquires gpusvm->notifier_lock in read mode.
+ * Must be called in the invalidate() callback of the corresponding notifier for
+ * IOMMU security model.
+ */
+static void drm_gpusvm_unmap_pages(struct drm_gpusvm *gpusvm,
+				   struct drm_gpusvm_pages *svm_pages,
+				   unsigned long npages,
+				   const struct drm_gpusvm_ctx *ctx)
+{
+	if (ctx->in_notifier)
+		lockdep_assert_held_write(&gpusvm->notifier_lock);
+	else
+		drm_gpusvm_notifier_lock(gpusvm);
+
+	__drm_gpusvm_unmap_pages(gpusvm, svm_pages, npages);
+
+	if (!ctx->in_notifier)
+		drm_gpusvm_notifier_unlock(gpusvm);
+}
+
 /**
  * drm_gpusvm_range_unmap_pages() - Unmap pages associated with a GPU SVM range
  * @gpusvm: Pointer to the GPU SVM structure
@@ -1419,15 +1486,7 @@ void drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,
 	unsigned long npages = npages_in_range(drm_gpusvm_range_start(range),
 					       drm_gpusvm_range_end(range));
 
-	if (ctx->in_notifier)
-		lockdep_assert_held_write(&gpusvm->notifier_lock);
-	else
-		drm_gpusvm_notifier_lock(gpusvm);
-
-	__drm_gpusvm_range_unmap_pages(gpusvm, range, npages);
-
-	if (!ctx->in_notifier)
-		drm_gpusvm_notifier_unlock(gpusvm);
+	return drm_gpusvm_unmap_pages(gpusvm, &range->pages, npages, ctx);
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_unmap_pages);
 

From 83f706ecbde1dfdc377bafda773fdc57644cd479 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:36 +0100
Subject: [PATCH 0952/1455] drm/gpusvm: export drm_gpusvm_pages API
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Export get/unmap/free pages API. We also need to tweak the SVM init to
allow skipping much of the unneeded parts.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-15-matthew.auld@intel.com
---
 drivers/gpu/drm/drm_gpusvm.c | 66 ++++++++++++++++++++++++++++--------
 include/drm/drm_gpusvm.h     | 16 +++++++++
 2 files changed, 67 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index eb50ed0d0f55..eeeeb99cfdf6 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -373,6 +373,12 @@ static const struct mmu_interval_notifier_ops drm_gpusvm_notifier_ops = {
  *
  * This function initializes the GPU SVM.
  *
+ * Note: If only using the simple drm_gpusvm_pages API (get/unmap/free),
+ * then only @gpusvm, @name, and @drm are expected. However, the same base
+ * @gpusvm can also be used with both modes together in which case the full
+ * setup is needed, where the core drm_gpusvm_pages API will simply never use
+ * the other fields.
+ *
  * Return: 0 on success, a negative error code on failure.
  */
 int drm_gpusvm_init(struct drm_gpusvm *gpusvm,
@@ -383,8 +389,16 @@ int drm_gpusvm_init(struct drm_gpusvm *gpusvm,
 		    const struct drm_gpusvm_ops *ops,
 		    const unsigned long *chunk_sizes, int num_chunks)
 {
-	if (!ops->invalidate || !num_chunks)
-		return -EINVAL;
+	if (mm) {
+		if (!ops->invalidate || !num_chunks)
+			return -EINVAL;
+		mmgrab(mm);
+	} else {
+		/* No full SVM mode, only core drm_gpusvm_pages API. */
+		if (ops || num_chunks || mm_range || notifier_size ||
+		    device_private_page_owner)
+			return -EINVAL;
+	}
 
 	gpusvm->name = name;
 	gpusvm->drm = drm;
@@ -397,7 +411,6 @@ int drm_gpusvm_init(struct drm_gpusvm *gpusvm,
 	gpusvm->chunk_sizes = chunk_sizes;
 	gpusvm->num_chunks = num_chunks;
 
-	mmgrab(mm);
 	gpusvm->root = RB_ROOT_CACHED;
 	INIT_LIST_HEAD(&gpusvm->notifier_list);
 
@@ -489,7 +502,8 @@ void drm_gpusvm_fini(struct drm_gpusvm *gpusvm)
 			drm_gpusvm_range_remove(gpusvm, range);
 	}
 
-	mmdrop(gpusvm->mm);
+	if (gpusvm->mm)
+		mmdrop(gpusvm->mm);
 	WARN_ON(!RB_EMPTY_ROOT(&gpusvm->root.rb_root));
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_fini);
@@ -1044,6 +1058,27 @@ static void __drm_gpusvm_free_pages(struct drm_gpusvm *gpusvm,
 	}
 }
 
+/**
+ * drm_gpusvm_free_pages() - Free dma-mapping associated with GPU SVM pages
+ * struct
+ * @gpusvm: Pointer to the GPU SVM structure
+ * @svm_pages: Pointer to the GPU SVM pages structure
+ * @npages: Number of mapped pages
+ *
+ * This function unmaps and frees the dma address array associated with a GPU
+ * SVM pages struct.
+ */
+void drm_gpusvm_free_pages(struct drm_gpusvm *gpusvm,
+			   struct drm_gpusvm_pages *svm_pages,
+			   unsigned long npages)
+{
+	drm_gpusvm_notifier_lock(gpusvm);
+	__drm_gpusvm_unmap_pages(gpusvm, svm_pages, npages);
+	__drm_gpusvm_free_pages(gpusvm, svm_pages);
+	drm_gpusvm_notifier_unlock(gpusvm);
+}
+EXPORT_SYMBOL_GPL(drm_gpusvm_free_pages);
+
 /**
  * drm_gpusvm_range_remove() - Remove GPU SVM range
  * @gpusvm: Pointer to the GPU SVM structure
@@ -1220,13 +1255,12 @@ static bool drm_gpusvm_pages_valid_unlocked(struct drm_gpusvm *gpusvm,
  *
  * Return: 0 on success, negative error code on failure.
  */
-static int drm_gpusvm_get_pages(struct drm_gpusvm *gpusvm,
-				struct drm_gpusvm_pages *svm_pages,
-				struct mm_struct *mm,
-				struct mmu_interval_notifier *notifier,
-				unsigned long pages_start,
-				unsigned long pages_end,
-				const struct drm_gpusvm_ctx *ctx)
+int drm_gpusvm_get_pages(struct drm_gpusvm *gpusvm,
+			 struct drm_gpusvm_pages *svm_pages,
+			 struct mm_struct *mm,
+			 struct mmu_interval_notifier *notifier,
+			 unsigned long pages_start, unsigned long pages_end,
+			 const struct drm_gpusvm_ctx *ctx)
 {
 	struct hmm_range hmm_range = {
 		.default_flags = HMM_PFN_REQ_FAULT | (ctx->read_only ? 0 :
@@ -1415,6 +1449,7 @@ err_free:
 		goto retry;
 	return err;
 }
+EXPORT_SYMBOL_GPL(drm_gpusvm_get_pages);
 
 /**
  * drm_gpusvm_range_get_pages() - Get pages for a GPU SVM range
@@ -1451,10 +1486,10 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_get_pages);
  * Must be called in the invalidate() callback of the corresponding notifier for
  * IOMMU security model.
  */
-static void drm_gpusvm_unmap_pages(struct drm_gpusvm *gpusvm,
-				   struct drm_gpusvm_pages *svm_pages,
-				   unsigned long npages,
-				   const struct drm_gpusvm_ctx *ctx)
+void drm_gpusvm_unmap_pages(struct drm_gpusvm *gpusvm,
+			    struct drm_gpusvm_pages *svm_pages,
+			    unsigned long npages,
+			    const struct drm_gpusvm_ctx *ctx)
 {
 	if (ctx->in_notifier)
 		lockdep_assert_held_write(&gpusvm->notifier_lock);
@@ -1466,6 +1501,7 @@ static void drm_gpusvm_unmap_pages(struct drm_gpusvm *gpusvm,
 	if (!ctx->in_notifier)
 		drm_gpusvm_notifier_unlock(gpusvm);
 }
+EXPORT_SYMBOL_GPL(drm_gpusvm_unmap_pages);
 
 /**
  * drm_gpusvm_range_unmap_pages() - Unmap pages associated with a GPU SVM range
diff --git a/include/drm/drm_gpusvm.h b/include/drm/drm_gpusvm.h
index 1ee4188c3067..5434048a2ca4 100644
--- a/include/drm/drm_gpusvm.h
+++ b/include/drm/drm_gpusvm.h
@@ -307,6 +307,22 @@ drm_gpusvm_range_find(struct drm_gpusvm_notifier *notifier, unsigned long start,
 void drm_gpusvm_range_set_unmapped(struct drm_gpusvm_range *range,
 				   const struct mmu_notifier_range *mmu_range);
 
+int drm_gpusvm_get_pages(struct drm_gpusvm *gpusvm,
+			 struct drm_gpusvm_pages *svm_pages,
+			 struct mm_struct *mm,
+			 struct mmu_interval_notifier *notifier,
+			 unsigned long pages_start, unsigned long pages_end,
+			 const struct drm_gpusvm_ctx *ctx);
+
+void drm_gpusvm_unmap_pages(struct drm_gpusvm *gpusvm,
+			    struct drm_gpusvm_pages *svm_pages,
+			    unsigned long npages,
+			    const struct drm_gpusvm_ctx *ctx);
+
+void drm_gpusvm_free_pages(struct drm_gpusvm *gpusvm,
+			   struct drm_gpusvm_pages *svm_pages,
+			   unsigned long npages);
+
 #ifdef CONFIG_LOCKDEP
 /**
  * drm_gpusvm_driver_set_lock() - Set the lock protecting accesses to GPU SVM

From dd25b995a2711cb135dba9af7670c6debeea7b04 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:37 +0100
Subject: [PATCH 0953/1455] drm/xe/vm: split userptr bits into separate file
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This will simplify compiling out the bits that depend on DRM_GPUSVM in a
later patch. Without this we end up littering the code with ifdef
checks, plus it becomes hard to be sure that something won't blow at
runtime due to something not being initialised, even though it passed
the build. Should be no functional change here.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-16-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/Makefile      |   1 +
 drivers/gpu/drm/xe/xe_pt.c       |   1 +
 drivers/gpu/drm/xe/xe_userptr.c  | 305 +++++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_userptr.h  |  97 ++++++++++
 drivers/gpu/drm/xe/xe_vm.c       | 278 +---------------------------
 drivers/gpu/drm/xe/xe_vm.h       |  17 --
 drivers/gpu/drm/xe/xe_vm_types.h |  60 +-----
 7 files changed, 411 insertions(+), 348 deletions(-)
 create mode 100644 drivers/gpu/drm/xe/xe_userptr.c
 create mode 100644 drivers/gpu/drm/xe/xe_userptr.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 95c0f0a9b403..ae833a9e1bb1 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -130,6 +130,7 @@ xe-y += xe_bb.o \
 	xe_tuning.o \
 	xe_uc.o \
 	xe_uc_fw.o \
+	xe_userptr.o \
 	xe_vm.o \
 	xe_vm_madvise.o \
 	xe_vram.o \
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 029480c9cebd..9eac29cbdc48 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -24,6 +24,7 @@
 #include "xe_tlb_inval_job.h"
 #include "xe_trace.h"
 #include "xe_ttm_stolen_mgr.h"
+#include "xe_userptr.h"
 #include "xe_vm.h"
 
 struct xe_pt_dir {
diff --git a/drivers/gpu/drm/xe/xe_userptr.c b/drivers/gpu/drm/xe/xe_userptr.c
new file mode 100644
index 000000000000..17ecff17c993
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_userptr.c
@@ -0,0 +1,305 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include "xe_userptr.h"
+
+#include <linux/mm.h>
+
+#include "xe_hmm.h"
+#include "xe_trace_bo.h"
+
+/**
+ * xe_vma_userptr_check_repin() - Advisory check for repin needed
+ * @uvma: The userptr vma
+ *
+ * Check if the userptr vma has been invalidated since last successful
+ * repin. The check is advisory only and can the function can be called
+ * without the vm->svm.gpusvm.notifier_lock held. There is no guarantee that the
+ * vma userptr will remain valid after a lockless check, so typically
+ * the call needs to be followed by a proper check under the notifier_lock.
+ *
+ * Return: 0 if userptr vma is valid, -EAGAIN otherwise; repin recommended.
+ */
+int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma)
+{
+	return mmu_interval_check_retry(&uvma->userptr.notifier,
+					uvma->userptr.notifier_seq) ?
+		-EAGAIN : 0;
+}
+
+/**
+ * __xe_vm_userptr_needs_repin() - Check whether the VM does have userptrs
+ * that need repinning.
+ * @vm: The VM.
+ *
+ * This function checks for whether the VM has userptrs that need repinning,
+ * and provides a release-type barrier on the userptr.notifier_lock after
+ * checking.
+ *
+ * Return: 0 if there are no userptrs needing repinning, -EAGAIN if there are.
+ */
+int __xe_vm_userptr_needs_repin(struct xe_vm *vm)
+{
+	lockdep_assert_held_read(&vm->userptr.notifier_lock);
+
+	return (list_empty(&vm->userptr.repin_list) &&
+		list_empty(&vm->userptr.invalidated)) ? 0 : -EAGAIN;
+}
+
+int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma)
+{
+	struct xe_vma *vma = &uvma->vma;
+	struct xe_vm *vm = xe_vma_vm(vma);
+	struct xe_device *xe = vm->xe;
+
+	lockdep_assert_held(&vm->lock);
+	xe_assert(xe, xe_vma_is_userptr(vma));
+
+	return xe_hmm_userptr_populate_range(uvma, false);
+}
+
+static void __vma_userptr_invalidate(struct xe_vm *vm, struct xe_userptr_vma *uvma)
+{
+	struct xe_userptr *userptr = &uvma->userptr;
+	struct xe_vma *vma = &uvma->vma;
+	struct dma_resv_iter cursor;
+	struct dma_fence *fence;
+	long err;
+
+	/*
+	 * Tell exec and rebind worker they need to repin and rebind this
+	 * userptr.
+	 */
+	if (!xe_vm_in_fault_mode(vm) &&
+	    !(vma->gpuva.flags & XE_VMA_DESTROYED)) {
+		spin_lock(&vm->userptr.invalidated_lock);
+		list_move_tail(&userptr->invalidate_link,
+			       &vm->userptr.invalidated);
+		spin_unlock(&vm->userptr.invalidated_lock);
+	}
+
+	/*
+	 * Preempt fences turn into schedule disables, pipeline these.
+	 * Note that even in fault mode, we need to wait for binds and
+	 * unbinds to complete, and those are attached as BOOKMARK fences
+	 * to the vm.
+	 */
+	dma_resv_iter_begin(&cursor, xe_vm_resv(vm),
+			    DMA_RESV_USAGE_BOOKKEEP);
+	dma_resv_for_each_fence_unlocked(&cursor, fence)
+		dma_fence_enable_sw_signaling(fence);
+	dma_resv_iter_end(&cursor);
+
+	err = dma_resv_wait_timeout(xe_vm_resv(vm),
+				    DMA_RESV_USAGE_BOOKKEEP,
+				    false, MAX_SCHEDULE_TIMEOUT);
+	XE_WARN_ON(err <= 0);
+
+	if (xe_vm_in_fault_mode(vm) && userptr->initial_bind) {
+		err = xe_vm_invalidate_vma(vma);
+		XE_WARN_ON(err);
+	}
+
+	xe_hmm_userptr_unmap(uvma);
+}
+
+static bool vma_userptr_invalidate(struct mmu_interval_notifier *mni,
+				   const struct mmu_notifier_range *range,
+				   unsigned long cur_seq)
+{
+	struct xe_userptr_vma *uvma = container_of(mni, typeof(*uvma), userptr.notifier);
+	struct xe_vma *vma = &uvma->vma;
+	struct xe_vm *vm = xe_vma_vm(vma);
+
+	xe_assert(vm->xe, xe_vma_is_userptr(vma));
+	trace_xe_vma_userptr_invalidate(vma);
+
+	if (!mmu_notifier_range_blockable(range))
+		return false;
+
+	vm_dbg(&xe_vma_vm(vma)->xe->drm,
+	       "NOTIFIER: addr=0x%016llx, range=0x%016llx",
+		xe_vma_start(vma), xe_vma_size(vma));
+
+	down_write(&vm->userptr.notifier_lock);
+	mmu_interval_set_seq(mni, cur_seq);
+
+	__vma_userptr_invalidate(vm, uvma);
+	up_write(&vm->userptr.notifier_lock);
+	trace_xe_vma_userptr_invalidate_complete(vma);
+
+	return true;
+}
+
+static const struct mmu_interval_notifier_ops vma_userptr_notifier_ops = {
+	.invalidate = vma_userptr_invalidate,
+};
+
+#if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
+/**
+ * xe_vma_userptr_force_invalidate() - force invalidate a userptr
+ * @uvma: The userptr vma to invalidate
+ *
+ * Perform a forced userptr invalidation for testing purposes.
+ */
+void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma)
+{
+	struct xe_vm *vm = xe_vma_vm(&uvma->vma);
+
+	/* Protect against concurrent userptr pinning */
+	lockdep_assert_held(&vm->lock);
+	/* Protect against concurrent notifiers */
+	lockdep_assert_held(&vm->userptr.notifier_lock);
+	/*
+	 * Protect against concurrent instances of this function and
+	 * the critical exec sections
+	 */
+	xe_vm_assert_held(vm);
+
+	if (!mmu_interval_read_retry(&uvma->userptr.notifier,
+				     uvma->userptr.notifier_seq))
+		uvma->userptr.notifier_seq -= 2;
+	__vma_userptr_invalidate(vm, uvma);
+}
+#endif
+
+int xe_vm_userptr_pin(struct xe_vm *vm)
+{
+	struct xe_userptr_vma *uvma, *next;
+	int err = 0;
+
+	xe_assert(vm->xe, !xe_vm_in_fault_mode(vm));
+	lockdep_assert_held_write(&vm->lock);
+
+	/* Collect invalidated userptrs */
+	spin_lock(&vm->userptr.invalidated_lock);
+	xe_assert(vm->xe, list_empty(&vm->userptr.repin_list));
+	list_for_each_entry_safe(uvma, next, &vm->userptr.invalidated,
+				 userptr.invalidate_link) {
+		list_del_init(&uvma->userptr.invalidate_link);
+		list_add_tail(&uvma->userptr.repin_link,
+			      &vm->userptr.repin_list);
+	}
+	spin_unlock(&vm->userptr.invalidated_lock);
+
+	/* Pin and move to bind list */
+	list_for_each_entry_safe(uvma, next, &vm->userptr.repin_list,
+				 userptr.repin_link) {
+		err = xe_vma_userptr_pin_pages(uvma);
+		if (err == -EFAULT) {
+			list_del_init(&uvma->userptr.repin_link);
+			/*
+			 * We might have already done the pin once already, but
+			 * then had to retry before the re-bind happened, due
+			 * some other condition in the caller, but in the
+			 * meantime the userptr got dinged by the notifier such
+			 * that we need to revalidate here, but this time we hit
+			 * the EFAULT. In such a case make sure we remove
+			 * ourselves from the rebind list to avoid going down in
+			 * flames.
+			 */
+			if (!list_empty(&uvma->vma.combined_links.rebind))
+				list_del_init(&uvma->vma.combined_links.rebind);
+
+			/* Wait for pending binds */
+			xe_vm_lock(vm, false);
+			dma_resv_wait_timeout(xe_vm_resv(vm),
+					      DMA_RESV_USAGE_BOOKKEEP,
+					      false, MAX_SCHEDULE_TIMEOUT);
+
+			down_read(&vm->userptr.notifier_lock);
+			err = xe_vm_invalidate_vma(&uvma->vma);
+			up_read(&vm->userptr.notifier_lock);
+			xe_vm_unlock(vm);
+			if (err)
+				break;
+		} else {
+			if (err)
+				break;
+
+			list_del_init(&uvma->userptr.repin_link);
+			list_move_tail(&uvma->vma.combined_links.rebind,
+				       &vm->rebind_list);
+		}
+	}
+
+	if (err) {
+		down_write(&vm->userptr.notifier_lock);
+		spin_lock(&vm->userptr.invalidated_lock);
+		list_for_each_entry_safe(uvma, next, &vm->userptr.repin_list,
+					 userptr.repin_link) {
+			list_del_init(&uvma->userptr.repin_link);
+			list_move_tail(&uvma->userptr.invalidate_link,
+				       &vm->userptr.invalidated);
+		}
+		spin_unlock(&vm->userptr.invalidated_lock);
+		up_write(&vm->userptr.notifier_lock);
+	}
+	return err;
+}
+
+/**
+ * xe_vm_userptr_check_repin() - Check whether the VM might have userptrs
+ * that need repinning.
+ * @vm: The VM.
+ *
+ * This function does an advisory check for whether the VM has userptrs that
+ * need repinning.
+ *
+ * Return: 0 if there are no indications of userptrs needing repinning,
+ * -EAGAIN if there are.
+ */
+int xe_vm_userptr_check_repin(struct xe_vm *vm)
+{
+	return (list_empty_careful(&vm->userptr.repin_list) &&
+		list_empty_careful(&vm->userptr.invalidated)) ? 0 : -EAGAIN;
+}
+
+int xe_userptr_setup(struct xe_userptr_vma *uvma, unsigned long start,
+		     unsigned long range)
+{
+	struct xe_userptr *userptr = &uvma->userptr;
+	int err;
+
+	INIT_LIST_HEAD(&userptr->invalidate_link);
+	INIT_LIST_HEAD(&userptr->repin_link);
+	mutex_init(&userptr->unmap_mutex);
+
+	err = mmu_interval_notifier_insert(&userptr->notifier, current->mm,
+					   start, range,
+					   &vma_userptr_notifier_ops);
+	if (err)
+		return err;
+
+	userptr->notifier_seq = LONG_MAX;
+
+	return 0;
+}
+
+void xe_userptr_remove(struct xe_userptr_vma *uvma)
+{
+	struct xe_userptr *userptr = &uvma->userptr;
+
+	if (userptr->sg)
+		xe_hmm_userptr_free_sg(uvma);
+
+	/*
+	 * Since userptr pages are not pinned, we can't remove
+	 * the notifier until we're sure the GPU is not accessing
+	 * them anymore
+	 */
+	mmu_interval_notifier_remove(&userptr->notifier);
+	mutex_destroy(&userptr->unmap_mutex);
+}
+
+void xe_userptr_destroy(struct xe_userptr_vma *uvma)
+{
+	struct xe_vm *vm = xe_vma_vm(&uvma->vma);
+
+	spin_lock(&vm->userptr.invalidated_lock);
+	xe_assert(vm->xe, list_empty(&uvma->userptr.repin_link));
+	list_del(&uvma->userptr.invalidate_link);
+	spin_unlock(&vm->userptr.invalidated_lock);
+}
diff --git a/drivers/gpu/drm/xe/xe_userptr.h b/drivers/gpu/drm/xe/xe_userptr.h
new file mode 100644
index 000000000000..dc49348ac42d
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_userptr.h
@@ -0,0 +1,97 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_USERPTR_H_
+#define _XE_USERPTR_H_
+
+#include <linux/list.h>
+#include <linux/mutex.h>
+#include <linux/notifier.h>
+#include <linux/scatterlist.h>
+#include <linux/spinlock.h>
+
+struct xe_vm;
+struct xe_vma;
+struct xe_userptr_vma;
+
+/** struct xe_userptr_vm - User pointer VM level state */
+struct xe_userptr_vm {
+	/**
+	 * @userptr.repin_list: list of VMAs which are user pointers,
+	 * and needs repinning. Protected by @lock.
+	 */
+	struct list_head repin_list;
+	/**
+	 * @notifier_lock: protects notifier in write mode and
+	 * submission in read mode.o
+	 */
+	struct rw_semaphore notifier_lock;
+	/**
+	 * @userptr.invalidated_lock: Protects the
+	 * @userptr.invalidated list.
+	 */
+	spinlock_t invalidated_lock;
+	/**
+	 * @userptr.invalidated: List of invalidated userptrs, not yet
+	 * picked
+	 * up for revalidation. Protected from access with the
+	 * @invalidated_lock. Removing items from the list
+	 * additionally requires @lock in write mode, and adding
+	 * items to the list requires either the @userptr.notifier_lock in
+	 * write mode, OR @lock in write mode.
+	 */
+	struct list_head invalidated;
+};
+
+/** struct xe_userptr - User pointer */
+struct xe_userptr {
+	/** @invalidate_link: Link for the vm::userptr.invalidated list */
+	struct list_head invalidate_link;
+	/** @userptr: link into VM repin list if userptr. */
+	struct list_head repin_link;
+	/**
+	 * @notifier: MMU notifier for user pointer (invalidation call back)
+	 */
+	struct mmu_interval_notifier notifier;
+	/** @sgt: storage for a scatter gather table */
+	struct sg_table sgt;
+	/** @sg: allocated scatter gather table */
+	struct sg_table *sg;
+	/** @notifier_seq: notifier sequence number */
+	unsigned long notifier_seq;
+	/** @unmap_mutex: Mutex protecting dma-unmapping */
+	struct mutex unmap_mutex;
+	/**
+	 * @initial_bind: user pointer has been bound at least once.
+	 * write: vm->userptr.notifier_lock in read mode and vm->resv held.
+	 * read: vm->userptr.notifier_lock in write mode or vm->resv held.
+	 */
+	bool initial_bind;
+	/** @mapped: Whether the @sgt sg-table is dma-mapped. Protected by @unmap_mutex. */
+	bool mapped;
+#if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
+	u32 divisor;
+#endif
+};
+
+void xe_userptr_remove(struct xe_userptr_vma *uvma);
+int xe_userptr_setup(struct xe_userptr_vma *uvma, unsigned long start,
+		     unsigned long range);
+void xe_userptr_destroy(struct xe_userptr_vma *uvma);
+
+int xe_vm_userptr_pin(struct xe_vm *vm);
+int __xe_vm_userptr_needs_repin(struct xe_vm *vm);
+int xe_vm_userptr_check_repin(struct xe_vm *vm);
+int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma);
+int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma);
+
+#if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
+void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma);
+#else
+static inline void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma)
+{
+}
+#endif
+#endif
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index f55f96bb240a..be3d4d023fa6 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -40,6 +40,7 @@
 #include "xe_tile.h"
 #include "xe_tlb_inval.h"
 #include "xe_trace_bo.h"
+#include "xe_userptr.h"
 #include "xe_wa.h"
 #include "xe_hmm.h"
 
@@ -48,37 +49,6 @@ static struct drm_gem_object *xe_vm_obj(struct xe_vm *vm)
 	return vm->gpuvm.r_obj;
 }
 
-/**
- * xe_vma_userptr_check_repin() - Advisory check for repin needed
- * @uvma: The userptr vma
- *
- * Check if the userptr vma has been invalidated since last successful
- * repin. The check is advisory only and can the function can be called
- * without the vm->userptr.notifier_lock held. There is no guarantee that the
- * vma userptr will remain valid after a lockless check, so typically
- * the call needs to be followed by a proper check under the notifier_lock.
- *
- * Return: 0 if userptr vma is valid, -EAGAIN otherwise; repin recommended.
- */
-int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma)
-{
-	return mmu_interval_check_retry(&uvma->userptr.notifier,
-					uvma->userptr.notifier_seq) ?
-		-EAGAIN : 0;
-}
-
-int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma)
-{
-	struct xe_vma *vma = &uvma->vma;
-	struct xe_vm *vm = xe_vma_vm(vma);
-	struct xe_device *xe = vm->xe;
-
-	lockdep_assert_held(&vm->lock);
-	xe_assert(xe, xe_vma_is_userptr(vma));
-
-	return xe_hmm_userptr_populate_range(uvma, false);
-}
-
 static bool preempt_fences_waiting(struct xe_vm *vm)
 {
 	struct xe_exec_queue *q;
@@ -300,25 +270,6 @@ void xe_vm_remove_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 	up_write(&vm->lock);
 }
 
-/**
- * __xe_vm_userptr_needs_repin() - Check whether the VM does have userptrs
- * that need repinning.
- * @vm: The VM.
- *
- * This function checks for whether the VM has userptrs that need repinning,
- * and provides a release-type barrier on the userptr.notifier_lock after
- * checking.
- *
- * Return: 0 if there are no userptrs needing repinning, -EAGAIN if there are.
- */
-int __xe_vm_userptr_needs_repin(struct xe_vm *vm)
-{
-	lockdep_assert_held_read(&vm->userptr.notifier_lock);
-
-	return (list_empty(&vm->userptr.repin_list) &&
-		list_empty(&vm->userptr.invalidated)) ? 0 : -EAGAIN;
-}
-
 #define XE_VM_REBIND_RETRY_TIMEOUT_MS 1000
 
 /**
@@ -584,203 +535,6 @@ out_unlock_outer:
 	trace_xe_vm_rebind_worker_exit(vm);
 }
 
-static void __vma_userptr_invalidate(struct xe_vm *vm, struct xe_userptr_vma *uvma)
-{
-	struct xe_userptr *userptr = &uvma->userptr;
-	struct xe_vma *vma = &uvma->vma;
-	struct dma_resv_iter cursor;
-	struct dma_fence *fence;
-	long err;
-
-	/*
-	 * Tell exec and rebind worker they need to repin and rebind this
-	 * userptr.
-	 */
-	if (!xe_vm_in_fault_mode(vm) &&
-	    !(vma->gpuva.flags & XE_VMA_DESTROYED)) {
-		spin_lock(&vm->userptr.invalidated_lock);
-		list_move_tail(&userptr->invalidate_link,
-			       &vm->userptr.invalidated);
-		spin_unlock(&vm->userptr.invalidated_lock);
-	}
-
-	/*
-	 * Preempt fences turn into schedule disables, pipeline these.
-	 * Note that even in fault mode, we need to wait for binds and
-	 * unbinds to complete, and those are attached as BOOKMARK fences
-	 * to the vm.
-	 */
-	dma_resv_iter_begin(&cursor, xe_vm_resv(vm),
-			    DMA_RESV_USAGE_BOOKKEEP);
-	dma_resv_for_each_fence_unlocked(&cursor, fence)
-		dma_fence_enable_sw_signaling(fence);
-	dma_resv_iter_end(&cursor);
-
-	err = dma_resv_wait_timeout(xe_vm_resv(vm),
-				    DMA_RESV_USAGE_BOOKKEEP,
-				    false, MAX_SCHEDULE_TIMEOUT);
-	XE_WARN_ON(err <= 0);
-
-	if (xe_vm_in_fault_mode(vm) && userptr->initial_bind) {
-		err = xe_vm_invalidate_vma(vma);
-		XE_WARN_ON(err);
-	}
-
-	xe_hmm_userptr_unmap(uvma);
-}
-
-static bool vma_userptr_invalidate(struct mmu_interval_notifier *mni,
-				   const struct mmu_notifier_range *range,
-				   unsigned long cur_seq)
-{
-	struct xe_userptr_vma *uvma = container_of(mni, typeof(*uvma), userptr.notifier);
-	struct xe_vma *vma = &uvma->vma;
-	struct xe_vm *vm = xe_vma_vm(vma);
-
-	xe_assert(vm->xe, xe_vma_is_userptr(vma));
-	trace_xe_vma_userptr_invalidate(vma);
-
-	if (!mmu_notifier_range_blockable(range))
-		return false;
-
-	vm_dbg(&xe_vma_vm(vma)->xe->drm,
-	       "NOTIFIER: addr=0x%016llx, range=0x%016llx",
-		xe_vma_start(vma), xe_vma_size(vma));
-
-	down_write(&vm->userptr.notifier_lock);
-	mmu_interval_set_seq(mni, cur_seq);
-
-	__vma_userptr_invalidate(vm, uvma);
-	up_write(&vm->userptr.notifier_lock);
-	trace_xe_vma_userptr_invalidate_complete(vma);
-
-	return true;
-}
-
-static const struct mmu_interval_notifier_ops vma_userptr_notifier_ops = {
-	.invalidate = vma_userptr_invalidate,
-};
-
-#if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
-/**
- * xe_vma_userptr_force_invalidate() - force invalidate a userptr
- * @uvma: The userptr vma to invalidate
- *
- * Perform a forced userptr invalidation for testing purposes.
- */
-void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma)
-{
-	struct xe_vm *vm = xe_vma_vm(&uvma->vma);
-
-	/* Protect against concurrent userptr pinning */
-	lockdep_assert_held(&vm->lock);
-	/* Protect against concurrent notifiers */
-	lockdep_assert_held(&vm->userptr.notifier_lock);
-	/*
-	 * Protect against concurrent instances of this function and
-	 * the critical exec sections
-	 */
-	xe_vm_assert_held(vm);
-
-	if (!mmu_interval_read_retry(&uvma->userptr.notifier,
-				     uvma->userptr.notifier_seq))
-		uvma->userptr.notifier_seq -= 2;
-	__vma_userptr_invalidate(vm, uvma);
-}
-#endif
-
-int xe_vm_userptr_pin(struct xe_vm *vm)
-{
-	struct xe_userptr_vma *uvma, *next;
-	int err = 0;
-
-	xe_assert(vm->xe, !xe_vm_in_fault_mode(vm));
-	lockdep_assert_held_write(&vm->lock);
-
-	/* Collect invalidated userptrs */
-	spin_lock(&vm->userptr.invalidated_lock);
-	xe_assert(vm->xe, list_empty(&vm->userptr.repin_list));
-	list_for_each_entry_safe(uvma, next, &vm->userptr.invalidated,
-				 userptr.invalidate_link) {
-		list_del_init(&uvma->userptr.invalidate_link);
-		list_add_tail(&uvma->userptr.repin_link,
-			      &vm->userptr.repin_list);
-	}
-	spin_unlock(&vm->userptr.invalidated_lock);
-
-	/* Pin and move to bind list */
-	list_for_each_entry_safe(uvma, next, &vm->userptr.repin_list,
-				 userptr.repin_link) {
-		err = xe_vma_userptr_pin_pages(uvma);
-		if (err == -EFAULT) {
-			list_del_init(&uvma->userptr.repin_link);
-			/*
-			 * We might have already done the pin once already, but
-			 * then had to retry before the re-bind happened, due
-			 * some other condition in the caller, but in the
-			 * meantime the userptr got dinged by the notifier such
-			 * that we need to revalidate here, but this time we hit
-			 * the EFAULT. In such a case make sure we remove
-			 * ourselves from the rebind list to avoid going down in
-			 * flames.
-			 */
-			if (!list_empty(&uvma->vma.combined_links.rebind))
-				list_del_init(&uvma->vma.combined_links.rebind);
-
-			/* Wait for pending binds */
-			xe_vm_lock(vm, false);
-			dma_resv_wait_timeout(xe_vm_resv(vm),
-					      DMA_RESV_USAGE_BOOKKEEP,
-					      false, MAX_SCHEDULE_TIMEOUT);
-
-			down_read(&vm->userptr.notifier_lock);
-			err = xe_vm_invalidate_vma(&uvma->vma);
-			up_read(&vm->userptr.notifier_lock);
-			xe_vm_unlock(vm);
-			if (err)
-				break;
-		} else {
-			if (err)
-				break;
-
-			list_del_init(&uvma->userptr.repin_link);
-			list_move_tail(&uvma->vma.combined_links.rebind,
-				       &vm->rebind_list);
-		}
-	}
-
-	if (err) {
-		down_write(&vm->userptr.notifier_lock);
-		spin_lock(&vm->userptr.invalidated_lock);
-		list_for_each_entry_safe(uvma, next, &vm->userptr.repin_list,
-					 userptr.repin_link) {
-			list_del_init(&uvma->userptr.repin_link);
-			list_move_tail(&uvma->userptr.invalidate_link,
-				       &vm->userptr.invalidated);
-		}
-		spin_unlock(&vm->userptr.invalidated_lock);
-		up_write(&vm->userptr.notifier_lock);
-	}
-	return err;
-}
-
-/**
- * xe_vm_userptr_check_repin() - Check whether the VM might have userptrs
- * that need repinning.
- * @vm: The VM.
- *
- * This function does an advisory check for whether the VM has userptrs that
- * need repinning.
- *
- * Return: 0 if there are no indications of userptrs needing repinning,
- * -EAGAIN if there are.
- */
-int xe_vm_userptr_check_repin(struct xe_vm *vm)
-{
-	return (list_empty_careful(&vm->userptr.repin_list) &&
-		list_empty_careful(&vm->userptr.invalidated)) ? 0 : -EAGAIN;
-}
-
 static int xe_vma_ops_alloc(struct xe_vma_ops *vops, bool array_of_binds)
 {
 	int i;
@@ -1245,25 +999,17 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm,
 		drm_gpuvm_bo_put(vm_bo);
 	} else /* userptr or null */ {
 		if (!is_null && !is_cpu_addr_mirror) {
-			struct xe_userptr *userptr = &to_userptr_vma(vma)->userptr;
+			struct xe_userptr_vma *uvma = to_userptr_vma(vma);
 			u64 size = end - start + 1;
 			int err;
 
-			INIT_LIST_HEAD(&userptr->invalidate_link);
-			INIT_LIST_HEAD(&userptr->repin_link);
 			vma->gpuva.gem.offset = bo_offset_or_userptr;
-			mutex_init(&userptr->unmap_mutex);
 
-			err = mmu_interval_notifier_insert(&userptr->notifier,
-							   current->mm,
-							   xe_vma_userptr(vma), size,
-							   &vma_userptr_notifier_ops);
+			err = xe_userptr_setup(uvma, xe_vma_userptr(vma), size);
 			if (err) {
 				xe_vma_free(vma);
 				return ERR_PTR(err);
 			}
-
-			userptr->notifier_seq = LONG_MAX;
 		}
 
 		xe_vm_get(vm);
@@ -1283,18 +1029,8 @@ static void xe_vma_destroy_late(struct xe_vma *vma)
 
 	if (xe_vma_is_userptr(vma)) {
 		struct xe_userptr_vma *uvma = to_userptr_vma(vma);
-		struct xe_userptr *userptr = &uvma->userptr;
 
-		if (userptr->sg)
-			xe_hmm_userptr_free_sg(uvma);
-
-		/*
-		 * Since userptr pages are not pinned, we can't remove
-		 * the notifier until we're sure the GPU is not accessing
-		 * them anymore
-		 */
-		mmu_interval_notifier_remove(&userptr->notifier);
-		mutex_destroy(&userptr->unmap_mutex);
+		xe_userptr_remove(uvma);
 		xe_vm_put(vm);
 	} else if (xe_vma_is_null(vma) || xe_vma_is_cpu_addr_mirror(vma)) {
 		xe_vm_put(vm);
@@ -1331,11 +1067,7 @@ static void xe_vma_destroy(struct xe_vma *vma, struct dma_fence *fence)
 
 	if (xe_vma_is_userptr(vma)) {
 		xe_assert(vm->xe, vma->gpuva.flags & XE_VMA_DESTROYED);
-
-		spin_lock(&vm->userptr.invalidated_lock);
-		xe_assert(vm->xe, list_empty(&to_userptr_vma(vma)->userptr.repin_link));
-		list_del(&to_userptr_vma(vma)->userptr.invalidate_link);
-		spin_unlock(&vm->userptr.invalidated_lock);
+		xe_userptr_destroy(to_userptr_vma(vma));
 	} else if (!xe_vma_is_null(vma) && !xe_vma_is_cpu_addr_mirror(vma)) {
 		xe_bo_assert_held(xe_vma_bo(vma));
 
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index b3e5bec0fa58..bec9bf89d937 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -220,12 +220,6 @@ static inline bool xe_vm_in_preempt_fence_mode(struct xe_vm *vm)
 int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q);
 void xe_vm_remove_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q);
 
-int xe_vm_userptr_pin(struct xe_vm *vm);
-
-int __xe_vm_userptr_needs_repin(struct xe_vm *vm);
-
-int xe_vm_userptr_check_repin(struct xe_vm *vm);
-
 int xe_vm_rebind(struct xe_vm *vm, bool rebind_worker);
 struct dma_fence *xe_vma_rebind(struct xe_vm *vm, struct xe_vma *vma,
 				u8 tile_mask);
@@ -266,10 +260,6 @@ static inline void xe_vm_reactivate_rebind(struct xe_vm *vm)
 	}
 }
 
-int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma);
-
-int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma);
-
 bool xe_vm_validate_should_retry(struct drm_exec *exec, int err, ktime_t *end);
 
 int xe_vm_lock_vma(struct drm_exec *exec, struct xe_vma *vma);
@@ -391,11 +381,4 @@ static inline bool xe_vm_is_validating(struct xe_vm *vm)
 #define xe_vm_has_valid_gpu_mapping(tile, tile_present, tile_invalidated)	\
 	((READ_ONCE(tile_present) & ~READ_ONCE(tile_invalidated)) & BIT((tile)->id))
 
-#if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
-void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma);
-#else
-static inline void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma)
-{
-}
-#endif
 #endif
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index b5108d010786..aee050c295db 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -17,6 +17,7 @@
 #include "xe_device_types.h"
 #include "xe_pt_types.h"
 #include "xe_range_fence.h"
+#include "xe_userptr.h"
 
 struct xe_bo;
 struct xe_svm_range;
@@ -46,37 +47,6 @@ struct xe_vm_pgtable_update_op;
 #define XE_VMA_DUMPABLE		(DRM_GPUVA_USERBITS << 8)
 #define XE_VMA_SYSTEM_ALLOCATOR	(DRM_GPUVA_USERBITS << 9)
 
-/** struct xe_userptr - User pointer */
-struct xe_userptr {
-	/** @invalidate_link: Link for the vm::userptr.invalidated list */
-	struct list_head invalidate_link;
-	/** @userptr: link into VM repin list if userptr. */
-	struct list_head repin_link;
-	/**
-	 * @notifier: MMU notifier for user pointer (invalidation call back)
-	 */
-	struct mmu_interval_notifier notifier;
-	/** @sgt: storage for a scatter gather table */
-	struct sg_table sgt;
-	/** @sg: allocated scatter gather table */
-	struct sg_table *sg;
-	/** @notifier_seq: notifier sequence number */
-	unsigned long notifier_seq;
-	/** @unmap_mutex: Mutex protecting dma-unmapping */
-	struct mutex unmap_mutex;
-	/**
-	 * @initial_bind: user pointer has been bound at least once.
-	 * write: vm->userptr.notifier_lock in read mode and vm->resv held.
-	 * read: vm->userptr.notifier_lock in write mode or vm->resv held.
-	 */
-	bool initial_bind;
-	/** @mapped: Whether the @sgt sg-table is dma-mapped. Protected by @unmap_mutex. */
-	bool mapped;
-#if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
-	u32 divisor;
-#endif
-};
-
 /**
  * struct xe_vma_mem_attr - memory attributes associated with vma
  */
@@ -289,33 +259,7 @@ struct xe_vm {
 	const struct xe_pt_ops *pt_ops;
 
 	/** @userptr: user pointer state */
-	struct {
-		/**
-		 * @userptr.repin_list: list of VMAs which are user pointers,
-		 * and needs repinning. Protected by @lock.
-		 */
-		struct list_head repin_list;
-		/**
-		 * @notifier_lock: protects notifier in write mode and
-		 * submission in read mode.
-		 */
-		struct rw_semaphore notifier_lock;
-		/**
-		 * @userptr.invalidated_lock: Protects the
-		 * @userptr.invalidated list.
-		 */
-		spinlock_t invalidated_lock;
-		/**
-		 * @userptr.invalidated: List of invalidated userptrs, not yet
-		 * picked
-		 * up for revalidation. Protected from access with the
-		 * @invalidated_lock. Removing items from the list
-		 * additionally requires @lock in write mode, and adding
-		 * items to the list requires either the @userptr.notifier_lock in
-		 * write mode, OR @lock in write mode.
-		 */
-		struct list_head invalidated;
-	} userptr;
+	struct xe_userptr_vm userptr;
 
 	/** @preempt: preempt state */
 	struct {

From 9e978741488261e117bb50e5dfcf8e4080990958 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:38 +0100
Subject: [PATCH 0954/1455] drm/xe/userptr: replace xe_hmm with gpusvm
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Goal here is cut over to gpusvm and remove xe_hmm, relying instead on
common code. The core facilities we need are get_pages(), unmap_pages()
and free_pages() for a given useptr range, plus a vm level notifier
lock, which is now provided by gpusvm.

v2:
  - Reuse the same SVM vm struct we use for full SVM, that way we can
    use the same lock (Matt B & Himal)
v3:
  - Re-use svm_init/fini for userptr.
v4:
  - Allow building xe without userptr if we are missing DRM_GPUSVM
    config. (Matt B)
  - Always make .read_only match xe_vma_read_only() for the ctx. (Dafna)
v5:
  - Fix missing conversion with CONFIG_DRM_XE_USERPTR_INVAL_INJECT
v6:
  - Convert the new user in xe_vm_madise.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Dafna Hirschfeld <dafna.hirschfeld@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-17-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/Kconfig         |   2 +-
 drivers/gpu/drm/xe/Kconfig.debug   |   1 +
 drivers/gpu/drm/xe/Makefile        |   3 +-
 drivers/gpu/drm/xe/xe_exec.c       |   5 +-
 drivers/gpu/drm/xe/xe_hmm.c        | 325 -----------------------------
 drivers/gpu/drm/xe/xe_hmm.h        |  18 --
 drivers/gpu/drm/xe/xe_pt.c         |  24 +--
 drivers/gpu/drm/xe/xe_svm.c        |  32 +--
 drivers/gpu/drm/xe/xe_svm.h        |  57 +++--
 drivers/gpu/drm/xe/xe_userptr.c    |  54 +++--
 drivers/gpu/drm/xe/xe_userptr.h    |  46 ++--
 drivers/gpu/drm/xe/xe_vm.c         |  48 ++---
 drivers/gpu/drm/xe/xe_vm_madvise.c |  24 +--
 drivers/gpu/drm/xe/xe_vm_types.h   |   8 +-
 14 files changed, 173 insertions(+), 474 deletions(-)
 delete mode 100644 drivers/gpu/drm/xe/xe_hmm.c
 delete mode 100644 drivers/gpu/drm/xe/xe_hmm.h

diff --git a/drivers/gpu/drm/xe/Kconfig b/drivers/gpu/drm/xe/Kconfig
index 714d5702dfd7..7219f6b884b6 100644
--- a/drivers/gpu/drm/xe/Kconfig
+++ b/drivers/gpu/drm/xe/Kconfig
@@ -40,12 +40,12 @@ config DRM_XE
 	select DRM_TTM
 	select DRM_TTM_HELPER
 	select DRM_EXEC
+	select DRM_GPUSVM if !UML && DEVICE_PRIVATE
 	select DRM_GPUVM
 	select DRM_SCHED
 	select MMU_NOTIFIER
 	select WANT_DEV_COREDUMP
 	select AUXILIARY_BUS
-	select HMM_MIRROR
 	select REGMAP if I2C
 	help
 	  Driver for Intel Xe2 series GPUs and later. Experimental support
diff --git a/drivers/gpu/drm/xe/Kconfig.debug b/drivers/gpu/drm/xe/Kconfig.debug
index 01735c6ece8b..87902b4bd6d3 100644
--- a/drivers/gpu/drm/xe/Kconfig.debug
+++ b/drivers/gpu/drm/xe/Kconfig.debug
@@ -104,6 +104,7 @@ config DRM_XE_DEBUG_GUC
 
 config DRM_XE_USERPTR_INVAL_INJECT
        bool "Inject userptr invalidation -EINVAL errors"
+       depends on DRM_GPUSVM
        default n
        help
          Choose this option when debugging error paths that
diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index ae833a9e1bb1..d6bd139c5839 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -130,7 +130,6 @@ xe-y += xe_bb.o \
 	xe_tuning.o \
 	xe_uc.o \
 	xe_uc_fw.o \
-	xe_userptr.o \
 	xe_vm.o \
 	xe_vm_madvise.o \
 	xe_vram.o \
@@ -141,8 +140,8 @@ xe-y += xe_bb.o \
 	xe_wopcm.o
 
 xe-$(CONFIG_I2C)	+= xe_i2c.o
-xe-$(CONFIG_HMM_MIRROR) += xe_hmm.o
 xe-$(CONFIG_DRM_XE_GPUSVM) += xe_svm.o
+xe-$(CONFIG_DRM_GPUSVM) += xe_userptr.o
 
 # graphics hardware monitoring (HWMON) support
 xe-$(CONFIG_HWMON) += xe_hwmon.o
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index 44364c042ad7..25a59b6934f6 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -19,6 +19,7 @@
 #include "xe_ring_ops_types.h"
 #include "xe_sched_job.h"
 #include "xe_sync.h"
+#include "xe_svm.h"
 #include "xe_vm.h"
 
 /**
@@ -294,7 +295,7 @@ retry:
 		if (err)
 			goto err_put_job;
 
-		err = down_read_interruptible(&vm->userptr.notifier_lock);
+		err = xe_svm_notifier_lock_interruptible(vm);
 		if (err)
 			goto err_put_job;
 
@@ -336,7 +337,7 @@ retry:
 
 err_repin:
 	if (!xe_vm_in_lr_mode(vm))
-		up_read(&vm->userptr.notifier_lock);
+		xe_svm_notifier_unlock(vm);
 err_put_job:
 	if (err)
 		xe_sched_job_put(job);
diff --git a/drivers/gpu/drm/xe/xe_hmm.c b/drivers/gpu/drm/xe/xe_hmm.c
deleted file mode 100644
index 57b71956ddf4..000000000000
--- a/drivers/gpu/drm/xe/xe_hmm.c
+++ /dev/null
@@ -1,325 +0,0 @@
-// SPDX-License-Identifier: MIT
-/*
- * Copyright © 2024 Intel Corporation
- */
-
-#include <linux/scatterlist.h>
-#include <linux/mmu_notifier.h>
-#include <linux/dma-mapping.h>
-#include <linux/memremap.h>
-#include <linux/swap.h>
-#include <linux/hmm.h>
-#include <linux/mm.h>
-#include "xe_hmm.h"
-#include "xe_vm.h"
-#include "xe_bo.h"
-
-static u64 xe_npages_in_range(unsigned long start, unsigned long end)
-{
-	return (end - start) >> PAGE_SHIFT;
-}
-
-static int xe_alloc_sg(struct xe_device *xe, struct sg_table *st,
-		       struct hmm_range *range, struct rw_semaphore *notifier_sem)
-{
-	unsigned long i, npages, hmm_pfn;
-	unsigned long num_chunks = 0;
-	int ret;
-
-	/* HMM docs says this is needed. */
-	ret = down_read_interruptible(notifier_sem);
-	if (ret)
-		return ret;
-
-	if (mmu_interval_read_retry(range->notifier, range->notifier_seq)) {
-		up_read(notifier_sem);
-		return -EAGAIN;
-	}
-
-	npages = xe_npages_in_range(range->start, range->end);
-	for (i = 0; i < npages;) {
-		unsigned long len;
-
-		hmm_pfn = range->hmm_pfns[i];
-		xe_assert(xe, hmm_pfn & HMM_PFN_VALID);
-
-		len = 1UL << hmm_pfn_to_map_order(hmm_pfn);
-
-		/* If order > 0 the page may extend beyond range->start */
-		len -= (hmm_pfn & ~HMM_PFN_FLAGS) & (len - 1);
-		i += len;
-		num_chunks++;
-	}
-	up_read(notifier_sem);
-
-	return sg_alloc_table(st, num_chunks, GFP_KERNEL);
-}
-
-/**
- * xe_build_sg() - build a scatter gather table for all the physical pages/pfn
- * in a hmm_range. dma-map pages if necessary. dma-address is save in sg table
- * and will be used to program GPU page table later.
- * @xe: the xe device who will access the dma-address in sg table
- * @range: the hmm range that we build the sg table from. range->hmm_pfns[]
- * has the pfn numbers of pages that back up this hmm address range.
- * @st: pointer to the sg table.
- * @notifier_sem: The xe notifier lock.
- * @write: whether we write to this range. This decides dma map direction
- * for system pages. If write we map it bi-diretional; otherwise
- * DMA_TO_DEVICE
- *
- * All the contiguous pfns will be collapsed into one entry in
- * the scatter gather table. This is for the purpose of efficiently
- * programming GPU page table.
- *
- * The dma_address in the sg table will later be used by GPU to
- * access memory. So if the memory is system memory, we need to
- * do a dma-mapping so it can be accessed by GPU/DMA.
- *
- * FIXME: This function currently only support pages in system
- * memory. If the memory is GPU local memory (of the GPU who
- * is going to access memory), we need gpu dpa (device physical
- * address), and there is no need of dma-mapping. This is TBD.
- *
- * FIXME: dma-mapping for peer gpu device to access remote gpu's
- * memory. Add this when you support p2p
- *
- * This function allocates the storage of the sg table. It is
- * caller's responsibility to free it calling sg_free_table.
- *
- * Returns 0 if successful; -ENOMEM if fails to allocate memory
- */
-static int xe_build_sg(struct xe_device *xe, struct hmm_range *range,
-		       struct sg_table *st,
-		       struct rw_semaphore *notifier_sem,
-		       bool write)
-{
-	unsigned long npages = xe_npages_in_range(range->start, range->end);
-	struct device *dev = xe->drm.dev;
-	struct scatterlist *sgl;
-	struct page *page;
-	unsigned long i, j;
-
-	lockdep_assert_held(notifier_sem);
-
-	i = 0;
-	for_each_sg(st->sgl, sgl, st->nents, j) {
-		unsigned long hmm_pfn, size;
-
-		hmm_pfn = range->hmm_pfns[i];
-		page = hmm_pfn_to_page(hmm_pfn);
-		xe_assert(xe, !is_device_private_page(page));
-
-		size = 1UL << hmm_pfn_to_map_order(hmm_pfn);
-		size -= page_to_pfn(page) & (size - 1);
-		i += size;
-
-		if (unlikely(j == st->nents - 1)) {
-			xe_assert(xe, i >= npages);
-			if (i > npages)
-				size -= (i - npages);
-
-			sg_mark_end(sgl);
-		} else {
-			xe_assert(xe, i < npages);
-		}
-
-		sg_set_page(sgl, page, size << PAGE_SHIFT, 0);
-	}
-
-	return dma_map_sgtable(dev, st, write ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE,
-			       DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_NO_KERNEL_MAPPING);
-}
-
-static void xe_hmm_userptr_set_mapped(struct xe_userptr_vma *uvma)
-{
-	struct xe_userptr *userptr = &uvma->userptr;
-	struct xe_vm *vm = xe_vma_vm(&uvma->vma);
-
-	lockdep_assert_held_write(&vm->lock);
-	lockdep_assert_held(&vm->userptr.notifier_lock);
-
-	mutex_lock(&userptr->unmap_mutex);
-	xe_assert(vm->xe, !userptr->mapped);
-	userptr->mapped = true;
-	mutex_unlock(&userptr->unmap_mutex);
-}
-
-void xe_hmm_userptr_unmap(struct xe_userptr_vma *uvma)
-{
-	struct xe_userptr *userptr = &uvma->userptr;
-	struct xe_vma *vma = &uvma->vma;
-	bool write = !xe_vma_read_only(vma);
-	struct xe_vm *vm = xe_vma_vm(vma);
-	struct xe_device *xe = vm->xe;
-
-	if (!lockdep_is_held_type(&vm->userptr.notifier_lock, 0) &&
-	    !lockdep_is_held_type(&vm->lock, 0) &&
-	    !(vma->gpuva.flags & XE_VMA_DESTROYED)) {
-		/* Don't unmap in exec critical section. */
-		xe_vm_assert_held(vm);
-		/* Don't unmap while mapping the sg. */
-		lockdep_assert_held(&vm->lock);
-	}
-
-	mutex_lock(&userptr->unmap_mutex);
-	if (userptr->sg && userptr->mapped)
-		dma_unmap_sgtable(xe->drm.dev, userptr->sg,
-				  write ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE, 0);
-	userptr->mapped = false;
-	mutex_unlock(&userptr->unmap_mutex);
-}
-
-/**
- * xe_hmm_userptr_free_sg() - Free the scatter gather table of userptr
- * @uvma: the userptr vma which hold the scatter gather table
- *
- * With function xe_userptr_populate_range, we allocate storage of
- * the userptr sg table. This is a helper function to free this
- * sg table, and dma unmap the address in the table.
- */
-void xe_hmm_userptr_free_sg(struct xe_userptr_vma *uvma)
-{
-	struct xe_userptr *userptr = &uvma->userptr;
-
-	xe_assert(xe_vma_vm(&uvma->vma)->xe, userptr->sg);
-	xe_hmm_userptr_unmap(uvma);
-	sg_free_table(userptr->sg);
-	userptr->sg = NULL;
-}
-
-/**
- * xe_hmm_userptr_populate_range() - Populate physical pages of a virtual
- * address range
- *
- * @uvma: userptr vma which has information of the range to populate.
- * @is_mm_mmap_locked: True if mmap_read_lock is already acquired by caller.
- *
- * This function populate the physical pages of a virtual
- * address range. The populated physical pages is saved in
- * userptr's sg table. It is similar to get_user_pages but call
- * hmm_range_fault.
- *
- * This function also read mmu notifier sequence # (
- * mmu_interval_read_begin), for the purpose of later
- * comparison (through mmu_interval_read_retry).
- *
- * This must be called with mmap read or write lock held.
- *
- * This function allocates the storage of the userptr sg table.
- * It is caller's responsibility to free it calling sg_free_table.
- *
- * returns: 0 for success; negative error no on failure
- */
-int xe_hmm_userptr_populate_range(struct xe_userptr_vma *uvma,
-				  bool is_mm_mmap_locked)
-{
-	unsigned long timeout =
-		jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
-	unsigned long *pfns;
-	struct xe_userptr *userptr;
-	struct xe_vma *vma = &uvma->vma;
-	u64 userptr_start = xe_vma_userptr(vma);
-	u64 userptr_end = userptr_start + xe_vma_size(vma);
-	struct xe_vm *vm = xe_vma_vm(vma);
-	struct hmm_range hmm_range = {
-		.pfn_flags_mask = 0, /* ignore pfns */
-		.default_flags = HMM_PFN_REQ_FAULT,
-		.start = userptr_start,
-		.end = userptr_end,
-		.notifier = &uvma->userptr.notifier,
-		.dev_private_owner = vm->xe,
-	};
-	bool write = !xe_vma_read_only(vma);
-	unsigned long notifier_seq;
-	u64 npages;
-	int ret;
-
-	userptr = &uvma->userptr;
-
-	if (is_mm_mmap_locked)
-		mmap_assert_locked(userptr->notifier.mm);
-
-	if (vma->gpuva.flags & XE_VMA_DESTROYED)
-		return 0;
-
-	notifier_seq = mmu_interval_read_begin(&userptr->notifier);
-	if (notifier_seq == userptr->notifier_seq)
-		return 0;
-
-	if (userptr->sg)
-		xe_hmm_userptr_free_sg(uvma);
-
-	npages = xe_npages_in_range(userptr_start, userptr_end);
-	pfns = kvmalloc_array(npages, sizeof(*pfns), GFP_KERNEL);
-	if (unlikely(!pfns))
-		return -ENOMEM;
-
-	if (write)
-		hmm_range.default_flags |= HMM_PFN_REQ_WRITE;
-
-	if (!mmget_not_zero(userptr->notifier.mm)) {
-		ret = -EFAULT;
-		goto free_pfns;
-	}
-
-	hmm_range.hmm_pfns = pfns;
-
-	while (true) {
-		hmm_range.notifier_seq = mmu_interval_read_begin(&userptr->notifier);
-
-		if (!is_mm_mmap_locked)
-			mmap_read_lock(userptr->notifier.mm);
-
-		ret = hmm_range_fault(&hmm_range);
-
-		if (!is_mm_mmap_locked)
-			mmap_read_unlock(userptr->notifier.mm);
-
-		if (ret == -EBUSY) {
-			if (time_after(jiffies, timeout))
-				break;
-
-			continue;
-		}
-		break;
-	}
-
-	mmput(userptr->notifier.mm);
-
-	if (ret)
-		goto free_pfns;
-
-	ret = xe_alloc_sg(vm->xe, &userptr->sgt, &hmm_range, &vm->userptr.notifier_lock);
-	if (ret)
-		goto free_pfns;
-
-	ret = down_read_interruptible(&vm->userptr.notifier_lock);
-	if (ret)
-		goto free_st;
-
-	if (mmu_interval_read_retry(hmm_range.notifier, hmm_range.notifier_seq)) {
-		ret = -EAGAIN;
-		goto out_unlock;
-	}
-
-	ret = xe_build_sg(vm->xe, &hmm_range, &userptr->sgt,
-			  &vm->userptr.notifier_lock, write);
-	if (ret)
-		goto out_unlock;
-
-	userptr->sg = &userptr->sgt;
-	xe_hmm_userptr_set_mapped(uvma);
-	userptr->notifier_seq = hmm_range.notifier_seq;
-	up_read(&vm->userptr.notifier_lock);
-	kvfree(pfns);
-	return 0;
-
-out_unlock:
-	up_read(&vm->userptr.notifier_lock);
-free_st:
-	sg_free_table(&userptr->sgt);
-free_pfns:
-	kvfree(pfns);
-	return ret;
-}
diff --git a/drivers/gpu/drm/xe/xe_hmm.h b/drivers/gpu/drm/xe/xe_hmm.h
deleted file mode 100644
index 0ea98d8e7bbc..000000000000
--- a/drivers/gpu/drm/xe/xe_hmm.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/* SPDX-License-Identifier: MIT
- *
- * Copyright © 2024 Intel Corporation
- */
-
-#ifndef _XE_HMM_H_
-#define _XE_HMM_H_
-
-#include <linux/types.h>
-
-struct xe_userptr_vma;
-
-int xe_hmm_userptr_populate_range(struct xe_userptr_vma *uvma, bool is_mm_mmap_locked);
-
-void xe_hmm_userptr_free_sg(struct xe_userptr_vma *uvma);
-
-void xe_hmm_userptr_unmap(struct xe_userptr_vma *uvma);
-#endif
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 9eac29cbdc48..382b25798bc3 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -761,8 +761,8 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
 
 	if (!xe_vma_is_null(vma) && !range) {
 		if (xe_vma_is_userptr(vma))
-			xe_res_first_sg(to_userptr_vma(vma)->userptr.sg, 0,
-					xe_vma_size(vma), &curs);
+			xe_res_first_dma(to_userptr_vma(vma)->userptr.pages.dma_addr, 0,
+					 xe_vma_size(vma), &curs);
 		else if (xe_bo_is_vram(bo) || xe_bo_is_stolen(bo))
 			xe_res_first(bo->ttm.resource, xe_vma_bo_offset(vma),
 				     xe_vma_size(vma), &curs);
@@ -915,7 +915,7 @@ bool xe_pt_zap_ptes(struct xe_tile *tile, struct xe_vma *vma)
 	if (xe_vma_bo(vma))
 		xe_bo_assert_held(xe_vma_bo(vma));
 	else if (xe_vma_is_userptr(vma))
-		lockdep_assert_held(&xe_vma_vm(vma)->userptr.notifier_lock);
+		lockdep_assert_held(&xe_vma_vm(vma)->svm.gpusvm.notifier_lock);
 
 	if (!(pt_mask & BIT(tile->id)))
 		return false;
@@ -1050,7 +1050,7 @@ static void xe_pt_commit_locks_assert(struct xe_vma *vma)
 	xe_pt_commit_prepare_locks_assert(vma);
 
 	if (xe_vma_is_userptr(vma))
-		lockdep_assert_held_read(&vm->userptr.notifier_lock);
+		xe_svm_assert_held_read(vm);
 }
 
 static void xe_pt_commit(struct xe_vma *vma,
@@ -1407,7 +1407,7 @@ static int vma_check_userptr(struct xe_vm *vm, struct xe_vma *vma,
 	struct xe_userptr_vma *uvma;
 	unsigned long notifier_seq;
 
-	lockdep_assert_held_read(&vm->userptr.notifier_lock);
+	xe_svm_assert_held_read(vm);
 
 	if (!xe_vma_is_userptr(vma))
 		return 0;
@@ -1416,7 +1416,7 @@ static int vma_check_userptr(struct xe_vm *vm, struct xe_vma *vma,
 	if (xe_pt_userptr_inject_eagain(uvma))
 		xe_vma_userptr_force_invalidate(uvma);
 
-	notifier_seq = uvma->userptr.notifier_seq;
+	notifier_seq = uvma->userptr.pages.notifier_seq;
 
 	if (!mmu_interval_read_retry(&uvma->userptr.notifier,
 				     notifier_seq))
@@ -1437,7 +1437,7 @@ static int op_check_userptr(struct xe_vm *vm, struct xe_vma_op *op,
 {
 	int err = 0;
 
-	lockdep_assert_held_read(&vm->userptr.notifier_lock);
+	xe_svm_assert_held_read(vm);
 
 	switch (op->base.op) {
 	case DRM_GPUVA_OP_MAP:
@@ -1478,12 +1478,12 @@ static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
 	if (err)
 		return err;
 
-	down_read(&vm->userptr.notifier_lock);
+	down_read(&vm->svm.gpusvm.notifier_lock);
 
 	list_for_each_entry(op, &vops->list, link) {
 		err = op_check_userptr(vm, op, pt_update_ops);
 		if (err) {
-			up_read(&vm->userptr.notifier_lock);
+			up_read(&vm->svm.gpusvm.notifier_lock);
 			break;
 		}
 	}
@@ -2200,7 +2200,7 @@ static void bind_op_commit(struct xe_vm *vm, struct xe_tile *tile,
 			   vma->tile_invalidated & ~BIT(tile->id));
 	vma->tile_staged &= ~BIT(tile->id);
 	if (xe_vma_is_userptr(vma)) {
-		lockdep_assert_held_read(&vm->userptr.notifier_lock);
+		xe_svm_assert_held_read(vm);
 		to_userptr_vma(vma)->userptr.initial_bind = true;
 	}
 
@@ -2236,7 +2236,7 @@ static void unbind_op_commit(struct xe_vm *vm, struct xe_tile *tile,
 	if (!vma->tile_present) {
 		list_del_init(&vma->combined_links.rebind);
 		if (xe_vma_is_userptr(vma)) {
-			lockdep_assert_held_read(&vm->userptr.notifier_lock);
+			xe_svm_assert_held_read(vm);
 
 			spin_lock(&vm->userptr.invalidated_lock);
 			list_del_init(&to_userptr_vma(vma)->userptr.invalidate_link);
@@ -2535,7 +2535,7 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	if (pt_update_ops->needs_svm_lock)
 		xe_svm_notifier_unlock(vm);
 	if (pt_update_ops->needs_userptr_lock)
-		up_read(&vm->userptr.notifier_lock);
+		up_read(&vm->svm.gpusvm.notifier_lock);
 
 	xe_tlb_inval_job_put(mjob);
 	xe_tlb_inval_job_put(ijob);
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index f30c89316812..c67967cdec79 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -739,22 +739,26 @@ int xe_svm_init(struct xe_vm *vm)
 {
 	int err;
 
-	spin_lock_init(&vm->svm.garbage_collector.lock);
-	INIT_LIST_HEAD(&vm->svm.garbage_collector.range_list);
-	INIT_WORK(&vm->svm.garbage_collector.work,
-		  xe_svm_garbage_collector_work_func);
+	if (vm->flags & XE_VM_FLAG_FAULT_MODE) {
+		spin_lock_init(&vm->svm.garbage_collector.lock);
+		INIT_LIST_HEAD(&vm->svm.garbage_collector.range_list);
+		INIT_WORK(&vm->svm.garbage_collector.work,
+			  xe_svm_garbage_collector_work_func);
 
-	err = drm_gpusvm_init(&vm->svm.gpusvm, "Xe SVM", &vm->xe->drm,
-			      current->mm, xe_svm_devm_owner(vm->xe), 0,
-			      vm->size, xe_modparam.svm_notifier_size * SZ_1M,
-			      &gpusvm_ops, fault_chunk_sizes,
-			      ARRAY_SIZE(fault_chunk_sizes));
-	if (err)
-		return err;
+		err = drm_gpusvm_init(&vm->svm.gpusvm, "Xe SVM", &vm->xe->drm,
+				      current->mm, xe_svm_devm_owner(vm->xe), 0,
+				      vm->size,
+				      xe_modparam.svm_notifier_size * SZ_1M,
+				      &gpusvm_ops, fault_chunk_sizes,
+				      ARRAY_SIZE(fault_chunk_sizes));
+		drm_gpusvm_driver_set_lock(&vm->svm.gpusvm, &vm->lock);
+	} else {
+		err = drm_gpusvm_init(&vm->svm.gpusvm, "Xe SVM (simple)",
+				      &vm->xe->drm, NULL, NULL, 0, 0, 0, NULL,
+				      NULL, 0);
+	}
 
-	drm_gpusvm_driver_set_lock(&vm->svm.gpusvm, &vm->lock);
-
-	return 0;
+	return err;
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index f0e0e009c7fc..cef6ee7d6fe3 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -155,19 +155,11 @@ static inline unsigned long xe_svm_range_size(struct xe_svm_range *range)
 	return drm_gpusvm_range_size(&range->base);
 }
 
-#define xe_svm_assert_in_notifier(vm__) \
-	lockdep_assert_held_write(&(vm__)->svm.gpusvm.notifier_lock)
-
-#define xe_svm_notifier_lock(vm__)	\
-	drm_gpusvm_notifier_lock(&(vm__)->svm.gpusvm)
-
-#define xe_svm_notifier_unlock(vm__)	\
-	drm_gpusvm_notifier_unlock(&(vm__)->svm.gpusvm)
-
 void xe_svm_flush(struct xe_vm *vm);
 
 #else
 #include <linux/interval_tree.h>
+#include "xe_vm.h"
 
 struct drm_pagemap_addr;
 struct drm_gpusvm_ctx;
@@ -206,12 +198,21 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr)
 static inline
 int xe_svm_init(struct xe_vm *vm)
 {
+#if IS_ENABLED(CONFIG_DRM_GPUSVM)
+	return drm_gpusvm_init(&vm->svm.gpusvm, "Xe SVM (simple)", &vm->xe->drm,
+			       NULL, NULL, 0, 0, 0, NULL, NULL, 0);
+#else
 	return 0;
+#endif
 }
 
 static inline
 void xe_svm_fini(struct xe_vm *vm)
 {
+#if IS_ENABLED(CONFIG_DRM_GPUSVM)
+	xe_assert(vm->xe, xe_vm_is_closed(vm));
+	drm_gpusvm_fini(&vm->svm.gpusvm);
+#endif
 }
 
 static inline
@@ -328,19 +329,47 @@ struct drm_pagemap *xe_vma_resolve_pagemap(struct xe_vma *vma, struct xe_tile *t
 	return NULL;
 }
 
-#define xe_svm_assert_in_notifier(...) do {} while (0)
+static inline void xe_svm_flush(struct xe_vm *vm)
+{
+}
 #define xe_svm_range_has_dma_mapping(...) false
+#endif /* CONFIG_DRM_XE_GPUSVM */
+
+#if IS_ENABLED(CONFIG_DRM_GPUSVM) /* Need to support userptr without XE_GPUSVM */
+#define xe_svm_assert_in_notifier(vm__) \
+	lockdep_assert_held_write(&(vm__)->svm.gpusvm.notifier_lock)
+
+#define xe_svm_assert_held_read(vm__) \
+	lockdep_assert_held_read(&(vm__)->svm.gpusvm.notifier_lock)
+
+#define xe_svm_notifier_lock(vm__)	\
+	drm_gpusvm_notifier_lock(&(vm__)->svm.gpusvm)
+
+#define xe_svm_notifier_lock_interruptible(vm__)	\
+	down_read_interruptible(&(vm__)->svm.gpusvm.notifier_lock)
+
+#define xe_svm_notifier_unlock(vm__)	\
+	drm_gpusvm_notifier_unlock(&(vm__)->svm.gpusvm)
+
+#else
+#define xe_svm_assert_in_notifier(...) do {} while (0)
+
+static inline void xe_svm_assert_held_read(struct xe_vm *vm)
+{
+}
 
 static inline void xe_svm_notifier_lock(struct xe_vm *vm)
 {
 }
 
+static inline int xe_svm_notifier_lock_interruptible(struct xe_vm *vm)
+{
+	return 0;
+}
+
 static inline void xe_svm_notifier_unlock(struct xe_vm *vm)
 {
 }
+#endif /* CONFIG_DRM_GPUSVM */
 
-static inline void xe_svm_flush(struct xe_vm *vm)
-{
-}
-#endif
 #endif
diff --git a/drivers/gpu/drm/xe/xe_userptr.c b/drivers/gpu/drm/xe/xe_userptr.c
index 17ecff17c993..91d09af71ced 100644
--- a/drivers/gpu/drm/xe/xe_userptr.c
+++ b/drivers/gpu/drm/xe/xe_userptr.c
@@ -7,7 +7,6 @@
 
 #include <linux/mm.h>
 
-#include "xe_hmm.h"
 #include "xe_trace_bo.h"
 
 /**
@@ -25,7 +24,7 @@
 int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma)
 {
 	return mmu_interval_check_retry(&uvma->userptr.notifier,
-					uvma->userptr.notifier_seq) ?
+					uvma->userptr.pages.notifier_seq) ?
 		-EAGAIN : 0;
 }
 
@@ -35,14 +34,14 @@ int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma)
  * @vm: The VM.
  *
  * This function checks for whether the VM has userptrs that need repinning,
- * and provides a release-type barrier on the userptr.notifier_lock after
+ * and provides a release-type barrier on the svm.gpusvm.notifier_lock after
  * checking.
  *
  * Return: 0 if there are no userptrs needing repinning, -EAGAIN if there are.
  */
 int __xe_vm_userptr_needs_repin(struct xe_vm *vm)
 {
-	lockdep_assert_held_read(&vm->userptr.notifier_lock);
+	lockdep_assert_held_read(&vm->svm.gpusvm.notifier_lock);
 
 	return (list_empty(&vm->userptr.repin_list) &&
 		list_empty(&vm->userptr.invalidated)) ? 0 : -EAGAIN;
@@ -53,11 +52,22 @@ int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma)
 	struct xe_vma *vma = &uvma->vma;
 	struct xe_vm *vm = xe_vma_vm(vma);
 	struct xe_device *xe = vm->xe;
+	struct drm_gpusvm_ctx ctx = {
+		.read_only = xe_vma_read_only(vma),
+	};
 
 	lockdep_assert_held(&vm->lock);
 	xe_assert(xe, xe_vma_is_userptr(vma));
 
-	return xe_hmm_userptr_populate_range(uvma, false);
+	if (vma->gpuva.flags & XE_VMA_DESTROYED)
+		return 0;
+
+	return drm_gpusvm_get_pages(&vm->svm.gpusvm, &uvma->userptr.pages,
+				    uvma->userptr.notifier.mm,
+				    &uvma->userptr.notifier,
+				    xe_vma_userptr(vma),
+				    xe_vma_userptr(vma) + xe_vma_size(vma),
+				    &ctx);
 }
 
 static void __vma_userptr_invalidate(struct xe_vm *vm, struct xe_userptr_vma *uvma)
@@ -66,6 +76,10 @@ static void __vma_userptr_invalidate(struct xe_vm *vm, struct xe_userptr_vma *uv
 	struct xe_vma *vma = &uvma->vma;
 	struct dma_resv_iter cursor;
 	struct dma_fence *fence;
+	struct drm_gpusvm_ctx ctx = {
+		.in_notifier = true,
+		.read_only = xe_vma_read_only(vma),
+	};
 	long err;
 
 	/*
@@ -102,7 +116,8 @@ static void __vma_userptr_invalidate(struct xe_vm *vm, struct xe_userptr_vma *uv
 		XE_WARN_ON(err);
 	}
 
-	xe_hmm_userptr_unmap(uvma);
+	drm_gpusvm_unmap_pages(&vm->svm.gpusvm, &uvma->userptr.pages,
+			       xe_vma_size(vma) >> PAGE_SHIFT, &ctx);
 }
 
 static bool vma_userptr_invalidate(struct mmu_interval_notifier *mni,
@@ -123,11 +138,11 @@ static bool vma_userptr_invalidate(struct mmu_interval_notifier *mni,
 	       "NOTIFIER: addr=0x%016llx, range=0x%016llx",
 		xe_vma_start(vma), xe_vma_size(vma));
 
-	down_write(&vm->userptr.notifier_lock);
+	down_write(&vm->svm.gpusvm.notifier_lock);
 	mmu_interval_set_seq(mni, cur_seq);
 
 	__vma_userptr_invalidate(vm, uvma);
-	up_write(&vm->userptr.notifier_lock);
+	up_write(&vm->svm.gpusvm.notifier_lock);
 	trace_xe_vma_userptr_invalidate_complete(vma);
 
 	return true;
@@ -151,7 +166,7 @@ void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma)
 	/* Protect against concurrent userptr pinning */
 	lockdep_assert_held(&vm->lock);
 	/* Protect against concurrent notifiers */
-	lockdep_assert_held(&vm->userptr.notifier_lock);
+	lockdep_assert_held(&vm->svm.gpusvm.notifier_lock);
 	/*
 	 * Protect against concurrent instances of this function and
 	 * the critical exec sections
@@ -159,8 +174,8 @@ void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma)
 	xe_vm_assert_held(vm);
 
 	if (!mmu_interval_read_retry(&uvma->userptr.notifier,
-				     uvma->userptr.notifier_seq))
-		uvma->userptr.notifier_seq -= 2;
+				     uvma->userptr.pages.notifier_seq))
+		uvma->userptr.pages.notifier_seq -= 2;
 	__vma_userptr_invalidate(vm, uvma);
 }
 #endif
@@ -209,9 +224,9 @@ int xe_vm_userptr_pin(struct xe_vm *vm)
 					      DMA_RESV_USAGE_BOOKKEEP,
 					      false, MAX_SCHEDULE_TIMEOUT);
 
-			down_read(&vm->userptr.notifier_lock);
+			down_read(&vm->svm.gpusvm.notifier_lock);
 			err = xe_vm_invalidate_vma(&uvma->vma);
-			up_read(&vm->userptr.notifier_lock);
+			up_read(&vm->svm.gpusvm.notifier_lock);
 			xe_vm_unlock(vm);
 			if (err)
 				break;
@@ -226,7 +241,7 @@ int xe_vm_userptr_pin(struct xe_vm *vm)
 	}
 
 	if (err) {
-		down_write(&vm->userptr.notifier_lock);
+		down_write(&vm->svm.gpusvm.notifier_lock);
 		spin_lock(&vm->userptr.invalidated_lock);
 		list_for_each_entry_safe(uvma, next, &vm->userptr.repin_list,
 					 userptr.repin_link) {
@@ -235,7 +250,7 @@ int xe_vm_userptr_pin(struct xe_vm *vm)
 				       &vm->userptr.invalidated);
 		}
 		spin_unlock(&vm->userptr.invalidated_lock);
-		up_write(&vm->userptr.notifier_lock);
+		up_write(&vm->svm.gpusvm.notifier_lock);
 	}
 	return err;
 }
@@ -265,7 +280,6 @@ int xe_userptr_setup(struct xe_userptr_vma *uvma, unsigned long start,
 
 	INIT_LIST_HEAD(&userptr->invalidate_link);
 	INIT_LIST_HEAD(&userptr->repin_link);
-	mutex_init(&userptr->unmap_mutex);
 
 	err = mmu_interval_notifier_insert(&userptr->notifier, current->mm,
 					   start, range,
@@ -273,17 +287,18 @@ int xe_userptr_setup(struct xe_userptr_vma *uvma, unsigned long start,
 	if (err)
 		return err;
 
-	userptr->notifier_seq = LONG_MAX;
+	userptr->pages.notifier_seq = LONG_MAX;
 
 	return 0;
 }
 
 void xe_userptr_remove(struct xe_userptr_vma *uvma)
 {
+	struct xe_vm *vm = xe_vma_vm(&uvma->vma);
 	struct xe_userptr *userptr = &uvma->userptr;
 
-	if (userptr->sg)
-		xe_hmm_userptr_free_sg(uvma);
+	drm_gpusvm_free_pages(&vm->svm.gpusvm, &uvma->userptr.pages,
+			      xe_vma_size(&uvma->vma) >> PAGE_SHIFT);
 
 	/*
 	 * Since userptr pages are not pinned, we can't remove
@@ -291,7 +306,6 @@ void xe_userptr_remove(struct xe_userptr_vma *uvma)
 	 * them anymore
 	 */
 	mmu_interval_notifier_remove(&userptr->notifier);
-	mutex_destroy(&userptr->unmap_mutex);
 }
 
 void xe_userptr_destroy(struct xe_userptr_vma *uvma)
diff --git a/drivers/gpu/drm/xe/xe_userptr.h b/drivers/gpu/drm/xe/xe_userptr.h
index dc49348ac42d..ef801234991e 100644
--- a/drivers/gpu/drm/xe/xe_userptr.h
+++ b/drivers/gpu/drm/xe/xe_userptr.h
@@ -12,6 +12,8 @@
 #include <linux/scatterlist.h>
 #include <linux/spinlock.h>
 
+#include <drm/drm_gpusvm.h>
+
 struct xe_vm;
 struct xe_vma;
 struct xe_userptr_vma;
@@ -23,11 +25,6 @@ struct xe_userptr_vm {
 	 * and needs repinning. Protected by @lock.
 	 */
 	struct list_head repin_list;
-	/**
-	 * @notifier_lock: protects notifier in write mode and
-	 * submission in read mode.o
-	 */
-	struct rw_semaphore notifier_lock;
 	/**
 	 * @userptr.invalidated_lock: Protects the
 	 * @userptr.invalidated list.
@@ -39,7 +36,7 @@ struct xe_userptr_vm {
 	 * up for revalidation. Protected from access with the
 	 * @invalidated_lock. Removing items from the list
 	 * additionally requires @lock in write mode, and adding
-	 * items to the list requires either the @userptr.notifier_lock in
+	 * items to the list requires either the @svm.gpusvm.notifier_lock in
 	 * write mode, OR @lock in write mode.
 	 */
 	struct list_head invalidated;
@@ -51,31 +48,27 @@ struct xe_userptr {
 	struct list_head invalidate_link;
 	/** @userptr: link into VM repin list if userptr. */
 	struct list_head repin_link;
+	/**
+	 * @pages: gpusvm pages for this user pointer.
+	 */
+	struct drm_gpusvm_pages pages;
 	/**
 	 * @notifier: MMU notifier for user pointer (invalidation call back)
 	 */
 	struct mmu_interval_notifier notifier;
-	/** @sgt: storage for a scatter gather table */
-	struct sg_table sgt;
-	/** @sg: allocated scatter gather table */
-	struct sg_table *sg;
-	/** @notifier_seq: notifier sequence number */
-	unsigned long notifier_seq;
-	/** @unmap_mutex: Mutex protecting dma-unmapping */
-	struct mutex unmap_mutex;
+
 	/**
 	 * @initial_bind: user pointer has been bound at least once.
-	 * write: vm->userptr.notifier_lock in read mode and vm->resv held.
-	 * read: vm->userptr.notifier_lock in write mode or vm->resv held.
+	 * write: vm->svm.gpusvm.notifier_lock in read mode and vm->resv held.
+	 * read: vm->svm.gpusvm.notifier_lock in write mode or vm->resv held.
 	 */
 	bool initial_bind;
-	/** @mapped: Whether the @sgt sg-table is dma-mapped. Protected by @unmap_mutex. */
-	bool mapped;
 #if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
 	u32 divisor;
 #endif
 };
 
+#if IS_ENABLED(CONFIG_DRM_GPUSVM)
 void xe_userptr_remove(struct xe_userptr_vma *uvma);
 int xe_userptr_setup(struct xe_userptr_vma *uvma, unsigned long start,
 		     unsigned long range);
@@ -86,6 +79,23 @@ int __xe_vm_userptr_needs_repin(struct xe_vm *vm);
 int xe_vm_userptr_check_repin(struct xe_vm *vm);
 int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma);
 int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma);
+#else
+static inline void xe_userptr_remove(struct xe_userptr_vma *uvma) {}
+
+static inline int xe_userptr_setup(struct xe_userptr_vma *uvma,
+				   unsigned long start, unsigned long range)
+{
+	return -ENODEV;
+}
+
+static inline void xe_userptr_destroy(struct xe_userptr_vma *uvma) {}
+
+static inline int xe_vm_userptr_pin(struct xe_vm *vm) { return 0; }
+static inline int __xe_vm_userptr_needs_repin(struct xe_vm *vm) { return 0; }
+static inline int xe_vm_userptr_check_repin(struct xe_vm *vm) { return 0; }
+static inline int xe_vma_userptr_pin_pages(struct xe_userptr_vma *uvma) { return -ENODEV; }
+static inline int xe_vma_userptr_check_repin(struct xe_userptr_vma *uvma) { return -ENODEV; };
+#endif
 
 #if IS_ENABLED(CONFIG_DRM_XE_USERPTR_INVAL_INJECT)
 void xe_vma_userptr_force_invalidate(struct xe_userptr_vma *uvma);
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index be3d4d023fa6..869f1db87d74 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -40,9 +40,7 @@
 #include "xe_tile.h"
 #include "xe_tlb_inval.h"
 #include "xe_trace_bo.h"
-#include "xe_userptr.h"
 #include "xe_wa.h"
-#include "xe_hmm.h"
 
 static struct drm_gem_object *xe_vm_obj(struct xe_vm *vm)
 {
@@ -220,7 +218,7 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 	++vm->preempt.num_exec_queues;
 	q->lr.pfence = pfence;
 
-	down_read(&vm->userptr.notifier_lock);
+	xe_svm_notifier_lock(vm);
 
 	drm_gpuvm_resv_add_fence(&vm->gpuvm, exec, pfence,
 				 DMA_RESV_USAGE_BOOKKEEP, DMA_RESV_USAGE_BOOKKEEP);
@@ -234,7 +232,7 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 	if (wait)
 		dma_fence_enable_sw_signaling(pfence);
 
-	up_read(&vm->userptr.notifier_lock);
+	xe_svm_notifier_unlock(vm);
 
 out_fini:
 	drm_exec_fini(exec);
@@ -498,9 +496,9 @@ retry:
 	(!(__tries)++ || __xe_vm_userptr_needs_repin(__vm)) : \
 	__xe_vm_userptr_needs_repin(__vm))
 
-	down_read(&vm->userptr.notifier_lock);
+	xe_svm_notifier_lock(vm);
 	if (retry_required(tries, vm)) {
-		up_read(&vm->userptr.notifier_lock);
+		xe_svm_notifier_unlock(vm);
 		err = -EAGAIN;
 		goto out_unlock;
 	}
@@ -514,7 +512,7 @@ retry:
 	/* Point of no return. */
 	arm_preempt_fences(vm, &preempt_fences);
 	resume_and_reinstall_preempt_fences(vm, &exec);
-	up_read(&vm->userptr.notifier_lock);
+	xe_svm_notifier_unlock(vm);
 
 out_unlock:
 	drm_exec_fini(&exec);
@@ -1449,7 +1447,6 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 
 	INIT_LIST_HEAD(&vm->userptr.repin_list);
 	INIT_LIST_HEAD(&vm->userptr.invalidated);
-	init_rwsem(&vm->userptr.notifier_lock);
 	spin_lock_init(&vm->userptr.invalidated_lock);
 
 	ttm_lru_bulk_move_init(&vm->lru_bulk_move);
@@ -1475,11 +1472,9 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 		xe_pm_runtime_get_noresume(xe);
 	}
 
-	if (flags & XE_VM_FLAG_FAULT_MODE) {
-		err = xe_svm_init(vm);
-		if (err)
-			goto err_no_resv;
-	}
+	err = xe_svm_init(vm);
+	if (err)
+		goto err_no_resv;
 
 	vm_resv_obj = drm_gpuvm_resv_object_alloc(&xe->drm);
 	if (!vm_resv_obj) {
@@ -1680,9 +1675,9 @@ void xe_vm_close_and_put(struct xe_vm *vm)
 		vma = gpuva_to_vma(gpuva);
 
 		if (xe_vma_has_no_bo(vma)) {
-			down_read(&vm->userptr.notifier_lock);
+			xe_svm_notifier_lock(vm);
 			vma->gpuva.flags |= XE_VMA_DESTROYED;
-			up_read(&vm->userptr.notifier_lock);
+			xe_svm_notifier_unlock(vm);
 		}
 
 		xe_vm_remove_vma(vm, vma);
@@ -1726,8 +1721,7 @@ void xe_vm_close_and_put(struct xe_vm *vm)
 		xe_vma_destroy_unlocked(vma);
 	}
 
-	if (xe_vm_in_fault_mode(vm))
-		xe_svm_fini(vm);
+	xe_svm_fini(vm);
 
 	up_write(&vm->lock);
 
@@ -2070,9 +2064,9 @@ static const u32 region_to_mem_type[] = {
 static void prep_vma_destroy(struct xe_vm *vm, struct xe_vma *vma,
 			     bool post_commit)
 {
-	down_read(&vm->userptr.notifier_lock);
+	xe_svm_notifier_lock(vm);
 	vma->gpuva.flags |= XE_VMA_DESTROYED;
-	up_read(&vm->userptr.notifier_lock);
+	xe_svm_notifier_unlock(vm);
 	if (post_commit)
 		xe_vm_remove_vma(vm, vma);
 }
@@ -2715,9 +2709,9 @@ static void xe_vma_op_unwind(struct xe_vm *vm, struct xe_vma_op *op,
 		struct xe_vma *vma = gpuva_to_vma(op->base.unmap.va);
 
 		if (vma) {
-			down_read(&vm->userptr.notifier_lock);
+			xe_svm_notifier_lock(vm);
 			vma->gpuva.flags &= ~XE_VMA_DESTROYED;
-			up_read(&vm->userptr.notifier_lock);
+			xe_svm_notifier_unlock(vm);
 			if (post_commit)
 				xe_vm_insert_vma(vm, vma);
 		}
@@ -2736,9 +2730,9 @@ static void xe_vma_op_unwind(struct xe_vm *vm, struct xe_vma_op *op,
 			xe_vma_destroy_unlocked(op->remap.next);
 		}
 		if (vma) {
-			down_read(&vm->userptr.notifier_lock);
+			xe_svm_notifier_lock(vm);
 			vma->gpuva.flags &= ~XE_VMA_DESTROYED;
-			up_read(&vm->userptr.notifier_lock);
+			xe_svm_notifier_unlock(vm);
 			if (post_commit)
 				xe_vm_insert_vma(vm, vma);
 		}
@@ -3313,6 +3307,8 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
 				 op == DRM_XE_VM_BIND_OP_MAP_USERPTR) ||
 		    XE_IOCTL_DBG(xe, coh_mode == XE_COH_NONE &&
 				 op == DRM_XE_VM_BIND_OP_MAP_USERPTR) ||
+		    XE_IOCTL_DBG(xe, op == DRM_XE_VM_BIND_OP_MAP_USERPTR &&
+				 !IS_ENABLED(CONFIG_DRM_GPUSVM)) ||
 		    XE_IOCTL_DBG(xe, obj &&
 				 op == DRM_XE_VM_BIND_OP_PREFETCH) ||
 		    XE_IOCTL_DBG(xe, prefetch_region &&
@@ -3858,13 +3854,13 @@ int xe_vm_invalidate_vma(struct xe_vma *vma)
 	 */
 	if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
 		if (xe_vma_is_userptr(vma)) {
-			lockdep_assert(lockdep_is_held_type(&vm->userptr.notifier_lock, 0) ||
-				       (lockdep_is_held_type(&vm->userptr.notifier_lock, 1) &&
+			lockdep_assert(lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 0) ||
+				       (lockdep_is_held_type(&vm->svm.gpusvm.notifier_lock, 1) &&
 					lockdep_is_held(&xe_vm_resv(vm)->lock.base)));
 
 			WARN_ON_ONCE(!mmu_interval_check_retry
 				     (&to_userptr_vma(vma)->userptr.notifier,
-				      to_userptr_vma(vma)->userptr.notifier_seq));
+				      to_userptr_vma(vma)->userptr.pages.notifier_seq));
 			WARN_ON_ONCE(!dma_resv_test_signaled(xe_vm_resv(vm),
 							     DMA_RESV_USAGE_BOOKKEEP));
 
diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 732577dfe519..153700743545 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -18,9 +18,8 @@ struct xe_vmas_in_madvise_range {
 	u64 range;
 	struct xe_vma **vmas;
 	int num_vmas;
-	bool has_svm_vmas;
 	bool has_bo_vmas;
-	bool has_userptr_vmas;
+	bool has_svm_userptr_vmas;
 };
 
 static int get_vmas(struct xe_vm *vm, struct xe_vmas_in_madvise_range *madvise_range)
@@ -46,10 +45,8 @@ static int get_vmas(struct xe_vm *vm, struct xe_vmas_in_madvise_range *madvise_r
 
 		if (xe_vma_bo(vma))
 			madvise_range->has_bo_vmas = true;
-		else if (xe_vma_is_cpu_addr_mirror(vma))
-			madvise_range->has_svm_vmas = true;
-		else if (xe_vma_is_userptr(vma))
-			madvise_range->has_userptr_vmas = true;
+		else if (xe_vma_is_cpu_addr_mirror(vma) || xe_vma_is_userptr(vma))
+			madvise_range->has_svm_userptr_vmas = true;
 
 		if (madvise_range->num_vmas == max_vmas) {
 			max_vmas <<= 1;
@@ -409,16 +406,10 @@ int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *fil
 		}
 	}
 
-	if (madvise_range.has_userptr_vmas) {
-		err = down_read_interruptible(&vm->userptr.notifier_lock);
-		if (err)
-			goto err_fini;
-	}
-
-	if (madvise_range.has_svm_vmas) {
+	if (madvise_range.has_svm_userptr_vmas) {
 		err = down_read_interruptible(&vm->svm.gpusvm.notifier_lock);
 		if (err)
-			goto unlock_userptr;
+			goto err_fini;
 	}
 
 	attr_type = array_index_nospec(args->type, ARRAY_SIZE(madvise_funcs));
@@ -426,12 +417,9 @@ int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *fil
 
 	err = xe_vm_invalidate_madvise_range(vm, args->start, args->start + args->range);
 
-	if (madvise_range.has_svm_vmas)
+	if (madvise_range.has_svm_userptr_vmas)
 		xe_svm_notifier_unlock(vm);
 
-unlock_userptr:
-	if (madvise_range.has_userptr_vmas)
-		up_read(&vm->userptr.notifier_lock);
 err_fini:
 	if (madvise_range.has_bo_vmas)
 		drm_exec_fini(&exec);
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index aee050c295db..f6dd964be42f 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -110,10 +110,10 @@ struct xe_vma {
 
 	/**
 	 * @tile_invalidated: Tile mask of binding are invalidated for this VMA.
-	 * protected by BO's resv and for userptrs, vm->userptr.notifier_lock in
-	 * write mode for writing or vm->userptr.notifier_lock in read mode and
+	 * protected by BO's resv and for userptrs, vm->svm.gpusvm.notifier_lock in
+	 * write mode for writing or vm->svm.gpusvm.notifier_lock in read mode and
 	 * the vm->resv. For stable reading, BO's resv or userptr
-	 * vm->userptr.notifier_lock in read mode is required. Can be
+	 * vm->svm.gpusvm.notifier_lock in read mode is required. Can be
 	 * opportunistically read with READ_ONCE outside of locks.
 	 */
 	u8 tile_invalidated;
@@ -124,7 +124,7 @@ struct xe_vma {
 	/**
 	 * @tile_present: Tile mask of binding are present for this VMA.
 	 * protected by vm->lock, vm->resv and for userptrs,
-	 * vm->userptr.notifier_lock for writing. Needs either for reading,
+	 * vm->svm.gpusvm.notifier_lock for writing. Needs either for reading,
 	 * but if reading is done under the vm->lock only, it needs to be held
 	 * in write mode.
 	 */

From 7477c4bd20dc1aba02ef2d955591826da24e2b04 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Thu, 28 Aug 2025 15:24:39 +0100
Subject: [PATCH 0955/1455] drm/xe/pt: unify xe_pt_svm_pre_commit with userptr
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We now use the same notifier lock for SVM and userptr, with that we can
combine xe_pt_userptr_pre_commit and xe_pt_svm_pre_commit.

v2: (Matt B)
  - Re-use xe_svm_notifier_lock/unlock for userptr.
  - Combine svm/userptr handling further down into op_check_svm_userptr.
v3:
  - Only hide the ops if we lack DRM_GPUSVM, since we also need them for
    userptr.

Suggested-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250828142430.615826-18-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_pt.c       | 123 ++++++++++++-------------------
 drivers/gpu/drm/xe/xe_pt_types.h |   2 -
 2 files changed, 47 insertions(+), 78 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 382b25798bc3..07ecf4132d41 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -1377,6 +1377,7 @@ static int xe_pt_pre_commit(struct xe_migrate_pt_update *pt_update)
 				     pt_update_ops, rftree);
 }
 
+#if IS_ENABLED(CONFIG_DRM_GPUSVM)
 #ifdef CONFIG_DRM_XE_USERPTR_INVAL_INJECT
 
 static bool xe_pt_userptr_inject_eagain(struct xe_userptr_vma *uvma)
@@ -1432,8 +1433,8 @@ static int vma_check_userptr(struct xe_vm *vm, struct xe_vma *vma,
 	return 0;
 }
 
-static int op_check_userptr(struct xe_vm *vm, struct xe_vma_op *op,
-			    struct xe_vm_pgtable_update_ops *pt_update)
+static int op_check_svm_userptr(struct xe_vm *vm, struct xe_vma_op *op,
+				struct xe_vm_pgtable_update_ops *pt_update)
 {
 	int err = 0;
 
@@ -1455,9 +1456,40 @@ static int op_check_userptr(struct xe_vm *vm, struct xe_vma_op *op,
 	case DRM_GPUVA_OP_UNMAP:
 		break;
 	case DRM_GPUVA_OP_PREFETCH:
-		err = vma_check_userptr(vm, gpuva_to_vma(op->base.prefetch.va),
-					pt_update);
+		if (xe_vma_is_cpu_addr_mirror(gpuva_to_vma(op->base.prefetch.va))) {
+			struct xe_svm_range *range = op->map_range.range;
+			unsigned long i;
+
+			xe_assert(vm->xe,
+				  xe_vma_is_cpu_addr_mirror(gpuva_to_vma(op->base.prefetch.va)));
+			xa_for_each(&op->prefetch_range.range, i, range) {
+				xe_svm_range_debug(range, "PRE-COMMIT");
+
+				if (!xe_svm_range_pages_valid(range)) {
+					xe_svm_range_debug(range, "PRE-COMMIT - RETRY");
+					return -ENODATA;
+				}
+			}
+		} else {
+			err = vma_check_userptr(vm, gpuva_to_vma(op->base.prefetch.va), pt_update);
+		}
 		break;
+#if IS_ENABLED(CONFIG_DRM_XE_GPUSVM)
+	case DRM_GPUVA_OP_DRIVER:
+		if (op->subop == XE_VMA_SUBOP_MAP_RANGE) {
+			struct xe_svm_range *range = op->map_range.range;
+
+			xe_assert(vm->xe, xe_vma_is_cpu_addr_mirror(op->map_range.vma));
+
+			xe_svm_range_debug(range, "PRE-COMMIT");
+
+			if (!xe_svm_range_pages_valid(range)) {
+				xe_svm_range_debug(range, "PRE-COMMIT - RETRY");
+				return -EAGAIN;
+			}
+		}
+		break;
+#endif
 	default:
 		drm_warn(&vm->xe->drm, "NOT POSSIBLE");
 	}
@@ -1465,7 +1497,7 @@ static int op_check_userptr(struct xe_vm *vm, struct xe_vma_op *op,
 	return err;
 }
 
-static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
+static int xe_pt_svm_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
 {
 	struct xe_vm *vm = pt_update->vops->vm;
 	struct xe_vma_ops *vops = pt_update->vops;
@@ -1478,69 +1510,18 @@ static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
 	if (err)
 		return err;
 
-	down_read(&vm->svm.gpusvm.notifier_lock);
+	xe_svm_notifier_lock(vm);
 
 	list_for_each_entry(op, &vops->list, link) {
-		err = op_check_userptr(vm, op, pt_update_ops);
+		err = op_check_svm_userptr(vm, op, pt_update_ops);
 		if (err) {
-			up_read(&vm->svm.gpusvm.notifier_lock);
+			xe_svm_notifier_unlock(vm);
 			break;
 		}
 	}
 
 	return err;
 }
-
-#if IS_ENABLED(CONFIG_DRM_XE_GPUSVM)
-static int xe_pt_svm_pre_commit(struct xe_migrate_pt_update *pt_update)
-{
-	struct xe_vm *vm = pt_update->vops->vm;
-	struct xe_vma_ops *vops = pt_update->vops;
-	struct xe_vma_op *op;
-	unsigned long i;
-	int err;
-
-	err = xe_pt_pre_commit(pt_update);
-	if (err)
-		return err;
-
-	xe_svm_notifier_lock(vm);
-
-	list_for_each_entry(op, &vops->list, link) {
-		struct xe_svm_range *range = NULL;
-
-		if (op->subop == XE_VMA_SUBOP_UNMAP_RANGE)
-			continue;
-
-		if (op->base.op == DRM_GPUVA_OP_PREFETCH) {
-			xe_assert(vm->xe,
-				  xe_vma_is_cpu_addr_mirror(gpuva_to_vma(op->base.prefetch.va)));
-			xa_for_each(&op->prefetch_range.range, i, range) {
-				xe_svm_range_debug(range, "PRE-COMMIT");
-
-				if (!xe_svm_range_pages_valid(range)) {
-					xe_svm_range_debug(range, "PRE-COMMIT - RETRY");
-					xe_svm_notifier_unlock(vm);
-					return -ENODATA;
-				}
-			}
-		} else {
-			xe_assert(vm->xe, xe_vma_is_cpu_addr_mirror(op->map_range.vma));
-			xe_assert(vm->xe, op->subop == XE_VMA_SUBOP_MAP_RANGE);
-			range = op->map_range.range;
-
-			xe_svm_range_debug(range, "PRE-COMMIT");
-
-			if (!xe_svm_range_pages_valid(range)) {
-				xe_svm_range_debug(range, "PRE-COMMIT - RETRY");
-				xe_svm_notifier_unlock(vm);
-				return -EAGAIN;
-			}
-		}
-	}
-
-	return 0;
-}
 #endif
 
 struct xe_pt_stage_unbind_walk {
@@ -1844,7 +1825,7 @@ static int bind_op_prepare(struct xe_vm *vm, struct xe_tile *tile,
 						 xe_vma_start(vma),
 						 xe_vma_end(vma));
 		++pt_update_ops->current_op;
-		pt_update_ops->needs_userptr_lock |= xe_vma_is_userptr(vma);
+		pt_update_ops->needs_svm_lock |= xe_vma_is_userptr(vma);
 
 		/*
 		 * If rebind, we have to invalidate TLB on !LR vms to invalidate
@@ -1952,7 +1933,7 @@ static int unbind_op_prepare(struct xe_tile *tile,
 	xe_pt_update_ops_rfence_interval(pt_update_ops, xe_vma_start(vma),
 					 xe_vma_end(vma));
 	++pt_update_ops->current_op;
-	pt_update_ops->needs_userptr_lock |= xe_vma_is_userptr(vma);
+	pt_update_ops->needs_svm_lock |= xe_vma_is_userptr(vma);
 	pt_update_ops->needs_invalidation = true;
 
 	xe_pt_commit_prepare_unbind(vma, pt_op->entries, pt_op->num_entries);
@@ -2339,20 +2320,14 @@ static const struct xe_migrate_pt_update_ops migrate_ops = {
 	.pre_commit = xe_pt_pre_commit,
 };
 
-static const struct xe_migrate_pt_update_ops userptr_migrate_ops = {
+#if IS_ENABLED(CONFIG_DRM_GPUSVM)
+static const struct xe_migrate_pt_update_ops svm_userptr_migrate_ops = {
 	.populate = xe_vm_populate_pgtable,
 	.clear = xe_migrate_clear_pgtable_callback,
-	.pre_commit = xe_pt_userptr_pre_commit,
-};
-
-#if IS_ENABLED(CONFIG_DRM_XE_GPUSVM)
-static const struct xe_migrate_pt_update_ops svm_migrate_ops = {
-	.populate = xe_vm_populate_pgtable,
-	.clear = xe_migrate_clear_pgtable_callback,
-	.pre_commit = xe_pt_svm_pre_commit,
+	.pre_commit = xe_pt_svm_userptr_pre_commit,
 };
 #else
-static const struct xe_migrate_pt_update_ops svm_migrate_ops;
+static const struct xe_migrate_pt_update_ops svm_userptr_migrate_ops;
 #endif
 
 static struct xe_dep_scheduler *to_dep_scheduler(struct xe_exec_queue *q,
@@ -2390,9 +2365,7 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 	int err = 0, i;
 	struct xe_migrate_pt_update update = {
 		.ops = pt_update_ops->needs_svm_lock ?
-			&svm_migrate_ops :
-			pt_update_ops->needs_userptr_lock ?
-			&userptr_migrate_ops :
+			&svm_userptr_migrate_ops :
 			&migrate_ops,
 		.vops = vops,
 		.tile_id = tile->id,
@@ -2534,8 +2507,6 @@ xe_pt_update_ops_run(struct xe_tile *tile, struct xe_vma_ops *vops)
 
 	if (pt_update_ops->needs_svm_lock)
 		xe_svm_notifier_unlock(vm);
-	if (pt_update_ops->needs_userptr_lock)
-		up_read(&vm->svm.gpusvm.notifier_lock);
 
 	xe_tlb_inval_job_put(mjob);
 	xe_tlb_inval_job_put(ijob);
diff --git a/drivers/gpu/drm/xe/xe_pt_types.h b/drivers/gpu/drm/xe/xe_pt_types.h
index 17cdd7c7e9f5..881f01e14db8 100644
--- a/drivers/gpu/drm/xe/xe_pt_types.h
+++ b/drivers/gpu/drm/xe/xe_pt_types.h
@@ -105,8 +105,6 @@ struct xe_vm_pgtable_update_ops {
 	u32 current_op;
 	/** @needs_svm_lock: Needs SVM lock */
 	bool needs_svm_lock;
-	/** @needs_userptr_lock: Needs userptr lock */
-	bool needs_userptr_lock;
 	/** @needs_invalidation: Needs invalidation */
 	bool needs_invalidation;
 	/**

From edb1745fc618ba8ef63a45ce3ae60de1bdf29231 Mon Sep 17 00:00:00 2001
From: Matthew Auld <matthew.auld@intel.com>
Date: Fri, 29 Aug 2025 17:47:16 +0100
Subject: [PATCH 0956/1455] drm/xe: improve dma-resv handling for backup object
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Since the dma-resv is shared we don't need to reserve and add a fence
slot fence twice, plus no need to loop through the dependencies.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/20250829164715.720735-2-matthew.auld@intel.com
---
 drivers/gpu/drm/xe/xe_bo.c      | 13 +------------
 drivers/gpu/drm/xe/xe_migrate.c |  2 +-
 2 files changed, 2 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 4faf15d5fa6d..49911247f6cb 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1260,14 +1260,11 @@ int xe_bo_evict_pinned(struct xe_bo *bo)
 		else
 			migrate = mem_type_to_migrate(xe, bo->ttm.resource->mem_type);
 
+		xe_assert(xe, bo->ttm.base.resv == backup->ttm.base.resv);
 		ret = dma_resv_reserve_fences(bo->ttm.base.resv, 1);
 		if (ret)
 			goto out_backup;
 
-		ret = dma_resv_reserve_fences(backup->ttm.base.resv, 1);
-		if (ret)
-			goto out_backup;
-
 		fence = xe_migrate_copy(migrate, bo, backup, bo->ttm.resource,
 					backup->ttm.resource, false);
 		if (IS_ERR(fence)) {
@@ -1277,8 +1274,6 @@ int xe_bo_evict_pinned(struct xe_bo *bo)
 
 		dma_resv_add_fence(bo->ttm.base.resv, fence,
 				   DMA_RESV_USAGE_KERNEL);
-		dma_resv_add_fence(backup->ttm.base.resv, fence,
-				   DMA_RESV_USAGE_KERNEL);
 		dma_fence_put(fence);
 	} else {
 		ret = xe_bo_vmap(backup);
@@ -1356,10 +1351,6 @@ int xe_bo_restore_pinned(struct xe_bo *bo)
 		if (ret)
 			goto out_unlock_bo;
 
-		ret = dma_resv_reserve_fences(backup->ttm.base.resv, 1);
-		if (ret)
-			goto out_unlock_bo;
-
 		fence = xe_migrate_copy(migrate, backup, bo,
 					backup->ttm.resource, bo->ttm.resource,
 					false);
@@ -1370,8 +1361,6 @@ int xe_bo_restore_pinned(struct xe_bo *bo)
 
 		dma_resv_add_fence(bo->ttm.base.resv, fence,
 				   DMA_RESV_USAGE_KERNEL);
-		dma_resv_add_fence(backup->ttm.base.resv, fence,
-				   DMA_RESV_USAGE_KERNEL);
 		dma_fence_put(fence);
 	} else {
 		ret = xe_bo_vmap(backup);
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 9643442ef101..861d9d0633d1 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -908,7 +908,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 		if (!fence) {
 			err = xe_sched_job_add_deps(job, src_bo->ttm.base.resv,
 						    DMA_RESV_USAGE_BOOKKEEP);
-			if (!err && src_bo != dst_bo)
+			if (!err && src_bo->ttm.base.resv != dst_bo->ttm.base.resv)
 				err = xe_sched_job_add_deps(job, dst_bo->ttm.base.resv,
 							    DMA_RESV_USAGE_BOOKKEEP);
 			if (err)

From 981daf1046ef3776ce63fbf11b0261569e858e46 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 19:19:16 +0200
Subject: [PATCH 0957/1455] drm/xe: Allow to stub lookup for graphics and media
 IP

In upcoming patch we will want to replace lookup code during the
test to relax the strict match that we use in production.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250829171922.572-2-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_pci.c | 38 +++++++++++++++++++++----------------
 1 file changed, 22 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index 046d330bad34..dea027e175b6 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -510,6 +510,26 @@ static void read_gmdid(struct xe_device *xe, enum xe_gmdid_type type, u32 *ver,
 	*revid = REG_FIELD_GET(GMD_ID_REVID, val);
 }
 
+static const struct xe_ip *find_graphics_ip(unsigned int verx100)
+{
+	KUNIT_STATIC_STUB_REDIRECT(find_graphics_ip, verx100);
+
+	for (int i = 0; i < ARRAY_SIZE(graphics_ips); i++)
+		if (graphics_ips[i].verx100 == verx100)
+			return &graphics_ips[i];
+	return NULL;
+}
+
+static const struct xe_ip *find_media_ip(unsigned int verx100)
+{
+	KUNIT_STATIC_STUB_REDIRECT(find_media_ip, verx100);
+
+	for (int i = 0; i < ARRAY_SIZE(media_ips); i++)
+		if (media_ips[i].verx100 == verx100)
+			return &media_ips[i];
+	return NULL;
+}
+
 /*
  * Read IP version from hardware and select graphics/media IP descriptors
  * based on the result.
@@ -527,14 +547,7 @@ static void handle_gmdid(struct xe_device *xe,
 
 	read_gmdid(xe, GMDID_GRAPHICS, &ver, graphics_revid);
 
-	for (int i = 0; i < ARRAY_SIZE(graphics_ips); i++) {
-		if (ver == graphics_ips[i].verx100) {
-			*graphics_ip = &graphics_ips[i];
-
-			break;
-		}
-	}
-
+	*graphics_ip = find_graphics_ip(ver);
 	if (!*graphics_ip) {
 		drm_err(&xe->drm, "Hardware reports unknown graphics version %u.%02u\n",
 			ver / 100, ver % 100);
@@ -545,14 +558,7 @@ static void handle_gmdid(struct xe_device *xe,
 	if (ver == 0)
 		return;
 
-	for (int i = 0; i < ARRAY_SIZE(media_ips); i++) {
-		if (ver == media_ips[i].verx100) {
-			*media_ip = &media_ips[i];
-
-			break;
-		}
-	}
-
+	*media_ip = find_media_ip(ver);
 	if (!*media_ip) {
 		drm_err(&xe->drm, "Hardware reports unknown media version %u.%02u\n",
 			ver / 100, ver % 100);

From 42367babd859fb75c3737c520f6541d542f19ba5 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 19:19:17 +0200
Subject: [PATCH 0958/1455] drm/xe/kunit: Update struct xe_pci_fake_data step
 declarations

The struct xe_pci_fake_data has fields that specify graphics and
media stepping of the fake PCI device used during KUnit testing.

Change definitions of those separate step fields and use existing
struct xe_step_info definition that already have required fields.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250829171922.572-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_pci.c      | 4 ++--
 drivers/gpu/drm/xe/tests/xe_pci_test.h | 4 ++--
 drivers/gpu/drm/xe/tests/xe_wa_test.c  | 3 +--
 3 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c
index db30c5156d0c..bb0ce0dc9b31 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci.c
+++ b/drivers/gpu/drm/xe/tests/xe_pci.c
@@ -94,10 +94,10 @@ static void fake_read_gmdid(struct xe_device *xe, enum xe_gmdid_type type,
 
 	if (type == GMDID_MEDIA) {
 		*ver = data->media_verx100;
-		*revid = xe_step_to_gmdid(data->media_step);
+		*revid = xe_step_to_gmdid(data->step.media);
 	} else {
 		*ver = data->graphics_verx100;
-		*revid = xe_step_to_gmdid(data->graphics_step);
+		*revid = xe_step_to_gmdid(data->step.graphics);
 	}
 }
 
diff --git a/drivers/gpu/drm/xe/tests/xe_pci_test.h b/drivers/gpu/drm/xe/tests/xe_pci_test.h
index ce4d2b86b778..32452bd65bff 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci_test.h
+++ b/drivers/gpu/drm/xe/tests/xe_pci_test.h
@@ -10,6 +10,7 @@
 
 #include "xe_platform_types.h"
 #include "xe_sriov_types.h"
+#include "xe_step_types.h"
 
 struct xe_device;
 
@@ -17,10 +18,9 @@ struct xe_pci_fake_data {
 	enum xe_sriov_mode sriov_mode;
 	enum xe_platform platform;
 	enum xe_subplatform subplatform;
+	struct xe_step_info step;
 	u32 graphics_verx100;
 	u32 media_verx100;
-	u32 graphics_step;
-	u32 media_step;
 };
 
 int xe_pci_fake_device_init(struct xe_device *xe);
diff --git a/drivers/gpu/drm/xe/tests/xe_wa_test.c b/drivers/gpu/drm/xe/tests/xe_wa_test.c
index 416258c193f6..a0de55f09267 100644
--- a/drivers/gpu/drm/xe/tests/xe_wa_test.c
+++ b/drivers/gpu/drm/xe/tests/xe_wa_test.c
@@ -93,8 +93,7 @@ static int xe_wa_test_init(struct kunit *test)
 		.subplatform = param->subplatform,
 		.graphics_verx100 = param->graphics_verx100,
 		.media_verx100 = param->media_verx100,
-		.graphics_step = param->step.graphics,
-		.media_step = param->step.media,
+		.step = param->step,
 	};
 	struct xe_device *xe;
 	struct device *dev;

From b1ee6558433be01cf15c180aec54cb79ccaa64e8 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 19:19:18 +0200
Subject: [PATCH 0959/1455] drm/xe/kunit: Introduce xe_pci_fake_data_desc()

We already use struct xe_pci_fake_data to provide custom config of
the fake PCI device and soon we will be using this struct also as
direct parameter for the parameterized Xe KUnit tests.

Add function to generate description based on that config data.

For platform or subplatform name lookup pciidlist which already
have definitions of all supported platforms.

Examples:

  TIGERLAKE
  TIGERLAKE A0
  TIGERLAKE SR-IOV PF
  ...
  PANTHERLAKE 30.00(Xe3_LPG) 30.00(Xe3_LPM)
  PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
  PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0 SR-IOV VF

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250829171922.572-4-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_pci.c      | 117 +++++++++++++++++++++++++
 drivers/gpu/drm/xe/tests/xe_pci_test.h |   1 +
 2 files changed, 118 insertions(+)

diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c
index bb0ce0dc9b31..a8b35439131f 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci.c
+++ b/drivers/gpu/drm/xe/tests/xe_pci.c
@@ -12,6 +12,123 @@
 #include <kunit/test-bug.h>
 #include <kunit/visibility.h>
 
+static const struct xe_device_desc *lookup_desc(enum xe_platform p)
+{
+	const struct xe_device_desc *desc;
+	const struct pci_device_id *ids;
+
+	for (ids = pciidlist; ids->driver_data; ids++) {
+		desc = (const void *)ids->driver_data;
+		if (desc->platform == p)
+			return desc;
+	}
+	return NULL;
+}
+
+static const struct xe_subplatform_desc *lookup_sub_desc(enum xe_platform p, enum xe_subplatform s)
+{
+	const struct xe_device_desc *desc = lookup_desc(p);
+	const struct xe_subplatform_desc *spd;
+
+	if (desc && desc->subplatforms)
+		for (spd = desc->subplatforms; spd->subplatform; spd++)
+			if (spd->subplatform == s)
+				return spd;
+	return NULL;
+}
+
+static const char *lookup_platform_name(enum xe_platform p)
+{
+	const struct xe_device_desc *desc = lookup_desc(p);
+
+	return desc ? desc->platform_name : "INVALID";
+}
+
+static const char *__lookup_subplatform_name(enum xe_platform p, enum xe_subplatform s)
+{
+	const struct xe_subplatform_desc *desc = lookup_sub_desc(p, s);
+
+	return desc ? desc->name : "INVALID";
+}
+
+static const char *lookup_subplatform_name(enum xe_platform p, enum xe_subplatform s)
+{
+	return s == XE_SUBPLATFORM_NONE ? "" : __lookup_subplatform_name(p, s);
+}
+
+static const char *subplatform_prefix(enum xe_subplatform s)
+{
+	return s == XE_SUBPLATFORM_NONE ? "" : " ";
+}
+
+static const char *step_prefix(enum xe_step step)
+{
+	return step == STEP_NONE ? "" : " ";
+}
+
+static const char *step_name(enum xe_step step)
+{
+	return step == STEP_NONE ? "" : xe_step_name(step);
+}
+
+static const char *sriov_prefix(enum xe_sriov_mode mode)
+{
+	return mode <= XE_SRIOV_MODE_NONE ? "" : " ";
+}
+
+static const char *sriov_name(enum xe_sriov_mode mode)
+{
+	return mode <= XE_SRIOV_MODE_NONE ? "" : xe_sriov_mode_to_string(mode);
+}
+
+static const char *lookup_graphics_name(unsigned int verx100)
+{
+	const struct xe_ip *ip = find_graphics_ip(verx100);
+
+	return ip ? ip->name : "";
+}
+
+static const char *lookup_media_name(unsigned int verx100)
+{
+	const struct xe_ip *ip = find_media_ip(verx100);
+
+	return ip ? ip->name : "";
+}
+
+/**
+ * xe_pci_fake_data_desc - Describe struct xe_pci_fake_data parameter
+ * @param: the &struct xe_pci_fake_data parameter to describe
+ * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE
+ *
+ * This function prepares description of the struct xe_pci_fake_data parameter.
+ *
+ * It is tailored for use in parameterized KUnit tests where parameter generator
+ * is based on the struct xe_pci_fake_data arrays.
+ */
+void xe_pci_fake_data_desc(const struct xe_pci_fake_data *param, char *desc)
+{
+	if (param->graphics_verx100 || param->media_verx100)
+		snprintf(desc, KUNIT_PARAM_DESC_SIZE, "%s%s%s %u.%02u(%s)%s%s %u.%02u(%s)%s%s%s%s",
+			 lookup_platform_name(param->platform),
+			 subplatform_prefix(param->subplatform),
+			 lookup_subplatform_name(param->platform, param->subplatform),
+			 param->graphics_verx100 / 100, param->graphics_verx100 % 100,
+			 lookup_graphics_name(param->graphics_verx100),
+			 step_prefix(param->step.graphics), step_name(param->step.graphics),
+			 param->media_verx100 / 100, param->media_verx100 % 100,
+			 lookup_media_name(param->media_verx100),
+			 step_prefix(param->step.media), step_name(param->step.media),
+			 sriov_prefix(param->sriov_mode), sriov_name(param->sriov_mode));
+	else
+		snprintf(desc, KUNIT_PARAM_DESC_SIZE, "%s%s%s%s%s%s%s",
+			 lookup_platform_name(param->platform),
+			 subplatform_prefix(param->subplatform),
+			 lookup_subplatform_name(param->platform, param->subplatform),
+			 step_prefix(param->step.graphics), step_name(param->step.graphics),
+			 sriov_prefix(param->sriov_mode), sriov_name(param->sriov_mode));
+}
+EXPORT_SYMBOL_IF_KUNIT(xe_pci_fake_data_desc);
+
 static void xe_ip_kunit_desc(const struct xe_ip *param, char *desc)
 {
 	snprintf(desc, KUNIT_PARAM_DESC_SIZE, "%u.%02u %s",
diff --git a/drivers/gpu/drm/xe/tests/xe_pci_test.h b/drivers/gpu/drm/xe/tests/xe_pci_test.h
index 32452bd65bff..1195839999ab 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci_test.h
+++ b/drivers/gpu/drm/xe/tests/xe_pci_test.h
@@ -24,6 +24,7 @@ struct xe_pci_fake_data {
 };
 
 int xe_pci_fake_device_init(struct xe_device *xe);
+void xe_pci_fake_data_desc(const struct xe_pci_fake_data *param, char *desc);
 
 const void *xe_pci_graphics_ip_gen_param(const void *prev, char *desc);
 const void *xe_pci_media_ip_gen_param(const void *prev, char *desc);

From ddbe5aecea846fd8f7510540e720919d5d781bd9 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 19:19:19 +0200
Subject: [PATCH 0960/1455] drm/xe/kunit: Drop custom struct platform_test_case

Custom struct platform_test_case definition in xe_wa is now almost
identical to generic struct xe_pci_fake_data defintiion except the
.name member, which could be generated by xe_pci_fake_data_desc().

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250829171922.572-5-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_wa_test.c | 31 ++++-----------------------
 1 file changed, 4 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_wa_test.c b/drivers/gpu/drm/xe/tests/xe_wa_test.c
index a0de55f09267..7bf34fd0d189 100644
--- a/drivers/gpu/drm/xe/tests/xe_wa_test.c
+++ b/drivers/gpu/drm/xe/tests/xe_wa_test.c
@@ -15,18 +15,8 @@
 #include "xe_tuning.h"
 #include "xe_wa.h"
 
-struct platform_test_case {
-	const char *name;
-	enum xe_platform platform;
-	enum xe_subplatform subplatform;
-	u32 graphics_verx100;
-	u32 media_verx100;
-	struct xe_step_info step;
-};
-
 #define PLATFORM_CASE(platform__, graphics_step__)				\
 	{									\
-		.name = #platform__ " (" #graphics_step__ ")",			\
 		.platform = XE_ ## platform__,					\
 		.subplatform = XE_SUBPLATFORM_NONE,				\
 		.step = { .graphics = STEP_ ## graphics_step__ }		\
@@ -35,7 +25,6 @@ struct platform_test_case {
 
 #define SUBPLATFORM_CASE(platform__, subplatform__, graphics_step__)			\
 	{										\
-		.name = #platform__ "_" #subplatform__ " (" #graphics_step__ ")",	\
 		.platform = XE_ ## platform__,						\
 		.subplatform = XE_SUBPLATFORM_ ## platform__ ## _ ## subplatform__,	\
 		.step = { .graphics = STEP_ ## graphics_step__ }			\
@@ -44,7 +33,6 @@ struct platform_test_case {
 #define GMDID_CASE(platform__, graphics_verx100__, graphics_step__,		\
 		   media_verx100__, media_step__)				\
 	{									\
-		.name = #platform__ " (g:" #graphics_step__ ", m:" #media_step__ ")",\
 		.platform = XE_ ## platform__,					\
 		.subplatform = XE_SUBPLATFORM_NONE,				\
 		.graphics_verx100 = graphics_verx100__,				\
@@ -53,7 +41,7 @@ struct platform_test_case {
 			   .media = STEP_ ## media_step__ }			\
 	}
 
-static const struct platform_test_case cases[] = {
+static const struct xe_pci_fake_data cases[] = {
 	PLATFORM_CASE(TIGERLAKE, B0),
 	PLATFORM_CASE(DG1, A0),
 	PLATFORM_CASE(DG1, B0),
@@ -78,23 +66,12 @@ static const struct platform_test_case cases[] = {
 	GMDID_CASE(PANTHERLAKE, 3000, A0, 3000, A0),
 };
 
-static void platform_desc(const struct platform_test_case *t, char *desc)
-{
-	strscpy(desc, t->name, KUNIT_PARAM_DESC_SIZE);
-}
-
-KUNIT_ARRAY_PARAM(platform, cases, platform_desc);
+KUNIT_ARRAY_PARAM(platform, cases, xe_pci_fake_data_desc);
 
 static int xe_wa_test_init(struct kunit *test)
 {
-	const struct platform_test_case *param = test->param_value;
-	struct xe_pci_fake_data data = {
-		.platform = param->platform,
-		.subplatform = param->subplatform,
-		.graphics_verx100 = param->graphics_verx100,
-		.media_verx100 = param->media_verx100,
-		.step = param->step,
-	};
+	const struct xe_pci_fake_data *param = test->param_value;
+	struct xe_pci_fake_data data = *param;
 	struct xe_device *xe;
 	struct device *dev;
 	int ret;

From a9c8517058cc425dc983ad0f22e6dc3b810ff773 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 19:19:20 +0200
Subject: [PATCH 0961/1455] drm/xe/kunit: Promote fake platform parameter list

The list of all known representative platforms defined in xe_wa
could be used in more places by other test suites.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250829171922.572-6-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_pci.c      | 69 ++++++++++++++++++++++++++
 drivers/gpu/drm/xe/tests/xe_pci_test.h |  1 +
 drivers/gpu/drm/xe/tests/xe_wa_test.c  | 55 +-------------------
 3 files changed, 71 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c
index a8b35439131f..f19b1a64128b 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci.c
+++ b/drivers/gpu/drm/xe/tests/xe_pci.c
@@ -12,6 +12,75 @@
 #include <kunit/test-bug.h>
 #include <kunit/visibility.h>
 
+#define PLATFORM_CASE(platform__, graphics_step__)					\
+	{										\
+		.platform = XE_ ## platform__,						\
+		.subplatform = XE_SUBPLATFORM_NONE,					\
+		.step = { .graphics = STEP_ ## graphics_step__ }			\
+	}
+
+#define SUBPLATFORM_CASE(platform__, subplatform__, graphics_step__)			\
+	{										\
+		.platform = XE_ ## platform__,						\
+		.subplatform = XE_SUBPLATFORM_ ## platform__ ## _ ## subplatform__,	\
+		.step = { .graphics = STEP_ ## graphics_step__ }			\
+	}
+
+#define GMDID_CASE(platform__, graphics_verx100__, graphics_step__,			\
+		   media_verx100__, media_step__)					\
+	{										\
+		.platform = XE_ ## platform__,						\
+		.subplatform = XE_SUBPLATFORM_NONE,					\
+		.graphics_verx100 = graphics_verx100__,					\
+		.media_verx100 = media_verx100__,					\
+		.step = { .graphics = STEP_ ## graphics_step__,				\
+			   .media = STEP_ ## media_step__ }				\
+	}
+
+static const struct xe_pci_fake_data cases[] = {
+	PLATFORM_CASE(TIGERLAKE, B0),
+	PLATFORM_CASE(DG1, A0),
+	PLATFORM_CASE(DG1, B0),
+	PLATFORM_CASE(ALDERLAKE_S, A0),
+	PLATFORM_CASE(ALDERLAKE_S, B0),
+	PLATFORM_CASE(ALDERLAKE_S, C0),
+	PLATFORM_CASE(ALDERLAKE_S, D0),
+	PLATFORM_CASE(ALDERLAKE_P, A0),
+	PLATFORM_CASE(ALDERLAKE_P, B0),
+	PLATFORM_CASE(ALDERLAKE_P, C0),
+	SUBPLATFORM_CASE(ALDERLAKE_S, RPLS, D0),
+	SUBPLATFORM_CASE(ALDERLAKE_P, RPLU, E0),
+	SUBPLATFORM_CASE(DG2, G10, C0),
+	SUBPLATFORM_CASE(DG2, G11, B1),
+	SUBPLATFORM_CASE(DG2, G12, A1),
+	GMDID_CASE(METEORLAKE, 1270, A0, 1300, A0),
+	GMDID_CASE(METEORLAKE, 1271, A0, 1300, A0),
+	GMDID_CASE(METEORLAKE, 1274, A0, 1300, A0),
+	GMDID_CASE(LUNARLAKE, 2004, A0, 2000, A0),
+	GMDID_CASE(LUNARLAKE, 2004, B0, 2000, A0),
+	GMDID_CASE(BATTLEMAGE, 2001, A0, 1301, A1),
+	GMDID_CASE(PANTHERLAKE, 3000, A0, 3000, A0),
+};
+
+KUNIT_ARRAY_PARAM(platform, cases, xe_pci_fake_data_desc);
+
+/**
+ * xe_pci_fake_data_gen_params - Generate struct xe_pci_fake_data parameters
+ * @prev: the pointer to the previous parameter to iterate from or NULL
+ * @desc: output buffer with minimum size of KUNIT_PARAM_DESC_SIZE
+ *
+ * This function prepares struct xe_pci_fake_data parameter.
+ *
+ * To be used only as a parameter generator function in &KUNIT_CASE_PARAM.
+ *
+ * Return: pointer to the next parameter or NULL if no more parameters
+ */
+const void *xe_pci_fake_data_gen_params(const void *prev, char *desc)
+{
+	return platform_gen_params(prev, desc);
+}
+EXPORT_SYMBOL_IF_KUNIT(xe_pci_fake_data_gen_params);
+
 static const struct xe_device_desc *lookup_desc(enum xe_platform p)
 {
 	const struct xe_device_desc *desc;
diff --git a/drivers/gpu/drm/xe/tests/xe_pci_test.h b/drivers/gpu/drm/xe/tests/xe_pci_test.h
index 1195839999ab..5e9a7ffc747f 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci_test.h
+++ b/drivers/gpu/drm/xe/tests/xe_pci_test.h
@@ -24,6 +24,7 @@ struct xe_pci_fake_data {
 };
 
 int xe_pci_fake_device_init(struct xe_device *xe);
+const void *xe_pci_fake_data_gen_params(const void *prev, char *desc);
 void xe_pci_fake_data_desc(const struct xe_pci_fake_data *param, char *desc);
 
 const void *xe_pci_graphics_ip_gen_param(const void *prev, char *desc);
diff --git a/drivers/gpu/drm/xe/tests/xe_wa_test.c b/drivers/gpu/drm/xe/tests/xe_wa_test.c
index 7bf34fd0d189..d67bfa3ec494 100644
--- a/drivers/gpu/drm/xe/tests/xe_wa_test.c
+++ b/drivers/gpu/drm/xe/tests/xe_wa_test.c
@@ -15,59 +15,6 @@
 #include "xe_tuning.h"
 #include "xe_wa.h"
 
-#define PLATFORM_CASE(platform__, graphics_step__)				\
-	{									\
-		.platform = XE_ ## platform__,					\
-		.subplatform = XE_SUBPLATFORM_NONE,				\
-		.step = { .graphics = STEP_ ## graphics_step__ }		\
-	}
-
-
-#define SUBPLATFORM_CASE(platform__, subplatform__, graphics_step__)			\
-	{										\
-		.platform = XE_ ## platform__,						\
-		.subplatform = XE_SUBPLATFORM_ ## platform__ ## _ ## subplatform__,	\
-		.step = { .graphics = STEP_ ## graphics_step__ }			\
-	}
-
-#define GMDID_CASE(platform__, graphics_verx100__, graphics_step__,		\
-		   media_verx100__, media_step__)				\
-	{									\
-		.platform = XE_ ## platform__,					\
-		.subplatform = XE_SUBPLATFORM_NONE,				\
-		.graphics_verx100 = graphics_verx100__,				\
-		.media_verx100 = media_verx100__,				\
-		.step = { .graphics = STEP_ ## graphics_step__,			\
-			   .media = STEP_ ## media_step__ }			\
-	}
-
-static const struct xe_pci_fake_data cases[] = {
-	PLATFORM_CASE(TIGERLAKE, B0),
-	PLATFORM_CASE(DG1, A0),
-	PLATFORM_CASE(DG1, B0),
-	PLATFORM_CASE(ALDERLAKE_S, A0),
-	PLATFORM_CASE(ALDERLAKE_S, B0),
-	PLATFORM_CASE(ALDERLAKE_S, C0),
-	PLATFORM_CASE(ALDERLAKE_S, D0),
-	PLATFORM_CASE(ALDERLAKE_P, A0),
-	PLATFORM_CASE(ALDERLAKE_P, B0),
-	PLATFORM_CASE(ALDERLAKE_P, C0),
-	SUBPLATFORM_CASE(ALDERLAKE_S, RPLS, D0),
-	SUBPLATFORM_CASE(ALDERLAKE_P, RPLU, E0),
-	SUBPLATFORM_CASE(DG2, G10, C0),
-	SUBPLATFORM_CASE(DG2, G11, B1),
-	SUBPLATFORM_CASE(DG2, G12, A1),
-	GMDID_CASE(METEORLAKE, 1270, A0, 1300, A0),
-	GMDID_CASE(METEORLAKE, 1271, A0, 1300, A0),
-	GMDID_CASE(METEORLAKE, 1274, A0, 1300, A0),
-	GMDID_CASE(LUNARLAKE, 2004, A0, 2000, A0),
-	GMDID_CASE(LUNARLAKE, 2004, B0, 2000, A0),
-	GMDID_CASE(BATTLEMAGE, 2001, A0, 1301, A1),
-	GMDID_CASE(PANTHERLAKE, 3000, A0, 3000, A0),
-};
-
-KUNIT_ARRAY_PARAM(platform, cases, xe_pci_fake_data_desc);
-
 static int xe_wa_test_init(struct kunit *test)
 {
 	const struct xe_pci_fake_data *param = test->param_value;
@@ -120,7 +67,7 @@ static void xe_wa_gt(struct kunit *test)
 }
 
 static struct kunit_case xe_wa_tests[] = {
-	KUNIT_CASE_PARAM(xe_wa_gt, platform_gen_params),
+	KUNIT_CASE_PARAM(xe_wa_gt, xe_pci_fake_data_gen_params),
 	{}
 };
 

From dcc38bc5e13701a2401f0873a38e447fefe1a1af Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 29 Aug 2025 19:19:21 +0200
Subject: [PATCH 0962/1455] drm/xe/kunit: Drop xe_wa_test_exit

Remove xe_wa_test_exit() as it could crach the KUnit kernel in
case of hitting some asserts in xe_wa_test_init() as test->priv
could not be pointing to expected data.

 |    # xe_wa_gt: ASSERTION FAILED at drivers/gpu/drm/xe/tests/xe_wa_test.c:34
 |    Expected ret == 0, but
 |        ret == -19 (0xffffffffffffffed)
 |Bus error - the host /dev/shm or /tmp mount likely just ran out of space
 |Kernel panic - not syncing: Kernel mode signal 7

Note that there is no need to call drm_kunit_helper_free_device()
since our fake device allocated by drm_kunit_helper_alloc_device()
will be cleaned up automatically.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250829171922.572-7-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_wa_test.c | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_wa_test.c b/drivers/gpu/drm/xe/tests/xe_wa_test.c
index d67bfa3ec494..49d191043dfa 100644
--- a/drivers/gpu/drm/xe/tests/xe_wa_test.c
+++ b/drivers/gpu/drm/xe/tests/xe_wa_test.c
@@ -43,13 +43,6 @@ static int xe_wa_test_init(struct kunit *test)
 	return 0;
 }
 
-static void xe_wa_test_exit(struct kunit *test)
-{
-	struct xe_device *xe = test->priv;
-
-	drm_kunit_helper_free_device(test, xe->drm.dev);
-}
-
 static void xe_wa_gt(struct kunit *test)
 {
 	struct xe_device *xe = test->priv;
@@ -74,7 +67,6 @@ static struct kunit_case xe_wa_tests[] = {
 static struct kunit_suite xe_rtp_test_suite = {
 	.name = "xe_wa",
 	.init = xe_wa_test_init,
-	.exit = xe_wa_test_exit,
 	.test_cases = xe_wa_tests,
 };
 

From 7a9b19cca035d9514ec1bb3d7fb65e56bbd18083 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 19:27:38 +0800
Subject: [PATCH 0963/1455] drm/sti: Remove redundant ternary operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Acked-by: Raphaël Gallais-Pou <rgallaispou@gmail.com>
Link: https://lore.kernel.org/r/20250904112738.350652-1-liaoyuanhong@vivo.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 drivers/gpu/drm/sti/sti_hqvdp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/sti/sti_hqvdp.c b/drivers/gpu/drm/sti/sti_hqvdp.c
index 03684062309b..b76606e9a82d 100644
--- a/drivers/gpu/drm/sti/sti_hqvdp.c
+++ b/drivers/gpu/drm/sti/sti_hqvdp.c
@@ -744,7 +744,7 @@ static bool sti_hqvdp_check_hw_scaling(struct sti_hqvdp *hqvdp,
 
 	inv_zy = DIV_ROUND_UP(src_h, dst_h);
 
-	return (inv_zy <= lfw) ? true : false;
+	return inv_zy <= lfw;
 }
 
 /**

From 2ce575d6a1be4ef03c804db0f668527af426472c Mon Sep 17 00:00:00 2001
From: Imre Deak <imre.deak@intel.com>
Date: Wed, 3 Sep 2025 15:21:52 +0300
Subject: [PATCH 0964/1455] drm/i915/display: Remove power state verification
 before HW readout

During system resume the display power state verification will print the

"power well x state mismatch (refcount 0/enabled 1)"

error message from the early resume sequence for a power well left
enabled by BIOS. This power well was probably left enabled by BIOS
inadvertently, since BIOS versions on current platforms do not leave any
display output enabled while resuming from an Sx power state, hence the
enabled display power well is unused. In theory however it is possible
that BIOS leaves a display output enabled, in that case the enabled
power well shouldn't be reported as an error.

According to the above, remove the display power state verification from
the early resume phase to avoid incorrectly reporting an enabled power
well without a power reference as an error.

Note: The refcount for any enabled and used power well (i.e. used for an
enabled display output) will be acquired following the early resume
sequence, after the HW state for display outputs (encoder/crtc etc.) is
read out. Any power well enabled but not used (hence not holding a
reference) will be disabled after the HW state readout. The display
power state will be verified afterwards in intel_power_domains_enable().

Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6012
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Imre Deak <imre.deak@intel.com>
Link: https://lore.kernel.org/r/20250903122152.2526050-1-imre.deak@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_power.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 7340d5a71673..7f9a2d743bac 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -2165,8 +2165,6 @@ void intel_power_domains_resume(struct intel_display *display)
 		power_domains->init_wakeref =
 			intel_display_power_get(display, POWER_DOMAIN_INIT);
 	}
-
-	intel_power_domains_verify_state(display);
 }
 
 #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_RUNTIME_PM)

From 2d1e962098e294330a07ab7e91ee0f35121a6964 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 4 Sep 2025 16:40:26 +0200
Subject: [PATCH 0965/1455] drm/xe: Fix broken kernel-doc for the struct xe_bo

Use correct multi-line kernel-doc style if required.
Some members were described only in the commit message.
Some other members were described using wrong names.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250904144026.7222-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_bo_types.h | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo_types.h b/drivers/gpu/drm/xe/xe_bo_types.h
index 314652afdca7..d4fe3c8dca5b 100644
--- a/drivers/gpu/drm/xe/xe_bo_types.h
+++ b/drivers/gpu/drm/xe/xe_bo_types.h
@@ -25,7 +25,9 @@ struct xe_vm;
 /* TODO: To be selected with VM_MADVISE */
 #define	XE_BO_PRIORITY_NORMAL	1
 
-/** @xe_bo: XE buffer object */
+/**
+ * struct xe_bo - Xe buffer object
+ */
 struct xe_bo {
 	/** @ttm: TTM base buffer object */
 	struct ttm_buffer_object ttm;
@@ -47,7 +49,7 @@ struct xe_bo {
 	struct xe_ggtt_node *ggtt_node[XE_MAX_TILES_PER_DEVICE];
 	/** @vmap: iosys map of this buffer */
 	struct iosys_map vmap;
-	/** @ttm_kmap: TTM bo kmap object for internal use only. Keep off. */
+	/** @kmap: TTM bo kmap object for internal use only. Keep off. */
 	struct ttm_bo_kmap_obj kmap;
 	/** @pinned_link: link to present / evicted list of pinned BO */
 	struct list_head pinned_link;
@@ -82,10 +84,10 @@ struct xe_bo {
 	/** @created: Whether the bo has passed initial creation */
 	bool created;
 
-	/** @ccs_cleared */
+	/** @ccs_cleared: true means that CCS region of BO is already cleared */
 	bool ccs_cleared;
 
-	/** @bb_ccs_rw: BB instructions of CCS read/write. Valid only for VF */
+	/** @bb_ccs: BB instructions of CCS read/write. Valid only for VF */
 	struct xe_bb *bb_ccs[XE_SRIOV_VF_CCS_CTX_COUNT];
 
 	/**
@@ -99,9 +101,10 @@ struct xe_bo {
 	struct drm_pagemap_devmem devmem_allocation;
 
 	/** @vram_userfault_link: Link into @mem_access.vram_userfault.list */
-		struct list_head vram_userfault_link;
+	struct list_head vram_userfault_link;
 
-	/** @min_align: minimum alignment needed for this BO if different
+	/**
+	 * @min_align: minimum alignment needed for this BO if different
 	 * from default
 	 */
 	u64 min_align;

From 8d87f08ba08646d21b9d6913fe63c463c9025250 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 18:39:15 +0300
Subject: [PATCH 0966/1455] drm/i915/hpd: Fix mtp_tc_hpd_enable_detection()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Set the MTP TC hotplug bits in the correct register.

Shouldn't matter really as this only gets used for eDP
detection and there should be never eDP on TC ports on
current hw.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902153915.4423-1-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_hotplug_irq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
index 43aee70597bf..4f72f3fb9af5 100644
--- a/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
+++ b/drivers/gpu/drm/i915/display/intel_hotplug_irq.c
@@ -1025,7 +1025,7 @@ static void mtp_tc_hpd_enable_detection(struct intel_encoder *encoder)
 {
 	struct intel_display *display = to_intel_display(encoder);
 
-	intel_de_rmw(display, SHOTPLUG_CTL_DDI,
+	intel_de_rmw(display, SHOTPLUG_CTL_TC,
 		     mtp_tc_hotplug_mask(encoder->hpd_pin),
 		     mtp_tc_hotplug_enables(encoder));
 }

From 3a64c199634457f0e159690510321e4778fb45a0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:06 +0300
Subject: [PATCH 0967/1455] drm/i915/dram: Populate PNV memory type accurately
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If PNV doesn't have DDR3 then it has DDR2. Add the appropriate memory
type for it.

No functional change since we currently only care about the
DDR3 vs. not difference.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-2-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 10 ++++++----
 drivers/gpu/drm/i915/soc/intel_dram.h |  1 +
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 3eb748ab44d9..7d8b8f81e215 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -35,6 +35,7 @@ static const char *intel_dram_type_str(enum intel_dram_type type)
 {
 	static const char * const str[] = {
 		DRAM_TYPE_STR(UNKNOWN),
+		DRAM_TYPE_STR(DDR2),
 		DRAM_TYPE_STR(DDR3),
 		DRAM_TYPE_STR(DDR4),
 		DRAM_TYPE_STR(LPDDR3),
@@ -55,9 +56,10 @@ static const char *intel_dram_type_str(enum intel_dram_type type)
 
 #undef DRAM_TYPE_STR
 
-static bool pnv_is_ddr3(struct drm_i915_private *i915)
+static enum intel_dram_type pnv_dram_type(struct drm_i915_private *i915)
 {
-	return intel_uncore_read(&i915->uncore, CSHRDDR3CTL) & CSHRDDR3CTL_DDR3;
+	return intel_uncore_read(&i915->uncore, CSHRDDR3CTL) & CSHRDDR3CTL_DDR3 ?
+		INTEL_DRAM_DDR3 : INTEL_DRAM_DDR2;
 }
 
 static unsigned int pnv_mem_freq(struct drm_i915_private *dev_priv)
@@ -252,8 +254,8 @@ static int i915_get_dram_info(struct drm_i915_private *i915, struct dram_info *d
 	if (dram_info->mem_freq)
 		drm_dbg(&i915->drm, "DDR speed: %d kHz\n", dram_info->mem_freq);
 
-	if (IS_PINEVIEW(i915) && pnv_is_ddr3(i915))
-		dram_info->type = INTEL_DRAM_DDR3;
+	if (IS_PINEVIEW(i915))
+		dram_info->type = pnv_dram_type(i915);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 97d21894abdc..7e3b3d961fcb 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -17,6 +17,7 @@ struct dram_info {
 	bool symmetric_memory;
 	enum intel_dram_type {
 		INTEL_DRAM_UNKNOWN,
+		INTEL_DRAM_DDR2,
 		INTEL_DRAM_DDR3,
 		INTEL_DRAM_DDR4,
 		INTEL_DRAM_LPDDR3,

From 413439008ca409da8d46957cf598a125aa32fdd4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:07 +0300
Subject: [PATCH 0968/1455] drm/i915/dram: Use intel_dram_type_str() for pnv
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Replace the hand rolled PNV memory type printk string
stuff with intel_dram_type_str().

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-3-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/i9xx_wm.c | 5 +++--
 drivers/gpu/drm/i915/soc/intel_dram.c  | 2 +-
 drivers/gpu/drm/i915/soc/intel_dram.h  | 1 +
 3 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_wm.c b/drivers/gpu/drm/i915/display/i9xx_wm.c
index 060aff765994..fd3b7b35f351 100644
--- a/drivers/gpu/drm/i915/display/i9xx_wm.c
+++ b/drivers/gpu/drm/i915/display/i9xx_wm.c
@@ -105,8 +105,9 @@ static const struct cxsr_latency *pnv_get_cxsr_latency(struct intel_display *dis
 	}
 
 	drm_dbg_kms(display->drm,
-		    "Could not find CxSR latency for DDR%s, FSB %u kHz, MEM %u kHz\n",
-		    is_ddr3 ? "3" : "2", dram_info->fsb_freq, dram_info->mem_freq);
+		    "Could not find CxSR latency for %s, FSB %u kHz, MEM %u kHz\n",
+		    intel_dram_type_str(dram_info->type),
+		    dram_info->fsb_freq, dram_info->mem_freq);
 
 	return NULL;
 }
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 7d8b8f81e215..b4f0793f778d 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -31,7 +31,7 @@ struct dram_channel_info {
 
 #define DRAM_TYPE_STR(type) [INTEL_DRAM_ ## type] = #type
 
-static const char *intel_dram_type_str(enum intel_dram_type type)
+const char *intel_dram_type_str(enum intel_dram_type type)
 {
 	static const char * const str[] = {
 		DRAM_TYPE_STR(UNKNOWN),
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 7e3b3d961fcb..846a77b1aa90 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -39,5 +39,6 @@ int intel_dram_detect(struct drm_i915_private *i915);
 unsigned int intel_fsb_freq(struct drm_i915_private *i915);
 unsigned int intel_mem_freq(struct drm_i915_private *i915);
 const struct dram_info *intel_dram_info(struct drm_device *drm);
+const char *intel_dram_type_str(enum intel_dram_type type);
 
 #endif /* __INTEL_DRAM_H__ */

From da548f7e781d3abc6ecbc08cc8b18ffaa3211653 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:08 +0300
Subject: [PATCH 0969/1455] drm/i915/dram: Pack dram_info better
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

struct dram_info has two holes in the middle. Shuffle things
around to plug them.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-4-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 846a77b1aa90..6212944d44aa 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -12,9 +12,6 @@ struct drm_i915_private;
 struct drm_device;
 
 struct dram_info {
-	bool wm_lv_0_adjust_needed;
-	u8 num_channels;
-	bool symmetric_memory;
 	enum intel_dram_type {
 		INTEL_DRAM_UNKNOWN,
 		INTEL_DRAM_DDR2,
@@ -28,10 +25,13 @@ struct dram_info {
 		INTEL_DRAM_GDDR_ECC,
 		__INTEL_DRAM_TYPE_MAX,
 	} type;
-	u8 num_qgv_points;
-	u8 num_psf_gv_points;
 	unsigned int fsb_freq;
 	unsigned int mem_freq;
+	u8 num_channels;
+	u8 num_qgv_points;
+	u8 num_psf_gv_points;
+	bool symmetric_memory;
+	bool wm_lv_0_adjust_needed;
 };
 
 void intel_dram_edram_detect(struct drm_i915_private *i915);

From 115cebc303db60ed7df85366aa10ef2a1780e3b1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:09 +0300
Subject: [PATCH 0970/1455] drm/i915/dram: s/wm_lv0.../has_16gb_dimms/
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The DRAM code shouldn't know anything about watermarks. Rename
wm_lv_0_adjust_needed to has_16gb_dimms. How this gets used is
up to the watermark code.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-5-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/skl_watermark.c |  2 +-
 drivers/gpu/drm/i915/soc/intel_dram.c        | 12 ++++++------
 drivers/gpu/drm/i915/soc/intel_dram.h        |  2 +-
 3 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index 33885d619a97..ae3ce0d65cfc 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -3214,7 +3214,7 @@ adjust_wm_latency(struct intel_display *display,
 	 * any underrun. If not able to get Dimm info assume 16GB dimm
 	 * to avoid any underrun.
 	 */
-	if (!display->platform.dg2 && dram_info->wm_lv_0_adjust_needed)
+	if (!display->platform.dg2 && dram_info->has_16gb_dimms)
 		wm[0] += 1;
 }
 
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index b4f0793f778d..efb72e137748 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -428,7 +428,7 @@ skl_dram_get_channels_info(struct drm_i915_private *i915, struct dram_info *dram
 		return -EINVAL;
 	}
 
-	dram_info->wm_lv_0_adjust_needed = ch0.is_16gb_dimm || ch1.is_16gb_dimm;
+	dram_info->has_16gb_dimms = ch0.is_16gb_dimm || ch1.is_16gb_dimm;
 
 	dram_info->symmetric_memory = intel_is_dram_symmetric(&ch0, &ch1);
 
@@ -673,7 +673,7 @@ static int gen11_get_dram_info(struct drm_i915_private *i915, struct dram_info *
 
 static int gen12_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info)
 {
-	dram_info->wm_lv_0_adjust_needed = false;
+	dram_info->has_16gb_dimms = false;
 
 	return icl_pcode_read_mem_global_info(i915, dram_info);
 }
@@ -737,10 +737,10 @@ int intel_dram_detect(struct drm_i915_private *i915)
 	i915->dram_info = dram_info;
 
 	/*
-	 * Assume level 0 watermark latency adjustment is needed until proven
+	 * Assume 16Gb DIMMs are present until proven
 	 * otherwise, this w/a is not needed by bxt/glk.
 	 */
-	dram_info->wm_lv_0_adjust_needed = !IS_BROXTON(i915) && !IS_GEMINILAKE(i915);
+	dram_info->has_16gb_dimms = !IS_BROXTON(i915) && !IS_GEMINILAKE(i915);
 
 	if (DISPLAY_VER(display) >= 14)
 		ret = xelpdp_get_dram_info(i915, dram_info);
@@ -766,8 +766,8 @@ int intel_dram_detect(struct drm_i915_private *i915)
 
 	drm_dbg_kms(&i915->drm, "DRAM channels: %u\n", dram_info->num_channels);
 
-	drm_dbg_kms(&i915->drm, "Watermark level 0 adjustment needed: %s\n",
-		    str_yes_no(dram_info->wm_lv_0_adjust_needed));
+	drm_dbg_kms(&i915->drm, "16Gb DIMMs: %s\n",
+		    str_yes_no(dram_info->has_16gb_dimms));
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.h b/drivers/gpu/drm/i915/soc/intel_dram.h
index 6212944d44aa..03a973f1c941 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.h
+++ b/drivers/gpu/drm/i915/soc/intel_dram.h
@@ -31,7 +31,7 @@ struct dram_info {
 	u8 num_qgv_points;
 	u8 num_psf_gv_points;
 	bool symmetric_memory;
-	bool wm_lv_0_adjust_needed;
+	bool has_16gb_dimms;
 };
 
 void intel_dram_edram_detect(struct drm_i915_private *i915);

From 84b72b545c0fe1fc3ecbe80a499f27da1de9a9e5 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:10 +0300
Subject: [PATCH 0971/1455] drm/i915/dram: Move 16Gb DIMM detection fully to
 the skl/icl codepaths
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are incorrectly applying the 16Gb DIMM w/a (adding 1 extra
usec to WM0 latency) on MTL+ even though the w/a is only needed
for SKL/ICL.

The current way of setting this is up is a bit of a disaster:
1. always set has_16gb_dimms=true for all platforms except BXT/GLK
2. has_16gb_dimms potentially gets overwritten with something else
 * BXT/GLK don't do anything since we never set has_16gb_dimms to
   begin with
 * skl_get_dram_info() overwrites has_16gb_dimms with the actual
   detection results for SKL/ICL/derivatives
 * gen12_get_dram_info() (correctly) resets has_16gb_dimms
   for TGL/ADL/derivatives
 * xelpdp_get_dram_info() doesn't do anything, leaving
   has_16gb_dimms incorrectly set for MTL+

Clean up the whole mess by only setting has_16gb_dimms in the
SKL/ICL codepaths where we have the actual detection code for
it. This avois applying the w/a incorrectly on MTL+.

v2: Rewrite commit msg (Jani)

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-6-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index efb72e137748..bf3ba874f8c5 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -406,6 +406,9 @@ skl_dram_get_channels_info(struct drm_i915_private *i915, struct dram_info *dram
 	u32 val;
 	int ret;
 
+	/* Assume 16Gb DIMMs are present until proven otherwise */
+	dram_info->has_16gb_dimms = true;
+
 	val = intel_uncore_read(&i915->uncore,
 				SKL_MAD_DIMM_CH0_0_0_0_MCHBAR_MCMAIN);
 	ret = skl_dram_get_channel_info(i915, &ch0, 0, val);
@@ -435,6 +438,9 @@ skl_dram_get_channels_info(struct drm_i915_private *i915, struct dram_info *dram
 	drm_dbg_kms(&i915->drm, "Memory configuration is symmetric? %s\n",
 		    str_yes_no(dram_info->symmetric_memory));
 
+	drm_dbg_kms(&i915->drm, "16Gb DIMMs: %s\n",
+		    str_yes_no(dram_info->has_16gb_dimms));
+
 	return 0;
 }
 
@@ -673,8 +679,6 @@ static int gen11_get_dram_info(struct drm_i915_private *i915, struct dram_info *
 
 static int gen12_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info)
 {
-	dram_info->has_16gb_dimms = false;
-
 	return icl_pcode_read_mem_global_info(i915, dram_info);
 }
 
@@ -736,12 +740,6 @@ int intel_dram_detect(struct drm_i915_private *i915)
 
 	i915->dram_info = dram_info;
 
-	/*
-	 * Assume 16Gb DIMMs are present until proven
-	 * otherwise, this w/a is not needed by bxt/glk.
-	 */
-	dram_info->has_16gb_dimms = !IS_BROXTON(i915) && !IS_GEMINILAKE(i915);
-
 	if (DISPLAY_VER(display) >= 14)
 		ret = xelpdp_get_dram_info(i915, dram_info);
 	else if (GRAPHICS_VER(i915) >= 12)
@@ -766,9 +764,6 @@ int intel_dram_detect(struct drm_i915_private *i915)
 
 	drm_dbg_kms(&i915->drm, "DRAM channels: %u\n", dram_info->num_channels);
 
-	drm_dbg_kms(&i915->drm, "16Gb DIMMs: %s\n",
-		    str_yes_no(dram_info->has_16gb_dimms));
-
 	return 0;
 }
 

From 4dfd3a56d653a96dad964c3c6f2444adb3d9c4fe Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:11 +0300
Subject: [PATCH 0972/1455] drm/i915/dram: Fix some spelling around the 16Gb
 DIMM w/a
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use consistent spelling when talking about the 16Gb DIMM w/a.

Even currently language is a bit off as the w/a is actually
about DIMMs with 16Gb DRAM devices on them, not the total capacity
of the whole DIMM. But this language does more or less match how
Bspec talks about this stuff.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-7-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/skl_watermark.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index ae3ce0d65cfc..0b9c86042e78 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -3209,9 +3209,9 @@ adjust_wm_latency(struct intel_display *display,
 	}
 
 	/*
-	 * WA Level-0 adjustment for 16GB DIMMs: SKL+
+	 * WA Level-0 adjustment for 16Gb DIMMs: SKL+
 	 * If we could not get dimm info enable this WA to prevent from
-	 * any underrun. If not able to get Dimm info assume 16GB dimm
+	 * any underrun. If not able to get DIMM info assume 16Gb DIMM
 	 * to avoid any underrun.
 	 */
 	if (!display->platform.dg2 && dram_info->has_16gb_dimms)

From a6d31e18c01211b041d416661b9ccfd75475e01b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:12 +0300
Subject: [PATCH 0973/1455] drm/i915/dram: Don't call
 skl_get_dram_info()/skl_get_dram_type() on icl
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently the icl codepaths first determine the memory type from the
memory controller registers (via skl_get_dram_info()->skl_get_dram_type())
and then overwrite the results with icl_pcode_read_mem_global_info().
Get rid of the pointless (and potentially incorrect) skl_get_dram_type()
stuff.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-8-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index bf3ba874f8c5..38b7dd20b18d 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -669,8 +669,9 @@ static int icl_pcode_read_mem_global_info(struct drm_i915_private *dev_priv,
 
 static int gen11_get_dram_info(struct drm_i915_private *i915, struct dram_info *dram_info)
 {
-	int ret = skl_get_dram_info(i915, dram_info);
+	int ret;
 
+	ret = skl_dram_get_channels_info(i915, dram_info);
 	if (ret)
 		return ret;
 

From 0d4a57f08b8320f7f65b54e4ff3ca1f55c6d70b1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Tue, 2 Sep 2025 16:31:13 +0300
Subject: [PATCH 0974/1455] drm/i915/dram: Print memory details even if
 something went wrong
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Print the memory details even if the detection failed in some way
but we continued the driver initialization anyway. It'll be easier
to debug issues if we at least know what the final results were.

And while at it also print the number of PSF GV points. Previously
we only printed the QGV points.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250902133113.18778-9-ville.syrjala@linux.intel.com
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/soc/intel_dram.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 38b7dd20b18d..149527827624 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -757,14 +757,15 @@ int intel_dram_detect(struct drm_i915_private *i915)
 	drm_dbg_kms(&i915->drm, "DRAM type: %s\n",
 		    intel_dram_type_str(dram_info->type));
 
+	drm_dbg_kms(&i915->drm, "DRAM channels: %u\n", dram_info->num_channels);
+
+	drm_dbg_kms(&i915->drm, "Num QGV points %u\n", dram_info->num_qgv_points);
+	drm_dbg_kms(&i915->drm, "Num PSF GV points %u\n", dram_info->num_psf_gv_points);
+
 	/* TODO: Do we want to abort probe on dram detection failures? */
 	if (ret)
 		return 0;
 
-	drm_dbg_kms(&i915->drm, "Num qgv points %u\n", dram_info->num_qgv_points);
-
-	drm_dbg_kms(&i915->drm, "DRAM channels: %u\n", dram_info->num_channels);
-
 	return 0;
 }
 

From f4c45db8a10b270bdd4a1644f545576c03ead523 Mon Sep 17 00:00:00 2001
From: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
Date: Fri, 29 Aug 2025 11:13:25 +0200
Subject: [PATCH 0975/1455] dt-bindings: panel: lvds: Append edt,etml0700z8dha
 in panel-lvds

List EDT ETML0700Z8DHA in the LVDS panel enumeration.

Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20250829-drm-misc-next-v1-1-fedb48cf50dd@foss.st.com
Signed-off-by: Raphael Gallais-Pou <raphael.gallais-pou@foss.st.com>
---
 Documentation/devicetree/bindings/display/panel/panel-lvds.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml b/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml
index f8f95e772778..4388d5375851 100644
--- a/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml
+++ b/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml
@@ -48,6 +48,8 @@ properties:
           - auo,g084sn05
           # Chunghwa Picture Tubes Ltd. 7" WXGA (800x1280) TFT LCD LVDS panel
           - chunghwa,claa070wp03xg
+          # EDT ETML0700Z8DHA 7.0" Full HD (1920x1080) color TFT LCD LVDS panel
+          - edt,etml0700z8dha
           # EDT ETML0700Z9NDHA 7.0" WSVGA (1024x600) color TFT LCD LVDS panel
           - edt,etml0700z9ndha
           # HannStar Display Corp. HSD101PWW2 10.1" WXGA (1280x800) LVDS panel

From 7d8a7ec813f4ee1ec4b0f94a7ec2a4d4918f43f5 Mon Sep 17 00:00:00 2001
From: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Date: Fri, 5 Sep 2025 13:46:24 +0300
Subject: [PATCH 0976/1455] drm/i915/display: take out dead code

if __waitfor timeout, ret will have -ETIMEDOUT. Then if condition
was met, and read_ret will have error that's handled.
Then if ret was zero, read_ret was zero ksv_ready must have value.

Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250905104626.1274147-1-juhapekka.heikkila@gmail.com
---
 drivers/gpu/drm/i915/display/intel_hdcp.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_hdcp.c b/drivers/gpu/drm/i915/display/intel_hdcp.c
index 910e34bd73a9..531ee122bf82 100644
--- a/drivers/gpu/drm/i915/display/intel_hdcp.c
+++ b/drivers/gpu/drm/i915/display/intel_hdcp.c
@@ -334,8 +334,6 @@ static int intel_hdcp_poll_ksv_fifo(struct intel_digital_port *dig_port,
 		return ret;
 	if (read_ret)
 		return read_ret;
-	if (!ksv_ready)
-		return -ETIMEDOUT;
 
 	return 0;
 }

From 1de89eca2004ebf0858cde504c903430b60f0975 Mon Sep 17 00:00:00 2001
From: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Date: Fri, 5 Sep 2025 13:46:25 +0300
Subject: [PATCH 0977/1455] drm/i915/display: log fail from
 intel_sdvo_enable_hotplug

Report in log if intel_sdvo_enable_hotplug failed

Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250905104626.1274147-2-juhapekka.heikkila@gmail.com
---
 drivers/gpu/drm/i915/display/intel_sdvo.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_sdvo.c b/drivers/gpu/drm/i915/display/intel_sdvo.c
index 8848a33fe9cc..6c032d81e7ee 100644
--- a/drivers/gpu/drm/i915/display/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/display/intel_sdvo.c
@@ -2052,8 +2052,10 @@ static void intel_sdvo_enable_hotplug(struct intel_encoder *encoder)
 {
 	struct intel_sdvo *intel_sdvo = to_sdvo(encoder);
 
-	intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG,
-			     &intel_sdvo->hotplug_active, 2);
+	if (!intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG,
+				  &intel_sdvo->hotplug_active, 2))
+		drm_warn(intel_sdvo->base.base.dev,
+			 "Failed to enable hotplug on SDVO encoder\n");
 }
 
 static enum intel_hotplug_state

From 70a9b201cfa893fd0b7125c8f9205d9e12e02ba5 Mon Sep 17 00:00:00 2001
From: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Date: Fri, 5 Sep 2025 13:46:26 +0300
Subject: [PATCH 0978/1455] drm/i915/display: Avoid divide by zero

skl_crtc_allocate_plane_ddb allow iter.data_rate to be zero
which could cause divide by zero in skl_allocate_plane_ddb,
check against that.

Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250905104626.1274147-3-juhapekka.heikkila@gmail.com
---
 drivers/gpu/drm/i915/display/skl_watermark.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/skl_watermark.c b/drivers/gpu/drm/i915/display/skl_watermark.c
index 0b9c86042e78..d74cbb43ae6f 100644
--- a/drivers/gpu/drm/i915/display/skl_watermark.c
+++ b/drivers/gpu/drm/i915/display/skl_watermark.c
@@ -1388,7 +1388,7 @@ skl_allocate_plane_ddb(struct skl_plane_ddb_iter *iter,
 {
 	u16 size, extra = 0;
 
-	if (data_rate) {
+	if (data_rate && iter->data_rate) {
 		extra = min_t(u16, iter->size,
 			      DIV64_U64_ROUND_UP(iter->size * data_rate,
 						 iter->data_rate));

From c4dfa0bea23df9a6870df439f2bae43ecbb73822 Mon Sep 17 00:00:00 2001
From: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
Date: Thu, 4 Sep 2025 21:44:23 +0530
Subject: [PATCH 0979/1455] drm/xe/migrate: Remove unneeded emit_pte() when
 copying CCS only

In xe_migrate_copy(), when copy_only_ccs is true, we only need two
emit_pte() calls one for the BO and one for the raw CCS storage.
However, the current implementation issues three emit_pte() calls,
resulting in an unnecessary PTE programming job.

This fix removes the redundant emit_pte() call to avoid programming
the same PTEs twice and reducing overhead during CCS-only migration.

v2: Preserve correct behavior on DG2, which requires both CCS and
page copies.

Signed-off-by: Sanjay Yadav <sanjay.kumar.yadav@intel.com>
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250904161423.2448727-1-sanjay.kumar.yadav@intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 861d9d0633d1..3f0c8832120f 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -842,11 +842,15 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 		batch_size += pte_update_size(m, pte_flags, src, &src_it, &src_L0,
 					      &src_L0_ofs, &src_L0_pt, 0, 0,
 					      avail_pts);
-
-		pte_flags = dst_is_vram ? PTE_UPDATE_FLAG_IS_VRAM : 0;
-		batch_size += pte_update_size(m, pte_flags, dst, &dst_it, &src_L0,
-					      &dst_L0_ofs, &dst_L0_pt, 0,
-					      avail_pts, avail_pts);
+		if (copy_only_ccs) {
+			dst_L0_ofs = src_L0_ofs;
+		} else {
+			pte_flags = dst_is_vram ? PTE_UPDATE_FLAG_IS_VRAM : 0;
+			batch_size += pte_update_size(m, pte_flags, dst,
+						      &dst_it, &src_L0,
+						      &dst_L0_ofs, &dst_L0_pt,
+						      0, avail_pts, avail_pts);
+		}
 
 		if (copy_system_ccs) {
 			xe_assert(xe, type_device);
@@ -876,7 +880,7 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
 
 		if (dst_is_vram && xe_migrate_allow_identity(src_L0, &dst_it))
 			xe_res_next(&dst_it, src_L0);
-		else
+		else if (!copy_only_ccs)
 			emit_pte(m, bb, dst_L0_pt, dst_is_vram, copy_system_ccs,
 				 &dst_it, src_L0, dst);
 

From 87b0a0f6bc14a95101db1502ef0b8a2ddc378e8e Mon Sep 17 00:00:00 2001
From: Chen Ni <nichen@iscas.ac.cn>
Date: Fri, 5 Sep 2025 15:37:12 +0800
Subject: [PATCH 0980/1455] drm/ast: ast_2100: Remove unneeded semicolon

Remove unnecessary semicolons reported by Coccinelle/coccicheck and the
semantic patch at scripts/coccinelle/misc/semicolon.cocci.

Signed-off-by: Chen Ni <nichen@iscas.ac.cn>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250905073712.3791260-1-nichen@iscas.ac.cn
---
 drivers/gpu/drm/ast/ast_2100.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/ast/ast_2100.c b/drivers/gpu/drm/ast/ast_2100.c
index 91541c8eaff7..829e3b8b0d19 100644
--- a/drivers/gpu/drm/ast/ast_2100.c
+++ b/drivers/gpu/drm/ast/ast_2100.c
@@ -336,7 +336,7 @@ static void ast_post_chip_2100(struct ast_device *ast)
 				default:
 					data = dram_reg_info->data;
 					break;
-				};
+				}
 
 				temp = ast_read32(ast, 0x12070);
 				temp &= 0xc;

From 09b26dce32f0d3065e6cfbc1095fc21fd5159921 Mon Sep 17 00:00:00 2001
From: Wig Cheng <onlywig@gmail.com>
Date: Tue, 2 Sep 2025 14:53:18 +0800
Subject: [PATCH 0981/1455] dt-bindings: vendor-prefixes: Add Mayqueen name

Mayqueen is a Taiwan-based company primarily focused on the development
of arm64 development boards and e-paper displays.

Signed-off-by: Wig Cheng <onlywig@gmail.com>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250902-drm-v5-1-d77c678c4ae3@gmail.com
---
 Documentation/devicetree/bindings/vendor-prefixes.yaml | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml
index abc43c43813a..7d74ecfee675 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.yaml
+++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml
@@ -933,6 +933,8 @@ patternProperties:
     description: Maxim Integrated Products
   "^maxlinear,.*":
     description: MaxLinear Inc.
+  "^mayqueen,.*":
+    description: Mayqueen Technologies Ltd.
   "^mbvl,.*":
     description: Mobiveil Inc.
   "^mcube,.*":

From d309c5fdf4a795e587f1852e2caf7f205c1bdffc Mon Sep 17 00:00:00 2001
From: LiangCheng Wang <zaq14760@gmail.com>
Date: Tue, 2 Sep 2025 14:53:19 +0800
Subject: [PATCH 0982/1455] dt-bindings: display: Add Mayqueen Pixpaper e-ink
 panel

The binding is for the Mayqueen Pixpaper e-ink display panel,
controlled via an SPI interface.

Signed-off-by: LiangCheng Wang <zaq14760@gmail.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250902-drm-v5-2-d77c678c4ae3@gmail.com
---
 .../bindings/display/mayqueen,pixpaper.yaml   | 63 +++++++++++++++++++
 1 file changed, 63 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/display/mayqueen,pixpaper.yaml

diff --git a/Documentation/devicetree/bindings/display/mayqueen,pixpaper.yaml b/Documentation/devicetree/bindings/display/mayqueen,pixpaper.yaml
new file mode 100644
index 000000000000..cd27f8ba5ae1
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/mayqueen,pixpaper.yaml
@@ -0,0 +1,63 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/display/mayqueen,pixpaper.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Mayqueen Pixpaper e-ink display panel
+
+maintainers:
+  - LiangCheng Wang <zaq14760@gmail.com>
+
+description:
+  The Pixpaper is an e-ink display panel controlled via an SPI interface.
+  The panel has a resolution of 122x250 pixels and requires GPIO pins for
+  reset, busy, and data/command control.
+
+allOf:
+  - $ref: /schemas/spi/spi-peripheral-props.yaml#
+
+properties:
+  compatible:
+    const: mayqueen,pixpaper
+
+  reg:
+    maxItems: 1
+
+  spi-max-frequency:
+    maximum: 1000000
+    default: 1000000
+
+  reset-gpios:
+    maxItems: 1
+
+  busy-gpios:
+    maxItems: 1
+
+  dc-gpios:
+    maxItems: 1
+
+required:
+  - compatible
+  - reg
+  - reset-gpios
+  - busy-gpios
+  - dc-gpios
+
+unevaluatedProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/gpio/gpio.h>
+    spi {
+        #address-cells = <1>;
+        #size-cells = <0>;
+        display@0 {
+            compatible = "mayqueen,pixpaper";
+            reg = <0>;
+            spi-max-frequency = <1000000>;
+            reset-gpios = <&gpio1 17 GPIO_ACTIVE_HIGH>;
+            busy-gpios = <&gpio1 18 GPIO_ACTIVE_HIGH>;
+            dc-gpios = <&gpio1 19 GPIO_ACTIVE_HIGH>;
+        };
+    };

From c9e70639f5915f2d084a0126e62fe53a52c29cea Mon Sep 17 00:00:00 2001
From: LiangCheng Wang <zaq14760@gmail.com>
Date: Tue, 2 Sep 2025 14:53:20 +0800
Subject: [PATCH 0983/1455] drm: tiny: Add support for Mayqueen Pixpaper e-ink
 panel

Introduce a DRM driver for the Mayqueen Pixpaper e-ink display panel,
which is controlled via SPI. The driver supports a 122x250 resolution
display with XRGB8888 format.

Also, add a MAINTAINERS entry for the Pixpaper driver.

Signed-off-by: LiangCheng Wang <zaq14760@gmail.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250902-drm-v5-3-d77c678c4ae3@gmail.com
---
 MAINTAINERS                     |    7 +
 drivers/gpu/drm/tiny/Kconfig    |   15 +
 drivers/gpu/drm/tiny/Makefile   |    1 +
 drivers/gpu/drm/tiny/pixpaper.c | 1165 +++++++++++++++++++++++++++++++
 4 files changed, 1188 insertions(+)
 create mode 100644 drivers/gpu/drm/tiny/pixpaper.c

diff --git a/MAINTAINERS b/MAINTAINERS
index c6f78b3702f6..402fe14091f1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7877,6 +7877,13 @@ T:	git https://gitlab.freedesktop.org/drm/misc/kernel.git
 F:	Documentation/devicetree/bindings/display/repaper.txt
 F:	drivers/gpu/drm/tiny/repaper.c
 
+DRM DRIVER FOR PIXPAPER E-INK PANEL
+M:	LiangCheng Wang <zaq14760@gmail.com>
+L:	dri-devel@lists.freedesktop.org
+S:	Maintained
+F:	Documentation/devicetree/bindings/display/mayqueen,pixpaper.yaml
+F:	drivers/gpu/drm/tiny/pixpaper.c
+
 DRM DRIVER FOR QEMU'S CIRRUS DEVICE
 M:	Dave Airlie <airlied@redhat.com>
 M:	Gerd Hoffmann <kraxel@redhat.com>
diff --git a/drivers/gpu/drm/tiny/Kconfig b/drivers/gpu/drm/tiny/Kconfig
index 06e54694a7f2..94a5bf61a115 100644
--- a/drivers/gpu/drm/tiny/Kconfig
+++ b/drivers/gpu/drm/tiny/Kconfig
@@ -82,6 +82,21 @@ config DRM_PANEL_MIPI_DBI
 	  https://github.com/notro/panel-mipi-dbi/wiki.
 	  To compile this driver as a module, choose M here.
 
+config DRM_PIXPAPER
+        tristate "DRM support for PIXPAPER display panels"
+        depends on DRM && SPI
+        select DRM_CLIENT_SELECTION
+        select DRM_GEM_DMA_HELPER
+        select DRM_KMS_HELPER
+        help
+	  DRM driver for the Mayqueen Pixpaper e-ink display panel.
+
+	  This driver supports small e-paper displays connected over SPI,
+	  with a resolution of 122x250 and XRGB8888 framebuffer format.
+	  It is intended for low-power embedded applications.
+
+	  If M is selected, the module will be built as pixpaper.ko.
+
 config TINYDRM_HX8357D
 	tristate "DRM support for HX8357D display panels"
 	depends on DRM && SPI
diff --git a/drivers/gpu/drm/tiny/Makefile b/drivers/gpu/drm/tiny/Makefile
index 4a9ff61ec254..48d30bf6152f 100644
--- a/drivers/gpu/drm/tiny/Makefile
+++ b/drivers/gpu/drm/tiny/Makefile
@@ -6,6 +6,7 @@ obj-$(CONFIG_DRM_BOCHS)			+= bochs.o
 obj-$(CONFIG_DRM_CIRRUS_QEMU)		+= cirrus-qemu.o
 obj-$(CONFIG_DRM_GM12U320)		+= gm12u320.o
 obj-$(CONFIG_DRM_PANEL_MIPI_DBI)	+= panel-mipi-dbi.o
+obj-$(CONFIG_DRM_PIXPAPER)              += pixpaper.o
 obj-$(CONFIG_TINYDRM_HX8357D)		+= hx8357d.o
 obj-$(CONFIG_TINYDRM_ILI9163)		+= ili9163.o
 obj-$(CONFIG_TINYDRM_ILI9225)		+= ili9225.o
diff --git a/drivers/gpu/drm/tiny/pixpaper.c b/drivers/gpu/drm/tiny/pixpaper.c
new file mode 100644
index 000000000000..b1379cb5f030
--- /dev/null
+++ b/drivers/gpu/drm/tiny/pixpaper.c
@@ -0,0 +1,1165 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * DRM driver for PIXPAPER e-ink panel
+ *
+ * Author: LiangCheng Wang <zaq14760@gmail.com>,
+ */
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/spi/spi.h>
+
+#include <drm/clients/drm_client_setup.h>
+#include <drm/drm_atomic.h>
+#include <drm/drm_atomic_helper.h>
+#include <drm/drm_drv.h>
+#include <drm/drm_fbdev_shmem.h>
+#include <drm/drm_framebuffer.h>
+#include <drm/drm_gem_atomic_helper.h>
+#include <drm/drm_gem_shmem_helper.h>
+#include <drm/drm_gem_framebuffer_helper.h>
+#include <drm/drm_probe_helper.h>
+
+/*
+ * Note on Undocumented Commands/Registers:
+ *
+ * Several commands and register parameters defined in this header are not
+ * documented in the datasheet. Their values and usage have been derived
+ * through analysis of existing userspace example programs.
+ *
+ * These 'unknown' definitions are crucial for the proper initialization
+ * and stable operation of the panel. Modifying these values without
+ * thorough understanding may lead to display anomalies, panel damage,
+ * or unexpected behavior.
+ */
+
+/* Command definitions */
+#define PIXPAPER_CMD_PANEL_SETTING 0x00	/* R00H: Panel settings */
+#define PIXPAPER_CMD_POWER_SETTING 0x01	/* R01H: Power settings */
+#define PIXPAPER_CMD_POWER_OFF 0x02		/* R02H: Power off */
+#define PIXPAPER_CMD_POWER_OFF_SEQUENCE 0x03	/* R03H: Power off sequence */
+#define PIXPAPER_CMD_POWER_ON 0x04		/* R04H: Power on */
+#define PIXPAPER_CMD_BOOSTER_SOFT_START 0x06	/* R06H: Booster soft start */
+#define PIXPAPER_CMD_DEEP_SLEEP 0x07		/* R07H: Deep sleep */
+#define PIXPAPER_CMD_DATA_START_TRANSMISSION 0x10
+/* R10H: Data transmission start */
+#define PIXPAPER_CMD_DISPLAY_REFRESH 0x12	/* R12H: Display refresh */
+#define PIXPAPER_CMD_PLL_CONTROL 0x30		/* R30H: PLL control */
+#define PIXPAPER_CMD_TEMP_SENSOR_CALIB 0x41
+/* R41H: Temperature sensor calibration */
+#define PIXPAPER_CMD_UNKNOWN_4D 0x4D		/* R4DH: Unknown command */
+#define PIXPAPER_CMD_VCOM_INTERVAL 0x50	/* R50H: VCOM interval */
+#define PIXPAPER_CMD_UNKNOWN_60 0x60		/* R60H: Unknown command */
+#define PIXPAPER_CMD_RESOLUTION_SETTING 0x61	/* R61H: Resolution settings */
+#define PIXPAPER_CMD_GATE_SOURCE_START 0x65	/* R65H: Gate/source start */
+#define PIXPAPER_CMD_UNKNOWN_B4 0xB4		/* RB4H: Unknown command */
+#define PIXPAPER_CMD_UNKNOWN_B5 0xB5		/* RB5H: Unknown command */
+#define PIXPAPER_CMD_UNKNOWN_E0 0xE0		/* RE0H: Unknown command */
+#define PIXPAPER_CMD_POWER_SAVING 0xE3		/* RE3H: Power saving */
+#define PIXPAPER_CMD_UNKNOWN_E7 0xE7		/* RE7H: Unknown command */
+#define PIXPAPER_CMD_UNKNOWN_E9 0xE9		/* RE9H: Unknown command */
+
+/* R00H PSR - First Parameter */
+#define PIXPAPER_PSR_RST_N BIT(0)
+/* Bit 0: RST_N, 1=no effect (default), 0=reset with booster OFF */
+#define PIXPAPER_PSR_SHD_N BIT(1)
+/* Bit 1: SHD_N, 1=booster ON (default), 0=booster OFF */
+#define PIXPAPER_PSR_SHL BIT(2)
+/* Bit 2: SHL, 1=shift right (default), 0=shift left */
+#define PIXPAPER_PSR_UD BIT(3)
+/* Bit 3: UD, 1=scan up (default), 0=scan down */
+#define PIXPAPER_PSR_PST_MODE BIT(5)
+/* Bit 5: PST_MODE, 0=frame scanning (default), 1=external */
+#define PIXPAPER_PSR_RES_MASK (3 << 6)
+/* Bits 7-6: RES[1:0], resolution setting */
+#define PIXPAPER_PSR_RES_176x296 (0x0 << 6)	/* 00: 176x296 */
+#define PIXPAPER_PSR_RES_128x296 (0x1 << 6)	/* 01: 128x296 */
+#define PIXPAPER_PSR_RES_128x250 (0x2 << 6)	/* 10: 128x250 */
+#define PIXPAPER_PSR_RES_112x204 (0x3 << 6)	/* 11: 112x204 */
+#define PIXPAPER_PSR_CONFIG                                           \
+	(PIXPAPER_PSR_RST_N | PIXPAPER_PSR_SHD_N | PIXPAPER_PSR_SHL | \
+	 PIXPAPER_PSR_UD)
+/* 0x0F: Default settings, resolution set by R61H */
+
+/* R00H PSR - Second Parameter */
+#define PIXPAPER_PSR2_VC_LUTZ \
+	(1 << 0) /* Bit 0: VC_LUTZ, 1=VCOM float after refresh (default), 0=no effect */
+#define PIXPAPER_PSR2_NORG \
+	(1 << 1) /* Bit 1: NORG, 1=VCOM to GND before power off, 0=no effect (default) */
+#define PIXPAPER_PSR2_TIEG \
+	(1 << 2) /* Bit 2: TIEG, 1=VGN to GND on power off, 0=no effect (default) */
+#define PIXPAPER_PSR2_TS_AUTO \
+	(1 << 3) /* Bit 3: TS_AUTO, 1=sensor on RST_N low to high (default), 0=on booster */
+#define PIXPAPER_PSR2_VCMZ \
+	(1 << 4) /* Bit 4: VCMZ, 1=VCOM always floating, 0=no effect (default) */
+#define PIXPAPER_PSR2_FOPT \
+	(1 << 5) /* Bit 5: FOPT, 0=scan 1 frame (default), 1=no scan, HiZ */
+#define PIXPAPER_PSR_CONFIG2     \
+	(PIXPAPER_PSR2_VC_LUTZ | \
+	 PIXPAPER_PSR2_TS_AUTO) /* 0x09: Default VCOM and temp sensor settings */
+
+/* R01H PWR - Power Setting Register */
+/* First Parameter */
+#define PIXPAPER_PWR_VDG_EN \
+	(1 << 0) /* Bit 0: VDG_EN, 1=internal DCDC for VGP/VGN (default), 0=external */
+#define PIXPAPER_PWR_VDS_EN \
+	(1 << 1) /* Bit 1: VDS_EN, 1=internal regulator for VSP/VSN (default), 0=external */
+#define PIXPAPER_PWR_VSC_EN \
+	(1 << 2) /* Bit 2: VSC_EN, 1=internal regulator for VSPL (default), 0=external */
+#define PIXPAPER_PWR_V_MODE \
+	(1 << 3) /* Bit 3: V_MODE, 0=Mode0 (default), 1=Mode1 */
+#define PIXPAPER_PWR_CONFIG1                         \
+	(PIXPAPER_PWR_VDG_EN | PIXPAPER_PWR_VDS_EN | \
+	 PIXPAPER_PWR_VSC_EN) /* 0x07: Internal power for VGP/VGN, VSP/VSN, VSPL */
+
+/* Second Parameter */
+#define PIXPAPER_PWR_VGPN_MASK \
+	(3 << 0) /* Bits 1-0: VGPN, VGP/VGN voltage levels */
+#define PIXPAPER_PWR_VGPN_20V (0x0 << 0) /* 00: VGP=20V, VGN=-20V (default) */
+#define PIXPAPER_PWR_VGPN_17V (0x1 << 0) /* 01: VGP=17V, VGN=-17V */
+#define PIXPAPER_PWR_VGPN_15V (0x2 << 0) /* 10: VGP=15V, VGN=-15V */
+#define PIXPAPER_PWR_VGPN_10V (0x3 << 0) /* 11: VGP=10V, VGN=-10V */
+#define PIXPAPER_PWR_CONFIG2 PIXPAPER_PWR_VGPN_20V /* 0x00: VGP=20V, VGN=-20V */
+
+/* Third, Fourth, Sixth Parameters (VSP_1, VSPL_0, VSPL_1) */
+#define PIXPAPER_PWR_VSP_8_2V 0x22 /* VSP_1/VSPL_1: 8.2V (34 decimal) */
+#define PIXPAPER_PWR_VSPL_15V 0x78 /* VSPL_0: 15V (120 decimal) */
+
+/* Fifth Parameter (VSN_1) */
+#define PIXPAPER_PWR_VSN_4V 0x0A /* VSN_1: -4V (10 decimal) */
+
+/* R03H PFS - Power Off Sequence Setting Register */
+/* First Parameter */
+#define PIXPAPER_PFS_T_VDS_OFF_MASK \
+	(3 << 0) /* Bits 1-0: T_VDS_OFF, VSP/VSN power-off sequence */
+#define PIXPAPER_PFS_T_VDS_OFF_20MS (0x0 << 0) /* 00: 20 ms (default) */
+#define PIXPAPER_PFS_T_VDS_OFF_40MS (0x1 << 0) /* 01: 40 ms */
+#define PIXPAPER_PFS_T_VDS_OFF_60MS (0x2 << 0) /* 10: 60 ms */
+#define PIXPAPER_PFS_T_VDS_OFF_80MS (0x3 << 0) /* 11: 80 ms */
+#define PIXPAPER_PFS_T_VDPG_OFF_MASK \
+	(3 << 4) /* Bits 5-4: T_VDPG_OFF, VGP/VGN power-off sequence */
+#define PIXPAPER_PFS_T_VDPG_OFF_20MS (0x0 << 4) /* 00: 20 ms (default) */
+#define PIXPAPER_PFS_T_VDPG_OFF_40MS (0x1 << 4) /* 01: 40 ms */
+#define PIXPAPER_PFS_T_VDPG_OFF_60MS (0x2 << 4) /* 10: 60 ms */
+#define PIXPAPER_PFS_T_VDPG_OFF_80MS (0x3 << 4) /* 11: 80 ms */
+#define PIXPAPER_PFS_CONFIG1           \
+	(PIXPAPER_PFS_T_VDS_OFF_20MS | \
+	 PIXPAPER_PFS_T_VDPG_OFF_20MS) /* 0x10: Default 20 ms for VSP/VSN and VGP/VGN */
+
+/* Second Parameter */
+#define PIXPAPER_PFS_VGP_EXT_MASK \
+	(0xF << 0) /* Bits 3-0: VGP_EXT, VGP extension time */
+#define PIXPAPER_PFS_VGP_EXT_0MS (0x0 << 0) /* 0000: 0 ms */
+#define PIXPAPER_PFS_VGP_EXT_500MS (0x1 << 0) /* 0001: 500 ms */
+#define PIXPAPER_PFS_VGP_EXT_1000MS (0x2 << 0) /* 0010: 1000 ms */
+#define PIXPAPER_PFS_VGP_EXT_1500MS (0x3 << 0) /* 0011: 1500 ms */
+#define PIXPAPER_PFS_VGP_EXT_2000MS (0x4 << 0) /* 0100: 2000 ms (default) */
+#define PIXPAPER_PFS_VGP_EXT_2500MS (0x5 << 0) /* 0101: 2500 ms */
+#define PIXPAPER_PFS_VGP_EXT_3000MS (0x6 << 0) /* 0110: 3000 ms */
+#define PIXPAPER_PFS_VGP_EXT_3500MS (0x7 << 0) /* 0111: 3500 ms */
+#define PIXPAPER_PFS_VGP_EXT_4000MS (0x8 << 0) /* 1000: 4000 ms */
+#define PIXPAPER_PFS_VGP_EXT_4500MS (0x9 << 0) /* 1001: 4500 ms */
+#define PIXPAPER_PFS_VGP_EXT_5000MS (0xA << 0) /* 1010: 5000 ms */
+#define PIXPAPER_PFS_VGP_EXT_5500MS (0xB << 0) /* 1011: 5500 ms */
+#define PIXPAPER_PFS_VGP_EXT_6000MS (0xC << 0) /* 1100: 6000 ms */
+#define PIXPAPER_PFS_VGP_EXT_6500MS (0xD << 0) /* 1101: 6500 ms */
+#define PIXPAPER_PFS_VGP_LEN_MASK \
+	(0xF << 4) /* Bits 7-4: VGP_LEN, VGP at 10V during power-off */
+#define PIXPAPER_PFS_VGP_LEN_0MS (0x0 << 4) /* 0000: 0 ms */
+#define PIXPAPER_PFS_VGP_LEN_500MS (0x1 << 4) /* 0001: 500 ms */
+#define PIXPAPER_PFS_VGP_LEN_1000MS (0x2 << 4) /* 0010: 1000 ms */
+#define PIXPAPER_PFS_VGP_LEN_1500MS (0x3 << 4) /* 0011: 1500 ms */
+#define PIXPAPER_PFS_VGP_LEN_2000MS (0x4 << 4) /* 0100: 2000 ms */
+#define PIXPAPER_PFS_VGP_LEN_2500MS (0x5 << 4) /* 0101: 2500 ms (default) */
+#define PIXPAPER_PFS_VGP_LEN_3000MS (0x6 << 4) /* 0110: 3000 ms */
+#define PIXPAPER_PFS_VGP_LEN_3500MS (0x7 << 4) /* 0111: 3500 ms */
+#define PIXPAPER_PFS_VGP_LEN_4000MS (0x8 << 4) /* 1000: 4000 ms */
+#define PIXPAPER_PFS_VGP_LEN_4500MS (0x9 << 4) /* 1001: 4500 ms */
+#define PIXPAPER_PFS_VGP_LEN_5000MS (0xA << 4) /* 1010: 5000 ms */
+#define PIXPAPER_PFS_VGP_LEN_5500MS (0xB << 4) /* 1011: 5500 ms */
+#define PIXPAPER_PFS_VGP_LEN_6000MS (0xC << 4) /* 1100: 6000 ms */
+#define PIXPAPER_PFS_VGP_LEN_6500MS (0xD << 4) /* 1101: 6500 ms */
+#define PIXPAPER_PFS_CONFIG2           \
+	(PIXPAPER_PFS_VGP_EXT_1000MS | \
+	 PIXPAPER_PFS_VGP_LEN_2500MS) /* 0x54: VGP extension 1000 ms, VGP at 10V for 2500 ms */
+
+/* Third Parameter */
+#define PIXPAPER_PFS_XON_LEN_MASK \
+	(0xF << 0) /* Bits 3-0: XON_LEN, XON enable time */
+#define PIXPAPER_PFS_XON_LEN_0MS (0x0 << 0) /* 0000: 0 ms */
+#define PIXPAPER_PFS_XON_LEN_500MS (0x1 << 0) /* 0001: 500 ms */
+#define PIXPAPER_PFS_XON_LEN_1000MS (0x2 << 0) /* 0010: 1000 ms */
+#define PIXPAPER_PFS_XON_LEN_1500MS (0x3 << 0) /* 0011: 1500 ms */
+#define PIXPAPER_PFS_XON_LEN_2000MS (0x4 << 0) /* 0100: 2000 ms (default) */
+#define PIXPAPER_PFS_XON_LEN_2500MS (0x5 << 0) /* 0101: 2500 ms */
+#define PIXPAPER_PFS_XON_LEN_3000MS (0x6 << 0) /* 0110: 3000 ms */
+#define PIXPAPER_PFS_XON_LEN_3500MS (0x7 << 0) /* 0111: 3500 ms */
+#define PIXPAPER_PFS_XON_LEN_4000MS (0x8 << 0) /* 1000: 4000 ms */
+#define PIXPAPER_PFS_XON_LEN_4500MS (0x9 << 0) /* 1001: 4500 ms */
+#define PIXPAPER_PFS_XON_LEN_5000MS (0xA << 0) /* 1010: 5000 ms */
+#define PIXPAPER_PFS_XON_LEN_5500MS (0xB << 0) /* 1011: 5500 ms */
+#define PIXPAPER_PFS_XON_LEN_6000MS (0xC << 0) /* 1100: 6000 ms */
+#define PIXPAPER_PFS_XON_DLY_MASK \
+	(0xF << 4) /* Bits 7-4: XON_DLY, XON delay time */
+#define PIXPAPER_PFS_XON_DLY_0MS (0x0 << 4) /* 0000: 0 ms */
+#define PIXPAPER_PFS_XON_DLY_500MS (0x1 << 4) /* 0001: 500 ms */
+#define PIXPAPER_PFS_XON_DLY_1000MS (0x2 << 4) /* 0010: 1000 ms */
+#define PIXPAPER_PFS_XON_DLY_1500MS (0x3 << 4) /* 0011: 1500 ms */
+#define PIXPAPER_PFS_XON_DLY_2000MS (0x4 << 4) /* 0100: 2000 ms (default) */
+#define PIXPAPER_PFS_XON_DLY_2500MS (0x5 << 4) /* 0101: 2500 ms */
+#define PIXPAPER_PFS_XON_DLY_3000MS (0x6 << 4) /* 0110: 3000 ms */
+#define PIXPAPER_PFS_XON_DLY_3500MS (0x7 << 4) /* 0111: 3500 ms */
+#define PIXPAPER_PFS_XON_DLY_4000MS (0x8 << 4) /* 1000: 4000 ms */
+#define PIXPAPER_PFS_XON_DLY_4500MS (0x9 << 4) /* 1001: 4500 ms */
+#define PIXPAPER_PFS_XON_DLY_5000MS (0xA << 4) /* 1010: 5000 ms */
+#define PIXPAPER_PFS_XON_DLY_5500MS (0xB << 4) /* 1011: 5500 ms */
+#define PIXPAPER_PFS_XON_DLY_6000MS (0xC << 4) /* 1100: 6000 ms */
+#define PIXPAPER_PFS_CONFIG3           \
+	(PIXPAPER_PFS_XON_LEN_2000MS | \
+	 PIXPAPER_PFS_XON_DLY_2000MS) /* 0x44: XON enable and delay at 2000 ms */
+
+/* R06H BTST - Booster Soft Start Command */
+/* First Parameter */
+#define PIXPAPER_BTST_PHA_SFT_MASK \
+	(3 << 0) /* Bits 1-0: PHA_SFT, soft start period for phase A */
+#define PIXPAPER_BTST_PHA_SFT_10MS (0x0 << 0) /* 00: 10 ms (default) */
+#define PIXPAPER_BTST_PHA_SFT_20MS (0x1 << 0) /* 01: 20 ms */
+#define PIXPAPER_BTST_PHA_SFT_30MS (0x2 << 0) /* 10: 30 ms */
+#define PIXPAPER_BTST_PHA_SFT_40MS (0x3 << 0) /* 11: 40 ms */
+#define PIXPAPER_BTST_PHB_SFT_MASK \
+	(3 << 2) /* Bits 3-2: PHB_SFT, soft start period for phase B */
+#define PIXPAPER_BTST_PHB_SFT_10MS (0x0 << 2) /* 00: 10 ms (default) */
+#define PIXPAPER_BTST_PHB_SFT_20MS (0x1 << 2) /* 01: 20 ms */
+#define PIXPAPER_BTST_PHB_SFT_30MS (0x2 << 2) /* 10: 30 ms */
+#define PIXPAPER_BTST_PHB_SFT_40MS (0x3 << 2) /* 11: 40 ms */
+#define PIXPAPER_BTST_CONFIG1         \
+	(PIXPAPER_BTST_PHA_SFT_40MS | \
+	 PIXPAPER_BTST_PHB_SFT_40MS) /* 0x0F: 40 ms for phase A and B */
+
+/* Second to Seventh Parameters (Driving Strength or Minimum OFF Time) */
+#define PIXPAPER_BTST_CONFIG2 0x0A /* Strength11 */
+#define PIXPAPER_BTST_CONFIG3 0x2F /* Period48 */
+#define PIXPAPER_BTST_CONFIG4 0x25 /* Strength38 */
+#define PIXPAPER_BTST_CONFIG5 0x22 /* Period35 */
+#define PIXPAPER_BTST_CONFIG6 0x2E /* Strength47 */
+#define PIXPAPER_BTST_CONFIG7 0x21 /* Period34 */
+
+/* R12H: DRF (Display Refresh) */
+#define PIXPAPER_DRF_VCOM_AC 0x00 /* AC VCOM: VCOM follows LUTC (default) */
+#define PIXPAPER_DRF_VCOM_DC 0x01 /* DC VCOM: VCOM fixed to VCOMDC */
+
+/* R30H PLL - PLL Control Register */
+/* First Parameter */
+#define PIXPAPER_PLL_FR_MASK (0x7 << 0) /* Bits 2-0: FR, frame rate */
+#define PIXPAPER_PLL_FR_12_5HZ (0x0 << 0) /* 000: 12.5 Hz */
+#define PIXPAPER_PLL_FR_25HZ (0x1 << 0)	/* 001: 25 Hz */
+#define PIXPAPER_PLL_FR_50HZ (0x2 << 0) /* 010: 50 Hz (default) */
+#define PIXPAPER_PLL_FR_65HZ (0x3 << 0) /* 011: 65 Hz */
+#define PIXPAPER_PLL_FR_75HZ (0x4 << 0) /* 100: 75 Hz */
+#define PIXPAPER_PLL_FR_85HZ (0x5 << 0) /* 101: 85 Hz */
+#define PIXPAPER_PLL_FR_100HZ (0x6 << 0) /* 110: 100 Hz */
+#define PIXPAPER_PLL_FR_120HZ (0x7 << 0) /* 111: 120 Hz */
+#define PIXPAPER_PLL_DFR \
+	(1 << 3) /* Bit 3: Dynamic frame rate, 0=disabled (default), 1=enabled */
+#define PIXPAPER_PLL_CONFIG \
+	(PIXPAPER_PLL_FR_50HZ) /* 0x02: 50 Hz, dynamic frame rate disabled */
+
+/* R41H TSE - Temperature Sensor Calibration Register */
+/* First Parameter */
+#define PIXPAPER_TSE_TO_MASK \
+	(0xF << 0) /* Bits 3-0: TO[3:0], temperature offset */
+#define PIXPAPER_TSE_TO_POS_0C (0x0 << 0) /* 0000: +0°C (default) */
+#define PIXPAPER_TSE_TO_POS_0_5C (0x1 << 0) /* 0001: +0.5°C */
+#define PIXPAPER_TSE_TO_POS_1C (0x2 << 0) /* 0010: +1°C */
+#define PIXPAPER_TSE_TO_POS_1_5C (0x3 << 0) /* 0011: +1.5°C */
+#define PIXPAPER_TSE_TO_POS_2C (0x4 << 0) /* 0100: +2°C */
+#define PIXPAPER_TSE_TO_POS_2_5C (0x5 << 0) /* 0101: +2.5°C */
+#define PIXPAPER_TSE_TO_POS_3C (0x6 << 0) /* 0110: +3°C */
+#define PIXPAPER_TSE_TO_POS_3_5C (0x7 << 0) /* 0111: +3.5°C */
+#define PIXPAPER_TSE_TO_NEG_4C (0x8 << 0) /* 1000: -4°C */
+#define PIXPAPER_TSE_TO_NEG_3_5C (0x9 << 0) /* 1001: -3.5°C */
+#define PIXPAPER_TSE_TO_NEG_3C (0xA << 0) /* 1010: -3°C */
+#define PIXPAPER_TSE_TO_NEG_2_5C (0xB << 0) /* 1011: -2.5°C */
+#define PIXPAPER_TSE_TO_NEG_2C (0xC << 0) /* 1100: -2°C */
+#define PIXPAPER_TSE_TO_NEG_1_5C (0xD << 0) /* 1101: -1.5°C */
+#define PIXPAPER_TSE_TO_NEG_1C (0xE << 0) /* 1110: -1°C */
+#define PIXPAPER_TSE_TO_NEG_0_5C (0xF << 0) /* 1111: -0.5°C */
+#define PIXPAPER_TSE_TO_FINE_MASK \
+	(0x3 << 4) /* Bits 5-4: TO[5:4], fine adjustment for positive offsets */
+#define PIXPAPER_TSE_TO_FINE_0C (0x0 << 4) /* 00: +0.0°C (default) */
+#define PIXPAPER_TSE_TO_FINE_0_25C (0x1 << 4) /* 01: +0.25°C */
+#define PIXPAPER_TSE_ENABLE \
+	(0 << 7) /* Bit 7: TSE, 0=internal sensor enabled (default), 1=disabled (external) */
+#define PIXPAPER_TSE_DISABLE \
+	(1 << 7) /* Bit 7: TSE, 1=internal sensor disabled, use external */
+#define PIXPAPER_TSE_CONFIG                                 \
+	(PIXPAPER_TSE_TO_POS_0C | PIXPAPER_TSE_TO_FINE_0C | \
+	 PIXPAPER_TSE_ENABLE) /* 0x00: Internal sensor enabled, +0°C offset */
+
+/* R4DH */
+#define PIXPAPER_UNKNOWN_4D_CONFIG \
+	0x78 /* This value is essential for initialization, derived from userspace examples. */
+
+/* R50H CDI - VCOM and DATA Interval Setting Register */
+/* First Parameter */
+#define PIXPAPER_CDI_INTERVAL_MASK \
+	(0xF << 0) /* Bits 3-0: CDI[3:0], VCOM and data interval (hsync) */
+#define PIXPAPER_CDI_17_HSYNC (0x0 << 0) /* 0000: 17 hsync */
+#define PIXPAPER_CDI_16_HSYNC (0x1 << 0) /* 0001: 16 hsync */
+#define PIXPAPER_CDI_15_HSYNC (0x2 << 0) /* 0010: 15 hsync */
+#define PIXPAPER_CDI_14_HSYNC (0x3 << 0) /* 0011: 14 hsync */
+#define PIXPAPER_CDI_13_HSYNC (0x4 << 0) /* 0100: 13 hsync */
+#define PIXPAPER_CDI_12_HSYNC (0x5 << 0) /* 0101: 12 hsync */
+#define PIXPAPER_CDI_11_HSYNC (0x6 << 0) /* 0110: 11 hsync */
+#define PIXPAPER_CDI_10_HSYNC (0x7 << 0) /* 0111: 10 hsync (default) */
+#define PIXPAPER_CDI_9_HSYNC (0x8 << 0) /* 1000: 9 hsync */
+#define PIXPAPER_CDI_8_HSYNC (0x9 << 0) /* 1001: 8 hsync */
+#define PIXPAPER_CDI_7_HSYNC (0xA << 0) /* 1010: 7 hsync */
+#define PIXPAPER_CDI_6_HSYNC (0xB << 0) /* 1011: 6 hsync */
+#define PIXPAPER_CDI_5_HSYNC (0xC << 0) /* 1100: 5 hsync */
+#define PIXPAPER_CDI_4_HSYNC (0xD << 0) /* 1101: 4 hsync */
+#define PIXPAPER_CDI_3_HSYNC (0xE << 0) /* 1110: 3 hsync */
+#define PIXPAPER_CDI_2_HSYNC (0xF << 0) /* 1111: 2 hsync */
+#define PIXPAPER_CDI_DDX \
+	(1 << 4) /* Bit 4: DDX, 0=grayscale mapping 0, 1=grayscale mapping 1 (default) */
+#define PIXPAPER_CDI_VBD_MASK \
+	(0x7 << 5) /* Bits 7-5: VBD[2:0], border data selection */
+#define PIXPAPER_CDI_VBD_FLOAT (0x0 << 5) /* 000: Floating (DDX=0 or 1) */
+#define PIXPAPER_CDI_VBD_GRAY3_DDX0 \
+	(0x1 << 5) /* 001: Gray3 (border_buf=011) when DDX=0 */
+#define PIXPAPER_CDI_VBD_GRAY2_DDX0 \
+	(0x2 << 5) /* 010: Gray2 (border_buf=010) when DDX=0 */
+#define PIXPAPER_CDI_VBD_GRAY1_DDX0 \
+	(0x3 << 5) /* 011: Gray1 (border_buf=001) when DDX=0 */
+#define PIXPAPER_CDI_VBD_GRAY0_DDX0 \
+	(0x4 << 5) /* 100: Gray0 (border_buf=000) when DDX=0 */
+#define PIXPAPER_CDI_VBD_GRAY0_DDX1 \
+	(0x0 << 5) /* 000: Gray0 (border_buf=000) when DDX=1 */
+#define PIXPAPER_CDI_VBD_GRAY1_DDX1 \
+	(0x1 << 5) /* 001: Gray1 (border_buf=001) when DDX=1 */
+#define PIXPAPER_CDI_VBD_GRAY2_DDX1 \
+	(0x2 << 5) /* 010: Gray2 (border_buf=010) when DDX=1 */
+#define PIXPAPER_CDI_VBD_GRAY3_DDX1 \
+	(0x3 << 5) /* 011: Gray3 (border_buf=011) when DDX=1 */
+#define PIXPAPER_CDI_VBD_FLOAT_DDX1 (0x4 << 5) /* 100: Floating when DDX=1 */
+#define PIXPAPER_CDI_CONFIG                         \
+	(PIXPAPER_CDI_10_HSYNC | PIXPAPER_CDI_DDX | \
+	 PIXPAPER_CDI_VBD_GRAY1_DDX1) /* 0x37: 10 hsync, DDX=1, border Gray1 */
+
+/* R60H */
+#define PIXPAPER_UNKNOWN_60_CONFIG1 \
+	0x02 /* This value is essential for initialization, derived from userspace examples. */
+#define PIXPAPER_UNKNOWN_60_CONFIG2 \
+	0x02 /* This value is essential for initialization, derived from userspace examples. */
+
+/* R61H TRES - Resolution Setting Register */
+#define PIXPAPER_TRES_HRES_H                  \
+	((PIXPAPER_PANEL_BUFFER_WIDTH >> 8) & \
+	 0xFF) /* HRES[9:8]: High byte of horizontal resolution (128) */
+#define PIXPAPER_TRES_HRES_L           \
+	(PIXPAPER_PANEL_BUFFER_WIDTH & \
+	 0xFF) /* HRES[7:0]: Low byte of horizontal resolution (128 = 0x80) */
+#define PIXPAPER_TRES_VRES_H      \
+	((PIXPAPER_HEIGHT >> 8) & \
+	 0xFF) /* VRES[9:8]: High byte of vertical resolution (250) */
+#define PIXPAPER_TRES_VRES_L \
+	(PIXPAPER_HEIGHT &   \
+	 0xFF) /* VRES[7:0]: Low byte of vertical resolution (250 = 0xFA) */
+
+/* R65H GSST - Gate/Source Start Setting Register */
+#define PIXPAPER_GSST_S_START 0x00 /* S_Start[7:0]: First source line (S0) */
+#define PIXPAPER_GSST_RESERVED 0x00 /* Reserved byte */
+#define PIXPAPER_GSST_G_START_H \
+	0x00 /* G_Start[8]: High bit of first gate line (G0) */
+#define PIXPAPER_GSST_G_START_L \
+	0x00 /* G_Start[7:0]: Low byte of first gate line (G0) */
+
+/* RB4H */
+#define PIXPAPER_UNKNOWN_B4_CONFIG \
+	0xD0 /* This value is essential for initialization, derived from userspace examples. */
+
+/* RB5H */
+#define PIXPAPER_UNKNOWN_B5_CONFIG \
+	0x03 /* This value is essential for initialization, derived from userspace examples. */
+
+/* RE0H */
+#define PIXPAPER_UNKNOWN_E0_CONFIG \
+	0x00 /* This value is essential for initialization, derived from userspace examples. */
+
+/* RE3H PWS - Power Saving Register */
+/* First Parameter */
+#define PIXPAPER_PWS_VCOM_W_MASK \
+	(0xF                     \
+	 << 4) /* Bits 7-4: VCOM_W[3:0], VCOM power-saving width (line periods) */
+#define PIXPAPER_PWS_VCOM_W_0 (0x0 << 4) /* 0000: 0 line periods */
+#define PIXPAPER_PWS_VCOM_W_1 (0x1 << 4) /* 0001: 1 line period */
+#define PIXPAPER_PWS_VCOM_W_2 (0x2 << 4) /* 0010: 2 line periods */
+#define PIXPAPER_PWS_VCOM_W_3 (0x3 << 4) /* 0011: 3 line periods */
+#define PIXPAPER_PWS_VCOM_W_4 (0x4 << 4) /* 0100: 4 line periods */
+#define PIXPAPER_PWS_VCOM_W_5 (0x5 << 4) /* 0101: 5 line periods */
+#define PIXPAPER_PWS_VCOM_W_6 (0x6 << 4) /* 0110: 6 line periods */
+#define PIXPAPER_PWS_VCOM_W_7 (0x7 << 4) /* 0111: 7 line periods */
+#define PIXPAPER_PWS_VCOM_W_8 (0x8 << 4) /* 1000: 8 line periods */
+#define PIXPAPER_PWS_VCOM_W_9 (0x9 << 4) /* 1001: 9 line periods */
+#define PIXPAPER_PWS_VCOM_W_10 (0xA << 4) /* 1010: 10 line periods */
+#define PIXPAPER_PWS_VCOM_W_11 (0xB << 4) /* 1011: 11 line periods */
+#define PIXPAPER_PWS_VCOM_W_12 (0xC << 4) /* 1100: 12 line periods */
+#define PIXPAPER_PWS_VCOM_W_13 (0xD << 4) /* 1101: 13 line periods */
+#define PIXPAPER_PWS_VCOM_W_14 (0xE << 4) /* 1110: 14 line periods */
+#define PIXPAPER_PWS_VCOM_W_15 (0xF << 4) /* 1111: 15 line periods */
+#define PIXPAPER_PWS_SD_W_MASK \
+	(0xF << 0) /* Bits 3-0: SD_W[3:0], source power-saving width (660 ns units) */
+#define PIXPAPER_PWS_SD_W_0 (0x0 << 0) /* 0000: 0 ns */
+#define PIXPAPER_PWS_SD_W_1 (0x1 << 0) /* 0001: 660 ns */
+#define PIXPAPER_PWS_SD_W_2 (0x2 << 0) /* 0010: 1320 ns */
+#define PIXPAPER_PWS_SD_W_3 (0x3 << 0) /* 0011: 1980 ns */
+#define PIXPAPER_PWS_SD_W_4 (0x4 << 0) /* 0100: 2640 ns */
+#define PIXPAPER_PWS_SD_W_5 (0x5 << 0) /* 0101: 3300 ns */
+#define PIXPAPER_PWS_SD_W_6 (0x6 << 0) /* 0110: 3960 ns */
+#define PIXPAPER_PWS_SD_W_7 (0x7 << 0) /* 0111: 4620 ns */
+#define PIXPAPER_PWS_SD_W_8 (0x8 << 0) /* 1000: 5280 ns */
+#define PIXPAPER_PWS_SD_W_9 (0x9 << 0) /* 1001: 5940 ns */
+#define PIXPAPER_PWS_SD_W_10 (0xA << 0) /* 1010: 6600 ns */
+#define PIXPAPER_PWS_SD_W_11 (0xB << 0) /* 1011: 7260 ns */
+#define PIXPAPER_PWS_SD_W_12 (0xC << 0) /* 1100: 7920 ns */
+#define PIXPAPER_PWS_SD_W_13 (0xD << 0) /* 1101: 8580 ns */
+#define PIXPAPER_PWS_SD_W_14 (0xE << 0) /* 1110: 9240 ns */
+#define PIXPAPER_PWS_SD_W_15 (0xF << 0) /* 1111: 9900 ns */
+#define PIXPAPER_PWS_CONFIG      \
+	(PIXPAPER_PWS_VCOM_W_2 | \
+	 PIXPAPER_PWS_SD_W_2) /* 0x22: VCOM 2 line periods (160 µs), source 1320 ns */
+
+/* RE7H */
+#define PIXPAPER_UNKNOWN_E7_CONFIG \
+	0x1C /* This value is essential for initialization, derived from userspace examples. */
+
+/* RE9H */
+#define PIXPAPER_UNKNOWN_E9_CONFIG \
+	0x01 /* This value is essential for initialization, derived from userspace examples. */
+
+MODULE_IMPORT_NS("DMA_BUF");
+
+/*
+ * The panel has a visible resolution of 122x250.
+ * However, the controller requires the horizontal resolution to be aligned to 128 pixels.
+ * No porch or sync timing values are provided in the datasheet, so we define minimal
+ * placeholder values to satisfy the DRM framework.
+ */
+
+/* Panel visible resolution */
+#define PIXPAPER_WIDTH           122
+#define PIXPAPER_HEIGHT          250
+
+/* Controller requires 128 horizontal pixels total (for memory alignment) */
+#define PIXPAPER_HTOTAL          128
+#define PIXPAPER_HFP             2
+#define PIXPAPER_HSYNC           2
+#define PIXPAPER_HBP             (PIXPAPER_HTOTAL - PIXPAPER_WIDTH - PIXPAPER_HFP - PIXPAPER_HSYNC)
+
+/*
+ * According to the datasheet, the total vertical blanking must be 55 lines,
+ * regardless of how the vertical back porch is set.
+ * Here we allocate VFP=2, VSYNC=2, and VBP=51 to sum up to 55 lines.
+ * Total vertical lines = 250 (visible) + 55 (blanking) = 305.
+ */
+#define PIXPAPER_VTOTAL  (250 + 55)
+#define PIXPAPER_VFP     2
+#define PIXPAPER_VSYNC   2
+#define PIXPAPER_VBP     (55 - PIXPAPER_VFP - PIXPAPER_VSYNC)
+
+/*
+ * Pixel clock calculation:
+ * pixel_clock = htotal * vtotal * refresh_rate
+ *             = 128 * 305 * 50
+ *             = 1,952,000 Hz = 1952 kHz
+ */
+#define PIXPAPER_PIXEL_CLOCK     1952
+
+#define PIXPAPER_WIDTH_MM        24    /* approximate from 23.7046mm */
+#define PIXPAPER_HEIGHT_MM       49    /* approximate from 48.55mm */
+
+#define PIXPAPER_SPI_BITS_PER_WORD	8
+#define PIXPAPER_SPI_SPEED_DEFAULT      1000000
+
+#define PIXPAPER_PANEL_BUFFER_WIDTH	128
+#define PIXPAPER_PANEL_BUFFER_TWO_BYTES_PER_ROW (PIXPAPER_PANEL_BUFFER_WIDTH / 4)
+
+#define PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL		60
+#define PIXPAPER_COLOR_THRESHOLD_HIGH_CHANNEL		200
+#define PIXPAPER_COLOR_THRESHOLD_YELLOW_MIN_GREEN	180
+
+struct pixpaper_error_ctx {
+	int errno_code;
+};
+
+struct pixpaper_panel {
+	struct drm_device drm;
+	struct drm_plane plane;
+	struct drm_crtc crtc;
+	struct drm_encoder encoder;
+	struct drm_connector connector;
+
+	struct spi_device *spi;
+	struct gpio_desc *reset;
+	struct gpio_desc *busy;
+	struct gpio_desc *dc;
+};
+
+static inline struct pixpaper_panel *to_pixpaper_panel(struct drm_device *drm)
+{
+	return container_of(drm, struct pixpaper_panel, drm);
+}
+
+static void pixpaper_wait_for_panel(struct pixpaper_panel *panel)
+{
+	unsigned int timeout_ms = 10000;
+	unsigned long timeout_jiffies = jiffies + msecs_to_jiffies(timeout_ms);
+
+	usleep_range(1000, 1500);
+	while (gpiod_get_value_cansleep(panel->busy) != 1) {
+		if (time_after(jiffies, timeout_jiffies)) {
+			drm_warn(&panel->drm, "Busy wait timed out\n");
+			return;
+		}
+		usleep_range(100, 200);
+	}
+}
+
+static void pixpaper_spi_sync(struct spi_device *spi, struct spi_message *msg,
+			      struct pixpaper_error_ctx *err)
+{
+	if (err->errno_code)
+		return;
+
+	int ret = spi_sync(spi, msg);
+
+	if (ret < 0)
+		err->errno_code = ret;
+}
+
+static void pixpaper_send_cmd(struct pixpaper_panel *panel, u8 cmd,
+			      struct pixpaper_error_ctx *err)
+{
+	if (err->errno_code)
+		return;
+
+	struct spi_transfer xfer = {
+		.tx_buf = &cmd,
+		.len = 1,
+	};
+	struct spi_message msg;
+
+	spi_message_init(&msg);
+	spi_message_add_tail(&xfer, &msg);
+
+	gpiod_set_value_cansleep(panel->dc, 0);
+	usleep_range(1, 5);
+	pixpaper_spi_sync(panel->spi, &msg, err);
+}
+
+static void pixpaper_send_data(struct pixpaper_panel *panel, u8 data,
+			       struct pixpaper_error_ctx *err)
+{
+	if (err->errno_code)
+		return;
+
+	struct spi_transfer xfer = {
+		.tx_buf = &data,
+		.len = 1,
+	};
+	struct spi_message msg;
+
+	spi_message_init(&msg);
+	spi_message_add_tail(&xfer, &msg);
+
+	gpiod_set_value_cansleep(panel->dc, 1);
+	usleep_range(1, 5);
+	pixpaper_spi_sync(panel->spi, &msg, err);
+}
+
+static int pixpaper_panel_hw_init(struct pixpaper_panel *panel)
+{
+	struct pixpaper_error_ctx err = { .errno_code = 0 };
+
+	gpiod_set_value_cansleep(panel->reset, 0);
+	msleep(50);
+	gpiod_set_value_cansleep(panel->reset, 1);
+	msleep(50);
+
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_4D, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_4D_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_PANEL_SETTING, &err);
+	pixpaper_send_data(panel, PIXPAPER_PSR_CONFIG, &err);
+	pixpaper_send_data(panel, PIXPAPER_PSR_CONFIG2, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_POWER_SETTING, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWR_CONFIG1, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWR_CONFIG2, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWR_VSP_8_2V, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWR_VSPL_15V, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWR_VSN_4V, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWR_VSP_8_2V, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_POWER_OFF_SEQUENCE, &err);
+	pixpaper_send_data(panel, PIXPAPER_PFS_CONFIG1, &err);
+	pixpaper_send_data(panel, PIXPAPER_PFS_CONFIG2, &err);
+	pixpaper_send_data(panel, PIXPAPER_PFS_CONFIG3, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_BOOSTER_SOFT_START, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG1, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG2, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG3, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG4, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG5, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG6, &err);
+	pixpaper_send_data(panel, PIXPAPER_BTST_CONFIG7, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_PLL_CONTROL, &err);
+	pixpaper_send_data(panel, PIXPAPER_PLL_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_TEMP_SENSOR_CALIB, &err);
+	pixpaper_send_data(panel, PIXPAPER_TSE_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_VCOM_INTERVAL, &err);
+	pixpaper_send_data(panel, PIXPAPER_CDI_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_60, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_60_CONFIG1, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_60_CONFIG2, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_RESOLUTION_SETTING, &err);
+	pixpaper_send_data(panel, PIXPAPER_TRES_HRES_H, &err);
+	pixpaper_send_data(panel, PIXPAPER_TRES_HRES_L, &err);
+	pixpaper_send_data(panel, PIXPAPER_TRES_VRES_H, &err);
+	pixpaper_send_data(panel, PIXPAPER_TRES_VRES_L, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_GATE_SOURCE_START, &err);
+	pixpaper_send_data(panel, PIXPAPER_GSST_S_START, &err);
+	pixpaper_send_data(panel, PIXPAPER_GSST_RESERVED, &err);
+	pixpaper_send_data(panel, PIXPAPER_GSST_G_START_H, &err);
+	pixpaper_send_data(panel, PIXPAPER_GSST_G_START_L, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_E7, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_E7_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_POWER_SAVING, &err);
+	pixpaper_send_data(panel, PIXPAPER_PWS_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_E0, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_E0_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_B4, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_B4_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_B5, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_B5_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_UNKNOWN_E9, &err);
+	pixpaper_send_data(panel, PIXPAPER_UNKNOWN_E9_CONFIG, &err);
+	if (err.errno_code)
+		goto init_fail;
+	pixpaper_wait_for_panel(panel);
+
+	return 0;
+
+init_fail:
+	drm_err(&panel->drm, "Hardware initialization failed (err=%d)\n",
+		err.errno_code);
+	return err.errno_code;
+}
+
+/*
+ * Convert framebuffer pixels to 2-bit e-paper format:
+ *   00 - White
+ *   01 - Black
+ *   10 - Yellow
+ *   11 - Red
+ */
+static u8 pack_pixels_to_byte(__le32 *src_pixels, int i, int j,
+			      struct drm_framebuffer *fb)
+{
+	u8 packed_byte = 0;
+	int k;
+
+	for (k = 0; k < 4; k++) {
+		int current_pixel_x = j * 4 + k;
+		u8 two_bit_val;
+
+		if (current_pixel_x < PIXPAPER_WIDTH) {
+			u32 pixel_offset =
+				(i * (fb->pitches[0] / 4)) + current_pixel_x;
+			u32 pixel = le32_to_cpu(src_pixels[pixel_offset]);
+			u32 r = (pixel >> 16) & 0xFF;
+			u32 g = (pixel >> 8) & 0xFF;
+			u32 b = pixel & 0xFF;
+
+			if (r < PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL &&
+			    g < PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL &&
+			    b < PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL) {
+				two_bit_val = 0b00;
+			} else if (r > PIXPAPER_COLOR_THRESHOLD_HIGH_CHANNEL &&
+				   g > PIXPAPER_COLOR_THRESHOLD_HIGH_CHANNEL &&
+				   b > PIXPAPER_COLOR_THRESHOLD_HIGH_CHANNEL) {
+				two_bit_val = 0b01;
+			} else if (r > PIXPAPER_COLOR_THRESHOLD_HIGH_CHANNEL &&
+				   g < PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL &&
+				   b < PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL) {
+				two_bit_val = 0b11;
+			} else if (r > PIXPAPER_COLOR_THRESHOLD_HIGH_CHANNEL &&
+				   g > PIXPAPER_COLOR_THRESHOLD_YELLOW_MIN_GREEN &&
+				   b < PIXPAPER_COLOR_THRESHOLD_LOW_CHANNEL) {
+				two_bit_val = 0b10;
+			} else {
+				two_bit_val = 0b01;
+			}
+		} else {
+			two_bit_val = 0b01;
+		}
+
+		packed_byte |= two_bit_val << ((3 - k) * 2);
+	}
+
+	return packed_byte;
+}
+
+static int pixpaper_plane_helper_atomic_check(struct drm_plane *plane,
+					      struct drm_atomic_state *state)
+{
+	struct drm_plane_state *new_plane_state =
+		drm_atomic_get_new_plane_state(state, plane);
+	struct drm_crtc *new_crtc = new_plane_state->crtc;
+	struct drm_crtc_state *new_crtc_state = NULL;
+	int ret;
+
+	if (new_crtc)
+		new_crtc_state = drm_atomic_get_new_crtc_state(state, new_crtc);
+
+	ret = drm_atomic_helper_check_plane_state(new_plane_state,
+						  new_crtc_state, DRM_PLANE_NO_SCALING,
+						  DRM_PLANE_NO_SCALING, false, false);
+	if (ret)
+		return ret;
+	else if (!new_plane_state->visible)
+		return 0;
+
+	return 0;
+}
+
+static int pixpaper_crtc_helper_atomic_check(struct drm_crtc *crtc,
+					     struct drm_atomic_state *state)
+{
+	struct drm_crtc_state *crtc_state =
+		drm_atomic_get_new_crtc_state(state, crtc);
+
+	if (!crtc_state->enable)
+		return 0;
+
+	return drm_atomic_helper_check_crtc_primary_plane(crtc_state);
+}
+
+static void pixpaper_crtc_atomic_enable(struct drm_crtc *crtc,
+					struct drm_atomic_state *state)
+{
+	struct pixpaper_panel *panel = to_pixpaper_panel(crtc->dev);
+	struct drm_device *drm = &panel->drm;
+	int idx;
+	struct pixpaper_error_ctx err = { .errno_code = 0 };
+
+	if (!drm_dev_enter(drm, &idx))
+		return;
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_POWER_ON, &err);
+	if (err.errno_code) {
+		drm_err_once(drm, "Failed to send PON command: %d\n", err.errno_code);
+		goto exit_drm_dev;
+	}
+
+	pixpaper_wait_for_panel(panel);
+
+	drm_dbg(drm, "Panel enabled and powered on\n");
+
+exit_drm_dev:
+	drm_dev_exit(idx);
+}
+
+static void pixpaper_crtc_atomic_disable(struct drm_crtc *crtc,
+					 struct drm_atomic_state *state)
+{
+	struct pixpaper_panel *panel = to_pixpaper_panel(crtc->dev);
+	struct drm_device *drm = &panel->drm;
+	struct pixpaper_error_ctx err = { .errno_code = 0 };
+	int idx;
+
+	if (!drm_dev_enter(drm, &idx))
+		return;
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_POWER_OFF, &err);
+	if (err.errno_code) {
+		drm_err_once(drm, "Failed to send POF command: %d\n", err.errno_code);
+		goto exit_drm_dev;
+	}
+	pixpaper_wait_for_panel(panel);
+
+	drm_dbg(drm, "Panel disabled\n");
+
+exit_drm_dev:
+	drm_dev_exit(idx);
+}
+
+static void pixpaper_plane_atomic_update(struct drm_plane *plane,
+					 struct drm_atomic_state *state)
+{
+	struct drm_plane_state *plane_state =
+		drm_atomic_get_new_plane_state(state, plane);
+	struct drm_shadow_plane_state *shadow_plane_state =
+		to_drm_shadow_plane_state(plane_state);
+	struct drm_crtc *crtc = plane_state->crtc;
+	struct pixpaper_panel *panel = to_pixpaper_panel(crtc->dev);
+
+	struct drm_device *drm = &panel->drm;
+	struct drm_framebuffer *fb = plane_state->fb;
+	struct iosys_map map = shadow_plane_state->data[0];
+	void *vaddr = map.vaddr;
+	int i, j, idx;
+	__le32 *src_pixels = NULL;
+	struct pixpaper_error_ctx err = { .errno_code = 0 };
+
+	if (!drm_dev_enter(drm, &idx))
+		return;
+
+	drm_dbg(drm, "Starting frame update (phys=%dx%d, buf_w=%d)\n",
+		PIXPAPER_WIDTH, PIXPAPER_HEIGHT, PIXPAPER_PANEL_BUFFER_WIDTH);
+
+	if (!fb || !plane_state->visible) {
+		drm_err_once(drm, "No framebuffer or plane not visible, skipping update\n");
+		goto update_cleanup;
+	}
+
+	src_pixels = (__le32 *)vaddr;
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_DATA_START_TRANSMISSION, &err);
+	if (err.errno_code)
+		goto update_cleanup;
+
+	pixpaper_wait_for_panel(panel);
+
+	for (i = 0; i < PIXPAPER_HEIGHT; i++) {
+		for (j = 0; j < PIXPAPER_PANEL_BUFFER_TWO_BYTES_PER_ROW; j++) {
+			u8 packed_byte =
+				pack_pixels_to_byte(src_pixels, i, j, fb);
+
+			pixpaper_wait_for_panel(panel);
+			pixpaper_send_data(panel, packed_byte, &err);
+		}
+	}
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_POWER_ON, &err);
+	if (err.errno_code) {
+		drm_err_once(drm, "Failed to send PON command: %d\n", err.errno_code);
+		goto update_cleanup;
+	}
+	pixpaper_wait_for_panel(panel);
+
+	pixpaper_send_cmd(panel, PIXPAPER_CMD_DISPLAY_REFRESH, &err);
+	pixpaper_send_data(panel, PIXPAPER_DRF_VCOM_AC, &err);
+	if (err.errno_code) {
+		drm_err_once(drm, "Failed sending data after DRF: %d\n", err.errno_code);
+		goto update_cleanup;
+	}
+	pixpaper_wait_for_panel(panel);
+
+update_cleanup:
+	if (err.errno_code && err.errno_code != -ETIMEDOUT)
+		drm_err_once(drm, "Frame update function failed with error %d\n", err.errno_code);
+
+	drm_dev_exit(idx);
+}
+
+static const struct drm_display_mode pixpaper_mode = {
+	.clock = PIXPAPER_PIXEL_CLOCK,
+	.hdisplay = PIXPAPER_WIDTH,
+	.hsync_start = PIXPAPER_WIDTH + PIXPAPER_HFP,
+	.hsync_end = PIXPAPER_WIDTH + PIXPAPER_HFP + PIXPAPER_HSYNC,
+	.htotal = PIXPAPER_HTOTAL,
+	.vdisplay = PIXPAPER_HEIGHT,
+	.vsync_start = PIXPAPER_HEIGHT + PIXPAPER_VFP,
+	.vsync_end = PIXPAPER_HEIGHT + PIXPAPER_VFP + PIXPAPER_VSYNC,
+	.vtotal = PIXPAPER_VTOTAL,
+	.width_mm = PIXPAPER_WIDTH_MM,
+	.height_mm = PIXPAPER_HEIGHT_MM,
+	.type = DRM_MODE_TYPE_DRIVER | DRM_MODE_TYPE_PREFERRED,
+};
+
+static int pixpaper_connector_get_modes(struct drm_connector *connector)
+{
+	return drm_connector_helper_get_modes_fixed(connector, &pixpaper_mode);
+}
+
+static const struct drm_plane_funcs pixpaper_plane_funcs = {
+	.update_plane = drm_atomic_helper_update_plane,
+	.disable_plane = drm_atomic_helper_disable_plane,
+	.destroy = drm_plane_cleanup,
+	DRM_GEM_SHADOW_PLANE_FUNCS,
+};
+
+static const struct drm_plane_helper_funcs pixpaper_plane_helper_funcs = {
+	DRM_GEM_SHADOW_PLANE_HELPER_FUNCS,
+	.atomic_check = pixpaper_plane_helper_atomic_check,
+	.atomic_update = pixpaper_plane_atomic_update,
+};
+
+static const struct drm_crtc_funcs pixpaper_crtc_funcs = {
+	.set_config = drm_atomic_helper_set_config,
+	.page_flip = drm_atomic_helper_page_flip,
+	.reset = drm_atomic_helper_crtc_reset,
+	.destroy = drm_crtc_cleanup,
+	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
+};
+
+static int pixpaper_mode_valid(struct drm_crtc *crtc,
+			       const struct drm_display_mode *mode)
+{
+	if (mode->hdisplay == PIXPAPER_WIDTH &&
+	    mode->vdisplay == PIXPAPER_HEIGHT) {
+		return MODE_OK;
+	}
+	return MODE_BAD;
+}
+
+static const struct drm_crtc_helper_funcs pixpaper_crtc_helper_funcs = {
+	.mode_valid = pixpaper_mode_valid,
+	.atomic_check = pixpaper_crtc_helper_atomic_check,
+	.atomic_enable = pixpaper_crtc_atomic_enable,
+	.atomic_disable = pixpaper_crtc_atomic_disable,
+};
+
+static const struct drm_encoder_funcs pixpaper_encoder_funcs = {
+	.destroy = drm_encoder_cleanup,
+};
+
+static const struct drm_connector_funcs pixpaper_connector_funcs = {
+	.reset = drm_atomic_helper_connector_reset,
+	.fill_modes = drm_helper_probe_single_connector_modes,
+	.destroy = drm_connector_cleanup,
+	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
+	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
+};
+
+static const struct drm_connector_helper_funcs pixpaper_connector_helper_funcs = {
+	.get_modes = pixpaper_connector_get_modes,
+};
+
+DEFINE_DRM_GEM_FOPS(pixpaper_fops);
+
+static struct drm_driver pixpaper_drm_driver = {
+	.driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC,
+	.fops = &pixpaper_fops,
+	.name = "pixpaper",
+	.desc = "DRM driver for PIXPAPER e-ink",
+	.major = 1,
+	.minor = 0,
+	DRM_GEM_SHMEM_DRIVER_OPS,
+	DRM_FBDEV_SHMEM_DRIVER_OPS,
+};
+
+static const struct drm_mode_config_funcs pixpaper_mode_config_funcs = {
+	.fb_create = drm_gem_fb_create_with_dirty,
+	.atomic_check = drm_atomic_helper_check,
+	.atomic_commit = drm_atomic_helper_commit,
+};
+
+static int pixpaper_probe(struct spi_device *spi)
+{
+	struct device *dev = &spi->dev;
+	struct pixpaper_panel *panel;
+	struct drm_device *drm;
+	int ret;
+
+	panel = devm_drm_dev_alloc(dev, &pixpaper_drm_driver,
+				   struct pixpaper_panel, drm);
+	if (IS_ERR(panel))
+		return PTR_ERR(panel);
+
+	drm = &panel->drm;
+	panel->spi = spi;
+	spi_set_drvdata(spi, panel);
+
+	spi->mode = SPI_MODE_0;
+	spi->bits_per_word = PIXPAPER_SPI_BITS_PER_WORD;
+
+	if (!spi->max_speed_hz) {
+		drm_warn(drm,
+			 "spi-max-frequency not specified in DT, using default %u Hz\n",
+			 PIXPAPER_SPI_SPEED_DEFAULT);
+		spi->max_speed_hz = PIXPAPER_SPI_SPEED_DEFAULT;
+	}
+
+	ret = spi_setup(spi);
+	if (ret < 0) {
+		drm_err(drm, "SPI setup failed: %d\n", ret);
+		return ret;
+	}
+
+	if (!dev->dma_mask)
+		dev->dma_mask = &dev->coherent_dma_mask;
+	ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));
+	if (ret) {
+		drm_err(drm, "Failed to set DMA mask: %d\n", ret);
+		return ret;
+	}
+
+	panel->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	if (IS_ERR(panel->reset))
+		return PTR_ERR(panel->reset);
+
+	panel->busy = devm_gpiod_get(dev, "busy", GPIOD_IN);
+	if (IS_ERR(panel->busy))
+		return PTR_ERR(panel->busy);
+
+	panel->dc = devm_gpiod_get(dev, "dc", GPIOD_OUT_HIGH);
+	if (IS_ERR(panel->dc))
+		return PTR_ERR(panel->dc);
+
+	ret = pixpaper_panel_hw_init(panel);
+	if (ret) {
+		drm_err(drm, "Panel hardware initialization failed: %d\n", ret);
+		return ret;
+	}
+
+	ret = drmm_mode_config_init(drm);
+	if (ret)
+		return ret;
+	drm->mode_config.funcs = &pixpaper_mode_config_funcs;
+	drm->mode_config.min_width = PIXPAPER_WIDTH;
+	drm->mode_config.max_width = PIXPAPER_WIDTH;
+	drm->mode_config.min_height = PIXPAPER_HEIGHT;
+	drm->mode_config.max_height = PIXPAPER_HEIGHT;
+
+	ret = drm_universal_plane_init(drm, &panel->plane, 1,
+				       &pixpaper_plane_funcs,
+				       (const uint32_t[]){ DRM_FORMAT_XRGB8888 },
+				       1, NULL, DRM_PLANE_TYPE_PRIMARY, NULL);
+	if (ret)
+		return ret;
+	drm_plane_helper_add(&panel->plane, &pixpaper_plane_helper_funcs);
+
+	ret = drm_crtc_init_with_planes(drm, &panel->crtc, &panel->plane, NULL,
+					&pixpaper_crtc_funcs, NULL);
+	if (ret)
+		return ret;
+	drm_crtc_helper_add(&panel->crtc, &pixpaper_crtc_helper_funcs);
+
+	ret = drm_encoder_init(drm, &panel->encoder, &pixpaper_encoder_funcs,
+			       DRM_MODE_ENCODER_NONE, NULL);
+	if (ret)
+		return ret;
+	panel->encoder.possible_crtcs = drm_crtc_mask(&panel->crtc);
+
+	ret = drm_connector_init(drm, &panel->connector,
+				 &pixpaper_connector_funcs,
+				 DRM_MODE_CONNECTOR_SPI);
+	if (ret)
+		return ret;
+
+	drm_connector_helper_add(&panel->connector,
+				 &pixpaper_connector_helper_funcs);
+	drm_connector_attach_encoder(&panel->connector, &panel->encoder);
+
+	drm_mode_config_reset(drm);
+
+	ret = drm_dev_register(drm, 0);
+	if (ret)
+		return ret;
+
+	drm_client_setup(drm, NULL);
+
+	return 0;
+}
+
+static void pixpaper_remove(struct spi_device *spi)
+{
+	struct pixpaper_panel *panel = spi_get_drvdata(spi);
+
+	if (!panel)
+		return;
+
+	drm_dev_unplug(&panel->drm);
+	drm_atomic_helper_shutdown(&panel->drm);
+}
+
+static const struct spi_device_id pixpaper_ids[] = { { "pixpaper", 0 }, {} };
+MODULE_DEVICE_TABLE(spi, pixpaper_ids);
+
+static const struct of_device_id pixpaper_dt_ids[] = {
+	{ .compatible = "mayqueen,pixpaper" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, pixpaper_dt_ids);
+
+static struct spi_driver pixpaper_spi_driver = {
+	.driver = {
+		.name = "pixpaper",
+		.of_match_table = pixpaper_dt_ids,
+	},
+	.id_table = pixpaper_ids,
+	.probe = pixpaper_probe,
+	.remove = pixpaper_remove,
+};
+
+module_spi_driver(pixpaper_spi_driver);
+
+MODULE_AUTHOR("LiangCheng Wang");
+MODULE_DESCRIPTION("DRM SPI driver for PIXPAPER e-ink panel");
+MODULE_LICENSE("GPL");

From 9e05c8dc4e8bb14bdb86eccff1d07169cfd69df8 Mon Sep 17 00:00:00 2001
From: Luca Ceresoli <luca.ceresoli@bootlin.com>
Date: Fri, 8 Aug 2025 16:49:09 +0200
Subject: [PATCH 0984/1455] drm/display: bridge-connector: remove unused
 variable assignment

The 'bridge' pointer started being assigned and used within this 'if' scope
in commit 0beba3f9d366 ("drm/bridge: connector: add support for HDMI codec
framework").

After that, commit 5d04b4188959 ("drm/bridge: split HDMI Audio from
DRM_BRIDGE_OP_HDMI") removed the code dereferencing it from the same 'if'
scope, but did not remove the assignment.

Reviewed-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250808-drm-bridge-alloc-getput-for_each_bridge-v2-2-edb6ee81edf1@bootlin.com
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
---
 drivers/gpu/drm/display/drm_bridge_connector.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/display/drm_bridge_connector.c b/drivers/gpu/drm/display/drm_bridge_connector.c
index 54f3f16d64c7..baacd21e7341 100644
--- a/drivers/gpu/drm/display/drm_bridge_connector.c
+++ b/drivers/gpu/drm/display/drm_bridge_connector.c
@@ -776,8 +776,6 @@ struct drm_connector *drm_bridge_connector_init(struct drm_device *drm,
 		if (!connector->ycbcr_420_allowed)
 			supported_formats &= ~BIT(HDMI_COLORSPACE_YUV420);
 
-		bridge = bridge_connector->bridge_hdmi;
-
 		ret = drmm_connector_hdmi_init(drm, connector,
 					       bridge_connector->bridge_hdmi->vendor,
 					       bridge_connector->bridge_hdmi->product,

From cb3d7b3b46b799c96b54f8e8fe36794a55a77f0b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 4 Sep 2025 18:07:13 +0200
Subject: [PATCH 0985/1455] drm/xe: Attempt to bring bos back to VRAM after
 eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

VRAM+TT bos that are evicted from VRAM to TT may remain in
TT also after a revalidation following eviction or suspend.

This manifests itself as applications becoming sluggish
after buffer objects get evicted or after a resume from
suspend or hibernation.

If the bo supports placement in both VRAM and TT, and
we are on DGFX, mark the TT placement as fallback. This means
that it is tried only after VRAM + eviction.

This flaw has probably been present since the xe module was
upstreamed but use a Fixes: commit below where backporting is
likely to be simple. For earlier versions we need to open-
code the fallback algorithm in the driver.

v2:
- Remove check for dgfx. (Matthew Auld)
- Update the xe_dma_buf kunit test for the new strategy (CI)
- Allow dma-buf to pin in current placement (CI)
- Make xe_bo_validate() for pinned bos a NOP.

Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/5995
Fixes: a78a8da51b36 ("drm/ttm: replace busy placement with flags v6")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: <stable@vger.kernel.org> # v6.9+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250904160715.2613-2-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/tests/xe_bo.c      |  2 +-
 drivers/gpu/drm/xe/tests/xe_dma_buf.c | 10 +---------
 drivers/gpu/drm/xe/xe_bo.c            | 16 ++++++++++++----
 drivers/gpu/drm/xe/xe_bo.h            |  2 +-
 drivers/gpu/drm/xe/xe_dma_buf.c       |  2 +-
 5 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c
index bb469096d072..7b40cc8be1c9 100644
--- a/drivers/gpu/drm/xe/tests/xe_bo.c
+++ b/drivers/gpu/drm/xe/tests/xe_bo.c
@@ -236,7 +236,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 		}
 
 		xe_bo_lock(external, false);
-		err = xe_bo_pin_external(external);
+		err = xe_bo_pin_external(external, false);
 		xe_bo_unlock(external);
 		if (err) {
 			KUNIT_FAIL(test, "external bo pin err=%pe\n",
diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
index 3c5ad8cc65a0..5baeab6b6fb7 100644
--- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
@@ -85,15 +85,7 @@ static void check_residency(struct kunit *test, struct xe_bo *exported,
 		return;
 	}
 
-	/*
-	 * If on different devices, the exporter is kept in system  if
-	 * possible, saving a migration step as the transfer is just
-	 * likely as fast from system memory.
-	 */
-	if (params->mem_mask & XE_BO_FLAG_SYSTEM)
-		KUNIT_EXPECT_TRUE(test, xe_bo_is_mem_type(exported, XE_PL_TT));
-	else
-		KUNIT_EXPECT_TRUE(test, xe_bo_is_mem_type(exported, mem_type));
+	KUNIT_EXPECT_TRUE(test, xe_bo_is_mem_type(exported, mem_type));
 
 	if (params->force_different_devices)
 		KUNIT_EXPECT_TRUE(test, xe_bo_is_mem_type(imported, XE_PL_TT));
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 49911247f6cb..f03f7432c6fa 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -188,6 +188,8 @@ static void try_add_system(struct xe_device *xe, struct xe_bo *bo,
 
 		bo->placements[*c] = (struct ttm_place) {
 			.mem_type = XE_PL_TT,
+			.flags = (bo_flags & XE_BO_FLAG_VRAM_MASK) ?
+			TTM_PL_FLAG_FALLBACK : 0,
 		};
 		*c += 1;
 	}
@@ -2311,6 +2313,7 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res)
 /**
  * xe_bo_pin_external - pin an external BO
  * @bo: buffer object to be pinned
+ * @in_place: Pin in current placement, don't attempt to migrate.
  *
  * Pin an external (not tied to a VM, can be exported via dma-buf / prime FD)
  * BO. Unique call compared to xe_bo_pin as this function has it own set of
@@ -2318,7 +2321,7 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res)
  *
  * Returns 0 for success, negative error code otherwise.
  */
-int xe_bo_pin_external(struct xe_bo *bo)
+int xe_bo_pin_external(struct xe_bo *bo, bool in_place)
 {
 	struct xe_device *xe = xe_bo_device(bo);
 	int err;
@@ -2327,9 +2330,11 @@ int xe_bo_pin_external(struct xe_bo *bo)
 	xe_assert(xe, xe_bo_is_user(bo));
 
 	if (!xe_bo_is_pinned(bo)) {
-		err = xe_bo_validate(bo, NULL, false);
-		if (err)
-			return err;
+		if (!in_place) {
+			err = xe_bo_validate(bo, NULL, false);
+			if (err)
+				return err;
+		}
 
 		spin_lock(&xe->pinned.lock);
 		list_add_tail(&bo->pinned_link, &xe->pinned.late.external);
@@ -2482,6 +2487,9 @@ int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict)
 	};
 	int ret;
 
+	if (xe_bo_is_pinned(bo))
+		return 0;
+
 	if (vm) {
 		lockdep_assert_held(&vm->lock);
 		xe_vm_assert_held(vm);
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 8cce413b5235..cfb1ec266a6d 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -200,7 +200,7 @@ static inline void xe_bo_unlock_vm_held(struct xe_bo *bo)
 	}
 }
 
-int xe_bo_pin_external(struct xe_bo *bo);
+int xe_bo_pin_external(struct xe_bo *bo, bool in_place);
 int xe_bo_pin(struct xe_bo *bo);
 void xe_bo_unpin_external(struct xe_bo *bo);
 void xe_bo_unpin(struct xe_bo *bo);
diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index 346f857f3837..af64baf872ef 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -72,7 +72,7 @@ static int xe_dma_buf_pin(struct dma_buf_attachment *attach)
 		return ret;
 	}
 
-	ret = xe_bo_pin_external(bo);
+	ret = xe_bo_pin_external(bo, true);
 	xe_assert(xe, !ret);
 
 	return 0;

From ebd546fdffddfcaeab08afdd68ec93052c8fa740 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 4 Sep 2025 18:07:14 +0200
Subject: [PATCH 0986/1455] drm/xe: Allow the pm notifier to continue on
 failure
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Its actions are opportunistic anyway and will be completed
on device suspend.

Marking as a fix to simplify backporting of the fix
that follows in the series.

v2:
- Keep the runtime pm reference over suspend / hibernate and
  document why. (Matt Auld, Rodrigo Vivi):

Fixes: c6a4d46ec1d7 ("drm/xe: evict user memory in PM notifier")
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: <stable@vger.kernel.org> # v6.16+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250904160715.2613-3-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_pm.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index d4b54cd34cff..ecfa7d77a2a3 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -312,17 +312,17 @@ static int xe_pm_notifier_callback(struct notifier_block *nb,
 	case PM_SUSPEND_PREPARE:
 		xe_pm_runtime_get(xe);
 		err = xe_bo_evict_all_user(xe);
-		if (err) {
+		if (err)
 			drm_dbg(&xe->drm, "Notifier evict user failed (%d)\n", err);
-			xe_pm_runtime_put(xe);
-			break;
-		}
 
 		err = xe_bo_notifier_prepare_all_pinned(xe);
-		if (err) {
+		if (err)
 			drm_dbg(&xe->drm, "Notifier prepare pin failed (%d)\n", err);
-			xe_pm_runtime_put(xe);
-		}
+		/*
+		 * Keep the runtime pm reference until post hibernation / post suspend to
+		 * avoid a runtime suspend interfering with evicted objects or backup
+		 * allocations.
+		 */
 		break;
 	case PM_POST_HIBERNATION:
 	case PM_POST_SUSPEND:
@@ -331,9 +331,6 @@ static int xe_pm_notifier_callback(struct notifier_block *nb,
 		break;
 	}
 
-	if (err)
-		return NOTIFY_BAD;
-
 	return NOTIFY_DONE;
 }
 

From 599334572a5a99111015fbbd5152ce4dedc2f8b7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 4 Sep 2025 18:07:15 +0200
Subject: [PATCH 0987/1455] drm/xe: Block exec and rebind worker while evicting
 for suspend / hibernate
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When the xe pm_notifier evicts for suspend / hibernate, there might be
racing tasks trying to re-validate again. This can lead to suspend taking
excessive time or get stuck in a live-lock. This behaviour becomes
much worse with the fix that actually makes re-validation bring back
bos to VRAM rather than letting them remain in TT.

Prevent that by having exec and the rebind worker waiting for a completion
that is set to block by the pm_notifier before suspend and is signaled
by the pm_notifier after resume / wakeup.

It's probably still possible to craft malicious applications that block
suspending. More work is pending to fix that.

v3:
- Avoid wait_for_completion() in the kernel worker since it could
  potentially cause work item flushes from freezable processes to
  wait forever. Instead terminate the rebind workers if needed and
  re-launch at resume. (Matt Auld)
v4:
- Fix some bad naming and leftover debug printouts.
- Fix kerneldoc.
- Use drmm_mutex_init() for the xe->rebind_resume_lock (Matt Auld).
- Rework the interface of xe_vm_rebind_resume_worker (Matt Auld).

Link: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/4288
Fixes: c6a4d46ec1d7 ("drm/xe: evict user memory in PM notifier")
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: <stable@vger.kernel.org> # v6.16+
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Link: https://lore.kernel.org/r/20250904160715.2613-4-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_device_types.h |  6 ++++
 drivers/gpu/drm/xe/xe_exec.c         |  9 ++++++
 drivers/gpu/drm/xe/xe_pm.c           | 25 +++++++++++++++++
 drivers/gpu/drm/xe/xe_vm.c           | 42 +++++++++++++++++++++++++++-
 drivers/gpu/drm/xe/xe_vm.h           |  2 ++
 drivers/gpu/drm/xe/xe_vm_types.h     |  5 ++++
 6 files changed, 88 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 646d04aec68c..9e30dc7d6e58 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -507,6 +507,12 @@ struct xe_device {
 
 	/** @pm_notifier: Our PM notifier to perform actions in response to various PM events. */
 	struct notifier_block pm_notifier;
+	/** @pm_block: Completion to block validating tasks on suspend / hibernate prepare */
+	struct completion pm_block;
+	/** @rebind_resume_list: List of wq items to kick on resume. */
+	struct list_head rebind_resume_list;
+	/** @rebind_resume_lock: Lock to protect the rebind_resume_list */
+	struct mutex rebind_resume_lock;
 
 	/** @pmt: Support the PMT driver callback interface */
 	struct {
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index 25a59b6934f6..1dc0c54cac5f 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -238,6 +238,15 @@ retry:
 		goto err_unlock_list;
 	}
 
+	/*
+	 * It's OK to block interruptible here with the vm lock held, since
+	 * on task freezing during suspend / hibernate, the call will
+	 * return -ERESTARTSYS and the IOCTL will be rerun.
+	 */
+	err = wait_for_completion_interruptible(&xe->pm_block);
+	if (err)
+		goto err_unlock_list;
+
 	vm_exec.vm = &vm->gpuvm;
 	vm_exec.flags = DRM_EXEC_INTERRUPTIBLE_WAIT;
 	if (xe_vm_in_lr_mode(vm)) {
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index ecfa7d77a2a3..6eea4190bbd2 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -25,6 +25,7 @@
 #include "xe_pxp.h"
 #include "xe_sriov_vf_ccs.h"
 #include "xe_trace.h"
+#include "xe_vm.h"
 #include "xe_wa.h"
 
 /**
@@ -301,6 +302,19 @@ static u32 vram_threshold_value(struct xe_device *xe)
 	return DEFAULT_VRAM_THRESHOLD;
 }
 
+static void xe_pm_wake_rebind_workers(struct xe_device *xe)
+{
+	struct xe_vm *vm, *next;
+
+	mutex_lock(&xe->rebind_resume_lock);
+	list_for_each_entry_safe(vm, next, &xe->rebind_resume_list,
+				 preempt.pm_activate_link) {
+		list_del_init(&vm->preempt.pm_activate_link);
+		xe_vm_resume_rebind_worker(vm);
+	}
+	mutex_unlock(&xe->rebind_resume_lock);
+}
+
 static int xe_pm_notifier_callback(struct notifier_block *nb,
 				   unsigned long action, void *data)
 {
@@ -310,6 +324,7 @@ static int xe_pm_notifier_callback(struct notifier_block *nb,
 	switch (action) {
 	case PM_HIBERNATION_PREPARE:
 	case PM_SUSPEND_PREPARE:
+		reinit_completion(&xe->pm_block);
 		xe_pm_runtime_get(xe);
 		err = xe_bo_evict_all_user(xe);
 		if (err)
@@ -326,6 +341,8 @@ static int xe_pm_notifier_callback(struct notifier_block *nb,
 		break;
 	case PM_POST_HIBERNATION:
 	case PM_POST_SUSPEND:
+		complete_all(&xe->pm_block);
+		xe_pm_wake_rebind_workers(xe);
 		xe_bo_notifier_unprepare_all_pinned(xe);
 		xe_pm_runtime_put(xe);
 		break;
@@ -352,6 +369,14 @@ int xe_pm_init(struct xe_device *xe)
 	if (err)
 		return err;
 
+	err = drmm_mutex_init(&xe->drm, &xe->rebind_resume_lock);
+	if (err)
+		goto err_unregister;
+
+	init_completion(&xe->pm_block);
+	complete_all(&xe->pm_block);
+	INIT_LIST_HEAD(&xe->rebind_resume_list);
+
 	/* For now suspend/resume is only allowed with GuC */
 	if (!xe_device_uc_enabled(xe))
 		return 0;
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 869f1db87d74..346975bc1065 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -343,6 +343,9 @@ static int xe_gpuvm_validate(struct drm_gpuvm_bo *vm_bo, struct drm_exec *exec)
 		list_move_tail(&gpuva_to_vma(gpuva)->combined_links.rebind,
 			       &vm->rebind_list);
 
+	if (!try_wait_for_completion(&vm->xe->pm_block))
+		return -EAGAIN;
+
 	ret = xe_bo_validate(gem_to_xe_bo(vm_bo->obj), vm, false);
 	if (ret)
 		return ret;
@@ -429,6 +432,33 @@ static int xe_preempt_work_begin(struct drm_exec *exec, struct xe_vm *vm,
 	return xe_vm_validate_rebind(vm, exec, vm->preempt.num_exec_queues);
 }
 
+static bool vm_suspend_rebind_worker(struct xe_vm *vm)
+{
+	struct xe_device *xe = vm->xe;
+	bool ret = false;
+
+	mutex_lock(&xe->rebind_resume_lock);
+	if (!try_wait_for_completion(&vm->xe->pm_block)) {
+		ret = true;
+		list_move_tail(&vm->preempt.pm_activate_link, &xe->rebind_resume_list);
+	}
+	mutex_unlock(&xe->rebind_resume_lock);
+
+	return ret;
+}
+
+/**
+ * xe_vm_resume_rebind_worker() - Resume the rebind worker.
+ * @vm: The vm whose preempt worker to resume.
+ *
+ * Resume a preempt worker that was previously suspended by
+ * vm_suspend_rebind_worker().
+ */
+void xe_vm_resume_rebind_worker(struct xe_vm *vm)
+{
+	queue_work(vm->xe->ordered_wq, &vm->preempt.rebind_work);
+}
+
 static void preempt_rebind_work_func(struct work_struct *w)
 {
 	struct xe_vm *vm = container_of(w, struct xe_vm, preempt.rebind_work);
@@ -452,6 +482,11 @@ static void preempt_rebind_work_func(struct work_struct *w)
 	}
 
 retry:
+	if (!try_wait_for_completion(&vm->xe->pm_block) && vm_suspend_rebind_worker(vm)) {
+		up_write(&vm->lock);
+		return;
+	}
+
 	if (xe_vm_userptr_check_repin(vm)) {
 		err = xe_vm_userptr_pin(vm);
 		if (err)
@@ -1470,6 +1505,7 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 	if (flags & XE_VM_FLAG_LR_MODE) {
 		INIT_WORK(&vm->preempt.rebind_work, preempt_rebind_work_func);
 		xe_pm_runtime_get_noresume(xe);
+		INIT_LIST_HEAD(&vm->preempt.pm_activate_link);
 	}
 
 	err = xe_svm_init(vm);
@@ -1649,8 +1685,12 @@ void xe_vm_close_and_put(struct xe_vm *vm)
 	xe_assert(xe, !vm->preempt.num_exec_queues);
 
 	xe_vm_close(vm);
-	if (xe_vm_in_preempt_fence_mode(vm))
+	if (xe_vm_in_preempt_fence_mode(vm)) {
+		mutex_lock(&xe->rebind_resume_lock);
+		list_del_init(&vm->preempt.pm_activate_link);
+		mutex_unlock(&xe->rebind_resume_lock);
 		flush_work(&vm->preempt.rebind_work);
+	}
 	if (xe_vm_in_fault_mode(vm))
 		xe_svm_close(vm);
 
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index bec9bf89d937..f29a4baee05d 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -271,6 +271,8 @@ struct dma_fence *xe_vm_bind_kernel_bo(struct xe_vm *vm, struct xe_bo *bo,
 				       struct xe_exec_queue *q, u64 addr,
 				       enum xe_cache_level cache_lvl);
 
+void xe_vm_resume_rebind_worker(struct xe_vm *vm);
+
 /**
  * xe_vm_resv() - Return's the vm's reservation object
  * @vm: The vm
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index f6dd964be42f..8e188b83ef30 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -282,6 +282,11 @@ struct xe_vm {
 		 * BOs
 		 */
 		struct work_struct rebind_work;
+		/**
+		 * @preempt.pm_activate_link: Link to list of rebind workers to be
+		 * kicked on resume.
+		 */
+		struct list_head pm_activate_link;
 	} preempt;
 
 	/** @um: unified memory state */

From 0d2902dfa31535c08d4976995d5b6890a8f43399 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jos=C3=A9=20Exp=C3=B3sito?= <jose.exposito89@gmail.com>
Date: Mon, 11 Aug 2025 12:15:17 +0200
Subject: [PATCH 0988/1455] drm/vkms: Assert if vkms_config_create_*() fails
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Check that the value returned by the vkms_config_create_*() functions is
valid. Otherwise, assert and finish the KUnit test.

Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/dri-devel/aJTL6IFEBaI8gqtH@stanley.mountain/
Signed-off-by: José Expósito <jose.exposito89@gmail.com>
Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20250811101529.150716-1-jose.exposito89@gmail.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/tests/vkms_config_test.c | 51 +++++++++++++++++--
 1 file changed, 46 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/vkms/tests/vkms_config_test.c b/drivers/gpu/drm/vkms/tests/vkms_config_test.c
index ff4566cf9925..b0d78a81d2df 100644
--- a/drivers/gpu/drm/vkms/tests/vkms_config_test.c
+++ b/drivers/gpu/drm/vkms/tests/vkms_config_test.c
@@ -200,6 +200,7 @@ static void vkms_config_test_get_planes(struct kunit *test)
 	KUNIT_ASSERT_EQ(test, n_planes, 0);
 
 	plane_cfg1 = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg1);
 	vkms_config_for_each_plane(config, plane_cfg) {
 		n_planes++;
 		if (plane_cfg != plane_cfg1)
@@ -209,6 +210,7 @@ static void vkms_config_test_get_planes(struct kunit *test)
 	n_planes = 0;
 
 	plane_cfg2 = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg2);
 	vkms_config_for_each_plane(config, plane_cfg) {
 		n_planes++;
 		if (plane_cfg != plane_cfg1 && plane_cfg != plane_cfg2)
@@ -242,6 +244,7 @@ static void vkms_config_test_get_crtcs(struct kunit *test)
 		KUNIT_FAIL(test, "Unexpected CRTC");
 
 	crtc_cfg1 = vkms_config_create_crtc(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg1);
 	KUNIT_ASSERT_EQ(test, vkms_config_get_num_crtcs(config), 1);
 	vkms_config_for_each_crtc(config, crtc_cfg) {
 		if (crtc_cfg != crtc_cfg1)
@@ -249,6 +252,7 @@ static void vkms_config_test_get_crtcs(struct kunit *test)
 	}
 
 	crtc_cfg2 = vkms_config_create_crtc(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg2);
 	KUNIT_ASSERT_EQ(test, vkms_config_get_num_crtcs(config), 2);
 	vkms_config_for_each_crtc(config, crtc_cfg) {
 		if (crtc_cfg != crtc_cfg1 && crtc_cfg != crtc_cfg2)
@@ -280,6 +284,7 @@ static void vkms_config_test_get_encoders(struct kunit *test)
 	KUNIT_ASSERT_EQ(test, n_encoders, 0);
 
 	encoder_cfg1 = vkms_config_create_encoder(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg1);
 	vkms_config_for_each_encoder(config, encoder_cfg) {
 		n_encoders++;
 		if (encoder_cfg != encoder_cfg1)
@@ -289,6 +294,7 @@ static void vkms_config_test_get_encoders(struct kunit *test)
 	n_encoders = 0;
 
 	encoder_cfg2 = vkms_config_create_encoder(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg2);
 	vkms_config_for_each_encoder(config, encoder_cfg) {
 		n_encoders++;
 		if (encoder_cfg != encoder_cfg1 && encoder_cfg != encoder_cfg2)
@@ -324,6 +330,7 @@ static void vkms_config_test_get_connectors(struct kunit *test)
 	KUNIT_ASSERT_EQ(test, n_connectors, 0);
 
 	connector_cfg1 = vkms_config_create_connector(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, connector_cfg1);
 	vkms_config_for_each_connector(config, connector_cfg) {
 		n_connectors++;
 		if (connector_cfg != connector_cfg1)
@@ -333,6 +340,7 @@ static void vkms_config_test_get_connectors(struct kunit *test)
 	n_connectors = 0;
 
 	connector_cfg2 = vkms_config_create_connector(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, connector_cfg2);
 	vkms_config_for_each_connector(config, connector_cfg) {
 		n_connectors++;
 		if (connector_cfg != connector_cfg1 &&
@@ -370,7 +378,7 @@ static void vkms_config_test_invalid_plane_number(struct kunit *test)
 
 	/* Invalid: Too many planes */
 	for (n = 0; n <= 32; n++)
-		vkms_config_create_plane(config);
+		KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vkms_config_create_plane(config));
 
 	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
 
@@ -395,6 +403,7 @@ static void vkms_config_test_valid_plane_type(struct kunit *test)
 
 	/* Invalid: No primary plane */
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_OVERLAY);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
@@ -402,11 +411,13 @@ static void vkms_config_test_valid_plane_type(struct kunit *test)
 
 	/* Invalid: Multiple primary planes */
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
 
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
@@ -419,11 +430,13 @@ static void vkms_config_test_valid_plane_type(struct kunit *test)
 
 	/* Invalid: Multiple cursor planes */
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_CURSOR);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
 
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_CURSOR);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
@@ -437,12 +450,16 @@ static void vkms_config_test_valid_plane_type(struct kunit *test)
 	/* Invalid: Second CRTC without primary plane */
 	crtc_cfg = vkms_config_create_crtc(config);
 	encoder_cfg = vkms_config_create_encoder(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg);
+
 	err = vkms_config_encoder_attach_crtc(encoder_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
 	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
 
 	/* Valid: Second CRTC with a primary plane */
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg);
 	KUNIT_EXPECT_EQ(test, err, 0);
@@ -486,7 +503,7 @@ static void vkms_config_test_invalid_crtc_number(struct kunit *test)
 
 	/* Invalid: Too many CRTCs */
 	for (n = 0; n <= 32; n++)
-		vkms_config_create_crtc(config);
+		KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vkms_config_create_crtc(config));
 
 	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
 
@@ -509,7 +526,7 @@ static void vkms_config_test_invalid_encoder_number(struct kunit *test)
 
 	/* Invalid: Too many encoders */
 	for (n = 0; n <= 32; n++)
-		vkms_config_create_encoder(config);
+		KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vkms_config_create_encoder(config));
 
 	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
 
@@ -531,12 +548,15 @@ static void vkms_config_test_valid_encoder_possible_crtcs(struct kunit *test)
 
 	/* Invalid: Encoder without a possible CRTC */
 	encoder_cfg = vkms_config_create_encoder(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg);
 	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
 
 	/* Valid: Second CRTC with shared encoder */
 	crtc_cfg2 = vkms_config_create_crtc(config);
-
 	plane_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg);
+
 	vkms_config_plane_set_type(plane_cfg, DRM_PLANE_TYPE_PRIMARY);
 	err = vkms_config_plane_attach_crtc(plane_cfg, crtc_cfg2);
 	KUNIT_EXPECT_EQ(test, err, 0);
@@ -577,7 +597,7 @@ static void vkms_config_test_invalid_connector_number(struct kunit *test)
 
 	/* Invalid: Too many connectors */
 	for (n = 0; n <= 32; n++)
-		connector_cfg = vkms_config_create_connector(config);
+		KUNIT_ASSERT_NOT_ERR_OR_NULL(test, vkms_config_create_connector(config));
 
 	KUNIT_EXPECT_FALSE(test, vkms_config_is_valid(config));
 
@@ -669,13 +689,19 @@ static void vkms_config_test_plane_attach_crtc(struct kunit *test)
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, config);
 
 	overlay_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, overlay_cfg);
 	vkms_config_plane_set_type(overlay_cfg, DRM_PLANE_TYPE_OVERLAY);
+
 	primary_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, primary_cfg);
 	vkms_config_plane_set_type(primary_cfg, DRM_PLANE_TYPE_PRIMARY);
+
 	cursor_cfg = vkms_config_create_plane(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, cursor_cfg);
 	vkms_config_plane_set_type(cursor_cfg, DRM_PLANE_TYPE_CURSOR);
 
 	crtc_cfg = vkms_config_create_crtc(config);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg);
 
 	/* No primary or cursor planes */
 	KUNIT_EXPECT_NULL(test, vkms_config_crtc_primary_plane(config, crtc_cfg));
@@ -735,6 +761,11 @@ static void vkms_config_test_plane_get_possible_crtcs(struct kunit *test)
 	crtc_cfg1 = vkms_config_create_crtc(config);
 	crtc_cfg2 = vkms_config_create_crtc(config);
 
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, plane_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg2);
+
 	/* No possible CRTCs */
 	vkms_config_plane_for_each_possible_crtc(plane_cfg1, idx, possible_crtc)
 		KUNIT_FAIL(test, "Unexpected possible CRTC");
@@ -799,6 +830,11 @@ static void vkms_config_test_encoder_get_possible_crtcs(struct kunit *test)
 	crtc_cfg1 = vkms_config_create_crtc(config);
 	crtc_cfg2 = vkms_config_create_crtc(config);
 
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, crtc_cfg2);
+
 	/* No possible CRTCs */
 	vkms_config_encoder_for_each_possible_crtc(encoder_cfg1, idx, possible_crtc)
 		KUNIT_FAIL(test, "Unexpected possible CRTC");
@@ -863,6 +899,11 @@ static void vkms_config_test_connector_get_possible_encoders(struct kunit *test)
 	encoder_cfg1 = vkms_config_create_encoder(config);
 	encoder_cfg2 = vkms_config_create_encoder(config);
 
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, connector_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, connector_cfg2);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg1);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, encoder_cfg2);
+
 	/* No possible encoders */
 	vkms_config_connector_for_each_possible_encoder(connector_cfg1, idx,
 							possible_encoder)

From 544f161a88af0370ced87783246418aed63b619c Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:56:57 +0200
Subject: [PATCH 0989/1455] drm/vkms: Create helpers macro to avoid code
 duplication in format callbacks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The callback functions for line conversion are almost identical for
some format. The generic READ_LINE macro generate all the required
boilerplate to process a line.

Two overrides of this macro have been added to avoid duplication of
the same arguments every time.

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-1-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 192 ++++++++++------------------
 1 file changed, 65 insertions(+), 127 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index 6d0227c6635a..07b13481378b 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -292,6 +292,64 @@ VISIBLE_IF_KUNIT struct pixel_argb_u16 argb_u16_from_yuv888(u8 y, u8 channel_1,
 }
 EXPORT_SYMBOL_IF_KUNIT(argb_u16_from_yuv888);
 
+/**
+ * READ_LINE() - Generic generator for a read_line function which can be used for format with one
+ * plane and a block_h == block_w == 1.
+ *
+ * @function_name: Function name to generate
+ * @pixel_name: Temporary pixel name used in the @__VA_ARGS__ parameters
+ * @pixel_type: Used to specify the type you want to cast the pixel pointer
+ * @callback: Callback to call for each pixels. This fonction should take @__VA_ARGS__ as parameter
+ *            and return a pixel_argb_u16
+ * __VA_ARGS__: Argument to pass inside the callback. You can use @pixel_name to access current
+ *  pixel.
+ */
+#define READ_LINE(function_name, pixel_name, pixel_type, callback, ...)				\
+static void function_name(const struct vkms_plane_state *plane, int x_start,			\
+			      int y_start, enum pixel_read_direction direction, int count,	\
+			      struct pixel_argb_u16 out_pixel[])				\
+{												\
+	struct pixel_argb_u16 *end = out_pixel + count;						\
+	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);			\
+	u8 *src_pixels;										\
+												\
+	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);		\
+												\
+	while (out_pixel < end) {								\
+		pixel_type *(pixel_name) = (pixel_type *)src_pixels;				\
+		*out_pixel = (callback)(__VA_ARGS__);						\
+		out_pixel += 1;									\
+		src_pixels += step;								\
+	}											\
+}
+
+/**
+ * READ_LINE_ARGB8888() - Generic generator for ARGB8888 formats.
+ * The pixel type used is u8, so pixel_name[0]..pixel_name[n] are the n components of the pixel.
+ *
+ * @function_name: Function name to generate
+ * @pixel_name: temporary pixel to use in @a, @r, @g and @b parameters
+ * @a: alpha value
+ * @r: red value
+ * @g: green value
+ * @b: blue value
+ */
+#define READ_LINE_ARGB8888(function_name, pixel_name, a, r, g, b) \
+	READ_LINE(function_name, pixel_name, u8, argb_u16_from_u8888, a, r, g, b)
+/**
+ * READ_LINE_le16161616() - Generic generator for ARGB16161616 formats.
+ * The pixel type used is u16, so pixel_name[0]..pixel_name[n] are the n components of the pixel.
+ *
+ * @function_name: Function name to generate
+ * @pixel_name: temporary pixel to use in @a, @r, @g and @b parameters
+ * @a: alpha value
+ * @r: red value
+ * @g: green value
+ * @b: blue value
+ */
+#define READ_LINE_le16161616(function_name, pixel_name, a, r, g, b) \
+	READ_LINE(function_name, pixel_name, __le16, argb_u16_from_le16161616, a, r, g, b)
+
 /*
  * The following functions are read_line function for each pixel format supported by VKMS.
  *
@@ -378,138 +436,18 @@ static void R4_read_line(const struct vkms_plane_state *plane, int x_start,
 	Rx_read_line(plane, x_start, y_start, direction, count, out_pixel);
 }
 
-static void R8_read_line(const struct vkms_plane_state *plane, int x_start,
-			 int y_start, enum pixel_read_direction direction, int count,
-			 struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
 
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
+READ_LINE_ARGB8888(XRGB8888_read_line, px, 0xFF, px[2], px[1], px[0])
 
-	while (out_pixel < end) {
-		*out_pixel = argb_u16_from_gray8(*src_pixels);
-		src_pixels += step;
-		out_pixel += 1;
-	}
-}
+READ_LINE_ARGB8888(ARGB8888_read_line, px, px[3], px[2], px[1], px[0])
+READ_LINE_ARGB8888(ABGR8888_read_line, px, px[3], px[0], px[1], px[2])
 
-static void ARGB8888_read_line(const struct vkms_plane_state *plane, int x_start, int y_start,
-			       enum pixel_read_direction direction, int count,
-			       struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
+READ_LINE_le16161616(ARGB16161616_read_line, px, px[3], px[2], px[1], px[0])
+READ_LINE_le16161616(XRGB16161616_read_line, px, cpu_to_le16(0xFFFF), px[2], px[1], px[0])
 
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
+READ_LINE(RGB565_read_line, px, __le16, argb_u16_from_RGB565, px)
 
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-
-	while (out_pixel < end) {
-		u8 *px = (u8 *)src_pixels;
-		*out_pixel = argb_u16_from_u8888(px[3], px[2], px[1], px[0]);
-		out_pixel += 1;
-		src_pixels += step;
-	}
-}
-
-static void XRGB8888_read_line(const struct vkms_plane_state *plane, int x_start, int y_start,
-			       enum pixel_read_direction direction, int count,
-			       struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
-
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
-
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-
-	while (out_pixel < end) {
-		u8 *px = (u8 *)src_pixels;
-		*out_pixel = argb_u16_from_u8888(255, px[2], px[1], px[0]);
-		out_pixel += 1;
-		src_pixels += step;
-	}
-}
-
-static void ABGR8888_read_line(const struct vkms_plane_state *plane, int x_start, int y_start,
-			       enum pixel_read_direction direction, int count,
-			       struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
-
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
-
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-
-	while (out_pixel < end) {
-		u8 *px = (u8 *)src_pixels;
-		/* Switch blue and red pixels. */
-		*out_pixel = argb_u16_from_u8888(px[3], px[0], px[1], px[2]);
-		out_pixel += 1;
-		src_pixels += step;
-	}
-}
-
-static void ARGB16161616_read_line(const struct vkms_plane_state *plane, int x_start,
-				   int y_start, enum pixel_read_direction direction, int count,
-				   struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
-
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
-
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-
-	while (out_pixel < end) {
-		u16 *px = (u16 *)src_pixels;
-		*out_pixel = argb_u16_from_u16161616(px[3], px[2], px[1], px[0]);
-		out_pixel += 1;
-		src_pixels += step;
-	}
-}
-
-static void XRGB16161616_read_line(const struct vkms_plane_state *plane, int x_start,
-				   int y_start, enum pixel_read_direction direction, int count,
-				   struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
-
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
-
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-
-	while (out_pixel < end) {
-		__le16 *px = (__le16 *)src_pixels;
-		*out_pixel = argb_u16_from_le16161616(cpu_to_le16(0xFFFF), px[2], px[1], px[0]);
-		out_pixel += 1;
-		src_pixels += step;
-	}
-}
-
-static void RGB565_read_line(const struct vkms_plane_state *plane, int x_start,
-			     int y_start, enum pixel_read_direction direction, int count,
-			     struct pixel_argb_u16 out_pixel[])
-{
-	struct pixel_argb_u16 *end = out_pixel + count;
-	u8 *src_pixels;
-
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0, &src_pixels);
-
-	int step = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-
-	while (out_pixel < end) {
-		__le16 *px = (__le16 *)src_pixels;
-
-		*out_pixel = argb_u16_from_RGB565(px);
-		out_pixel += 1;
-		src_pixels += step;
-	}
-}
+READ_LINE(R8_read_line, px, u8, argb_u16_from_gray8, *px)
 
 /*
  * This callback can be used for YUV formats where U and V values are

From 985769b8b5c42666be91f5e60aeb93365c934b88 Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:56:58 +0200
Subject: [PATCH 0990/1455] drm/vkms: Add support for ARGB8888 formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The formats XRGB8888 and ARGB8888 were already supported. Add the
support for:
- XBGR8888
- ABGR8888
- RGBA8888
- BGRA8888

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-2-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 13 +++++++++++--
 drivers/gpu/drm/vkms/vkms_plane.c   |  5 ++++-
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index 07b13481378b..b11cb2955943 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -438,9 +438,12 @@ static void R4_read_line(const struct vkms_plane_state *plane, int x_start,
 
 
 READ_LINE_ARGB8888(XRGB8888_read_line, px, 0xFF, px[2], px[1], px[0])
+READ_LINE_ARGB8888(XBGR8888_read_line, px, 0xFF, px[0], px[1], px[2])
 
 READ_LINE_ARGB8888(ARGB8888_read_line, px, px[3], px[2], px[1], px[0])
 READ_LINE_ARGB8888(ABGR8888_read_line, px, px[3], px[0], px[1], px[2])
+READ_LINE_ARGB8888(RGBA8888_read_line, px, px[0], px[3], px[2], px[1])
+READ_LINE_ARGB8888(BGRA8888_read_line, px, px[0], px[1], px[2], px[3])
 
 READ_LINE_le16161616(ARGB16161616_read_line, px, px[3], px[2], px[1], px[0])
 READ_LINE_le16161616(XRGB16161616_read_line, px, cpu_to_le16(0xFFFF), px[2], px[1], px[0])
@@ -650,10 +653,16 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 	switch (format) {
 	case DRM_FORMAT_ARGB8888:
 		return &ARGB8888_read_line;
-	case DRM_FORMAT_XRGB8888:
-		return &XRGB8888_read_line;
 	case DRM_FORMAT_ABGR8888:
 		return &ABGR8888_read_line;
+	case DRM_FORMAT_BGRA8888:
+		return &BGRA8888_read_line;
+	case DRM_FORMAT_RGBA8888:
+		return &RGBA8888_read_line;
+	case DRM_FORMAT_XRGB8888:
+		return &XRGB8888_read_line;
+	case DRM_FORMAT_XBGR8888:
+		return &XBGR8888_read_line;
 	case DRM_FORMAT_ARGB16161616:
 		return &ARGB16161616_read_line;
 	case DRM_FORMAT_XRGB16161616:
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index e3fdd161d0f0..01fed722808b 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -14,8 +14,11 @@
 
 static const u32 vkms_formats[] = {
 	DRM_FORMAT_ARGB8888,
-	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_ABGR8888,
+	DRM_FORMAT_BGRA8888,
+	DRM_FORMAT_RGBA8888,
+	DRM_FORMAT_XRGB8888,
+	DRM_FORMAT_XBGR8888,
 	DRM_FORMAT_XRGB16161616,
 	DRM_FORMAT_ARGB16161616,
 	DRM_FORMAT_RGB565,

From bac5c590d42ce8a930950da716b7537e60153196 Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:56:59 +0200
Subject: [PATCH 0991/1455] drm/vkms: Add support for ARGB16161616 formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The formats XRGB16161616 and ARGB16161616 were already supported.
Add the support for:
- ABGR16161616
- XBGR16161616

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-3-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 6 ++++++
 drivers/gpu/drm/vkms/vkms_plane.c   | 2 ++
 2 files changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index b11cb2955943..c37d1aae6eac 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -446,7 +446,9 @@ READ_LINE_ARGB8888(RGBA8888_read_line, px, px[0], px[3], px[2], px[1])
 READ_LINE_ARGB8888(BGRA8888_read_line, px, px[0], px[1], px[2], px[3])
 
 READ_LINE_le16161616(ARGB16161616_read_line, px, px[3], px[2], px[1], px[0])
+READ_LINE_le16161616(ABGR16161616_read_line, px, px[3], px[0], px[1], px[2])
 READ_LINE_le16161616(XRGB16161616_read_line, px, cpu_to_le16(0xFFFF), px[2], px[1], px[0])
+READ_LINE_le16161616(XBGR16161616_read_line, px, cpu_to_le16(0xFFFF), px[0], px[1], px[2])
 
 READ_LINE(RGB565_read_line, px, __le16, argb_u16_from_RGB565, px)
 
@@ -665,8 +667,12 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 		return &XBGR8888_read_line;
 	case DRM_FORMAT_ARGB16161616:
 		return &ARGB16161616_read_line;
+	case DRM_FORMAT_ABGR16161616:
+		return &ABGR16161616_read_line;
 	case DRM_FORMAT_XRGB16161616:
 		return &XRGB16161616_read_line;
+	case DRM_FORMAT_XBGR16161616:
+		return &XBGR16161616_read_line;
 	case DRM_FORMAT_RGB565:
 		return &RGB565_read_line;
 	case DRM_FORMAT_NV12:
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 01fed722808b..d4c51b4486f7 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -20,7 +20,9 @@ static const u32 vkms_formats[] = {
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_XBGR8888,
 	DRM_FORMAT_XRGB16161616,
+	DRM_FORMAT_XBGR16161616,
 	DRM_FORMAT_ARGB16161616,
+	DRM_FORMAT_ABGR16161616,
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_NV12,
 	DRM_FORMAT_NV16,

From 39d80cee4207d937daf58179ae3c572bcf541f8a Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:57:00 +0200
Subject: [PATCH 0992/1455] drm/vkms: Add support for RGB565 formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The format RGB565 was already supported. Add the support for:
- BGR565

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-4-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 13 +++++++++++++
 drivers/gpu/drm/vkms/vkms_plane.c   |  1 +
 2 files changed, 14 insertions(+)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index c37d1aae6eac..544bb6795805 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -259,6 +259,16 @@ static struct pixel_argb_u16 argb_u16_from_grayu16(u16 gray)
 	return argb_u16_from_u16161616(0xFFFF, gray, gray, gray);
 }
 
+static struct pixel_argb_u16 argb_u16_from_BGR565(const __le16 *pixel)
+{
+	struct pixel_argb_u16 out_pixel;
+
+	out_pixel = argb_u16_from_RGB565(pixel);
+	swap(out_pixel.r, out_pixel.b);
+
+	return out_pixel;
+}
+
 VISIBLE_IF_KUNIT struct pixel_argb_u16 argb_u16_from_yuv888(u8 y, u8 channel_1, u8 channel_2,
 							    const struct conversion_matrix *matrix)
 {
@@ -451,6 +461,7 @@ READ_LINE_le16161616(XRGB16161616_read_line, px, cpu_to_le16(0xFFFF), px[2], px[
 READ_LINE_le16161616(XBGR16161616_read_line, px, cpu_to_le16(0xFFFF), px[0], px[1], px[2])
 
 READ_LINE(RGB565_read_line, px, __le16, argb_u16_from_RGB565, px)
+READ_LINE(BGR565_read_line, px, __le16, argb_u16_from_BGR565, px)
 
 READ_LINE(R8_read_line, px, u8, argb_u16_from_gray8, *px)
 
@@ -675,6 +686,8 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 		return &XBGR16161616_read_line;
 	case DRM_FORMAT_RGB565:
 		return &RGB565_read_line;
+	case DRM_FORMAT_BGR565:
+		return &BGR565_read_line;
 	case DRM_FORMAT_NV12:
 	case DRM_FORMAT_NV16:
 	case DRM_FORMAT_NV24:
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index d4c51b4486f7..6e7597ab935d 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -24,6 +24,7 @@ static const u32 vkms_formats[] = {
 	DRM_FORMAT_ARGB16161616,
 	DRM_FORMAT_ABGR16161616,
 	DRM_FORMAT_RGB565,
+	DRM_FORMAT_BGR565,
 	DRM_FORMAT_NV12,
 	DRM_FORMAT_NV16,
 	DRM_FORMAT_NV24,

From 50c58f4f0562504256477ad153463d86500aa337 Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:57:01 +0200
Subject: [PATCH 0993/1455] drm/vkms: Add support for RGB888 formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add the support for:
- RGB888
- BGR888

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-5-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 7 +++++++
 drivers/gpu/drm/vkms/vkms_plane.c   | 2 ++
 2 files changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index 544bb6795805..f1649d9bf430 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -455,6 +455,9 @@ READ_LINE_ARGB8888(ABGR8888_read_line, px, px[3], px[0], px[1], px[2])
 READ_LINE_ARGB8888(RGBA8888_read_line, px, px[0], px[3], px[2], px[1])
 READ_LINE_ARGB8888(BGRA8888_read_line, px, px[0], px[1], px[2], px[3])
 
+READ_LINE_ARGB8888(RGB888_read_line, px, 0xFF, px[2], px[1], px[0])
+READ_LINE_ARGB8888(BGR888_read_line, px, 0xFF, px[0], px[1], px[2])
+
 READ_LINE_le16161616(ARGB16161616_read_line, px, px[3], px[2], px[1], px[0])
 READ_LINE_le16161616(ABGR16161616_read_line, px, px[3], px[0], px[1], px[2])
 READ_LINE_le16161616(XRGB16161616_read_line, px, cpu_to_le16(0xFFFF), px[2], px[1], px[0])
@@ -676,6 +679,10 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 		return &XRGB8888_read_line;
 	case DRM_FORMAT_XBGR8888:
 		return &XBGR8888_read_line;
+	case DRM_FORMAT_RGB888:
+		return &RGB888_read_line;
+	case DRM_FORMAT_BGR888:
+		return &BGR888_read_line;
 	case DRM_FORMAT_ARGB16161616:
 		return &ARGB16161616_read_line;
 	case DRM_FORMAT_ABGR16161616:
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 6e7597ab935d..9f34f3a18d8c 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -19,6 +19,8 @@ static const u32 vkms_formats[] = {
 	DRM_FORMAT_RGBA8888,
 	DRM_FORMAT_XRGB8888,
 	DRM_FORMAT_XBGR8888,
+	DRM_FORMAT_RGB888,
+	DRM_FORMAT_BGR888,
 	DRM_FORMAT_XRGB16161616,
 	DRM_FORMAT_XBGR16161616,
 	DRM_FORMAT_ARGB16161616,

From 7766ae8de472be3e0167cf3652e6ae8d74ef0221 Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:57:02 +0200
Subject: [PATCH 0994/1455] drm/vkms: Change YUV helpers to support u16 inputs
 for conversion
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Some YUV format uses 16 bit values, so change the helper function for
conversion to support those new formats.

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-6-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/tests/vkms_format_test.c | 143 +++++++++---------
 drivers/gpu/drm/vkms/vkms_formats.c           |  22 +--
 drivers/gpu/drm/vkms/vkms_formats.h           |   4 +-
 3 files changed, 85 insertions(+), 84 deletions(-)

diff --git a/drivers/gpu/drm/vkms/tests/vkms_format_test.c b/drivers/gpu/drm/vkms/tests/vkms_format_test.c
index 2e1daef94831..a7788fbc45dc 100644
--- a/drivers/gpu/drm/vkms/tests/vkms_format_test.c
+++ b/drivers/gpu/drm/vkms/tests/vkms_format_test.c
@@ -14,20 +14,20 @@
 MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");
 
 /**
- * struct pixel_yuv_u8 - Internal representation of a pixel color.
- * @y: Luma value, stored in 8 bits, without padding, using
+ * struct pixel_yuv_u16 - Internal representation of a pixel color.
+ * @y: Luma value, stored in 16 bits, without padding, using
  *     machine endianness
- * @u: Blue difference chroma value, stored in 8 bits, without padding, using
+ * @u: Blue difference chroma value, stored in 16 bits, without padding, using
  *     machine endianness
- * @v: Red difference chroma value, stored in 8 bits, without padding, using
+ * @v: Red difference chroma value, stored in 16 bits, without padding, using
  *     machine endianness
  */
-struct pixel_yuv_u8 {
-	u8 y, u, v;
+struct pixel_yuv_u16 {
+	u16 y, u, v;
 };
 
 /*
- * struct yuv_u8_to_argb_u16_case - Reference values to test the color
+ * struct yuv_u16_to_argb_u16_case - Reference values to test the color
  * conversions in VKMS between YUV to ARGB
  *
  * @encoding: Encoding used to convert RGB to YUV
@@ -39,13 +39,13 @@ struct pixel_yuv_u8 {
  * @format_pair.yuv: Same color as @format_pair.rgb, but converted to
  *                   YUV using @encoding and @range.
  */
-struct yuv_u8_to_argb_u16_case {
+struct yuv_u16_to_argb_u16_case {
 	enum drm_color_encoding encoding;
 	enum drm_color_range range;
 	size_t n_colors;
 	struct format_pair {
 		char *name;
-		struct pixel_yuv_u8 yuv;
+		struct pixel_yuv_u16 yuv;
 		struct pixel_argb_u16 argb;
 	} colors[TEST_BUFF_SIZE];
 };
@@ -57,14 +57,14 @@ struct yuv_u8_to_argb_u16_case {
  * For more information got to the docs:
  * https://colour.readthedocs.io/en/master/generated/colour.RGB_to_YCbCr.html
  */
-static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
+static struct yuv_u16_to_argb_u16_case yuv_u16_to_argb_u16_cases[] = {
 	/*
 	 * colour.RGB_to_YCbCr(<rgb color in 16 bit form>,
 	 *                     K=colour.WEIGHTS_YCBCR["ITU-R BT.601"],
 	 *                     in_bits = 16,
 	 *                     in_legal = False,
 	 *                     in_int = True,
-	 *                     out_bits = 8,
+	 *                     out_bits = 16,
 	 *                     out_legal = False,
 	 *                     out_int = True)
 	 *
@@ -76,13 +76,13 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 		.range = DRM_COLOR_YCBCR_FULL_RANGE,
 		.n_colors = 6,
 		.colors = {
-			{ "white", { 0xff, 0x80, 0x80 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
-			{ "gray",  { 0x80, 0x80, 0x80 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
-			{ "black", { 0x00, 0x80, 0x80 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
-			{ "red",   { 0x4c, 0x55, 0xff }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
-			{ "green", { 0x96, 0x2c, 0x15 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
-			{ "blue",  { 0x1d, 0xff, 0x6b }, { 0xffff, 0x0000, 0x0000, 0xffff }},
-		},
+			{ "white", { 0xffff, 0x8000, 0x8000 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
+			{ "gray",  { 0x8080, 0x8000, 0x8000 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
+			{ "black", { 0x0000, 0x8000, 0x8000 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
+			{ "red",   { 0x4c8b, 0x54ce, 0xffff }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
+			{ "green", { 0x9645, 0x2b33, 0x14d1 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
+			{ "blue",  { 0x1d2f, 0xffff, 0x6b2f }, { 0xffff, 0x0000, 0x0000, 0xffff }},
+		}
 	},
 	/*
 	 * colour.RGB_to_YCbCr(<rgb color in 16 bit form>,
@@ -90,7 +90,7 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 	 *                     in_bits = 16,
 	 *                     in_legal = False,
 	 *                     in_int = True,
-	 *                     out_bits = 8,
+	 *                     out_bits = 16,
 	 *                     out_legal = True,
 	 *                     out_int = True)
 	 * Tests cases for color conversion generated by converting RGB
@@ -101,13 +101,13 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 		.range = DRM_COLOR_YCBCR_LIMITED_RANGE,
 		.n_colors = 6,
 		.colors = {
-			{ "white", { 0xeb, 0x80, 0x80 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
-			{ "gray",  { 0x7e, 0x80, 0x80 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
-			{ "black", { 0x10, 0x80, 0x80 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
-			{ "red",   { 0x51, 0x5a, 0xf0 }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
-			{ "green", { 0x91, 0x36, 0x22 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
-			{ "blue",  { 0x29, 0xf0, 0x6e }, { 0xffff, 0x0000, 0x0000, 0xffff }},
-		},
+			{ "white", { 0xeb00, 0x8000, 0x8000 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
+			{ "gray",  { 0x7dee, 0x8000, 0x8000 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
+			{ "black", { 0x1000, 0x8000, 0x8000 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
+			{ "red",   { 0x517b, 0x5a34, 0xf000 }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
+			{ "green", { 0x908e, 0x35cc, 0x2237 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
+			{ "blue",  { 0x28f7, 0xf000, 0x6dc9 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
+		}
 	},
 	/*
 	 * colour.RGB_to_YCbCr(<rgb color in 16 bit form>,
@@ -115,7 +115,7 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 	 *                     in_bits = 16,
 	 *                     in_legal = False,
 	 *                     in_int = True,
-	 *                     out_bits = 8,
+	 *                     out_bits = 16,
 	 *                     out_legal = False,
 	 *                     out_int = True)
 	 * Tests cases for color conversion generated by converting RGB
@@ -126,21 +126,21 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 		.range = DRM_COLOR_YCBCR_FULL_RANGE,
 		.n_colors = 6,
 		.colors = {
-			{ "white", { 0xff, 0x80, 0x80 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
-			{ "gray",  { 0x80, 0x80, 0x80 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
-			{ "black", { 0x00, 0x80, 0x80 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
-			{ "red",   { 0x36, 0x63, 0xff }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
-			{ "green", { 0xb6, 0x1e, 0x0c }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
-			{ "blue",  { 0x12, 0xff, 0x74 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
-		},
+			{ "white", { 0xffff, 0x8000, 0x8000 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
+			{ "gray",  { 0x8080, 0x8000, 0x8000 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
+			{ "black", { 0x0000, 0x8000, 0x8000 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
+			{ "red",   { 0x366d, 0x62ac, 0xffff }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
+			{ "green", { 0xb717, 0x1d55, 0x0bbd }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
+			{ "blue",  { 0x127c, 0xffff, 0x7443 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
+		}
 	},
 	/*
 	 * colour.RGB_to_YCbCr(<rgb color in 16 bit form>,
 	 *                     K=colour.WEIGHTS_YCBCR["ITU-R BT.709"],
 	 *                     in_bits = 16,
-	 *                     int_legal = False,
+	 *                     in_legal = False,
 	 *                     in_int = True,
-	 *                     out_bits = 8,
+	 *                     out_bits = 16,
 	 *                     out_legal = True,
 	 *                     out_int = True)
 	 * Tests cases for color conversion generated by converting RGB
@@ -151,13 +151,13 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 		.range = DRM_COLOR_YCBCR_LIMITED_RANGE,
 		.n_colors = 6,
 		.colors = {
-			{ "white", { 0xeb, 0x80, 0x80 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
-			{ "gray",  { 0x7e, 0x80, 0x80 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
-			{ "black", { 0x10, 0x80, 0x80 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
-			{ "red",   { 0x3f, 0x66, 0xf0 }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
-			{ "green", { 0xad, 0x2a, 0x1a }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
-			{ "blue",  { 0x20, 0xf0, 0x76 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
-		},
+			{ "white", { 0xeb00, 0x8000, 0x8000 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
+			{ "gray",  { 0x7dee, 0x8000, 0x8000 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
+			{ "black", { 0x1000, 0x8000, 0x8000 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
+			{ "red",   { 0x3e8f, 0x6656, 0xf000 }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
+			{ "green", { 0xaca1, 0x29aa, 0x1a45 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
+			{ "blue",  { 0x1fd0, 0xf000, 0x75bb }, { 0xffff, 0x0000, 0x0000, 0xffff }},
+		}
 	},
 	/*
 	 * colour.RGB_to_YCbCr(<rgb color in 16 bit form>,
@@ -165,7 +165,7 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 	 *                     in_bits = 16,
 	 *                     in_legal = False,
 	 *                     in_int = True,
-	 *                     out_bits = 8,
+	 *                     out_bits = 16,
 	 *                     out_legal = False,
 	 *                     out_int = True)
 	 * Tests cases for color conversion generated by converting RGB
@@ -176,13 +176,13 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 		.range = DRM_COLOR_YCBCR_FULL_RANGE,
 		.n_colors = 6,
 		.colors = {
-			{ "white", { 0xff, 0x80, 0x80 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
-			{ "gray",  { 0x80, 0x80, 0x80 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
-			{ "black", { 0x00, 0x80, 0x80 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
-			{ "red",   { 0x43, 0x5c, 0xff }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
-			{ "green", { 0xad, 0x24, 0x0b }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
-			{ "blue",  { 0x0f, 0xff, 0x76 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
-		},
+			{ "white", { 0xffff, 0x8000, 0x8000 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
+			{ "gray",  { 0x8080, 0x8000, 0x8000 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
+			{ "black", { 0x0000, 0x8000, 0x8000 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
+			{ "red",   { 0x4340, 0x5c41, 0xffff }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
+			{ "green", { 0xad91, 0x23bf, 0x0a4c }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
+			{ "blue",  { 0x0f2e, 0xffff, 0x75b5 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
+		}
 	},
 	/*
 	 * colour.RGB_to_YCbCr(<rgb color in 16 bit form>,
@@ -190,7 +190,7 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 	 *                     in_bits = 16,
 	 *                     in_legal = False,
 	 *                     in_int = True,
-	 *                     out_bits = 8,
+	 *                     out_bits = 16,
 	 *                     out_legal = True,
 	 *                     out_int = True)
 	 * Tests cases for color conversion generated by converting RGB
@@ -201,32 +201,30 @@ static struct yuv_u8_to_argb_u16_case yuv_u8_to_argb_u16_cases[] = {
 		.range = DRM_COLOR_YCBCR_LIMITED_RANGE,
 		.n_colors = 6,
 		.colors = {
-			{ "white", { 0xeb, 0x80, 0x80 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
-			{ "gray",  { 0x7e, 0x80, 0x80 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
-			{ "black", { 0x10, 0x80, 0x80 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
-			{ "red",   { 0x4a, 0x61, 0xf0 }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
-			{ "green", { 0xa4, 0x2f, 0x19 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
-			{ "blue",  { 0x1d, 0xf0, 0x77 }, { 0xffff, 0x0000, 0x0000, 0xffff }},
-		},
+			{ "white", { 0xeb00, 0x8000, 0x8000 }, { 0xffff, 0xffff, 0xffff, 0xffff }},
+			{ "gray",  { 0x7dee, 0x8000, 0x8000 }, { 0xffff, 0x8080, 0x8080, 0x8080 }},
+			{ "black", { 0x1000, 0x8000, 0x8000 }, { 0xffff, 0x0000, 0x0000, 0x0000 }},
+			{ "red",   { 0x4988, 0x60b9, 0xf000 }, { 0xffff, 0xffff, 0x0000, 0x0000 }},
+			{ "green", { 0xa47b, 0x2f47, 0x1902 }, { 0xffff, 0x0000, 0xffff, 0x0000 }},
+			{ "blue",  { 0x1cfd, 0xf000, 0x76fe }, { 0xffff, 0x0000, 0x0000, 0xffff }},
+		}
 	},
 };
 
 /*
- * vkms_format_test_yuv_u8_to_argb_u16 - Testing the conversion between YUV
+ * vkms_format_test_yuv_u16_to_argb_u16 - Testing the conversion between YUV
  * colors to ARGB colors in VKMS
  *
  * This test will use the functions get_conversion_matrix_to_argb_u16 and
- * argb_u16_from_yuv888 to convert YUV colors (stored in
- * yuv_u8_to_argb_u16_cases) into ARGB colors.
+ * argb_u16_from_yuv161616 to convert YUV colors (stored in
+ * yuv_u16_to_argb_u16_cases) into ARGB colors.
  *
  * The conversion between YUV and RGB is not totally reversible, so there may be
  * some difference between the expected value and the result.
- * In addition, there may be some rounding error as the input color is 8 bits
- * and output color is 16 bits.
  */
-static void vkms_format_test_yuv_u8_to_argb_u16(struct kunit *test)
+static void vkms_format_test_yuv_u16_to_argb_u16(struct kunit *test)
 {
-	const struct yuv_u8_to_argb_u16_case *param = test->param_value;
+	const struct yuv_u16_to_argb_u16_case *param = test->param_value;
 	struct pixel_argb_u16 argb;
 
 	for (size_t i = 0; i < param->n_colors; i++) {
@@ -236,7 +234,8 @@ static void vkms_format_test_yuv_u8_to_argb_u16(struct kunit *test)
 		get_conversion_matrix_to_argb_u16
 			(DRM_FORMAT_NV12, param->encoding, param->range, &matrix);
 
-		argb = argb_u16_from_yuv888(color->yuv.y, color->yuv.u, color->yuv.v, &matrix);
+		argb = argb_u16_from_yuv161616(&matrix, color->yuv.y, color->yuv.u,
+					       color->yuv.v);
 
 		KUNIT_EXPECT_LE_MSG(test, abs_diff(argb.a, color->argb.a), 0x1ff,
 				    "On the A channel of the color %s expected 0x%04x, got 0x%04x",
@@ -253,19 +252,19 @@ static void vkms_format_test_yuv_u8_to_argb_u16(struct kunit *test)
 	}
 }
 
-static void vkms_format_test_yuv_u8_to_argb_u16_case_desc(struct yuv_u8_to_argb_u16_case *t,
-							  char *desc)
+static void vkms_format_test_yuv_u16_to_argb_u16_case_desc(struct yuv_u16_to_argb_u16_case *t,
+							   char *desc)
 {
 	snprintf(desc, KUNIT_PARAM_DESC_SIZE, "%s - %s",
 		 drm_get_color_encoding_name(t->encoding), drm_get_color_range_name(t->range));
 }
 
-KUNIT_ARRAY_PARAM(yuv_u8_to_argb_u16, yuv_u8_to_argb_u16_cases,
-		  vkms_format_test_yuv_u8_to_argb_u16_case_desc
+KUNIT_ARRAY_PARAM(yuv_u16_to_argb_u16, yuv_u16_to_argb_u16_cases,
+		  vkms_format_test_yuv_u16_to_argb_u16_case_desc
 );
 
 static struct kunit_case vkms_format_test_cases[] = {
-	KUNIT_CASE_PARAM(vkms_format_test_yuv_u8_to_argb_u16, yuv_u8_to_argb_u16_gen_params),
+	KUNIT_CASE_PARAM(vkms_format_test_yuv_u16_to_argb_u16, yuv_u16_to_argb_u16_gen_params),
 	{}
 };
 
diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index f1649d9bf430..cec27a72bc48 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -269,16 +269,17 @@ static struct pixel_argb_u16 argb_u16_from_BGR565(const __le16 *pixel)
 	return out_pixel;
 }
 
-VISIBLE_IF_KUNIT struct pixel_argb_u16 argb_u16_from_yuv888(u8 y, u8 channel_1, u8 channel_2,
-							    const struct conversion_matrix *matrix)
+VISIBLE_IF_KUNIT
+struct pixel_argb_u16 argb_u16_from_yuv161616(const struct conversion_matrix *matrix,
+					      u16 y, u16 channel_1, u16 channel_2)
 {
 	u16 r, g, b;
 	s64 fp_y, fp_channel_1, fp_channel_2;
 	s64 fp_r, fp_g, fp_b;
 
-	fp_y = drm_int2fixp(((int)y - matrix->y_offset) * 257);
-	fp_channel_1 = drm_int2fixp(((int)channel_1 - 128) * 257);
-	fp_channel_2 = drm_int2fixp(((int)channel_2 - 128) * 257);
+	fp_y = drm_int2fixp((int)y - matrix->y_offset * 257);
+	fp_channel_1 = drm_int2fixp((int)channel_1 - 128 * 257);
+	fp_channel_2 = drm_int2fixp((int)channel_2 - 128 * 257);
 
 	fp_r = drm_fixp_mul(matrix->matrix[0][0], fp_y) +
 	       drm_fixp_mul(matrix->matrix[0][1], fp_channel_1) +
@@ -300,7 +301,7 @@ VISIBLE_IF_KUNIT struct pixel_argb_u16 argb_u16_from_yuv888(u8 y, u8 channel_1,
 
 	return argb_u16_from_u16161616(0xffff, r, g, b);
 }
-EXPORT_SYMBOL_IF_KUNIT(argb_u16_from_yuv888);
+EXPORT_SYMBOL_IF_KUNIT(argb_u16_from_yuv161616);
 
 /**
  * READ_LINE() - Generic generator for a read_line function which can be used for format with one
@@ -498,8 +499,8 @@ static void semi_planar_yuv_read_line(const struct vkms_plane_state *plane, int
 	const struct conversion_matrix *conversion_matrix = &plane->conversion_matrix;
 
 	for (int i = 0; i < count; i++) {
-		*out_pixel = argb_u16_from_yuv888(y_plane[0], uv_plane[0], uv_plane[1],
-						  conversion_matrix);
+		*out_pixel = argb_u16_from_yuv161616(conversion_matrix, y_plane[0] * 257,
+						     uv_plane[0] * 257, uv_plane[1] * 257);
 		out_pixel += 1;
 		y_plane += step_y;
 		if ((i + subsampling_offset + 1) % subsampling == 0)
@@ -543,8 +544,9 @@ static void planar_yuv_read_line(const struct vkms_plane_state *plane, int x_sta
 	const struct conversion_matrix *conversion_matrix = &plane->conversion_matrix;
 
 	for (int i = 0; i < count; i++) {
-		*out_pixel = argb_u16_from_yuv888(*y_plane, *channel_1_plane, *channel_2_plane,
-						  conversion_matrix);
+		*out_pixel = argb_u16_from_yuv161616(conversion_matrix,
+						     *y_plane * 257, *channel_1_plane * 257,
+						     *channel_2_plane * 257);
 		out_pixel += 1;
 		y_plane += step_y;
 		if ((i + subsampling_offset + 1) % subsampling == 0) {
diff --git a/drivers/gpu/drm/vkms/vkms_formats.h b/drivers/gpu/drm/vkms/vkms_formats.h
index b4fe62ab9c65..eeb208cdd6b1 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.h
+++ b/drivers/gpu/drm/vkms/vkms_formats.h
@@ -14,8 +14,8 @@ void get_conversion_matrix_to_argb_u16(u32 format, enum drm_color_encoding encod
 				       struct conversion_matrix *matrix);
 
 #if IS_ENABLED(CONFIG_KUNIT)
-struct pixel_argb_u16 argb_u16_from_yuv888(u8 y, u8 channel_1, u8 channel_2,
-					   const struct conversion_matrix *matrix);
+struct pixel_argb_u16 argb_u16_from_yuv161616(const struct conversion_matrix *matrix,
+					      u16 y, u16 channel_1, u16 channel_2);
 #endif
 
 #endif /* _VKMS_FORMATS_H_ */

From 9e6600e9d3683cdf0196fde7439dba4c50946906 Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:57:03 +0200
Subject: [PATCH 0995/1455] drm/vkms: Create helper macro for YUV formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The callback functions for line conversion are almost identical for
semi-planar formats. The generic READ_LINE_YUV_SEMIPLANAR macro
generate all the required boilerplate to process a line from a
semi-planar format.

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-7-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 75 ++++++++++++++++++-----------
 1 file changed, 48 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index cec27a72bc48..639b1c7ca43e 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -479,35 +479,56 @@ READ_LINE(R8_read_line, px, u8, argb_u16_from_gray8, *px)
  * - Convert YUV and YVU with the same function (a column swap is needed when setting up
  * plane->conversion_matrix)
  */
-static void semi_planar_yuv_read_line(const struct vkms_plane_state *plane, int x_start,
-				      int y_start, enum pixel_read_direction direction, int count,
-				      struct pixel_argb_u16 out_pixel[])
-{
-	u8 *y_plane;
-	u8 *uv_plane;
 
-	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0,
-			       &y_plane);
-	packed_pixels_addr_1x1(plane->frame_info,
-			       x_start / plane->frame_info->fb->format->hsub,
-			       y_start / plane->frame_info->fb->format->vsub, 1,
-			       &uv_plane);
-	int step_y = get_block_step_bytes(plane->frame_info->fb, direction, 0);
-	int step_uv = get_block_step_bytes(plane->frame_info->fb, direction, 1);
-	int subsampling = get_subsampling(plane->frame_info->fb->format, direction);
-	int subsampling_offset = get_subsampling_offset(direction, x_start, y_start);
-	const struct conversion_matrix *conversion_matrix = &plane->conversion_matrix;
-
-	for (int i = 0; i < count; i++) {
-		*out_pixel = argb_u16_from_yuv161616(conversion_matrix, y_plane[0] * 257,
-						     uv_plane[0] * 257, uv_plane[1] * 257);
-		out_pixel += 1;
-		y_plane += step_y;
-		if ((i + subsampling_offset + 1) % subsampling == 0)
-			uv_plane += step_uv;
-	}
+/**
+ * READ_LINE_YUV_SEMIPLANAR() - Generic generator for a read_line function which can be used for yuv
+ * formats with two planes and block_w == block_h == 1.
+ *
+ * @function_name: Function name to generate
+ * @pixel_1_name: temporary pixel name for the first plane used in the @__VA_ARGS__ parameters
+ * @pixel_2_name: temporary pixel name for the second plane used in the @__VA_ARGS__ parameters
+ * @pixel_1_type: Used to specify the type you want to cast the pixel pointer on the plane 1
+ * @pixel_2_type: Used to specify the type you want to cast the pixel pointer on the plane 2
+ * @callback: Callback to call for each pixels. This function should take
+ *            (struct conversion_matrix*, @__VA_ARGS__) as parameter and return a pixel_argb_u16
+ * __VA_ARGS__: Argument to pass inside the callback. You can use @pixel_1_name and @pixel_2_name
+ *               to access current pixel values
+ */
+#define READ_LINE_YUV_SEMIPLANAR(function_name, pixel_1_name, pixel_2_name, pixel_1_type,	\
+				 pixel_2_type, callback, ...)					\
+static void function_name(const struct vkms_plane_state *plane, int x_start,			\
+		 int y_start, enum pixel_read_direction direction, int count,			\
+		 struct pixel_argb_u16 out_pixel[])						\
+{												\
+	u8 *plane_1;										\
+	u8 *plane_2;										\
+												\
+	packed_pixels_addr_1x1(plane->frame_info, x_start, y_start, 0,				\
+			       &plane_1);							\
+	packed_pixels_addr_1x1(plane->frame_info,						\
+			       x_start / plane->frame_info->fb->format->hsub,			\
+			       y_start / plane->frame_info->fb->format->vsub, 1,		\
+			       &plane_2);							\
+	int step_1 = get_block_step_bytes(plane->frame_info->fb, direction, 0);			\
+	int step_2 = get_block_step_bytes(plane->frame_info->fb, direction, 1);			\
+	int subsampling = get_subsampling(plane->frame_info->fb->format, direction);		\
+	int subsampling_offset = get_subsampling_offset(direction, x_start, y_start);		\
+	const struct conversion_matrix *conversion_matrix = &plane->conversion_matrix;		\
+												\
+	for (int i = 0; i < count; i++) {							\
+		pixel_1_type *(pixel_1_name) = (pixel_1_type *)plane_1;				\
+		pixel_2_type *(pixel_2_name) = (pixel_2_type *)plane_2;				\
+		*out_pixel = (callback)(conversion_matrix, __VA_ARGS__);			\
+		out_pixel += 1;									\
+		plane_1 += step_1;								\
+		if ((i + subsampling_offset + 1) % subsampling == 0)				\
+			plane_2 += step_2;							\
+	}											\
 }
 
+READ_LINE_YUV_SEMIPLANAR(YUV888_semiplanar_read_line, y, uv, u8, u8, argb_u16_from_yuv161616,
+			 y[0] * 257, uv[0] * 257, uv[1] * 257)
+
 /*
  * This callback can be used for YUV format where each color component is
  * stored in a different plane (often called planar formats). It will
@@ -703,7 +724,7 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 	case DRM_FORMAT_NV21:
 	case DRM_FORMAT_NV61:
 	case DRM_FORMAT_NV42:
-		return &semi_planar_yuv_read_line;
+		return &YUV888_semiplanar_read_line;
 	case DRM_FORMAT_YUV420:
 	case DRM_FORMAT_YUV422:
 	case DRM_FORMAT_YUV444:

From 9cd6b43a0f7fc0dfb779c73f4aeb4d3fe8820698 Mon Sep 17 00:00:00 2001
From: Louis Chauvet <louis.chauvet@bootlin.com>
Date: Thu, 3 Jul 2025 09:57:04 +0200
Subject: [PATCH 0996/1455] drm/vkms: Add P01* formats
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The formats NV 12/16/24/21/61/42 were already supported.
Add support for:
- P010
- P012
- P016

Reviewed-by: Maíra Canal <mcanal@igalia.com>
Acked-by: Daniel Stone <daniels@collabora.com>
Link: https://lore.kernel.org/r/20250703-b4-new-color-formats-v7-8-15fd8fd2e15c@bootlin.com
Signed-off-by: Louis Chauvet <louis.chauvet@bootlin.com>
---
 drivers/gpu/drm/vkms/vkms_formats.c | 7 ++++++-
 drivers/gpu/drm/vkms/vkms_plane.c   | 3 +++
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/vkms/vkms_formats.c b/drivers/gpu/drm/vkms/vkms_formats.c
index 639b1c7ca43e..dfb8e13cba87 100644
--- a/drivers/gpu/drm/vkms/vkms_formats.c
+++ b/drivers/gpu/drm/vkms/vkms_formats.c
@@ -528,7 +528,8 @@ static void function_name(const struct vkms_plane_state *plane, int x_start,			\
 
 READ_LINE_YUV_SEMIPLANAR(YUV888_semiplanar_read_line, y, uv, u8, u8, argb_u16_from_yuv161616,
 			 y[0] * 257, uv[0] * 257, uv[1] * 257)
-
+READ_LINE_YUV_SEMIPLANAR(YUV161616_semiplanar_read_line, y, uv, u16, u16, argb_u16_from_yuv161616,
+			 y[0], uv[0], uv[1])
 /*
  * This callback can be used for YUV format where each color component is
  * stored in a different plane (often called planar formats). It will
@@ -725,6 +726,10 @@ pixel_read_line_t get_pixel_read_line_function(u32 format)
 	case DRM_FORMAT_NV61:
 	case DRM_FORMAT_NV42:
 		return &YUV888_semiplanar_read_line;
+	case DRM_FORMAT_P010:
+	case DRM_FORMAT_P012:
+	case DRM_FORMAT_P016:
+		return &YUV161616_semiplanar_read_line;
 	case DRM_FORMAT_YUV420:
 	case DRM_FORMAT_YUV422:
 	case DRM_FORMAT_YUV444:
diff --git a/drivers/gpu/drm/vkms/vkms_plane.c b/drivers/gpu/drm/vkms/vkms_plane.c
index 9f34f3a18d8c..e592e47a5736 100644
--- a/drivers/gpu/drm/vkms/vkms_plane.c
+++ b/drivers/gpu/drm/vkms/vkms_plane.c
@@ -39,6 +39,9 @@ static const u32 vkms_formats[] = {
 	DRM_FORMAT_YVU420,
 	DRM_FORMAT_YVU422,
 	DRM_FORMAT_YVU444,
+	DRM_FORMAT_P010,
+	DRM_FORMAT_P012,
+	DRM_FORMAT_P016,
 	DRM_FORMAT_R1,
 	DRM_FORMAT_R2,
 	DRM_FORMAT_R4,

From 6fc957185e1691bb6dfa4193698a229db537c2a2 Mon Sep 17 00:00:00 2001
From: Julia Filipchuk <julia.filipchuk@intel.com>
Date: Wed, 3 Sep 2025 12:00:38 -0700
Subject: [PATCH 0997/1455] drm/xe: Extend Wa_13011645652 to PTL-H, WCL
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Expand workaround to additional graphics architectures.

Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: intel-xe@lists.freedesktop.org
Cc: <stable@vger.kernel.org> # v6.17+
Signed-off-by: Julia Filipchuk <julia.filipchuk@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250903190122.1028373-2-julia.filipchuk@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_wa_oob.rules | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 0cf2b3c03532..338c344dcd7d 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -32,7 +32,8 @@
 16022287689	GRAPHICS_VERSION(2001)
 		GRAPHICS_VERSION(2004)
 13011645652	GRAPHICS_VERSION(2004)
-		GRAPHICS_VERSION(3001)
+		GRAPHICS_VERSION_RANGE(3000, 3001)
+		GRAPHICS_VERSION(3003)
 14022293748	GRAPHICS_VERSION_RANGE(2001, 2002)
 		GRAPHICS_VERSION(2004)
 		GRAPHICS_VERSION_RANGE(3000, 3001)

From 4320fd9e0d817f63b31dc380a1127335c445083c Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Wed, 3 Sep 2025 09:20:18 +0100
Subject: [PATCH 0998/1455] drm/amd/amdgpu: Fix a less than zero check on a
 uint32_t struct field

Currently the error check from the call to mes_v12_inv_tlb_convert_hub_id
is always false because a uint32_t struct field hub_id is being used to
to perform the less than zero error check. Fix this by using the int
variable ret to perform the check.

Fixes: 87e65052616c ("drm/amd/amdgpu : Use the MES INV_TLBS API for tlb invalidation on gfx12")
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_v12_0.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index cd5c966cee95..ff5df28b57ec 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -899,6 +899,7 @@ static int mes_v12_0_inv_tlbs_pasid(struct amdgpu_mes *mes,
 				    struct mes_inv_tlbs_pasid_input *input)
 {
 	union MESAPI__INV_TLBS mes_inv_tlbs;
+	int ret;
 
 	memset(&mes_inv_tlbs, 0, sizeof(mes_inv_tlbs));
 
@@ -911,9 +912,10 @@ static int mes_v12_0_inv_tlbs_pasid(struct amdgpu_mes *mes,
 	mes_inv_tlbs.invalidate_tlbs.inv_sel_id = input->pasid;
 
 	/*convert amdgpu_mes_hub_id to mes expected hub_id */
-	mes_inv_tlbs.invalidate_tlbs.hub_id = mes_v12_inv_tlb_convert_hub_id(input->hub_id);
-	if (mes_inv_tlbs.invalidate_tlbs.hub_id < 0)
+	ret = mes_v12_inv_tlb_convert_hub_id(input->hub_id);
+	if (ret < 0)
 		return -EINVAL;
+	mes_inv_tlbs.invalidate_tlbs.hub_id = ret;
 	return mes_v12_0_submit_pkt_and_poll_completion(mes, AMDGPU_MES_KIQ_PIPE,
 			&mes_inv_tlbs, sizeof(mes_inv_tlbs),
 			offsetof(union MESAPI__INV_TLBS, api_status));

From 7670daf65afae2d6d08b300df64c96439e813b3a Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 3 Sep 2025 20:03:45 +0800
Subject: [PATCH 0999/1455] drm/amdgpu/amdgpu_cper: Remove redundant ternary
 operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? false : true", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
index 6c266f18c598..c4590ec35cad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
@@ -174,7 +174,7 @@ int amdgpu_cper_entry_fill_runtime_section(struct amdgpu_device *adev,
 	struct cper_sec_nonstd_err *section;
 	bool poison;
 
-	poison = (sev == CPER_SEV_NON_FATAL_CORRECTED) ? false : true;
+	poison = sev != CPER_SEV_NON_FATAL_CORRECTED;
 	section_desc = (struct cper_sec_desc *)((uint8_t *)hdr + SEC_DESC_OFFSET(idx));
 	section = (struct cper_sec_nonstd_err *)((uint8_t *)hdr +
 		   NONSTD_SEC_OFFSET(hdr->sec_cnt, idx));

From d261e744af55179fcdd71fa767088421e6dea158 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 3 Sep 2025 20:03:46 +0800
Subject: [PATCH 1000/1455] drm/amdgpu/gfx: Remove redundant ternary operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? false : true", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 3 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index c85de8c8f6f5..3d9c045a8a64 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -4643,8 +4643,7 @@ static int gfx_v11_0_gfxhub_enable(struct amdgpu_device *adev)
 
 	amdgpu_device_flush_hdp(adev, NULL);
 
-	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
-		false : true;
+	value = amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS;
 
 	adev->gfxhub.funcs->set_fault_enable_default(adev, value);
 	/* TODO investigate why this and the hdp flush above is needed,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index fd44d5503e28..5dbc5dbc694a 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -3524,8 +3524,7 @@ static int gfx_v12_0_gfxhub_enable(struct amdgpu_device *adev)
 
 	amdgpu_device_flush_hdp(adev, NULL);
 
-	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
-		false : true;
+	value = amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS;
 
 	adev->gfxhub.funcs->set_fault_enable_default(adev, value);
 	/* TODO investigate why this and the hdp flush above is needed,

From 60df3e81d742e95805ca5e32ef336f87ab35dc37 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 3 Sep 2025 20:03:47 +0800
Subject: [PATCH 1001/1455] drm/amdgpu/gmc: Remove redundant ternary operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? false : true", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 3 +--
 drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 3 +--
 drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c | 3 +--
 3 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
index 7031dd8c3c5e..d7499be8c4bf 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c
@@ -963,8 +963,7 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev)
 	/* Flush HDP after it is initialized */
 	amdgpu_device_flush_hdp(adev, NULL);
 
-	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
-		false : true;
+	value = amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS;
 
 	if (!adev->in_s0ix)
 		adev->gfxhub.funcs->set_fault_enable_default(adev, value);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
index 93d2b0bbe641..7bc389d9f5c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c
@@ -905,8 +905,7 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev)
 	/* Flush HDP after it is initialized */
 	amdgpu_device_flush_hdp(adev, NULL);
 
-	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
-		false : true;
+	value = amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS;
 
 	adev->mmhub.funcs->set_fault_enable_default(adev, value);
 	gmc_v11_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
index 9ba055ddc00f..76d3c40735b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
@@ -893,8 +893,7 @@ static int gmc_v12_0_gart_enable(struct amdgpu_device *adev)
 	/* Flush HDP after it is initialized */
 	amdgpu_device_flush_hdp(adev, NULL);
 
-	value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
-		false : true;
+	value = amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS;
 
 	adev->mmhub.funcs->set_fault_enable_default(adev, value);
 	gmc_v12_0_flush_gpu_tlb(adev, 0, AMDGPU_MMHUB0(0), 0);

From 8a4bc4508c6b104210b85f380d3237d3a2582f37 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 3 Sep 2025 20:03:48 +0800
Subject: [PATCH 1002/1455] drm/amdgpu/ih: Remove redundant ternary operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? false : true", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/ih_v6_0.c | 3 +--
 drivers/gpu/drm/amd/amdgpu/ih_v6_1.c | 3 +--
 drivers/gpu/drm/amd/amdgpu/ih_v7_0.c | 3 +--
 3 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
index 5900b560b7de..333e9c30c091 100644
--- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c
@@ -587,8 +587,7 @@ static int ih_v6_0_sw_init(struct amdgpu_ip_block *ip_block)
 	/* use gpu virtual address for ih ring
 	 * until ih_checken is programmed to allow
 	 * use bus address for ih ring by psp bl */
-	use_bus_addr =
-		(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) ? false : true;
+	use_bus_addr = adev->firmware.load_type != AMDGPU_FW_LOAD_PSP;
 	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, IH_RING_SIZE, use_bus_addr);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c
index 068ed849dbad..95b3f4e55ec3 100644
--- a/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c
@@ -562,8 +562,7 @@ static int ih_v6_1_sw_init(struct amdgpu_ip_block *ip_block)
 	/* use gpu virtual address for ih ring
 	 * until ih_checken is programmed to allow
 	 * use bus address for ih ring by psp bl */
-	use_bus_addr =
-		(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) ? false : true;
+	use_bus_addr = adev->firmware.load_type != AMDGPU_FW_LOAD_PSP;
 	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c
index 40a3530e0453..b32ea4129c61 100644
--- a/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c
@@ -552,8 +552,7 @@ static int ih_v7_0_sw_init(struct amdgpu_ip_block *ip_block)
 	/* use gpu virtual address for ih ring
 	 * until ih_checken is programmed to allow
 	 * use bus address for ih ring by psp bl */
-	use_bus_addr =
-		(adev->firmware.load_type == AMDGPU_FW_LOAD_PSP) ? false : true;
+	use_bus_addr = adev->firmware.load_type != AMDGPU_FW_LOAD_PSP;
 	r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 256 * 1024, use_bus_addr);
 	if (r)
 		return r;

From 9502b09933ed8debe9888c69b10e79e3a2b66800 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 3 Sep 2025 20:03:49 +0800
Subject: [PATCH 1003/1455] drm/amdgpu/jpeg: Remove redundant ternary operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
index 481d1a2dbe5a..5d86e1d846eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
@@ -686,7 +686,7 @@ static int jpeg_v4_0_5_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = ip_block->adev;
-	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	bool enable = state == AMD_CG_STATE_GATE;
 	int i;
 
 	for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
index e0a71909252b..34c70270ea1d 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
@@ -584,7 +584,7 @@ static int jpeg_v5_0_0_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = ip_block->adev;
-	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	bool enable = state == AMD_CG_STATE_GATE;
 
 	if (enable) {
 		if (!jpeg_v5_0_0_is_idle(ip_block))
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
index 47bde62b3909..baf097d2e1ac 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_1.c
@@ -697,7 +697,7 @@ static int jpeg_v5_0_1_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 					     enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = ip_block->adev;
-	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	bool enable = state == AMD_CG_STATE_GATE;
 
 	int i;
 

From 086f66edf9244d8c05b0ddf24631fe7f3472ac45 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 3 Sep 2025 20:03:50 +0800
Subject: [PATCH 1004/1455] drm/amdgpu/vcn: Remove redundant ternary operators
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index 75c884a8f556..6dbf33b26ee2 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -1591,7 +1591,7 @@ static int vcn_v4_0_5_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = ip_block->adev;
-	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	bool enable = state == AMD_CG_STATE_GATE;
 	int i;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index 455f829b8bb9..536f06b81706 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -1311,7 +1311,7 @@ static int vcn_v5_0_0_set_clockgating_state(struct amdgpu_ip_block *ip_block,
 					  enum amd_clockgating_state state)
 {
 	struct amdgpu_device *adev = ip_block->adev;
-	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
+	bool enable = state == AMD_CG_STATE_GATE;
 	int i;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {

From 3f36e712a8b17d49c089706e229849d3f745a540 Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Wed, 3 Sep 2025 20:21:10 +0800
Subject: [PATCH 1005/1455] drm/radeon/ci_dpm: Use int type to store negative
 error codes

Change the 'ret' variable in ci_populate_all_graphic_levels()
and ci_populate_all_memory_levels() from u32 to int, as it needs to store
either negative error codes or zero returned by other functions.

Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but can be confusing.  Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.

No effect on runtime.

Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/ci_dpm.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 7c3a960f486a..1f9f4424055a 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -3238,7 +3238,8 @@ static int ci_populate_all_graphic_levels(struct radeon_device *rdev)
 	u32 level_array_size = sizeof(SMU7_Discrete_GraphicsLevel) *
 		SMU7_MAX_LEVELS_GRAPHICS;
 	SMU7_Discrete_GraphicsLevel *levels = pi->smc_state_table.GraphicsLevel;
-	u32 i, ret;
+	int ret;
+	u32 i;
 
 	memset(levels, 0, level_array_size);
 
@@ -3285,7 +3286,8 @@ static int ci_populate_all_memory_levels(struct radeon_device *rdev)
 	u32 level_array_size = sizeof(SMU7_Discrete_MemoryLevel) *
 		SMU7_MAX_LEVELS_MEMORY;
 	SMU7_Discrete_MemoryLevel *levels = pi->smc_state_table.MemoryLevel;
-	u32 i, ret;
+	int ret;
+	u32 i;
 
 	memset(levels, 0, level_array_size);
 

From 36cc7d13178d901982da7a122c883861d98da624 Mon Sep 17 00:00:00 2001
From: Eric Huang <jinhuieric.huang@amd.com>
Date: Mon, 25 Aug 2025 09:50:49 -0400
Subject: [PATCH 1006/1455] drm/amdkfd: fix p2p links bug in topology

When creating p2p links, KFD needs to check XGMI link
with two conditions, hive_id and is_sharing_enabled,
but it is missing to check is_sharing_enabled, so add
it to fix the error.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index 82dbd68d8c99..5c98746eb72d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -1589,7 +1589,8 @@ static int kfd_dev_create_p2p_links(void)
 			break;
 		if (!dev->gpu || !dev->gpu->adev ||
 		    (dev->gpu->kfd->hive_id &&
-		     dev->gpu->kfd->hive_id == new_dev->gpu->kfd->hive_id))
+		     dev->gpu->kfd->hive_id == new_dev->gpu->kfd->hive_id &&
+		     amdgpu_xgmi_get_is_sharing_enabled(dev->gpu->adev, new_dev->gpu->adev)))
 			goto next;
 
 		/* check if node(s) is/are peer accessible in one direction or bi-direction */

From f320ed01cf5f2259e2035a56900952cb3cc77e7a Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Tue, 2 Sep 2025 22:13:44 +0800
Subject: [PATCH 1007/1455] drm/amdgpu: Correct info field of bad page
 threshold exceed CPER

Correct valid_bits and ms_chk_bits of section info field for bad page
threshold exceed CPER to match OOB's behavior.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
index c4590ec35cad..ef996493115f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cper.c
@@ -68,7 +68,6 @@ void amdgpu_cper_entry_fill_hdr(struct amdgpu_device *adev,
 	hdr->error_severity		= sev;
 
 	hdr->valid_bits.platform_id	= 1;
-	hdr->valid_bits.partition_id	= 1;
 	hdr->valid_bits.timestamp	= 1;
 
 	amdgpu_cper_get_timestamp(&hdr->timestamp);
@@ -220,7 +219,10 @@ int amdgpu_cper_entry_fill_bad_page_threshold_section(struct amdgpu_device *adev
 	section->hdr.valid_bits.err_context_cnt = 1;
 
 	section->info.error_type = RUNTIME;
+	section->info.valid_bits.ms_chk = 1;
 	section->info.ms_chk_bits.err_type_valid = 1;
+	section->info.ms_chk_bits.err_type = 1;
+	section->info.ms_chk_bits.pcc = 1;
 	section->ctx.reg_ctx_type = CPER_CTX_TYPE_CRASH;
 	section->ctx.reg_arr_size = sizeof(section->ctx.reg_dump);
 

From 7e0fc7b2b7f7e030f5886fa21e002f2edb6c12be Mon Sep 17 00:00:00 2001
From: Sunil Khatri <sunil.khatri@amd.com>
Date: Thu, 14 Aug 2025 13:23:32 +0530
Subject: [PATCH 1008/1455] drm/amdgpu: add more information in debugfs to
 pagetable dump
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add more information in the debugfs which is needed to dump
a pagetable correctly for userqueues where vmid is not known
in the kernel.

Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index 0e6e2e2acf5b..127091de0f34 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -2136,12 +2136,14 @@ static int amdgpu_pt_info_read(struct seq_file *m, void *unused)
 	struct drm_file *file;
 	struct amdgpu_fpriv *fpriv;
 	struct amdgpu_bo *root_bo;
+	struct amdgpu_device *adev;
 	int r;
 
 	file = m->private;
 	if (!file)
 		return -EINVAL;
 
+	adev = drm_to_adev(file->minor->dev);
 	fpriv = file->driver_priv;
 	if (!fpriv || !fpriv->vm.root.bo)
 		return -ENODEV;
@@ -2154,6 +2156,10 @@ static int amdgpu_pt_info_read(struct seq_file *m, void *unused)
 	}
 
 	seq_printf(m, "gpu_address: 0x%llx\n", amdgpu_bo_gpu_offset(fpriv->vm.root.bo));
+	seq_printf(m, "max_pfn: 0x%llx\n", adev->vm_manager.max_pfn);
+	seq_printf(m, "num_level: 0x%x\n", adev->vm_manager.num_level);
+	seq_printf(m, "block_size: 0x%x\n", adev->vm_manager.block_size);
+	seq_printf(m, "fragment_size: 0x%x\n", adev->vm_manager.fragment_size);
 
 	amdgpu_bo_unreserve(root_bo);
 	amdgpu_bo_unref(&root_bo);

From 1e18746381793bef7c715fc5ec5611a422a75c4c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 27 Jun 2025 10:09:06 -0400
Subject: [PATCH 1009/1455] drm/amd: add more cyan skillfish PCI ids

Add additional PCI IDs to the cyan skillfish family.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index b42a2a32b0b7..ece251cbe8c3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -2172,6 +2172,11 @@ static const struct pci_device_id pciidlist[] = {
 	{0x1002, 0x7410, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_ALDEBARAN},
 
 	/* CYAN_SKILLFISH */
+	{0x1002, 0x13DB, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
+	{0x1002, 0x13F9, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
+	{0x1002, 0x13FA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
+	{0x1002, 0x13FB, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
+	{0x1002, 0x13FC, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
 	{0x1002, 0x13FE, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
 	{0x1002, 0x143F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYAN_SKILLFISH|AMD_IS_APU},
 

From 878f33f3907ae2e5751c1a3c4397f37b86e53494 Mon Sep 17 00:00:00 2001
From: Sunil Khatri <sunil.khatri@amd.com>
Date: Thu, 14 Aug 2025 13:29:04 +0530
Subject: [PATCH 1010/1455] drm/amdgpu: fix the formating for debugfs print
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the format of debugfs print in the mqd. Need to
add a colon so parser can parse it properly.

Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index f72de06a4ac8..467e8fa6cb8b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -364,7 +364,7 @@ static int amdgpu_mqd_info_read(struct seq_file *m, void *unused)
 		return -EINVAL;
 	}
 
-	seq_printf(m, "queue_type %d\n", queue->queue_type);
+	seq_printf(m, "queue_type: %d\n", queue->queue_type);
 	seq_printf(m, "mqd_gpu_address: 0x%llx\n", amdgpu_bo_gpu_offset(queue->mqd.obj));
 
 	amdgpu_bo_unreserve(bo);

From 0b05857dc11c97635e3cc80b231019307ec9d704 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Thu, 4 Sep 2025 12:57:50 -0700
Subject: [PATCH 1011/1455] drm/xe/guc: Clean up of GuC 'CTL' defines

All the field generation for the CTL defines (used for GuC init data)
were hand-rolled rather than using FIELD_PREP/REG_GENMASK/BIT macros.

Also, there were a bunch of macros defined for verbosity settings that
were never used.

So fix that all up.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250904195752.3846138-2-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/xe_guc.c      | 21 ++++++---------------
 drivers/gpu/drm/xe/xe_guc_fwif.h | 28 +++++++++-------------------
 2 files changed, 15 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index b3a6408a5760..e56c9c5c8845 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -74,8 +74,7 @@ static u32 guc_ctl_debug_flags(struct xe_guc *guc)
 	if (!GUC_LOG_LEVEL_IS_VERBOSE(level))
 		flags |= GUC_LOG_DISABLED;
 	else
-		flags |= GUC_LOG_LEVEL_TO_VERBOSITY(level) <<
-			 GUC_LOG_VERBOSITY_SHIFT;
+		flags |= FIELD_PREP(GUC_LOG_VERBOSITY, GUC_LOG_LEVEL_TO_VERBOSITY(level));
 
 	return flags;
 }
@@ -122,22 +121,14 @@ static u32 guc_ctl_log_params_flags(struct xe_guc *guc)
 	BUILD_BUG_ON(!CAPTURE_BUFFER_SIZE);
 	BUILD_BUG_ON(!IS_ALIGNED(CAPTURE_BUFFER_SIZE, CAPTURE_UNIT));
 
-	BUILD_BUG_ON((CRASH_BUFFER_SIZE / LOG_UNIT - 1) >
-			(GUC_LOG_CRASH_MASK >> GUC_LOG_CRASH_SHIFT));
-	BUILD_BUG_ON((DEBUG_BUFFER_SIZE / LOG_UNIT - 1) >
-			(GUC_LOG_DEBUG_MASK >> GUC_LOG_DEBUG_SHIFT));
-	BUILD_BUG_ON((CAPTURE_BUFFER_SIZE / CAPTURE_UNIT - 1) >
-			(GUC_LOG_CAPTURE_MASK >> GUC_LOG_CAPTURE_SHIFT));
-
 	flags = GUC_LOG_VALID |
 		GUC_LOG_NOTIFY_ON_HALF_FULL |
 		CAPTURE_FLAG |
 		LOG_FLAG |
-		((CRASH_BUFFER_SIZE / LOG_UNIT - 1) << GUC_LOG_CRASH_SHIFT) |
-		((DEBUG_BUFFER_SIZE / LOG_UNIT - 1) << GUC_LOG_DEBUG_SHIFT) |
-		((CAPTURE_BUFFER_SIZE / CAPTURE_UNIT - 1) <<
-		 GUC_LOG_CAPTURE_SHIFT) |
-		(offset << GUC_LOG_BUF_ADDR_SHIFT);
+		FIELD_PREP(GUC_LOG_CRASH, CRASH_BUFFER_SIZE / LOG_UNIT - 1) |
+		FIELD_PREP(GUC_LOG_DEBUG, DEBUG_BUFFER_SIZE / LOG_UNIT - 1) |
+		FIELD_PREP(GUC_LOG_CAPTURE, CAPTURE_BUFFER_SIZE / CAPTURE_UNIT - 1) |
+		FIELD_PREP(GUC_LOG_BUF_ADDR, offset);
 
 	#undef LOG_UNIT
 	#undef LOG_FLAG
@@ -150,7 +141,7 @@ static u32 guc_ctl_log_params_flags(struct xe_guc *guc)
 static u32 guc_ctl_ads_flags(struct xe_guc *guc)
 {
 	u32 ads = guc_bo_ggtt_addr(guc, guc->ads.bo) >> PAGE_SHIFT;
-	u32 flags = ads << GUC_ADS_ADDR_SHIFT;
+	u32 flags = FIELD_PREP(GUC_ADS_ADDR, ads);
 
 	return flags;
 }
diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index 0508f1064178..457d914097c8 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -89,13 +89,10 @@ struct guc_update_exec_queue_policy {
 #define   GUC_LOG_NOTIFY_ON_HALF_FULL	BIT(1)
 #define   GUC_LOG_CAPTURE_ALLOC_UNITS	BIT(2)
 #define   GUC_LOG_LOG_ALLOC_UNITS	BIT(3)
-#define   GUC_LOG_CRASH_SHIFT		4
-#define   GUC_LOG_CRASH_MASK		(0x3 << GUC_LOG_CRASH_SHIFT)
-#define   GUC_LOG_DEBUG_SHIFT		6
-#define   GUC_LOG_DEBUG_MASK	        (0xF << GUC_LOG_DEBUG_SHIFT)
-#define   GUC_LOG_CAPTURE_SHIFT		10
-#define   GUC_LOG_CAPTURE_MASK	        (0x3 << GUC_LOG_CAPTURE_SHIFT)
-#define   GUC_LOG_BUF_ADDR_SHIFT	12
+#define   GUC_LOG_CRASH			REG_GENMASK(5, 4)
+#define   GUC_LOG_DEBUG			REG_GENMASK(9, 6)
+#define   GUC_LOG_CAPTURE		REG_GENMASK(11, 10)
+#define   GUC_LOG_BUF_ADDR		REG_GENMASK(31, 12)
 
 #define GUC_CTL_WA			1
 #define   GUC_WA_GAM_CREDITS		BIT(10)
@@ -117,21 +114,14 @@ struct guc_update_exec_queue_policy {
 #define   GUC_CTL_DISABLE_SCHEDULER	BIT(14)
 
 #define GUC_CTL_DEBUG			3
-#define   GUC_LOG_VERBOSITY_SHIFT	0
-#define   GUC_LOG_VERBOSITY_LOW		(0 << GUC_LOG_VERBOSITY_SHIFT)
-#define   GUC_LOG_VERBOSITY_MED		(1 << GUC_LOG_VERBOSITY_SHIFT)
-#define   GUC_LOG_VERBOSITY_HIGH	(2 << GUC_LOG_VERBOSITY_SHIFT)
-#define   GUC_LOG_VERBOSITY_ULTRA	(3 << GUC_LOG_VERBOSITY_SHIFT)
-#define	  GUC_LOG_VERBOSITY_MIN		0
+#define   GUC_LOG_VERBOSITY		REG_GENMASK(1, 0)
 #define	  GUC_LOG_VERBOSITY_MAX		3
-#define	  GUC_LOG_VERBOSITY_MASK	0x0000000f
-#define	  GUC_LOG_DESTINATION_MASK	(3 << 4)
-#define   GUC_LOG_DISABLED		(1 << 6)
-#define   GUC_PROFILE_ENABLED		(1 << 7)
+#define	  GUC_LOG_DESTINATION		REG_GENMASK(5, 4)
+#define   GUC_LOG_DISABLED		BIT(6)
+#define   GUC_PROFILE_ENABLED		BIT(7)
 
 #define GUC_CTL_ADS			4
-#define   GUC_ADS_ADDR_SHIFT		1
-#define   GUC_ADS_ADDR_MASK		(0xFFFFF << GUC_ADS_ADDR_SHIFT)
+#define   GUC_ADS_ADDR			REG_GENMASK(21, 1)
 
 #define GUC_CTL_DEVID			5
 

From cf423b928f27f5228c7942c1b270988c87b2f25b Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Thu, 4 Sep 2025 12:57:51 -0700
Subject: [PATCH 1012/1455] drm/xe/guc: Fix badly worded error message

If a GuC id lookup failed, the error message was 'Not engine present',
which is bad in multiple ways - incorrect English and 'engines' are
now called 'exec queues' in this context. So fix it.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250904195752.3846138-3-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/xe_guc_submit.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 1185b23b1384..897f33ded8ec 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2029,7 +2029,7 @@ g2h_exec_queue_lookup(struct xe_guc *guc, u32 guc_id)
 
 	q = xa_load(&guc->submission_state.exec_queue_lookup, guc_id);
 	if (unlikely(!q)) {
-		xe_gt_err(gt, "Not engine present for guc_id %u\n", guc_id);
+		xe_gt_err(gt, "No exec queue found for guc_id %u\n", guc_id);
 		return NULL;
 	}
 

From 2318336573c51748bd2089a835242c70be6850cf Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Tue, 5 Aug 2025 11:26:15 +0800
Subject: [PATCH 1013/1455] drm/amdgpu: Add preempt and restore callbacks to
 userq funcs

Add two new function pointers to struct amdgpu_userq_funcs:
- preempt: To handle preemption of user mode queues
- restore: To restore preempted user mode queues

These callbacks will allow the driver to properly manage queue
preemption and restoration when needed, such as during context
switching or priority changes.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
index b1ca91b7cda4..017efd2075df 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
@@ -78,6 +78,10 @@ struct amdgpu_userq_funcs {
 		     struct amdgpu_usermode_queue *queue);
 	int (*map)(struct amdgpu_userq_mgr *uq_mgr,
 		   struct amdgpu_usermode_queue *queue);
+	int (*preempt)(struct amdgpu_userq_mgr *uq_mgr,
+		   struct amdgpu_usermode_queue *queue);
+	int (*restore)(struct amdgpu_userq_mgr *uq_mgr,
+		   struct amdgpu_usermode_queue *queue);
 };
 
 /* Usermode queues for gfx */

From 6abd725fdfc418abdf3aae0a1b19cb483305c723 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Sun, 3 Aug 2025 21:21:42 +0800
Subject: [PATCH 1014/1455] drm/amd/amdgpu: Implement MES suspend/resume gang
 functionality for v12

This commit implements the actual MES (Micro Engine Scheduler) suspend
and resume gang operations for version 12 hardware. Previously these
functions were just stubs returning success.

v2: Always use AMDGPU_MES_SCHED_PIPE

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_v12_0.c | 32 ++++++++++++++++++++++++--
 1 file changed, 30 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index ff5df28b57ec..0075bb2ed66c 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -568,13 +568,41 @@ static int mes_v12_0_unmap_legacy_queue(struct amdgpu_mes *mes,
 static int mes_v12_0_suspend_gang(struct amdgpu_mes *mes,
 				  struct mes_suspend_gang_input *input)
 {
-	return 0;
+	union MESAPI__SUSPEND mes_suspend_gang_pkt;
+
+	memset(&mes_suspend_gang_pkt, 0, sizeof(mes_suspend_gang_pkt));
+
+	mes_suspend_gang_pkt.header.type = MES_API_TYPE_SCHEDULER;
+	mes_suspend_gang_pkt.header.opcode = MES_SCH_API_SUSPEND;
+	mes_suspend_gang_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
+
+	mes_suspend_gang_pkt.suspend_all_gangs = input->suspend_all_gangs;
+	mes_suspend_gang_pkt.gang_context_addr = input->gang_context_addr;
+	mes_suspend_gang_pkt.suspend_fence_addr = input->suspend_fence_addr;
+	mes_suspend_gang_pkt.suspend_fence_value = input->suspend_fence_value;
+
+	return mes_v12_0_submit_pkt_and_poll_completion(mes, AMDGPU_MES_SCHED_PIPE,
+			&mes_suspend_gang_pkt, sizeof(mes_suspend_gang_pkt),
+			offsetof(union MESAPI__SUSPEND, api_status));
 }
 
 static int mes_v12_0_resume_gang(struct amdgpu_mes *mes,
 				 struct mes_resume_gang_input *input)
 {
-	return 0;
+	union MESAPI__RESUME mes_resume_gang_pkt;
+
+	memset(&mes_resume_gang_pkt, 0, sizeof(mes_resume_gang_pkt));
+
+	mes_resume_gang_pkt.header.type = MES_API_TYPE_SCHEDULER;
+	mes_resume_gang_pkt.header.opcode = MES_SCH_API_RESUME;
+	mes_resume_gang_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
+
+	mes_resume_gang_pkt.resume_all_gangs = input->resume_all_gangs;
+	mes_resume_gang_pkt.gang_context_addr = input->gang_context_addr;
+
+	return mes_v12_0_submit_pkt_and_poll_completion(mes, AMDGPU_MES_SCHED_PIPE,
+			&mes_resume_gang_pkt, sizeof(mes_resume_gang_pkt),
+			offsetof(union MESAPI__RESUME, api_status));
 }
 
 static int mes_v12_0_query_sched_status(struct amdgpu_mes *mes, int pipe)

From 78e1222fbf0026de99420bc35c6c12931e0f36ee Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Thu, 4 Sep 2025 09:39:34 +0800
Subject: [PATCH 1015/1455] drm/amdgpu/mes: add front end for detect and reset
 hung queue

Helper function to detect and reset hung queues.  MES will
return an array of doorbell indices of which queues are hung
and were optionally reset.

v2:  Clear the doorbell array before detection

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 65 +++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h | 21 ++++++++
 2 files changed, 86 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index 135598502c8d..5bf9be073cdd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -191,6 +191,20 @@ int amdgpu_mes_init(struct amdgpu_device *adev)
 	if (r)
 		goto error_doorbell;
 
+	if (adev->mes.hung_queue_db_array_size) {
+		r = amdgpu_bo_create_kernel(adev,
+					    adev->mes.hung_queue_db_array_size * sizeof(u32),
+					    PAGE_SIZE,
+					    AMDGPU_GEM_DOMAIN_GTT,
+					    &adev->mes.hung_queue_db_array_gpu_obj,
+					    &adev->mes.hung_queue_db_array_gpu_addr,
+					    &adev->mes.hung_queue_db_array_cpu_addr);
+		if (r) {
+			dev_warn(adev->dev, "failed to create MES hung db array buffer (%d)", r);
+			goto error_doorbell;
+		}
+	}
+
 	return 0;
 
 error_doorbell:
@@ -216,6 +230,10 @@ void amdgpu_mes_fini(struct amdgpu_device *adev)
 {
 	int i;
 
+	amdgpu_bo_free_kernel(&adev->mes.hung_queue_db_array_gpu_obj,
+			      &adev->mes.hung_queue_db_array_gpu_addr,
+			      &adev->mes.hung_queue_db_array_cpu_addr);
+
 	amdgpu_bo_free_kernel(&adev->mes.event_log_gpu_obj,
 			      &adev->mes.event_log_gpu_addr,
 			      &adev->mes.event_log_cpu_addr);
@@ -366,6 +384,53 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
 	return r;
 }
 
+int amdgpu_mes_get_hung_queue_db_array_size(struct amdgpu_device *adev)
+{
+	return adev->mes.hung_queue_db_array_size;
+}
+
+int amdgpu_mes_detect_and_reset_hung_queues(struct amdgpu_device *adev,
+					    int queue_type,
+					    bool detect_only,
+					    unsigned int *hung_db_num,
+					    u32 *hung_db_array)
+
+{
+	struct mes_detect_and_reset_queue_input input;
+	u32 *db_array = adev->mes.hung_queue_db_array_cpu_addr;
+	int r, i;
+
+	if (!hung_db_num || !hung_db_array)
+		return -EINVAL;
+
+	if ((queue_type != AMDGPU_RING_TYPE_GFX) &&
+	    (queue_type != AMDGPU_RING_TYPE_COMPUTE) &&
+	    (queue_type != AMDGPU_RING_TYPE_SDMA))
+		return -EINVAL;
+
+	/* Clear the doorbell array before detection */
+	memset(adev->mes.hung_queue_db_array_cpu_addr, 0,
+		adev->mes.hung_queue_db_array_size * sizeof(u32));
+	input.queue_type = queue_type;
+	input.detect_only = detect_only;
+
+	r = adev->mes.funcs->detect_and_reset_hung_queues(&adev->mes,
+							  &input);
+	if (r) {
+		dev_err(adev->dev, "failed to detect and reset\n");
+	} else {
+		*hung_db_num = 0;
+		for (i = 0; i < adev->mes.hung_queue_db_array_size; i++) {
+			if (db_array[i] != AMDGPU_MES_INVALID_DB_OFFSET) {
+				hung_db_array[i] = db_array[i];
+				*hung_db_num += 1;
+			}
+		}
+	}
+
+	return r;
+}
+
 uint32_t amdgpu_mes_rreg(struct amdgpu_device *adev, uint32_t reg)
 {
 	struct mes_misc_op_input op_input;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
index 489a4a0f0610..6b506fc72f58 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
@@ -41,6 +41,7 @@
 #define AMDGPU_MES_API_VERSION_MASK	0x00fff000
 #define AMDGPU_MES_FEAT_VERSION_MASK	0xff000000
 #define AMDGPU_MES_MSCRATCH_SIZE	0x40000
+#define AMDGPU_MES_INVALID_DB_OFFSET	0xffffffff
 
 enum amdgpu_mes_priority_level {
 	AMDGPU_MES_PRIORITY_LEVEL_LOW       = 0,
@@ -147,6 +148,10 @@ struct amdgpu_mes {
 	uint64_t            resource_1_gpu_addr[AMDGPU_MAX_MES_PIPES];
 	void                *resource_1_addr[AMDGPU_MAX_MES_PIPES];
 
+	int				hung_queue_db_array_size;
+	struct amdgpu_bo		*hung_queue_db_array_gpu_obj;
+	uint64_t			hung_queue_db_array_gpu_addr;
+	void				*hung_queue_db_array_cpu_addr;
 };
 
 struct amdgpu_mes_gang {
@@ -280,6 +285,11 @@ struct mes_reset_queue_input {
 	bool                               is_kq;
 };
 
+struct mes_detect_and_reset_queue_input {
+	uint32_t                           queue_type;
+	bool                               detect_only;
+};
+
 struct mes_inv_tlbs_pasid_input {
 	uint32_t        xcc_id;
 	uint16_t        pasid;
@@ -375,6 +385,10 @@ struct amdgpu_mes_funcs {
 	int (*reset_hw_queue)(struct amdgpu_mes *mes,
 			      struct mes_reset_queue_input *input);
 
+	int (*detect_and_reset_hung_queues)(struct amdgpu_mes *mes,
+			      struct mes_detect_and_reset_queue_input *input);
+
+
 	int (*invalidate_tlbs_pasid)(struct amdgpu_mes *mes,
 			      struct mes_inv_tlbs_pasid_input *input);
 };
@@ -400,6 +414,13 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev,
 				  unsigned int vmid,
 				  bool use_mmio);
 
+int amdgpu_mes_get_hung_queue_db_array_size(struct amdgpu_device *adev);
+int amdgpu_mes_detect_and_reset_hung_queues(struct amdgpu_device *adev,
+					    int queue_type,
+					    bool detect_only,
+					    unsigned int *hung_db_num,
+					    u32 *hung_db_array);
+
 uint32_t amdgpu_mes_rreg(struct amdgpu_device *adev, uint32_t reg);
 int amdgpu_mes_wreg(struct amdgpu_device *adev,
 		    uint32_t reg, uint32_t val);

From b28cfc8305f7f58d261d558ea699be14c03537fd Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Thu, 4 Sep 2025 09:44:51 +0800
Subject: [PATCH 1016/1455] drm/amdgpu/mes11: implement detect and reset
 callback

Implement support for the hung queue detect and reset
functionality.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 31 ++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 3f6a828cad8a..3b91ea601add 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -66,6 +66,8 @@ static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev);
 #define GFX_MES_DRAM_SIZE	0x80000
 #define MES11_HW_RESOURCE_1_SIZE (128 * AMDGPU_GPU_PAGE_SIZE)
 
+#define MES11_HUNG_DB_OFFSET_ARRAY_SIZE 4
+
 static void mes_v11_0_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -784,6 +786,32 @@ static int mes_v11_0_reset_hw_queue(struct amdgpu_mes *mes,
 			offsetof(union MESAPI__RESET, api_status));
 }
 
+static int mes_v11_0_detect_and_reset_hung_queues(struct amdgpu_mes *mes,
+						  struct mes_detect_and_reset_queue_input *input)
+{
+	union MESAPI__RESET mes_reset_queue_pkt;
+
+	memset(&mes_reset_queue_pkt, 0, sizeof(mes_reset_queue_pkt));
+
+	mes_reset_queue_pkt.header.type = MES_API_TYPE_SCHEDULER;
+	mes_reset_queue_pkt.header.opcode = MES_SCH_API_RESET;
+	mes_reset_queue_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
+
+	mes_reset_queue_pkt.queue_type =
+		convert_to_mes_queue_type(input->queue_type);
+	mes_reset_queue_pkt.doorbell_offset_addr =
+		mes->hung_queue_db_array_gpu_addr;
+
+	if (input->detect_only)
+		mes_reset_queue_pkt.hang_detect_only = 1;
+	else
+		mes_reset_queue_pkt.hang_detect_then_reset = 1;
+
+	return mes_v11_0_submit_pkt_and_poll_completion(mes,
+			&mes_reset_queue_pkt, sizeof(mes_reset_queue_pkt),
+			offsetof(union MESAPI__RESET, api_status));
+}
+
 static const struct amdgpu_mes_funcs mes_v11_0_funcs = {
 	.add_hw_queue = mes_v11_0_add_hw_queue,
 	.remove_hw_queue = mes_v11_0_remove_hw_queue,
@@ -793,6 +821,7 @@ static const struct amdgpu_mes_funcs mes_v11_0_funcs = {
 	.resume_gang = mes_v11_0_resume_gang,
 	.misc_op = mes_v11_0_misc_op,
 	.reset_hw_queue = mes_v11_0_reset_hw_queue,
+	.detect_and_reset_hung_queues = mes_v11_0_detect_and_reset_hung_queues,
 };
 
 static int mes_v11_0_allocate_ucode_buffer(struct amdgpu_device *adev,
@@ -1685,6 +1714,8 @@ static int mes_v11_0_early_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	int pipe, r;
 
+	adev->mes.hung_queue_db_array_size =
+		MES11_HUNG_DB_OFFSET_ARRAY_SIZE;
 	for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) {
 		if (!adev->enable_mes_kiq && pipe == AMDGPU_MES_KIQ_PIPE)
 			continue;

From 724471254ee09949bd99358494e8f38d016616c1 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Thu, 4 Sep 2025 09:50:00 +0800
Subject: [PATCH 1017/1455] drm/amdgpu/mes12: implement detect and reset
 callback

Implement support for the hung queue detect and reset
functionality.

v2: Always use AMDGPU_MES_SCHED_PIPE

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_v12_0.c | 31 ++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index 0075bb2ed66c..998893dff08e 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -47,6 +47,8 @@ static int mes_v12_0_kiq_hw_fini(struct amdgpu_device *adev);
 
 #define MES_EOP_SIZE   2048
 
+#define MES12_HUNG_DB_OFFSET_ARRAY_SIZE 4
+
 static void mes_v12_0_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
@@ -908,6 +910,32 @@ static int mes_v12_0_reset_hw_queue(struct amdgpu_mes *mes,
 			offsetof(union MESAPI__RESET, api_status));
 }
 
+static int mes_v12_0_detect_and_reset_hung_queues(struct amdgpu_mes *mes,
+						  struct mes_detect_and_reset_queue_input *input)
+{
+	union MESAPI__RESET mes_reset_queue_pkt;
+
+	memset(&mes_reset_queue_pkt, 0, sizeof(mes_reset_queue_pkt));
+
+	mes_reset_queue_pkt.header.type = MES_API_TYPE_SCHEDULER;
+	mes_reset_queue_pkt.header.opcode = MES_SCH_API_RESET;
+	mes_reset_queue_pkt.header.dwsize = API_FRAME_SIZE_IN_DWORDS;
+
+	mes_reset_queue_pkt.queue_type =
+		convert_to_mes_queue_type(input->queue_type);
+	mes_reset_queue_pkt.doorbell_offset_addr =
+		mes->hung_queue_db_array_gpu_addr;
+
+	if (input->detect_only)
+		mes_reset_queue_pkt.hang_detect_only = 1;
+	else
+		mes_reset_queue_pkt.hang_detect_then_reset = 1;
+
+	return mes_v12_0_submit_pkt_and_poll_completion(mes, AMDGPU_MES_SCHED_PIPE,
+			&mes_reset_queue_pkt, sizeof(mes_reset_queue_pkt),
+			offsetof(union MESAPI__RESET, api_status));
+}
+
 static int mes_v12_inv_tlb_convert_hub_id(uint8_t id)
 {
 	/*
@@ -960,6 +988,7 @@ static const struct amdgpu_mes_funcs mes_v12_0_funcs = {
 	.misc_op = mes_v12_0_misc_op,
 	.reset_hw_queue = mes_v12_0_reset_hw_queue,
 	.invalidate_tlbs_pasid = mes_v12_0_inv_tlbs_pasid,
+	.detect_and_reset_hung_queues = mes_v12_0_detect_and_reset_hung_queues,
 };
 
 static int mes_v12_0_allocate_ucode_buffer(struct amdgpu_device *adev,
@@ -1865,6 +1894,8 @@ static int mes_v12_0_early_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 	int pipe, r;
 
+	adev->mes.hung_queue_db_array_size =
+		MES12_HUNG_DB_OFFSET_ARRAY_SIZE;
 	for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) {
 		r = amdgpu_mes_init_microcode(adev, pipe);
 		if (r)

From c5da9e9c023893ece97855097240247a590c0d16 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 16 Apr 2025 13:12:40 -0400
Subject: [PATCH 1018/1455] drm/amdgpu: add user queue reset source
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Track resets from user queues.

Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Sunil Khatri <sunil.khatri@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c | 3 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h | 1 +
 2 files changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
index dabfbdf6f1ce..28c4ad62f50e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c
@@ -340,6 +340,9 @@ void amdgpu_reset_get_desc(struct amdgpu_reset_context *rst_ctxt, char *buf,
 	case AMDGPU_RESET_SRC_USER:
 		strscpy(buf, "user trigger", len);
 		break;
+	case AMDGPU_RESET_SRC_USERQ:
+		strscpy(buf, "user queue trigger", len);
+		break;
 	default:
 		strscpy(buf, "unknown", len);
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
index 2f92b3be40f5..07b4d37f1db6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h
@@ -43,6 +43,7 @@ enum AMDGPU_RESET_SRCS {
 	AMDGPU_RESET_SRC_MES,
 	AMDGPU_RESET_SRC_HWS,
 	AMDGPU_RESET_SRC_USER,
+	AMDGPU_RESET_SRC_USERQ,
 };
 
 struct amdgpu_reset_context {

From 276e8beb2a0c390fdac8f11bb58d9fdf95a6d538 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 18 Apr 2025 11:35:49 -0400
Subject: [PATCH 1019/1455] drm/amdgpu/userq: add force completion helpers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add support for forcing completion of userq fences.
This is needed for userq resets and asic resets so that we
can set the error on the fence and force completion.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_userq_fence.c   | 42 +++++++++++++++++++
 .../gpu/drm/amd/amdgpu/amdgpu_userq_fence.h   |  1 +
 2 files changed, 43 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
index c2a983ff23c9..95e91d1dc58a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
@@ -67,6 +67,14 @@ static u64 amdgpu_userq_fence_read(struct amdgpu_userq_fence_driver *fence_drv)
 	return le64_to_cpu(*fence_drv->cpu_addr);
 }
 
+static void
+amdgpu_userq_fence_write(struct amdgpu_userq_fence_driver *fence_drv,
+			 u64 seq)
+{
+	if (fence_drv->cpu_addr)
+		*fence_drv->cpu_addr = cpu_to_le64(seq);
+}
+
 int amdgpu_userq_fence_driver_alloc(struct amdgpu_device *adev,
 				    struct amdgpu_usermode_queue *userq)
 {
@@ -408,6 +416,40 @@ static void amdgpu_userq_fence_cleanup(struct dma_fence *fence)
 	dma_fence_put(fence);
 }
 
+static void
+amdgpu_userq_fence_driver_set_error(struct amdgpu_userq_fence *fence,
+				    int error)
+{
+	struct amdgpu_userq_fence_driver *fence_drv = fence->fence_drv;
+	unsigned long flags;
+	struct dma_fence *f;
+
+	spin_lock_irqsave(&fence_drv->fence_list_lock, flags);
+
+	f = rcu_dereference_protected(&fence->base,
+				      lockdep_is_held(&fence_drv->fence_list_lock));
+	if (f && !dma_fence_is_signaled_locked(f))
+		dma_fence_set_error(f, error);
+	spin_unlock_irqrestore(&fence_drv->fence_list_lock, flags);
+}
+
+void
+amdgpu_userq_fence_driver_force_completion(struct amdgpu_usermode_queue *userq)
+{
+	struct dma_fence *f = userq->last_fence;
+
+	if (f) {
+		struct amdgpu_userq_fence *fence = to_amdgpu_userq_fence(f);
+		struct amdgpu_userq_fence_driver *fence_drv = fence->fence_drv;
+		u64 wptr = fence->base.seqno;
+
+		amdgpu_userq_fence_driver_set_error(fence, -ECANCELED);
+		amdgpu_userq_fence_write(fence_drv, wptr);
+		amdgpu_userq_fence_driver_process(fence_drv);
+
+	}
+}
+
 int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *filp)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h
index 97a125ab8a78..d76add2afc77 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.h
@@ -67,6 +67,7 @@ int amdgpu_userq_fence_driver_alloc(struct amdgpu_device *adev,
 				    struct amdgpu_usermode_queue *userq);
 void amdgpu_userq_fence_driver_free(struct amdgpu_usermode_queue *userq);
 void amdgpu_userq_fence_driver_process(struct amdgpu_userq_fence_driver *fence_drv);
+void amdgpu_userq_fence_driver_force_completion(struct amdgpu_usermode_queue *userq);
 void amdgpu_userq_fence_driver_destroy(struct kref *ref);
 int amdgpu_userq_signal_ioctl(struct drm_device *dev, void *data,
 			      struct drm_file *filp);

From 89923fb7ead4fdd37b78dd49962d9bb5892403e6 Mon Sep 17 00:00:00 2001
From: Geoffrey McRae <geoffrey.mcrae@amd.com>
Date: Thu, 28 Aug 2025 22:26:22 +1000
Subject: [PATCH 1020/1455] drm/amd/display: remove oem i2c adapter on finish

Fixes a bug where unbinding of the GPU would leave the oem i2c adapter
registered resulting in a null pointer dereference when applications try
to access the invalid device.

Fixes: 3d5470c97314 ("drm/amd/display/dm: add support for OEM i2c bus")
Cc: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Geoffrey McRae <geoffrey.mcrae@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 242f98564261..dc4d04a8ce6b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -2957,6 +2957,17 @@ static int dm_oem_i2c_hw_init(struct amdgpu_device *adev)
 	return 0;
 }
 
+static void dm_oem_i2c_hw_fini(struct amdgpu_device *adev)
+{
+	struct amdgpu_display_manager *dm = &adev->dm;
+
+	if (dm->oem_i2c) {
+		i2c_del_adapter(&dm->oem_i2c->base);
+		kfree(dm->oem_i2c);
+		dm->oem_i2c = NULL;
+	}
+}
+
 /**
  * dm_hw_init() - Initialize DC device
  * @ip_block: Pointer to the amdgpu_ip_block for this hw instance.
@@ -3007,7 +3018,7 @@ static int dm_hw_fini(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 
-	kfree(adev->dm.oem_i2c);
+	dm_oem_i2c_hw_fini(adev);
 
 	amdgpu_dm_hpd_fini(adev);
 

From fa7c99f04f6dd299388e9282812b14e95558ac8e Mon Sep 17 00:00:00 2001
From: Yifan Zhang <yifan1.zhang@amd.com>
Date: Wed, 20 Aug 2025 16:10:51 +0800
Subject: [PATCH 1021/1455] amd/amdkfd: correct mem limit calculation for small
 APUs

Current mem limit check leaks some GTT memory (reserved_for_pt
reserved_for_ras + adev->vram_pin_size) for small APUs.

Since carveout VRAM is tunable on APUs, there are three case
regarding the carveout VRAM size relative to GTT:

1. 0 < carveout < gtt
   apu_prefer_gtt = true, is_app_apu = false

2. carveout > gtt / 2
   apu_prefer_gtt = false, is_app_apu = false

3. 0 = carveout
   apu_prefer_gtt = true, is_app_apu = true

It doesn't make sense to check below limitation in case 1
(default case, small carveout) because the values in the below
expression are mixed with carveout and gtt.

adev->kfd.vram_used[xcp_id] + vram_needed >
    vram_size - reserved_for_pt - reserved_for_ras -
    atomic64_read(&adev->vram_pin_size)

gtt: kfd.vram_used, vram_needed, vram_size
carveout: reserved_for_pt, reserved_for_ras, adev->vram_pin_size

In case 1, vram allocation will go to gtt domain, skip vram check
since ttm_mem_limit check already cover this allocation.

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  | 44 ++++++++++++++-----
 1 file changed, 32 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index d478acb4568a..c3b34a410375 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -213,19 +213,35 @@ int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
 	spin_lock(&kfd_mem_limit.mem_limit_lock);
 
 	if (kfd_mem_limit.system_mem_used + system_mem_needed >
-	    kfd_mem_limit.max_system_mem_limit)
+	    kfd_mem_limit.max_system_mem_limit) {
 		pr_debug("Set no_system_mem_limit=1 if using shared memory\n");
+		if (!no_system_mem_limit) {
+			ret = -ENOMEM;
+			goto release;
+		}
+	}
 
-	if ((kfd_mem_limit.system_mem_used + system_mem_needed >
-	     kfd_mem_limit.max_system_mem_limit && !no_system_mem_limit) ||
-	    (kfd_mem_limit.ttm_mem_used + ttm_mem_needed >
-	     kfd_mem_limit.max_ttm_mem_limit) ||
-	    (adev && xcp_id >= 0 && adev->kfd.vram_used[xcp_id] + vram_needed >
-	     vram_size - reserved_for_pt - reserved_for_ras - atomic64_read(&adev->vram_pin_size))) {
+	if (kfd_mem_limit.ttm_mem_used + ttm_mem_needed >
+		kfd_mem_limit.max_ttm_mem_limit) {
 		ret = -ENOMEM;
 		goto release;
 	}
 
+	/*if is_app_apu is false and apu_prefer_gtt is true, it is an APU with
+	 * carve out < gtt. In that case, VRAM allocation will go to gtt domain, skip
+	 * VRAM check since ttm_mem_limit check already cover this allocation
+	 */
+
+	if (adev && xcp_id >= 0 && (!adev->apu_prefer_gtt || adev->gmc.is_app_apu)) {
+		uint64_t vram_available =
+			vram_size - reserved_for_pt - reserved_for_ras -
+			atomic64_read(&adev->vram_pin_size);
+		if (adev->kfd.vram_used[xcp_id] + vram_needed > vram_available) {
+			ret = -ENOMEM;
+			goto release;
+		}
+	}
+
 	/* Update memory accounting by decreasing available system
 	 * memory, TTM memory and GPU memory as computed above
 	 */
@@ -1627,11 +1643,15 @@ size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev,
 	uint64_t vram_available, system_mem_available, ttm_mem_available;
 
 	spin_lock(&kfd_mem_limit.mem_limit_lock);
-	vram_available = KFD_XCP_MEMORY_SIZE(adev, xcp_id)
-		- adev->kfd.vram_used_aligned[xcp_id]
-		- atomic64_read(&adev->vram_pin_size)
-		- reserved_for_pt
-		- reserved_for_ras;
+	if (adev->apu_prefer_gtt && !adev->gmc.is_app_apu)
+		vram_available = KFD_XCP_MEMORY_SIZE(adev, xcp_id)
+			- adev->kfd.vram_used_aligned[xcp_id];
+	else
+		vram_available = KFD_XCP_MEMORY_SIZE(adev, xcp_id)
+			- adev->kfd.vram_used_aligned[xcp_id]
+			- atomic64_read(&adev->vram_pin_size)
+			- reserved_for_pt
+			- reserved_for_ras;
 
 	if (adev->apu_prefer_gtt) {
 		system_mem_available = no_system_mem_limit ?

From 38ab33dbea594700c8d6cc81eec0a54e95d3eb2f Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sun, 31 Aug 2025 15:29:56 +0530
Subject: [PATCH 1022/1455] drm/amdgpu: Fix function header names in
 amdgpu_connectors.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Align the function headers for `amdgpu_max_hdmi_pixel_clock` and
`amdgpu_connector_dvi_mode_valid` with the function implementations so
they match the expected kdoc style.

Fixes the below:
drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c:1199: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Returns the maximum supported HDMI (TMDS) pixel clock in KHz.
drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c:1212: warning: This comment starts with '/**', but isn't a kernel-doc comment. Refer Documentation/doc-guide/kernel-doc.rst
 * Validates the given display mode on DVI and HDMI connectors.

Fixes: 585b2f685c56 ("drm/amdgpu: Respect max pixel clock for HDMI and DVI-D (v2)")
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index a381de8648e5..bf38fc69c1cf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -1196,7 +1196,10 @@ static void amdgpu_connector_dvi_force(struct drm_connector *connector)
 }
 
 /**
- * Returns the maximum supported HDMI (TMDS) pixel clock in KHz.
+ * amdgpu_max_hdmi_pixel_clock - Return max supported HDMI (TMDS) pixel clock
+ * @adev: pointer to amdgpu_device
+ *
+ * Return: maximum supported HDMI (TMDS) pixel clock in KHz.
  */
 static int amdgpu_max_hdmi_pixel_clock(const struct amdgpu_device *adev)
 {
@@ -1209,8 +1212,14 @@ static int amdgpu_max_hdmi_pixel_clock(const struct amdgpu_device *adev)
 }
 
 /**
- * Validates the given display mode on DVI and HDMI connectors,
- * including analog signals on DVI-I.
+ * amdgpu_connector_dvi_mode_valid - Validate a mode on DVI/HDMI connectors
+ * @connector: DRM connector to validate the mode on
+ * @mode: display mode to validate
+ *
+ * Validate the given display mode on DVI and HDMI connectors, including
+ * analog signals on DVI-I.
+ *
+ * Return: drm_mode_status indicating whether the mode is valid.
  */
 static enum drm_mode_status amdgpu_connector_dvi_mode_valid(struct drm_connector *connector,
 					    const struct drm_display_mode *mode)

From e8529dbc75cab56fc3c57830d0fd48cbd8911e6c Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 27 Jun 2025 10:12:36 -0400
Subject: [PATCH 1023/1455] drm/amdgpu: add ip offset support for cyan
 skillfish

For chips that don't have IP discovery tables.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile           |  3 +-
 .../drm/amd/amdgpu/cyan_skillfish_reg_init.c  | 56 +++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/nv.h               |  1 +
 3 files changed, 59 insertions(+), 1 deletion(-)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/cyan_skillfish_reg_init.c

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 930de203d533..2d0fea87af79 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -84,7 +84,8 @@ amdgpu-y += \
 	vega20_reg_init.o nbio_v7_4.o nbio_v2_3.o nv.o arct_reg_init.o mxgpu_nv.o \
 	nbio_v7_2.o hdp_v4_0.o hdp_v5_0.o aldebaran_reg_init.o aldebaran.o soc21.o soc24.o \
 	sienna_cichlid.o smu_v13_0_10.o nbio_v4_3.o hdp_v6_0.o nbio_v7_7.o hdp_v5_2.o lsdma_v6_0.o \
-	nbio_v7_9.o aqua_vanjaram.o nbio_v7_11.o lsdma_v7_0.o hdp_v7_0.o nbif_v6_3_1.o
+	nbio_v7_9.o aqua_vanjaram.o nbio_v7_11.o lsdma_v7_0.o hdp_v7_0.o nbif_v6_3_1.o \
+	cyan_skillfish_reg_init.o
 
 # add DF block
 amdgpu-y += \
diff --git a/drivers/gpu/drm/amd/amdgpu/cyan_skillfish_reg_init.c b/drivers/gpu/drm/amd/amdgpu/cyan_skillfish_reg_init.c
new file mode 100644
index 000000000000..96616a865aac
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/cyan_skillfish_reg_init.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include "amdgpu.h"
+#include "nv.h"
+
+#include "soc15_common.h"
+#include "soc15_hw_ip.h"
+#include "cyan_skillfish_ip_offset.h"
+
+int cyan_skillfish_reg_base_init(struct amdgpu_device *adev)
+{
+	/* HW has more IP blocks,  only initialized the blocke needed by driver */
+	uint32_t i;
+
+	adev->gfx.xcc_mask = 1;
+	for (i = 0 ; i < MAX_INSTANCE ; ++i) {
+		adev->reg_offset[GC_HWIP][i] = (uint32_t *)(&(GC_BASE.instance[i]));
+		adev->reg_offset[HDP_HWIP][i] = (uint32_t *)(&(HDP_BASE.instance[i]));
+		adev->reg_offset[MMHUB_HWIP][i] = (uint32_t *)(&(MMHUB_BASE.instance[i]));
+		adev->reg_offset[ATHUB_HWIP][i] = (uint32_t *)(&(ATHUB_BASE.instance[i]));
+		adev->reg_offset[NBIO_HWIP][i] = (uint32_t *)(&(NBIO_BASE.instance[i]));
+		adev->reg_offset[MP0_HWIP][i] = (uint32_t *)(&(MP0_BASE.instance[i]));
+		adev->reg_offset[MP1_HWIP][i] = (uint32_t *)(&(MP1_BASE.instance[i]));
+		adev->reg_offset[VCN_HWIP][i] = (uint32_t *)(&(UVD0_BASE.instance[i]));
+		adev->reg_offset[DF_HWIP][i] = (uint32_t *)(&(DF_BASE.instance[i]));
+		adev->reg_offset[DCE_HWIP][i] = (uint32_t *)(&(DMU_BASE.instance[i]));
+		adev->reg_offset[OSSSYS_HWIP][i] = (uint32_t *)(&(OSSSYS_BASE.instance[i]));
+		adev->reg_offset[SDMA0_HWIP][i] = (uint32_t *)(&(GC_BASE.instance[i]));
+		adev->reg_offset[SDMA1_HWIP][i] = (uint32_t *)(&(GC_BASE.instance[i]));
+		adev->reg_offset[SMUIO_HWIP][i] = (uint32_t *)(&(SMUIO_BASE.instance[i]));
+		adev->reg_offset[THM_HWIP][i] = (uint32_t *)(&(THM_BASE.instance[i]));
+		adev->reg_offset[CLK_HWIP][i] = (uint32_t *)(&(CLK_BASE.instance[i]));
+	}
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.h b/drivers/gpu/drm/amd/amdgpu/nv.h
index 83e9782aef39..8f4817404f10 100644
--- a/drivers/gpu/drm/amd/amdgpu/nv.h
+++ b/drivers/gpu/drm/amd/amdgpu/nv.h
@@ -31,5 +31,6 @@ extern const struct amdgpu_ip_block_version nv_common_ip_block;
 void nv_grbm_select(struct amdgpu_device *adev,
 		    u32 me, u32 pipe, u32 queue, u32 vmid);
 void nv_set_virt_ops(struct amdgpu_device *adev);
+int cyan_skillfish_reg_base_init(struct amdgpu_device *adev);
 
 #endif

From 9e6a5cf1a23bf575e93544ae05585659063b1c18 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 27 Jun 2025 10:18:46 -0400
Subject: [PATCH 1024/1455] drm/amdgpu: add support for cyan skillfish without
 IP discovery

For platforms without an IP discovery table.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 30 +++++++++++++++++++
 1 file changed, 30 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index efe0058b48ca..38c4ebc063db 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -2746,6 +2746,36 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev)
 		adev->ip_versions[UVD_HWIP][1] = IP_VERSION(2, 6, 0);
 		adev->ip_versions[XGMI_HWIP][0] = IP_VERSION(6, 1, 0);
 		break;
+	case CHIP_CYAN_SKILLFISH:
+		if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2) {
+			r = amdgpu_discovery_reg_base_init(adev);
+			if (r)
+				return -EINVAL;
+
+			amdgpu_discovery_harvest_ip(adev);
+			amdgpu_discovery_get_gfx_info(adev);
+			amdgpu_discovery_get_mall_info(adev);
+			amdgpu_discovery_get_vcn_info(adev);
+		} else {
+			cyan_skillfish_reg_base_init(adev);
+			adev->sdma.num_instances = 2;
+			adev->ip_versions[MMHUB_HWIP][0] = IP_VERSION(2, 0, 3);
+			adev->ip_versions[ATHUB_HWIP][0] = IP_VERSION(2, 0, 3);
+			adev->ip_versions[OSSSYS_HWIP][0] = IP_VERSION(5, 0, 1);
+			adev->ip_versions[HDP_HWIP][0] = IP_VERSION(5, 0, 1);
+			adev->ip_versions[SDMA0_HWIP][0] = IP_VERSION(5, 0, 1);
+			adev->ip_versions[SDMA1_HWIP][1] = IP_VERSION(5, 0, 1);
+			adev->ip_versions[DF_HWIP][0] = IP_VERSION(3, 5, 0);
+			adev->ip_versions[NBIO_HWIP][0] = IP_VERSION(2, 1, 1);
+			adev->ip_versions[UMC_HWIP][0] = IP_VERSION(8, 1, 1);
+			adev->ip_versions[MP0_HWIP][0] = IP_VERSION(11, 0, 8);
+			adev->ip_versions[MP1_HWIP][0] = IP_VERSION(11, 0, 8);
+			adev->ip_versions[THM_HWIP][0] = IP_VERSION(11, 0, 1);
+			adev->ip_versions[SMUIO_HWIP][0] = IP_VERSION(11, 0, 8);
+			adev->ip_versions[GC_HWIP][0] = IP_VERSION(10, 1, 3);
+			adev->ip_versions[UVD_HWIP][0] = IP_VERSION(2, 0, 3);
+		}
+		break;
 	default:
 		r = amdgpu_discovery_reg_base_init(adev);
 		if (r) {

From fa819e3a7c1ee994ce014cc5a991c7fd91bc00f1 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 27 Jun 2025 10:21:16 -0400
Subject: [PATCH 1025/1455] drm/amdgpu: add support for cyan skillfish gpu_info

Some SOCs which are part of the cyan skillfish family
rely on an explicit firmware for IP discovery.  Add support
for the gpu_info firmware.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index b9e2f0293d61..bdfb80377e6a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -95,6 +95,7 @@ MODULE_FIRMWARE("amdgpu/picasso_gpu_info.bin");
 MODULE_FIRMWARE("amdgpu/raven2_gpu_info.bin");
 MODULE_FIRMWARE("amdgpu/arcturus_gpu_info.bin");
 MODULE_FIRMWARE("amdgpu/navi12_gpu_info.bin");
+MODULE_FIRMWARE("amdgpu/cyan_skillfish_gpu_info.bin");
 
 #define AMDGPU_RESUME_MS		2000
 #define AMDGPU_MAX_RETRY_LIMIT		2
@@ -2629,6 +2630,9 @@ static int amdgpu_device_parse_gpu_info_fw(struct amdgpu_device *adev)
 			return 0;
 		chip_name = "navi12";
 		break;
+	case CHIP_CYAN_SKILLFISH:
+		chip_name = "cyan_skillfish";
+		break;
 	}
 
 	err = amdgpu_ucode_request(adev, &adev->firmware.gpu_info_fw,

From 94bd7bf2c920998b4c756bc8a54fd3dbdf7e4360 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 27 Jun 2025 10:25:09 -0400
Subject: [PATCH 1026/1455] drm/amdgpu: don't enable SMU on cyan skillfish

Cyan skillfish uses different SMU firmware.

Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
index 38c4ebc063db..73401f0aeb34 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c
@@ -2124,7 +2124,6 @@ static int amdgpu_discovery_set_smu_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(11, 0, 5):
 	case IP_VERSION(11, 0, 9):
 	case IP_VERSION(11, 0, 7):
-	case IP_VERSION(11, 0, 8):
 	case IP_VERSION(11, 0, 11):
 	case IP_VERSION(11, 0, 12):
 	case IP_VERSION(11, 0, 13):
@@ -2132,6 +2131,10 @@ static int amdgpu_discovery_set_smu_ip_blocks(struct amdgpu_device *adev)
 	case IP_VERSION(11, 5, 2):
 		amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
 		break;
+	case IP_VERSION(11, 0, 8):
+		if (adev->apu_flags & AMD_APU_IS_CYAN_SKILLFISH2)
+			amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
+		break;
 	case IP_VERSION(12, 0, 0):
 	case IP_VERSION(12, 0, 1):
 		amdgpu_device_ip_block_add(adev, &smu_v12_0_ip_block);

From cbda64f3f58027f68211dda8ea94d52d7e493995 Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Thu, 4 Sep 2025 20:36:46 +0800
Subject: [PATCH 1027/1455] drm/amdkfd: Fix error code sign for EINVAL in
 svm_ioctl()

Use negative error code -EINVAL instead of positive EINVAL in the default
case of svm_ioctl() to conform to Linux kernel error code conventions.

Fixes: 42de677f7999 ("drm/amdkfd: register svm range")
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 521c14c7a789..68ba239b2e5d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -4261,7 +4261,7 @@ svm_ioctl(struct kfd_process *p, enum kfd_ioctl_svm_op op, uint64_t start,
 		r = svm_range_get_attr(p, mm, start, size, nattrs, attrs);
 		break;
 	default:
-		r = EINVAL;
+		r = -EINVAL;
 		break;
 	}
 

From 54d18bc6003fad33eb901aa5fa3425449e53d5aa Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Thu, 4 Sep 2025 09:54:36 +0800
Subject: [PATCH 1028/1455] drm/amdgpu/userq: add a detect and reset callback

Add a detect and reset callback and add the implementation
for mes.  The callback will detect all hung queues of a
particular ip type (e.g., GFX or compute or SDMA) and
reset them.

v2: increase reset counter and set fence force completion
v3: Removed userq_mutex in mes_userq_detect_and_reset since the driver holds it when calling

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h  |  2 +
 drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 49 ++++++++++++++++++++++
 2 files changed, 51 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
index 017efd2075df..1bd84f4cce78 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
@@ -82,6 +82,8 @@ struct amdgpu_userq_funcs {
 		   struct amdgpu_usermode_queue *queue);
 	int (*restore)(struct amdgpu_userq_mgr *uq_mgr,
 		   struct amdgpu_usermode_queue *queue);
+	int (*detect_and_reset)(struct amdgpu_device *adev,
+		  int queue_type);
 };
 
 /* Usermode queues for gfx */
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
index d6f50b13e2ba..aee26f80bd53 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
@@ -21,6 +21,7 @@
  * OTHER DEALINGS IN THE SOFTWARE.
  *
  */
+#include <drm/drm_drv.h>
 #include "amdgpu.h"
 #include "amdgpu_gfx.h"
 #include "mes_userqueue.h"
@@ -198,6 +199,53 @@ static int mes_userq_create_ctx_space(struct amdgpu_userq_mgr *uq_mgr,
 	return 0;
 }
 
+static int mes_userq_detect_and_reset(struct amdgpu_device *adev,
+				      int queue_type)
+{
+	int db_array_size = amdgpu_mes_get_hung_queue_db_array_size(adev);
+	struct mes_detect_and_reset_queue_input input;
+	struct amdgpu_usermode_queue *queue;
+	struct amdgpu_userq_mgr *uqm, *tmp;
+	unsigned int hung_db_num = 0;
+	int queue_id, r, i;
+	u32 db_array[4];
+
+	if (db_array_size > 4) {
+		dev_err(adev->dev, "DB array size (%d vs 4) too small\n",
+			db_array_size);
+		return -EINVAL;
+	}
+
+	memset(&input, 0x0, sizeof(struct mes_detect_and_reset_queue_input));
+
+	input.queue_type = queue_type;
+
+	amdgpu_mes_lock(&adev->mes);
+	r = amdgpu_mes_detect_and_reset_hung_queues(adev, queue_type, false,
+						    &hung_db_num, db_array);
+	amdgpu_mes_unlock(&adev->mes);
+	if (r) {
+		dev_err(adev->dev, "Failed to detect and reset queues, err (%d)\n", r);
+	} else if (hung_db_num) {
+		list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
+			idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
+				if (queue->queue_type == queue_type) {
+					for (i = 0; i < hung_db_num; i++) {
+						if (queue->doorbell_index == db_array[i]) {
+							queue->state = AMDGPU_USERQ_STATE_HUNG;
+							atomic_inc(&adev->gpu_reset_counter);
+							amdgpu_userq_fence_driver_force_completion(queue);
+							drm_dev_wedged_event(adev_to_drm(adev), DRM_WEDGE_RECOVERY_NONE, NULL);
+						}
+					}
+				}
+			}
+		}
+	}
+
+	return r;
+}
+
 static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 				struct drm_amdgpu_userq_in *args_in,
 				struct amdgpu_usermode_queue *queue)
@@ -352,4 +400,5 @@ const struct amdgpu_userq_funcs userq_mes_funcs = {
 	.mqd_destroy = mes_userq_mqd_destroy,
 	.unmap = mes_userq_unmap,
 	.map = mes_userq_map,
+	.detect_and_reset = mes_userq_detect_and_reset,
 };

From 53c271b9a06ca307c2ce6994877d8d084d031962 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 15:10:15 +0800
Subject: [PATCH 1029/1455] drm/amd/display: Remove redundant ternary operators

For ternary operators in the form of "a ? true : false" or
"a ? false : true", if 'a' itself returns a boolean result, the ternary
operator can be omitted. Remove redundant ternary operators to clean up the
code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_link_encoder.c | 2 +-
 .../drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c            | 2 +-
 drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c      | 3 +--
 drivers/gpu/drm/amd/display/modules/freesync/freesync.c       | 4 ++--
 5 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_link_encoder.c
index e0558a78b11c..1c1228116487 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_link_encoder.c
@@ -812,7 +812,7 @@ bool dcn10_link_encoder_validate_output_with_stream(
 					enc10, &stream->timing);
 	break;
 	case SIGNAL_TYPE_EDP:
-		is_valid = (stream->timing.pixel_encoding == PIXEL_ENCODING_RGB) ? true : false;
+		is_valid = stream->timing.pixel_encoding == PIXEL_ENCODING_RGB;
 	break;
 	case SIGNAL_TYPE_VIRTUAL:
 		is_valid = true;
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c
index 6ab2a218b769..6f30b6cc3c76 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c
@@ -397,7 +397,7 @@ static bool enc35_is_fifo_enabled(struct stream_encoder *enc)
 	uint32_t reset_val;
 
 	REG_GET(DIG_FIFO_CTRL0, DIG_FIFO_ENABLE, &reset_val);
-	return (reset_val == 0) ? false : true;
+	return reset_val != 0;
 }
 void enc35_disable_fifo(struct stream_encoder *enc)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
index 0318260370ed..9deb03a18ccc 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c
@@ -535,7 +535,7 @@ static bool dml2_validate_only(struct dc_state *context, enum dc_validate_mode v
 	if (result)
 		result = does_configuration_meet_sw_policies(dml2, &dml2->v20.scratch.cur_display_config, &dml2->v20.scratch.mode_support_info);
 
-	return (result == 1) ? true : false;
+	return result == 1;
 }
 
 static void dml2_apply_debug_options(const struct dc *dc, struct dml2_context *dml2)
diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c
index b68bcc9fca0a..892907991f91 100644
--- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c
+++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.c
@@ -138,8 +138,7 @@ void setup_dio_stream_attribute(struct pipe_ctx *pipe_ctx)
 		stream_encoder->funcs->dvi_set_stream_attribute(
 				stream_encoder,
 				&stream->timing,
-				(stream->signal == SIGNAL_TYPE_DVI_DUAL_LINK) ?
-						true : false);
+				stream->signal == SIGNAL_TYPE_DVI_DUAL_LINK);
 	else if (dc_is_lvds_signal(stream->signal))
 		stream_encoder->funcs->lvds_set_stream_attribute(
 				stream_encoder,
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 71efd2770c99..ce421bcddcb0 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -226,8 +226,8 @@ static void update_v_total_for_static_ramp(
 	unsigned int target_duration_in_us =
 			calc_duration_in_us_from_refresh_in_uhz(
 				in_out_vrr->fixed.target_refresh_in_uhz);
-	bool ramp_direction_is_up = (current_duration_in_us >
-				target_duration_in_us) ? true : false;
+	bool ramp_direction_is_up = current_duration_in_us >
+				target_duration_in_us;
 
 	/* Calculate ratio between new and current frame duration with 3 digit */
 	unsigned int frame_duration_ratio = div64_u64(1000000,

From db51c5d98b03940d865002adba8a16939dc4c26a Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 16:21:11 +0800
Subject: [PATCH 1030/1455] amdgpu/pm/legacy: remove redundant ternary
 operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
index ea3ace882a10..52dbf6d0469d 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
@@ -771,8 +771,7 @@ static struct amdgpu_ps *amdgpu_dpm_pick_power_state(struct amdgpu_device *adev,
 	int i;
 	struct amdgpu_ps *ps;
 	u32 ui_class;
-	bool single_display = (adev->pm.dpm.new_active_crtc_count < 2) ?
-		true : false;
+	bool single_display = adev->pm.dpm.new_active_crtc_count < 2;
 
 	/* check if the vblank period is too short to adjust the mclk */
 	if (single_display && adev->powerplay.pp_funcs->vblank_too_short) {

From fa740b115b467f48128c727cd4e19a2b7345a18a Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 16:21:12 +0800
Subject: [PATCH 1031/1455] drm/amd/pm/powerplay/hwmgr/ppatomctrl: Remove
 redundant ternary operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code. Swap variable positions
on either side of '!=' to enhance readability.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
index 8d40ed0f0e83..ce166a7f8e42 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c
@@ -563,8 +563,8 @@ bool atomctrl_is_voltage_controlled_by_gpio_v3(
 	PP_ASSERT_WITH_CODE((NULL != voltage_info),
 			"Could not find Voltage Table in BIOS.", return false;);
 
-	ret = (NULL != atomctrl_lookup_voltage_type_v3
-			(voltage_info, voltage_type, voltage_mode)) ? true : false;
+	ret = atomctrl_lookup_voltage_type_v3
+			(voltage_info, voltage_type, voltage_mode) != NULL;
 
 	return ret;
 }

From 9ab06ab36d4a84d087e995233fe10d5d9496f908 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 16:21:13 +0800
Subject: [PATCH 1032/1455] drm/amd/pm/powerplay/smumgr: remove redundant
 ternary operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code. Swap variable positions
on either side of '==' to enhance readability.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c      | 5 ++---
 drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c   | 5 ++---
 drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c | 5 ++---
 drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c     | 5 ++---
 4 files changed, 8 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c
index 5e43ad2b2956..d2dbd90bb427 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/fiji_smumgr.c
@@ -2540,9 +2540,8 @@ static int fiji_initialize_mc_reg_table(struct pp_hwmgr *hwmgr)
 
 static bool fiji_is_dpm_running(struct pp_hwmgr *hwmgr)
 {
-	return (1 == PHM_READ_INDIRECT_FIELD(hwmgr->device,
-			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON))
-			? true : false;
+	return PHM_READ_INDIRECT_FIELD(hwmgr->device,
+			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON) == 1;
 }
 
 static int fiji_update_dpm_settings(struct pp_hwmgr *hwmgr,
diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c
index 17d2f5bff4a7..1f50f1e74c48 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/iceland_smumgr.c
@@ -2655,9 +2655,8 @@ static int iceland_initialize_mc_reg_table(struct pp_hwmgr *hwmgr)
 
 static bool iceland_is_dpm_running(struct pp_hwmgr *hwmgr)
 {
-	return (1 == PHM_READ_INDIRECT_FIELD(hwmgr->device,
-			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON))
-			? true : false;
+	return PHM_READ_INDIRECT_FIELD(hwmgr->device,
+			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON) == 1;
 }
 
 const struct pp_smumgr_func iceland_smu_funcs = {
diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c
index ff6b563ecbf5..bf6d09572cfc 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/polaris10_smumgr.c
@@ -2578,9 +2578,8 @@ static int polaris10_initialize_mc_reg_table(struct pp_hwmgr *hwmgr)
 
 static bool polaris10_is_dpm_running(struct pp_hwmgr *hwmgr)
 {
-	return (1 == PHM_READ_INDIRECT_FIELD(hwmgr->device,
-			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON))
-			? true : false;
+	return PHM_READ_INDIRECT_FIELD(hwmgr->device,
+			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON) == 1;
 }
 
 static int polaris10_update_dpm_settings(struct pp_hwmgr *hwmgr,
diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c
index 6fe6e6abb5d8..2e21f9d066cb 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/tonga_smumgr.c
@@ -3139,9 +3139,8 @@ static int tonga_initialize_mc_reg_table(struct pp_hwmgr *hwmgr)
 
 static bool tonga_is_dpm_running(struct pp_hwmgr *hwmgr)
 {
-	return (1 == PHM_READ_INDIRECT_FIELD(hwmgr->device,
-			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON))
-			? true : false;
+	return PHM_READ_INDIRECT_FIELD(hwmgr->device,
+			CGS_IND_REG__SMC, FEATURE_STATUS, VOLTAGE_CONTROLLER_ON) == 1;
 }
 
 static int tonga_update_dpm_settings(struct pp_hwmgr *hwmgr,

From 7a50377cea5f1f39a91d64eb8c522ee760c91b56 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 19:31:57 +0800
Subject: [PATCH 1033/1455] drm/radeon/atom: Remove redundant ternary operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/atombios_encoders.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index d1c5e471bdca..3d9f47bc807a 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -1832,7 +1832,7 @@ radeon_atom_encoder_dpms(struct drm_encoder *encoder, int mode)
 		return;
 	}
 
-	radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+	radeon_atombios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 
 }
 

From 60d6f01b5a7de4f1475774cc807840a002d601d6 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 19:31:58 +0800
Subject: [PATCH 1034/1455] drm/radeon/dpm: Remove redundant ternary operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/ci_dpm.c | 8 ++++----
 drivers/gpu/drm/radeon/ni_dpm.c | 2 +-
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 1f9f4424055a..ba8db1d07c07 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -2457,7 +2457,7 @@ static void ci_register_patching_mc_arb(struct radeon_device *rdev,
 	u32 tmp, tmp2;
 
 	tmp = RREG32(MC_SEQ_MISC0);
-	patch = ((tmp & 0x0000f00) == 0x300) ? true : false;
+	patch = (tmp & 0x0000f00) == 0x300;
 
 	if (patch &&
 	    ((rdev->pdev->device == 0x67B0) ||
@@ -3438,7 +3438,7 @@ static int ci_setup_default_dpm_tables(struct radeon_device *rdev)
 			pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count].value =
 				allowed_sclk_vddc_table->entries[i].clk;
 			pi->dpm_table.sclk_table.dpm_levels[pi->dpm_table.sclk_table.count].enabled =
-				(i == 0) ? true : false;
+				i == 0;
 			pi->dpm_table.sclk_table.count++;
 		}
 	}
@@ -3451,7 +3451,7 @@ static int ci_setup_default_dpm_tables(struct radeon_device *rdev)
 			pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count].value =
 				allowed_mclk_table->entries[i].clk;
 			pi->dpm_table.mclk_table.dpm_levels[pi->dpm_table.mclk_table.count].enabled =
-				(i == 0) ? true : false;
+				i == 0;
 			pi->dpm_table.mclk_table.count++;
 		}
 	}
@@ -4489,7 +4489,7 @@ static int ci_register_patching_mc_seq(struct radeon_device *rdev,
 	bool patch;
 
 	tmp = RREG32(MC_SEQ_MISC0);
-	patch = ((tmp & 0x0000f00) == 0x300) ? true : false;
+	patch = (tmp & 0x0000f00) == 0x300;
 
 	if (patch &&
 	    ((rdev->pdev->device == 0x67B0) ||
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index e08559c44a5c..82edbfb259bf 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -3397,7 +3397,7 @@ static int ni_enable_smc_cac(struct radeon_device *rdev,
 				if (PPSMC_Result_OK != smc_result)
 					ret = -EINVAL;
 
-				ni_pi->cac_enabled = (PPSMC_Result_OK == smc_result) ? true : false;
+				ni_pi->cac_enabled = PPSMC_Result_OK == smc_result;
 			}
 		} else if (ni_pi->cac_enabled) {
 			smc_result = rv770_send_msg_to_smc(rdev, PPSMC_MSG_DisableCac);

From 220c7a21cb0a570fd0a44ce71e2e4a3a5f342fde Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 19:31:59 +0800
Subject: [PATCH 1035/1455] drm/radeon/radeon_legacy_encoders: Remove redundant
 ternary operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/radeon/radeon_legacy_encoders.c   | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
index d6aa1a3012a8..d1e8b9757a65 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
@@ -136,9 +136,9 @@ static void radeon_legacy_lvds_update(struct drm_encoder *encoder, int mode)
 	}
 
 	if (rdev->is_atom_bios)
-		radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_atombios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 	else
-		radeon_combios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_combios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 
 }
 
@@ -545,9 +545,9 @@ static void radeon_legacy_primary_dac_dpms(struct drm_encoder *encoder, int mode
 	WREG32(RADEON_DAC_MACRO_CNTL, dac_macro_cntl);
 
 	if (rdev->is_atom_bios)
-		radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_atombios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 	else
-		radeon_combios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_combios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 
 }
 
@@ -742,9 +742,9 @@ static void radeon_legacy_tmds_int_dpms(struct drm_encoder *encoder, int mode)
 	WREG32(RADEON_FP_GEN_CNTL, fp_gen_cntl);
 
 	if (rdev->is_atom_bios)
-		radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_atombios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 	else
-		radeon_combios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_combios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 
 }
 
@@ -908,9 +908,9 @@ static void radeon_legacy_tmds_ext_dpms(struct drm_encoder *encoder, int mode)
 	WREG32(RADEON_FP2_GEN_CNTL, fp2_gen_cntl);
 
 	if (rdev->is_atom_bios)
-		radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_atombios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 	else
-		radeon_combios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_combios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 
 }
 
@@ -1113,9 +1113,9 @@ static void radeon_legacy_tv_dac_dpms(struct drm_encoder *encoder, int mode)
 	}
 
 	if (rdev->is_atom_bios)
-		radeon_atombios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_atombios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 	else
-		radeon_combios_encoder_dpms_scratch_regs(encoder, (mode == DRM_MODE_DPMS_ON) ? true : false);
+		radeon_combios_encoder_dpms_scratch_regs(encoder, mode == DRM_MODE_DPMS_ON);
 
 }
 

From a055658793ddce9f2554c0d77e4b404cd0ac4107 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 19:32:00 +0800
Subject: [PATCH 1036/1455] drm/radeon/pm: Remove redundant ternary operators

For ternary operators in the form of "a ? true : false", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/radeon_pm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index b4fb7e70320b..a855a96dd2ea 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -907,8 +907,7 @@ static void radeon_dpm_thermal_work_handler(struct work_struct *work)
 
 static bool radeon_dpm_single_display(struct radeon_device *rdev)
 {
-	bool single_display = (rdev->pm.dpm.new_active_crtc_count < 2) ?
-		true : false;
+	bool single_display = rdev->pm.dpm.new_active_crtc_count < 2;
 
 	/* check if the vblank period is too short to adjust the mclk */
 	if (single_display && rdev->asic->dpm.vblank_too_short) {

From bddf30944d42e1fb7f105fad4e5f9e7e86ea871e Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Thu, 4 Sep 2025 22:58:00 +0800
Subject: [PATCH 1037/1455] drm/amd/pm: use int type to store negative error
 codes

Use int instead of uint32_t for 'ret' variable to store negative error
codes or zero returned by other functions.

Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but can be confusing. Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.

No effect on runtime.

Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c  | 2 +-
 drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c | 2 +-
 drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c       | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c
index 9a821563bc8e..14ccd743ca1d 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c
@@ -1032,7 +1032,7 @@ static int smu10_print_clock_levels(struct pp_hwmgr *hwmgr,
 			data->clock_vol_info.vdd_dep_on_fclk;
 	uint32_t i, now, size = 0;
 	uint32_t min_freq, max_freq = 0;
-	uint32_t ret = 0;
+	int ret = 0;
 
 	switch (type) {
 	case PP_SCLK:
diff --git a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
index baf51cd82a35..0d4cbe4113a0 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/smumgr/smu7_smumgr.c
@@ -401,7 +401,7 @@ failed:
 int smu7_check_fw_load_finish(struct pp_hwmgr *hwmgr, uint32_t fw_type)
 {
 	struct smu7_smumgr *smu_data = (struct smu7_smumgr *)(hwmgr->smu_backend);
-	uint32_t ret;
+	int ret;
 
 	ret = phm_wait_on_indirect_register(hwmgr, mmSMC_IND_INDEX_11,
 					smu_data->soft_regs_start + smum_get_offsetof(hwmgr,
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
index e97b0cf19197..3baf20f4c373 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c
@@ -470,7 +470,7 @@ static int renoir_od_edit_dpm_table(struct smu_context *smu,
 static int renoir_set_fine_grain_gfx_freq_parameters(struct smu_context *smu)
 {
 	uint32_t min = 0, max = 0;
-	uint32_t ret = 0;
+	int ret = 0;
 
 	ret = smu_cmn_send_smc_msg_with_param(smu,
 								SMU_MSG_GetMinGfxclkFrequency,

From cf32515a70618c0fb2319bd4a855f4d9447940a8 Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Wed, 3 Sep 2025 14:29:12 +0800
Subject: [PATCH 1038/1455] drm/amd/pm: refine amdgpu pm sysfs node error code

v1:
Returns different error codes based on the scenario to help the user app understand
the AMDGPU device status when an exception occurs.

v2:
change -NODEV to -EBUSY.

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 5230276628a3..96590c1da553 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -110,9 +110,10 @@ static int amdgpu_pm_dev_state_check(struct amdgpu_device *adev, bool runpm)
 	bool runpm_check = runpm ? adev->in_runpm : false;
 
 	if (amdgpu_in_reset(adev))
-		return -EPERM;
+		return -EBUSY;
+
 	if (adev->in_suspend && !runpm_check)
-		return -EPERM;
+		return -EBUSY;
 
 	return 0;
 }

From b842b994ed0678faf73907a67fb13afcd5e50509 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 28 Aug 2025 10:02:35 +0530
Subject: [PATCH 1039/1455] drm/amd/pm: Add caching for SystemMetrics table

Implement caching separately for SystemMetrics table from PMFW. The same
table could be used for multiple interfaces. Hence, cache it internally
to avoid multiple queries to the firmware. For SystemMetrics table, 5ms
cache interval is sufficient.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h |  1 +
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 43 +++++++++++++------
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  |  4 ++
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h  |  1 +
 4 files changed, 36 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 5dd49eca598d..5976eda80035 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -333,6 +333,7 @@ enum smu_table_id {
 	SMU_TABLE_WIFIBAND,
 	SMU_TABLE_GPUBOARD_TEMP_METRICS,
 	SMU_TABLE_BASEBOARD_TEMP_METRICS,
+	SMU_TABLE_PMFW_SYSTEM_METRICS,
 	SMU_TABLE_COUNT,
 };
 
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 32fd0be05cff..0bec12b348ce 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -149,6 +149,12 @@ int smu_v13_0_12_tables_init(struct smu_context *smu)
 	struct smu_table_cache *cache;
 	int ret;
 
+	ret = smu_table_cache_init(smu, SMU_TABLE_PMFW_SYSTEM_METRICS,
+				   smu_v13_0_12_get_system_metrics_size(), 5);
+
+	if (ret)
+		return ret;
+
 	ret = smu_table_cache_init(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS,
 				   sizeof(*baseboard_temp_metrics), 50);
 	if (ret)
@@ -162,6 +168,7 @@ int smu_v13_0_12_tables_init(struct smu_context *smu)
 	ret = smu_table_cache_init(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS,
 				   sizeof(*gpuboard_temp_metrics), 50);
 	if (ret) {
+		smu_table_cache_fini(smu, SMU_TABLE_PMFW_SYSTEM_METRICS);
 		smu_table_cache_fini(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS);
 		return ret;
 	}
@@ -176,6 +183,7 @@ void smu_v13_0_12_tables_fini(struct smu_context *smu)
 {
 	smu_table_cache_fini(smu, SMU_TABLE_BASEBOARD_TEMP_METRICS);
 	smu_table_cache_fini(smu, SMU_TABLE_GPUBOARD_TEMP_METRICS);
+	smu_table_cache_fini(smu, SMU_TABLE_PMFW_SYSTEM_METRICS);
 }
 
 static int smu_v13_0_12_get_enabled_mask(struct smu_context *smu,
@@ -222,8 +230,12 @@ static int smu_v13_0_12_fru_get_product_info(struct smu_context *smu,
 
 int smu_v13_0_12_get_max_metrics_size(void)
 {
-	return max3(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t),
-		   sizeof(SystemMetricsTable_t));
+	return max(sizeof(StaticMetricsTable_t), sizeof(MetricsTable_t));
+}
+
+size_t smu_v13_0_12_get_system_metrics_size(void)
+{
+	return sizeof(SystemMetricsTable_t);
 }
 
 static void smu_v13_0_12_init_xgmi_data(struct smu_context *smu,
@@ -414,14 +426,18 @@ int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
 	return 0;
 }
 
-static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu,
-						 void *metrics_table)
+static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu)
 {
 	struct smu_table_context *smu_table = &smu->smu_table;
-	uint32_t table_size = smu_table->tables[SMU_TABLE_SMU_METRICS].size;
 	struct smu_table *table = &smu_table->driver_table;
+	struct smu_table *tables = smu_table->tables;
+	struct smu_table *sys_table;
 	int ret;
 
+	sys_table = &tables[SMU_TABLE_PMFW_SYSTEM_METRICS];
+	if (smu_table_cache_is_valid(sys_table))
+		return 0;
+
 	ret = smu_cmn_send_smc_msg(smu, SMU_MSG_GetSystemMetricsTable, NULL);
 	if (ret) {
 		dev_info(smu->adev->dev,
@@ -430,10 +446,9 @@ static int smu_v13_0_12_get_system_metrics_table(struct smu_context *smu,
 	}
 
 	amdgpu_asic_invalidate_hdp(smu->adev, NULL);
-	memcpy(smu_table->metrics_table, table->cpu_addr, table_size);
-
-	if (metrics_table)
-		memcpy(metrics_table, smu_table->metrics_table, sizeof(SystemMetricsTable_t));
+	smu_table_cache_update_time(sys_table, jiffies);
+	memcpy(sys_table->cache.buffer, table->cpu_addr,
+	       smu_v13_0_12_get_system_metrics_size());
 
 	return 0;
 }
@@ -571,10 +586,10 @@ static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
 	struct amdgpu_baseboard_temp_metrics_v1_0 *baseboard_temp_metrics;
 	struct amdgpu_gpuboard_temp_metrics_v1_0 *gpuboard_temp_metrics;
 	struct smu_table_context *smu_table = &smu->smu_table;
-	SystemMetricsTable_t *metrics =
-		(SystemMetricsTable_t *)smu_table->metrics_table;
-
+	struct smu_table *tables = smu_table->tables;
+	SystemMetricsTable_t *metrics;
 	struct smu_table *data_table;
+	struct smu_table *sys_table;
 	int ret, sensor_type;
 	u32 idx, sensors;
 	ssize_t size;
@@ -596,10 +611,12 @@ static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
 		size = sizeof(*baseboard_temp_metrics);
 	}
 
-	ret = smu_v13_0_12_get_system_metrics_table(smu, NULL);
+	ret = smu_v13_0_12_get_system_metrics_table(smu);
 	if (ret)
 		return ret;
 
+	sys_table = &tables[SMU_TABLE_PMFW_SYSTEM_METRICS];
+	metrics = (SystemMetricsTable_t *)sys_table->cache.buffer;
 	smu_table_cache_update_time(data_table, jiffies);
 
 	if (type == SMU_TEMP_METRIC_GPUBOARD) {
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index e37b7b5358ea..ebee659f8a1c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -559,6 +559,10 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu)
 		       PAGE_SIZE,
 		       AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT);
 
+	SMU_TABLE_INIT(tables, SMU_TABLE_PMFW_SYSTEM_METRICS,
+		       smu_v13_0_12_get_system_metrics_size(), PAGE_SIZE,
+		       AMDGPU_GEM_DOMAIN_VRAM | AMDGPU_GEM_DOMAIN_GTT);
+
 	metrics_table = kzalloc(METRICS_TABLE_SIZE, GFP_KERNEL);
 	if (!metrics_table)
 		return -ENOMEM;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index bcb8246c0804..aae9a546a67e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -82,6 +82,7 @@ int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table,
 
 bool smu_v13_0_12_is_dpm_running(struct smu_context *smu);
 int smu_v13_0_12_get_max_metrics_size(void);
+size_t smu_v13_0_12_get_system_metrics_size(void);
 int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu);
 int smu_v13_0_12_get_smu_metrics_data(struct smu_context *smu,
 				      MetricsMember_t member, uint32_t *value);

From c5f4fb40584ee591da9fa090c6f265d11cbb1acf Mon Sep 17 00:00:00 2001
From: Fangzhi Zuo <Jerry.Zuo@amd.com>
Date: Thu, 4 Sep 2025 15:13:51 -0400
Subject: [PATCH 1040/1455] drm/amd/display: Disable DPCD Probe Quirk

Disable dpcd probe quirk to native aux.

Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Reviewed-by: Imre Deak <imre.deak@intel.com>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4500
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20250904191351.746707-1-Jerry.Zuo@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 6a817508c826..6911c339bc61 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -841,6 +841,7 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
 	drm_dp_aux_init(&aconnector->dm_dp_aux.aux);
 	drm_dp_cec_register_connector(&aconnector->dm_dp_aux.aux,
 				      &aconnector->base);
+	drm_dp_dpcd_set_probe(&aconnector->dm_dp_aux.aux, false);
 
 	if (aconnector->base.connector_type == DRM_MODE_CONNECTOR_eDP)
 		return;

From d6fa8026617896c23cb165f09f1c1d32fe22a3df Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 28 Aug 2025 12:50:09 +0530
Subject: [PATCH 1041/1455] drm/amdgpu: Add vbios build number interface

Fetch VBIOS build number from atom rom image. Add a sysfs interface to
read the build number.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c | 33 ++++++++++++++++++--
 drivers/gpu/drm/amd/amdgpu/atom.c            | 22 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/atom.h            |  2 ++
 3 files changed, 54 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
index e476e45b996a..9dfdc08cc887 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
@@ -1816,16 +1816,43 @@ static ssize_t amdgpu_atombios_get_vbios_version(struct device *dev,
 	return sysfs_emit(buf, "%s\n", ctx->vbios_pn);
 }
 
+static ssize_t amdgpu_atombios_get_vbios_build(struct device *dev,
+					       struct device_attribute *attr,
+					       char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct atom_context *ctx = adev->mode_info.atom_context;
+
+	return sysfs_emit(buf, "%s\n", ctx->build_num);
+}
+
 static DEVICE_ATTR(vbios_version, 0444, amdgpu_atombios_get_vbios_version,
 		   NULL);
+static DEVICE_ATTR(vbios_build, 0444, amdgpu_atombios_get_vbios_build, NULL);
 
 static struct attribute *amdgpu_vbios_version_attrs[] = {
-	&dev_attr_vbios_version.attr,
-	NULL
+	&dev_attr_vbios_version.attr, &dev_attr_vbios_build.attr, NULL
 };
 
+static umode_t amdgpu_vbios_version_attrs_is_visible(struct kobject *kobj,
+						     struct attribute *attr,
+						     int index)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	struct atom_context *ctx = adev->mode_info.atom_context;
+
+	if (attr == &dev_attr_vbios_build.attr && !strlen(ctx->build_num))
+		return 0;
+
+	return attr->mode;
+}
+
 const struct attribute_group amdgpu_vbios_version_attr_group = {
-	.attrs = amdgpu_vbios_version_attrs
+	.attrs = amdgpu_vbios_version_attrs,
+	.is_visible = amdgpu_vbios_version_attrs_is_visible,
 };
 
 int amdgpu_atombios_sysfs_init(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index 427b073de2fc..1c994d0cc50b 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -1494,6 +1494,27 @@ static void atom_get_vbios_version(struct atom_context *ctx)
 	}
 }
 
+static void atom_get_vbios_build(struct atom_context *ctx)
+{
+	unsigned char *atom_rom_hdr;
+	unsigned char *str;
+	uint16_t base;
+
+	base = CU16(ATOM_ROM_TABLE_PTR);
+	atom_rom_hdr = CSTR(base);
+
+	str = CSTR(CU16(base + ATOM_ROM_CFG_PTR));
+	/* Skip config string */
+	while (str < atom_rom_hdr && *str++)
+		;
+	/* Skip change list string */
+	while (str < atom_rom_hdr && *str++)
+		;
+
+	if ((str + STRLEN_NORMAL) < atom_rom_hdr)
+		strscpy(ctx->build_num, str, STRLEN_NORMAL);
+}
+
 struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios)
 {
 	int base;
@@ -1554,6 +1575,7 @@ struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios)
 	atom_get_vbios_pn(ctx);
 	atom_get_vbios_date(ctx);
 	atom_get_vbios_version(ctx);
+	atom_get_vbios_build(ctx);
 
 	return ctx;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/atom.h b/drivers/gpu/drm/amd/amdgpu/atom.h
index b807f6639a4c..825ff28731f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.h
+++ b/drivers/gpu/drm/amd/amdgpu/atom.h
@@ -37,6 +37,7 @@ struct drm_device;
 #define ATOM_ROM_MAGIC		"ATOM"
 #define ATOM_ROM_MAGIC_PTR	4
 
+#define ATOM_ROM_CFG_PTR 0xC
 #define ATOM_ROM_MSG_PTR	0x10
 #define ATOM_ROM_CMD_PTR	0x1E
 #define ATOM_ROM_DATA_PTR	0x20
@@ -151,6 +152,7 @@ struct atom_context {
 	uint32_t version;
 	uint8_t vbios_ver_str[STRLEN_NORMAL];
 	uint8_t date[STRLEN_NORMAL];
+	uint8_t build_num[STRLEN_NORMAL];
 };
 
 extern int amdgpu_atom_debug;

From 9fd2da71c301184d98fe37674ca8d017d1ce6600 Mon Sep 17 00:00:00 2001
From: Oleh Kuzhylnyi <okuzhyln@amd.com>
Date: Mon, 18 Aug 2025 13:19:29 +0200
Subject: [PATCH 1042/1455] drm/amd/display: Add HDCP policy control

[Why]
DM should be able to control HDCP retry limit via configurable
parameter.

[How]
Expose a retry_limit parameter for controlling the maximum number of
retries and lift the hardcode out to DM.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Oleh Kuzhylnyi <okuzhyln@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c | 2 ++
 drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c        | 5 +++--
 drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h     | 1 +
 3 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
index 437174d4fed5..58e084f52526 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
@@ -223,6 +223,7 @@ void hdcp_update_display(struct hdcp_workqueue *hdcp_work,
 		display_adjust.disable = MOD_HDCP_DISPLAY_NOT_DISABLE;
 
 		link_adjust.auth_delay = 2;
+		link_adjust.retry_limit = MAX_NUM_OF_ATTEMPTS;
 
 		if (content_type == DRM_MODE_HDCP_CONTENT_TYPE0) {
 			link_adjust.hdcp2.force_type = MOD_HDCP_FORCE_TYPE_0;
@@ -572,6 +573,7 @@ static void update_config(void *handle, struct cp_psp_stream_config *config)
 	link->dp.usb4_enabled = config->usb4_enabled;
 	display->adjust.disable = MOD_HDCP_DISPLAY_DISABLE_AUTHENTICATION;
 	link->adjust.auth_delay = 2;
+	link->adjust.retry_limit = MAX_NUM_OF_ATTEMPTS;
 	link->adjust.hdcp1.disable = 0;
 	hdcp_w->encryption_status[display->index] = MOD_HDCP_ENCRYPTION_STATUS_HDCP_OFF;
 
diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c
index 5e01c6e24cbc..c760216a6240 100644
--- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c
+++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp.c
@@ -29,6 +29,7 @@ static void push_error_status(struct mod_hdcp *hdcp,
 		enum mod_hdcp_status status)
 {
 	struct mod_hdcp_trace *trace = &hdcp->connection.trace;
+	const uint8_t retry_limit = hdcp->connection.link.adjust.retry_limit;
 
 	if (trace->error_count < MAX_NUM_OF_ERROR_TRACE) {
 		trace->errors[trace->error_count].status = status;
@@ -39,11 +40,11 @@ static void push_error_status(struct mod_hdcp *hdcp,
 
 	if (is_hdcp1(hdcp)) {
 		hdcp->connection.hdcp1_retry_count++;
-		if (hdcp->connection.hdcp1_retry_count == MAX_NUM_OF_ATTEMPTS)
+		if (hdcp->connection.hdcp1_retry_count == retry_limit)
 			hdcp->connection.link.adjust.hdcp1.disable = 1;
 	} else if (is_hdcp2(hdcp)) {
 		hdcp->connection.hdcp2_retry_count++;
-		if (hdcp->connection.hdcp2_retry_count == MAX_NUM_OF_ATTEMPTS)
+		if (hdcp->connection.hdcp2_retry_count == retry_limit)
 			hdcp->connection.link.adjust.hdcp2.disable = 1;
 	}
 }
diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h
index c42468bb70ac..b51ddf2846df 100644
--- a/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h
+++ b/drivers/gpu/drm/amd/display/modules/inc/mod_hdcp.h
@@ -220,6 +220,7 @@ struct mod_hdcp_link_adjustment_hdcp2 {
 
 struct mod_hdcp_link_adjustment {
 	uint8_t auth_delay;
+	uint8_t retry_limit;
 	struct mod_hdcp_link_adjustment_hdcp1 hdcp1;
 	struct mod_hdcp_link_adjustment_hdcp2 hdcp2;
 };

From c2223b26869dde3f7610b5c9766bd5322cfa908f Mon Sep 17 00:00:00 2001
From: Cruise Hung <Cruise.Hung@amd.com>
Date: Fri, 22 Aug 2025 15:45:03 +0800
Subject: [PATCH 1043/1455] drm/amd/display: Add link index in AUX and dpms

[Why & How]
Add the link index in DP AUX transfer and DPMS functions.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dce_aux.c    | 17 +++++++++++++----
 .../drm/amd/display/dc/link/link_detection.c    |  2 +-
 drivers/gpu/drm/amd/display/dc/link/link_dpms.c |  9 +++++----
 3 files changed, 19 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
index bb4ac5042c80..673bb87d2c17 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_aux.c
@@ -725,14 +725,18 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
 	for (i = 0; i < AUX_MAX_RETRIES; i++) {
 		DC_TRACE_LEVEL_MESSAGE(DAL_TRACE_LEVEL_INFORMATION,
 					LOG_FLAG_I2cAux_DceAux,
-					"dce_aux_transfer_with_retries: link_index=%u: START: retry %d of %d: address=0x%04x length=%u write=%d mot=%d",
+					"dce_aux_transfer_with_retries: link_index=%u: START: retry %d of %d: "
+					"address=0x%04x length=%u write=%d mot=%d is_i2c=%d is_dpia=%d ddc_hw_inst=%d",
 					ddc && ddc->link ? ddc->link->link_index : UINT_MAX,
 					i + 1,
 					(int)AUX_MAX_RETRIES,
 					payload->address,
 					payload->length,
 					(unsigned int) payload->write,
-					(unsigned int) payload->mot);
+					(unsigned int) payload->mot,
+					payload->i2c_over_aux,
+					(ddc->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) ? true : false,
+					ddc->link->ddc_hw_inst);
 		if (payload->write)
 			dce_aux_log_payload("  write", payload->data, payload->length, 16);
 
@@ -746,7 +750,9 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
 
 		DC_TRACE_LEVEL_MESSAGE(DAL_TRACE_LEVEL_INFORMATION,
 					LOG_FLAG_I2cAux_DceAux,
-					"dce_aux_transfer_with_retries: link_index=%u: END: retry %d of %d: address=0x%04x length=%u write=%d mot=%d: ret=%d operation_result=%d payload->reply=%u",
+					"dce_aux_transfer_with_retries: link_index=%u: END: retry %d of %d: "
+					"address=0x%04x length=%u write=%d mot=%d: ret=%d operation_result=%d "
+					"payload->reply=%u  is_i2c=%d is_dpia=%d ddc_hw_inst=%d",
 					ddc && ddc->link ? ddc->link->link_index : UINT_MAX,
 					i + 1,
 					(int)AUX_MAX_RETRIES,
@@ -756,7 +762,10 @@ bool dce_aux_transfer_with_retries(struct ddc_service *ddc,
 					(unsigned int) payload->mot,
 					ret,
 					(int)operation_result,
-					(unsigned int) *payload->reply);
+					(unsigned int) *payload->reply,
+					payload->i2c_over_aux,
+					(ddc->link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA) ? true : false,
+					ddc->link->ddc_hw_inst);
 		if (!payload->write)
 			dce_aux_log_payload("  read", payload->data, ret > 0 ? ret : 0, 16);
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
index 827b630daf49..b717e430051a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
@@ -656,7 +656,7 @@ static bool wait_for_entering_dp_alt_mode(struct dc_link *link)
 		return true;
 
 	is_in_alt_mode = link->link_enc->funcs->is_in_alt_mode(link->link_enc);
-	DC_LOG_DC("DP Alt mode state on HPD: %d\n", is_in_alt_mode);
+	DC_LOG_DC("DP Alt mode state on HPD: %d  Link=%d\n", is_in_alt_mode, link->link_index);
 
 	if (is_in_alt_mode)
 		return true;
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index cb80b4599936..08ee8d2f777b 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -2358,9 +2358,9 @@ void link_set_dpms_off(struct pipe_ctx *pipe_ctx)
 	if (pipe_ctx->stream->sink) {
 		if (pipe_ctx->stream->sink->sink_signal != SIGNAL_TYPE_VIRTUAL &&
 			pipe_ctx->stream->sink->sink_signal != SIGNAL_TYPE_NONE) {
-			DC_LOG_DC("%s pipe_ctx dispname=%s signal=%x\n", __func__,
+			DC_LOG_DC("%s pipe_ctx dispname=%s signal=%x link=%d\n", __func__,
 			pipe_ctx->stream->sink->edid_caps.display_name,
-			pipe_ctx->stream->signal);
+			pipe_ctx->stream->signal, link->link_index);
 		}
 	}
 
@@ -2473,9 +2473,10 @@ void link_set_dpms_on(
 	if (pipe_ctx->stream->sink) {
 		if (pipe_ctx->stream->sink->sink_signal != SIGNAL_TYPE_VIRTUAL &&
 			pipe_ctx->stream->sink->sink_signal != SIGNAL_TYPE_NONE) {
-			DC_LOG_DC("%s pipe_ctx dispname=%s signal=%x\n", __func__,
+			DC_LOG_DC("%s pipe_ctx dispname=%s signal=%x link=%d\n", __func__,
 			pipe_ctx->stream->sink->edid_caps.display_name,
-			pipe_ctx->stream->signal);
+			pipe_ctx->stream->signal,
+			link->link_index);
 		}
 	}
 

From 895b61395eefd28376250778a741f11e12715a39 Mon Sep 17 00:00:00 2001
From: Ausef Yousof <Ausef.Yousof@amd.com>
Date: Thu, 21 Aug 2025 18:11:54 -0400
Subject: [PATCH 1044/1455] drm/amd/display: dont wait for pipe update during
 medupdate/highirq

[why&how]
control flag for the wait during pipe update wait for vupdate should
be set if update type is not fast or med to prevent an invalid sleep
operation

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Ausef Yousof <Ausef.Yousof@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 261b7d43e91d..5963019d1e74 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -4169,7 +4169,7 @@ static void commit_planes_for_stream(struct dc *dc,
 	}
 
 	if (dc->hwseq->funcs.wait_for_pipe_update_if_needed)
-		dc->hwseq->funcs.wait_for_pipe_update_if_needed(dc, top_pipe_to_program, update_type == UPDATE_TYPE_FAST);
+		dc->hwseq->funcs.wait_for_pipe_update_if_needed(dc, top_pipe_to_program, update_type < UPDATE_TYPE_FULL);
 
 	if (should_lock_all_pipes && dc->hwss.interdependent_update_lock) {
 		if (dc->hwss.subvp_pipe_control_lock)

From 68f3c044f37d9f50d67417fa8018d9cf16423458 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <superm1@kernel.org>
Date: Sun, 24 Aug 2025 15:20:58 -0500
Subject: [PATCH 1045/1455] drm/amd/display: Indicate when custom brightness
 curves are in use

[Why]
There is a `scale` sysfs attribute that can be used to indicate when
non-linear brightness scaling is in use.  As Custom brightness curves
work by linear interpolation of points the scale is no longer linear.

[How]
Indicate non-linear scaling when custom brightness curves in use and
linear scaling otherwise.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <superm1@kernel.org>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index dc4d04a8ce6b..8d13212ff4ab 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -5060,8 +5060,11 @@ amdgpu_dm_register_backlight_device(struct amdgpu_dm_connector *aconnector)
 	} else
 		props.brightness = props.max_brightness = MAX_BACKLIGHT_LEVEL;
 
-	if (caps->data_points && !(amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE))
+	if (caps->data_points && !(amdgpu_dc_debug_mask & DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE)) {
 		drm_info(drm, "Using custom brightness curve\n");
+		props.scale = BACKLIGHT_SCALE_NON_LINEAR;
+	} else
+		props.scale = BACKLIGHT_SCALE_LINEAR;
 	props.type = BACKLIGHT_RAW;
 
 	snprintf(bl_name, sizeof(bl_name), "amdgpu_bl%d",

From 215bed54897bae6352e4850a03063739c55f7e9a Mon Sep 17 00:00:00 2001
From: Allen Li <wei-guang.li@amd.com>
Date: Mon, 25 Aug 2025 14:23:14 +0800
Subject: [PATCH 1046/1455] drm/amd/display: Read DPCD to obtain eDP capability
 information.

[Why & How]
Extend to read eDP general capability 2 in detect_edp_sink_caps().

Reviewed-by: Robin Chen <robin.chen@amd.com>
Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Allen Li <wei-guang.li@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_dp_types.h                | 1 +
 .../drm/amd/display/dc/link/protocols/link_dp_capability.c  | 6 ++++++
 2 files changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
index 3a3ec38cdf8b..db669ccb1d58 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h
@@ -1284,6 +1284,7 @@ struct dpcd_caps {
 	union dp_receive_port0_cap receive_port0_cap;
 	/* Indicates the number of SST links supported by MSO (Multi-Stream Output) */
 	uint8_t mso_cap_sst_links_supported;
+	uint8_t dp_edp_general_cap_2;
 };
 
 union dpcd_sink_ext_caps {
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
index caddb7dfb133..b12c11bd6a14 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c
@@ -2195,6 +2195,12 @@ void detect_edp_sink_caps(struct dc_link *link)
 			DP_EDP_MSO_LINK_CAPABILITIES,
 			(uint8_t *)&link->dpcd_caps.mso_cap_sst_links_supported,
 			sizeof(link->dpcd_caps.mso_cap_sst_links_supported));
+	/*
+	 * Read eDP general capability 2
+	 */
+	core_link_read_dpcd(link, DP_EDP_GENERAL_CAP_2,
+			(uint8_t *)&link->dpcd_caps.dp_edp_general_cap_2,
+			sizeof(link->dpcd_caps.dp_edp_general_cap_2));
 }
 
 bool dp_get_max_link_enc_cap(const struct dc_link *link, struct dc_link_settings *max_link_enc_cap)

From f74291c1de320c3aa36f053280f30ebdc122576b Mon Sep 17 00:00:00 2001
From: Wenjing Liu <wenjing.liu@amd.com>
Date: Wed, 13 Aug 2025 16:18:53 -0400
Subject: [PATCH 1047/1455] drm/amd/display: Update dchubbub.h for hubbub
 perfmon support

[why]
dchubbub supports performance monitoring for hubbub.
The interfaces define the performance monitoring events and their
attributes.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/inc/hw/dchubbub.h  | 22 +++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
index 9bee45b36629..843a18287c83 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dchubbub.h
@@ -137,6 +137,19 @@ struct dcn_hubbub_state {
 	uint32_t dram_state_cntl;
 };
 
+struct hubbub_system_latencies {
+	uint32_t max_latency_ns;
+	uint32_t avg_latency_ns;
+	uint32_t min_latency_ns;
+};
+
+struct hubbub_urgent_latency_params {
+	uint32_t refclk_mhz;
+	uint32_t t_win_ns;
+	uint32_t bandwidth_mbps;
+	uint32_t bw_factor_x1000;
+};
+
 struct hubbub_funcs {
 	void (*update_dchub)(
 			struct hubbub *hubbub,
@@ -231,6 +244,15 @@ struct hubbub_funcs {
 	bool (*program_arbiter)(struct hubbub *hubbub, struct dml2_display_arb_regs *arb_regs, bool safe_to_lower);
 	void (*get_det_sizes)(struct hubbub *hubbub, uint32_t *curr_det_sizes, uint32_t *target_det_sizes);
 	uint32_t (*compbuf_config_error)(struct hubbub *hubbub);
+	struct hubbub_perfmon_funcs{
+		void (*start_system_latency_measurement)(struct hubbub *hubbub);
+		void (*get_system_latency_result)(struct hubbub *hubbub, uint32_t refclk_mhz, struct hubbub_system_latencies *latencies);
+		void (*start_in_order_bandwidth_measurement)(struct hubbub *hubbub);
+		void (*get_in_order_bandwidth_result)(struct hubbub *hubbub, uint32_t refclk_mhz, uint32_t *bandwidth_mbps);
+		void (*start_urgent_ramp_latency_measurement)(struct hubbub *hubbub, const struct hubbub_urgent_latency_params *params);
+		void (*get_urgent_ramp_latency_result)(struct hubbub *hubbub, uint32_t refclk_mhz, uint32_t *latency_ns);
+		void (*reset)(struct hubbub *hubbub);
+	} perfmon;
 };
 
 struct hubbub {

From 3b14fe98939b01aa72707bc3c2617de319839a65 Mon Sep 17 00:00:00 2001
From: Roman Li <Roman.Li@amd.com>
Date: Thu, 21 Aug 2025 15:43:31 -0400
Subject: [PATCH 1048/1455] drm/amd/display: Refine error message for vblank
 init failure

[Why]
The error message "failed to initialize sw for display support" is used
for both DRM device and vblank initialization failures, making it difficult
to identify the specific failure during troubleshooting.

[How]
Update the vblank initialization error message to
"failed to initialize vblank for display support"
to distinguish it from the DRM device init failure.

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Roman Li <Roman.Li@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 8d13212ff4ab..5ae01b0ac093 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -2186,7 +2186,7 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 
 	if (drm_vblank_init(adev_to_drm(adev), adev->dm.display_indexes_num)) {
 		drm_err(adev_to_drm(adev),
-		"failed to initialize sw for display support.\n");
+		"failed to initialize vblank for display support.\n");
 		goto error;
 	}
 

From 1bde5584e297921f45911ae874b0175dce5ed4b5 Mon Sep 17 00:00:00 2001
From: Ovidiu Bunea <ovidiu.bunea@amd.com>
Date: Mon, 25 Aug 2025 14:45:33 -0400
Subject: [PATCH 1049/1455] drm/amd/display: Correct sequences and delays for
 DCN35 PG & RCG

[why]
The current PG & RCG programming in driver has some gaps and incorrect
sequences.

[how]
Added delays after ungating clocks to allow ramp up, increased polling
to allow more time for power up, and removed the incorrect sequences.

Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Signed-off-by: Ovidiu Bunea <ovidiu.bunea@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h           |   1 +
 .../amd/display/dc/dccg/dcn35/dcn35_dccg.c    |  74 +++++------
 .../amd/display/dc/hwss/dcn35/dcn35_hwseq.c   | 115 +++---------------
 .../amd/display/dc/hwss/dcn35/dcn35_init.c    |   3 -
 .../amd/display/dc/hwss/dcn351/dcn351_init.c  |   3 -
 .../gpu/drm/amd/display/dc/inc/hw/pg_cntl.h   |   1 +
 .../amd/display/dc/pg/dcn35/dcn35_pg_cntl.c   |  78 +++++++-----
 7 files changed, 111 insertions(+), 164 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index b41e66c31e6a..09d705cf5c9b 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1162,6 +1162,7 @@ struct dc_debug_options {
 	unsigned int auxless_alpm_lfps_silence_ns;
 	unsigned int auxless_alpm_lfps_t1t2_us;
 	short auxless_alpm_lfps_t1t2_offset_us;
+	bool enable_pg_cntl_debug_logs;
 };
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
index 58c84f555c0f..0ce9489ac6b7 100644
--- a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
@@ -133,30 +133,34 @@ enum dsc_clk_source {
 };
 
 
-static void dccg35_set_dsc_clk_rcg(struct dccg *dccg, int inst, bool enable)
+static void dccg35_set_dsc_clk_rcg(struct dccg *dccg, int inst, bool allow_rcg)
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
-	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dsc && enable)
+	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dsc && allow_rcg)
 		return;
 
 	switch (inst) {
 	case 0:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK0_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK0_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	case 1:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK1_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK1_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	case 2:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK2_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK2_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	case 3:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK3_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK3_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+
+	/* Wait for clock to ramp */
+	if (!allow_rcg)
+		udelay(10);
 }
 
 static void dccg35_set_symclk32_se_rcg(
@@ -385,35 +389,34 @@ static void dccg35_set_dtbclk_p_rcg(struct dccg *dccg, int inst, bool enable)
 	}
 }
 
-static void dccg35_set_dppclk_rcg(struct dccg *dccg,
-												int inst, bool enable)
+static void dccg35_set_dppclk_rcg(struct dccg *dccg, int inst, bool allow_rcg)
 {
-
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
-
-	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dpp && enable)
+	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dpp && allow_rcg)
 		return;
 
 	switch (inst) {
 	case 0:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK0_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK0_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	case 1:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK1_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK1_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	case 2:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK2_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK2_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	case 3:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK3_ROOT_GATE_DISABLE, enable ? 0 : 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK3_ROOT_GATE_DISABLE, allow_rcg ? 0 : 1);
 		break;
 	default:
 	BREAK_TO_DEBUGGER();
 		break;
 	}
-	//DC_LOG_DEBUG("%s: inst(%d) DPPCLK rcg_disable: %d\n", __func__, inst, enable ? 0 : 1);
 
+	/* Wait for clock to ramp */
+	if (!allow_rcg)
+		udelay(10);
 }
 
 static void dccg35_set_dpstreamclk_rcg(
@@ -1177,32 +1180,34 @@ static void dccg35_update_dpp_dto(struct dccg *dccg, int dpp_inst,
 }
 
 static void dccg35_set_dppclk_root_clock_gating(struct dccg *dccg,
-		 uint32_t dpp_inst, uint32_t enable)
+		 uint32_t dpp_inst, uint32_t disallow_rcg)
 {
 	struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg);
 
-	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dpp)
+	if (!dccg->ctx->dc->debug.root_clock_optimization.bits.dpp && !disallow_rcg)
 		return;
 
 
 	switch (dpp_inst) {
 	case 0:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK0_ROOT_GATE_DISABLE, enable);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK0_ROOT_GATE_DISABLE, disallow_rcg);
 		break;
 	case 1:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK1_ROOT_GATE_DISABLE, enable);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK1_ROOT_GATE_DISABLE, disallow_rcg);
 		break;
 	case 2:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK2_ROOT_GATE_DISABLE, enable);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK2_ROOT_GATE_DISABLE, disallow_rcg);
 		break;
 	case 3:
-		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK3_ROOT_GATE_DISABLE, enable);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DPPCLK3_ROOT_GATE_DISABLE, disallow_rcg);
 		break;
 	default:
 		break;
 	}
-	//DC_LOG_DEBUG("%s: dpp_inst(%d) rcg: %d\n", __func__, dpp_inst, enable);
 
+	/* Wait for clock to ramp */
+	if (disallow_rcg)
+		udelay(10);
 }
 
 static void dccg35_get_pixel_rate_div(
@@ -1782,8 +1787,7 @@ static void dccg35_enable_dscclk(struct dccg *dccg, int inst)
 	//Disable DTO
 	switch (inst) {
 	case 0:
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK0_ROOT_GATE_DISABLE, 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK0_ROOT_GATE_DISABLE, 1);
 
 		REG_UPDATE_2(DSCCLK0_DTO_PARAM,
 				DSCCLK0_DTO_PHASE, 0,
@@ -1791,8 +1795,7 @@ static void dccg35_enable_dscclk(struct dccg *dccg, int inst)
 		REG_UPDATE(DSCCLK_DTO_CTRL,	DSCCLK0_EN, 1);
 		break;
 	case 1:
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK1_ROOT_GATE_DISABLE, 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK1_ROOT_GATE_DISABLE, 1);
 
 		REG_UPDATE_2(DSCCLK1_DTO_PARAM,
 				DSCCLK1_DTO_PHASE, 0,
@@ -1800,8 +1803,7 @@ static void dccg35_enable_dscclk(struct dccg *dccg, int inst)
 		REG_UPDATE(DSCCLK_DTO_CTRL, DSCCLK1_EN, 1);
 		break;
 	case 2:
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK2_ROOT_GATE_DISABLE, 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK2_ROOT_GATE_DISABLE, 1);
 
 		REG_UPDATE_2(DSCCLK2_DTO_PARAM,
 				DSCCLK2_DTO_PHASE, 0,
@@ -1809,8 +1811,7 @@ static void dccg35_enable_dscclk(struct dccg *dccg, int inst)
 		REG_UPDATE(DSCCLK_DTO_CTRL, DSCCLK2_EN, 1);
 		break;
 	case 3:
-		if (dccg->ctx->dc->debug.root_clock_optimization.bits.dsc)
-			REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK3_ROOT_GATE_DISABLE, 1);
+		REG_UPDATE(DCCG_GATE_DISABLE_CNTL6, DSCCLK3_ROOT_GATE_DISABLE, 1);
 
 		REG_UPDATE_2(DSCCLK3_DTO_PARAM,
 				DSCCLK3_DTO_PHASE, 0,
@@ -1821,6 +1822,9 @@ static void dccg35_enable_dscclk(struct dccg *dccg, int inst)
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+
+	/* Wait for clock to ramp */
+	udelay(10);
 }
 
 static void dccg35_disable_dscclk(struct dccg *dccg,
@@ -1864,6 +1868,9 @@ static void dccg35_disable_dscclk(struct dccg *dccg,
 	default:
 		return;
 	}
+
+	/* Wait for clock ramp */
+	udelay(10);
 }
 
 static void dccg35_enable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst)
@@ -2349,10 +2356,7 @@ static void dccg35_disable_symclk_se_cb(
 
 void dccg35_root_gate_disable_control(struct dccg *dccg, uint32_t pipe_idx, uint32_t disable_clock_gating)
 {
-
-	if (dccg->ctx->dc->debug.root_clock_optimization.bits.dpp) {
-		dccg35_set_dppclk_root_clock_gating(dccg, pipe_idx, disable_clock_gating);
-	}
+	dccg35_set_dppclk_root_clock_gating(dccg, pipe_idx, disable_clock_gating);
 }
 
 static const struct dccg_funcs dccg35_funcs_new = {
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
index a267f574b619..764eff6a4ec6 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
@@ -113,6 +113,14 @@ static void enable_memory_low_power(struct dc *dc)
 }
 #endif
 
+static void print_pg_status(struct dc *dc, const char *debug_func, const char *debug_log)
+{
+	if (dc->debug.enable_pg_cntl_debug_logs && dc->res_pool->pg_cntl) {
+		if (dc->res_pool->pg_cntl->funcs->print_pg_status)
+			dc->res_pool->pg_cntl->funcs->print_pg_status(dc->res_pool->pg_cntl, debug_func, debug_log);
+	}
+}
+
 void dcn35_set_dmu_fgcg(struct dce_hwseq *hws, bool enable)
 {
 	REG_UPDATE_3(DMU_CLK_CNTL,
@@ -137,6 +145,8 @@ void dcn35_init_hw(struct dc *dc)
 	uint32_t user_level = MAX_BACKLIGHT_LEVEL;
 	int i;
 
+	print_pg_status(dc, __func__, ": start");
+
 	if (dc->clk_mgr && dc->clk_mgr->funcs->init_clocks)
 		dc->clk_mgr->funcs->init_clocks(dc->clk_mgr);
 
@@ -200,10 +210,7 @@ void dcn35_init_hw(struct dc *dc)
 
 	/* we want to turn off all dp displays before doing detection */
 	dc->link_srv->blank_all_dp_displays(dc);
-/*
-	if (hws->funcs.enable_power_gating_plane)
-		hws->funcs.enable_power_gating_plane(dc->hwseq, true);
-*/
+
 	if (res_pool->hubbub && res_pool->hubbub->funcs->dchubbub_init)
 		res_pool->hubbub->funcs->dchubbub_init(dc->res_pool->hubbub);
 	/* If taking control over from VBIOS, we may want to optimize our first
@@ -236,6 +243,8 @@ void dcn35_init_hw(struct dc *dc)
 		}
 
 		hws->funcs.init_pipes(dc, dc->current_state);
+		print_pg_status(dc, __func__, ": after init_pipes");
+
 		if (dc->res_pool->hubbub->funcs->allow_self_refresh_control &&
 			!dc->res_pool->hubbub->ctx->dc->debug.disable_stutter)
 			dc->res_pool->hubbub->funcs->allow_self_refresh_control(dc->res_pool->hubbub,
@@ -312,6 +321,7 @@ void dcn35_init_hw(struct dc *dc)
 		if (dc->res_pool->pg_cntl->funcs->init_pg_status)
 			dc->res_pool->pg_cntl->funcs->init_pg_status(dc->res_pool->pg_cntl);
 	}
+	print_pg_status(dc, __func__, ": after init_pg_status");
 }
 
 static void update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
@@ -500,97 +510,6 @@ void dcn35_physymclk_root_clock_control(struct dce_hwseq *hws, unsigned int phy_
 	}
 }
 
-void dcn35_dsc_pg_control(
-		struct dce_hwseq *hws,
-		unsigned int dsc_inst,
-		bool power_on)
-{
-	uint32_t power_gate = power_on ? 0 : 1;
-	uint32_t pwr_status = power_on ? 0 : 2;
-	uint32_t org_ip_request_cntl = 0;
-
-	if (hws->ctx->dc->debug.disable_dsc_power_gate)
-		return;
-	if (hws->ctx->dc->debug.ignore_pg)
-		return;
-	REG_GET(DC_IP_REQUEST_CNTL, IP_REQUEST_EN, &org_ip_request_cntl);
-	if (org_ip_request_cntl == 0)
-		REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1);
-
-	switch (dsc_inst) {
-	case 0: /* DSC0 */
-		REG_UPDATE(DOMAIN16_PG_CONFIG,
-				DOMAIN_POWER_GATE, power_gate);
-
-		REG_WAIT(DOMAIN16_PG_STATUS,
-				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
-		break;
-	case 1: /* DSC1 */
-		REG_UPDATE(DOMAIN17_PG_CONFIG,
-				DOMAIN_POWER_GATE, power_gate);
-
-		REG_WAIT(DOMAIN17_PG_STATUS,
-				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
-		break;
-	case 2: /* DSC2 */
-		REG_UPDATE(DOMAIN18_PG_CONFIG,
-				DOMAIN_POWER_GATE, power_gate);
-
-		REG_WAIT(DOMAIN18_PG_STATUS,
-				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
-		break;
-	case 3: /* DSC3 */
-		REG_UPDATE(DOMAIN19_PG_CONFIG,
-				DOMAIN_POWER_GATE, power_gate);
-
-		REG_WAIT(DOMAIN19_PG_STATUS,
-				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
-		break;
-	default:
-		BREAK_TO_DEBUGGER();
-		break;
-	}
-
-	if (org_ip_request_cntl == 0)
-		REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 0);
-}
-
-void dcn35_enable_power_gating_plane(struct dce_hwseq *hws, bool enable)
-{
-	bool force_on = true; /* disable power gating */
-	uint32_t org_ip_request_cntl = 0;
-
-	if (hws->ctx->dc->debug.disable_hubp_power_gate)
-		return;
-	if (hws->ctx->dc->debug.ignore_pg)
-		return;
-	REG_GET(DC_IP_REQUEST_CNTL, IP_REQUEST_EN, &org_ip_request_cntl);
-	if (org_ip_request_cntl == 0)
-		REG_SET(DC_IP_REQUEST_CNTL, 0, IP_REQUEST_EN, 1);
-	/* DCHUBP0/1/2/3/4/5 */
-	REG_UPDATE(DOMAIN0_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-	REG_UPDATE(DOMAIN2_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-	/* DPP0/1/2/3/4/5 */
-	REG_UPDATE(DOMAIN1_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-	REG_UPDATE(DOMAIN3_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-
-	force_on = true; /* disable power gating */
-	if (enable && !hws->ctx->dc->debug.disable_dsc_power_gate)
-		force_on = false;
-
-	/* DCS0/1/2/3/4 */
-	REG_UPDATE(DOMAIN16_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-	REG_UPDATE(DOMAIN17_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-	REG_UPDATE(DOMAIN18_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-	REG_UPDATE(DOMAIN19_PG_CONFIG, DOMAIN_POWER_FORCEON, force_on);
-
-
-}
-
 /* In headless boot cases, DIG may be turned
  * on which causes HW/SW discrepancies.
  * To avoid this, power down hardware on boot
@@ -1453,6 +1372,8 @@ void dcn35_prepare_bandwidth(
 	}
 
 	dcn20_prepare_bandwidth(dc, context);
+
+	print_pg_status(dc, __func__, ": after rcg and power up");
 }
 
 void dcn35_optimize_bandwidth(
@@ -1461,6 +1382,8 @@ void dcn35_optimize_bandwidth(
 {
 	struct pg_block_update pg_update_state;
 
+	print_pg_status(dc, __func__, ": before rcg and power up");
+
 	dcn20_optimize_bandwidth(dc, context);
 
 	if (dc->hwss.calc_blocks_to_gate) {
@@ -1472,6 +1395,8 @@ void dcn35_optimize_bandwidth(
 		if (dc->hwss.root_clock_control)
 			dc->hwss.root_clock_control(dc, &pg_update_state, false);
 	}
+
+	print_pg_status(dc, __func__, ": after rcg and power up");
 }
 
 void dcn35_set_drr(struct pipe_ctx **pipe_ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
index 52cc488416ac..f2f16a0bdb4f 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c
@@ -115,7 +115,6 @@ static const struct hw_sequencer_funcs dcn35_funcs = {
 	.exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state,
 	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.apply_idle_power_optimizations = dcn35_apply_idle_power_optimizations,
-	.update_dsc_pg = dcn32_update_dsc_pg,
 	.calc_blocks_to_gate = dcn35_calc_blocks_to_gate,
 	.calc_blocks_to_ungate = dcn35_calc_blocks_to_ungate,
 	.hw_block_power_up = dcn35_hw_block_power_up,
@@ -151,7 +150,6 @@ static const struct hwseq_private_funcs dcn35_private_funcs = {
 	.plane_atomic_disable = dcn35_plane_atomic_disable,
 	//.plane_atomic_disable = dcn20_plane_atomic_disable,/*todo*/
 	//.hubp_pg_control = dcn35_hubp_pg_control,
-	.enable_power_gating_plane = dcn35_enable_power_gating_plane,
 	.dpp_root_clock_control = dcn35_dpp_root_clock_control,
 	.dpstream_root_clock_control = dcn35_dpstream_root_clock_control,
 	.physymclk_root_clock_control = dcn35_physymclk_root_clock_control,
@@ -166,7 +164,6 @@ static const struct hwseq_private_funcs dcn35_private_funcs = {
 	.calculate_dccg_k1_k2_values = dcn32_calculate_dccg_k1_k2_values,
 	.resync_fifo_dccg_dio = dcn314_resync_fifo_dccg_dio,
 	.is_dp_dig_pixel_rate_div_policy = dcn35_is_dp_dig_pixel_rate_div_policy,
-	.dsc_pg_control = dcn35_dsc_pg_control,
 	.dsc_pg_status = dcn32_dsc_pg_status,
 	.enable_plane = dcn35_enable_plane,
 	.wait_for_pipe_update_if_needed = dcn10_wait_for_pipe_update_if_needed,
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
index e34efcb7bde5..09e60158f0b5 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c
@@ -114,7 +114,6 @@ static const struct hw_sequencer_funcs dcn351_funcs = {
 	.exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state,
 	.update_visual_confirm_color = dcn10_update_visual_confirm_color,
 	.apply_idle_power_optimizations = dcn35_apply_idle_power_optimizations,
-	.update_dsc_pg = dcn32_update_dsc_pg,
 	.calc_blocks_to_gate = dcn351_calc_blocks_to_gate,
 	.calc_blocks_to_ungate = dcn351_calc_blocks_to_ungate,
 	.hw_block_power_up = dcn351_hw_block_power_up,
@@ -146,7 +145,6 @@ static const struct hwseq_private_funcs dcn351_private_funcs = {
 	.plane_atomic_disable = dcn35_plane_atomic_disable,
 	//.plane_atomic_disable = dcn20_plane_atomic_disable,/*todo*/
 	//.hubp_pg_control = dcn35_hubp_pg_control,
-	.enable_power_gating_plane = dcn35_enable_power_gating_plane,
 	.dpp_root_clock_control = dcn35_dpp_root_clock_control,
 	.dpstream_root_clock_control = dcn35_dpstream_root_clock_control,
 	.physymclk_root_clock_control = dcn35_physymclk_root_clock_control,
@@ -160,7 +158,6 @@ static const struct hwseq_private_funcs dcn351_private_funcs = {
 	.setup_hpo_hw_control = dcn35_setup_hpo_hw_control,
 	.calculate_dccg_k1_k2_values = dcn32_calculate_dccg_k1_k2_values,
 	.is_dp_dig_pixel_rate_div_policy = dcn35_is_dp_dig_pixel_rate_div_policy,
-	.dsc_pg_control = dcn35_dsc_pg_control,
 	.dsc_pg_status = dcn32_dsc_pg_status,
 	.enable_plane = dcn35_enable_plane,
 	.wait_for_pipe_update_if_needed = dcn10_wait_for_pipe_update_if_needed,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h b/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h
index 44f86cc2d1d6..227e3f8d7e5f 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/pg_cntl.h
@@ -49,6 +49,7 @@ struct pg_cntl_funcs {
 	void (*mem_pg_control)(struct pg_cntl *pg_cntl, bool power_on);
 	void (*dio_pg_control)(struct pg_cntl *pg_cntl, bool power_on);
 	void (*init_pg_status)(struct pg_cntl *pg_cntl);
+	void (*print_pg_status)(struct pg_cntl *pg_cntl, const char *debug_func, const char *debug_log);
 };
 
 #endif //__DC_PG_CNTL_H__
diff --git a/drivers/gpu/drm/amd/display/dc/pg/dcn35/dcn35_pg_cntl.c b/drivers/gpu/drm/amd/display/dc/pg/dcn35/dcn35_pg_cntl.c
index af21c0a27f86..72bd43f9bbe2 100644
--- a/drivers/gpu/drm/amd/display/dc/pg/dcn35/dcn35_pg_cntl.c
+++ b/drivers/gpu/drm/amd/display/dc/pg/dcn35/dcn35_pg_cntl.c
@@ -79,16 +79,12 @@ void pg_cntl35_dsc_pg_control(struct pg_cntl *pg_cntl, unsigned int dsc_inst, bo
 	uint32_t power_gate = power_on ? 0 : 1;
 	uint32_t pwr_status = power_on ? 0 : 2;
 	uint32_t org_ip_request_cntl = 0;
-	bool block_enabled;
+	bool block_enabled = false;
+	bool skip_pg = pg_cntl->ctx->dc->debug.ignore_pg ||
+		       pg_cntl->ctx->dc->debug.disable_dsc_power_gate ||
+		       pg_cntl->ctx->dc->idle_optimizations_allowed;
 
-	/*need to enable dscclk regardless DSC_PG*/
-	if (pg_cntl->ctx->dc->res_pool->dccg->funcs->enable_dsc && power_on)
-		pg_cntl->ctx->dc->res_pool->dccg->funcs->enable_dsc(
-				pg_cntl->ctx->dc->res_pool->dccg, dsc_inst);
-
-	if (pg_cntl->ctx->dc->debug.ignore_pg ||
-		pg_cntl->ctx->dc->debug.disable_dsc_power_gate ||
-		pg_cntl->ctx->dc->idle_optimizations_allowed)
+	if (skip_pg && !power_on)
 		return;
 
 	block_enabled = pg_cntl35_dsc_pg_status(pg_cntl, dsc_inst);
@@ -111,7 +107,7 @@ void pg_cntl35_dsc_pg_control(struct pg_cntl *pg_cntl, unsigned int dsc_inst, bo
 
 		REG_WAIT(DOMAIN16_PG_STATUS,
 				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
+				1, 10000);
 		break;
 	case 1: /* DSC1 */
 		REG_UPDATE(DOMAIN17_PG_CONFIG,
@@ -119,7 +115,7 @@ void pg_cntl35_dsc_pg_control(struct pg_cntl *pg_cntl, unsigned int dsc_inst, bo
 
 		REG_WAIT(DOMAIN17_PG_STATUS,
 				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
+				1, 10000);
 		break;
 	case 2: /* DSC2 */
 		REG_UPDATE(DOMAIN18_PG_CONFIG,
@@ -127,7 +123,7 @@ void pg_cntl35_dsc_pg_control(struct pg_cntl *pg_cntl, unsigned int dsc_inst, bo
 
 		REG_WAIT(DOMAIN18_PG_STATUS,
 				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
+				1, 10000);
 		break;
 	case 3: /* DSC3 */
 		REG_UPDATE(DOMAIN19_PG_CONFIG,
@@ -135,7 +131,7 @@ void pg_cntl35_dsc_pg_control(struct pg_cntl *pg_cntl, unsigned int dsc_inst, bo
 
 		REG_WAIT(DOMAIN19_PG_STATUS,
 				DOMAIN_PGFSM_PWR_STATUS, pwr_status,
-				1, 1000);
+				1, 10000);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
@@ -144,12 +140,6 @@ void pg_cntl35_dsc_pg_control(struct pg_cntl *pg_cntl, unsigned int dsc_inst, bo
 
 	if (dsc_inst < MAX_PIPES)
 		pg_cntl->pg_pipe_res_enable[PG_DSC][dsc_inst] = power_on;
-
-	if (pg_cntl->ctx->dc->res_pool->dccg->funcs->disable_dsc && !power_on) {
-		/*this is to disable dscclk*/
-		pg_cntl->ctx->dc->res_pool->dccg->funcs->disable_dsc(
-			pg_cntl->ctx->dc->res_pool->dccg, dsc_inst);
-	}
 }
 
 static bool pg_cntl35_hubp_dpp_pg_status(struct pg_cntl *pg_cntl, unsigned int hubp_dpp_inst)
@@ -189,11 +179,12 @@ void pg_cntl35_hubp_dpp_pg_control(struct pg_cntl *pg_cntl, unsigned int hubp_dp
 	uint32_t pwr_status = power_on ? 0 : 2;
 	uint32_t org_ip_request_cntl;
 	bool block_enabled;
+	bool skip_pg = pg_cntl->ctx->dc->debug.ignore_pg ||
+		       pg_cntl->ctx->dc->debug.disable_hubp_power_gate ||
+		       pg_cntl->ctx->dc->debug.disable_dpp_power_gate ||
+		       pg_cntl->ctx->dc->idle_optimizations_allowed;
 
-	if (pg_cntl->ctx->dc->debug.ignore_pg ||
-		pg_cntl->ctx->dc->debug.disable_hubp_power_gate ||
-		pg_cntl->ctx->dc->debug.disable_dpp_power_gate ||
-		pg_cntl->ctx->dc->idle_optimizations_allowed)
+	if (skip_pg && !power_on)
 		return;
 
 	block_enabled = pg_cntl35_hubp_dpp_pg_status(pg_cntl, hubp_dpp_inst);
@@ -213,22 +204,22 @@ void pg_cntl35_hubp_dpp_pg_control(struct pg_cntl *pg_cntl, unsigned int hubp_dp
 	case 0:
 		/* DPP0 & HUBP0 */
 		REG_UPDATE(DOMAIN0_PG_CONFIG, DOMAIN_POWER_GATE, power_gate);
-		REG_WAIT(DOMAIN0_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 1000);
+		REG_WAIT(DOMAIN0_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 10000);
 		break;
 	case 1:
 		/* DPP1 & HUBP1 */
 		REG_UPDATE(DOMAIN1_PG_CONFIG, DOMAIN_POWER_GATE, power_gate);
-		REG_WAIT(DOMAIN1_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 1000);
+		REG_WAIT(DOMAIN1_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 10000);
 		break;
 	case 2:
 		/* DPP2 & HUBP2 */
 		REG_UPDATE(DOMAIN2_PG_CONFIG, DOMAIN_POWER_GATE, power_gate);
-		REG_WAIT(DOMAIN2_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 1000);
+		REG_WAIT(DOMAIN2_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 10000);
 		break;
 	case 3:
 		/* DPP3 & HUBP3 */
 		REG_UPDATE(DOMAIN3_PG_CONFIG, DOMAIN_POWER_GATE, power_gate);
-		REG_WAIT(DOMAIN3_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 1000);
+		REG_WAIT(DOMAIN3_PG_STATUS, DOMAIN_PGFSM_PWR_STATUS, pwr_status, 1, 10000);
 		break;
 	default:
 		BREAK_TO_DEBUGGER();
@@ -501,6 +492,36 @@ void pg_cntl35_init_pg_status(struct pg_cntl *pg_cntl)
 	pg_cntl->pg_res_enable[PG_DWB] = block_enabled;
 }
 
+static void pg_cntl35_print_pg_status(struct pg_cntl *pg_cntl, const char *debug_func, const char *debug_log)
+{
+	int i = 0;
+	bool block_enabled = false;
+
+	DC_LOG_DEBUG("%s: %s", debug_func, debug_log);
+
+	DC_LOG_DEBUG("PG_CNTL status:\n");
+
+	block_enabled = pg_cntl35_io_clk_status(pg_cntl);
+	DC_LOG_DEBUG("ONO0=%d (DCCG, DIO, DCIO)\n", block_enabled ? 1 : 0);
+
+	block_enabled = pg_cntl35_mem_status(pg_cntl);
+	DC_LOG_DEBUG("ONO1=%d (DCHUBBUB, DCHVM, DCHUBBUBMEM)\n", block_enabled ? 1 : 0);
+
+	block_enabled = pg_cntl35_plane_otg_status(pg_cntl);
+	DC_LOG_DEBUG("ONO2=%d (MPC, OPP, OPTC, DWB)\n", block_enabled ? 1 : 0);
+
+	block_enabled = pg_cntl35_hpo_pg_status(pg_cntl);
+	DC_LOG_DEBUG("ONO3=%d (HPO)\n", block_enabled ? 1 : 0);
+
+	for (i = 0; i < pg_cntl->ctx->dc->res_pool->pipe_count; i++) {
+		block_enabled = pg_cntl35_hubp_dpp_pg_status(pg_cntl, i);
+		DC_LOG_DEBUG("ONO%d=%d (DCHUBP%d, DPP%d)\n", 4 + i * 2, block_enabled ? 1 : 0, i, i);
+
+		block_enabled = pg_cntl35_dsc_pg_status(pg_cntl, i);
+		DC_LOG_DEBUG("ONO%d=%d (DSC%d)\n", 5 + i * 2, block_enabled ? 1 : 0, i);
+	}
+}
+
 static const struct pg_cntl_funcs pg_cntl35_funcs = {
 	.init_pg_status = pg_cntl35_init_pg_status,
 	.dsc_pg_control = pg_cntl35_dsc_pg_control,
@@ -511,7 +532,8 @@ static const struct pg_cntl_funcs pg_cntl35_funcs = {
 	.mpcc_pg_control = pg_cntl35_mpcc_pg_control,
 	.opp_pg_control = pg_cntl35_opp_pg_control,
 	.optc_pg_control = pg_cntl35_optc_pg_control,
-	.dwb_pg_control = pg_cntl35_dwb_pg_control
+	.dwb_pg_control = pg_cntl35_dwb_pg_control,
+	.print_pg_status = pg_cntl35_print_pg_status
 };
 
 struct pg_cntl *pg_cntl35_create(

From 12cdfb61b32a7be581ec5932e0b6a482cb098204 Mon Sep 17 00:00:00 2001
From: Fangzhi Zuo <Jerry.Zuo@amd.com>
Date: Tue, 8 Apr 2025 15:11:43 -0400
Subject: [PATCH 1050/1455] drm/amd/display: Fix pbn_div Calculation Error

[Why]
dm_mst_get_pbn_divider() returns value integer coming from
the cast from fixed point, but the casted integer will then be used
in dfixed_const to be multiplied by 4096. The cast from fixed point to integer
causes the calculation error becomes bigger when multiplied by 4096.

That makes the calculated pbn_div value becomes smaller than
it should be, which leads to the req_slot number becomes bigger.

Such error is getting reflected in 8k30 timing,
where the correct and incorrect calculated req_slot 62.9 Vs 63.1.
That makes the wrong calculation failed to light up 8k30
after a dock under HBR3 x 4.

[How]
Restore the accuracy by keeping the fraction part
calculated for the left shift operation.

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c   |  2 +-
 .../drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c | 13 ++++++++++---
 .../drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h |  2 +-
 3 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 5ae01b0ac093..e34d98a945f2 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8015,7 +8015,7 @@ static int dm_encoder_helper_atomic_check(struct drm_encoder *encoder,
 	if (IS_ERR(mst_state))
 		return PTR_ERR(mst_state);
 
-	mst_state->pbn_div.full = dfixed_const(dm_mst_get_pbn_divider(aconnector->mst_root->dc_link));
+	mst_state->pbn_div.full = dm_mst_get_pbn_divider(aconnector->mst_root->dc_link);
 
 	if (!state->duplicated) {
 		int max_bpc = conn_state->max_requested_bpc;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 6911c339bc61..5e92eaa67aa3 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -854,13 +854,20 @@ void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
 	drm_connector_attach_dp_subconnector_property(&aconnector->base);
 }
 
-int dm_mst_get_pbn_divider(struct dc_link *link)
+uint32_t dm_mst_get_pbn_divider(struct dc_link *link)
 {
+	uint32_t pbn_div_x100;
+	uint64_t dividend, divisor;
+
 	if (!link)
 		return 0;
 
-	return dc_link_bandwidth_kbps(link,
-			dc_link_get_link_cap(link)) / (8 * 1000 * 54);
+	dividend = (uint64_t)dc_link_bandwidth_kbps(link, dc_link_get_link_cap(link)) * 100;
+	divisor = 8 * 1000 * 54;
+
+	pbn_div_x100 = div64_u64(dividend, divisor);
+
+	return dfixed_const(pbn_div_x100) / 100;
 }
 
 struct dsc_mst_fairness_params {
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
index 65f76a7d00db..6f7ea684b555 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
@@ -60,7 +60,7 @@ enum mst_msg_ready_type {
 struct amdgpu_display_manager;
 struct amdgpu_dm_connector;
 
-int dm_mst_get_pbn_divider(struct dc_link *link);
+uint32_t dm_mst_get_pbn_divider(struct dc_link *link);
 
 void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
 				       struct amdgpu_dm_connector *aconnector,

From aba4ead2690ed0bd529fcfeca22de04468f1bad8 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 29 Aug 2025 09:45:27 -0500
Subject: [PATCH 1051/1455] drm/amd/display: Promote DC to 3.2.349

This version brings along following updates:

- Disable stutter when programming watermarks on dcn32
- Fix pbn_div Calculation Error
- Correct sequences and delays for DCN35 PG & RCG
- Define interfaces for hubbub perfmance monitoring support
- Extend to read eDP general capability 2
- Indicate when custom brightness curves are in use
- Dont wait for pipe update during medupdate/highirq
- Add HDCP retry_limit control parameter

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Wayne Lin <wayne.lin@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 09d705cf5c9b..aa808675fead 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.348"
+#define DC_VER "3.2.349"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 476a4e10a3361509b1e436ec49e9a2896052a4b6 Mon Sep 17 00:00:00 2001
From: Sunil Khatri <sunil.khatri@amd.com>
Date: Mon, 18 Aug 2025 12:51:25 +0530
Subject: [PATCH 1052/1455] drm/amdgpu: print root PD address in PDE format
 instead of GPU
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Print PD address of VM root instead of GPU address in the debugfs.
On modern GPU's this is what UMR tool expects in the registers
as well.

Fixes: 719b378d3718 ("drm/amdgpu: add debugfs support for VM pagetable per client")
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
index 127091de0f34..a70651050acf 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c
@@ -2155,7 +2155,7 @@ static int amdgpu_pt_info_read(struct seq_file *m, void *unused)
 		return -EINVAL;
 	}
 
-	seq_printf(m, "gpu_address: 0x%llx\n", amdgpu_bo_gpu_offset(fpriv->vm.root.bo));
+	seq_printf(m, "pd_address: 0x%llx\n", amdgpu_gmc_pd_addr(fpriv->vm.root.bo));
 	seq_printf(m, "max_pfn: 0x%llx\n", adev->vm_manager.max_pfn);
 	seq_printf(m, "num_level: 0x%x\n", adev->vm_manager.num_level);
 	seq_printf(m, "block_size: 0x%x\n", adev->vm_manager.block_size);

From 86b6f02a6d504d035f617d3d5627d2a2099877d4 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Sun, 31 Aug 2025 15:16:04 +0530
Subject: [PATCH 1053/1455] drm/amdgpu: Correct misnamed function in
 amdgpu_gem.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The header comment above amdgpu_gem_list_handles_ioctl referenced
drm_amdgpu_gem_list_handles_ioctl. Update the comment to reflect the
actual function identifier to avoid misleading prototype warnings.

Fixes the below:
drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c:1106: warning: expecting prototype for drm_amdgpu_gem_list_handles_ioctl(). Prototype was for amdgpu_gem_list_handles_ioctl() instead

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index e587e5ddeb00..8bec41cdccd7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -1087,7 +1087,7 @@ out_exec:
 }
 
 /**
- * drm_amdgpu_gem_list_handles_ioctl - get information about a process' buffer objects
+ * amdgpu_gem_list_handles_ioctl - get information about a process' buffer objects
  *
  * @dev: drm device pointer
  * @data: drm_amdgpu_gem_list_handles

From b7c5334af69d359fcd7a39ed409a08138900ec18 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Tue, 2 Sep 2025 12:45:34 -0400
Subject: [PATCH 1054/1455] drm/radeon: use dev_warn_once() in CS parsers

Older GPUs did not support memory protection, so the kernel
driver would validate the command submissions (CS) from userspace
to avoid the GPU accessing any memory it shouldn't.

Change any error messages in that validation to dev_warn_once() to
avoid spamming the kernel log in the event of a bad CS.  If users
see any of these messages they should report them to the user space
component, which in most cases is mesa
(https://gitlab.freedesktop.org/mesa/mesa/-/issues).

Cc: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20250829171655.GBaLHgh3VOvuM1UfJg@fat_crate.local
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/radeon/evergreen_cs.c | 523 +++++++++++++-------------
 drivers/gpu/drm/radeon/r100.c         | 215 +++++------
 drivers/gpu/drm/radeon/r200.c         |  34 +-
 drivers/gpu/drm/radeon/r300.c         |  66 ++--
 drivers/gpu/drm/radeon/r600_cs.c      | 445 +++++++++++-----------
 drivers/gpu/drm/radeon/radeon_cs.c    |   2 +-
 6 files changed, 650 insertions(+), 635 deletions(-)

diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 266c57733136..1162cb5d75ed 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -951,13 +951,13 @@ static int evergreen_cs_track_check(struct radeon_cs_parser *p)
 					u64 offset = (u64)track->vgt_strmout_bo_offset[i] +
 							(u64)track->vgt_strmout_size[i];
 					if (offset > radeon_bo_size(track->vgt_strmout_bo[i])) {
-						DRM_ERROR("streamout %d bo too small: 0x%llx, 0x%lx\n",
-							  i, offset,
-							  radeon_bo_size(track->vgt_strmout_bo[i]));
+						dev_warn_once(p->dev, "streamout %d bo too small: 0x%llx, 0x%lx\n",
+							      i, offset,
+							      radeon_bo_size(track->vgt_strmout_bo[i]));
 						return -EINVAL;
 					}
 				} else {
-					dev_warn(p->dev, "No buffer for streamout %d\n", i);
+					dev_warn_once(p->dev, "No buffer for streamout %d\n", i);
 					return -EINVAL;
 				}
 			}
@@ -979,8 +979,8 @@ static int evergreen_cs_track_check(struct radeon_cs_parser *p)
 			    (tmp >> (i * 4)) & 0xF) {
 				/* at least one component is enabled */
 				if (track->cb_color_bo[i] == NULL) {
-					dev_warn(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n",
-						__func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i);
+					dev_warn_once(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n",
+						      __func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i);
 					return -EINVAL;
 				}
 				/* check cb */
@@ -1056,8 +1056,8 @@ static int evergreen_packet0_check(struct radeon_cs_parser *p,
 	case EVERGREEN_VLINE_START_END:
 		r = evergreen_cs_packet_parse_vline(p);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			return r;
 		}
 		break;
@@ -1143,8 +1143,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_VSTMP_RING_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1155,15 +1155,15 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		break;
 	case CAYMAN_DB_EQAA:
 		if (p->rdev->family < CHIP_CAYMAN) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-				 "0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		break;
 	case CAYMAN_DB_DEPTH_INFO:
 		if (p->rdev->family < CHIP_CAYMAN) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-				 "0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		break;
@@ -1172,8 +1172,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				dev_warn(p->dev, "bad SET_CONTEXT_REG "
-						"0x%04X\n", reg);
+				dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+					      "0x%04X\n", reg);
 				return -EINVAL;
 			}
 			ib[idx] &= ~Z_ARRAY_MODE(0xf);
@@ -1214,8 +1214,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_Z_READ_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->db_z_read_offset = radeon_get_ib_value(p, idx);
@@ -1226,8 +1226,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_Z_WRITE_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->db_z_write_offset = radeon_get_ib_value(p, idx);
@@ -1238,8 +1238,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_STENCIL_READ_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->db_s_read_offset = radeon_get_ib_value(p, idx);
@@ -1250,8 +1250,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_STENCIL_WRITE_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->db_s_write_offset = radeon_get_ib_value(p, idx);
@@ -1273,8 +1273,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case VGT_STRMOUT_BUFFER_BASE_3:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = (reg - VGT_STRMOUT_BUFFER_BASE_0) / 16;
@@ -1295,8 +1295,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CP_COHER_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "missing reloc for CP_COHER_BASE "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "missing reloc for CP_COHER_BASE "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1311,8 +1311,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		break;
 	case PA_SC_AA_CONFIG:
 		if (p->rdev->family >= CHIP_CAYMAN) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-				 "0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = radeon_get_ib_value(p, idx) & MSAA_NUM_SAMPLES_MASK;
@@ -1320,8 +1320,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		break;
 	case CAYMAN_PA_SC_AA_CONFIG:
 		if (p->rdev->family < CHIP_CAYMAN) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-				 "0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = radeon_get_ib_value(p, idx) & CAYMAN_MSAA_NUM_SAMPLES_MASK;
@@ -1360,8 +1360,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				dev_warn(p->dev, "bad SET_CONTEXT_REG "
-						"0x%04X\n", reg);
+				dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+					      "0x%04X\n", reg);
 				return -EINVAL;
 			}
 			ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->tiling_flags));
@@ -1378,8 +1378,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				dev_warn(p->dev, "bad SET_CONTEXT_REG "
-						"0x%04X\n", reg);
+				dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+					      "0x%04X\n", reg);
 				return -EINVAL;
 			}
 			ib[idx] |= CB_ARRAY_MODE(evergreen_cs_get_aray_mode(reloc->tiling_flags));
@@ -1439,8 +1439,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR7_ATTRIB:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
@@ -1467,8 +1467,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR11_ATTRIB:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
@@ -1555,8 +1555,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR7_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = (reg - CB_COLOR0_BASE) / 0x3c;
@@ -1571,8 +1571,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR11_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = ((reg - CB_COLOR8_BASE) / 0x1c) + 8;
@@ -1584,8 +1584,8 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_HTILE_DATA_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->htile_offset = radeon_get_ib_value(p, idx);
@@ -1702,36 +1702,36 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_ALU_CONST_CACHE_LS_15:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
 		break;
 	case SX_MEMORY_EXPORT_BASE:
 		if (p->rdev->family >= CHIP_CAYMAN) {
-			dev_warn(p->dev, "bad SET_CONFIG_REG "
-				 "0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONFIG_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONFIG_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONFIG_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
 		break;
 	case CAYMAN_SX_SCATTER_EXPORT_BASE:
 		if (p->rdev->family < CHIP_CAYMAN) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-				 "0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1740,7 +1740,7 @@ static int evergreen_cs_handle_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->sx_misc_kill_all_prims = (radeon_get_ib_value(p, idx) & 0x1) != 0;
 		break;
 	default:
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+		dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
 	}
 	return 0;
@@ -1795,7 +1795,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 1) {
-			DRM_ERROR("bad SET PREDICATION\n");
+			dev_warn_once(p->dev, "bad SET PREDICATION\n");
 			return -EINVAL;
 		}
 
@@ -1807,13 +1807,13 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			return 0;
 
 		if (pred_op > 2) {
-			DRM_ERROR("bad SET PREDICATION operation %d\n", pred_op);
+			dev_warn_once(p->dev, "bad SET PREDICATION operation %d\n", pred_op);
 			return -EINVAL;
 		}
 
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad SET PREDICATION\n");
+			dev_warn_once(p->dev, "bad SET PREDICATION\n");
 			return -EINVAL;
 		}
 
@@ -1827,7 +1827,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	break;
 	case PACKET3_CONTEXT_CONTROL:
 		if (pkt->count != 1) {
-			DRM_ERROR("bad CONTEXT_CONTROL\n");
+			dev_warn_once(p->dev, "bad CONTEXT_CONTROL\n");
 			return -EINVAL;
 		}
 		break;
@@ -1835,17 +1835,17 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_NUM_INSTANCES:
 	case PACKET3_CLEAR_STATE:
 		if (pkt->count) {
-			DRM_ERROR("bad INDEX_TYPE/NUM_INSTANCES/CLEAR_STATE\n");
+			dev_warn_once(p->dev, "bad INDEX_TYPE/NUM_INSTANCES/CLEAR_STATE\n");
 			return -EINVAL;
 		}
 		break;
 	case CAYMAN_PACKET3_DEALLOC_STATE:
 		if (p->rdev->family < CHIP_CAYMAN) {
-			DRM_ERROR("bad PACKET3_DEALLOC_STATE\n");
+			dev_warn_once(p->dev, "bad PACKET3_DEALLOC_STATE\n");
 			return -EINVAL;
 		}
 		if (pkt->count) {
-			DRM_ERROR("bad INDEX_TYPE/NUM_INSTANCES/CLEAR_STATE\n");
+			dev_warn_once(p->dev, "bad INDEX_TYPE/NUM_INSTANCES/CLEAR_STATE\n");
 			return -EINVAL;
 		}
 		break;
@@ -1854,12 +1854,12 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 1) {
-			DRM_ERROR("bad INDEX_BASE\n");
+			dev_warn_once(p->dev, "bad INDEX_BASE\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad INDEX_BASE\n");
+			dev_warn_once(p->dev, "bad INDEX_BASE\n");
 			return -EINVAL;
 		}
 
@@ -1872,7 +1872,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
@@ -1880,7 +1880,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_INDEX_BUFFER_SIZE:
 	{
 		if (pkt->count != 0) {
-			DRM_ERROR("bad INDEX_BUFFER_SIZE\n");
+			dev_warn_once(p->dev, "bad INDEX_BUFFER_SIZE\n");
 			return -EINVAL;
 		}
 		break;
@@ -1889,12 +1889,12 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	{
 		uint64_t offset;
 		if (pkt->count != 3) {
-			DRM_ERROR("bad DRAW_INDEX\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad DRAW_INDEX\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX\n");
 			return -EINVAL;
 		}
 
@@ -1907,7 +1907,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
@@ -1917,12 +1917,12 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 4) {
-			DRM_ERROR("bad DRAW_INDEX_2\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_2\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad DRAW_INDEX_2\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_2\n");
 			return -EINVAL;
 		}
 
@@ -1935,63 +1935,63 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	}
 	case PACKET3_DRAW_INDEX_AUTO:
 		if (pkt->count != 1) {
-			DRM_ERROR("bad DRAW_INDEX_AUTO\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_AUTO\n");
 			return -EINVAL;
 		}
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
 			return r;
 		}
 		break;
 	case PACKET3_DRAW_INDEX_MULTI_AUTO:
 		if (pkt->count != 2) {
-			DRM_ERROR("bad DRAW_INDEX_MULTI_AUTO\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_MULTI_AUTO\n");
 			return -EINVAL;
 		}
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
 			return r;
 		}
 		break;
 	case PACKET3_DRAW_INDEX_IMMD:
 		if (pkt->count < 2) {
-			DRM_ERROR("bad DRAW_INDEX_IMMD\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_IMMD\n");
 			return -EINVAL;
 		}
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	case PACKET3_DRAW_INDEX_OFFSET:
 		if (pkt->count != 2) {
-			DRM_ERROR("bad DRAW_INDEX_OFFSET\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_OFFSET\n");
 			return -EINVAL;
 		}
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	case PACKET3_DRAW_INDEX_OFFSET_2:
 		if (pkt->count != 3) {
-			DRM_ERROR("bad DRAW_INDEX_OFFSET_2\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_OFFSET_2\n");
 			return -EINVAL;
 		}
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
@@ -2005,19 +2005,19 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		   4 ADDRESS_HI Bits [31:8] - Reserved. Bits [7:0] - Upper bits of Address [47:32]
 		*/
 		if (pkt->count != 2) {
-			DRM_ERROR("bad SET_BASE\n");
+			dev_warn_once(p->dev, "bad SET_BASE\n");
 			return -EINVAL;
 		}
 
 		/* currently only supporting setting indirect draw buffer base address */
 		if (idx_value != 1) {
-			DRM_ERROR("bad SET_BASE\n");
+			dev_warn_once(p->dev, "bad SET_BASE\n");
 			return -EINVAL;
 		}
 
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad SET_BASE\n");
+			dev_warn_once(p->dev, "bad SET_BASE\n");
 			return -EINVAL;
 		}
 
@@ -2039,54 +2039,54 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		   3 DRAW_INITIATOR Draw Initiator Register. Written to the VGT_DRAW_INITIATOR register for the assigned context
 		*/
 		if (pkt->count != 1) {
-			DRM_ERROR("bad DRAW_INDIRECT\n");
+			dev_warn_once(p->dev, "bad DRAW_INDIRECT\n");
 			return -EINVAL;
 		}
 
 		if (idx_value + size > track->indirect_draw_buffer_size) {
-			dev_warn(p->dev, "DRAW_INDIRECT buffer too small %u + %llu > %lu\n",
-				idx_value, size, track->indirect_draw_buffer_size);
+			dev_warn_once(p->dev, "DRAW_INDIRECT buffer too small %u + %llu > %lu\n",
+				      idx_value, size, track->indirect_draw_buffer_size);
 			return -EINVAL;
 		}
 
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	}
 	case PACKET3_DISPATCH_DIRECT:
 		if (pkt->count != 3) {
-			DRM_ERROR("bad DISPATCH_DIRECT\n");
+			dev_warn_once(p->dev, "bad DISPATCH_DIRECT\n");
 			return -EINVAL;
 		}
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
 			return r;
 		}
 		break;
 	case PACKET3_DISPATCH_INDIRECT:
 		if (pkt->count != 1) {
-			DRM_ERROR("bad DISPATCH_INDIRECT\n");
+			dev_warn_once(p->dev, "bad DISPATCH_INDIRECT\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad DISPATCH_INDIRECT\n");
+			dev_warn_once(p->dev, "bad DISPATCH_INDIRECT\n");
 			return -EINVAL;
 		}
 		ib[idx+0] = idx_value + (u32)(reloc->gpu_offset & 0xffffffff);
 		r = evergreen_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	case PACKET3_WAIT_REG_MEM:
 		if (pkt->count != 5) {
-			DRM_ERROR("bad WAIT_REG_MEM\n");
+			dev_warn_once(p->dev, "bad WAIT_REG_MEM\n");
 			return -EINVAL;
 		}
 		/* bit 4 is reg (0) or mem (1) */
@@ -2095,7 +2095,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad WAIT_REG_MEM\n");
+				dev_warn_once(p->dev, "bad WAIT_REG_MEM\n");
 				return -EINVAL;
 			}
 
@@ -2106,7 +2106,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffffc);
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
 		} else if (idx_value & 0x100) {
-			DRM_ERROR("cannot use PFP on REG wait\n");
+			dev_warn_once(p->dev, "cannot use PFP on REG wait\n");
 			return -EINVAL;
 		}
 		break;
@@ -2115,7 +2115,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		u32 command, size, info;
 		u64 offset, tmp;
 		if (pkt->count != 4) {
-			DRM_ERROR("bad CP DMA\n");
+			dev_warn_once(p->dev, "bad CP DMA\n");
 			return -EINVAL;
 		}
 		command = radeon_get_ib_value(p, idx+4);
@@ -2129,7 +2129,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		     (command & PACKET3_CP_DMA_CMD_SAS))) { /* src = register */
 			/* non mem to mem copies requires dw aligned count */
 			if (size % 4) {
-				DRM_ERROR("CP DMA command requires dw count alignment\n");
+				dev_warn_once(p->dev, "CP DMA command requires dw count alignment\n");
 				return -EINVAL;
 			}
 		}
@@ -2137,19 +2137,19 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* src address space is register */
 			/* GDS is ok */
 			if (((info & 0x60000000) >> 29) != 1) {
-				DRM_ERROR("CP DMA SAS not supported\n");
+				dev_warn_once(p->dev, "CP DMA SAS not supported\n");
 				return -EINVAL;
 			}
 		} else {
 			if (command & PACKET3_CP_DMA_CMD_SAIC) {
-				DRM_ERROR("CP DMA SAIC only supported for registers\n");
+				dev_warn_once(p->dev, "CP DMA SAIC only supported for registers\n");
 				return -EINVAL;
 			}
 			/* src address space is memory */
 			if (((info & 0x60000000) >> 29) == 0) {
 				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
-					DRM_ERROR("bad CP DMA SRC\n");
+					dev_warn_once(p->dev, "bad CP DMA SRC\n");
 					return -EINVAL;
 				}
 
@@ -2159,15 +2159,15 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				offset = reloc->gpu_offset + tmp;
 
 				if ((tmp + size) > radeon_bo_size(reloc->robj)) {
-					dev_warn(p->dev, "CP DMA src buffer too small (%llu %lu)\n",
-						 tmp + size, radeon_bo_size(reloc->robj));
+					dev_warn_once(p->dev, "CP DMA src buffer too small (%llu %lu)\n",
+						      tmp + size, radeon_bo_size(reloc->robj));
 					return -EINVAL;
 				}
 
 				ib[idx] = offset;
 				ib[idx+1] = (ib[idx+1] & 0xffffff00) | (upper_32_bits(offset) & 0xff);
 			} else if (((info & 0x60000000) >> 29) != 2) {
-				DRM_ERROR("bad CP DMA SRC_SEL\n");
+				dev_warn_once(p->dev, "bad CP DMA SRC_SEL\n");
 				return -EINVAL;
 			}
 		}
@@ -2175,19 +2175,19 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* dst address space is register */
 			/* GDS is ok */
 			if (((info & 0x00300000) >> 20) != 1) {
-				DRM_ERROR("CP DMA DAS not supported\n");
+				dev_warn_once(p->dev, "CP DMA DAS not supported\n");
 				return -EINVAL;
 			}
 		} else {
 			/* dst address space is memory */
 			if (command & PACKET3_CP_DMA_CMD_DAIC) {
-				DRM_ERROR("CP DMA DAIC only supported for registers\n");
+				dev_warn_once(p->dev, "CP DMA DAIC only supported for registers\n");
 				return -EINVAL;
 			}
 			if (((info & 0x00300000) >> 20) == 0) {
 				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
-					DRM_ERROR("bad CP DMA DST\n");
+					dev_warn_once(p->dev, "bad CP DMA DST\n");
 					return -EINVAL;
 				}
 
@@ -2197,15 +2197,15 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				offset = reloc->gpu_offset + tmp;
 
 				if ((tmp + size) > radeon_bo_size(reloc->robj)) {
-					dev_warn(p->dev, "CP DMA dst buffer too small (%llu %lu)\n",
-						 tmp + size, radeon_bo_size(reloc->robj));
+					dev_warn_once(p->dev, "CP DMA dst buffer too small (%llu %lu)\n",
+						      tmp + size, radeon_bo_size(reloc->robj));
 					return -EINVAL;
 				}
 
 				ib[idx+2] = offset;
 				ib[idx+3] = upper_32_bits(offset) & 0xff;
 			} else {
-				DRM_ERROR("bad CP DMA DST_SEL\n");
+				dev_warn_once(p->dev, "bad CP DMA DST_SEL\n");
 				return -EINVAL;
 			}
 		}
@@ -2213,13 +2213,13 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	}
 	case PACKET3_PFP_SYNC_ME:
 		if (pkt->count) {
-			DRM_ERROR("bad PFP_SYNC_ME\n");
+			dev_warn_once(p->dev, "bad PFP_SYNC_ME\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_SURFACE_SYNC:
 		if (pkt->count != 3) {
-			DRM_ERROR("bad SURFACE_SYNC\n");
+			dev_warn_once(p->dev, "bad SURFACE_SYNC\n");
 			return -EINVAL;
 		}
 		/* 0xffffffff/0x0 is flush all cache flag */
@@ -2227,7 +2227,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		    radeon_get_ib_value(p, idx + 2) != 0) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad SURFACE_SYNC\n");
+				dev_warn_once(p->dev, "bad SURFACE_SYNC\n");
 				return -EINVAL;
 			}
 			ib[idx+2] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -2235,7 +2235,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case PACKET3_EVENT_WRITE:
 		if (pkt->count != 2 && pkt->count != 0) {
-			DRM_ERROR("bad EVENT_WRITE\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE\n");
 			return -EINVAL;
 		}
 		if (pkt->count) {
@@ -2243,7 +2243,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad EVENT_WRITE\n");
+				dev_warn_once(p->dev, "bad EVENT_WRITE\n");
 				return -EINVAL;
 			}
 			offset = reloc->gpu_offset +
@@ -2259,12 +2259,12 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 4) {
-			DRM_ERROR("bad EVENT_WRITE_EOP\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE_EOP\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad EVENT_WRITE_EOP\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE_EOP\n");
 			return -EINVAL;
 		}
 
@@ -2281,12 +2281,12 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 3) {
-			DRM_ERROR("bad EVENT_WRITE_EOS\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE_EOS\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad EVENT_WRITE_EOS\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE_EOS\n");
 			return -EINVAL;
 		}
 
@@ -2304,7 +2304,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_CONFIG_REG_START) ||
 		    (start_reg >= PACKET3_SET_CONFIG_REG_END) ||
 		    (end_reg >= PACKET3_SET_CONFIG_REG_END)) {
-			DRM_ERROR("bad PACKET3_SET_CONFIG_REG\n");
+			dev_warn_once(p->dev, "bad PACKET3_SET_CONFIG_REG\n");
 			return -EINVAL;
 		}
 		for (reg = start_reg, idx++; reg <= end_reg; reg += 4, idx++) {
@@ -2321,7 +2321,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_CONTEXT_REG_START) ||
 		    (start_reg >= PACKET3_SET_CONTEXT_REG_END) ||
 		    (end_reg >= PACKET3_SET_CONTEXT_REG_END)) {
-			DRM_ERROR("bad PACKET3_SET_CONTEXT_REG\n");
+			dev_warn_once(p->dev, "bad PACKET3_SET_CONTEXT_REG\n");
 			return -EINVAL;
 		}
 		for (reg = start_reg, idx++; reg <= end_reg; reg += 4, idx++) {
@@ -2334,7 +2334,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case PACKET3_SET_RESOURCE:
 		if (pkt->count % 8) {
-			DRM_ERROR("bad SET_RESOURCE\n");
+			dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 			return -EINVAL;
 		}
 		start_reg = (idx_value << 2) + PACKET3_SET_RESOURCE_START;
@@ -2342,7 +2342,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_RESOURCE_START) ||
 		    (start_reg >= PACKET3_SET_RESOURCE_END) ||
 		    (end_reg >= PACKET3_SET_RESOURCE_END)) {
-			DRM_ERROR("bad SET_RESOURCE\n");
+			dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 			return -EINVAL;
 		}
 		for (i = 0; i < (pkt->count / 8); i++) {
@@ -2355,7 +2355,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				/* tex base */
 				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
-					DRM_ERROR("bad SET_RESOURCE (tex)\n");
+					dev_warn_once(p->dev, "bad SET_RESOURCE (tex)\n");
 					return -EINVAL;
 				}
 				if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
@@ -2392,7 +2392,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				} else {
 					r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 					if (r) {
-						DRM_ERROR("bad SET_RESOURCE (tex)\n");
+						dev_warn_once(p->dev, "bad SET_RESOURCE (tex)\n");
 						return -EINVAL;
 					}
 					moffset = (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -2411,14 +2411,15 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				/* vtx base */
 				r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 				if (r) {
-					DRM_ERROR("bad SET_RESOURCE (vtx)\n");
+					dev_warn_once(p->dev, "bad SET_RESOURCE (vtx)\n");
 					return -EINVAL;
 				}
 				offset = radeon_get_ib_value(p, idx+1+(i*8)+0);
 				size = radeon_get_ib_value(p, idx+1+(i*8)+1);
 				if (p->rdev && (size + offset) > radeon_bo_size(reloc->robj)) {
 					/* force size to size of the buffer */
-					dev_warn_ratelimited(p->dev, "vbo resource seems too big for the bo\n");
+					dev_warn_once(p->dev, "vbo resource seems too big (%d) for the bo (%ld)\n",
+						      size + offset, radeon_bo_size(reloc->robj));
 					ib[idx+1+(i*8)+1] = radeon_bo_size(reloc->robj) - offset;
 				}
 
@@ -2431,7 +2432,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			case SQ_TEX_VTX_INVALID_TEXTURE:
 			case SQ_TEX_VTX_INVALID_BUFFER:
 			default:
-				DRM_ERROR("bad SET_RESOURCE\n");
+				dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 				return -EINVAL;
 			}
 		}
@@ -2445,7 +2446,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_BOOL_CONST_START) ||
 		    (start_reg >= PACKET3_SET_BOOL_CONST_END) ||
 		    (end_reg >= PACKET3_SET_BOOL_CONST_END)) {
-			DRM_ERROR("bad SET_BOOL_CONST\n");
+			dev_warn_once(p->dev, "bad SET_BOOL_CONST\n");
 			return -EINVAL;
 		}
 		break;
@@ -2455,7 +2456,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_LOOP_CONST_START) ||
 		    (start_reg >= PACKET3_SET_LOOP_CONST_END) ||
 		    (end_reg >= PACKET3_SET_LOOP_CONST_END)) {
-			DRM_ERROR("bad SET_LOOP_CONST\n");
+			dev_warn_once(p->dev, "bad SET_LOOP_CONST\n");
 			return -EINVAL;
 		}
 		break;
@@ -2465,13 +2466,13 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_CTL_CONST_START) ||
 		    (start_reg >= PACKET3_SET_CTL_CONST_END) ||
 		    (end_reg >= PACKET3_SET_CTL_CONST_END)) {
-			DRM_ERROR("bad SET_CTL_CONST\n");
+			dev_warn_once(p->dev, "bad SET_CTL_CONST\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_SET_SAMPLER:
 		if (pkt->count % 3) {
-			DRM_ERROR("bad SET_SAMPLER\n");
+			dev_warn_once(p->dev, "bad SET_SAMPLER\n");
 			return -EINVAL;
 		}
 		start_reg = (idx_value << 2) + PACKET3_SET_SAMPLER_START;
@@ -2479,13 +2480,13 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_SAMPLER_START) ||
 		    (start_reg >= PACKET3_SET_SAMPLER_END) ||
 		    (end_reg >= PACKET3_SET_SAMPLER_END)) {
-			DRM_ERROR("bad SET_SAMPLER\n");
+			dev_warn_once(p->dev, "bad SET_SAMPLER\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_STRMOUT_BUFFER_UPDATE:
 		if (pkt->count != 4) {
-			DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (invalid count)\n");
+			dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE (invalid count)\n");
 			return -EINVAL;
 		}
 		/* Updating memory at DST_ADDRESS. */
@@ -2493,14 +2494,14 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			u64 offset;
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n");
+				dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+1);
 			offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE dst bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE dst bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2512,14 +2513,14 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			u64 offset;
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n");
+				dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+3);
 			offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE src bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE src bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2532,23 +2533,23 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		u64 offset;
 
 		if (pkt->count != 3) {
-			DRM_ERROR("bad MEM_WRITE (invalid count)\n");
+			dev_warn_once(p->dev, "bad MEM_WRITE (invalid count)\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad MEM_WRITE (missing reloc)\n");
+			dev_warn_once(p->dev, "bad MEM_WRITE (missing reloc)\n");
 			return -EINVAL;
 		}
 		offset = radeon_get_ib_value(p, idx+0);
 		offset += ((u64)(radeon_get_ib_value(p, idx+1) & 0xff)) << 32UL;
 		if (offset & 0x7) {
-			DRM_ERROR("bad MEM_WRITE (address not qwords aligned)\n");
+			dev_warn_once(p->dev, "bad MEM_WRITE (address not qwords aligned)\n");
 			return -EINVAL;
 		}
 		if ((offset + 8) > radeon_bo_size(reloc->robj)) {
-			DRM_ERROR("bad MEM_WRITE bo too small: 0x%llx, 0x%lx\n",
-				  offset + 8, radeon_bo_size(reloc->robj));
+			dev_warn_once(p->dev, "bad MEM_WRITE bo too small: 0x%llx, 0x%lx\n",
+				      offset + 8, radeon_bo_size(reloc->robj));
 			return -EINVAL;
 		}
 		offset += reloc->gpu_offset;
@@ -2558,7 +2559,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	}
 	case PACKET3_COPY_DW:
 		if (pkt->count != 4) {
-			DRM_ERROR("bad COPY_DW (invalid count)\n");
+			dev_warn_once(p->dev, "bad COPY_DW (invalid count)\n");
 			return -EINVAL;
 		}
 		if (idx_value & 0x1) {
@@ -2566,14 +2567,14 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* SRC is memory. */
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad COPY_DW (missing src reloc)\n");
+				dev_warn_once(p->dev, "bad COPY_DW (missing src reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+1);
 			offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad COPY_DW src bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad COPY_DW src bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2583,8 +2584,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* SRC is a reg. */
 			reg = radeon_get_ib_value(p, idx+1) << 2;
 			if (!evergreen_is_safe_reg(p, reg)) {
-				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
-					 reg, idx + 1);
+				dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n",
+					      reg, idx + 1);
 				return -EINVAL;
 			}
 		}
@@ -2593,14 +2594,14 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* DST is memory. */
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad COPY_DW (missing dst reloc)\n");
+				dev_warn_once(p->dev, "bad COPY_DW (missing dst reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+3);
 			offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad COPY_DW dst bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad COPY_DW dst bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2610,8 +2611,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* DST is a reg. */
 			reg = radeon_get_ib_value(p, idx+3) << 2;
 			if (!evergreen_is_safe_reg(p, reg)) {
-				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
-					 reg, idx + 3);
+				dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n",
+					      reg, idx + 3);
 				return -EINVAL;
 			}
 		}
@@ -2622,7 +2623,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		uint32_t allowed_reg_base;
 		uint32_t source_sel;
 		if (pkt->count != 2) {
-			DRM_ERROR("bad SET_APPEND_CNT (invalid count)\n");
+			dev_warn_once(p->dev, "bad SET_APPEND_CNT (invalid count)\n");
 			return -EINVAL;
 		}
 
@@ -2632,8 +2633,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 
 		areg = idx_value >> 16;
 		if (areg < allowed_reg_base || areg > (allowed_reg_base + 11)) {
-			dev_warn(p->dev, "forbidden register for append cnt 0x%08x at %d\n",
-				 areg, idx);
+			dev_warn_once(p->dev, "forbidden register for append cnt 0x%08x at %d\n",
+				      areg, idx);
 			return -EINVAL;
 		}
 
@@ -2643,7 +2644,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			uint32_t swap;
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad SET_APPEND_CNT (missing reloc)\n");
+				dev_warn_once(p->dev, "bad SET_APPEND_CNT (missing reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx + 1);
@@ -2656,7 +2657,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			ib[idx+1] = (offset & 0xfffffffc) | swap;
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
 		} else {
-			DRM_ERROR("bad SET_APPEND_CNT (unsupported operation)\n");
+			dev_warn_once(p->dev, "bad SET_APPEND_CNT (unsupported operation)\n");
 			return -EINVAL;
 		}
 		break;
@@ -2666,23 +2667,23 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		u64 offset;
 
 		if (pkt->count != 2) {
-			DRM_ERROR("bad COND_EXEC (invalid count)\n");
+			dev_warn_once(p->dev, "bad COND_EXEC (invalid count)\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("bad COND_EXEC (missing reloc)\n");
+			dev_warn_once(p->dev, "bad COND_EXEC (missing reloc)\n");
 			return -EINVAL;
 		}
 		offset = radeon_get_ib_value(p, idx + 0);
 		offset += ((u64)(radeon_get_ib_value(p, idx + 1) & 0xff)) << 32UL;
 		if (offset & 0x7) {
-			DRM_ERROR("bad COND_EXEC (address not qwords aligned)\n");
+			dev_warn_once(p->dev, "bad COND_EXEC (address not qwords aligned)\n");
 			return -EINVAL;
 		}
 		if ((offset + 8) > radeon_bo_size(reloc->robj)) {
-			DRM_ERROR("bad COND_EXEC bo too small: 0x%llx, 0x%lx\n",
-				  offset + 8, radeon_bo_size(reloc->robj));
+			dev_warn_once(p->dev, "bad COND_EXEC bo too small: 0x%llx, 0x%lx\n",
+				      offset + 8, radeon_bo_size(reloc->robj));
 			return -EINVAL;
 		}
 		offset += reloc->gpu_offset;
@@ -2692,7 +2693,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	}
 	case PACKET3_COND_WRITE:
 		if (pkt->count != 7) {
-			DRM_ERROR("bad COND_WRITE (invalid count)\n");
+			dev_warn_once(p->dev, "bad COND_WRITE (invalid count)\n");
 			return -EINVAL;
 		}
 		if (idx_value & 0x10) {
@@ -2700,14 +2701,14 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* POLL is memory. */
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad COND_WRITE (missing src reloc)\n");
+				dev_warn_once(p->dev, "bad COND_WRITE (missing src reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx + 1);
 			offset += ((u64)(radeon_get_ib_value(p, idx + 2) & 0xff)) << 32;
 			if ((offset + 8) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad COND_WRITE src bo too small: 0x%llx, 0x%lx\n",
-					  offset + 8, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad COND_WRITE src bo too small: 0x%llx, 0x%lx\n",
+					      offset + 8, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2717,8 +2718,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* POLL is a reg. */
 			reg = radeon_get_ib_value(p, idx + 1) << 2;
 			if (!evergreen_is_safe_reg(p, reg)) {
-				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
-					 reg, idx + 1);
+				dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n",
+					      reg, idx + 1);
 				return -EINVAL;
 			}
 		}
@@ -2727,14 +2728,14 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* WRITE is memory. */
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("bad COND_WRITE (missing dst reloc)\n");
+				dev_warn_once(p->dev, "bad COND_WRITE (missing dst reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx + 5);
 			offset += ((u64)(radeon_get_ib_value(p, idx + 6) & 0xff)) << 32;
 			if ((offset + 8) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad COND_WRITE dst bo too small: 0x%llx, 0x%lx\n",
-					  offset + 8, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad COND_WRITE dst bo too small: 0x%llx, 0x%lx\n",
+					      offset + 8, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2744,8 +2745,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			/* WRITE is a reg. */
 			reg = radeon_get_ib_value(p, idx + 5) << 2;
 			if (!evergreen_is_safe_reg(p, reg)) {
-				dev_warn(p->dev, "forbidden register 0x%08x at %d\n",
-					 reg, idx + 5);
+				dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n",
+					      reg, idx + 5);
 				return -EINVAL;
 			}
 		}
@@ -2753,7 +2754,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_NOP:
 		break;
 	default:
-		DRM_ERROR("Packet3 opcode %x not supported\n", pkt->opcode);
+		dev_warn_once(p->dev, "Packet3 opcode %x not supported\n", pkt->opcode);
 		return -EINVAL;
 	}
 	return 0;
@@ -2853,7 +2854,7 @@ int evergreen_cs_parse(struct radeon_cs_parser *p)
 			r = evergreen_packet3_check(p, &pkt);
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d !\n", pkt.type);
+			dev_warn_once(p->dev, "Unknown packet type %d !\n", pkt.type);
 			kfree(p->track);
 			p->track = NULL;
 			return -EINVAL;
@@ -2896,8 +2897,8 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 
 	do {
 		if (p->idx >= ib_chunk->length_dw) {
-			DRM_ERROR("Can not parse packet at %d after CS end %d !\n",
-				  p->idx, ib_chunk->length_dw);
+			dev_warn_once(p->dev, "Can not parse packet at %d after CS end %d !\n",
+				      p->idx, ib_chunk->length_dw);
 			return -EINVAL;
 		}
 		idx = p->idx;
@@ -2910,7 +2911,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 		case DMA_PACKET_WRITE:
 			r = r600_dma_cs_next_reloc(p, &dst_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_WRITE\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_WRITE\n");
 				return -EINVAL;
 			}
 			switch (sub_cmd) {
@@ -2932,24 +2933,24 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				p->idx += count + 3;
 				break;
 			default:
-				DRM_ERROR("bad DMA_PACKET_WRITE [%6d] 0x%08x sub cmd is not 0 or 8\n", idx, header);
+				dev_warn_once(p->dev, "bad DMA_PACKET_WRITE [%6d] 0x%08x sub cmd is not 0 or 8\n", idx, header);
 				return -EINVAL;
 			}
 			if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-				dev_warn(p->dev, "DMA write buffer too small (%llu %lu)\n",
-					 dst_offset, radeon_bo_size(dst_reloc->robj));
+				dev_warn_once(p->dev, "DMA write buffer too small (%llu %lu)\n",
+					      dst_offset, radeon_bo_size(dst_reloc->robj));
 				return -EINVAL;
 			}
 			break;
 		case DMA_PACKET_COPY:
 			r = r600_dma_cs_next_reloc(p, &src_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_COPY\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_COPY\n");
 				return -EINVAL;
 			}
 			r = r600_dma_cs_next_reloc(p, &dst_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_COPY\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_COPY\n");
 				return -EINVAL;
 			}
 			switch (sub_cmd) {
@@ -2961,13 +2962,13 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				dst_offset = radeon_get_ib_value(p, idx+1);
 				dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, dw src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, dw src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, dw dst buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, dw dst buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(dst_reloc->gpu_offset & 0xfffffffc);
@@ -3001,13 +3002,13 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 					ib[idx+1] += (u32)(dst_reloc->gpu_offset >> 8);
 				}
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, dst buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, dst buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				p->idx += 9;
@@ -3020,13 +3021,13 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				dst_offset = radeon_get_ib_value(p, idx+1);
 				dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0xff)) << 32;
 				if ((src_offset + count) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, byte src buffer too small (%llu %lu)\n",
-							src_offset + count, radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, byte src buffer too small (%llu %lu)\n",
+						      src_offset + count, radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + count) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, byte dst buffer too small (%llu %lu)\n",
-							dst_offset + count, radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, byte dst buffer too small (%llu %lu)\n",
+						      dst_offset + count, radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(dst_reloc->gpu_offset & 0xffffffff);
@@ -3039,7 +3040,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			case 0x41:
 				/* L2L, partial */
 				if (p->family < CHIP_CAYMAN) {
-					DRM_ERROR("L2L Partial is cayman only !\n");
+					dev_warn_once(p->dev, "L2L Partial is cayman only !\n");
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(src_reloc->gpu_offset & 0xffffffff);
@@ -3054,7 +3055,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				/* L2L, dw, broadcast */
 				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
 				if (r) {
-					DRM_ERROR("bad L2L, dw, broadcast DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2L, dw, broadcast DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				dst_offset = radeon_get_ib_value(p, idx+1);
@@ -3064,18 +3065,18 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				src_offset = radeon_get_ib_value(p, idx+3);
 				src_offset |= ((u64)(radeon_get_ib_value(p, idx+6) & 0xff)) << 32;
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, dw, broadcast src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, dw, broadcast src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, dw, broadcast dst buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, dw, broadcast dst buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2L, dw, broadcast dst2 buffer too small (%llu %lu)\n",
-							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2L, dw, broadcast dst2 buffer too small (%llu %lu)\n",
+						      dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(dst_reloc->gpu_offset & 0xfffffffc);
@@ -3089,12 +3090,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			/* Copy L2T Frame to Field */
 			case 0x48:
 				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
-					DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2T, frame to fields DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
 				if (r) {
-					DRM_ERROR("bad L2T, frame to fields DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2T, frame to fields DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				dst_offset = radeon_get_ib_value(p, idx+1);
@@ -3104,18 +3105,18 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				src_offset = radeon_get_ib_value(p, idx+8);
 				src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, frame to fields src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, frame to fields src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
-							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, frame to fields buffer too small (%llu %lu)\n",
+						      dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(dst_reloc->gpu_offset >> 8);
@@ -3128,7 +3129,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			case 0x49:
 				/* L2T, T2L partial */
 				if (p->family < CHIP_CAYMAN) {
-					DRM_ERROR("L2T, T2L Partial is cayman only !\n");
+					dev_warn_once(p->dev, "L2T, T2L Partial is cayman only !\n");
 					return -EINVAL;
 				}
 				/* detile bit */
@@ -3151,12 +3152,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			case 0x4b:
 				/* L2T, broadcast */
 				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
-					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2T, broadcast DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
 				if (r) {
-					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2T, broadcast DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				dst_offset = radeon_get_ib_value(p, idx+1);
@@ -3166,18 +3167,18 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				src_offset = radeon_get_ib_value(p, idx+8);
 				src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
-							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
+						      dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(dst_reloc->gpu_offset >> 8);
@@ -3212,13 +3213,13 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 					ib[idx+1] += (u32)(dst_reloc->gpu_offset >> 8);
 				}
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, T2L src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, T2L src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, T2L dst buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, T2L dst buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				p->idx += 9;
@@ -3227,7 +3228,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			case 0x4d:
 				/* T2T partial */
 				if (p->family < CHIP_CAYMAN) {
-					DRM_ERROR("L2T, T2L Partial is cayman only !\n");
+					dev_warn_once(p->dev, "L2T, T2L Partial is cayman only !\n");
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(src_reloc->gpu_offset >> 8);
@@ -3238,12 +3239,12 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			case 0x4f:
 				/* L2T, broadcast */
 				if (radeon_get_ib_value(p, idx + 2) & (1 << 31)) {
-					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2T, broadcast DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				r = r600_dma_cs_next_reloc(p, &dst2_reloc);
 				if (r) {
-					DRM_ERROR("bad L2T, broadcast DMA_PACKET_COPY\n");
+					dev_warn_once(p->dev, "bad L2T, broadcast DMA_PACKET_COPY\n");
 					return -EINVAL;
 				}
 				dst_offset = radeon_get_ib_value(p, idx+1);
@@ -3253,18 +3254,18 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				src_offset = radeon_get_ib_value(p, idx+8);
 				src_offset |= ((u64)(radeon_get_ib_value(p, idx+9) & 0xff)) << 32;
 				if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
-							src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, broadcast src buffer too small (%llu %lu)\n",
+						      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
-							dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, broadcast dst buffer too small (%llu %lu)\n",
+						      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 					return -EINVAL;
 				}
 				if ((dst2_offset + (count * 4)) > radeon_bo_size(dst2_reloc->robj)) {
-					dev_warn(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
-							dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
+					dev_warn_once(p->dev, "DMA L2T, broadcast dst2 buffer too small (%llu %lu)\n",
+						      dst2_offset + (count * 4), radeon_bo_size(dst2_reloc->robj));
 					return -EINVAL;
 				}
 				ib[idx+1] += (u32)(dst_reloc->gpu_offset >> 8);
@@ -3274,21 +3275,21 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 				p->idx += 10;
 				break;
 			default:
-				DRM_ERROR("bad DMA_PACKET_COPY [%6d] 0x%08x invalid sub cmd\n", idx, header);
+				dev_warn_once(p->dev, "bad DMA_PACKET_COPY [%6d] 0x%08x invalid sub cmd\n", idx, header);
 				return -EINVAL;
 			}
 			break;
 		case DMA_PACKET_CONSTANT_FILL:
 			r = r600_dma_cs_next_reloc(p, &dst_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_CONSTANT_FILL\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_CONSTANT_FILL\n");
 				return -EINVAL;
 			}
 			dst_offset = radeon_get_ib_value(p, idx+1);
 			dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0x00ff0000)) << 16;
 			if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-				dev_warn(p->dev, "DMA constant fill buffer too small (%llu %lu)\n",
-					 dst_offset, radeon_bo_size(dst_reloc->robj));
+				dev_warn_once(p->dev, "DMA constant fill buffer too small (%llu %lu)\n",
+					      dst_offset, radeon_bo_size(dst_reloc->robj));
 				return -EINVAL;
 			}
 			ib[idx+1] += (u32)(dst_reloc->gpu_offset & 0xfffffffc);
@@ -3299,7 +3300,7 @@ int evergreen_dma_cs_parse(struct radeon_cs_parser *p)
 			p->idx += 1;
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx);
+			dev_warn_once(p->dev, "Unknown packet type %d at %d !\n", cmd, idx);
 			return -EINVAL;
 		}
 	} while (p->idx < p->chunk_ib->length_dw);
@@ -3430,7 +3431,7 @@ static bool evergreen_vm_reg_valid(u32 reg)
 	case CAYMAN_SQ_EX_ALLOC_TABLE_SLOTS:
 		return true;
 	default:
-		DRM_ERROR("Invalid register 0x%x in CS\n", reg);
+		DRM_DEBUG("Invalid register 0x%x in CS\n", reg);
 		return false;
 	}
 }
@@ -3448,7 +3449,7 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 		break;
 	case PACKET3_SET_BASE:
 		if (idx_value != 1) {
-			DRM_ERROR("bad SET_BASE");
+			dev_warn_once(rdev->dev, "bad SET_BASE");
 			return -EINVAL;
 		}
 		break;
@@ -3519,7 +3520,7 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 		if ((start_reg < PACKET3_SET_CONFIG_REG_START) ||
 		    (start_reg >= PACKET3_SET_CONFIG_REG_END) ||
 		    (end_reg >= PACKET3_SET_CONFIG_REG_END)) {
-			DRM_ERROR("bad PACKET3_SET_CONFIG_REG\n");
+			dev_warn_once(rdev->dev, "bad PACKET3_SET_CONFIG_REG\n");
 			return -EINVAL;
 		}
 		for (i = 0; i < pkt->count; i++) {
@@ -3539,7 +3540,7 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 		     (command & PACKET3_CP_DMA_CMD_SAS))) { /* src = register */
 			/* non mem to mem copies requires dw aligned count */
 			if ((command & 0x1fffff) % 4) {
-				DRM_ERROR("CP DMA command requires dw count alignment\n");
+				dev_warn_once(rdev->dev, "CP DMA command requires dw count alignment\n");
 				return -EINVAL;
 			}
 		}
@@ -3550,14 +3551,14 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 				if (command & PACKET3_CP_DMA_CMD_SAIC) {
 					reg = start_reg;
 					if (!evergreen_vm_reg_valid(reg)) {
-						DRM_ERROR("CP DMA Bad SRC register\n");
+						dev_warn_once(rdev->dev, "CP DMA Bad SRC register\n");
 						return -EINVAL;
 					}
 				} else {
 					for (i = 0; i < (command & 0x1fffff); i++) {
 						reg = start_reg + (4 * i);
 						if (!evergreen_vm_reg_valid(reg)) {
-							DRM_ERROR("CP DMA Bad SRC register\n");
+							dev_warn_once(rdev->dev, "CP DMA Bad SRC register\n");
 							return -EINVAL;
 						}
 					}
@@ -3571,14 +3572,14 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 				if (command & PACKET3_CP_DMA_CMD_DAIC) {
 					reg = start_reg;
 					if (!evergreen_vm_reg_valid(reg)) {
-						DRM_ERROR("CP DMA Bad DST register\n");
+						dev_warn_once(rdev->dev, "CP DMA Bad DST register\n");
 						return -EINVAL;
 					}
 				} else {
 					for (i = 0; i < (command & 0x1fffff); i++) {
 						reg = start_reg + (4 * i);
 						if (!evergreen_vm_reg_valid(reg)) {
-							DRM_ERROR("CP DMA Bad DST register\n");
+							dev_warn_once(rdev->dev, "CP DMA Bad DST register\n");
 							return -EINVAL;
 						}
 					}
@@ -3591,7 +3592,7 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 		uint32_t allowed_reg_base;
 
 		if (pkt->count != 2) {
-			DRM_ERROR("bad SET_APPEND_CNT (invalid count)\n");
+			dev_warn_once(rdev->dev, "bad SET_APPEND_CNT (invalid count)\n");
 			return -EINVAL;
 		}
 
@@ -3601,8 +3602,8 @@ static int evergreen_vm_packet3_check(struct radeon_device *rdev,
 
 		areg = idx_value >> 16;
 		if (areg < allowed_reg_base || areg > (allowed_reg_base + 11)) {
-			DRM_ERROR("forbidden register for append cnt 0x%08x at %d\n",
-				  areg, idx);
+			dev_warn_once(rdev->dev, "forbidden register for append cnt 0x%08x at %d\n",
+				      areg, idx);
 			return -EINVAL;
 		}
 		break;
@@ -3681,7 +3682,9 @@ int evergreen_dma_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 				idx += count + 3;
 				break;
 			default:
-				DRM_ERROR("bad DMA_PACKET_WRITE [%6d] 0x%08x sub cmd is not 0 or 8\n", idx, ib->ptr[idx]);
+				dev_warn_once(rdev->dev,
+					      "bad DMA_PACKET_WRITE [%6d] 0x%08x sub cmd is not 0 or 8\n",
+					      idx, ib->ptr[idx]);
 				return -EINVAL;
 			}
 			break;
@@ -3732,7 +3735,9 @@ int evergreen_dma_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 				idx += 10;
 				break;
 			default:
-				DRM_ERROR("bad DMA_PACKET_COPY [%6d] 0x%08x invalid sub cmd\n", idx, ib->ptr[idx]);
+				dev_warn_once(rdev->dev,
+					      "bad DMA_PACKET_COPY [%6d] 0x%08x invalid sub cmd\n",
+					      idx, ib->ptr[idx]);
 				return -EINVAL;
 			}
 			break;
@@ -3743,7 +3748,7 @@ int evergreen_dma_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
 			idx += 1;
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx);
+			dev_warn_once(rdev->dev, "Unknown packet type %d at %d !\n", cmd, idx);
 			return -EINVAL;
 		}
 	} while (idx < ib->length_dw);
diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
index 80703417d8a1..07a9c523a17a 100644
--- a/drivers/gpu/drm/radeon/r100.c
+++ b/drivers/gpu/drm/radeon/r100.c
@@ -1298,8 +1298,8 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p,
 
 	r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 	if (r) {
-		DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-			  idx, reg);
+		dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+			      idx, reg);
 		radeon_cs_dump_packet(p, pkt);
 		return r;
 	}
@@ -1313,7 +1313,7 @@ int r100_reloc_pitch_offset(struct radeon_cs_parser *p,
 			tile_flags |= RADEON_DST_TILE_MACRO;
 		if (reloc->tiling_flags & RADEON_TILING_MICRO) {
 			if (reg == RADEON_SRC_PITCH_OFFSET) {
-				DRM_ERROR("Cannot src blit from microtiled surface\n");
+				dev_warn_once(p->dev, "Cannot src blit from microtiled surface\n");
 				radeon_cs_dump_packet(p, pkt);
 				return -EINVAL;
 			}
@@ -1342,8 +1342,8 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 	track = (struct r100_cs_track *)p->track;
 	c = radeon_get_ib_value(p, idx++) & 0x1F;
 	if (c > 16) {
-	    DRM_ERROR("Only 16 vertex buffers are allowed %d\n",
-		      pkt->opcode);
+	    dev_warn_once(p->dev, "Only 16 vertex buffers are allowed %d\n",
+			  pkt->opcode);
 	    radeon_cs_dump_packet(p, pkt);
 	    return -EINVAL;
 	}
@@ -1351,8 +1351,8 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 	for (i = 0; i < (c - 1); i += 2, idx += 3) {
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for packet3 %d\n",
-				  pkt->opcode);
+			dev_warn_once(p->dev, "No reloc for packet3 %d\n",
+				      pkt->opcode);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1364,8 +1364,8 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 		track->arrays[i + 0].esize &= 0x7F;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for packet3 %d\n",
-				  pkt->opcode);
+			dev_warn_once(p->dev, "No reloc for packet3 %d\n",
+				      pkt->opcode);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1377,8 +1377,8 @@ int r100_packet3_load_vbpntr(struct radeon_cs_parser *p,
 	if (c & 1) {
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for packet3 %d\n",
-					  pkt->opcode);
+			dev_warn_once(p->dev, "No reloc for packet3 %d\n",
+				      pkt->opcode);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1470,12 +1470,12 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	/* check its a wait until and only 1 count */
 	if (waitreloc.reg != RADEON_WAIT_UNTIL ||
 	    waitreloc.count != 0) {
-		DRM_ERROR("vline wait had illegal wait until segment\n");
+		dev_warn_once(p->dev, "vline wait had illegal wait until segment\n");
 		return -EINVAL;
 	}
 
 	if (radeon_get_ib_value(p, waitreloc.idx + 1) != RADEON_WAIT_CRTC_VLINE) {
-		DRM_ERROR("vline wait had illegal wait until\n");
+		dev_warn_once(p->dev, "vline wait had illegal wait until\n");
 		return -EINVAL;
 	}
 
@@ -1493,7 +1493,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 	reg = R100_CP_PACKET0_GET_REG(header);
 	crtc = drm_crtc_find(rdev_to_drm(p->rdev), p->filp, crtc_id);
 	if (!crtc) {
-		DRM_ERROR("cannot find crtc %d\n", crtc_id);
+		dev_warn_once(p->dev, "cannot find crtc %d\n", crtc_id);
 		return -ENOENT;
 	}
 	radeon_crtc = to_radeon_crtc(crtc);
@@ -1514,7 +1514,7 @@ int r100_cs_packet_parse_vline(struct radeon_cs_parser *p)
 			header |= RADEON_CRTC2_GUI_TRIG_VLINE >> 2;
 			break;
 		default:
-			DRM_ERROR("unknown crtc reloc\n");
+			dev_warn_once(p->dev, "unknown crtc reloc\n");
 			return -EINVAL;
 		}
 		ib[h_idx] = header;
@@ -1599,7 +1599,7 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_CRTC_GUI_TRIG_VLINE:
 		r = r100_cs_packet_parse_vline(p);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
 				  idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
@@ -1616,8 +1616,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_DEPTHOFFSET:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1629,8 +1629,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_COLOROFFSET:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1645,8 +1645,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - RADEON_PP_TXOFFSET_0) / 24;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1672,8 +1672,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - RADEON_PP_CUBIC_OFFSET_T0_0) / 4;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1690,8 +1690,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - RADEON_PP_CUBIC_OFFSET_T1_0) / 4;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1708,8 +1708,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - RADEON_PP_CUBIC_OFFSET_T2_0) / 4;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1726,8 +1726,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_COLORPITCH:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1768,8 +1768,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 			track->cb[0].cpp = 4;
 			break;
 		default:
-			DRM_ERROR("Invalid color buffer format (%d) !\n",
-				  ((idx_value >> RADEON_RB3D_COLOR_FORMAT_SHIFT) & 0x1f));
+			dev_warn_once(p->dev, "Invalid color buffer format (%d) !\n",
+				      ((idx_value >> RADEON_RB3D_COLOR_FORMAT_SHIFT) & 0x1f));
 			return -EINVAL;
 		}
 		track->z_enabled = !!(idx_value & RADEON_Z_ENABLE);
@@ -1797,8 +1797,8 @@ static int r100_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_ZPASS_ADDR:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1927,10 +1927,10 @@ int r100_cs_track_check_pkt3_indx_buffer(struct radeon_cs_parser *p,
 	idx = pkt->idx + 1;
 	value = radeon_get_ib_value(p, idx + 2);
 	if ((value + 1) > radeon_bo_size(robj)) {
-		DRM_ERROR("[drm] Buffer too small for PACKET3 INDX_BUFFER "
-			  "(need %u have %lu) !\n",
-			  value + 1,
-			  radeon_bo_size(robj));
+		dev_warn_once(p->dev, "[drm] Buffer too small for PACKET3 INDX_BUFFER "
+			      "(need %u have %lu) !\n",
+			      value + 1,
+			      radeon_bo_size(robj));
 		return -EINVAL;
 	}
 	return 0;
@@ -1957,7 +1957,7 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_INDX_BUFFER:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for packet3 %d\n", pkt->opcode);
+			dev_warn_once(p->dev, "No reloc for packet3 %d\n", pkt->opcode);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1971,7 +1971,7 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 		/* 3D_RNDR_GEN_INDX_PRIM on r100/r200 */
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for packet3 %d\n", pkt->opcode);
+			dev_warn_once(p->dev, "No reloc for packet3 %d\n", pkt->opcode);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1992,7 +1992,7 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case PACKET3_3D_DRAW_IMMD:
 		if (((radeon_get_ib_value(p, idx + 1) >> 4) & 0x3) != 3) {
-			DRM_ERROR("PRIM_WALK must be 3 for IMMD draw\n");
+			dev_warn_once(p->dev, "PRIM_WALK must be 3 for IMMD draw\n");
 			return -EINVAL;
 		}
 		track->vtx_size = r100_get_vtx_size(radeon_get_ib_value(p, idx + 0));
@@ -2005,7 +2005,7 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 		/* triggers drawing using in-packet vertex data */
 	case PACKET3_3D_DRAW_IMMD_2:
 		if (((radeon_get_ib_value(p, idx) >> 4) & 0x3) != 3) {
-			DRM_ERROR("PRIM_WALK must be 3 for IMMD draw\n");
+			dev_warn_once(p->dev, "PRIM_WALK must be 3 for IMMD draw\n");
 			return -EINVAL;
 		}
 		track->vap_vf_cntl = radeon_get_ib_value(p, idx);
@@ -2051,7 +2051,7 @@ static int r100_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_NOP:
 		break;
 	default:
-		DRM_ERROR("Packet3 opcode %x not supported\n", pkt->opcode);
+		dev_warn_once(p->dev, "Packet3 opcode %x not supported\n", pkt->opcode);
 		return -EINVAL;
 	}
 	return 0;
@@ -2093,8 +2093,8 @@ int r100_cs_parse(struct radeon_cs_parser *p)
 			r = r100_packet3_check(p, &pkt);
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d !\n",
-				  pkt.type);
+			dev_warn_once(p->dev, "Unknown packet type %d !\n",
+				      pkt.type);
 			return -EINVAL;
 		}
 		if (r)
@@ -2105,19 +2105,19 @@ int r100_cs_parse(struct radeon_cs_parser *p)
 
 static void r100_cs_track_texture_print(struct r100_cs_track_texture *t)
 {
-	DRM_ERROR("pitch                      %d\n", t->pitch);
-	DRM_ERROR("use_pitch                  %d\n", t->use_pitch);
-	DRM_ERROR("width                      %d\n", t->width);
-	DRM_ERROR("width_11                   %d\n", t->width_11);
-	DRM_ERROR("height                     %d\n", t->height);
-	DRM_ERROR("height_11                  %d\n", t->height_11);
-	DRM_ERROR("num levels                 %d\n", t->num_levels);
-	DRM_ERROR("depth                      %d\n", t->txdepth);
-	DRM_ERROR("bpp                        %d\n", t->cpp);
-	DRM_ERROR("coordinate type            %d\n", t->tex_coord_type);
-	DRM_ERROR("width round to power of 2  %d\n", t->roundup_w);
-	DRM_ERROR("height round to power of 2 %d\n", t->roundup_h);
-	DRM_ERROR("compress format            %d\n", t->compress_format);
+	DRM_DEBUG("pitch                      %d\n", t->pitch);
+	DRM_DEBUG("use_pitch                  %d\n", t->use_pitch);
+	DRM_DEBUG("width                      %d\n", t->width);
+	DRM_DEBUG("width_11                   %d\n", t->width_11);
+	DRM_DEBUG("height                     %d\n", t->height);
+	DRM_DEBUG("height_11                  %d\n", t->height_11);
+	DRM_DEBUG("num levels                 %d\n", t->num_levels);
+	DRM_DEBUG("depth                      %d\n", t->txdepth);
+	DRM_DEBUG("bpp                        %d\n", t->cpp);
+	DRM_DEBUG("coordinate type            %d\n", t->tex_coord_type);
+	DRM_DEBUG("width round to power of 2  %d\n", t->roundup_w);
+	DRM_DEBUG("height round to power of 2 %d\n", t->roundup_h);
+	DRM_DEBUG("compress format            %d\n", t->compress_format);
 }
 
 static int r100_track_compress_size(int compress_format, int w, int h)
@@ -2172,8 +2172,9 @@ static int r100_cs_track_cube(struct radeon_device *rdev,
 		size += track->textures[idx].cube_info[face].offset;
 
 		if (size > radeon_bo_size(cube_robj)) {
-			DRM_ERROR("Cube texture offset greater than object size %lu %lu\n",
-				  size, radeon_bo_size(cube_robj));
+			dev_warn_once(rdev->dev,
+				      "Cube texture offset greater than object size %lu %lu\n",
+				      size, radeon_bo_size(cube_robj));
 			r100_cs_track_texture_print(&track->textures[idx]);
 			return -1;
 		}
@@ -2196,7 +2197,7 @@ static int r100_cs_track_texture_check(struct radeon_device *rdev,
 			continue;
 		robj = track->textures[u].robj;
 		if (robj == NULL) {
-			DRM_ERROR("No texture bound to unit %u\n", u);
+			dev_warn_once(rdev->dev, "No texture bound to unit %u\n", u);
 			return -EINVAL;
 		}
 		size = 0;
@@ -2249,13 +2250,13 @@ static int r100_cs_track_texture_check(struct radeon_device *rdev,
 				size *= 6;
 			break;
 		default:
-			DRM_ERROR("Invalid texture coordinate type %u for unit "
-				  "%u\n", track->textures[u].tex_coord_type, u);
+			dev_warn_once(rdev->dev, "Invalid texture coordinate type %u for unit "
+				      "%u\n", track->textures[u].tex_coord_type, u);
 			return -EINVAL;
 		}
 		if (size > radeon_bo_size(robj)) {
-			DRM_ERROR("Texture of unit %u needs %lu bytes but is "
-				  "%lu\n", u, size, radeon_bo_size(robj));
+			dev_warn_once(rdev->dev, "Texture of unit %u needs %lu bytes but is "
+				      "%lu\n", u, size, radeon_bo_size(robj));
 			r100_cs_track_texture_print(&track->textures[u]);
 			return -EINVAL;
 		}
@@ -2277,18 +2278,18 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
 
 	for (i = 0; i < num_cb; i++) {
 		if (track->cb[i].robj == NULL) {
-			DRM_ERROR("[drm] No buffer for color buffer %d !\n", i);
+			dev_warn_once(rdev->dev, "[drm] No buffer for color buffer %d !\n", i);
 			return -EINVAL;
 		}
 		size = track->cb[i].pitch * track->cb[i].cpp * track->maxy;
 		size += track->cb[i].offset;
 		if (size > radeon_bo_size(track->cb[i].robj)) {
-			DRM_ERROR("[drm] Buffer too small for color buffer %d "
-				  "(need %lu have %lu) !\n", i, size,
-				  radeon_bo_size(track->cb[i].robj));
-			DRM_ERROR("[drm] color buffer %d (%u %u %u %u)\n",
-				  i, track->cb[i].pitch, track->cb[i].cpp,
-				  track->cb[i].offset, track->maxy);
+			dev_warn_once(rdev->dev, "[drm] Buffer too small for color buffer %d "
+				      "(need %lu have %lu) !\n", i, size,
+				      radeon_bo_size(track->cb[i].robj));
+			dev_warn_once(rdev->dev, "[drm] color buffer %d (%u %u %u %u)\n",
+				      i, track->cb[i].pitch, track->cb[i].cpp,
+				      track->cb[i].offset, track->maxy);
 			return -EINVAL;
 		}
 	}
@@ -2296,18 +2297,18 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
 
 	if (track->zb_dirty && track->z_enabled) {
 		if (track->zb.robj == NULL) {
-			DRM_ERROR("[drm] No buffer for z buffer !\n");
+			dev_warn_once(rdev->dev, "[drm] No buffer for z buffer !\n");
 			return -EINVAL;
 		}
 		size = track->zb.pitch * track->zb.cpp * track->maxy;
 		size += track->zb.offset;
 		if (size > radeon_bo_size(track->zb.robj)) {
-			DRM_ERROR("[drm] Buffer too small for z buffer "
-				  "(need %lu have %lu) !\n", size,
-				  radeon_bo_size(track->zb.robj));
-			DRM_ERROR("[drm] zbuffer (%u %u %u %u)\n",
-				  track->zb.pitch, track->zb.cpp,
-				  track->zb.offset, track->maxy);
+			dev_warn_once(rdev->dev, "[drm] Buffer too small for z buffer "
+				      "(need %lu have %lu) !\n", size,
+				      radeon_bo_size(track->zb.robj));
+			dev_warn_once(rdev->dev, "[drm] zbuffer (%u %u %u %u)\n",
+				      track->zb.pitch, track->zb.cpp,
+				      track->zb.offset, track->maxy);
 			return -EINVAL;
 		}
 	}
@@ -2315,19 +2316,19 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
 
 	if (track->aa_dirty && track->aaresolve) {
 		if (track->aa.robj == NULL) {
-			DRM_ERROR("[drm] No buffer for AA resolve buffer %d !\n", i);
+			dev_warn_once(rdev->dev, "[drm] No buffer for AA resolve buffer %d !\n", i);
 			return -EINVAL;
 		}
 		/* I believe the format comes from colorbuffer0. */
 		size = track->aa.pitch * track->cb[0].cpp * track->maxy;
 		size += track->aa.offset;
 		if (size > radeon_bo_size(track->aa.robj)) {
-			DRM_ERROR("[drm] Buffer too small for AA resolve buffer %d "
-				  "(need %lu have %lu) !\n", i, size,
-				  radeon_bo_size(track->aa.robj));
-			DRM_ERROR("[drm] AA resolve buffer %d (%u %u %u %u)\n",
-				  i, track->aa.pitch, track->cb[0].cpp,
-				  track->aa.offset, track->maxy);
+			dev_warn_once(rdev->dev, "[drm] Buffer too small for AA resolve buffer %d "
+				      "(need %lu have %lu) !\n", i, size,
+				      radeon_bo_size(track->aa.robj));
+			dev_warn_once(rdev->dev, "[drm] AA resolve buffer %d (%u %u %u %u)\n",
+				      i, track->aa.pitch, track->cb[0].cpp,
+				      track->aa.offset, track->maxy);
 			return -EINVAL;
 		}
 	}
@@ -2344,17 +2345,17 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
 		for (i = 0; i < track->num_arrays; i++) {
 			size = track->arrays[i].esize * track->max_indx * 4UL;
 			if (track->arrays[i].robj == NULL) {
-				DRM_ERROR("(PW %u) Vertex array %u no buffer "
-					  "bound\n", prim_walk, i);
+				dev_warn_once(rdev->dev, "(PW %u) Vertex array %u no buffer "
+					      "bound\n", prim_walk, i);
 				return -EINVAL;
 			}
 			if (size > radeon_bo_size(track->arrays[i].robj)) {
-				dev_err(rdev->dev, "(PW %u) Vertex array %u "
-					"need %lu dwords have %lu dwords\n",
-					prim_walk, i, size >> 2,
-					radeon_bo_size(track->arrays[i].robj)
-					>> 2);
-				DRM_ERROR("Max indices %u\n", track->max_indx);
+				dev_warn_once(rdev->dev, "(PW %u) Vertex array %u "
+					      "need %lu dwords have %lu dwords\n",
+					      prim_walk, i, size >> 2,
+					      radeon_bo_size(track->arrays[i].robj)
+					      >> 2);
+				dev_warn_once(rdev->dev, "Max indices %u\n", track->max_indx);
 				return -EINVAL;
 			}
 		}
@@ -2363,16 +2364,16 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
 		for (i = 0; i < track->num_arrays; i++) {
 			size = track->arrays[i].esize * (nverts - 1) * 4UL;
 			if (track->arrays[i].robj == NULL) {
-				DRM_ERROR("(PW %u) Vertex array %u no buffer "
-					  "bound\n", prim_walk, i);
+				dev_warn_once(rdev->dev, "(PW %u) Vertex array %u no buffer "
+					      "bound\n", prim_walk, i);
 				return -EINVAL;
 			}
 			if (size > radeon_bo_size(track->arrays[i].robj)) {
-				dev_err(rdev->dev, "(PW %u) Vertex array %u "
-					"need %lu dwords have %lu dwords\n",
-					prim_walk, i, size >> 2,
-					radeon_bo_size(track->arrays[i].robj)
-					>> 2);
+				dev_warn_once(rdev->dev, "(PW %u) Vertex array %u "
+					      "need %lu dwords have %lu dwords\n",
+					      prim_walk, i, size >> 2,
+					      radeon_bo_size(track->arrays[i].robj)
+					      >> 2);
 				return -EINVAL;
 			}
 		}
@@ -2380,16 +2381,16 @@ int r100_cs_track_check(struct radeon_device *rdev, struct r100_cs_track *track)
 	case 3:
 		size = track->vtx_size * nverts;
 		if (size != track->immd_dwords) {
-			DRM_ERROR("IMMD draw %u dwors but needs %lu dwords\n",
-				  track->immd_dwords, size);
-			DRM_ERROR("VAP_VF_CNTL.NUM_VERTICES %u, VTX_SIZE %u\n",
-				  nverts, track->vtx_size);
+			dev_warn_once(rdev->dev, "IMMD draw %u dwors but needs %lu dwords\n",
+				      track->immd_dwords, size);
+			dev_warn_once(rdev->dev, "VAP_VF_CNTL.NUM_VERTICES %u, VTX_SIZE %u\n",
+				      nverts, track->vtx_size);
 			return -EINVAL;
 		}
 		break;
 	default:
-		DRM_ERROR("[drm] Invalid primitive walk %d for VAP_VF_CNTL\n",
-			  prim_walk);
+		dev_warn_once(rdev->dev, "[drm] Invalid primitive walk %d for VAP_VF_CNTL\n",
+			      prim_walk);
 		return -EINVAL;
 	}
 
diff --git a/drivers/gpu/drm/radeon/r200.c b/drivers/gpu/drm/radeon/r200.c
index f5f2ffea5ab2..10a65a71de31 100644
--- a/drivers/gpu/drm/radeon/r200.c
+++ b/drivers/gpu/drm/radeon/r200.c
@@ -163,8 +163,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_CRTC_GUI_TRIG_VLINE:
 		r = r100_cs_packet_parse_vline(p);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -180,8 +180,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_DEPTHOFFSET:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -193,8 +193,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_COLOROFFSET:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -212,8 +212,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - R200_PP_TXOFFSET_0) / 24;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -265,8 +265,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 		face = (reg - ((i * 24) + R200_PP_TXOFFSET_0)) / 4;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -283,8 +283,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_COLORPITCH:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -326,12 +326,12 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 			track->cb[0].cpp = 4;
 			break;
 		default:
-			DRM_ERROR("Invalid color buffer format (%d) !\n",
-				  ((idx_value >> RADEON_RB3D_COLOR_FORMAT_SHIFT) & 0x1f));
+			dev_warn_once(p->dev, "Invalid color buffer format (%d) !\n",
+				      ((idx_value >> RADEON_RB3D_COLOR_FORMAT_SHIFT) & 0x1f));
 			return -EINVAL;
 		}
 		if (idx_value & RADEON_DEPTHXY_OFFSET_ENABLE) {
-			DRM_ERROR("No support for depth xy offset in kms\n");
+			dev_warn_once(p->dev, "No support for depth xy offset in kms\n");
 			return -EINVAL;
 		}
 
@@ -360,8 +360,8 @@ int r200_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_RB3D_ZPASS_ADDR:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index d22889fbfa9c..d2ee6deec039 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -645,8 +645,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case RADEON_CRTC_GUI_TRIG_VLINE:
 		r = r100_cs_packet_parse_vline(p);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -664,8 +664,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - R300_RB3D_COLOROFFSET0) >> 2;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -677,8 +677,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case R300_ZB_DEPTHOFFSET:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -706,8 +706,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		i = (reg - R300_TX_OFFSET_0) >> 2;
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -762,7 +762,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		/* RB3D_CCTL */
 		if ((idx_value & (1 << 10)) && /* CMASK_ENABLE */
 		    p->rdev->cmask_filp != p->filp) {
-			DRM_ERROR("Invalid RB3D_CCTL: Cannot enable CMASK.\n");
+			dev_warn_once(p->dev, "Invalid RB3D_CCTL: Cannot enable CMASK.\n");
 			return -EINVAL;
 		}
 		track->num_cb = ((idx_value >> 5) & 0x3) + 1;
@@ -779,8 +779,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					  idx, reg);
+				dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+					      idx, reg);
 				radeon_cs_dump_packet(p, pkt);
 				return r;
 			}
@@ -812,8 +812,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 			break;
 		case 5:
 			if (p->rdev->family < CHIP_RV515) {
-				DRM_ERROR("Invalid color buffer format (%d)!\n",
-					  ((idx_value >> 21) & 0xF));
+				dev_warn_once(p->dev, "Invalid color buffer format (%d)!\n",
+					      ((idx_value >> 21) & 0xF));
 				return -EINVAL;
 			}
 			fallthrough;
@@ -827,8 +827,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 			track->cb[i].cpp = 16;
 			break;
 		default:
-			DRM_ERROR("Invalid color buffer format (%d) !\n",
-				  ((idx_value >> 21) & 0xF));
+			dev_warn_once(p->dev, "Invalid color buffer format (%d) !\n",
+				      ((idx_value >> 21) & 0xF));
 			return -EINVAL;
 		}
 		track->cb_dirty = true;
@@ -853,8 +853,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 			track->zb.cpp = 4;
 			break;
 		default:
-			DRM_ERROR("Invalid z buffer format (%d) !\n",
-				  (idx_value & 0xF));
+			dev_warn_once(p->dev, "Invalid z buffer format (%d) !\n",
+				      (idx_value & 0xF));
 			return -EINVAL;
 		}
 		track->zb_dirty = true;
@@ -864,8 +864,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 		if (!(p->cs_flags & RADEON_CS_KEEP_TILING_FLAGS)) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 			if (r) {
-				DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					  idx, reg);
+				dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+					      idx, reg);
 				radeon_cs_dump_packet(p, pkt);
 				return r;
 			}
@@ -962,8 +962,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 			break;
 		case R300_TX_FORMAT_ATI2N:
 			if (p->rdev->family < CHIP_R420) {
-				DRM_ERROR("Invalid texture format %u\n",
-					  (idx_value & 0x1F));
+				dev_warn_once(p->dev, "Invalid texture format %u\n",
+					      (idx_value & 0x1F));
 				return -EINVAL;
 			}
 			/* The same rules apply as for DXT3/5. */
@@ -974,8 +974,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 			track->textures[i].compress_format = R100_TRACK_COMP_DXT35;
 			break;
 		default:
-			DRM_ERROR("Invalid texture format %u\n",
-				  (idx_value & 0x1F));
+			dev_warn_once(p->dev, "Invalid texture format %u\n",
+				      (idx_value & 0x1F));
 			return -EINVAL;
 		}
 		track->tex_dirty = true;
@@ -1041,7 +1041,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 					R100_TRACK_COMP_DXT1;
 			}
 		} else if (idx_value & (1 << 14)) {
-			DRM_ERROR("Forbidden bit TXFORMAT_MSB\n");
+			dev_warn_once(p->dev, "Forbidden bit TXFORMAT_MSB\n");
 			return -EINVAL;
 		}
 		track->tex_dirty = true;
@@ -1079,8 +1079,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case R300_ZB_ZPASS_ADDR:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1121,8 +1121,8 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
 	case R300_RB3D_AARESOLVE_OFFSET:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-				  idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1191,7 +1191,7 @@ static int r300_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_INDX_BUFFER:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			DRM_ERROR("No reloc for packet3 %d\n", pkt->opcode);
+			dev_warn_once(p->dev, "No reloc for packet3 %d\n", pkt->opcode);
 			radeon_cs_dump_packet(p, pkt);
 			return r;
 		}
@@ -1207,7 +1207,7 @@ static int r300_packet3_check(struct radeon_cs_parser *p,
 		 * PRIM_WALK must be equal to 3 vertex data in embedded
 		 * in cmd stream */
 		if (((radeon_get_ib_value(p, idx + 1) >> 4) & 0x3) != 3) {
-			DRM_ERROR("PRIM_WALK must be 3 for IMMD draw\n");
+			dev_warn_once(p->dev, "PRIM_WALK must be 3 for IMMD draw\n");
 			return -EINVAL;
 		}
 		track->vap_vf_cntl = radeon_get_ib_value(p, idx + 1);
@@ -1222,7 +1222,7 @@ static int r300_packet3_check(struct radeon_cs_parser *p,
 		 * PRIM_WALK must be equal to 3 vertex data in embedded
 		 * in cmd stream */
 		if (((radeon_get_ib_value(p, idx) >> 4) & 0x3) != 3) {
-			DRM_ERROR("PRIM_WALK must be 3 for IMMD draw\n");
+			dev_warn_once(p->dev, "PRIM_WALK must be 3 for IMMD draw\n");
 			return -EINVAL;
 		}
 		track->vap_vf_cntl = radeon_get_ib_value(p, idx);
@@ -1272,7 +1272,7 @@ static int r300_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_NOP:
 		break;
 	default:
-		DRM_ERROR("Packet3 opcode %x not supported\n", pkt->opcode);
+		dev_warn_once(p->dev, "Packet3 opcode %x not supported\n", pkt->opcode);
 		return -EINVAL;
 	}
 	return 0;
@@ -1308,7 +1308,7 @@ int r300_cs_parse(struct radeon_cs_parser *p)
 			r = r300_packet3_check(p, &pkt);
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d !\n", pkt.type);
+			dev_warn_once(p->dev, "Unknown packet type %d !\n", pkt.type);
 			return -EINVAL;
 		}
 		if (r) {
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 811265648a58..8eeceeeca362 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -361,9 +361,9 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
 
 	format = G_0280A0_FORMAT(track->cb_color_info[i]);
 	if (!r600_fmt_is_valid_color(format)) {
-		dev_warn(p->dev, "%s:%d cb invalid format %d for %d (0x%08X)\n",
-			 __func__, __LINE__, format,
-			i, track->cb_color_info[i]);
+		dev_warn_once(p->dev, "%s:%d cb invalid format %d for %d (0x%08X)\n",
+			      __func__, __LINE__, format,
+			      i, track->cb_color_info[i]);
 		return -EINVAL;
 	}
 	/* pitch in pixels */
@@ -384,9 +384,9 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
 	array_check.blocksize = r600_fmt_get_blocksize(format);
 	if (r600_get_array_mode_alignment(&array_check,
 					  &pitch_align, &height_align, &depth_align, &base_align)) {
-		dev_warn(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__,
-			 G_0280A0_ARRAY_MODE(track->cb_color_info[i]), i,
-			 track->cb_color_info[i]);
+		dev_warn_once(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__,
+			      G_0280A0_ARRAY_MODE(track->cb_color_info[i]), i,
+			      track->cb_color_info[i]);
 		return -EINVAL;
 	}
 	switch (array_mode) {
@@ -402,25 +402,26 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
 	case V_0280A0_ARRAY_2D_TILED_THIN1:
 		break;
 	default:
-		dev_warn(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__,
-			G_0280A0_ARRAY_MODE(track->cb_color_info[i]), i,
-			track->cb_color_info[i]);
+		dev_warn_once(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__,
+			      G_0280A0_ARRAY_MODE(track->cb_color_info[i]), i,
+			      track->cb_color_info[i]);
 		return -EINVAL;
 	}
 
 	if (!IS_ALIGNED(pitch, pitch_align)) {
-		dev_warn(p->dev, "%s:%d cb pitch (%d, 0x%x, %d) invalid\n",
-			 __func__, __LINE__, pitch, pitch_align, array_mode);
+		dev_warn_once(p->dev, "%s:%d cb pitch (%d, 0x%x, %d) invalid\n",
+			      __func__, __LINE__, pitch, pitch_align, array_mode);
 		return -EINVAL;
 	}
 	if (!IS_ALIGNED(height, height_align)) {
-		dev_warn(p->dev, "%s:%d cb height (%d, 0x%x, %d) invalid\n",
-			 __func__, __LINE__, height, height_align, array_mode);
+		dev_warn_once(p->dev, "%s:%d cb height (%d, 0x%x, %d) invalid\n",
+			      __func__, __LINE__, height, height_align, array_mode);
 		return -EINVAL;
 	}
 	if (!IS_ALIGNED(base_offset, base_align)) {
-		dev_warn(p->dev, "%s offset[%d] 0x%llx 0x%llx, %d not aligned\n", __func__, i,
-			 base_offset, base_align, array_mode);
+		dev_warn_once(p->dev,
+			      "%s offset[%d] 0x%llx 0x%llx, %d not aligned\n", __func__, i,
+			      base_offset, base_align, array_mode);
 		return -EINVAL;
 	}
 
@@ -447,13 +448,14 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
 			 * broken userspace.
 			 */
 		} else {
-			dev_warn(p->dev, "%s offset[%d] %d %llu %d %lu too big (%d %d) (%d %d %d)\n",
-				 __func__, i, array_mode,
-				 track->cb_color_bo_offset[i], tmp,
-				 radeon_bo_size(track->cb_color_bo[i]),
-				 pitch, height, r600_fmt_get_nblocksx(format, pitch),
-				 r600_fmt_get_nblocksy(format, height),
-				 r600_fmt_get_blocksize(format));
+			dev_warn_once(p->dev,
+				      "%s offset[%d] %d %llu %d %lu too big (%d %d) (%d %d %d)\n",
+				      __func__, i, array_mode,
+				      track->cb_color_bo_offset[i], tmp,
+				      radeon_bo_size(track->cb_color_bo[i]),
+				      pitch, height, r600_fmt_get_nblocksx(format, pitch),
+				      r600_fmt_get_nblocksy(format, height),
+				      r600_fmt_get_blocksize(format));
 			return -EINVAL;
 		}
 	}
@@ -478,11 +480,11 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
 
 			if (bytes + track->cb_color_frag_offset[i] >
 			    radeon_bo_size(track->cb_color_frag_bo[i])) {
-				dev_warn(p->dev, "%s FMASK_TILE_MAX too large "
-					 "(tile_max=%u, bytes=%u, offset=%llu, bo_size=%lu)\n",
-					 __func__, tile_max, bytes,
-					 track->cb_color_frag_offset[i],
-					 radeon_bo_size(track->cb_color_frag_bo[i]));
+				dev_warn_once(p->dev, "%s FMASK_TILE_MAX too large "
+					      "(tile_max=%u, bytes=%u, offset=%llu, bo_size=%lu)\n",
+					      __func__, tile_max, bytes,
+					      track->cb_color_frag_offset[i],
+					      radeon_bo_size(track->cb_color_frag_bo[i]));
 				return -EINVAL;
 			}
 		}
@@ -496,17 +498,17 @@ static int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
 
 		if (bytes + track->cb_color_tile_offset[i] >
 		    radeon_bo_size(track->cb_color_tile_bo[i])) {
-			dev_warn(p->dev, "%s CMASK_BLOCK_MAX too large "
-				 "(block_max=%u, bytes=%u, offset=%llu, bo_size=%lu)\n",
-				 __func__, block_max, bytes,
-				 track->cb_color_tile_offset[i],
-				 radeon_bo_size(track->cb_color_tile_bo[i]));
+			dev_warn_once(p->dev, "%s CMASK_BLOCK_MAX too large "
+				      "(block_max=%u, bytes=%u, offset=%llu, bo_size=%lu)\n",
+				      __func__, block_max, bytes,
+				      track->cb_color_tile_offset[i],
+				      radeon_bo_size(track->cb_color_tile_bo[i]));
 			return -EINVAL;
 		}
 		break;
 	}
 	default:
-		dev_warn(p->dev, "%s invalid tile mode\n", __func__);
+		dev_warn_once(p->dev, "%s invalid tile mode\n", __func__);
 		return -EINVAL;
 	}
 	return 0;
@@ -526,7 +528,7 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 
 
 	if (track->db_bo == NULL) {
-		dev_warn(p->dev, "z/stencil with no depth buffer\n");
+		dev_warn_once(p->dev, "z/stencil with no depth buffer\n");
 		return -EINVAL;
 	}
 	switch (G_028010_FORMAT(track->db_depth_info)) {
@@ -544,20 +546,22 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 		bpe = 8;
 		break;
 	default:
-		dev_warn(p->dev, "z/stencil with invalid format %d\n", G_028010_FORMAT(track->db_depth_info));
+		dev_warn_once(p->dev,
+			      "z/stencil with invalid format %d\n",
+			      G_028010_FORMAT(track->db_depth_info));
 		return -EINVAL;
 	}
 	if ((track->db_depth_size & 0xFFFFFC00) == 0xFFFFFC00) {
 		if (!track->db_depth_size_idx) {
-			dev_warn(p->dev, "z/stencil buffer size not set\n");
+			dev_warn_once(p->dev, "z/stencil buffer size not set\n");
 			return -EINVAL;
 		}
 		tmp = radeon_bo_size(track->db_bo) - track->db_offset;
 		tmp = (tmp / bpe) >> 6;
 		if (!tmp) {
-			dev_warn(p->dev, "z/stencil buffer too small (0x%08X %d %d %ld)\n",
-					track->db_depth_size, bpe, track->db_offset,
-					radeon_bo_size(track->db_bo));
+			dev_warn_once(p->dev, "z/stencil buffer too small (0x%08X %d %d %ld)\n",
+				      track->db_depth_size, bpe, track->db_offset,
+				      radeon_bo_size(track->db_bo));
 			return -EINVAL;
 		}
 		ib[track->db_depth_size_idx] = S_028000_SLICE_TILE_MAX(tmp - 1) | (track->db_depth_size & 0x3FF);
@@ -579,9 +583,9 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 		array_check.blocksize = bpe;
 		if (r600_get_array_mode_alignment(&array_check,
 					&pitch_align, &height_align, &depth_align, &base_align)) {
-			dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__,
-					G_028010_ARRAY_MODE(track->db_depth_info),
-					track->db_depth_info);
+			dev_warn_once(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__,
+				      G_028010_ARRAY_MODE(track->db_depth_info),
+				      track->db_depth_info);
 			return -EINVAL;
 		}
 		switch (array_mode) {
@@ -592,24 +596,24 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 		case V_028010_ARRAY_2D_TILED_THIN1:
 			break;
 		default:
-			dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__,
-					G_028010_ARRAY_MODE(track->db_depth_info),
-					track->db_depth_info);
+			dev_warn_once(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__,
+				      G_028010_ARRAY_MODE(track->db_depth_info),
+				      track->db_depth_info);
 			return -EINVAL;
 		}
 
 		if (!IS_ALIGNED(pitch, pitch_align)) {
-			dev_warn(p->dev, "%s:%d db pitch (%d, 0x%x, %d) invalid\n",
+			dev_warn_once(p->dev, "%s:%d db pitch (%d, 0x%x, %d) invalid\n",
 					__func__, __LINE__, pitch, pitch_align, array_mode);
 			return -EINVAL;
 		}
 		if (!IS_ALIGNED(height, height_align)) {
-			dev_warn(p->dev, "%s:%d db height (%d, 0x%x, %d) invalid\n",
+			dev_warn_once(p->dev, "%s:%d db height (%d, 0x%x, %d) invalid\n",
 					__func__, __LINE__, height, height_align, array_mode);
 			return -EINVAL;
 		}
 		if (!IS_ALIGNED(base_offset, base_align)) {
-			dev_warn(p->dev, "%s offset 0x%llx, 0x%llx, %d not aligned\n", __func__,
+			dev_warn_once(p->dev, "%s offset 0x%llx, 0x%llx, %d not aligned\n", __func__,
 					base_offset, base_align, array_mode);
 			return -EINVAL;
 		}
@@ -618,10 +622,11 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 		nviews = G_028004_SLICE_MAX(track->db_depth_view) + 1;
 		tmp = ntiles * bpe * 64 * nviews * track->nsamples;
 		if ((tmp + track->db_offset) > radeon_bo_size(track->db_bo)) {
-			dev_warn(p->dev, "z/stencil buffer (%d) too small (0x%08X %d %d %d -> %u have %lu)\n",
-					array_mode,
-					track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset,
-					radeon_bo_size(track->db_bo));
+			dev_warn_once(p->dev,
+				      "z/stencil buffer (%d) too small (0x%08X %d %d %d -> %u have %lu)\n",
+				      array_mode,
+				      track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset,
+				      radeon_bo_size(track->db_bo));
 			return -EINVAL;
 		}
 	}
@@ -632,13 +637,13 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 		unsigned nbx, nby;
 
 		if (track->htile_bo == NULL) {
-			dev_warn(p->dev, "%s:%d htile enabled without htile surface 0x%08x\n",
-				 __func__, __LINE__, track->db_depth_info);
+			dev_warn_once(p->dev, "%s:%d htile enabled without htile surface 0x%08x\n",
+				      __func__, __LINE__, track->db_depth_info);
 			return -EINVAL;
 		}
 		if ((track->db_depth_size & 0xFFFFFC00) == 0xFFFFFC00) {
-			dev_warn(p->dev, "%s:%d htile can't be enabled with bogus db_depth_size 0x%08x\n",
-				 __func__, __LINE__, track->db_depth_size);
+			dev_warn_once(p->dev, "%s:%d htile can't be enabled with bogus db_depth_size 0x%08x\n",
+				      __func__, __LINE__, track->db_depth_size);
 			return -EINVAL;
 		}
 
@@ -676,8 +681,8 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 				nby = round_up(nby, 16 * 8);
 				break;
 			default:
-				dev_warn(p->dev, "%s:%d invalid num pipes %d\n",
-					 __func__, __LINE__, track->npipes);
+				dev_warn_once(p->dev, "%s:%d invalid num pipes %d\n",
+					      __func__, __LINE__, track->npipes);
 				return -EINVAL;
 			}
 		}
@@ -689,9 +694,9 @@ static int r600_cs_track_validate_db(struct radeon_cs_parser *p)
 		size += track->htile_offset;
 
 		if (size > radeon_bo_size(track->htile_bo)) {
-			dev_warn(p->dev, "%s:%d htile surface too small %ld for %ld (%d %d)\n",
-				 __func__, __LINE__, radeon_bo_size(track->htile_bo),
-				 size, nbx, nby);
+			dev_warn_once(p->dev, "%s:%d htile surface too small %ld for %ld (%d %d)\n",
+				      __func__, __LINE__, radeon_bo_size(track->htile_bo),
+				      size, nbx, nby);
 			return -EINVAL;
 		}
 	}
@@ -718,13 +723,13 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
 					u64 offset = (u64)track->vgt_strmout_bo_offset[i] +
 						(u64)track->vgt_strmout_size[i];
 					if (offset > radeon_bo_size(track->vgt_strmout_bo[i])) {
-						DRM_ERROR("streamout %d bo too small: 0x%llx, 0x%lx\n",
-							  i, offset,
-							  radeon_bo_size(track->vgt_strmout_bo[i]));
+						dev_warn_once(p->dev, "streamout %d bo too small: 0x%llx, 0x%lx\n",
+							      i, offset,
+							      radeon_bo_size(track->vgt_strmout_bo[i]));
 						return -EINVAL;
 					}
 				} else {
-					dev_warn(p->dev, "No buffer for streamout %d\n", i);
+					dev_warn_once(p->dev, "No buffer for streamout %d\n", i);
 					return -EINVAL;
 				}
 			}
@@ -753,8 +758,8 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
 			    (tmp >> (i * 4)) & 0xF) {
 				/* at least one component is enabled */
 				if (track->cb_color_bo[i] == NULL) {
-					dev_warn(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n",
-						__func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i);
+					dev_warn_once(p->dev, "%s:%d mask 0x%08X | 0x%08X no cb for %d\n",
+						      __func__, __LINE__, track->cb_target_mask, track->cb_shader_mask, i);
 					return -EINVAL;
 				}
 				/* perform rewrite of CB_COLOR[0-7]_SIZE */
@@ -841,33 +846,33 @@ int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
 	/* check its a WAIT_REG_MEM */
 	if (wait_reg_mem.type != RADEON_PACKET_TYPE3 ||
 	    wait_reg_mem.opcode != PACKET3_WAIT_REG_MEM) {
-		DRM_ERROR("vline wait missing WAIT_REG_MEM segment\n");
+		dev_warn_once(p->dev, "vline wait missing WAIT_REG_MEM segment\n");
 		return -EINVAL;
 	}
 
 	wait_reg_mem_info = radeon_get_ib_value(p, wait_reg_mem.idx + 1);
 	/* bit 4 is reg (0) or mem (1) */
 	if (wait_reg_mem_info & 0x10) {
-		DRM_ERROR("vline WAIT_REG_MEM waiting on MEM instead of REG\n");
+		dev_warn_once(p->dev, "vline WAIT_REG_MEM waiting on MEM instead of REG\n");
 		return -EINVAL;
 	}
 	/* bit 8 is me (0) or pfp (1) */
 	if (wait_reg_mem_info & 0x100) {
-		DRM_ERROR("vline WAIT_REG_MEM waiting on PFP instead of ME\n");
+		dev_warn_once(p->dev, "vline WAIT_REG_MEM waiting on PFP instead of ME\n");
 		return -EINVAL;
 	}
 	/* waiting for value to be equal */
 	if ((wait_reg_mem_info & 0x7) != 0x3) {
-		DRM_ERROR("vline WAIT_REG_MEM function not equal\n");
+		dev_warn_once(p->dev, "vline WAIT_REG_MEM function not equal\n");
 		return -EINVAL;
 	}
 	if ((radeon_get_ib_value(p, wait_reg_mem.idx + 2) << 2) != vline_status[0]) {
-		DRM_ERROR("vline WAIT_REG_MEM bad reg\n");
+		dev_warn_once(p->dev, "vline WAIT_REG_MEM bad reg\n");
 		return -EINVAL;
 	}
 
 	if (radeon_get_ib_value(p, wait_reg_mem.idx + 5) != RADEON_VLINE_STAT) {
-		DRM_ERROR("vline WAIT_REG_MEM bad bit mask\n");
+		dev_warn_once(p->dev, "vline WAIT_REG_MEM bad bit mask\n");
 		return -EINVAL;
 	}
 
@@ -886,7 +891,7 @@ int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
 
 	crtc = drm_crtc_find(rdev_to_drm(p->rdev), p->filp, crtc_id);
 	if (!crtc) {
-		DRM_ERROR("cannot find crtc %d\n", crtc_id);
+		dev_warn_once(p->dev, "cannot find crtc %d\n", crtc_id);
 		return -ENOENT;
 	}
 	radeon_crtc = to_radeon_crtc(crtc);
@@ -907,7 +912,7 @@ int r600_cs_common_vline_parse(struct radeon_cs_parser *p,
 		ib[h_idx] = header;
 		ib[h_idx + 4] = vline_status[crtc_id] >> 2;
 	} else {
-		DRM_ERROR("unknown crtc reloc\n");
+		dev_warn_once(p->dev, "unknown crtc reloc\n");
 		return -EINVAL;
 	}
 	return 0;
@@ -923,8 +928,8 @@ static int r600_packet0_check(struct radeon_cs_parser *p,
 	case AVIVO_D1MODE_VLINE_START_END:
 		r = r600_cs_packet_parse_vline(p);
 		if (r) {
-			DRM_ERROR("No reloc for ib[%d]=0x%04X\n",
-					idx, reg);
+			dev_warn_once(p->dev, "No reloc for ib[%d]=0x%04X\n",
+				      idx, reg);
 			return r;
 		}
 		break;
@@ -972,7 +977,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 
 	i = (reg >> 7);
 	if (i >= ARRAY_SIZE(r600_reg_safe_bm)) {
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+		dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
 	}
 	m = 1 << ((reg >> 2) & 31);
@@ -1013,8 +1018,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_VSTMP_RING_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, 0);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1031,8 +1036,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		    radeon_cs_packet_next_is_pkt3_nop(p)) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					 "0x%04X\n", reg);
+				dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+					      "0x%04X\n", reg);
 				return -EINVAL;
 			}
 			track->db_depth_info = radeon_get_ib_value(p, idx);
@@ -1073,8 +1078,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case VGT_STRMOUT_BUFFER_BASE_3:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = (reg - VGT_STRMOUT_BUFFER_BASE_0) / 16;
@@ -1096,8 +1101,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CP_COHER_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "missing reloc for CP_COHER_BASE "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "missing reloc for CP_COHER_BASE "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1270,8 +1275,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case CB_COLOR7_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		tmp = (reg - CB_COLOR0_BASE) / 4;
@@ -1285,8 +1290,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_DEPTH_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->db_offset = radeon_get_ib_value(p, idx) << 8;
@@ -1298,8 +1303,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case DB_HTILE_DATA_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		track->htile_offset = (u64)radeon_get_ib_value(p, idx) << 8;
@@ -1368,8 +1373,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SQ_ALU_CONST_CACHE_VS_15:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONTEXT_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONTEXT_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1377,8 +1382,8 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 	case SX_MEMORY_EXPORT_BASE:
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			dev_warn(p->dev, "bad SET_CONFIG_REG "
-					"0x%04X\n", reg);
+			dev_warn_once(p->dev, "bad SET_CONFIG_REG "
+				      "0x%04X\n", reg);
 			return -EINVAL;
 		}
 		ib[idx] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1387,7 +1392,7 @@ static int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 		track->sx_misc_kill_all_prims = (radeon_get_ib_value(p, idx) & 0x1) != 0;
 		break;
 	default:
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+		dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return -EINVAL;
 	}
 	return 0;
@@ -1543,43 +1548,43 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p,  u32 idx,
 		llevel = 0;
 		break;
 	default:
-		dev_warn(p->dev, "this kernel doesn't support %d texture dim\n", G_038000_DIM(word0));
+		dev_warn_once(p->dev, "this kernel doesn't support %d texture dim\n", G_038000_DIM(word0));
 		return -EINVAL;
 	}
 	if (!r600_fmt_is_valid_texture(format, p->family)) {
-		dev_warn(p->dev, "%s:%d texture invalid format %d\n",
-			 __func__, __LINE__, format);
+		dev_warn_once(p->dev, "%s:%d texture invalid format %d\n",
+			      __func__, __LINE__, format);
 		return -EINVAL;
 	}
 
 	if (r600_get_array_mode_alignment(&array_check,
 					  &pitch_align, &height_align, &depth_align, &base_align)) {
-		dev_warn(p->dev, "%s:%d tex array mode (%d) invalid\n",
-			 __func__, __LINE__, G_038000_TILE_MODE(word0));
+		dev_warn_once(p->dev, "%s:%d tex array mode (%d) invalid\n",
+			      __func__, __LINE__, G_038000_TILE_MODE(word0));
 		return -EINVAL;
 	}
 
 	/* XXX check height as well... */
 
 	if (!IS_ALIGNED(pitch, pitch_align)) {
-		dev_warn(p->dev, "%s:%d tex pitch (%d, 0x%x, %d) invalid\n",
-			 __func__, __LINE__, pitch, pitch_align, G_038000_TILE_MODE(word0));
+		dev_warn_once(p->dev, "%s:%d tex pitch (%d, 0x%x, %d) invalid\n",
+			      __func__, __LINE__, pitch, pitch_align, G_038000_TILE_MODE(word0));
 		return -EINVAL;
 	}
 	if (!IS_ALIGNED(base_offset, base_align)) {
-		dev_warn(p->dev, "%s:%d tex base offset (0x%llx, 0x%llx, %d) invalid\n",
-			 __func__, __LINE__, base_offset, base_align, G_038000_TILE_MODE(word0));
+		dev_warn_once(p->dev, "%s:%d tex base offset (0x%llx, 0x%llx, %d) invalid\n",
+			      __func__, __LINE__, base_offset, base_align, G_038000_TILE_MODE(word0));
 		return -EINVAL;
 	}
 	if (!IS_ALIGNED(mip_offset, base_align)) {
-		dev_warn(p->dev, "%s:%d tex mip offset (0x%llx, 0x%llx, %d) invalid\n",
-			 __func__, __LINE__, mip_offset, base_align, G_038000_TILE_MODE(word0));
+		dev_warn_once(p->dev, "%s:%d tex mip offset (0x%llx, 0x%llx, %d) invalid\n",
+			      __func__, __LINE__, mip_offset, base_align, G_038000_TILE_MODE(word0));
 		return -EINVAL;
 	}
 
 	if (blevel > llevel) {
-		dev_warn(p->dev, "texture blevel %d > llevel %d\n",
-			 blevel, llevel);
+		dev_warn_once(p->dev, "texture blevel %d > llevel %d\n",
+			      blevel, llevel);
 	}
 	if (is_array) {
 		barray = G_038014_BASE_ARRAY(word5);
@@ -1592,16 +1597,16 @@ static int r600_check_texture_resource(struct radeon_cs_parser *p,  u32 idx,
 			  &l0_size, &mipmap_size);
 	/* using get ib will give us the offset into the texture bo */
 	if ((l0_size + word2) > radeon_bo_size(texture)) {
-		dev_warn(p->dev, "texture bo too small ((%d %d) (%d %d) %d %d %d -> %d have %ld)\n",
-			 w0, h0, pitch_align, height_align,
-			 array_check.array_mode, format, word2,
-			 l0_size, radeon_bo_size(texture));
-		dev_warn(p->dev, "alignments %d %d %d %lld\n", pitch, pitch_align, height_align, base_align);
+		dev_warn_once(p->dev, "texture bo too small ((%d %d) (%d %d) %d %d %d -> %d have %ld)\n",
+			      w0, h0, pitch_align, height_align,
+			      array_check.array_mode, format, word2,
+			      l0_size, radeon_bo_size(texture));
+		dev_warn_once(p->dev, "alignments %d %d %d %lld\n", pitch, pitch_align, height_align, base_align);
 		return -EINVAL;
 	}
 	/* using get ib will give us the offset into the mipmap bo */
 	if ((mipmap_size + word3) > radeon_bo_size(mipmap)) {
-		/*dev_warn(p->dev, "mipmap bo too small (%d %d %d %d %d %d -> %d have %ld)\n",
+		/*dev_warn_once(p->dev, "mipmap bo too small (%d %d %d %d %d %d -> %d have %ld)\n",
 		  w0, h0, format, blevel, nlevels, word3, mipmap_size, radeon_bo_size(texture));*/
 	}
 	return 0;
@@ -1613,13 +1618,13 @@ static bool r600_is_safe_reg(struct radeon_cs_parser *p, u32 reg, u32 idx)
 
 	i = (reg >> 7);
 	if (i >= ARRAY_SIZE(r600_reg_safe_bm)) {
-		dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+		dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 		return false;
 	}
 	m = 1 << ((reg >> 2) & 31);
 	if (!(r600_reg_safe_bm[i] & m))
 		return true;
-	dev_warn(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
+	dev_warn_once(p->dev, "forbidden register 0x%08x at %d\n", reg, idx);
 	return false;
 }
 
@@ -1648,7 +1653,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 1) {
-			DRM_ERROR("bad SET PREDICATION\n");
+			dev_warn_once(p->dev, "bad SET PREDICATION\n");
 			return -EINVAL;
 		}
 
@@ -1660,13 +1665,13 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			return 0;
 
 		if (pred_op > 2) {
-			DRM_ERROR("bad SET PREDICATION operation %d\n", pred_op);
+			dev_warn_once(p->dev, "bad SET PREDICATION operation %d\n", pred_op);
 			return -EINVAL;
 		}
 
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			DRM_ERROR("bad SET PREDICATION\n");
+			dev_warn_once(p->dev, "bad SET PREDICATION\n");
 			return -EINVAL;
 		}
 
@@ -1681,20 +1686,20 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 
 	case PACKET3_START_3D_CMDBUF:
 		if (p->family >= CHIP_RV770 || pkt->count) {
-			DRM_ERROR("bad START_3D\n");
+			dev_warn_once(p->dev, "bad START_3D\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_CONTEXT_CONTROL:
 		if (pkt->count != 1) {
-			DRM_ERROR("bad CONTEXT_CONTROL\n");
+			dev_warn_once(p->dev, "bad CONTEXT_CONTROL\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_INDEX_TYPE:
 	case PACKET3_NUM_INSTANCES:
 		if (pkt->count) {
-			DRM_ERROR("bad INDEX_TYPE/NUM_INSTANCES\n");
+			dev_warn_once(p->dev, "bad INDEX_TYPE/NUM_INSTANCES\n");
 			return -EINVAL;
 		}
 		break;
@@ -1702,12 +1707,12 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 	{
 		uint64_t offset;
 		if (pkt->count != 3) {
-			DRM_ERROR("bad DRAW_INDEX\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			DRM_ERROR("bad DRAW_INDEX\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX\n");
 			return -EINVAL;
 		}
 
@@ -1720,37 +1725,37 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 
 		r = r600_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	}
 	case PACKET3_DRAW_INDEX_AUTO:
 		if (pkt->count != 1) {
-			DRM_ERROR("bad DRAW_INDEX_AUTO\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_AUTO\n");
 			return -EINVAL;
 		}
 		r = r600_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream %d\n", __func__, __LINE__, idx);
 			return r;
 		}
 		break;
 	case PACKET3_DRAW_INDEX_IMMD_BE:
 	case PACKET3_DRAW_INDEX_IMMD:
 		if (pkt->count < 2) {
-			DRM_ERROR("bad DRAW_INDEX_IMMD\n");
+			dev_warn_once(p->dev, "bad DRAW_INDEX_IMMD\n");
 			return -EINVAL;
 		}
 		r = r600_cs_track_check(p);
 		if (r) {
-			dev_warn(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
+			dev_warn_once(p->dev, "%s:%d invalid cmd stream\n", __func__, __LINE__);
 			return r;
 		}
 		break;
 	case PACKET3_WAIT_REG_MEM:
 		if (pkt->count != 5) {
-			DRM_ERROR("bad WAIT_REG_MEM\n");
+			dev_warn_once(p->dev, "bad WAIT_REG_MEM\n");
 			return -EINVAL;
 		}
 		/* bit 4 is reg (0) or mem (1) */
@@ -1759,7 +1764,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad WAIT_REG_MEM\n");
+				dev_warn_once(p->dev, "bad WAIT_REG_MEM\n");
 				return -EINVAL;
 			}
 
@@ -1770,7 +1775,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffff0);
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
 		} else if (idx_value & 0x100) {
-			DRM_ERROR("cannot use PFP on REG wait\n");
+			dev_warn_once(p->dev, "cannot use PFP on REG wait\n");
 			return -EINVAL;
 		}
 		break;
@@ -1779,24 +1784,24 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		u32 command, size;
 		u64 offset, tmp;
 		if (pkt->count != 4) {
-			DRM_ERROR("bad CP DMA\n");
+			dev_warn_once(p->dev, "bad CP DMA\n");
 			return -EINVAL;
 		}
 		command = radeon_get_ib_value(p, idx+4);
 		size = command & 0x1fffff;
 		if (command & PACKET3_CP_DMA_CMD_SAS) {
 			/* src address space is register */
-			DRM_ERROR("CP DMA SAS not supported\n");
+			dev_warn_once(p->dev, "CP DMA SAS not supported\n");
 			return -EINVAL;
 		} else {
 			if (command & PACKET3_CP_DMA_CMD_SAIC) {
-				DRM_ERROR("CP DMA SAIC only supported for registers\n");
+				dev_warn_once(p->dev, "CP DMA SAIC only supported for registers\n");
 				return -EINVAL;
 			}
 			/* src address space is memory */
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad CP DMA SRC\n");
+				dev_warn_once(p->dev, "bad CP DMA SRC\n");
 				return -EINVAL;
 			}
 
@@ -1806,8 +1811,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			offset = reloc->gpu_offset + tmp;
 
 			if ((tmp + size) > radeon_bo_size(reloc->robj)) {
-				dev_warn(p->dev, "CP DMA src buffer too small (%llu %lu)\n",
-					 tmp + size, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "CP DMA src buffer too small (%llu %lu)\n",
+					      tmp + size, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 
@@ -1816,17 +1821,17 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		}
 		if (command & PACKET3_CP_DMA_CMD_DAS) {
 			/* dst address space is register */
-			DRM_ERROR("CP DMA DAS not supported\n");
+			dev_warn_once(p->dev, "CP DMA DAS not supported\n");
 			return -EINVAL;
 		} else {
 			/* dst address space is memory */
 			if (command & PACKET3_CP_DMA_CMD_DAIC) {
-				DRM_ERROR("CP DMA DAIC only supported for registers\n");
+				dev_warn_once(p->dev, "CP DMA DAIC only supported for registers\n");
 				return -EINVAL;
 			}
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad CP DMA DST\n");
+				dev_warn_once(p->dev, "bad CP DMA DST\n");
 				return -EINVAL;
 			}
 
@@ -1836,8 +1841,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			offset = reloc->gpu_offset + tmp;
 
 			if ((tmp + size) > radeon_bo_size(reloc->robj)) {
-				dev_warn(p->dev, "CP DMA dst buffer too small (%llu %lu)\n",
-					 tmp + size, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "CP DMA dst buffer too small (%llu %lu)\n",
+					      tmp + size, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 
@@ -1848,7 +1853,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 	}
 	case PACKET3_SURFACE_SYNC:
 		if (pkt->count != 3) {
-			DRM_ERROR("bad SURFACE_SYNC\n");
+			dev_warn_once(p->dev, "bad SURFACE_SYNC\n");
 			return -EINVAL;
 		}
 		/* 0xffffffff/0x0 is flush all cache flag */
@@ -1856,7 +1861,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		    radeon_get_ib_value(p, idx + 2) != 0) {
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad SURFACE_SYNC\n");
+				dev_warn_once(p->dev, "bad SURFACE_SYNC\n");
 				return -EINVAL;
 			}
 			ib[idx+2] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1864,7 +1869,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case PACKET3_EVENT_WRITE:
 		if (pkt->count != 2 && pkt->count != 0) {
-			DRM_ERROR("bad EVENT_WRITE\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE\n");
 			return -EINVAL;
 		}
 		if (pkt->count) {
@@ -1872,7 +1877,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad EVENT_WRITE\n");
+				dev_warn_once(p->dev, "bad EVENT_WRITE\n");
 				return -EINVAL;
 			}
 			offset = reloc->gpu_offset +
@@ -1888,12 +1893,12 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		uint64_t offset;
 
 		if (pkt->count != 4) {
-			DRM_ERROR("bad EVENT_WRITE_EOP\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE_EOP\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			DRM_ERROR("bad EVENT_WRITE\n");
+			dev_warn_once(p->dev, "bad EVENT_WRITE\n");
 			return -EINVAL;
 		}
 
@@ -1911,7 +1916,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_CONFIG_REG_OFFSET) ||
 		    (start_reg >= PACKET3_SET_CONFIG_REG_END) ||
 		    (end_reg >= PACKET3_SET_CONFIG_REG_END)) {
-			DRM_ERROR("bad PACKET3_SET_CONFIG_REG\n");
+			dev_warn_once(p->dev, "bad PACKET3_SET_CONFIG_REG\n");
 			return -EINVAL;
 		}
 		for (i = 0; i < pkt->count; i++) {
@@ -1927,7 +1932,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_CONTEXT_REG_OFFSET) ||
 		    (start_reg >= PACKET3_SET_CONTEXT_REG_END) ||
 		    (end_reg >= PACKET3_SET_CONTEXT_REG_END)) {
-			DRM_ERROR("bad PACKET3_SET_CONTEXT_REG\n");
+			dev_warn_once(p->dev, "bad PACKET3_SET_CONTEXT_REG\n");
 			return -EINVAL;
 		}
 		for (i = 0; i < pkt->count; i++) {
@@ -1939,7 +1944,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case PACKET3_SET_RESOURCE:
 		if (pkt->count % 7) {
-			DRM_ERROR("bad SET_RESOURCE\n");
+			dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 			return -EINVAL;
 		}
 		start_reg = (idx_value << 2) + PACKET3_SET_RESOURCE_OFFSET;
@@ -1947,7 +1952,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_RESOURCE_OFFSET) ||
 		    (start_reg >= PACKET3_SET_RESOURCE_END) ||
 		    (end_reg >= PACKET3_SET_RESOURCE_END)) {
-			DRM_ERROR("bad SET_RESOURCE\n");
+			dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 			return -EINVAL;
 		}
 		for (i = 0; i < (pkt->count / 7); i++) {
@@ -1959,7 +1964,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				/* tex base */
 				r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 				if (r) {
-					DRM_ERROR("bad SET_RESOURCE\n");
+					dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 					return -EINVAL;
 				}
 				base_offset = (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1973,7 +1978,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				/* tex mip base */
 				r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 				if (r) {
-					DRM_ERROR("bad SET_RESOURCE\n");
+					dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 					return -EINVAL;
 				}
 				mip_offset = (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -1994,15 +1999,15 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				/* vtx base */
 				r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 				if (r) {
-					DRM_ERROR("bad SET_RESOURCE\n");
+					dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 					return -EINVAL;
 				}
 				offset = radeon_get_ib_value(p, idx+1+(i*7)+0);
 				size = radeon_get_ib_value(p, idx+1+(i*7)+1) + 1;
 				if (p->rdev && (size + offset) > radeon_bo_size(reloc->robj)) {
 					/* force size to size of the buffer */
-					dev_warn(p->dev, "vbo resource seems too big (%d) for the bo (%ld)\n",
-						 size + offset, radeon_bo_size(reloc->robj));
+					dev_warn_once(p->dev, "vbo resource seems too big (%d) for the bo (%ld)\n",
+						      size + offset, radeon_bo_size(reloc->robj));
 					ib[idx+1+(i*7)+1] = radeon_bo_size(reloc->robj) - offset;
 				}
 
@@ -2015,7 +2020,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			case SQ_TEX_VTX_INVALID_TEXTURE:
 			case SQ_TEX_VTX_INVALID_BUFFER:
 			default:
-				DRM_ERROR("bad SET_RESOURCE\n");
+				dev_warn_once(p->dev, "bad SET_RESOURCE\n");
 				return -EINVAL;
 			}
 		}
@@ -2027,7 +2032,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			if ((start_reg < PACKET3_SET_ALU_CONST_OFFSET) ||
 			    (start_reg >= PACKET3_SET_ALU_CONST_END) ||
 			    (end_reg >= PACKET3_SET_ALU_CONST_END)) {
-				DRM_ERROR("bad SET_ALU_CONST\n");
+				dev_warn_once(p->dev, "bad SET_ALU_CONST\n");
 				return -EINVAL;
 			}
 		}
@@ -2038,7 +2043,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_BOOL_CONST_OFFSET) ||
 		    (start_reg >= PACKET3_SET_BOOL_CONST_END) ||
 		    (end_reg >= PACKET3_SET_BOOL_CONST_END)) {
-			DRM_ERROR("bad SET_BOOL_CONST\n");
+			dev_warn_once(p->dev, "bad SET_BOOL_CONST\n");
 			return -EINVAL;
 		}
 		break;
@@ -2048,7 +2053,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_LOOP_CONST_OFFSET) ||
 		    (start_reg >= PACKET3_SET_LOOP_CONST_END) ||
 		    (end_reg >= PACKET3_SET_LOOP_CONST_END)) {
-			DRM_ERROR("bad SET_LOOP_CONST\n");
+			dev_warn_once(p->dev, "bad SET_LOOP_CONST\n");
 			return -EINVAL;
 		}
 		break;
@@ -2058,13 +2063,13 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_CTL_CONST_OFFSET) ||
 		    (start_reg >= PACKET3_SET_CTL_CONST_END) ||
 		    (end_reg >= PACKET3_SET_CTL_CONST_END)) {
-			DRM_ERROR("bad SET_CTL_CONST\n");
+			dev_warn_once(p->dev, "bad SET_CTL_CONST\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_SET_SAMPLER:
 		if (pkt->count % 3) {
-			DRM_ERROR("bad SET_SAMPLER\n");
+			dev_warn_once(p->dev, "bad SET_SAMPLER\n");
 			return -EINVAL;
 		}
 		start_reg = (idx_value << 2) + PACKET3_SET_SAMPLER_OFFSET;
@@ -2072,22 +2077,22 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		if ((start_reg < PACKET3_SET_SAMPLER_OFFSET) ||
 		    (start_reg >= PACKET3_SET_SAMPLER_END) ||
 		    (end_reg >= PACKET3_SET_SAMPLER_END)) {
-			DRM_ERROR("bad SET_SAMPLER\n");
+			dev_warn_once(p->dev, "bad SET_SAMPLER\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_STRMOUT_BASE_UPDATE:
 		/* RS780 and RS880 also need this */
 		if (p->family < CHIP_RS780) {
-			DRM_ERROR("STRMOUT_BASE_UPDATE only supported on 7xx\n");
+			dev_warn_once(p->dev, "STRMOUT_BASE_UPDATE only supported on 7xx\n");
 			return -EINVAL;
 		}
 		if (pkt->count != 1) {
-			DRM_ERROR("bad STRMOUT_BASE_UPDATE packet count\n");
+			dev_warn_once(p->dev, "bad STRMOUT_BASE_UPDATE packet count\n");
 			return -EINVAL;
 		}
 		if (idx_value > 3) {
-			DRM_ERROR("bad STRMOUT_BASE_UPDATE index\n");
+			dev_warn_once(p->dev, "bad STRMOUT_BASE_UPDATE index\n");
 			return -EINVAL;
 		}
 		{
@@ -2095,25 +2100,27 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad STRMOUT_BASE_UPDATE reloc\n");
+				dev_warn_once(p->dev, "bad STRMOUT_BASE_UPDATE reloc\n");
 				return -EINVAL;
 			}
 
 			if (reloc->robj != track->vgt_strmout_bo[idx_value]) {
-				DRM_ERROR("bad STRMOUT_BASE_UPDATE, bo does not match\n");
+				dev_warn_once(p->dev, "bad STRMOUT_BASE_UPDATE, bo does not match\n");
 				return -EINVAL;
 			}
 
 			offset = (u64)radeon_get_ib_value(p, idx+1) << 8;
 			if (offset != track->vgt_strmout_bo_offset[idx_value]) {
-				DRM_ERROR("bad STRMOUT_BASE_UPDATE, bo offset does not match: 0x%llx, 0x%x\n",
-					  offset, track->vgt_strmout_bo_offset[idx_value]);
+				dev_warn_once(p->dev,
+					      "bad STRMOUT_BASE_UPDATE, bo offset does not match: 0x%llx, 0x%x\n",
+					      offset, track->vgt_strmout_bo_offset[idx_value]);
 				return -EINVAL;
 			}
 
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad STRMOUT_BASE_UPDATE bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev,
+					      "bad STRMOUT_BASE_UPDATE bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			ib[idx+1] += (u32)((reloc->gpu_offset >> 8) & 0xffffffff);
@@ -2121,17 +2128,17 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		break;
 	case PACKET3_SURFACE_BASE_UPDATE:
 		if (p->family >= CHIP_RV770 || p->family == CHIP_R600) {
-			DRM_ERROR("bad SURFACE_BASE_UPDATE\n");
+			dev_warn_once(p->dev, "bad SURFACE_BASE_UPDATE\n");
 			return -EINVAL;
 		}
 		if (pkt->count) {
-			DRM_ERROR("bad SURFACE_BASE_UPDATE\n");
+			dev_warn_once(p->dev, "bad SURFACE_BASE_UPDATE\n");
 			return -EINVAL;
 		}
 		break;
 	case PACKET3_STRMOUT_BUFFER_UPDATE:
 		if (pkt->count != 4) {
-			DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (invalid count)\n");
+			dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE (invalid count)\n");
 			return -EINVAL;
 		}
 		/* Updating memory at DST_ADDRESS. */
@@ -2139,14 +2146,15 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			u64 offset;
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n");
+				dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE (missing dst reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+1);
 			offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE dst bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev,
+					      "bad STRMOUT_BUFFER_UPDATE dst bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2158,14 +2166,15 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			u64 offset;
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n");
+				dev_warn_once(p->dev, "bad STRMOUT_BUFFER_UPDATE (missing src reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+3);
 			offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad STRMOUT_BUFFER_UPDATE src bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev,
+					      "bad STRMOUT_BUFFER_UPDATE src bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2178,23 +2187,23 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		u64 offset;
 
 		if (pkt->count != 3) {
-			DRM_ERROR("bad MEM_WRITE (invalid count)\n");
+			dev_warn_once(p->dev, "bad MEM_WRITE (invalid count)\n");
 			return -EINVAL;
 		}
 		r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 		if (r) {
-			DRM_ERROR("bad MEM_WRITE (missing reloc)\n");
+			dev_warn_once(p->dev, "bad MEM_WRITE (missing reloc)\n");
 			return -EINVAL;
 		}
 		offset = radeon_get_ib_value(p, idx+0);
 		offset += ((u64)(radeon_get_ib_value(p, idx+1) & 0xff)) << 32UL;
 		if (offset & 0x7) {
-			DRM_ERROR("bad MEM_WRITE (address not qwords aligned)\n");
+			dev_warn_once(p->dev, "bad MEM_WRITE (address not qwords aligned)\n");
 			return -EINVAL;
 		}
 		if ((offset + 8) > radeon_bo_size(reloc->robj)) {
-			DRM_ERROR("bad MEM_WRITE bo too small: 0x%llx, 0x%lx\n",
-				  offset + 8, radeon_bo_size(reloc->robj));
+			dev_warn_once(p->dev, "bad MEM_WRITE bo too small: 0x%llx, 0x%lx\n",
+				      offset + 8, radeon_bo_size(reloc->robj));
 			return -EINVAL;
 		}
 		offset += reloc->gpu_offset;
@@ -2204,7 +2213,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 	}
 	case PACKET3_COPY_DW:
 		if (pkt->count != 4) {
-			DRM_ERROR("bad COPY_DW (invalid count)\n");
+			dev_warn_once(p->dev, "bad COPY_DW (invalid count)\n");
 			return -EINVAL;
 		}
 		if (idx_value & 0x1) {
@@ -2212,14 +2221,14 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			/* SRC is memory. */
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad COPY_DW (missing src reloc)\n");
+				dev_warn_once(p->dev, "bad COPY_DW (missing src reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+1);
 			offset += ((u64)(radeon_get_ib_value(p, idx+2) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad COPY_DW src bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad COPY_DW src bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2236,14 +2245,14 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			/* DST is memory. */
 			r = radeon_cs_packet_next_reloc(p, &reloc, r600_nomm);
 			if (r) {
-				DRM_ERROR("bad COPY_DW (missing dst reloc)\n");
+				dev_warn_once(p->dev, "bad COPY_DW (missing dst reloc)\n");
 				return -EINVAL;
 			}
 			offset = radeon_get_ib_value(p, idx+3);
 			offset += ((u64)(radeon_get_ib_value(p, idx+4) & 0xff)) << 32;
 			if ((offset + 4) > radeon_bo_size(reloc->robj)) {
-				DRM_ERROR("bad COPY_DW dst bo too small: 0x%llx, 0x%lx\n",
-					  offset + 4, radeon_bo_size(reloc->robj));
+				dev_warn_once(p->dev, "bad COPY_DW dst bo too small: 0x%llx, 0x%lx\n",
+					      offset + 4, radeon_bo_size(reloc->robj));
 				return -EINVAL;
 			}
 			offset += reloc->gpu_offset;
@@ -2259,7 +2268,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 	case PACKET3_NOP:
 		break;
 	default:
-		DRM_ERROR("Packet3 opcode %x not supported\n", pkt->opcode);
+		dev_warn_once(p->dev, "Packet3 opcode %x not supported\n", pkt->opcode);
 		return -EINVAL;
 	}
 	return 0;
@@ -2306,7 +2315,7 @@ int r600_cs_parse(struct radeon_cs_parser *p)
 			r = r600_packet3_check(p, &pkt);
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d !\n", pkt.type);
+			dev_warn_once(p->dev, "Unknown packet type %d !\n", pkt.type);
 			kfree(p->track);
 			p->track = NULL;
 			return -EINVAL;
@@ -2346,13 +2355,13 @@ int r600_dma_cs_next_reloc(struct radeon_cs_parser *p,
 
 	*cs_reloc = NULL;
 	if (p->chunk_relocs == NULL) {
-		DRM_ERROR("No relocation chunk !\n");
+		dev_warn_once(p->dev, "No relocation chunk !\n");
 		return -EINVAL;
 	}
 	idx = p->dma_reloc_idx;
 	if (idx >= p->nrelocs) {
-		DRM_ERROR("Relocs at %d after relocations chunk end %d !\n",
-			  idx, p->nrelocs);
+		dev_warn_once(p->dev, "Relocs at %d after relocations chunk end %d !\n",
+			      idx, p->nrelocs);
 		return -EINVAL;
 	}
 	*cs_reloc = &p->relocs[idx];
@@ -2385,8 +2394,8 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
 
 	do {
 		if (p->idx >= ib_chunk->length_dw) {
-			DRM_ERROR("Can not parse packet at %d after CS end %d !\n",
-				  p->idx, ib_chunk->length_dw);
+			dev_warn_once(p->dev, "Can not parse packet at %d after CS end %d !\n",
+				      p->idx, ib_chunk->length_dw);
 			return -EINVAL;
 		}
 		idx = p->idx;
@@ -2399,7 +2408,7 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
 		case DMA_PACKET_WRITE:
 			r = r600_dma_cs_next_reloc(p, &dst_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_WRITE\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_WRITE\n");
 				return -EINVAL;
 			}
 			if (tiled) {
@@ -2417,20 +2426,20 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
 				p->idx += count + 3;
 			}
 			if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-				dev_warn(p->dev, "DMA write buffer too small (%llu %lu)\n",
-					 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+				dev_warn_once(p->dev, "DMA write buffer too small (%llu %lu)\n",
+					      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 				return -EINVAL;
 			}
 			break;
 		case DMA_PACKET_COPY:
 			r = r600_dma_cs_next_reloc(p, &src_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_COPY\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_COPY\n");
 				return -EINVAL;
 			}
 			r = r600_dma_cs_next_reloc(p, &dst_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_COPY\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_COPY\n");
 				return -EINVAL;
 			}
 			if (tiled) {
@@ -2484,31 +2493,31 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
 				}
 			}
 			if ((src_offset + (count * 4)) > radeon_bo_size(src_reloc->robj)) {
-				dev_warn(p->dev, "DMA copy src buffer too small (%llu %lu)\n",
-					 src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
+				dev_warn_once(p->dev, "DMA copy src buffer too small (%llu %lu)\n",
+					      src_offset + (count * 4), radeon_bo_size(src_reloc->robj));
 				return -EINVAL;
 			}
 			if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-				dev_warn(p->dev, "DMA write dst buffer too small (%llu %lu)\n",
-					 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+				dev_warn_once(p->dev, "DMA write dst buffer too small (%llu %lu)\n",
+					      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 				return -EINVAL;
 			}
 			break;
 		case DMA_PACKET_CONSTANT_FILL:
 			if (p->family < CHIP_RV770) {
-				DRM_ERROR("Constant Fill is 7xx only !\n");
+				dev_warn_once(p->dev, "Constant Fill is 7xx only !\n");
 				return -EINVAL;
 			}
 			r = r600_dma_cs_next_reloc(p, &dst_reloc);
 			if (r) {
-				DRM_ERROR("bad DMA_PACKET_WRITE\n");
+				dev_warn_once(p->dev, "bad DMA_PACKET_WRITE\n");
 				return -EINVAL;
 			}
 			dst_offset = radeon_get_ib_value(p, idx+1);
 			dst_offset |= ((u64)(radeon_get_ib_value(p, idx+3) & 0x00ff0000)) << 16;
 			if ((dst_offset + (count * 4)) > radeon_bo_size(dst_reloc->robj)) {
-				dev_warn(p->dev, "DMA constant fill buffer too small (%llu %lu)\n",
-					 dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
+				dev_warn_once(p->dev, "DMA constant fill buffer too small (%llu %lu)\n",
+					      dst_offset + (count * 4), radeon_bo_size(dst_reloc->robj));
 				return -EINVAL;
 			}
 			ib[idx+1] += (u32)(dst_reloc->gpu_offset & 0xfffffffc);
@@ -2519,7 +2528,7 @@ int r600_dma_cs_parse(struct radeon_cs_parser *p)
 			p->idx += 1;
 			break;
 		default:
-			DRM_ERROR("Unknown packet type %d at %d !\n", cmd, idx);
+			dev_warn_once(p->dev, "Unknown packet type %d at %d !\n", cmd, idx);
 			return -EINVAL;
 		}
 	} while (p->idx < p->chunk_ib->length_dw);
diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c
index b8e6202f1d5b..3f9c0011244f 100644
--- a/drivers/gpu/drm/radeon/radeon_cs.c
+++ b/drivers/gpu/drm/radeon/radeon_cs.c
@@ -834,7 +834,7 @@ void radeon_cs_dump_packet(struct radeon_cs_parser *p,
 	ib = p->ib.ptr;
 	idx = pkt->idx;
 	for (i = 0; i <= (pkt->count + 1); i++, idx++)
-		DRM_INFO("ib[%d]=0x%08X\n", idx, ib[idx]);
+		dev_dbg(p->dev, "ib[%d]=0x%08X\n", idx, ib[idx]);
 }
 
 /**

From 64cc12f9798f146f2c616db4885d653741d3b3c1 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 4 Sep 2025 21:58:49 +0300
Subject: [PATCH 1055/1455] drm/amdgpu: Fix error codes if copy_to_user() fails

The copy_to_user() function returns the number of bytes that it wasn't
able to copy, but we should return -EFAULT to the user.

Fixes: 4d82724f7f2b ("drm/amdgpu: Add mapping info option for GEM_OP ioctl")
Fixes: f9db1fc52ceb ("drm/amdgpu: Add ioctl to get all gem handles for a process")
Reviewed-By: David Francis <David.Francis@amd.com>
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 8bec41cdccd7..630175746780 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -1066,7 +1066,8 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
 		drm_exec_fini(&exec);
 
 		if (num_mappings > 0 && num_mappings <= args->num_entries)
-			r = copy_to_user(u64_to_user_ptr(args->value), vm_entries, num_mappings * sizeof(*vm_entries));
+			if (copy_to_user(u64_to_user_ptr(args->value), vm_entries, num_mappings * sizeof(*vm_entries)))
+				r = -EFAULT;
 
 		args->num_entries = num_mappings;
 
@@ -1158,7 +1159,8 @@ int amdgpu_gem_list_handles_ioctl(struct drm_device *dev, void *data,
 	args->num_entries = bo_index;
 
 	if (!ret)
-		ret = copy_to_user(u64_to_user_ptr(args->entries), bo_entries, num_bos * sizeof(*bo_entries));
+		if (copy_to_user(u64_to_user_ptr(args->entries), bo_entries, num_bos * sizeof(*bo_entries)))
+			ret = -EFAULT;
 
 	kvfree(bo_entries);
 

From 2fd653b9bb5aacec5d4c421ab290905898fe85a2 Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <mario.limonciello@amd.com>
Date: Fri, 5 Sep 2025 10:36:27 -0500
Subject: [PATCH 1056/1455] drm/amd/display: Drop dm_prepare_suspend() and
 dm_complete()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
dm_prepare_suspend() was added in commit 50e0bae34fa6b
("drm/amd/display: Add and use new dm_prepare_suspend() callback")
to allow display to turn off earlier in the suspend sequence.

This caused a regression that HDMI audio sometimes didn't work
properly after resume unless audio was playing during suspend.

[How]
Drop dm_prepare_suspend() callback. All code in it will still run
during dm_suspend(). Also drop unnecessary dm_complete() callback.
dm_complete() was used for failed prepare and also for any case
of successful resume.  The code in it already runs in dm_resume().

This change will introduce more time that the display is turned on
during suspend sequence. The compositor can turn it off sooner if
desired.

Cc: Harry Wentland <harry.wentland@amd.com>
Reported-by: Przemysław Kopa <prz.kopa@gmail.com>
Closes: https://lore.kernel.org/amd-gfx/1cea0d56-7739-4ad9-bf8e-c9330faea2bb@kernel.org/T/#m383d9c08397043a271b36c32b64bb80e524e4b0f
Reported-by: Kalvin <hikaph+oss@gmail.com>
Closes: https://github.com/alsa-project/alsa-lib/issues/465
Closes: https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/4809
Fixes: 50e0bae34fa6b ("drm/amd/display: Add and use new dm_prepare_suspend() callback")
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 21 -------------------
 1 file changed, 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index e34d98a945f2..fadc6098eaee 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3182,25 +3182,6 @@ static void dm_destroy_cached_state(struct amdgpu_device *adev)
 	dm->cached_state = NULL;
 }
 
-static void dm_complete(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-
-	dm_destroy_cached_state(adev);
-}
-
-static int dm_prepare_suspend(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-
-	if (amdgpu_in_reset(adev))
-		return 0;
-
-	WARN_ON(adev->dm.cached_state);
-
-	return dm_cache_state(adev);
-}
-
 static int dm_suspend(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -3626,10 +3607,8 @@ static const struct amd_ip_funcs amdgpu_dm_funcs = {
 	.early_fini = amdgpu_dm_early_fini,
 	.hw_init = dm_hw_init,
 	.hw_fini = dm_hw_fini,
-	.prepare_suspend = dm_prepare_suspend,
 	.suspend = dm_suspend,
 	.resume = dm_resume,
-	.complete = dm_complete,
 	.is_idle = dm_is_idle,
 	.wait_for_idle = dm_wait_for_idle,
 	.check_soft_reset = dm_check_soft_reset,

From e2580413a83680f679904ad2f2c1aa6969876469 Mon Sep 17 00:00:00 2001
From: Danilo Krummrich <dakr@kernel.org>
Date: Mon, 1 Sep 2025 17:01:53 +0200
Subject: [PATCH 1057/1455] gpu: nova-core: take advantage of
 pci::Device::unbind()

Now that we have pci::Device::unbind() we can unregister the sysmem
flush page with a direct access the I/O resource, i.e. without RCU read
side critical section.

Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Alexandre Courbot <acourbot@nvidia.com>
Link: https://lore.kernel.org/r/20250901150207.63094-1-dakr@kernel.org
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/driver.rs |  4 ++++
 drivers/gpu/nova-core/gpu.rs    | 22 ++++++++++++----------
 2 files changed, 16 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/nova-core/driver.rs b/drivers/gpu/nova-core/driver.rs
index 274989ea1fb4..02b3edd7bbdc 100644
--- a/drivers/gpu/nova-core/driver.rs
+++ b/drivers/gpu/nova-core/driver.rs
@@ -54,4 +54,8 @@ impl pci::Driver for NovaCore {
 
         Ok(this)
     }
+
+    fn unbind(pdev: &pci::Device<Core>, this: Pin<&Self>) {
+        this.gpu.unbind(pdev.as_ref());
+    }
 }
diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index 4d0f759610ec..a0878ecdc18b 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -163,7 +163,7 @@ impl Spec {
 }
 
 /// Structure holding the resources required to operate the GPU.
-#[pin_data(PinnedDrop)]
+#[pin_data]
 pub(crate) struct Gpu {
     spec: Spec,
     /// MMIO mapping of PCI BAR 0
@@ -174,15 +174,6 @@ pub(crate) struct Gpu {
     sysmem_flush: SysmemFlush,
 }
 
-#[pinned_drop]
-impl PinnedDrop for Gpu {
-    fn drop(self: Pin<&mut Self>) {
-        // Unregister the sysmem flush page before we release it.
-        self.bar
-            .try_access_with(|b| self.sysmem_flush.unregister(b));
-    }
-}
-
 impl Gpu {
     /// Helper function to load and run the FWSEC-FRTS firmware and confirm that it has properly
     /// created the WPR2 region.
@@ -309,4 +300,15 @@ impl Gpu {
             sysmem_flush,
         }))
     }
+
+    /// Called when the corresponding [`Device`](device::Device) is unbound.
+    ///
+    /// Note: This method must only be called from `Driver::unbind`.
+    pub(crate) fn unbind(&self, dev: &device::Device<device::Core>) {
+        kernel::warn_on!(self
+            .bar
+            .access(dev)
+            .inspect(|bar| self.sysmem_flush.unregister(bar))
+            .is_err());
+    }
 }

From 278f8904434aa96055e793936b5977c010549e28 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Wed, 20 Aug 2025 17:04:25 -0700
Subject: [PATCH 1058/1455] drm/msm: Fix obj leak in VM_BIND error path

If we fail a handle-lookup part way thru, we need to drop the already
obtained obj references.

Fixes: 2e6a8a1fe2b2 ("drm/msm: Add VM_BIND ioctl")
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Tested-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/669784/
---
 drivers/gpu/drm/msm/msm_gem_vma.c | 25 +++++++++++++++++++------
 1 file changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 00d0f3b7ba32..209154be5efc 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -1023,6 +1023,7 @@ vm_bind_job_lookup_ops(struct msm_vm_bind_job *job, struct drm_msm_vm_bind *args
 	struct drm_device *dev = job->vm->drm;
 	int ret = 0;
 	int cnt = 0;
+	int i = -1;
 
 	if (args->nr_ops == 1) {
 		/* Single op case, the op is inlined: */
@@ -1056,11 +1057,12 @@ vm_bind_job_lookup_ops(struct msm_vm_bind_job *job, struct drm_msm_vm_bind *args
 
 	spin_lock(&file->table_lock);
 
-	for (unsigned i = 0; i < args->nr_ops; i++) {
+	for (i = 0; i < args->nr_ops; i++) {
+		struct msm_vm_bind_op *op = &job->ops[i];
 		struct drm_gem_object *obj;
 
-		if (!job->ops[i].handle) {
-			job->ops[i].obj = NULL;
+		if (!op->handle) {
+			op->obj = NULL;
 			continue;
 		}
 
@@ -1068,15 +1070,15 @@ vm_bind_job_lookup_ops(struct msm_vm_bind_job *job, struct drm_msm_vm_bind *args
 		 * normally use drm_gem_object_lookup(), but for bulk lookup
 		 * all under single table_lock just hit object_idr directly:
 		 */
-		obj = idr_find(&file->object_idr, job->ops[i].handle);
+		obj = idr_find(&file->object_idr, op->handle);
 		if (!obj) {
-			ret = UERR(EINVAL, dev, "invalid handle %u at index %u\n", job->ops[i].handle, i);
+			ret = UERR(EINVAL, dev, "invalid handle %u at index %u\n", op->handle, i);
 			goto out_unlock;
 		}
 
 		drm_gem_object_get(obj);
 
-		job->ops[i].obj = obj;
+		op->obj = obj;
 		cnt++;
 	}
 
@@ -1085,6 +1087,17 @@ vm_bind_job_lookup_ops(struct msm_vm_bind_job *job, struct drm_msm_vm_bind *args
 out_unlock:
 	spin_unlock(&file->table_lock);
 
+	if (ret) {
+		for (; i >= 0; i--) {
+			struct msm_vm_bind_op *op = &job->ops[i];
+
+			if (!op->obj)
+				continue;
+
+			drm_gem_object_put(op->obj);
+			op->obj = NULL;
+		}
+	}
 out:
 	return ret;
 }

From 3a3bef68a6c15d079646a964ebc4dc8bb0aedb06 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Wed, 20 Aug 2025 17:04:26 -0700
Subject: [PATCH 1059/1455] drm/msm: Fix missing VM_BIND offset/range
 validation

We need to reject the MAP op if offset+range is larger than the BO size.

Reported-by: Connor Abbott <cwabbott0@gmail.com>
Fixes: 2e6a8a1fe2b2 ("drm/msm: Add VM_BIND ioctl")
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Tested-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/669781/
---
 drivers/gpu/drm/msm/msm_gem_vma.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c
index 209154be5efc..381a0853c05b 100644
--- a/drivers/gpu/drm/msm/msm_gem_vma.c
+++ b/drivers/gpu/drm/msm/msm_gem_vma.c
@@ -1080,6 +1080,12 @@ vm_bind_job_lookup_ops(struct msm_vm_bind_job *job, struct drm_msm_vm_bind *args
 
 		op->obj = obj;
 		cnt++;
+
+		if ((op->range + op->obj_offset) > obj->size) {
+			ret = UERR(EINVAL, dev, "invalid range: %016llx + %016llx > %016zx\n",
+				   op->range, op->obj_offset, obj->size);
+			goto out_unlock;
+		}
 	}
 
 	*nr_bos = cnt;

From 6d6a29a19b232e29b61a14d58b71fefc2e003fa4 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Wed, 20 Aug 2025 17:04:27 -0700
Subject: [PATCH 1060/1455] drm/msm: Fix 32b size truncation

Somehow we never noticed this when arm64 became a thing, many years ago.

v2: also fix npages

Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Tested-by: Connor Abbott <cwabbott0@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/669785/
---
 drivers/gpu/drm/msm/msm_gem.c       | 21 ++++++++++-----------
 drivers/gpu/drm/msm/msm_gem.h       |  6 +++---
 drivers/gpu/drm/msm/msm_gem_prime.c |  2 +-
 3 files changed, 14 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index 9f0f5b77f1bd..e6cb51f21d97 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -192,7 +192,7 @@ static struct page **get_pages(struct drm_gem_object *obj)
 	if (!msm_obj->pages) {
 		struct drm_device *dev = obj->dev;
 		struct page **p;
-		int npages = obj->size >> PAGE_SHIFT;
+		size_t npages = obj->size >> PAGE_SHIFT;
 
 		p = drm_gem_get_pages(obj);
 
@@ -1149,7 +1149,7 @@ static int msm_gem_object_mmap(struct drm_gem_object *obj, struct vm_area_struct
 
 /* convenience method to construct a GEM buffer object, and userspace handle */
 int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
-		uint32_t size, uint32_t flags, uint32_t *handle,
+		size_t size, uint32_t flags, uint32_t *handle,
 		char *name)
 {
 	struct drm_gem_object *obj;
@@ -1215,9 +1215,8 @@ static const struct drm_gem_object_funcs msm_gem_object_funcs = {
 	.vm_ops = &vm_ops,
 };
 
-static int msm_gem_new_impl(struct drm_device *dev,
-		uint32_t size, uint32_t flags,
-		struct drm_gem_object **obj)
+static int msm_gem_new_impl(struct drm_device *dev, uint32_t flags,
+			    struct drm_gem_object **obj)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_gem_object *msm_obj;
@@ -1251,7 +1250,7 @@ static int msm_gem_new_impl(struct drm_device *dev,
 	return 0;
 }
 
-struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32_t flags)
+struct drm_gem_object *msm_gem_new(struct drm_device *dev, size_t size, uint32_t flags)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_gem_object *msm_obj;
@@ -1266,7 +1265,7 @@ struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32
 	if (size == 0)
 		return ERR_PTR(-EINVAL);
 
-	ret = msm_gem_new_impl(dev, size, flags, &obj);
+	ret = msm_gem_new_impl(dev, flags, &obj);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -1306,12 +1305,12 @@ struct drm_gem_object *msm_gem_import(struct drm_device *dev,
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_gem_object *msm_obj;
 	struct drm_gem_object *obj;
-	uint32_t size;
-	int ret, npages;
+	size_t size, npages;
+	int ret;
 
 	size = PAGE_ALIGN(dmabuf->size);
 
-	ret = msm_gem_new_impl(dev, size, MSM_BO_WC, &obj);
+	ret = msm_gem_new_impl(dev, MSM_BO_WC, &obj);
 	if (ret)
 		return ERR_PTR(ret);
 
@@ -1354,7 +1353,7 @@ fail:
 	return ERR_PTR(ret);
 }
 
-void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, uint32_t flags,
+void *msm_gem_kernel_new(struct drm_device *dev, size_t size, uint32_t flags,
 			 struct drm_gpuvm *vm, struct drm_gem_object **bo,
 			 uint64_t *iova)
 {
diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h
index 751c3b4965bc..a4cf31853c50 100644
--- a/drivers/gpu/drm/msm/msm_gem.h
+++ b/drivers/gpu/drm/msm/msm_gem.h
@@ -297,10 +297,10 @@ bool msm_gem_active(struct drm_gem_object *obj);
 int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout);
 int msm_gem_cpu_fini(struct drm_gem_object *obj);
 int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file,
-		uint32_t size, uint32_t flags, uint32_t *handle, char *name);
+		size_t size, uint32_t flags, uint32_t *handle, char *name);
 struct drm_gem_object *msm_gem_new(struct drm_device *dev,
-		uint32_t size, uint32_t flags);
-void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, uint32_t flags,
+		size_t size, uint32_t flags);
+void *msm_gem_kernel_new(struct drm_device *dev, size_t size, uint32_t flags,
 			 struct drm_gpuvm *vm, struct drm_gem_object **bo,
 			 uint64_t *iova);
 void msm_gem_kernel_put(struct drm_gem_object *bo, struct drm_gpuvm *vm);
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
index c0a33ac839cb..036d34c674d9 100644
--- a/drivers/gpu/drm/msm/msm_gem_prime.c
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -15,7 +15,7 @@
 struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-	int npages = obj->size >> PAGE_SHIFT;
+	size_t npages = obj->size >> PAGE_SHIFT;
 
 	if (msm_obj->flags & MSM_BO_NO_SHARE)
 		return ERR_PTR(-EINVAL);

From 9e710a2a2fd97052592bc5cad37bbe2e6968e5ce Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Thu, 17 Jul 2025 08:12:01 -0700
Subject: [PATCH 1061/1455] drm/msm: Drop unneeded NULL check

This is always set in msm_gpu_init(), and can never be NULL.

Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/664431/
---
 drivers/gpu/drm/msm/msm_iommu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 76cdd5ea06a0..89cdda73711a 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -776,7 +776,7 @@ struct msm_mmu *msm_iommu_gpu_new(struct device *dev, struct msm_gpu *gpu, unsig
 		return mmu;
 
 	iommu = to_msm_iommu(mmu);
-	if (adreno_smmu && adreno_smmu->cookie) {
+	if (adreno_smmu->cookie) {
 		const struct io_pgtable_cfg *cfg =
 			adreno_smmu->get_ttbr1_cfg(adreno_smmu->cookie);
 		size_t tblsz = get_tblsz(cfg);

From b4e3429a04e2aaf166270089a01832d80f534da7 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Fri, 25 Jul 2025 10:35:39 +0200
Subject: [PATCH 1062/1455] drm/msm: adreno: a6xx: enable GMU bandwidth voting
 for x1e80100 GPU

The Adreno GPU Management Unit (GMU) can also scale DDR Bandwidth along
the Frequency and Power Domain level, but by default we leave the
OPP core scale the interconnect ddr path.

Declare the Bus Control Modules (BCMs) and the corresponding parameters
in the GPU info struct to allow the GMU to vote for the bandwidth.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Reviewed-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/665778/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index 00e1afd46b81..892f98b1f2ae 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -1440,6 +1440,17 @@ static const struct adreno_info a7xx_gpus[] = {
 			.pwrup_reglist = &a7xx_pwrup_reglist,
 			.gmu_chipid = 0x7050001,
 			.gmu_cgc_mode = 0x00020202,
+			.bcms = (const struct a6xx_bcm[]) {
+				{ .name = "SH0", .buswidth = 16 },
+				{ .name = "MC0", .buswidth = 4 },
+				{
+					.name = "ACV",
+					.fixed = true,
+					.perfmode = BIT(3),
+					.perfmode_bw = 16500000,
+				},
+				{ /* sentinel */ },
+			},
 		},
 		.preempt_record_size = 4192 * SZ_1K,
 		.speedbins = ADRENO_SPEEDBINS(

From 10fb1b2fcaee5545a5e54db1ed4d7b15c2db50c8 Mon Sep 17 00:00:00 2001
From: Antonino Maniscalco <antomani103@gmail.com>
Date: Thu, 21 Aug 2025 15:06:34 +0200
Subject: [PATCH 1063/1455] drm/msm: make sure to not queue up recovery more
 than once

If two fault IRQs arrive in short succession recovery work will be
queued up twice.

When recovery runs a second time it may end up killing an unrelated
context.

Prevent this by masking off interrupts when triggering recovery.

Signed-off-by: Antonino Maniscalco <antomani103@gmail.com>
Reviewed-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/670023/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 45dd5fd1c2bf..f8992a68df7f 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1727,6 +1727,9 @@ static void a6xx_fault_detect_irq(struct msm_gpu *gpu)
 	/* Turn off the hangcheck timer to keep it from bothering us */
 	timer_delete(&gpu->hangcheck_timer);
 
+	/* Turn off interrupts to avoid triggering recovery again */
+	gpu_write(gpu, REG_A6XX_RBBM_INT_0_MASK, 0);
+
 	kthread_queue_work(gpu->worker, &gpu->recover_work);
 }
 

From 0c5300343d0c622f7852145a763c570fbaf68a48 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Fri, 22 Aug 2025 00:25:26 +0530
Subject: [PATCH 1064/1455] drm/msm/adreno: Add speedbins for A663 GPU

Add speedbin mappings for A663 GPU.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/670096/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index 892f98b1f2ae..fcfe89eeccf2 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -1024,6 +1024,11 @@ static const struct adreno_info a6xx_gpus[] = {
 			.gmu_cgc_mode = 0x00020200,
 			.prim_fifo_threshold = 0x00300200,
 		},
+		.speedbins = ADRENO_SPEEDBINS(
+			{ 0,   0 },
+			{ 169, 0 },
+			{ 113, 1 },
+		),
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(0x06030500),
 		.family = ADRENO_6XX_GEN4,

From c2cc1e60c1afff4f23c22561b57a5d5157dde20d Mon Sep 17 00:00:00 2001
From: Jie Zhang <quic_jiezh@quicinc.com>
Date: Wed, 3 Sep 2025 12:49:52 +0530
Subject: [PATCH 1065/1455] dt-bindings: display/msm/gmu: Update Adreno 623
 bindings

Update Adreno 623's dt-binding to remove smmu_clk which is not required
for this GMU.

Signed-off-by: Jie Zhang <quic_jiezh@quicinc.com>
Signed-off-by: Akhil P Oommen <quic_akhilpo@quicinc.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Patchwork: https://patchwork.freedesktop.org/patch/672455/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 .../devicetree/bindings/display/msm/gmu.yaml  | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/msm/gmu.yaml b/Documentation/devicetree/bindings/display/msm/gmu.yaml
index 4392aa7a4ffe..afc187935744 100644
--- a/Documentation/devicetree/bindings/display/msm/gmu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gmu.yaml
@@ -124,6 +124,40 @@ allOf:
           contains:
             enum:
               - qcom,adreno-gmu-623.0
+    then:
+      properties:
+        reg:
+          items:
+            - description: Core GMU registers
+            - description: Resource controller registers
+            - description: GMU PDC registers
+        reg-names:
+          items:
+            - const: gmu
+            - const: rscc
+            - const: gmu_pdc
+        clocks:
+          items:
+            - description: GMU clock
+            - description: GPU CX clock
+            - description: GPU AXI clock
+            - description: GPU MEMNOC clock
+            - description: GPU AHB clock
+            - description: GPU HUB CX clock
+        clock-names:
+          items:
+            - const: gmu
+            - const: cxo
+            - const: axi
+            - const: memnoc
+            - const: ahb
+            - const: hub
+
+  - if:
+      properties:
+        compatible:
+          contains:
+            enum:
               - qcom,adreno-gmu-635.0
               - qcom,adreno-gmu-660.1
               - qcom,adreno-gmu-663.0

From 485aaebd738bfd666cf446e3a6166faf4a5dde3f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 5 Sep 2025 10:27:05 +0300
Subject: [PATCH 1066/1455] drm/i915/psr: Pass intel_crtc_state instead of
 intel_dp in wait_for_idle
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is preparation to add own function for polling PSR being ready for
update when doing dsb commit.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250905072708.2659411-2-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 22433fe2ee14..6ab5c028845a 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -2997,10 +2997,11 @@ void intel_psr_post_plane_update(struct intel_atomic_state *state,
 	}
 }
 
-static int _psr2_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
+static int
+_psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
+	struct intel_display *display = to_intel_display(new_crtc_state);
+	enum transcoder cpu_transcoder = new_crtc_state->cpu_transcoder;
 
 	/*
 	 * Any state lower than EDP_PSR2_STATUS_STATE_DEEP_SLEEP is enough.
@@ -3012,10 +3013,11 @@ static int _psr2_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
 				       EDP_PSR2_STATUS_STATE_DEEP_SLEEP, 50);
 }
 
-static int _psr1_ready_for_pipe_update_locked(struct intel_dp *intel_dp)
+static int
+_psr1_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state)
 {
-	struct intel_display *display = to_intel_display(intel_dp);
-	enum transcoder cpu_transcoder = intel_dp->psr.transcoder;
+	struct intel_display *display = to_intel_display(new_crtc_state);
+	enum transcoder cpu_transcoder = new_crtc_state->cpu_transcoder;
 
 	/*
 	 * From bspec: Panel Self Refresh (BDW+)
@@ -3054,9 +3056,9 @@ void intel_psr_wait_for_idle_locked(const struct intel_crtc_state *new_crtc_stat
 			continue;
 
 		if (intel_dp->psr.sel_update_enabled)
-			ret = _psr2_ready_for_pipe_update_locked(intel_dp);
+			ret = _psr2_ready_for_pipe_update_locked(new_crtc_state);
 		else
-			ret = _psr1_ready_for_pipe_update_locked(intel_dp);
+			ret = _psr1_ready_for_pipe_update_locked(new_crtc_state);
 
 		if (ret)
 			drm_err(display->drm,

From a0f7f3b1654c28739e1ae338afa354c83946c3d3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 5 Sep 2025 10:27:06 +0300
Subject: [PATCH 1067/1455] drm/i915/psr: Add new define for PSR idle timeout
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we are using value 50ms as timeout for waiting PSR to idle. Add
own define for this purpose.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Link: https://lore.kernel.org/r/20250905072708.2659411-3-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 6ab5c028845a..eae1eedbbb26 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -2997,6 +2997,14 @@ void intel_psr_post_plane_update(struct intel_atomic_state *state,
 	}
 }
 
+/*
+ * From bspec: Panel Self Refresh (BDW+)
+ * Max. time for PSR to idle = Inverse of the refresh rate + 6 ms of
+ * exit training time + 1.5 ms of aux channel handshake. 50 ms is
+ * defensive enough to cover everything.
+ */
+#define PSR_IDLE_TIMEOUT_MS 50
+
 static int
 _psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state)
 {
@@ -3010,7 +3018,8 @@ _psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state
 	 */
 	return intel_de_wait_for_clear(display,
 				       EDP_PSR2_STATUS(display, cpu_transcoder),
-				       EDP_PSR2_STATUS_STATE_DEEP_SLEEP, 50);
+				       EDP_PSR2_STATUS_STATE_DEEP_SLEEP,
+				       PSR_IDLE_TIMEOUT_MS);
 }
 
 static int
@@ -3019,15 +3028,10 @@ _psr1_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state
 	struct intel_display *display = to_intel_display(new_crtc_state);
 	enum transcoder cpu_transcoder = new_crtc_state->cpu_transcoder;
 
-	/*
-	 * From bspec: Panel Self Refresh (BDW+)
-	 * Max. time for PSR to idle = Inverse of the refresh rate + 6 ms of
-	 * exit training time + 1.5 ms of aux channel handshake. 50 ms is
-	 * defensive enough to cover everything.
-	 */
 	return intel_de_wait_for_clear(display,
 				       psr_status_reg(display, cpu_transcoder),
-				       EDP_PSR_STATUS_STATE_MASK, 50);
+				       EDP_PSR_STATUS_STATE_MASK,
+				       PSR_IDLE_TIMEOUT_MS);
 }
 
 /**

From 11b0567ee605e95becc02f858f8f42706687a068 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 5 Sep 2025 10:27:07 +0300
Subject: [PATCH 1068/1455] drm/i915/psr: New interface adding PSR idle poll
 into dsb commit
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are currently observing crc failures after we started using dsb for PSR
updates as well. This seems to happen because PSR HW is still sending
couple of updates using old framebuffers on wake-up.

This patch is preparing to fix that by adding interface which can be used
to add poll ensuring PSR HW is idle into dsb commit.

v3: add intel_dsb as a parameter to intel_psr_wait_for_idle_dsb
v2: add pass crtc_state->dsb_commit as parameter

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250905072708.2659411-4-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 39 +++++++++++++++++++++---
 drivers/gpu/drm/i915/display/intel_psr.h |  2 ++
 2 files changed, 37 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index eae1eedbbb26..1aa4a127afaf 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -42,6 +42,7 @@
 #include "intel_dmc.h"
 #include "intel_dp.h"
 #include "intel_dp_aux.h"
+#include "intel_dsb.h"
 #include "intel_frontbuffer.h"
 #include "intel_hdmi.h"
 #include "intel_psr.h"
@@ -3006,7 +3007,8 @@ void intel_psr_post_plane_update(struct intel_atomic_state *state,
 #define PSR_IDLE_TIMEOUT_MS 50
 
 static int
-_psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state)
+_psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state,
+				   struct intel_dsb *dsb)
 {
 	struct intel_display *display = to_intel_display(new_crtc_state);
 	enum transcoder cpu_transcoder = new_crtc_state->cpu_transcoder;
@@ -3016,6 +3018,13 @@ _psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state
 	 * As all higher states has bit 4 of PSR2 state set we can just wait for
 	 * EDP_PSR2_STATUS_STATE_DEEP_SLEEP to be cleared.
 	 */
+	if (dsb) {
+		intel_dsb_poll(dsb, EDP_PSR2_STATUS(display, cpu_transcoder),
+			       EDP_PSR2_STATUS_STATE_DEEP_SLEEP, 0, 200,
+			       PSR_IDLE_TIMEOUT_MS * 1000 / 200);
+		return true;
+	}
+
 	return intel_de_wait_for_clear(display,
 				       EDP_PSR2_STATUS(display, cpu_transcoder),
 				       EDP_PSR2_STATUS_STATE_DEEP_SLEEP,
@@ -3023,11 +3032,19 @@ _psr2_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state
 }
 
 static int
-_psr1_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state)
+_psr1_ready_for_pipe_update_locked(const struct intel_crtc_state *new_crtc_state,
+				   struct intel_dsb *dsb)
 {
 	struct intel_display *display = to_intel_display(new_crtc_state);
 	enum transcoder cpu_transcoder = new_crtc_state->cpu_transcoder;
 
+	if (dsb) {
+		intel_dsb_poll(dsb, psr_status_reg(display, cpu_transcoder),
+			       EDP_PSR_STATUS_STATE_MASK, 0, 200,
+			       PSR_IDLE_TIMEOUT_MS * 1000 / 200);
+		return true;
+	}
+
 	return intel_de_wait_for_clear(display,
 				       psr_status_reg(display, cpu_transcoder),
 				       EDP_PSR_STATUS_STATE_MASK,
@@ -3060,9 +3077,11 @@ void intel_psr_wait_for_idle_locked(const struct intel_crtc_state *new_crtc_stat
 			continue;
 
 		if (intel_dp->psr.sel_update_enabled)
-			ret = _psr2_ready_for_pipe_update_locked(new_crtc_state);
+			ret = _psr2_ready_for_pipe_update_locked(new_crtc_state,
+								 NULL);
 		else
-			ret = _psr1_ready_for_pipe_update_locked(new_crtc_state);
+			ret = _psr1_ready_for_pipe_update_locked(new_crtc_state,
+								 NULL);
 
 		if (ret)
 			drm_err(display->drm,
@@ -3070,6 +3089,18 @@ void intel_psr_wait_for_idle_locked(const struct intel_crtc_state *new_crtc_stat
 	}
 }
 
+void intel_psr_wait_for_idle_dsb(struct intel_dsb *dsb,
+				 const struct intel_crtc_state *new_crtc_state)
+{
+	if (!new_crtc_state->has_psr || new_crtc_state->has_panel_replay)
+		return;
+
+	if (new_crtc_state->has_sel_update)
+		_psr2_ready_for_pipe_update_locked(new_crtc_state, dsb);
+	else
+		_psr1_ready_for_pipe_update_locked(new_crtc_state, dsb);
+}
+
 static bool __psr_wait_for_idle_locked(struct intel_dp *intel_dp)
 {
 	struct intel_display *display = to_intel_display(intel_dp);
diff --git a/drivers/gpu/drm/i915/display/intel_psr.h b/drivers/gpu/drm/i915/display/intel_psr.h
index 9b061a22361f..077751aa599f 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.h
+++ b/drivers/gpu/drm/i915/display/intel_psr.h
@@ -52,6 +52,8 @@ void intel_psr_get_config(struct intel_encoder *encoder,
 void intel_psr_irq_handler(struct intel_dp *intel_dp, u32 psr_iir);
 void intel_psr_short_pulse(struct intel_dp *intel_dp);
 void intel_psr_wait_for_idle_locked(const struct intel_crtc_state *new_crtc_state);
+void intel_psr_wait_for_idle_dsb(struct intel_dsb *dsb,
+				 const struct intel_crtc_state *new_crtc_state);
 bool intel_psr_enabled(struct intel_dp *intel_dp);
 int intel_psr2_sel_fetch_update(struct intel_atomic_state *state,
 				struct intel_crtc *crtc);

From 8bc70f2ea932976604876756d6d868eca3ac4cbb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Fri, 5 Sep 2025 10:27:08 +0300
Subject: [PATCH 1069/1455] drm/i915/psr: Add poll for checking PSR is idle
 before starting update
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We are currently observing crc failures after we started using dsb for PSR
updates as well. This seems to happen because PSR HW is still sending
couple of updates using old framebuffers on wake-up.

Fix this by adding poll ensuring PSR is idle before starting update.

v2: pass new_crtc_state->dsb_commit to intel_psr_wait_for_idle_dsb

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250905072708.2659411-5-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_display.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index c1a3a95c65f0..5dca7f96b425 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -7271,6 +7271,9 @@ static void intel_atomic_dsb_finish(struct intel_atomic_state *state,
 		intel_psr_trigger_frame_change_event(new_crtc_state->dsb_commit,
 						     state, crtc);
 
+		intel_psr_wait_for_idle_dsb(new_crtc_state->dsb_commit,
+					    new_crtc_state);
+
 		if (new_crtc_state->use_dsb)
 			intel_dsb_vblank_evade(state, new_crtc_state->dsb_commit);
 

From b5ddb9ab3d83fe0d93d38de442c45610a5846d06 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert+renesas@glider.be>
Date: Thu, 4 Sep 2025 17:31:00 +0200
Subject: [PATCH 1070/1455] drm: rcar-du: lvds: Convert to RUNTIME_PM_OPS()

Convert the Renesas R-Car Display Unit LVDS driver from
SET_RUNTIME_PM_OPS() to RUNTIME_PM_OPS(), and pm_ptr().  This reduces
kernel size in case CONFIG_PM is disabled.  While DRM_RCAR_LVDS depends
on PM, the code may still serve as an example for new drivers.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Link: https://lore.kernel.org/r/2264ff4f21a7e17384822e0efba176cf78ae184d.1756999823.git.geert+renesas@glider.be
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
index af58b814e588..001b3543924a 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_lvds.c
@@ -1013,7 +1013,7 @@ err_reset_assert:
 }
 
 static const struct dev_pm_ops rcar_lvds_pm_ops = {
-	SET_RUNTIME_PM_OPS(rcar_lvds_runtime_suspend, rcar_lvds_runtime_resume, NULL)
+	RUNTIME_PM_OPS(rcar_lvds_runtime_suspend, rcar_lvds_runtime_resume, NULL)
 };
 
 static struct platform_driver rcar_lvds_platform_driver = {
@@ -1021,7 +1021,7 @@ static struct platform_driver rcar_lvds_platform_driver = {
 	.remove		= rcar_lvds_remove,
 	.driver		= {
 		.name	= "rcar-lvds",
-		.pm	= &rcar_lvds_pm_ops,
+		.pm	= pm_ptr(&rcar_lvds_pm_ops),
 		.of_match_table = rcar_lvds_of_table,
 	},
 };

From 685e8dae19df73d5400734ee5ad9e96470f9c0b4 Mon Sep 17 00:00:00 2001
From: Marek Vasut <marek.vasut+renesas@mailbox.org>
Date: Sun, 31 Aug 2025 21:04:25 +0200
Subject: [PATCH 1071/1455] drm/rcar-du: dsi: Implement DSI command support

Implement support for DSI command transfer. Transmission of both Short
Packet and Long Packet is implemented, so is command transmission to
request response from peripheral device and transmission of non-read
command with BTA.

The AXI memory access mode is currently not implemented, each transfer
is performed purely using controller register interface. Short Packet
transfer can transfer up to 2 Bytes of data, Long Packet transfer can
transfer up to 16 Bytes of data.

Signed-off-by: Marek Vasut <marek.vasut+renesas@mailbox.org>
Link: https://lore.kernel.org/r/20250831190507.327848-1-marek.vasut+renesas@mailbox.org
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
---
 .../gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c   | 225 ++++++++++++++++++
 .../drm/renesas/rcar-du/rcar_mipi_dsi_regs.h  | 125 ++++++++++
 2 files changed, 350 insertions(+)

diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
index 952c3efb74da..5c73a513f678 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi.c
@@ -937,9 +937,234 @@ static int rcar_mipi_dsi_host_detach(struct mipi_dsi_host *host,
 	return 0;
 }
 
+static ssize_t rcar_mipi_dsi_host_tx_transfer(struct mipi_dsi_host *host,
+					      const struct mipi_dsi_msg *msg,
+					      bool is_rx_xfer)
+{
+	const bool is_tx_long = mipi_dsi_packet_format_is_long(msg->type);
+	struct rcar_mipi_dsi *dsi = host_to_rcar_mipi_dsi(host);
+	struct mipi_dsi_packet packet;
+	u8 payload[16] = { 0 };
+	u32 status;
+	int ret;
+
+	ret = mipi_dsi_create_packet(&packet, msg);
+	if (ret)
+		return ret;
+
+	/* Configure LP or HS command transfer. */
+	rcar_mipi_dsi_write(dsi, TXCMSETR, (msg->flags & MIPI_DSI_MSG_USE_LPM) ?
+					   TXCMSETR_SPDTYP : 0);
+
+	/* Register access mode for RX transfer. */
+	if (is_rx_xfer)
+		rcar_mipi_dsi_write(dsi, RXPSETR, 0);
+
+	/* Do not use IRQ, poll for completion, the completion is quick. */
+	rcar_mipi_dsi_write(dsi, TXCMIER, 0);
+
+	/*
+	 * Send the header:
+	 * header[0] = Virtual Channel + Data Type
+	 * header[1] = Word Count LSB (LP) or first param (SP)
+	 * header[2] = Word Count MSB (LP) or second param (SP)
+	 */
+	rcar_mipi_dsi_write(dsi, TXCMPHDR,
+			    (is_tx_long ? TXCMPHDR_FMT : 0) |
+			    TXCMPHDR_VC(msg->channel) |
+			    TXCMPHDR_DT(msg->type) |
+			    TXCMPHDR_DATA1(packet.header[2]) |
+			    TXCMPHDR_DATA0(packet.header[1]));
+
+	if (is_tx_long) {
+		memcpy(payload, packet.payload,
+		       min(msg->tx_len, sizeof(payload)));
+
+		rcar_mipi_dsi_write(dsi, TXCMPPD0R,
+				    (payload[3] << 24) | (payload[2] << 16) |
+				    (payload[1] << 8) | payload[0]);
+		rcar_mipi_dsi_write(dsi, TXCMPPD1R,
+				    (payload[7] << 24) | (payload[6] << 16) |
+				    (payload[5] << 8) | payload[4]);
+		rcar_mipi_dsi_write(dsi, TXCMPPD2R,
+				    (payload[11] << 24) | (payload[10] << 16) |
+				    (payload[9] << 8) | payload[8]);
+		rcar_mipi_dsi_write(dsi, TXCMPPD3R,
+				    (payload[15] << 24) | (payload[14] << 16) |
+				    (payload[13] << 8) | payload[12]);
+	}
+
+	/* Start the transfer, RX with BTA, TX without BTA. */
+	if (is_rx_xfer) {
+		rcar_mipi_dsi_write(dsi, TXCMCR, TXCMCR_BTAREQ);
+
+		/* Wait until the transmission, BTA, reception completed. */
+		ret = read_poll_timeout(rcar_mipi_dsi_read, status,
+					(status & RXPSR_BTAREQEND),
+					2000, 50000, false, dsi, RXPSR);
+	} else {
+		rcar_mipi_dsi_write(dsi, TXCMCR, TXCMCR_TXREQ);
+
+		/* Wait until the transmission completed. */
+		ret = read_poll_timeout(rcar_mipi_dsi_read, status,
+					(status & TXCMSR_TXREQEND),
+					2000, 50000, false, dsi, TXCMSR);
+	}
+
+	if (ret < 0) {
+		dev_err(dsi->dev, "Command transfer timeout (0x%08x)\n",
+			status);
+		return ret;
+	}
+
+	return packet.size;
+}
+
+static ssize_t rcar_mipi_dsi_host_rx_transfer(struct mipi_dsi_host *host,
+					      const struct mipi_dsi_msg *msg)
+{
+	struct rcar_mipi_dsi *dsi = host_to_rcar_mipi_dsi(host);
+	u8 *rx_buf = (u8 *)(msg->rx_buf);
+	u32 reg, data, status, wc;
+	int i, ret;
+
+	/* RX transfer received data validation and parsing starts here. */
+	reg = rcar_mipi_dsi_read(dsi, TOSR);
+	if (reg & TOSR_TATO) {	/* Turn-Around TimeOut. */
+		/* Clear TATO Turn-Around TimeOut bit. */
+		rcar_mipi_dsi_write(dsi, TOSR, TOSR_TATO);
+		return -ETIMEDOUT;
+	}
+
+	reg = rcar_mipi_dsi_read(dsi, RXPSR);
+
+	if (msg->flags & MIPI_DSI_MSG_REQ_ACK) {
+		/* Transfer with zero-length RX. */
+		if (!(reg & RXPSR_RCVACK)) {
+			/* No ACK on RX response received. */
+			return -EINVAL;
+		}
+	} else {
+		/* Transfer with non-zero-length RX. */
+		if (!(reg & RXPSR_RCVRESP)) {
+			/* No packet header of RX response received. */
+			return -EINVAL;
+		}
+
+		if (reg & (RXPSR_CRCERR | RXPSR_WCERR | RXPSR_AXIERR | RXPSR_OVRERR)) {
+			/* Incorrect response payload. */
+			return -ENODATA;
+		}
+
+		data = rcar_mipi_dsi_read(dsi, RXPHDR);
+		if (data & RXPHDR_FMT) {	/* Long Packet Response. */
+			/* Read Long Packet Response length from packet header. */
+			wc = data & 0xffff;
+			if (wc > msg->rx_len) {
+				dev_warn(dsi->dev,
+					 "Long Packet Response longer than RX buffer (%d), limited to %zu Bytes\n",
+					 wc, msg->rx_len);
+				wc = msg->rx_len;
+			}
+
+			if (wc > 16) {
+				dev_warn(dsi->dev,
+					 "Long Packet Response too long (%d), limited to 16 Bytes\n",
+					 wc);
+				wc = 16;
+			}
+
+			for (i = 0; i < msg->rx_len; i++) {
+				if (!(i % 4))
+					data = rcar_mipi_dsi_read(dsi, RXPPD0R + i);
+
+				rx_buf[i] = data & 0xff;
+				data >>= 8;
+			}
+		} else {	/* Short Packet Response. */
+			if (msg->rx_len >= 1)
+				rx_buf[0] = data & 0xff;
+			if (msg->rx_len >= 2)
+				rx_buf[1] = (data >> 8) & 0xff;
+			if (msg->rx_len >= 3) {
+				dev_warn(dsi->dev,
+					 "Expected Short Packet Response too long (%zu), limited to 2 Bytes\n",
+					 msg->rx_len);
+			}
+		}
+	}
+
+	if (reg & RXPSR_RCVAKE) {
+		/* Acknowledge and Error report received. */
+		return -EFAULT;
+	}
+
+	/* Wait until the bus handover to host processor completed. */
+	ret = read_poll_timeout(rcar_mipi_dsi_read, status,
+				!(status & PPIDL0SR_DIR),
+				2000, 50000, false, dsi, PPIDL0SR);
+	if (ret < 0) {
+		dev_err(dsi->dev, "Command RX DIR timeout (0x%08x)\n", status);
+		return ret;
+	}
+
+	/* Wait until the data lane is in LP11 stop state. */
+	ret = read_poll_timeout(rcar_mipi_dsi_read, status,
+				status & PPIDL0SR_STPST,
+				2000, 50000, false, dsi, PPIDL0SR);
+	if (ret < 0) {
+		dev_err(dsi->dev, "Command RX STPST timeout (0x%08x)\n", status);
+		return ret;
+	}
+
+	return 0;
+}
+
+static ssize_t rcar_mipi_dsi_host_transfer(struct mipi_dsi_host *host,
+					   const struct mipi_dsi_msg *msg)
+{
+	const bool is_rx_xfer = (msg->flags & MIPI_DSI_MSG_REQ_ACK) || msg->rx_len;
+	struct rcar_mipi_dsi *dsi = host_to_rcar_mipi_dsi(host);
+	int ret;
+
+	if (msg->tx_len > 16 || msg->rx_len > 16) {
+		/* ToDo: Implement Memory on AXI bus command mode. */
+		dev_warn(dsi->dev,
+			 "Register-based command mode supports only up to 16 Bytes long payload\n");
+		return -EOPNOTSUPP;
+	}
+
+	ret = rcar_mipi_dsi_host_tx_transfer(host, msg, is_rx_xfer);
+
+	/* If TX transfer succeeded and this transfer has RX part. */
+	if (ret >= 0 && is_rx_xfer) {
+		ret = rcar_mipi_dsi_host_rx_transfer(host, msg);
+		if (ret)
+			return ret;
+
+		ret = msg->rx_len;
+	}
+
+	/*
+	 * Wait a bit between commands, otherwise panels based on ILI9881C
+	 * TCON may fail to correctly receive all commands sent to them.
+	 * Until we can actually test with another DSI device, keep the
+	 * delay here, but eventually this delay might have to be moved
+	 * into the ILI9881C panel driver.
+	 */
+	usleep_range(1000, 2000);
+
+	/* Clear the completion interrupt. */
+	if (!msg->rx_len)
+		rcar_mipi_dsi_write(dsi, TXCMSR, TXCMSR_TXREQEND);
+
+	return ret;
+}
+
 static const struct mipi_dsi_host_ops rcar_mipi_dsi_host_ops = {
 	.attach = rcar_mipi_dsi_host_attach,
 	.detach = rcar_mipi_dsi_host_detach,
+	.transfer = rcar_mipi_dsi_host_transfer
 };
 
 /* -----------------------------------------------------------------------------
diff --git a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
index a54c7eb4113b..76521276e2af 100644
--- a/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
+++ b/drivers/gpu/drm/renesas/rcar-du/rcar_mipi_dsi_regs.h
@@ -15,6 +15,127 @@
 #define TXSETR				0x100
 #define TXSETR_LANECNT_MASK		(0x3 << 0)
 
+/*
+ * DSI Command Transfer Registers
+ */
+#define TXCMSETR			0x110
+#define TXCMSETR_SPDTYP			(1 << 8)	/* 0:HS 1:LP */
+#define TXCMSETR_LPPDACC		(1 << 0)
+#define TXCMCR				0x120
+#define TXCMCR_BTATYP			(1 << 2)
+#define TXCMCR_BTAREQ			(1 << 1)
+#define TXCMCR_TXREQ			(1 << 0)
+#define TXCMSR				0x130
+#define TXCMSR_CLSNERR			(1 << 18)
+#define TXCMSR_AXIERR			(1 << 16)
+#define TXCMSR_TXREQEND			(1 << 0)
+#define TXCMSCR				0x134
+#define TXCMSCR_CLSNERR			(1 << 18)
+#define TXCMSCR_AXIERR			(1 << 16)
+#define TXCMSCR_TXREQEND		(1 << 0)
+#define TXCMIER				0x138
+#define TXCMIER_CLSNERR			(1 << 18)
+#define TXCMIER_AXIERR			(1 << 16)
+#define TXCMIER_TXREQEND		(1 << 0)
+#define TXCMADDRSET0R			0x140
+#define TXCMPHDR			0x150
+#define TXCMPHDR_FMT			(1 << 24)	/* 0:SP 1:LP */
+#define TXCMPHDR_VC(n)			(((n) & 0x3) << 22)
+#define TXCMPHDR_DT(n)			(((n) & 0x3f) << 16)
+#define TXCMPHDR_DATA1(n)		(((n) & 0xff) << 8)
+#define TXCMPHDR_DATA0(n)		(((n) & 0xff) << 0)
+#define TXCMPPD0R			0x160
+#define TXCMPPD1R			0x164
+#define TXCMPPD2R			0x168
+#define TXCMPPD3R			0x16c
+
+#define RXSETR				0x200
+#define RXSETR_CRCEN			(((n) & 0xf) << 24)
+#define RXSETR_ECCEN			(((n) & 0xf) << 16)
+#define RXPSETR				0x210
+#define RXPSETR_LPPDACC			(1 << 0)
+#define RXPSR				0x220
+#define RXPSR_ECCERR1B			(1 << 28)
+#define RXPSR_UEXTRGERR			(1 << 25)
+#define RXPSR_RESPTOERR			(1 << 24)
+#define RXPSR_OVRERR			(1 << 23)
+#define RXPSR_AXIERR			(1 << 22)
+#define RXPSR_CRCERR			(1 << 21)
+#define RXPSR_WCERR			(1 << 20)
+#define RXPSR_UEXDTERR			(1 << 19)
+#define RXPSR_UEXPKTERR			(1 << 18)
+#define RXPSR_ECCERR			(1 << 17)
+#define RXPSR_MLFERR			(1 << 16)
+#define RXPSR_RCVACK			(1 << 14)
+#define RXPSR_RCVEOT			(1 << 10)
+#define RXPSR_RCVAKE			(1 << 9)
+#define RXPSR_RCVRESP			(1 << 8)
+#define RXPSR_BTAREQEND			(1 << 0)
+#define RXPSCR				0x224
+#define RXPSCR_ECCERR1B			(1 << 28)
+#define RXPSCR_UEXTRGERR		(1 << 25)
+#define RXPSCR_RESPTOERR		(1 << 24)
+#define RXPSCR_OVRERR			(1 << 23)
+#define RXPSCR_AXIERR			(1 << 22)
+#define RXPSCR_CRCERR			(1 << 21)
+#define RXPSCR_WCERR			(1 << 20)
+#define RXPSCR_UEXDTERR			(1 << 19)
+#define RXPSCR_UEXPKTERR		(1 << 18)
+#define RXPSCR_ECCERR			(1 << 17)
+#define RXPSCR_MLFERR			(1 << 16)
+#define RXPSCR_RCVACK			(1 << 14)
+#define RXPSCR_RCVEOT			(1 << 10)
+#define RXPSCR_RCVAKE			(1 << 9)
+#define RXPSCR_RCVRESP			(1 << 8)
+#define RXPSCR_BTAREQEND		(1 << 0)
+#define RXPIER				0x228
+#define RXPIER_ECCERR1B			(1 << 28)
+#define RXPIER_UEXTRGERR		(1 << 25)
+#define RXPIER_RESPTOERR		(1 << 24)
+#define RXPIER_OVRERR			(1 << 23)
+#define RXPIER_AXIERR			(1 << 22)
+#define RXPIER_CRCERR			(1 << 21)
+#define RXPIER_WCERR			(1 << 20)
+#define RXPIER_UEXDTERR			(1 << 19)
+#define RXPIER_UEXPKTERR		(1 << 18)
+#define RXPIER_ECCERR			(1 << 17)
+#define RXPIER_MLFERR			(1 << 16)
+#define RXPIER_RCVACK			(1 << 14)
+#define RXPIER_RCVEOT			(1 << 10)
+#define RXPIER_RCVAKE			(1 << 9)
+#define RXPIER_RCVRESP			(1 << 8)
+#define RXPIER_BTAREQEND		(1 << 0)
+#define RXPADDRSET0R			0x230
+#define RXPSIZESETR			0x238
+#define RXPSIZESETR_SIZE(n)		(((n) & 0xf) << 3)
+#define RXPHDR				0x240
+#define RXPHDR_FMT			(1 << 24)	/* 0:SP 1:LP */
+#define RXPHDR_VC(n)			(((n) & 0x3) << 22)
+#define RXPHDR_DT(n)			(((n) & 0x3f) << 16)
+#define RXPHDR_DATA1(n)			(((n) & 0xff) << 8)
+#define RXPHDR_DATA0(n)			(((n) & 0xff) << 0)
+#define RXPPD0R				0x250
+#define RXPPD1R				0x254
+#define RXPPD2R				0x258
+#define RXPPD3R				0x25c
+#define AKEPR				0x300
+#define AKEPR_VC(n)			(((n) & 0x3) << 22)
+#define AKEPR_DT(n)			(((n) & 0x3f) << 16)
+#define AKEPR_ERRRPT(n)			(((n) & 0xffff) << 0)
+#define RXRESPTOSETR			0x400
+#define TACR				0x500
+#define TASR				0x510
+#define TASCR				0x514
+#define TAIER				0x518
+#define TOSR				0x610
+#define TOSR_TATO			(1 << 2)
+#define TOSR_LRXHTO			(1 << 1)
+#define TOSR_HRXTO			(1 << 0)
+#define TOSCR				0x614
+#define TOSCR_TATO			(1 << 2)
+#define TOSCR_LRXHTO			(1 << 1)
+#define TOSCR_HRXTO			(1 << 0)
+
 /*
  * Video Mode Register
  */
@@ -100,6 +221,10 @@
 #define PPICLSCR_HSTOLP			(1 << 27)
 #define PPICLSCR_TOHS			(1 << 26)
 
+#define PPIDL0SR			0x740
+#define PPIDL0SR_DIR			(1 << 10)
+#define PPIDL0SR_STPST			(1 << 6)
+
 #define PPIDLSR				0x760
 #define PPIDLSR_STPST			(0xf << 0)
 

From 90d8aad11f613e13e1eddc35c0225cce9e54ba94 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Jouni=20H=C3=B6gander?= <jouni.hogander@intel.com>
Date: Wed, 27 Aug 2025 09:08:09 +0300
Subject: [PATCH 1072/1455] drm/i915/psr: Panel Replay SU cap dpcd read return
 value
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently return value of drm_dpcd_readb is not checked when reading sink
Panel Replay Selective Update capabilities.

Fix this and switch to drm_dpcd_read_byte.

Signed-off-by: Jouni Högander <jouni.hogander@intel.com>
Reviewed-by: Mika Kahola <mika.kahola@intel.com>
Reviewed-by: Michał Grzelak <michal.grzelak@intel.com>
Link: https://lore.kernel.org/r/20250827060809.2461725-1-jouni.hogander@intel.com
---
 drivers/gpu/drm/i915/display/intel_psr.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c
index 1aa4a127afaf..01bf304c705f 100644
--- a/drivers/gpu/drm/i915/display/intel_psr.c
+++ b/drivers/gpu/drm/i915/display/intel_psr.c
@@ -495,12 +495,14 @@ static u8 intel_dp_get_su_capability(struct intel_dp *intel_dp)
 {
 	u8 su_capability = 0;
 
-	if (intel_dp->psr.sink_panel_replay_su_support)
-		drm_dp_dpcd_readb(&intel_dp->aux,
-				  DP_PANEL_REPLAY_CAP_CAPABILITY,
-				  &su_capability);
-	else
+	if (intel_dp->psr.sink_panel_replay_su_support) {
+		if (drm_dp_dpcd_read_byte(&intel_dp->aux,
+					  DP_PANEL_REPLAY_CAP_CAPABILITY,
+					  &su_capability) < 0)
+			return 0;
+	} else {
 		su_capability = intel_dp->psr_dpcd[1];
+	}
 
 	return su_capability;
 }

From 7ea3baa6efe4bb93d11e1c0e6528b1468d7debf6 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 5 Sep 2025 13:41:49 +0300
Subject: [PATCH 1073/1455] drm/i915/power: fix size for for_each_set_bit() in
 abox iteration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

for_each_set_bit() expects size to be in bits, not bytes. The abox mask
iteration uses bytes, but it works by coincidence, because the local
variable holding the mask is unsigned long, and the mask only ever has
bit 2 as the highest bit. Using a smaller type could lead to subtle and
very hard to track bugs.

Fixes: 62afef2811e4 ("drm/i915/rkl: RKL uses ABOX0 for pixel transfers")
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: stable@vger.kernel.org # v5.9+
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20250905104149.1144751-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_display_power.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
index 7f9a2d743bac..da4babfd6bcb 100644
--- a/drivers/gpu/drm/i915/display/intel_display_power.c
+++ b/drivers/gpu/drm/i915/display/intel_display_power.c
@@ -1174,7 +1174,7 @@ static void icl_mbus_init(struct intel_display *display)
 	if (DISPLAY_VER(display) == 12)
 		abox_regs |= BIT(0);
 
-	for_each_set_bit(i, &abox_regs, sizeof(abox_regs))
+	for_each_set_bit(i, &abox_regs, BITS_PER_TYPE(abox_regs))
 		intel_de_rmw(display, MBUS_ABOX_CTL(i), mask, val);
 }
 
@@ -1639,11 +1639,11 @@ static void tgl_bw_buddy_init(struct intel_display *display)
 	if (table[config].page_mask == 0) {
 		drm_dbg_kms(display->drm,
 			    "Unknown memory configuration; disabling address buddy logic.\n");
-		for_each_set_bit(i, &abox_mask, sizeof(abox_mask))
+		for_each_set_bit(i, &abox_mask, BITS_PER_TYPE(abox_mask))
 			intel_de_write(display, BW_BUDDY_CTL(i),
 				       BW_BUDDY_DISABLE);
 	} else {
-		for_each_set_bit(i, &abox_mask, sizeof(abox_mask)) {
+		for_each_set_bit(i, &abox_mask, BITS_PER_TYPE(abox_mask)) {
 			intel_de_write(display, BW_BUDDY_PAGE_MASK(i),
 				       table[config].page_mask);
 

From 55c771b0d8a70a10d549d5296cf5e3342bef548f Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 3 Sep 2025 13:10:50 +0300
Subject: [PATCH 1074/1455] drm/i915/ddi: abstract figuring out encoder name

The encoder name deduction has become a bit cumbersome within
intel_ddi_init(). Split it out to a separate function to declutter
intel_ddi_init(), even if that means having to use a temp seq buffer for
the name.

Cc: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Reviewed-by: Ankit Nautiyal <ankit.k.nautiyal@intel.com>
Link: https://lore.kernel.org/r/20250903101050.3671305-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_ddi.c | 69 +++++++++++++-----------
 1 file changed, 38 insertions(+), 31 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
index 46017091bb0b..08083ac83a74 100644
--- a/drivers/gpu/drm/i915/display/intel_ddi.c
+++ b/drivers/gpu/drm/i915/display/intel_ddi.c
@@ -26,6 +26,7 @@
  */
 
 #include <linux/iopoll.h>
+#include <linux/seq_buf.h>
 #include <linux/string_helpers.h>
 
 #include <drm/display/drm_dp_helper.h>
@@ -5067,11 +5068,45 @@ static bool port_in_use(struct intel_display *display, enum port port)
 	return false;
 }
 
+static const char *intel_ddi_encoder_name(struct intel_display *display,
+					  enum port port, enum phy phy,
+					  struct seq_buf *s)
+{
+	if (DISPLAY_VER(display) >= 13 && port >= PORT_D_XELPD) {
+		seq_buf_printf(s, "DDI %c/PHY %c",
+			       port_name(port - PORT_D_XELPD + PORT_D),
+			       phy_name(phy));
+	} else if (DISPLAY_VER(display) >= 12) {
+		enum tc_port tc_port = intel_port_to_tc(display, port);
+
+		seq_buf_printf(s, "DDI %s%c/PHY %s%c",
+			       port >= PORT_TC1 ? "TC" : "",
+			       port >= PORT_TC1 ? port_tc_name(port) : port_name(port),
+			       tc_port != TC_PORT_NONE ? "TC" : "",
+			       tc_port != TC_PORT_NONE ? tc_port_name(tc_port) : phy_name(phy));
+	} else if (DISPLAY_VER(display) >= 11) {
+		enum tc_port tc_port = intel_port_to_tc(display, port);
+
+		seq_buf_printf(s, "DDI %c%s/PHY %s%c",
+			       port_name(port),
+			       port >= PORT_C ? " (TC)" : "",
+			       tc_port != TC_PORT_NONE ? "TC" : "",
+			       tc_port != TC_PORT_NONE ? tc_port_name(tc_port) : phy_name(phy));
+	} else {
+		seq_buf_printf(s, "DDI %c/PHY %c", port_name(port),  phy_name(phy));
+	}
+
+	drm_WARN_ON(display->drm, seq_buf_has_overflowed(s));
+
+	return seq_buf_str(s);
+}
+
 void intel_ddi_init(struct intel_display *display,
 		    const struct intel_bios_encoder_data *devdata)
 {
 	struct intel_digital_port *dig_port;
 	struct intel_encoder *encoder;
+	DECLARE_SEQ_BUF(encoder_name, 20);
 	bool init_hdmi, init_dp;
 	enum port port;
 	enum phy phy;
@@ -5156,37 +5191,9 @@ void intel_ddi_init(struct intel_display *display,
 	encoder = &dig_port->base;
 	encoder->devdata = devdata;
 
-	if (DISPLAY_VER(display) >= 13 && port >= PORT_D_XELPD) {
-		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
-				 DRM_MODE_ENCODER_TMDS,
-				 "DDI %c/PHY %c",
-				 port_name(port - PORT_D_XELPD + PORT_D),
-				 phy_name(phy));
-	} else if (DISPLAY_VER(display) >= 12) {
-		enum tc_port tc_port = intel_port_to_tc(display, port);
-
-		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
-				 DRM_MODE_ENCODER_TMDS,
-				 "DDI %s%c/PHY %s%c",
-				 port >= PORT_TC1 ? "TC" : "",
-				 port >= PORT_TC1 ? port_tc_name(port) : port_name(port),
-				 tc_port != TC_PORT_NONE ? "TC" : "",
-				 tc_port != TC_PORT_NONE ? tc_port_name(tc_port) : phy_name(phy));
-	} else if (DISPLAY_VER(display) >= 11) {
-		enum tc_port tc_port = intel_port_to_tc(display, port);
-
-		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
-				 DRM_MODE_ENCODER_TMDS,
-				 "DDI %c%s/PHY %s%c",
-				 port_name(port),
-				 port >= PORT_C ? " (TC)" : "",
-				 tc_port != TC_PORT_NONE ? "TC" : "",
-				 tc_port != TC_PORT_NONE ? tc_port_name(tc_port) : phy_name(phy));
-	} else {
-		drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
-				 DRM_MODE_ENCODER_TMDS,
-				 "DDI %c/PHY %c", port_name(port), phy_name(phy));
-	}
+	drm_encoder_init(display->drm, &encoder->base, &intel_ddi_funcs,
+			 DRM_MODE_ENCODER_TMDS, "%s",
+			 intel_ddi_encoder_name(display, port, phy, &encoder_name));
 
 	intel_encoder_link_check_init(encoder, intel_ddi_link_check);
 

From 30c3ffb880315b907ccd682ed4e17d20f786fdf9 Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Thu, 4 Sep 2025 19:26:40 +0800
Subject: [PATCH 1075/1455] drm/i915/gvt: Remove redundant ternary operators

For ternary operators in the form of "a ? false : true", if 'a' itself
returns a boolean result, the ternary operator can be omitted. Remove
redundant ternary operators to clean up the code.

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Link: https://lore.kernel.org/r/20250904112644.350512-1-liaoyuanhong@vivo.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/gvt/cmd_parser.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c
index a91e23c22ea1..d432fdd69833 100644
--- a/drivers/gpu/drm/i915/gvt/cmd_parser.c
+++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c
@@ -1921,7 +1921,7 @@ static int perform_bb_shadow(struct parser_exec_state *s)
 	if (!bb)
 		return -ENOMEM;
 
-	bb->ppgtt = (s->buf_addr_type == GTT_BUFFER) ? false : true;
+	bb->ppgtt = s->buf_addr_type != GTT_BUFFER;
 
 	/*
 	 * The start_offset stores the batch buffer's start gma's

From 45b95459802afe9b7d7add22db98176749c88d6c Mon Sep 17 00:00:00 2001
From: Uma Shankar <uma.shankar@intel.com>
Date: Thu, 4 Sep 2025 15:23:38 +0530
Subject: [PATCH 1076/1455] drm/i915/display: Remove FBC modulo 4 restriction
 for ADL-P+

WA:22010751166 does not apply past display version 12.  Or, in
other words, the FBC restriction where FBC is disabled for
non-modulo 4 plane sizes (including plane size + yoffset) is fixed
from display version 13 and onwards. Relax the restriction for the same.

v4: Dropped redundant commit message

v3: Update comments for clarity (Jonathan Cavitt)

v2: Update the macro for display version check (Vinod)

Suggested-by: Vidya Srinivas <vidya.srinivas@intel.com>
Signed-off-by: Uma Shankar <uma.shankar@intel.com>
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250904095338.300813-2-uma.shankar@intel.com
---
 drivers/gpu/drm/i915/display/intel_fbc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index d4c5deff9cbe..9e097ed80bd1 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -1550,14 +1550,14 @@ static int intel_fbc_check_plane(struct intel_atomic_state *state,
 	 * having a Y offset that isn't divisible by 4 causes FIFO underrun
 	 * and screen flicker.
 	 */
-	if (DISPLAY_VER(display) >= 9 &&
+	if (IS_DISPLAY_VER(display, 9, 12) &&
 	    plane_state->view.color_plane[0].y & 3) {
 		plane_state->no_fbc_reason = "plane start Y offset misaligned";
 		return 0;
 	}
 
 	/* Wa_22010751166: icl, ehl, tgl, dg1, rkl */
-	if (DISPLAY_VER(display) >= 11 &&
+	if (IS_DISPLAY_VER(display, 9, 12) &&
 	    (plane_state->view.color_plane[0].y +
 	     (drm_rect_height(&plane_state->uapi.src) >> 16)) & 3) {
 		plane_state->no_fbc_reason = "plane end Y offset misaligned";

From cef2c7266feaf15676fb5e47cbb6d7d832b61563 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:47 +0300
Subject: [PATCH 1077/1455] drm/i915/fb: add intel_framebuffer_alloc()

Add intel_framebuffer_alloc() to hide intel_bo_alloc_framebuffer(), as
that doesn't feel like the correct abstraction.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/379c306c692c50f6af3b6f2488c213f12627954f.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/i9xx_plane.c          | 3 +--
 drivers/gpu/drm/i915/display/intel_fb.c            | 7 ++++++-
 drivers/gpu/drm/i915/display/intel_fb.h            | 3 +++
 drivers/gpu/drm/i915/display/skl_universal_plane.c | 2 +-
 4 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index 3eb96d8abba8..9a39e9efbfaf 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -15,7 +15,6 @@
 #include "i9xx_plane.h"
 #include "i9xx_plane_regs.h"
 #include "intel_atomic.h"
-#include "intel_bo.h"
 #include "intel_de.h"
 #include "intel_display_irq.h"
 #include "intel_display_regs.h"
@@ -1178,7 +1177,7 @@ i9xx_get_initial_plane_config(struct intel_crtc *crtc,
 
 	drm_WARN_ON(display->drm, pipe != crtc->pipe);
 
-	intel_fb = intel_bo_alloc_framebuffer();
+	intel_fb = intel_framebuffer_alloc();
 	if (!intel_fb) {
 		drm_dbg_kms(display->drm, "failed to alloc fb\n");
 		return;
diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 71b4fa19f1e5..3f2cbcd2626d 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -2340,6 +2340,11 @@ intel_user_framebuffer_create(struct drm_device *dev,
 	return fb;
 }
 
+struct intel_framebuffer *intel_framebuffer_alloc(void)
+{
+	return intel_bo_alloc_framebuffer();
+}
+
 struct drm_framebuffer *
 intel_framebuffer_create(struct drm_gem_object *obj,
 			 struct drm_mode_fb_cmd2 *mode_cmd)
@@ -2347,7 +2352,7 @@ intel_framebuffer_create(struct drm_gem_object *obj,
 	struct intel_framebuffer *intel_fb;
 	int ret;
 
-	intel_fb = intel_bo_alloc_framebuffer();
+	intel_fb = intel_framebuffer_alloc();
 	if (!intel_fb)
 		return ERR_PTR(-ENOMEM);
 
diff --git a/drivers/gpu/drm/i915/display/intel_fb.h b/drivers/gpu/drm/i915/display/intel_fb.h
index bdd76b372957..7d77022235a5 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.h
+++ b/drivers/gpu/drm/i915/display/intel_fb.h
@@ -103,6 +103,9 @@ void intel_add_fb_offsets(int *x, int *y,
 int intel_framebuffer_init(struct intel_framebuffer *ifb,
 			   struct drm_gem_object *obj,
 			   struct drm_mode_fb_cmd2 *mode_cmd);
+
+struct intel_framebuffer *intel_framebuffer_alloc(void);
+
 struct drm_framebuffer *
 intel_framebuffer_create(struct drm_gem_object *obj,
 			 struct drm_mode_fb_cmd2 *mode_cmd);
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 950dc79dbdd4..59ee0830de02 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -3028,7 +3028,7 @@ skl_get_initial_plane_config(struct intel_crtc *crtc,
 		return;
 	}
 
-	intel_fb = intel_bo_alloc_framebuffer();
+	intel_fb = intel_framebuffer_alloc();
 	if (!intel_fb) {
 		drm_dbg_kms(display->drm, "failed to alloc fb\n");
 		return;

From 5f54b2196c674adc02a22f99da414b503daccb4c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:48 +0300
Subject: [PATCH 1078/1455] drm/{i915,xe}/panic: split out intel_panic.[ch]

intel_bo.[ch] is not the appropriate location for the panic
functionality. Split out intel_panic.[ch] and xe_panic.c in i915 and
xe. Keep the function names for now.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/d98b831a011a028ffd33ce99b0ba62be061ee235.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/Makefile                 |  1 +
 drivers/gpu/drm/i915/display/i9xx_plane.c     |  1 +
 drivers/gpu/drm/i915/display/intel_bo.c       | 15 ---
 drivers/gpu/drm/i915/display/intel_bo.h       |  3 -
 drivers/gpu/drm/i915/display/intel_fb.c       |  1 +
 drivers/gpu/drm/i915/display/intel_panic.c    | 20 ++++
 drivers/gpu/drm/i915/display/intel_panic.h    | 14 +++
 drivers/gpu/drm/i915/display/intel_plane.c    |  2 +-
 .../drm/i915/display/skl_universal_plane.c    |  1 +
 drivers/gpu/drm/xe/Makefile                   |  1 +
 drivers/gpu/drm/xe/display/intel_bo.c         | 91 ------------------
 drivers/gpu/drm/xe/display/xe_panic.c         | 96 +++++++++++++++++++
 12 files changed, 136 insertions(+), 110 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/display/intel_panic.c
 create mode 100644 drivers/gpu/drm/i915/display/intel_panic.h
 create mode 100644 drivers/gpu/drm/xe/display/xe_panic.c

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index 853543443072..db150a0c33ce 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -280,6 +280,7 @@ i915-y += \
 	display/intel_modeset_setup.o \
 	display/intel_modeset_verify.o \
 	display/intel_overlay.o \
+	display/intel_panic.o \
 	display/intel_pch.o \
 	display/intel_pch_display.o \
 	display/intel_pch_refclk.o \
diff --git a/drivers/gpu/drm/i915/display/i9xx_plane.c b/drivers/gpu/drm/i915/display/i9xx_plane.c
index 9a39e9efbfaf..407deb5dfb57 100644
--- a/drivers/gpu/drm/i915/display/i9xx_plane.c
+++ b/drivers/gpu/drm/i915/display/i9xx_plane.c
@@ -22,6 +22,7 @@
 #include "intel_fb.h"
 #include "intel_fbc.h"
 #include "intel_frontbuffer.h"
+#include "intel_panic.h"
 #include "intel_plane.h"
 #include "intel_sprite.h"
 
diff --git a/drivers/gpu/drm/i915/display/intel_bo.c b/drivers/gpu/drm/i915/display/intel_bo.c
index d29c1508ccb9..6ae1374d5c2b 100644
--- a/drivers/gpu/drm/i915/display/intel_bo.c
+++ b/drivers/gpu/drm/i915/display/intel_bo.c
@@ -59,18 +59,3 @@ void intel_bo_describe(struct seq_file *m, struct drm_gem_object *obj)
 {
 	i915_debugfs_describe_obj(m, to_intel_bo(obj));
 }
-
-struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
-{
-	return i915_gem_object_alloc_framebuffer();
-}
-
-int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
-{
-	return i915_gem_object_panic_setup(sb);
-}
-
-void intel_bo_panic_finish(struct intel_framebuffer *fb)
-{
-	return i915_gem_object_panic_finish(fb);
-}
diff --git a/drivers/gpu/drm/i915/display/intel_bo.h b/drivers/gpu/drm/i915/display/intel_bo.h
index 97087a64d23b..48d87019e48a 100644
--- a/drivers/gpu/drm/i915/display/intel_bo.h
+++ b/drivers/gpu/drm/i915/display/intel_bo.h
@@ -25,8 +25,5 @@ struct intel_frontbuffer *intel_bo_set_frontbuffer(struct drm_gem_object *obj,
 						   struct intel_frontbuffer *front);
 
 void intel_bo_describe(struct seq_file *m, struct drm_gem_object *obj);
-struct intel_framebuffer *intel_bo_alloc_framebuffer(void);
-int intel_bo_panic_setup(struct drm_scanout_buffer *sb);
-void intel_bo_panic_finish(struct intel_framebuffer *fb);
 
 #endif /* __INTEL_BO__ */
diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 3f2cbcd2626d..4b415dfb5a66 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -20,6 +20,7 @@
 #include "intel_fb.h"
 #include "intel_fb_bo.h"
 #include "intel_frontbuffer.h"
+#include "intel_panic.h"
 #include "intel_plane.h"
 
 #define check_array_bounds(display, a, i) drm_WARN_ON((display)->drm, (i) >= ARRAY_SIZE(a))
diff --git a/drivers/gpu/drm/i915/display/intel_panic.c b/drivers/gpu/drm/i915/display/intel_panic.c
new file mode 100644
index 000000000000..9eae210bb4cd
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_panic.c
@@ -0,0 +1,20 @@
+// SPDX-License-Identifier: MIT
+/* Copyright © 2025 Intel Corporation */
+
+#include "gem/i915_gem_object.h"
+#include "intel_panic.h"
+
+struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
+{
+	return i915_gem_object_alloc_framebuffer();
+}
+
+int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
+{
+	return i915_gem_object_panic_setup(sb);
+}
+
+void intel_bo_panic_finish(struct intel_framebuffer *fb)
+{
+	return i915_gem_object_panic_finish(fb);
+}
diff --git a/drivers/gpu/drm/i915/display/intel_panic.h b/drivers/gpu/drm/i915/display/intel_panic.h
new file mode 100644
index 000000000000..07e72240cbd7
--- /dev/null
+++ b/drivers/gpu/drm/i915/display/intel_panic.h
@@ -0,0 +1,14 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __INTEL_PANIC_H__
+#define __INTEL_PANIC_H__
+
+struct drm_scanout_buffer;
+struct intel_framebuffer;
+
+struct intel_framebuffer *intel_bo_alloc_framebuffer(void);
+int intel_bo_panic_setup(struct drm_scanout_buffer *sb);
+void intel_bo_panic_finish(struct intel_framebuffer *fb);
+
+#endif /* __INTEL_PANIC_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c
index 81f05ee9a21a..a467e14e6388 100644
--- a/drivers/gpu/drm/i915/display/intel_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_plane.c
@@ -47,7 +47,6 @@
 #include "gem/i915_gem_object.h"
 #include "i915_scheduler_types.h"
 #include "i9xx_plane_regs.h"
-#include "intel_bo.h"
 #include "intel_cdclk.h"
 #include "intel_cursor.h"
 #include "intel_display_rps.h"
@@ -56,6 +55,7 @@
 #include "intel_fb.h"
 #include "intel_fb_pin.h"
 #include "intel_fbdev.h"
+#include "intel_panic.h"
 #include "intel_plane.h"
 #include "intel_psr.h"
 #include "skl_scaler.h"
diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index 59ee0830de02..e13fb781e7b2 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -20,6 +20,7 @@
 #include "intel_fb.h"
 #include "intel_fbc.h"
 #include "intel_frontbuffer.h"
+#include "intel_panic.h"
 #include "intel_plane.h"
 #include "intel_psr.h"
 #include "intel_psr_regs.h"
diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index c83d72de81dd..d2202f2a2374 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -192,6 +192,7 @@ xe-$(CONFIG_DRM_XE_DISPLAY) += \
 	display/xe_dsb_buffer.o \
 	display/xe_fb_pin.o \
 	display/xe_hdcp_gsc.o \
+	display/xe_panic.o \
 	display/xe_plane_initial.o \
 	display/xe_tdf.o
 
diff --git a/drivers/gpu/drm/xe/display/intel_bo.c b/drivers/gpu/drm/xe/display/intel_bo.c
index 910632f57c3d..27437c22bd70 100644
--- a/drivers/gpu/drm/xe/display/intel_bo.c
+++ b/drivers/gpu/drm/xe/display/intel_bo.c
@@ -1,12 +1,7 @@
 // SPDX-License-Identifier: MIT
 /* Copyright © 2024 Intel Corporation */
 
-#include <drm/drm_cache.h>
 #include <drm/drm_gem.h>
-#include <drm/drm_panic.h>
-
-#include "intel_fb.h"
-#include "intel_display_types.h"
 
 #include "xe_bo.h"
 #include "intel_bo.h"
@@ -64,89 +59,3 @@ void intel_bo_describe(struct seq_file *m, struct drm_gem_object *obj)
 {
 	/* FIXME */
 }
-
-struct xe_panic_data {
-	struct page **pages;
-	int page;
-	void *vaddr;
-};
-
-struct xe_framebuffer {
-	struct intel_framebuffer base;
-	struct xe_panic_data panic;
-};
-
-static inline struct xe_panic_data *to_xe_panic_data(struct intel_framebuffer *fb)
-{
-	return &container_of_const(fb, struct xe_framebuffer, base)->panic;
-}
-
-static void xe_panic_kunmap(struct xe_panic_data *panic)
-{
-	if (panic->vaddr) {
-		drm_clflush_virt_range(panic->vaddr, PAGE_SIZE);
-		kunmap_local(panic->vaddr);
-		panic->vaddr = NULL;
-	}
-}
-
-/*
- * The scanout buffer pages are not mapped, so for each pixel,
- * use kmap_local_page_try_from_panic() to map the page, and write the pixel.
- * Try to keep the map from the previous pixel, to avoid too much map/unmap.
- */
-static void xe_panic_page_set_pixel(struct drm_scanout_buffer *sb, unsigned int x,
-				    unsigned int y, u32 color)
-{
-	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct xe_panic_data *panic = to_xe_panic_data(fb);
-	struct xe_bo *bo = gem_to_xe_bo(intel_fb_bo(&fb->base));
-	unsigned int new_page;
-	unsigned int offset;
-
-	if (fb->panic_tiling)
-		offset = fb->panic_tiling(sb->width, x, y);
-	else
-		offset = y * sb->pitch[0] + x * sb->format->cpp[0];
-
-	new_page = offset >> PAGE_SHIFT;
-	offset = offset % PAGE_SIZE;
-	if (new_page != panic->page) {
-		xe_panic_kunmap(panic);
-		panic->page = new_page;
-		panic->vaddr = ttm_bo_kmap_try_from_panic(&bo->ttm,
-							  panic->page);
-	}
-	if (panic->vaddr) {
-		u32 *pix = panic->vaddr + offset;
-		*pix = color;
-	}
-}
-
-struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
-{
-	struct xe_framebuffer *xe_fb;
-
-	xe_fb = kzalloc(sizeof(*xe_fb), GFP_KERNEL);
-	if (xe_fb)
-		return &xe_fb->base;
-	return NULL;
-}
-
-int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
-{
-	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct xe_panic_data *panic = to_xe_panic_data(fb);
-
-	panic->page = -1;
-	sb->set_pixel = xe_panic_page_set_pixel;
-	return 0;
-}
-
-void intel_bo_panic_finish(struct intel_framebuffer *fb)
-{
-	struct xe_panic_data *panic = to_xe_panic_data(fb);
-
-	xe_panic_kunmap(panic);
-	panic->page = -1;
-}
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
new file mode 100644
index 000000000000..14fba8aa4e5f
--- /dev/null
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -0,0 +1,96 @@
+// SPDX-License-Identifier: MIT
+/* Copyright © 2025 Intel Corporation */
+
+#include <drm/drm_cache.h>
+#include <drm/drm_panic.h>
+
+#include "intel_display_types.h"
+#include "intel_fb.h"
+#include "intel_panic.h"
+#include "xe_bo.h"
+
+struct xe_panic_data {
+	struct page **pages;
+	int page;
+	void *vaddr;
+};
+
+struct xe_framebuffer {
+	struct intel_framebuffer base;
+	struct xe_panic_data panic;
+};
+
+static inline struct xe_panic_data *to_xe_panic_data(struct intel_framebuffer *fb)
+{
+	return &container_of_const(fb, struct xe_framebuffer, base)->panic;
+}
+
+static void xe_panic_kunmap(struct xe_panic_data *panic)
+{
+	if (panic->vaddr) {
+		drm_clflush_virt_range(panic->vaddr, PAGE_SIZE);
+		kunmap_local(panic->vaddr);
+		panic->vaddr = NULL;
+	}
+}
+
+/*
+ * The scanout buffer pages are not mapped, so for each pixel,
+ * use kmap_local_page_try_from_panic() to map the page, and write the pixel.
+ * Try to keep the map from the previous pixel, to avoid too much map/unmap.
+ */
+static void xe_panic_page_set_pixel(struct drm_scanout_buffer *sb, unsigned int x,
+				    unsigned int y, u32 color)
+{
+	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
+	struct xe_panic_data *panic = to_xe_panic_data(fb);
+	struct xe_bo *bo = gem_to_xe_bo(intel_fb_bo(&fb->base));
+	unsigned int new_page;
+	unsigned int offset;
+
+	if (fb->panic_tiling)
+		offset = fb->panic_tiling(sb->width, x, y);
+	else
+		offset = y * sb->pitch[0] + x * sb->format->cpp[0];
+
+	new_page = offset >> PAGE_SHIFT;
+	offset = offset % PAGE_SIZE;
+	if (new_page != panic->page) {
+		xe_panic_kunmap(panic);
+		panic->page = new_page;
+		panic->vaddr = ttm_bo_kmap_try_from_panic(&bo->ttm,
+							  panic->page);
+	}
+	if (panic->vaddr) {
+		u32 *pix = panic->vaddr + offset;
+		*pix = color;
+	}
+}
+
+struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
+{
+	struct xe_framebuffer *xe_fb;
+
+	xe_fb = kzalloc(sizeof(*xe_fb), GFP_KERNEL);
+	if (xe_fb)
+		return &xe_fb->base;
+	return NULL;
+}
+
+int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
+{
+	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
+	struct xe_panic_data *panic = to_xe_panic_data(fb);
+
+	panic->page = -1;
+	sb->set_pixel = xe_panic_page_set_pixel;
+	return 0;
+}
+
+void intel_bo_panic_finish(struct intel_framebuffer *fb)
+{
+	struct xe_panic_data *panic = to_xe_panic_data(fb);
+
+	xe_panic_kunmap(panic);
+	panic->page = -1;
+}

From 05a85e781ff0e37ccadcac133aa355b81db86ada Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:49 +0300
Subject: [PATCH 1079/1455] drm/{i915,xe}/panic: rename intel_bo_panic_*() to
 intel_panic_*()

Rename the intel_bo_panic_*() functions according to the functionality,
dropping the misleading intel_bo reference.

Keep intel_bo_alloc_framebuffer() for now; it'll be refactored later.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/a3be8f8b5e7dd208027a1131e89452d9a214054f.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_panic.c | 4 ++--
 drivers/gpu/drm/i915/display/intel_panic.h | 4 ++--
 drivers/gpu/drm/i915/display/intel_plane.c | 4 ++--
 drivers/gpu/drm/xe/display/xe_panic.c      | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_panic.c b/drivers/gpu/drm/i915/display/intel_panic.c
index 9eae210bb4cd..20eecb0f168f 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.c
+++ b/drivers/gpu/drm/i915/display/intel_panic.c
@@ -9,12 +9,12 @@ struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
 	return i915_gem_object_alloc_framebuffer();
 }
 
-int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
+int intel_panic_setup(struct drm_scanout_buffer *sb)
 {
 	return i915_gem_object_panic_setup(sb);
 }
 
-void intel_bo_panic_finish(struct intel_framebuffer *fb)
+void intel_panic_finish(struct intel_framebuffer *fb)
 {
 	return i915_gem_object_panic_finish(fb);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_panic.h b/drivers/gpu/drm/i915/display/intel_panic.h
index 07e72240cbd7..67ce253fcdf5 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.h
+++ b/drivers/gpu/drm/i915/display/intel_panic.h
@@ -8,7 +8,7 @@ struct drm_scanout_buffer;
 struct intel_framebuffer;
 
 struct intel_framebuffer *intel_bo_alloc_framebuffer(void);
-int intel_bo_panic_setup(struct drm_scanout_buffer *sb);
-void intel_bo_panic_finish(struct intel_framebuffer *fb);
+int intel_panic_setup(struct drm_scanout_buffer *sb);
+void intel_panic_finish(struct intel_framebuffer *fb);
 
 #endif /* __INTEL_PANIC_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c
index a467e14e6388..890d897bda10 100644
--- a/drivers/gpu/drm/i915/display/intel_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_plane.c
@@ -1326,7 +1326,7 @@ static void intel_panic_flush(struct drm_plane *plane)
 	struct drm_framebuffer *fb = plane_state->hw.fb;
 	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
 
-	intel_bo_panic_finish(intel_fb);
+	intel_panic_finish(intel_fb);
 
 	if (crtc_state->enable_psr2_sel_fetch) {
 		/* Force a full update for psr2 */
@@ -1409,7 +1409,7 @@ static int intel_get_scanout_buffer(struct drm_plane *plane,
 				return -EOPNOTSUPP;
 		}
 		sb->private = intel_fb;
-		ret = intel_bo_panic_setup(sb);
+		ret = intel_panic_setup(sb);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
index 14fba8aa4e5f..18b27a06b260 100644
--- a/drivers/gpu/drm/xe/display/xe_panic.c
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -77,7 +77,7 @@ struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
 	return NULL;
 }
 
-int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
+int intel_panic_setup(struct drm_scanout_buffer *sb)
 {
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
 	struct xe_panic_data *panic = to_xe_panic_data(fb);
@@ -87,7 +87,7 @@ int intel_bo_panic_setup(struct drm_scanout_buffer *sb)
 	return 0;
 }
 
-void intel_bo_panic_finish(struct intel_framebuffer *fb)
+void intel_panic_finish(struct intel_framebuffer *fb)
 {
 	struct xe_panic_data *panic = to_xe_panic_data(fb);
 

From 4e839f2c7bafb18c450af78a1b6fc78e423a2039 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:50 +0300
Subject: [PATCH 1080/1455] drm/{i915,xe}/fb: add panic pointer member to
 struct intel_framebuffer

Add a panic data pointer member in struct intel_framebuffer in
preparation for breaking the artificial subclassing between
intel_framebuffer and panic structures.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/41f42e1de8545409274d54854aa12e0fb390e394.1756835342.git.jani.nikula@intel.com
---
 .../drm/i915/display/intel_display_types.h    |  1 +
 drivers/gpu/drm/i915/gem/i915_gem_pages.c     | 20 +++++++++----------
 drivers/gpu/drm/xe/display/xe_panic.c         | 20 +++++++++----------
 3 files changed, 19 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index fd9d2527889b..608fd69af657 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -149,6 +149,7 @@ struct intel_framebuffer {
 	unsigned int vtd_guard;
 
 	unsigned int (*panic_tiling)(unsigned int x, unsigned int y, unsigned int width);
+	void *panic;
 };
 
 enum intel_hotplug_state {
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index c16a57160b26..c54ed1b33e60 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -368,11 +368,6 @@ struct i915_framebuffer {
 	struct i915_panic_data panic;
 };
 
-static inline struct i915_panic_data *to_i915_panic_data(struct intel_framebuffer *fb)
-{
-	return &container_of_const(fb, struct i915_framebuffer, base)->panic;
-}
-
 static void i915_panic_kunmap(struct i915_panic_data *panic)
 {
 	if (panic->vaddr) {
@@ -420,7 +415,7 @@ static void i915_gem_object_panic_page_set_pixel(struct drm_scanout_buffer *sb,
 	unsigned int new_page;
 	unsigned int offset;
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct i915_panic_data *panic = to_i915_panic_data(fb);
+	struct i915_panic_data *panic = fb->panic;
 
 	if (fb->panic_tiling)
 		offset = fb->panic_tiling(sb->width, x, y);
@@ -446,9 +441,12 @@ struct intel_framebuffer *i915_gem_object_alloc_framebuffer(void)
 	struct i915_framebuffer *i915_fb;
 
 	i915_fb = kzalloc(sizeof(*i915_fb), GFP_KERNEL);
-	if (i915_fb)
-		return &i915_fb->base;
-	return NULL;
+	if (!i915_fb)
+		return NULL;
+
+	i915_fb->base.panic = &i915_fb->panic;
+
+	return &i915_fb->base;
 }
 
 /*
@@ -460,7 +458,7 @@ int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
 {
 	enum i915_map_type has_type;
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct i915_panic_data *panic = to_i915_panic_data(fb);
+	struct i915_panic_data *panic = fb->panic;
 	struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base));
 	void *ptr;
 
@@ -488,7 +486,7 @@ int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
 
 void i915_gem_object_panic_finish(struct intel_framebuffer *fb)
 {
-	struct i915_panic_data *panic = to_i915_panic_data(fb);
+	struct i915_panic_data *panic = fb->panic;
 
 	i915_panic_kunmap(panic);
 	panic->page = -1;
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
index 18b27a06b260..fc1804d330e4 100644
--- a/drivers/gpu/drm/xe/display/xe_panic.c
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -20,11 +20,6 @@ struct xe_framebuffer {
 	struct xe_panic_data panic;
 };
 
-static inline struct xe_panic_data *to_xe_panic_data(struct intel_framebuffer *fb)
-{
-	return &container_of_const(fb, struct xe_framebuffer, base)->panic;
-}
-
 static void xe_panic_kunmap(struct xe_panic_data *panic)
 {
 	if (panic->vaddr) {
@@ -43,7 +38,7 @@ static void xe_panic_page_set_pixel(struct drm_scanout_buffer *sb, unsigned int
 				    unsigned int y, u32 color)
 {
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct xe_panic_data *panic = to_xe_panic_data(fb);
+	struct xe_panic_data *panic = fb->panic;
 	struct xe_bo *bo = gem_to_xe_bo(intel_fb_bo(&fb->base));
 	unsigned int new_page;
 	unsigned int offset;
@@ -72,15 +67,18 @@ struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
 	struct xe_framebuffer *xe_fb;
 
 	xe_fb = kzalloc(sizeof(*xe_fb), GFP_KERNEL);
-	if (xe_fb)
-		return &xe_fb->base;
-	return NULL;
+	if (!xe_fb)
+		return NULL;
+
+	xe_fb->base.panic = &xe_fb->panic;
+
+	return &xe_fb->base;
 }
 
 int intel_panic_setup(struct drm_scanout_buffer *sb)
 {
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct xe_panic_data *panic = to_xe_panic_data(fb);
+	struct xe_panic_data *panic = fb->panic;
 
 	panic->page = -1;
 	sb->set_pixel = xe_panic_page_set_pixel;
@@ -89,7 +87,7 @@ int intel_panic_setup(struct drm_scanout_buffer *sb)
 
 void intel_panic_finish(struct intel_framebuffer *fb)
 {
-	struct xe_panic_data *panic = to_xe_panic_data(fb);
+	struct xe_panic_data *panic = fb->panic;
 
 	xe_panic_kunmap(panic);
 	panic->page = -1;

From fa48f6335af9b063303d9e9a036514bbdc774d5b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:51 +0300
Subject: [PATCH 1081/1455] drm/{i915,xe}/panic: rename struct
 {i915,xe}_panic_data to struct intel_panic

Prepare for better shared interfaces between panic implementations. The
struct intel_panic remains an opaque data type, with unique
implementations in i915 and xe.

This allows us to change the panic data pointer from void * to struct
intel_panic *, helping type safety.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/884ffc80c8b5fef1b92956e644a4e559560cc2ba.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_display_types.h |  3 ++-
 drivers/gpu/drm/i915/gem/i915_gem_pages.c          | 12 ++++++------
 drivers/gpu/drm/xe/display/xe_panic.c              | 12 ++++++------
 3 files changed, 14 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h
index 608fd69af657..358ab922d7a7 100644
--- a/drivers/gpu/drm/i915/display/intel_display_types.h
+++ b/drivers/gpu/drm/i915/display/intel_display_types.h
@@ -60,6 +60,7 @@ struct intel_ddi_buf_trans;
 struct intel_fbc;
 struct intel_global_objs_state;
 struct intel_hdcp_shim;
+struct intel_panic;
 struct intel_tc_port;
 
 /*
@@ -149,7 +150,7 @@ struct intel_framebuffer {
 	unsigned int vtd_guard;
 
 	unsigned int (*panic_tiling)(unsigned int x, unsigned int y, unsigned int width);
-	void *panic;
+	struct intel_panic *panic;
 };
 
 enum intel_hotplug_state {
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index c54ed1b33e60..e36d60b785b1 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -357,7 +357,7 @@ static void *i915_gem_object_map_pfn(struct drm_i915_gem_object *obj,
 	return vaddr ?: ERR_PTR(-ENOMEM);
 }
 
-struct i915_panic_data {
+struct intel_panic {
 	struct page **pages;
 	int page;
 	void *vaddr;
@@ -365,10 +365,10 @@ struct i915_panic_data {
 
 struct i915_framebuffer {
 	struct intel_framebuffer base;
-	struct i915_panic_data panic;
+	struct intel_panic panic;
 };
 
-static void i915_panic_kunmap(struct i915_panic_data *panic)
+static void i915_panic_kunmap(struct intel_panic *panic)
 {
 	if (panic->vaddr) {
 		drm_clflush_virt_range(panic->vaddr, PAGE_SIZE);
@@ -415,7 +415,7 @@ static void i915_gem_object_panic_page_set_pixel(struct drm_scanout_buffer *sb,
 	unsigned int new_page;
 	unsigned int offset;
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct i915_panic_data *panic = fb->panic;
+	struct intel_panic *panic = fb->panic;
 
 	if (fb->panic_tiling)
 		offset = fb->panic_tiling(sb->width, x, y);
@@ -458,7 +458,7 @@ int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
 {
 	enum i915_map_type has_type;
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct i915_panic_data *panic = fb->panic;
+	struct intel_panic *panic = fb->panic;
 	struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base));
 	void *ptr;
 
@@ -486,7 +486,7 @@ int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
 
 void i915_gem_object_panic_finish(struct intel_framebuffer *fb)
 {
-	struct i915_panic_data *panic = fb->panic;
+	struct intel_panic *panic = fb->panic;
 
 	i915_panic_kunmap(panic);
 	panic->page = -1;
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
index fc1804d330e4..3ef23a6795b3 100644
--- a/drivers/gpu/drm/xe/display/xe_panic.c
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -9,7 +9,7 @@
 #include "intel_panic.h"
 #include "xe_bo.h"
 
-struct xe_panic_data {
+struct intel_panic {
 	struct page **pages;
 	int page;
 	void *vaddr;
@@ -17,10 +17,10 @@ struct xe_panic_data {
 
 struct xe_framebuffer {
 	struct intel_framebuffer base;
-	struct xe_panic_data panic;
+	struct intel_panic panic;
 };
 
-static void xe_panic_kunmap(struct xe_panic_data *panic)
+static void xe_panic_kunmap(struct intel_panic *panic)
 {
 	if (panic->vaddr) {
 		drm_clflush_virt_range(panic->vaddr, PAGE_SIZE);
@@ -38,7 +38,7 @@ static void xe_panic_page_set_pixel(struct drm_scanout_buffer *sb, unsigned int
 				    unsigned int y, u32 color)
 {
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct xe_panic_data *panic = fb->panic;
+	struct intel_panic *panic = fb->panic;
 	struct xe_bo *bo = gem_to_xe_bo(intel_fb_bo(&fb->base));
 	unsigned int new_page;
 	unsigned int offset;
@@ -78,7 +78,7 @@ struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
 int intel_panic_setup(struct drm_scanout_buffer *sb)
 {
 	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct xe_panic_data *panic = fb->panic;
+	struct intel_panic *panic = fb->panic;
 
 	panic->page = -1;
 	sb->set_pixel = xe_panic_page_set_pixel;
@@ -87,7 +87,7 @@ int intel_panic_setup(struct drm_scanout_buffer *sb)
 
 void intel_panic_finish(struct intel_framebuffer *fb)
 {
-	struct xe_panic_data *panic = fb->panic;
+	struct intel_panic *panic = fb->panic;
 
 	xe_panic_kunmap(panic);
 	panic->page = -1;

From 729c5f7ffa83391b44939efc26b7ea34a08aea74 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:52 +0300
Subject: [PATCH 1082/1455] drm/{i915,xe}/panic: move framebuffer allocation
 where it belongs

The struct intel_framebuffer allocation naturally belongs in intel_fb.c,
not hidden inside panic implementation. Separate the panic
allocation. Drop the unnecessary struct i915_framebuffer and struct
xe_framebuffer types.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/d29f63e0118d002fc8edd368caea7e8185418e17.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_fb.c    | 17 ++++++++++++++++-
 drivers/gpu/drm/i915/display/intel_panic.c |  4 ++--
 drivers/gpu/drm/i915/display/intel_panic.h |  3 ++-
 drivers/gpu/drm/i915/gem/i915_gem_object.h |  5 +++--
 drivers/gpu/drm/i915/gem/i915_gem_pages.c  | 17 ++++-------------
 drivers/gpu/drm/xe/display/xe_panic.c      | 17 ++++-------------
 6 files changed, 31 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fb.c b/drivers/gpu/drm/i915/display/intel_fb.c
index 4b415dfb5a66..c55855729628 100644
--- a/drivers/gpu/drm/i915/display/intel_fb.c
+++ b/drivers/gpu/drm/i915/display/intel_fb.c
@@ -2343,7 +2343,22 @@ intel_user_framebuffer_create(struct drm_device *dev,
 
 struct intel_framebuffer *intel_framebuffer_alloc(void)
 {
-	return intel_bo_alloc_framebuffer();
+	struct intel_framebuffer *intel_fb;
+	struct intel_panic *panic;
+
+	intel_fb = kzalloc(sizeof(*intel_fb), GFP_KERNEL);
+	if (!intel_fb)
+		return NULL;
+
+	panic = intel_panic_alloc();
+	if (!panic) {
+		kfree(intel_fb);
+		return NULL;
+	}
+
+	intel_fb->panic = panic;
+
+	return intel_fb;
 }
 
 struct drm_framebuffer *
diff --git a/drivers/gpu/drm/i915/display/intel_panic.c b/drivers/gpu/drm/i915/display/intel_panic.c
index 20eecb0f168f..5431bd4d3a7d 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.c
+++ b/drivers/gpu/drm/i915/display/intel_panic.c
@@ -4,9 +4,9 @@
 #include "gem/i915_gem_object.h"
 #include "intel_panic.h"
 
-struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
+struct intel_panic *intel_panic_alloc(void)
 {
-	return i915_gem_object_alloc_framebuffer();
+	return i915_gem_object_alloc_panic();
 }
 
 int intel_panic_setup(struct drm_scanout_buffer *sb)
diff --git a/drivers/gpu/drm/i915/display/intel_panic.h b/drivers/gpu/drm/i915/display/intel_panic.h
index 67ce253fcdf5..45ce6104e6fb 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.h
+++ b/drivers/gpu/drm/i915/display/intel_panic.h
@@ -6,8 +6,9 @@
 
 struct drm_scanout_buffer;
 struct intel_framebuffer;
+struct intel_panic;
 
-struct intel_framebuffer *intel_bo_alloc_framebuffer(void);
+struct intel_panic *intel_panic_alloc(void);
 int intel_panic_setup(struct drm_scanout_buffer *sb);
 void intel_panic_finish(struct intel_framebuffer *fb);
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 565f8fa330db..9b3f25cb48db 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -16,9 +16,10 @@
 #include "i915_gem_ww.h"
 #include "i915_vma_types.h"
 
-struct drm_scanout_buffer;
 enum intel_region_id;
+struct drm_scanout_buffer;
 struct intel_framebuffer;
+struct intel_panic;
 
 #define obj_to_i915(obj__) to_i915((obj__)->base.dev)
 
@@ -693,7 +694,7 @@ i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj)
 int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
 int i915_gem_object_truncate(struct drm_i915_gem_object *obj);
 
-struct intel_framebuffer *i915_gem_object_alloc_framebuffer(void);
+struct intel_panic *i915_gem_object_alloc_panic(void);
 int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb);
 void i915_gem_object_panic_finish(struct intel_framebuffer *fb);
 
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index e36d60b785b1..b219474aecc7 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -363,11 +363,6 @@ struct intel_panic {
 	void *vaddr;
 };
 
-struct i915_framebuffer {
-	struct intel_framebuffer base;
-	struct intel_panic panic;
-};
-
 static void i915_panic_kunmap(struct intel_panic *panic)
 {
 	if (panic->vaddr) {
@@ -436,17 +431,13 @@ static void i915_gem_object_panic_page_set_pixel(struct drm_scanout_buffer *sb,
 	}
 }
 
-struct intel_framebuffer *i915_gem_object_alloc_framebuffer(void)
+struct intel_panic *i915_gem_object_alloc_panic(void)
 {
-	struct i915_framebuffer *i915_fb;
+	struct intel_panic *panic;
 
-	i915_fb = kzalloc(sizeof(*i915_fb), GFP_KERNEL);
-	if (!i915_fb)
-		return NULL;
+	panic = kzalloc(sizeof(*panic), GFP_KERNEL);
 
-	i915_fb->base.panic = &i915_fb->panic;
-
-	return &i915_fb->base;
+	return panic;
 }
 
 /*
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
index 3ef23a6795b3..b5a7615708a1 100644
--- a/drivers/gpu/drm/xe/display/xe_panic.c
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -15,11 +15,6 @@ struct intel_panic {
 	void *vaddr;
 };
 
-struct xe_framebuffer {
-	struct intel_framebuffer base;
-	struct intel_panic panic;
-};
-
 static void xe_panic_kunmap(struct intel_panic *panic)
 {
 	if (panic->vaddr) {
@@ -62,17 +57,13 @@ static void xe_panic_page_set_pixel(struct drm_scanout_buffer *sb, unsigned int
 	}
 }
 
-struct intel_framebuffer *intel_bo_alloc_framebuffer(void)
+struct intel_panic *intel_panic_alloc(void)
 {
-	struct xe_framebuffer *xe_fb;
+	struct intel_panic *panic;
 
-	xe_fb = kzalloc(sizeof(*xe_fb), GFP_KERNEL);
-	if (!xe_fb)
-		return NULL;
+	panic = kzalloc(sizeof(*panic), GFP_KERNEL);
 
-	xe_fb->base.panic = &xe_fb->panic;
-
-	return &xe_fb->base;
+	return panic;
 }
 
 int intel_panic_setup(struct drm_scanout_buffer *sb)

From 427a3f939aa3d0c77beb34b469d17244f775b14b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:53 +0300
Subject: [PATCH 1083/1455] drm/{i915,xe}/panic: convert intel_panic_finish()
 to struct intel_panic

The intel_panic_finish() function really needs the struct intel_panic
pointer, not struct intel_framebuffer. Make it so.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/3fdbcbe17e0e90c4a590f2a2486a9ec79a90cf62.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_panic.c | 4 ++--
 drivers/gpu/drm/i915/display/intel_panic.h | 3 +--
 drivers/gpu/drm/i915/display/intel_plane.c | 2 +-
 drivers/gpu/drm/i915/gem/i915_gem_object.h | 3 +--
 drivers/gpu/drm/i915/gem/i915_gem_pages.c  | 4 +---
 drivers/gpu/drm/xe/display/xe_panic.c      | 4 +---
 6 files changed, 7 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_panic.c b/drivers/gpu/drm/i915/display/intel_panic.c
index 5431bd4d3a7d..c8ac8b5a39f3 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.c
+++ b/drivers/gpu/drm/i915/display/intel_panic.c
@@ -14,7 +14,7 @@ int intel_panic_setup(struct drm_scanout_buffer *sb)
 	return i915_gem_object_panic_setup(sb);
 }
 
-void intel_panic_finish(struct intel_framebuffer *fb)
+void intel_panic_finish(struct intel_panic *panic)
 {
-	return i915_gem_object_panic_finish(fb);
+	return i915_gem_object_panic_finish(panic);
 }
diff --git a/drivers/gpu/drm/i915/display/intel_panic.h b/drivers/gpu/drm/i915/display/intel_panic.h
index 45ce6104e6fb..cec193e725a4 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.h
+++ b/drivers/gpu/drm/i915/display/intel_panic.h
@@ -5,11 +5,10 @@
 #define __INTEL_PANIC_H__
 
 struct drm_scanout_buffer;
-struct intel_framebuffer;
 struct intel_panic;
 
 struct intel_panic *intel_panic_alloc(void);
 int intel_panic_setup(struct drm_scanout_buffer *sb);
-void intel_panic_finish(struct intel_framebuffer *fb);
+void intel_panic_finish(struct intel_panic *panic);
 
 #endif /* __INTEL_PANIC_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c
index 890d897bda10..6a3f4a7b6dfd 100644
--- a/drivers/gpu/drm/i915/display/intel_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_plane.c
@@ -1326,7 +1326,7 @@ static void intel_panic_flush(struct drm_plane *plane)
 	struct drm_framebuffer *fb = plane_state->hw.fb;
 	struct intel_framebuffer *intel_fb = to_intel_framebuffer(fb);
 
-	intel_panic_finish(intel_fb);
+	intel_panic_finish(intel_fb->panic);
 
 	if (crtc_state->enable_psr2_sel_fetch) {
 		/* Force a full update for psr2 */
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 9b3f25cb48db..08da7ec670fe 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -18,7 +18,6 @@
 
 enum intel_region_id;
 struct drm_scanout_buffer;
-struct intel_framebuffer;
 struct intel_panic;
 
 #define obj_to_i915(obj__) to_i915((obj__)->base.dev)
@@ -696,7 +695,7 @@ int i915_gem_object_truncate(struct drm_i915_gem_object *obj);
 
 struct intel_panic *i915_gem_object_alloc_panic(void);
 int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb);
-void i915_gem_object_panic_finish(struct intel_framebuffer *fb);
+void i915_gem_object_panic_finish(struct intel_panic *panic);
 
 /**
  * i915_gem_object_pin_map - return a contiguous mapping of the entire object
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index b219474aecc7..3505b1842c40 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -475,10 +475,8 @@ int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
 	return -EOPNOTSUPP;
 }
 
-void i915_gem_object_panic_finish(struct intel_framebuffer *fb)
+void i915_gem_object_panic_finish(struct intel_panic *panic)
 {
-	struct intel_panic *panic = fb->panic;
-
 	i915_panic_kunmap(panic);
 	panic->page = -1;
 	kfree(panic->pages);
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
index b5a7615708a1..ea9583473ea3 100644
--- a/drivers/gpu/drm/xe/display/xe_panic.c
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -76,10 +76,8 @@ int intel_panic_setup(struct drm_scanout_buffer *sb)
 	return 0;
 }
 
-void intel_panic_finish(struct intel_framebuffer *fb)
+void intel_panic_finish(struct intel_panic *panic)
 {
-	struct intel_panic *panic = fb->panic;
-
 	xe_panic_kunmap(panic);
 	panic->page = -1;
 }

From 0b470321e931432157820ab417f54ec4deb50ffa Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Tue, 2 Sep 2025 20:51:54 +0300
Subject: [PATCH 1084/1455] drm/{i915,xe}/panic: pass struct intel_panic to
 intel_panic_setup()

Reduce the struct intel_framebuffer usage within the panic
implementation.

Cc: Jocelyn Falempe <jfalempe@redhat.com>
Cc: Maarten Lankhorst <dev@lankhorst.se>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com>
Link: https://lore.kernel.org/r/2a016167b1f6f0b432aed0a630f9dbcd07fadb7b.1756835342.git.jani.nikula@intel.com
---
 drivers/gpu/drm/i915/display/intel_panic.c | 11 +++++++++--
 drivers/gpu/drm/i915/display/intel_panic.h |  2 +-
 drivers/gpu/drm/i915/display/intel_plane.c |  2 +-
 drivers/gpu/drm/i915/gem/i915_gem_object.h |  3 ++-
 drivers/gpu/drm/i915/gem/i915_gem_pages.c  |  9 ++++-----
 drivers/gpu/drm/xe/display/xe_panic.c      |  5 +----
 6 files changed, 18 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_panic.c b/drivers/gpu/drm/i915/display/intel_panic.c
index c8ac8b5a39f3..7311ce4e8b6c 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.c
+++ b/drivers/gpu/drm/i915/display/intel_panic.c
@@ -1,7 +1,11 @@
 // SPDX-License-Identifier: MIT
 /* Copyright © 2025 Intel Corporation */
 
+#include <drm/drm_panic.h>
+
 #include "gem/i915_gem_object.h"
+#include "intel_display_types.h"
+#include "intel_fb.h"
 #include "intel_panic.h"
 
 struct intel_panic *intel_panic_alloc(void)
@@ -9,9 +13,12 @@ struct intel_panic *intel_panic_alloc(void)
 	return i915_gem_object_alloc_panic();
 }
 
-int intel_panic_setup(struct drm_scanout_buffer *sb)
+int intel_panic_setup(struct intel_panic *panic, struct drm_scanout_buffer *sb)
 {
-	return i915_gem_object_panic_setup(sb);
+	struct intel_framebuffer *fb = sb->private;
+	struct drm_gem_object *obj = intel_fb_bo(&fb->base);
+
+	return i915_gem_object_panic_setup(panic, sb, obj, fb->panic_tiling);
 }
 
 void intel_panic_finish(struct intel_panic *panic)
diff --git a/drivers/gpu/drm/i915/display/intel_panic.h b/drivers/gpu/drm/i915/display/intel_panic.h
index cec193e725a4..afb472e924aa 100644
--- a/drivers/gpu/drm/i915/display/intel_panic.h
+++ b/drivers/gpu/drm/i915/display/intel_panic.h
@@ -8,7 +8,7 @@ struct drm_scanout_buffer;
 struct intel_panic;
 
 struct intel_panic *intel_panic_alloc(void);
-int intel_panic_setup(struct drm_scanout_buffer *sb);
+int intel_panic_setup(struct intel_panic *panic, struct drm_scanout_buffer *sb);
 void intel_panic_finish(struct intel_panic *panic);
 
 #endif /* __INTEL_PANIC_H__ */
diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c
index 6a3f4a7b6dfd..2329f09d413d 100644
--- a/drivers/gpu/drm/i915/display/intel_plane.c
+++ b/drivers/gpu/drm/i915/display/intel_plane.c
@@ -1409,7 +1409,7 @@ static int intel_get_scanout_buffer(struct drm_plane *plane,
 				return -EOPNOTSUPP;
 		}
 		sb->private = intel_fb;
-		ret = intel_panic_setup(sb);
+		ret = intel_panic_setup(intel_fb->panic, sb);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h
index 08da7ec670fe..148034ef504d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.h
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h
@@ -694,7 +694,8 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj);
 int i915_gem_object_truncate(struct drm_i915_gem_object *obj);
 
 struct intel_panic *i915_gem_object_alloc_panic(void);
-int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb);
+int i915_gem_object_panic_setup(struct intel_panic *panic, struct drm_scanout_buffer *sb,
+				struct drm_gem_object *_obj, bool panic_tiling);
 void i915_gem_object_panic_finish(struct intel_panic *panic);
 
 /**
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
index 3505b1842c40..76d2178572b6 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c
@@ -445,12 +445,11 @@ struct intel_panic *i915_gem_object_alloc_panic(void)
  * Use current vaddr if it exists, or setup a list of pages.
  * pfn is not supported yet.
  */
-int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
+int i915_gem_object_panic_setup(struct intel_panic *panic, struct drm_scanout_buffer *sb,
+				struct drm_gem_object *_obj, bool panic_tiling)
 {
 	enum i915_map_type has_type;
-	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct intel_panic *panic = fb->panic;
-	struct drm_i915_gem_object *obj = to_intel_bo(intel_fb_bo(&fb->base));
+	struct drm_i915_gem_object *obj = to_intel_bo(_obj);
 	void *ptr;
 
 	ptr = page_unpack_bits(obj->mm.mapping, &has_type);
@@ -460,7 +459,7 @@ int i915_gem_object_panic_setup(struct drm_scanout_buffer *sb)
 		else
 			iosys_map_set_vaddr(&sb->map[0], ptr);
 
-		if (fb->panic_tiling)
+		if (panic_tiling)
 			sb->set_pixel = i915_gem_object_panic_map_set_pixel;
 		return 0;
 	}
diff --git a/drivers/gpu/drm/xe/display/xe_panic.c b/drivers/gpu/drm/xe/display/xe_panic.c
index ea9583473ea3..f32b23338331 100644
--- a/drivers/gpu/drm/xe/display/xe_panic.c
+++ b/drivers/gpu/drm/xe/display/xe_panic.c
@@ -66,11 +66,8 @@ struct intel_panic *intel_panic_alloc(void)
 	return panic;
 }
 
-int intel_panic_setup(struct drm_scanout_buffer *sb)
+int intel_panic_setup(struct intel_panic *panic, struct drm_scanout_buffer *sb)
 {
-	struct intel_framebuffer *fb = (struct intel_framebuffer *)sb->private;
-	struct intel_panic *panic = fb->panic;
-
 	panic->page = -1;
 	sb->set_pixel = xe_panic_page_set_pixel;
 	return 0;

From 1e4c85167cf672312a178b2af0b058add77e448b Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 29 Aug 2025 20:45:59 +0300
Subject: [PATCH 1085/1455] drm/i915: rename range_overflows_end() to
 range_end_overflows()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename range_overflows_end() to range_end_overflows(), along with the _t
variant.

It's all rather subjective, but I think range_end_overflows() reads
better.

Cc: Kees Cook <kees@kernel.org>
Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Cc: linux-hardening@vger.kernel.org
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/20250829174601.2163064-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/display/intel_fbc.c | 4 ++--
 drivers/gpu/drm/i915/gt/intel_rc6.c      | 2 +-
 drivers/gpu/drm/i915/i915_utils.h        | 6 +++---
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 9e097ed80bd1..06c35c7ab4e2 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -383,11 +383,11 @@ static void i8xx_fbc_program_cfb(struct intel_fbc *fbc)
 	struct drm_i915_private *i915 = to_i915(display->drm);
 
 	drm_WARN_ON(display->drm,
-		    range_overflows_end_t(u64, i915_gem_stolen_area_address(i915),
+		    range_end_overflows_t(u64, i915_gem_stolen_area_address(i915),
 					  i915_gem_stolen_node_offset(&fbc->compressed_fb),
 					  U32_MAX));
 	drm_WARN_ON(display->drm,
-		    range_overflows_end_t(u64, i915_gem_stolen_area_address(i915),
+		    range_end_overflows_t(u64, i915_gem_stolen_area_address(i915),
 					  i915_gem_stolen_node_offset(&fbc->compressed_llb),
 					  U32_MAX));
 	intel_de_write(display, FBC_CFB_BASE,
diff --git a/drivers/gpu/drm/i915/gt/intel_rc6.c b/drivers/gpu/drm/i915/gt/intel_rc6.c
index 9ca42589da4d..bf38cc5fe872 100644
--- a/drivers/gpu/drm/i915/gt/intel_rc6.c
+++ b/drivers/gpu/drm/i915/gt/intel_rc6.c
@@ -341,7 +341,7 @@ static int vlv_rc6_init(struct intel_rc6 *rc6)
 		return PTR_ERR(pctx);
 	}
 
-	GEM_BUG_ON(range_overflows_end_t(u64,
+	GEM_BUG_ON(range_end_overflows_t(u64,
 					 i915->dsm.stolen.start,
 					 pctx->stolen->start,
 					 U32_MAX));
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 9cb40c2c4b12..fdac9a158b53 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -79,7 +79,7 @@ bool i915_error_injected(void);
 #define range_overflows_t(type, start, size, max) \
 	range_overflows((type)(start), (type)(size), (type)(max))
 
-#define range_overflows_end(start, size, max) ({ \
+#define range_end_overflows(start, size, max) ({ \
 	typeof(start) start__ = (start); \
 	typeof(size) size__ = (size); \
 	typeof(max) max__ = (max); \
@@ -88,8 +88,8 @@ bool i915_error_injected(void);
 	start__ > max__ || size__ > max__ - start__; \
 })
 
-#define range_overflows_end_t(type, start, size, max) \
-	range_overflows_end((type)(start), (type)(size), (type)(max))
+#define range_end_overflows_t(type, start, size, max) \
+	range_end_overflows((type)(start), (type)(size), (type)(max))
 
 #define ptr_mask_bits(ptr, n) ({					\
 	unsigned long __v = (unsigned long)(ptr);			\

From 59bf37add1d2223d699d1855354277b7330765e4 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 29 Aug 2025 20:46:00 +0300
Subject: [PATCH 1086/1455] drm/i915: document range_overflows() and
 range_end_overflows() macros
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Document the macros in preparation for making them more generally
available.

Cc: Kees Cook <kees@kernel.org>
Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Cc: linux-hardening@vger.kernel.org
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Link: https://lore.kernel.org/r/20250829174601.2163064-2-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/i915_utils.h | 46 +++++++++++++++++++++++++++++++
 1 file changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index fdac9a158b53..968dae941532 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -67,6 +67,18 @@ bool i915_error_injected(void);
 		drm_err(&(i915)->drm, fmt, ##__VA_ARGS__); \
 })
 
+/**
+ * range_overflows() - Check if a range is out of bounds
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * A strict check to determine if the range [@start, @start + @size) is
+ * invalid with respect to the allowable range [0, @max). Any range
+ * starting at or beyond @max is considered an overflow, even if @size is 0.
+ *
+ * Returns: true if the range is out of bounds.
+ */
 #define range_overflows(start, size, max) ({ \
 	typeof(start) start__ = (start); \
 	typeof(size) size__ = (size); \
@@ -76,9 +88,32 @@ bool i915_error_injected(void);
 	start__ >= max__ || size__ > max__ - start__; \
 })
 
+/**
+ * range_overflows_t() - Check if a range is out of bounds
+ * @type:  Data type to use.
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * Same as range_overflows() but forcing the parameters to @type.
+ *
+ * Returns: true if the range is out of bounds.
+ */
 #define range_overflows_t(type, start, size, max) \
 	range_overflows((type)(start), (type)(size), (type)(max))
 
+/**
+ * range_end_overflows() - Check if a range's endpoint is out of bounds
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * Checks only if the endpoint of a range (@start + @size) exceeds @max.
+ * Unlike range_overflows(), a zero-sized range at the boundary (@start == @max)
+ * is not considered an overflow. Useful for iterator-style checks.
+ *
+ * Returns: true if the endpoint exceeds the boundary.
+ */
 #define range_end_overflows(start, size, max) ({ \
 	typeof(start) start__ = (start); \
 	typeof(size) size__ = (size); \
@@ -88,6 +123,17 @@ bool i915_error_injected(void);
 	start__ > max__ || size__ > max__ - start__; \
 })
 
+/**
+ * range_end_overflows_t() - Check if a range's endpoint is out of bounds
+ * @type:  Data type to use.
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * Same as range_end_overflows() but forcing the parameters to @type.
+ *
+ * Returns: true if the endpoint exceeds the boundary.
+ */
 #define range_end_overflows_t(type, start, size, max) \
 	range_end_overflows((type)(start), (type)(size), (type)(max))
 

From 5f3cec21f6d57336a2cbfa9ee428ac66c77a7211 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Fri, 29 Aug 2025 20:46:01 +0300
Subject: [PATCH 1087/1455] overflow: add range_overflows() and
 range_end_overflows()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Move the range_overflows() and range_end_overflows() along with the _t
variants over from drm/i915 and drm/buddy to overflow.h.

Cc: Kees Cook <kees@kernel.org>
Cc: "Gustavo A. R. Silva" <gustavoars@kernel.org>
Cc: linux-hardening@vger.kernel.org
Reviewed-by: Kees Cook <kees@kernel.org>
Reviewed-by: Jouni Högander <jouni.hogander@intel.com>
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250829174601.2163064-3-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/i915_utils.h | 70 -------------------------------
 include/drm/drm_buddy.h           |  9 ----
 include/linux/overflow.h          | 70 +++++++++++++++++++++++++++++++
 3 files changed, 70 insertions(+), 79 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 968dae941532..eb4d43c40009 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -67,76 +67,6 @@ bool i915_error_injected(void);
 		drm_err(&(i915)->drm, fmt, ##__VA_ARGS__); \
 })
 
-/**
- * range_overflows() - Check if a range is out of bounds
- * @start: Start of the range.
- * @size:  Size of the range.
- * @max:   Exclusive upper boundary.
- *
- * A strict check to determine if the range [@start, @start + @size) is
- * invalid with respect to the allowable range [0, @max). Any range
- * starting at or beyond @max is considered an overflow, even if @size is 0.
- *
- * Returns: true if the range is out of bounds.
- */
-#define range_overflows(start, size, max) ({ \
-	typeof(start) start__ = (start); \
-	typeof(size) size__ = (size); \
-	typeof(max) max__ = (max); \
-	(void)(&start__ == &size__); \
-	(void)(&start__ == &max__); \
-	start__ >= max__ || size__ > max__ - start__; \
-})
-
-/**
- * range_overflows_t() - Check if a range is out of bounds
- * @type:  Data type to use.
- * @start: Start of the range.
- * @size:  Size of the range.
- * @max:   Exclusive upper boundary.
- *
- * Same as range_overflows() but forcing the parameters to @type.
- *
- * Returns: true if the range is out of bounds.
- */
-#define range_overflows_t(type, start, size, max) \
-	range_overflows((type)(start), (type)(size), (type)(max))
-
-/**
- * range_end_overflows() - Check if a range's endpoint is out of bounds
- * @start: Start of the range.
- * @size:  Size of the range.
- * @max:   Exclusive upper boundary.
- *
- * Checks only if the endpoint of a range (@start + @size) exceeds @max.
- * Unlike range_overflows(), a zero-sized range at the boundary (@start == @max)
- * is not considered an overflow. Useful for iterator-style checks.
- *
- * Returns: true if the endpoint exceeds the boundary.
- */
-#define range_end_overflows(start, size, max) ({ \
-	typeof(start) start__ = (start); \
-	typeof(size) size__ = (size); \
-	typeof(max) max__ = (max); \
-	(void)(&start__ == &size__); \
-	(void)(&start__ == &max__); \
-	start__ > max__ || size__ > max__ - start__; \
-})
-
-/**
- * range_end_overflows_t() - Check if a range's endpoint is out of bounds
- * @type:  Data type to use.
- * @start: Start of the range.
- * @size:  Size of the range.
- * @max:   Exclusive upper boundary.
- *
- * Same as range_end_overflows() but forcing the parameters to @type.
- *
- * Returns: true if the endpoint exceeds the boundary.
- */
-#define range_end_overflows_t(type, start, size, max) \
-	range_end_overflows((type)(start), (type)(size), (type)(max))
-
 #define ptr_mask_bits(ptr, n) ({					\
 	unsigned long __v = (unsigned long)(ptr);			\
 	(typeof(ptr))(__v & -BIT(n));					\
diff --git a/include/drm/drm_buddy.h b/include/drm/drm_buddy.h
index 9689a7c5dd36..236971681514 100644
--- a/include/drm/drm_buddy.h
+++ b/include/drm/drm_buddy.h
@@ -13,15 +13,6 @@
 
 #include <drm/drm_print.h>
 
-#define range_overflows(start, size, max) ({ \
-	typeof(start) start__ = (start); \
-	typeof(size) size__ = (size); \
-	typeof(max) max__ = (max); \
-	(void)(&start__ == &size__); \
-	(void)(&start__ == &max__); \
-	start__ >= max__ || size__ > max__ - start__; \
-})
-
 #define DRM_BUDDY_RANGE_ALLOCATION		BIT(0)
 #define DRM_BUDDY_TOPDOWN_ALLOCATION		BIT(1)
 #define DRM_BUDDY_CONTIGUOUS_ALLOCATION		BIT(2)
diff --git a/include/linux/overflow.h b/include/linux/overflow.h
index 154ed0dbb43f..725f95f7e416 100644
--- a/include/linux/overflow.h
+++ b/include/linux/overflow.h
@@ -238,6 +238,76 @@ static inline bool __must_check __must_check_overflow(bool overflow)
 			      __overflows_type_constexpr(n, T),	\
 			      __overflows_type(n, T))
 
+/**
+ * range_overflows() - Check if a range is out of bounds
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * A strict check to determine if the range [@start, @start + @size) is
+ * invalid with respect to the allowable range [0, @max). Any range
+ * starting at or beyond @max is considered an overflow, even if @size is 0.
+ *
+ * Returns: true if the range is out of bounds.
+ */
+#define range_overflows(start, size, max) ({ \
+	typeof(start) start__ = (start); \
+	typeof(size) size__ = (size); \
+	typeof(max) max__ = (max); \
+	(void)(&start__ == &size__); \
+	(void)(&start__ == &max__); \
+	start__ >= max__ || size__ > max__ - start__; \
+})
+
+/**
+ * range_overflows_t() - Check if a range is out of bounds
+ * @type:  Data type to use.
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * Same as range_overflows() but forcing the parameters to @type.
+ *
+ * Returns: true if the range is out of bounds.
+ */
+#define range_overflows_t(type, start, size, max) \
+	range_overflows((type)(start), (type)(size), (type)(max))
+
+/**
+ * range_end_overflows() - Check if a range's endpoint is out of bounds
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * Checks only if the endpoint of a range (@start + @size) exceeds @max.
+ * Unlike range_overflows(), a zero-sized range at the boundary (@start == @max)
+ * is not considered an overflow. Useful for iterator-style checks.
+ *
+ * Returns: true if the endpoint exceeds the boundary.
+ */
+#define range_end_overflows(start, size, max) ({ \
+	typeof(start) start__ = (start); \
+	typeof(size) size__ = (size); \
+	typeof(max) max__ = (max); \
+	(void)(&start__ == &size__); \
+	(void)(&start__ == &max__); \
+	start__ > max__ || size__ > max__ - start__; \
+})
+
+/**
+ * range_end_overflows_t() - Check if a range's endpoint is out of bounds
+ * @type:  Data type to use.
+ * @start: Start of the range.
+ * @size:  Size of the range.
+ * @max:   Exclusive upper boundary.
+ *
+ * Same as range_end_overflows() but forcing the parameters to @type.
+ *
+ * Returns: true if the endpoint exceeds the boundary.
+ */
+#define range_end_overflows_t(type, start, size, max) \
+	range_end_overflows((type)(start), (type)(size), (type)(max))
+
 /**
  * castable_to_type - like __same_type(), but also allows for casted literals
  *

From 306c8959b5fdfdc3900aebe4e56257a349f5765f Mon Sep 17 00:00:00 2001
From: Leander Kieweg <kieweg.leander@gmail.com>
Date: Mon, 18 Aug 2025 13:35:29 +0200
Subject: [PATCH 1088/1455] drm/tiny/bochs: Convert dev_err() to drm_err()

The DRM subsystem has a set of preferred, prefixed logging functions
(drm_info, drm_warn, drm_err) which improve debuggability by including
the driver and function name in the log output.

As part of the ongoing effort to modernize logging calls,
convert a dev_err() call in the bochs hardware initialization
function to its drm_err() equivalent.

This work was suggested by the DRM TODO list.

Signed-off-by: Leander Kieweg <kieweg.leander@gmail.com>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250818113530.187440-1-kieweg.leander@gmail.com
---
 drivers/gpu/drm/tiny/bochs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/tiny/bochs.c b/drivers/gpu/drm/tiny/bochs.c
index 8d3b7c4fa6a4..d2d5e9f1269f 100644
--- a/drivers/gpu/drm/tiny/bochs.c
+++ b/drivers/gpu/drm/tiny/bochs.c
@@ -252,7 +252,7 @@ static int bochs_hw_init(struct bochs_device *bochs)
 		}
 		bochs->ioports = 1;
 	} else {
-		dev_err(dev->dev, "I/O ports are not supported\n");
+		drm_err(dev, "I/O ports are not supported\n");
 		return -EIO;
 	}
 

From 0584da4515dbb4fec69107ce837eef36a7be5d7d Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 12:49:53 +0530
Subject: [PATCH 1089/1455] drm/msm/adreno: Add speedbin data for A623 GPU

Add the speedbin mappings for Adreno 623 GPU.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672462/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index fcfe89eeccf2..4b3f18b2d2b9 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -913,6 +913,11 @@ static const struct adreno_info a6xx_gpus[] = {
 				{ /* sentinel */ },
 			},
 		},
+		.speedbins = ADRENO_SPEEDBINS(
+			{ 0,   0 },
+			{ 185, 0 },
+			{ 127, 1 },
+		),
 	}, {
 		.chip_ids = ADRENO_CHIP_IDS(
 			0x06030001,

From cc64568b522b64fb8f8c607da9eb4e2d9f72d0cf Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 13:51:49 +0300
Subject: [PATCH 1090/1455] drm/msm/mdp4: stop supporting no-IOMMU
 configuration

With the switch to GPUVM the msm driver no longer supports the no-IOMMU
configurations (even without the actual GPU). Return an error in case we
face the lack of the IOMMU for an MDP4 device.

Fixes: 111fdd2198e6 ("drm/msm: drm_gpuvm conversion")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672557/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
index 0952c7f18abd..4d1ea9b26191 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
@@ -463,9 +463,9 @@ static int mdp4_kms_init(struct drm_device *dev)
 		ret = PTR_ERR(mmu);
 		goto fail;
 	} else if (!mmu) {
-		DRM_DEV_INFO(dev->dev, "no iommu, fallback to phys "
-				"contig buffers for scanout\n");
-		vm = NULL;
+		DRM_DEV_INFO(dev->dev, "no IOMMU, bailing out\n");
+		ret = -ENODEV;
+		goto fail;
 	} else {
 		vm  = msm_gem_vm_create(dev, mmu, "mdp4",
 					0x1000, 0x100000000 - 0x1000,

From c94fc6d35685587aa0cb9a8d7d7062c73ab04d89 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 13:51:50 +0300
Subject: [PATCH 1091/1455] drm/msm: stop supporting no-IOMMU configuration

With the switch to GPUVM the msm driver no longer supports the no-IOMMU
configurations (even without the actual GPU). Return an error in case we
face the lack of the IOMMU.

Fixes: 111fdd2198e6 ("drm/msm: drm_gpuvm conversion")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672559/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/msm_kms.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/msm/msm_kms.c b/drivers/gpu/drm/msm/msm_kms.c
index 56828d218e88..4c4dcb095c4d 100644
--- a/drivers/gpu/drm/msm/msm_kms.c
+++ b/drivers/gpu/drm/msm/msm_kms.c
@@ -195,14 +195,13 @@ struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev)
 		iommu_dev = mdp_dev;
 	else
 		iommu_dev = mdss_dev;
-
 	mmu = msm_iommu_disp_new(iommu_dev, 0);
 	if (IS_ERR(mmu))
 		return ERR_CAST(mmu);
 
 	if (!mmu) {
-		drm_info(dev, "no IOMMU, fallback to phys contig buffers for scanout\n");
-		return NULL;
+		drm_info(dev, "no IOMMU, bailing out\n");
+		return ERR_PTR(-ENODEV);
 	}
 
 	vm = msm_gem_vm_create(dev, mmu, "mdp_kms",

From 618c11ea0b4a7c42eee9e5699d074c79be104f60 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 13:51:51 +0300
Subject: [PATCH 1092/1455] drm/msm: don't return NULL from msm_iommu_new()

As we've dropped no-IOMMU support, there is no reason to specially
handle the no-IOMMU case inside msm_iommu_new(). Make it return -ENODEV
if there is no IOMMU (instead of returning NULL) and simplify all
calling sites accordingly.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672561/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c    |  2 --
 drivers/gpu/drm/msm/adreno/adreno_gpu.c  |  4 +---
 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c | 10 ++++++----
 drivers/gpu/drm/msm/msm_iommu.c          |  6 +++---
 drivers/gpu/drm/msm/msm_kms.c            | 12 ++++++------
 5 files changed, 16 insertions(+), 18 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 28e6705c6da6..92958c73a5ee 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1318,8 +1318,6 @@ static int a6xx_gmu_memory_probe(struct drm_device *drm, struct a6xx_gmu *gmu)
 	struct msm_mmu *mmu;
 
 	mmu = msm_iommu_new(gmu->dev, 0);
-	if (!mmu)
-		return -ENODEV;
 	if (IS_ERR(mmu))
 		return PTR_ERR(mmu);
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 8c6336b007dc..afaa3cfefd35 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -202,9 +202,7 @@ adreno_iommu_create_vm(struct msm_gpu *gpu,
 	u64 start, size;
 
 	mmu = msm_iommu_gpu_new(&pdev->dev, gpu, quirks);
-	if (!mmu)
-		return ERR_PTR(-ENODEV);
-	else if (IS_ERR_OR_NULL(mmu))
+	if (IS_ERR(mmu))
 		return ERR_CAST(mmu);
 
 	geometry = msm_iommu_get_geometry(mmu);
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
index 4d1ea9b26191..66bfa96965be 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
@@ -458,14 +458,16 @@ static int mdp4_kms_init(struct drm_device *dev)
 	mdp4_disable(mdp4_kms);
 	mdelay(16);
 
+	if (!device_iommu_mapped(&pdev->dev)) {
+		DRM_DEV_INFO(dev->dev, "no IOMMU, bailing out\n");
+		ret = -ENODEV;
+		goto fail;
+	}
+
 	mmu = msm_iommu_new(&pdev->dev, 0);
 	if (IS_ERR(mmu)) {
 		ret = PTR_ERR(mmu);
 		goto fail;
-	} else if (!mmu) {
-		DRM_DEV_INFO(dev->dev, "no IOMMU, bailing out\n");
-		ret = -ENODEV;
-		goto fail;
 	} else {
 		vm  = msm_gem_vm_create(dev, mmu, "mdp4",
 					0x1000, 0x100000000 - 0x1000,
diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 89cdda73711a..0e18619f96cb 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -721,7 +721,7 @@ struct msm_mmu *msm_iommu_new(struct device *dev, unsigned long quirks)
 	int ret;
 
 	if (!device_iommu_mapped(dev))
-		return NULL;
+		return ERR_PTR(-ENODEV);
 
 	domain = iommu_paging_domain_alloc(dev);
 	if (IS_ERR(domain))
@@ -756,7 +756,7 @@ struct msm_mmu *msm_iommu_disp_new(struct device *dev, unsigned long quirks)
 	struct msm_mmu *mmu;
 
 	mmu = msm_iommu_new(dev, quirks);
-	if (IS_ERR_OR_NULL(mmu))
+	if (IS_ERR(mmu))
 		return mmu;
 
 	iommu = to_msm_iommu(mmu);
@@ -772,7 +772,7 @@ struct msm_mmu *msm_iommu_gpu_new(struct device *dev, struct msm_gpu *gpu, unsig
 	struct msm_mmu *mmu;
 
 	mmu = msm_iommu_new(dev, quirks);
-	if (IS_ERR_OR_NULL(mmu))
+	if (IS_ERR(mmu))
 		return mmu;
 
 	iommu = to_msm_iommu(mmu);
diff --git a/drivers/gpu/drm/msm/msm_kms.c b/drivers/gpu/drm/msm/msm_kms.c
index 4c4dcb095c4d..725367157d90 100644
--- a/drivers/gpu/drm/msm/msm_kms.c
+++ b/drivers/gpu/drm/msm/msm_kms.c
@@ -193,17 +193,17 @@ struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev)
 	 */
 	if (device_iommu_mapped(mdp_dev))
 		iommu_dev = mdp_dev;
-	else
+	else if (device_iommu_mapped(mdss_dev))
 		iommu_dev = mdss_dev;
-	mmu = msm_iommu_disp_new(iommu_dev, 0);
-	if (IS_ERR(mmu))
-		return ERR_CAST(mmu);
-
-	if (!mmu) {
+	else {
 		drm_info(dev, "no IOMMU, bailing out\n");
 		return ERR_PTR(-ENODEV);
 	}
 
+	mmu = msm_iommu_disp_new(iommu_dev, 0);
+	if (IS_ERR(mmu))
+		return ERR_CAST(mmu);
+
 	vm = msm_gem_vm_create(dev, mmu, "mdp_kms",
 			       0x1000, 0x100000000 - 0x1000, true);
 	if (IS_ERR(vm)) {

From e05edbcf3a82aa2549b96cfc470907d920a58b97 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 13:51:52 +0300
Subject: [PATCH 1093/1455] drm/msm/mdp4: use msm_kms_init_vm() instead of
 duplicating it

Use the msm_kms_init_vm() function to allocate memory manager instead of
hand-coding a copy of it. Although MDP4 platforms don't have MDSS
device, it's still safe to use the function as all MDP4 devices have
IOMMU and the parent of the MDP4 is the root SoC device.

Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672563/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c  |  2 +-
 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c | 27 ++++--------------------
 drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c |  2 +-
 drivers/gpu/drm/msm/msm_drv.h            |  2 +-
 drivers/gpu/drm/msm/msm_kms.c            |  5 ++---
 5 files changed, 9 insertions(+), 29 deletions(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
index a306077647c3..4e5a8ecd31f7 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c
@@ -1110,7 +1110,7 @@ static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms)
 {
 	struct drm_gpuvm *vm;
 
-	vm = msm_kms_init_vm(dpu_kms->dev);
+	vm = msm_kms_init_vm(dpu_kms->dev, dpu_kms->dev->dev->parent);
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
index 66bfa96965be..1d2a1e4f6ce7 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
@@ -391,11 +391,9 @@ static void read_mdp_hw_revision(struct mdp4_kms *mdp4_kms,
 
 static int mdp4_kms_init(struct drm_device *dev)
 {
-	struct platform_device *pdev = to_platform_device(dev->dev);
 	struct msm_drm_private *priv = dev->dev_private;
 	struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(priv->kms));
 	struct msm_kms *kms = NULL;
-	struct msm_mmu *mmu;
 	struct drm_gpuvm *vm;
 	int ret;
 	u32 major, minor;
@@ -458,30 +456,13 @@ static int mdp4_kms_init(struct drm_device *dev)
 	mdp4_disable(mdp4_kms);
 	mdelay(16);
 
-	if (!device_iommu_mapped(&pdev->dev)) {
-		DRM_DEV_INFO(dev->dev, "no IOMMU, bailing out\n");
-		ret = -ENODEV;
+	vm = msm_kms_init_vm(mdp4_kms->dev, NULL);
+	if (IS_ERR(vm)) {
+		ret = PTR_ERR(vm);
 		goto fail;
 	}
 
-	mmu = msm_iommu_new(&pdev->dev, 0);
-	if (IS_ERR(mmu)) {
-		ret = PTR_ERR(mmu);
-		goto fail;
-	} else {
-		vm  = msm_gem_vm_create(dev, mmu, "mdp4",
-					0x1000, 0x100000000 - 0x1000,
-					true);
-
-		if (IS_ERR(vm)) {
-			if (!IS_ERR(mmu))
-				mmu->funcs->destroy(mmu);
-			ret = PTR_ERR(vm);
-			goto fail;
-		}
-
-		kms->vm = vm;
-	}
+	kms->vm = vm;
 
 	ret = modeset_init(mdp4_kms);
 	if (ret) {
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c
index 5b6ca8dd929e..61edf6864092 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c
@@ -534,7 +534,7 @@ static int mdp5_kms_init(struct drm_device *dev)
 	}
 	mdelay(16);
 
-	vm = msm_kms_init_vm(mdp5_kms->dev);
+	vm = msm_kms_init_vm(mdp5_kms->dev, pdev->dev.parent);
 	if (IS_ERR(vm)) {
 		ret = PTR_ERR(vm);
 		goto fail;
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index 9875ca62e9ad..f4c20dc91cc6 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -229,7 +229,7 @@ void msm_crtc_disable_vblank(struct drm_crtc *crtc);
 int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu);
 void msm_unregister_mmu(struct drm_device *dev, struct msm_mmu *mmu);
 
-struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev);
+struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev, struct device *mdss_dev);
 bool msm_use_mmu(struct drm_device *dev);
 
 int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
diff --git a/drivers/gpu/drm/msm/msm_kms.c b/drivers/gpu/drm/msm/msm_kms.c
index 725367157d90..6e5e94f5c9a7 100644
--- a/drivers/gpu/drm/msm/msm_kms.c
+++ b/drivers/gpu/drm/msm/msm_kms.c
@@ -177,12 +177,11 @@ static int msm_kms_fault_handler(void *arg, unsigned long iova, int flags, void
 	return -ENOSYS;
 }
 
-struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev)
+struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev, struct device *mdss_dev)
 {
 	struct drm_gpuvm *vm;
 	struct msm_mmu *mmu;
 	struct device *mdp_dev = dev->dev;
-	struct device *mdss_dev = mdp_dev->parent;
 	struct msm_drm_private *priv = dev->dev_private;
 	struct msm_kms *kms = priv->kms;
 	struct device *iommu_dev;
@@ -193,7 +192,7 @@ struct drm_gpuvm *msm_kms_init_vm(struct drm_device *dev)
 	 */
 	if (device_iommu_mapped(mdp_dev))
 		iommu_dev = mdp_dev;
-	else if (device_iommu_mapped(mdss_dev))
+	else if (mdss_dev && device_iommu_mapped(mdss_dev))
 		iommu_dev = mdss_dev;
 	else {
 		drm_info(dev, "no IOMMU, bailing out\n");

From f1f514d730f62fffd8fc160e9fb6df719c7425a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Barnab=C3=A1s=20Cz=C3=A9m=C3=A1n?=
 <barnabas.czeman@mainlining.org>
Date: Wed, 3 Sep 2025 23:08:24 +0200
Subject: [PATCH 1094/1455] dt-bindings: display/msm/gpu: describe A505 clocks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Descirbe A505 clocks it is using same clocks like A506.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Barnabás Czémán <barnabas.czeman@mainlining.org>
Patchwork: https://patchwork.freedesktop.org/patch/672751/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 Documentation/devicetree/bindings/display/msm/gpu.yaml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/display/msm/gpu.yaml b/Documentation/devicetree/bindings/display/msm/gpu.yaml
index 7ef80f9fac8b..3696b083e353 100644
--- a/Documentation/devicetree/bindings/display/msm/gpu.yaml
+++ b/Documentation/devicetree/bindings/display/msm/gpu.yaml
@@ -252,7 +252,7 @@ allOf:
       properties:
         compatible:
           contains:
-            pattern: '^qcom,adreno-506\.[0-9]+$'
+            pattern: '^qcom,adreno-50[56]\.[0-9]+$'
     then:
       properties:
         clocks:

From f23e09a60d48195345aabdf007f9c372ed05163f Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:56:54 +0530
Subject: [PATCH 1095/1455] drm/msm: Update GMU register xml

Update GMU register xml with additional definitions for a7x family.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673356/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml
index 3d2cc339b8f1..b15a242d974d 100644
--- a/drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/a6xx_gmu.xml
@@ -99,6 +99,10 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 		<bitfield name="GX_HM_GDSC_POWER_OFF" pos="6" type="boolean"/>
 		<bitfield name="GX_HM_CLK_OFF" pos="7" type="boolean"/>
 	</reg32>
+	<reg32 offset="0x50d0" name="GMU_SPTPRAC_PWR_CLK_STATUS" variants="A7XX">
+		<bitfield name="GX_HM_GDSC_POWER_OFF" pos="0" type="boolean"/>
+		<bitfield name="GX_HM_CLK_OFF" pos="1" type="boolean"/>
+	</reg32>
 	<reg32 offset="0x50e4" name="GMU_GPU_NAP_CTRL">
 		<bitfield name="HW_NAP_ENABLE" pos="0"/>
 		<bitfield name="SID" low="4" high="8"/>
@@ -127,6 +131,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 	<reg32 offset="0x5088" name="GMU_ALWAYS_ON_COUNTER_L"/>
 	<reg32 offset="0x5089" name="GMU_ALWAYS_ON_COUNTER_H"/>
 	<reg32 offset="0x50c3" name="GMU_GMU_PWR_COL_KEEPALIVE"/>
+	<reg32 offset="0x50c4" name="GMU_PWR_COL_PREEMPT_KEEPALIVE"/>
 	<reg32 offset="0x5180" name="GMU_HFI_CTRL_STATUS"/>
 	<reg32 offset="0x5181" name="GMU_HFI_VERSION_INFO"/>
 	<reg32 offset="0x5182" name="GMU_HFI_SFR_ADDR"/>
@@ -228,6 +233,12 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 	<reg32 offset="0x03ee" name="RSCC_TCS1_DRV0_STATUS"/>
 	<reg32 offset="0x0496" name="RSCC_TCS2_DRV0_STATUS"/>
 	<reg32 offset="0x053e" name="RSCC_TCS3_DRV0_STATUS"/>
+	<reg32 offset="0x05e6" name="RSCC_TCS4_DRV0_STATUS" variants="A7XX"/>
+	<reg32 offset="0x068e" name="RSCC_TCS5_DRV0_STATUS" variants="A7XX"/>
+	<reg32 offset="0x0736" name="RSCC_TCS6_DRV0_STATUS" variants="A7XX"/>
+	<reg32 offset="0x07de" name="RSCC_TCS7_DRV0_STATUS" variants="A7XX"/>
+	<reg32 offset="0x0886" name="RSCC_TCS8_DRV0_STATUS" variants="A7XX"/>
+	<reg32 offset="0x092e" name="RSCC_TCS9_DRV0_STATUS" variants="A7XX"/>
 </domain>
 
 </database>

From 6bf8186d5d551430fdd1c6d5a4a412b30893856f Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:56:55 +0530
Subject: [PATCH 1096/1455] drm/msm: a6xx: Fix gx_is_on check for a7x family

Bitfield definition for REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS register is
different in A7XX family. Check the correct bits to see if GX is
collapsed on A7XX series.

Fixes: af66706accdf ("drm/msm/a6xx: Add skeleton A7xx support")
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673358/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 92958c73a5ee..3f8c43255b6e 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -93,6 +93,8 @@ bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu)
 /* Check to see if the GX rail is still powered */
 bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu)
 {
+	struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu);
+	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
 	u32 val;
 
 	/* This can be called from gpu state code so make sure GMU is valid */
@@ -101,6 +103,11 @@ bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu)
 
 	val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS);
 
+	if (adreno_is_a7xx(adreno_gpu))
+		return !(val &
+			(A7XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF |
+			A7XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_CLK_OFF));
+
 	return !(val &
 		(A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF |
 		A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_CLK_OFF));

From ea601501a0afac378d84e109c0e880038f3d6dec Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:56:56 +0530
Subject: [PATCH 1097/1455] drm/msm/a6xx: Poll additional DRV status

A7XX_GEN2 generation has additional TCS slots. Poll the respective
DRV status registers before pm suspend.

Fixes: 1f8c29e80066 ("drm/msm/a6xx: Add A740 support")
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673361/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 3f8c43255b6e..e74651d5bd7a 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -987,6 +987,22 @@ static void a6xx_gmu_rpmh_off(struct a6xx_gmu *gmu)
 		val, (val & 1), 100, 10000);
 	gmu_poll_timeout_rscc(gmu, REG_A6XX_RSCC_TCS3_DRV0_STATUS + seqmem_off,
 		val, (val & 1), 100, 1000);
+
+	if (!adreno_is_a740_family(adreno_gpu))
+		return;
+
+	gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS4_DRV0_STATUS + seqmem_off,
+		val, (val & 1), 100, 10000);
+	gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS5_DRV0_STATUS + seqmem_off,
+		val, (val & 1), 100, 10000);
+	gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS6_DRV0_STATUS + seqmem_off,
+		val, (val & 1), 100, 10000);
+	gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS7_DRV0_STATUS + seqmem_off,
+		val, (val & 1), 100, 1000);
+	gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS8_DRV0_STATUS + seqmem_off,
+		val, (val & 1), 100, 10000);
+	gmu_poll_timeout_rscc(gmu, REG_A7XX_RSCC_TCS9_DRV0_STATUS + seqmem_off,
+		val, (val & 1), 100, 1000);
 }
 
 /* Force the GMU off in case it isn't responsive */

From f248d5d5159a88ded55329f0b1b463d0f4094228 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:56:57 +0530
Subject: [PATCH 1098/1455] drm/msm/a6xx: Fix PDC sleep sequence

Since the PDC resides out of the GPU subsystem and cannot be reset in
case it enters bad state, utmost care must be taken to trigger the PDC
wake/sleep routines in the correct order.

The PDC wake sequence can be exercised only after a PDC sleep sequence.
Additionally, GMU firmware should initialize a few registers before the
KMD can trigger a PDC sleep sequence. So PDC sleep can't be done if the
GMU firmware has not initialized. Track these dependencies using a new
status variable and trigger PDC sleep/wake sequences appropriately.

Cc: stable@vger.kernel.org
Fixes: 4b565ca5a2cb ("drm/msm: Add A6XX device support")
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673362/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 28 ++++++++++++++++-----------
 drivers/gpu/drm/msm/adreno/a6xx_gmu.h |  6 ++++++
 2 files changed, 23 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index e74651d5bd7a..5594499ad2d1 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -279,6 +279,8 @@ static int a6xx_gmu_start(struct a6xx_gmu *gmu)
 	if (ret)
 		DRM_DEV_ERROR(gmu->dev, "GMU firmware initialization timed out\n");
 
+	set_bit(GMU_STATUS_FW_START, &gmu->status);
+
 	return ret;
 }
 
@@ -525,6 +527,9 @@ static int a6xx_rpmh_start(struct a6xx_gmu *gmu)
 	int ret;
 	u32 val;
 
+	if (!test_and_clear_bit(GMU_STATUS_PDC_SLEEP, &gmu->status))
+		return 0;
+
 	gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, BIT(1));
 
 	ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_RSCC_CONTROL_ACK, val,
@@ -552,6 +557,9 @@ static void a6xx_rpmh_stop(struct a6xx_gmu *gmu)
 	int ret;
 	u32 val;
 
+	if (test_and_clear_bit(GMU_STATUS_FW_START, &gmu->status))
+		return;
+
 	gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 1);
 
 	ret = gmu_poll_timeout_rscc(gmu, REG_A6XX_GPU_RSCC_RSC_STATUS0_DRV0,
@@ -560,6 +568,8 @@ static void a6xx_rpmh_stop(struct a6xx_gmu *gmu)
 		DRM_DEV_ERROR(gmu->dev, "Unable to power off the GPU RSC\n");
 
 	gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0);
+
+	set_bit(GMU_STATUS_PDC_SLEEP, &gmu->status);
 }
 
 static inline void pdc_write(void __iomem *ptr, u32 offset, u32 value)
@@ -688,8 +698,6 @@ setup_pdc:
 	/* ensure no writes happen before the uCode is fully written */
 	wmb();
 
-	a6xx_rpmh_stop(gmu);
-
 err:
 	if (!IS_ERR_OR_NULL(pdcptr))
 		iounmap(pdcptr);
@@ -849,19 +857,15 @@ static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state)
 	else
 		gmu_write(gmu, REG_A6XX_GMU_GENERAL_7, 1);
 
-	if (state == GMU_WARM_BOOT) {
-		ret = a6xx_rpmh_start(gmu);
-		if (ret)
-			return ret;
-	} else {
+	ret = a6xx_rpmh_start(gmu);
+	if (ret)
+		return ret;
+
+	if (state == GMU_COLD_BOOT) {
 		if (WARN(!adreno_gpu->fw[ADRENO_FW_GMU],
 			"GMU firmware is not loaded\n"))
 			return -ENOENT;
 
-		ret = a6xx_rpmh_start(gmu);
-		if (ret)
-			return ret;
-
 		ret = a6xx_gmu_fw_load(gmu);
 		if (ret)
 			return ret;
@@ -1046,6 +1050,8 @@ static void a6xx_gmu_force_off(struct a6xx_gmu *gmu)
 
 	/* Reset GPU core blocks */
 	a6xx_gpu_sw_reset(gpu, true);
+
+	a6xx_rpmh_stop(gmu);
 }
 
 static void a6xx_gmu_set_initial_freq(struct msm_gpu *gpu, struct a6xx_gmu *gmu)
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index d1ce11131ba6..069a8c9474e8 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -117,6 +117,12 @@ struct a6xx_gmu {
 
 	struct qmp *qmp;
 	struct a6xx_hfi_msg_bw_table *bw_table;
+
+/* To check if we can trigger sleep seq at PDC. Cleared in a6xx_rpmh_stop() */
+#define GMU_STATUS_FW_START	0
+/* To track if PDC sleep seq was done */
+#define GMU_STATUS_PDC_SLEEP	1
+	unsigned long status;
 };
 
 static inline u32 gmu_read(struct a6xx_gmu *gmu, u32 offset)

From a477aa67e0a3dafe90da70133272e22cd30cf65f Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:56:58 +0530
Subject: [PATCH 1099/1455] drm/msm: a6xx: Refactor a6xx_sptprac_enable()

A minor refactor to combine the subroutines for legacy a6xx GMUs under
a single check. This helps to avoid an unnecessary check and return
early from the subroutine for majority of a6xx gpus.

Also, document an intermediate unknown low power state which is not
exposed by the GMU firmware.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673364/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 8 ++++----
 drivers/gpu/drm/msm/adreno/a6xx_gmu.h | 3 +++
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 5594499ad2d1..da339459ae4b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -412,7 +412,10 @@ int a6xx_sptprac_enable(struct a6xx_gmu *gmu)
 	int ret;
 	u32 val;
 
-	if (!gmu->legacy)
+	WARN_ON(!gmu->legacy);
+
+	/* Nothing to do if GMU does the power management */
+	if (gmu->idle_level > GMU_IDLE_STATE_ACTIVE)
 		return 0;
 
 	gmu_write(gmu, REG_A6XX_GMU_GX_SPTPRAC_POWER_CONTROL, 0x778000);
@@ -936,10 +939,7 @@ static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state)
 		ret = a6xx_gmu_gfx_rail_on(gmu);
 		if (ret)
 			return ret;
-	}
 
-	/* Enable SPTP_PC if the CPU is responsible for it */
-	if (gmu->idle_level < GMU_IDLE_STATE_SPTP) {
 		ret = a6xx_sptprac_enable(gmu);
 		if (ret)
 			return ret;
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index 069a8c9474e8..9494bbed9a1f 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -50,6 +50,9 @@ struct a6xx_bcm {
 /* The GMU does not do any idle state management */
 #define GMU_IDLE_STATE_ACTIVE 0
 
+/* Unknown power state. Not exposed by the firmware. For documentation purpose only */
+#define GMU_IDLE_STATE_RESERVED 1
+
 /* The GMU manages SPTP power collapse */
 #define GMU_IDLE_STATE_SPTP 2
 

From ac9098b1794bc4db463241db7ed090a11dcaa541 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:56:59 +0530
Subject: [PATCH 1100/1455] drm/msm: Add an ftrace for gpu register access

With IFPC, there is a probability of accessing a GX domain register when
it is collapsed, which leads to gmu fence errors. To debug this, we need
to trace every gpu register accesses and identify the one just before a
gmu fence error. So, add an ftrace to track all gpu register accesses.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673366/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/msm_gpu.h       |  8 ++++++++
 drivers/gpu/drm/msm/msm_gpu_trace.h | 12 ++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index b2a96544f92a..866bb090af06 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -16,6 +16,7 @@
 
 #include "msm_drv.h"
 #include "msm_fence.h"
+#include "msm_gpu_trace.h"
 #include "msm_ringbuffer.h"
 #include "msm_gem.h"
 
@@ -613,16 +614,19 @@ struct msm_gpu_state {
 
 static inline void gpu_write(struct msm_gpu *gpu, u32 reg, u32 data)
 {
+	trace_msm_gpu_regaccess(reg);
 	writel(data, gpu->mmio + (reg << 2));
 }
 
 static inline u32 gpu_read(struct msm_gpu *gpu, u32 reg)
 {
+	trace_msm_gpu_regaccess(reg);
 	return readl(gpu->mmio + (reg << 2));
 }
 
 static inline void gpu_rmw(struct msm_gpu *gpu, u32 reg, u32 mask, u32 or)
 {
+	trace_msm_gpu_regaccess(reg);
 	msm_rmw(gpu->mmio + (reg << 2), mask, or);
 }
 
@@ -644,7 +648,9 @@ static inline u64 gpu_read64(struct msm_gpu *gpu, u32 reg)
 	 * when the lo is read, so make sure to read the lo first to trigger
 	 * that
 	 */
+	trace_msm_gpu_regaccess(reg);
 	val = (u64) readl(gpu->mmio + (reg << 2));
+	trace_msm_gpu_regaccess(reg+1);
 	val |= ((u64) readl(gpu->mmio + ((reg + 1) << 2)) << 32);
 
 	return val;
@@ -652,8 +658,10 @@ static inline u64 gpu_read64(struct msm_gpu *gpu, u32 reg)
 
 static inline void gpu_write64(struct msm_gpu *gpu, u32 reg, u64 val)
 {
+	trace_msm_gpu_regaccess(reg);
 	/* Why not a writeq here? Read the screed above */
 	writel(lower_32_bits(val), gpu->mmio + (reg << 2));
+	trace_msm_gpu_regaccess(reg+1);
 	writel(upper_32_bits(val), gpu->mmio + ((reg + 1) << 2));
 }
 
diff --git a/drivers/gpu/drm/msm/msm_gpu_trace.h b/drivers/gpu/drm/msm/msm_gpu_trace.h
index 781bbe5540bd..5417f8d389a3 100644
--- a/drivers/gpu/drm/msm/msm_gpu_trace.h
+++ b/drivers/gpu/drm/msm/msm_gpu_trace.h
@@ -219,6 +219,18 @@ TRACE_EVENT(msm_mmu_prealloc_cleanup,
 		TP_printk("count=%u, remaining=%u", __entry->count, __entry->remaining)
 );
 
+TRACE_EVENT(msm_gpu_regaccess,
+		TP_PROTO(u32 offset),
+		TP_ARGS(offset),
+		TP_STRUCT__entry(
+			__field(u32, offset)
+			),
+		TP_fast_assign(
+			__entry->offset = offset;
+			),
+		TP_printk("offset=0x%x", __entry->offset)
+);
+
 #endif
 
 #undef TRACE_INCLUDE_PATH

From a27d774045566b587bfc1ae9fb122642b06677b8 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:00 +0530
Subject: [PATCH 1101/1455] drm/msm/adreno: Add fenced regwrite support

There are some special registers which are accessible even when GX power
domain is collapsed during an IFPC sleep. Accessing these registers
wakes up GPU from power collapse and allow programming these registers
without additional handshake with GMU. This patch adds support for this
special register write sequence.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673368/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c     | 80 ++++++++++++++++++++++-
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h     |  1 +
 drivers/gpu/drm/msm/adreno/a6xx_preempt.c | 20 +++---
 3 files changed, 90 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index f8992a68df7f..536da1acf615 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -16,6 +16,84 @@
 
 #define GPU_PAS_ID 13
 
+static bool fence_status_check(struct msm_gpu *gpu, u32 offset, u32 value, u32 status, u32 mask)
+{
+	/* Success if !writedropped0/1 */
+	if (!(status & mask))
+		return true;
+
+	udelay(10);
+
+	/* Try to update fenced register again */
+	gpu_write(gpu, offset, value);
+
+	/* We can't do a posted write here because the power domain could be
+	 * in collapse state. So use the heaviest barrier instead
+	 */
+	mb();
+	return false;
+}
+
+static int fenced_write(struct a6xx_gpu *a6xx_gpu, u32 offset, u32 value, u32 mask)
+{
+	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
+	struct msm_gpu *gpu = &adreno_gpu->base;
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	u32 status;
+
+	gpu_write(gpu, offset, value);
+
+	/* Nothing else to be done in the case of no-GMU */
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		return 0;
+
+	/* We can't do a posted write here because the power domain could be
+	 * in collapse state. So use the heaviest barrier instead
+	 */
+	mb();
+
+	if (!gmu_poll_timeout(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS, status,
+			fence_status_check(gpu, offset, value, status, mask), 0, 1000))
+		return 0;
+
+	/* Try again for another 1ms before failing */
+	gpu_write(gpu, offset, value);
+	mb();
+
+	if (!gmu_poll_timeout(gmu, REG_A6XX_GMU_AHB_FENCE_STATUS, status,
+			fence_status_check(gpu, offset, value, status, mask), 0, 1000)) {
+		/*
+		 * The 'delay' warning is here because the pause to print this
+		 * warning will allow gpu to move to power collapse which
+		 * defeats the purpose of continuous polling for 2 ms
+		 */
+		dev_err_ratelimited(gmu->dev, "delay in fenced register write (0x%x)\n",
+				offset);
+		return 0;
+	}
+
+	dev_err_ratelimited(gmu->dev, "fenced register write (0x%x) fail\n",
+			offset);
+
+	return -ETIMEDOUT;
+}
+
+int a6xx_fenced_write(struct a6xx_gpu *a6xx_gpu, u32 offset, u64 value, u32 mask, bool is_64b)
+{
+	int ret;
+
+	ret = fenced_write(a6xx_gpu, offset, lower_32_bits(value), mask);
+	if (ret)
+		return ret;
+
+	if (!is_64b)
+		return 0;
+
+	ret = fenced_write(a6xx_gpu, offset + 1, upper_32_bits(value), mask);
+
+	return ret;
+}
+
 static inline bool _a6xx_check_idle(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -86,7 +164,7 @@ static void a6xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 	/* Update HW if this is the current ring and we are not in preempt*/
 	if (!a6xx_in_preempt(a6xx_gpu)) {
 		if (a6xx_gpu->cur_ring == ring)
-			gpu_write(gpu, REG_A6XX_CP_RB_WPTR, wptr);
+			a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_RB_WPTR, wptr, BIT(0), false);
 		else
 			ring->restore_wptr = true;
 	} else {
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index 6e71f617fc3d..e736c59d566b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -295,5 +295,6 @@ int a6xx_gpu_state_put(struct msm_gpu_state *state);
 
 void a6xx_bus_clear_pending_transactions(struct adreno_gpu *adreno_gpu, bool gx_off);
 void a6xx_gpu_sw_reset(struct msm_gpu *gpu, bool assert);
+int a6xx_fenced_write(struct a6xx_gpu *gpu, u32 offset, u64 value, u32 mask, bool is_64b);
 
 #endif /* __A6XX_GPU_H__ */
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
index 6a12a35dabff..10625ffbc4cf 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
@@ -41,7 +41,7 @@ static inline void set_preempt_state(struct a6xx_gpu *gpu,
 }
 
 /* Write the most recent wptr for the given ring into the hardware */
-static inline void update_wptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
+static inline void update_wptr(struct a6xx_gpu *a6xx_gpu, struct msm_ringbuffer *ring)
 {
 	unsigned long flags;
 	uint32_t wptr;
@@ -51,7 +51,7 @@ static inline void update_wptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 	if (ring->restore_wptr) {
 		wptr = get_wptr(ring);
 
-		gpu_write(gpu, REG_A6XX_CP_RB_WPTR, wptr);
+		a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_RB_WPTR, wptr, BIT(0), false);
 
 		ring->restore_wptr = false;
 	}
@@ -172,7 +172,7 @@ void a6xx_preempt_irq(struct msm_gpu *gpu)
 
 	set_preempt_state(a6xx_gpu, PREEMPT_FINISH);
 
-	update_wptr(gpu, a6xx_gpu->cur_ring);
+	update_wptr(a6xx_gpu, a6xx_gpu->cur_ring);
 
 	set_preempt_state(a6xx_gpu, PREEMPT_NONE);
 
@@ -268,7 +268,7 @@ void a6xx_preempt_trigger(struct msm_gpu *gpu)
 	 */
 	if (!ring || (a6xx_gpu->cur_ring == ring)) {
 		set_preempt_state(a6xx_gpu, PREEMPT_FINISH);
-		update_wptr(gpu, a6xx_gpu->cur_ring);
+		update_wptr(a6xx_gpu, a6xx_gpu->cur_ring);
 		set_preempt_state(a6xx_gpu, PREEMPT_NONE);
 		spin_unlock_irqrestore(&a6xx_gpu->eval_lock, flags);
 		return;
@@ -302,13 +302,13 @@ void a6xx_preempt_trigger(struct msm_gpu *gpu)
 
 	spin_unlock_irqrestore(&ring->preempt_lock, flags);
 
-	gpu_write64(gpu,
-		REG_A6XX_CP_CONTEXT_SWITCH_SMMU_INFO,
-		a6xx_gpu->preempt_smmu_iova[ring->id]);
+	a6xx_fenced_write(a6xx_gpu,
+		REG_A6XX_CP_CONTEXT_SWITCH_SMMU_INFO, a6xx_gpu->preempt_smmu_iova[ring->id],
+		BIT(1), true);
 
-	gpu_write64(gpu,
+	a6xx_fenced_write(a6xx_gpu,
 		REG_A6XX_CP_CONTEXT_SWITCH_PRIV_NON_SECURE_RESTORE_ADDR,
-		a6xx_gpu->preempt_iova[ring->id]);
+		a6xx_gpu->preempt_iova[ring->id], BIT(1), true);
 
 	a6xx_gpu->next_ring = ring;
 
@@ -328,7 +328,7 @@ void a6xx_preempt_trigger(struct msm_gpu *gpu)
 	set_preempt_state(a6xx_gpu, PREEMPT_TRIGGERED);
 
 	/* Trigger the preemption */
-	gpu_write(gpu, REG_A6XX_CP_CONTEXT_SWITCH_CNTL, cntl);
+	a6xx_fenced_write(a6xx_gpu, REG_A6XX_CP_CONTEXT_SWITCH_CNTL, cntl, BIT(1), false);
 }
 
 static int preempt_init_ring(struct a6xx_gpu *a6xx_gpu,

From 0e7107a84d391bfab72d7eb96802920b662812c9 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:01 +0530
Subject: [PATCH 1102/1455] drm/msm/a6xx: Set Keep-alive votes to block IFPC

Set Keepalive votes at appropriate places to block IFPC power collapse
until we access all the required registers. This is required during gpu
IRQ handling and also during preemption.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673369/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c     | 26 +++++++++++++++--------
 drivers/gpu/drm/msm/adreno/a6xx_preempt.c | 20 +++++++++++++++++
 2 files changed, 37 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 536da1acf615..2a2b29949e86 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1771,8 +1771,6 @@ static void a6xx_cp_hw_err_irq(struct msm_gpu *gpu)
 
 static void a6xx_fault_detect_irq(struct msm_gpu *gpu)
 {
-	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
-	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 	struct msm_ringbuffer *ring = gpu->funcs->active_ring(gpu);
 
 	/*
@@ -1784,13 +1782,6 @@ static void a6xx_fault_detect_irq(struct msm_gpu *gpu)
 	if (gpu_read(gpu, REG_A6XX_RBBM_STATUS3) & A6XX_RBBM_STATUS3_SMMU_STALLED_ON_FAULT)
 		return;
 
-	/*
-	 * Force the GPU to stay on until after we finish
-	 * collecting information
-	 */
-	if (!adreno_has_gmu_wrapper(adreno_gpu))
-		gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, 1);
-
 	DRM_DEV_ERROR(&gpu->pdev->dev,
 		"gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n",
 		ring ? ring->id : -1, ring ? ring->fctx->last_fence : 0,
@@ -1832,9 +1823,24 @@ static void a7xx_sw_fuse_violation_irq(struct msm_gpu *gpu)
 	}
 }
 
+static void a6xx_gpu_keepalive_vote(struct msm_gpu *gpu, bool on)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		return;
+
+	gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, on);
+}
+
 static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 {
 	struct msm_drm_private *priv = gpu->dev->dev_private;
+
+	/* Set keepalive vote to avoid power collapse after RBBM_INT_0_STATUS is read */
+	a6xx_gpu_keepalive_vote(gpu, true);
+
 	u32 status = gpu_read(gpu, REG_A6XX_RBBM_INT_0_STATUS);
 
 	gpu_write(gpu, REG_A6XX_RBBM_INT_CLEAR_CMD, status);
@@ -1871,6 +1877,8 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 	if (status & A6XX_RBBM_INT_0_MASK_CP_SW)
 		a6xx_preempt_irq(gpu);
 
+	a6xx_gpu_keepalive_vote(gpu, false);
+
 	return IRQ_HANDLED;
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
index 10625ffbc4cf..2ce073d1cf8c 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
@@ -136,6 +136,21 @@ static void preempt_disable_postamble(struct a6xx_gpu *a6xx_gpu)
 	a6xx_gpu->postamble_enabled = false;
 }
 
+/*
+ * Set preemption keepalive vote. Please note that this vote is different from the one used in
+ * a6xx_irq()
+ */
+static void a6xx_preempt_keepalive_vote(struct msm_gpu *gpu, bool on)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		return;
+
+	gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_PWR_COL_PREEMPT_KEEPALIVE, on);
+}
+
 void a6xx_preempt_irq(struct msm_gpu *gpu)
 {
 	uint32_t status;
@@ -176,6 +191,8 @@ void a6xx_preempt_irq(struct msm_gpu *gpu)
 
 	set_preempt_state(a6xx_gpu, PREEMPT_NONE);
 
+	a6xx_preempt_keepalive_vote(gpu, false);
+
 	trace_msm_gpu_preemption_irq(a6xx_gpu->cur_ring->id);
 
 	/*
@@ -302,6 +319,9 @@ void a6xx_preempt_trigger(struct msm_gpu *gpu)
 
 	spin_unlock_irqrestore(&ring->preempt_lock, flags);
 
+	/* Set the keepalive bit to keep the GPU ON until preemption is complete */
+	a6xx_preempt_keepalive_vote(gpu, true);
+
 	a6xx_fenced_write(a6xx_gpu,
 		REG_A6XX_CP_CONTEXT_SWITCH_SMMU_INFO, a6xx_gpu->preempt_smmu_iova[ring->id],
 		BIT(1), true);

From f195421318bd00151b3a111af6f315a25c3438a8 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:02 +0530
Subject: [PATCH 1103/1455] drm/msm/a6xx: Switch to GMU AO counter

CP_ALWAYS_ON counter falls under GX domain which is collapsed during
IFPC. So switch to GMU_ALWAYS_ON counter for any CPU reads since it is
not impacted by IFPC. Both counters are clocked by same xo clock source.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673373/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 30 ++++++++++++++-------------
 1 file changed, 16 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 2a2b29949e86..48aa07b6b2f5 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -16,6 +16,19 @@
 
 #define GPU_PAS_ID 13
 
+static u64 read_gmu_ao_counter(struct a6xx_gpu *a6xx_gpu)
+{
+	u64 count_hi, count_lo, temp;
+
+	do {
+		count_hi = gmu_read(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_H);
+		count_lo = gmu_read(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_L);
+		temp = gmu_read(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_H);
+	} while (unlikely(count_hi != temp));
+
+	return (count_hi << 32) | count_lo;
+}
+
 static bool fence_status_check(struct msm_gpu *gpu, u32 offset, u32 value, u32 status, u32 mask)
 {
 	/* Success if !writedropped0/1 */
@@ -376,8 +389,7 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 	OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence)));
 	OUT_RING(ring, submit->seqno);
 
-	trace_msm_gpu_submit_flush(submit,
-		gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER));
+	trace_msm_gpu_submit_flush(submit, read_gmu_ao_counter(a6xx_gpu));
 
 	a6xx_flush(gpu, ring);
 }
@@ -577,8 +589,7 @@ static void a7xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit)
 	}
 
 
-	trace_msm_gpu_submit_flush(submit,
-		gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER));
+	trace_msm_gpu_submit_flush(submit, read_gmu_ao_counter(a6xx_gpu));
 
 	a6xx_flush(gpu, ring);
 
@@ -2268,16 +2279,7 @@ static int a6xx_gmu_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 
-	mutex_lock(&a6xx_gpu->gmu.lock);
-
-	/* Force the GPU power on so we can read this register */
-	a6xx_gmu_set_oob(&a6xx_gpu->gmu, GMU_OOB_PERFCOUNTER_SET);
-
-	*value = gpu_read64(gpu, REG_A6XX_CP_ALWAYS_ON_COUNTER);
-
-	a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_PERFCOUNTER_SET);
-
-	mutex_unlock(&a6xx_gpu->gmu.lock);
+	*value = read_gmu_ao_counter(a6xx_gpu);
 
 	return 0;
 }

From 04ca842ef2473120a9587dfe49523c8c243159ea Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:03 +0530
Subject: [PATCH 1104/1455] drm/msm/a6xx: Poll AHB fence status in GPU IRQ
 handler

Even though the GX power domain is kept ON when there is a pending GPU
interrupt, there is a small window of potential race with GMU where it
may move the AHB fence to 'Drop' mode. Once the GMU sees the pending IRQ,
it will move back the fence state to ALLOW mode. Close this race window
by polling for AHB fence to ensure that it is in 'Allow' mode.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673377/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.h |  3 +++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 26 ++++++++++++++++++++++++++
 2 files changed, 29 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index 9494bbed9a1f..40c4c0445b8e 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -167,6 +167,9 @@ static inline u64 gmu_read64(struct a6xx_gmu *gmu, u32 lo, u32 hi)
 #define gmu_poll_timeout(gmu, addr, val, cond, interval, timeout) \
 	readl_poll_timeout((gmu)->mmio + ((addr) << 2), val, cond, \
 		interval, timeout)
+#define gmu_poll_timeout_atomic(gmu, addr, val, cond, interval, timeout) \
+	readl_poll_timeout_atomic((gmu)->mmio + ((addr) << 2), val, cond, \
+		interval, timeout)
 
 static inline u32 gmu_read_rscc(struct a6xx_gmu *gmu, u32 offset)
 {
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 48aa07b6b2f5..4cdbad05affc 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1845,6 +1845,28 @@ static void a6xx_gpu_keepalive_vote(struct msm_gpu *gpu, bool on)
 	gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_GMU_PWR_COL_KEEPALIVE, on);
 }
 
+static int irq_poll_fence(struct msm_gpu *gpu)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	u32 status;
+
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		return 0;
+
+	if (gmu_poll_timeout_atomic(gmu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, status, !status, 1, 100)) {
+		u32 rbbm_unmasked = gmu_read(gmu, REG_A6XX_GMU_RBBM_INT_UNMASKED_STATUS);
+
+		dev_err_ratelimited(&gpu->pdev->dev,
+				"irq fence poll timeout, fence_ctrl=0x%x, unmasked_status=0x%x\n",
+				status, rbbm_unmasked);
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
 static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 {
 	struct msm_drm_private *priv = gpu->dev->dev_private;
@@ -1852,6 +1874,9 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 	/* Set keepalive vote to avoid power collapse after RBBM_INT_0_STATUS is read */
 	a6xx_gpu_keepalive_vote(gpu, true);
 
+	if (irq_poll_fence(gpu))
+		goto done;
+
 	u32 status = gpu_read(gpu, REG_A6XX_RBBM_INT_0_STATUS);
 
 	gpu_write(gpu, REG_A6XX_RBBM_INT_CLEAR_CMD, status);
@@ -1888,6 +1913,7 @@ static irqreturn_t a6xx_irq(struct msm_gpu *gpu)
 	if (status & A6XX_RBBM_INT_0_MASK_CP_SW)
 		a6xx_preempt_irq(gpu);
 
+done:
 	a6xx_gpu_keepalive_vote(gpu, false);
 
 	return IRQ_HANDLED;

From 365075b7d0b95c737284d6970769e964c7e7052d Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:04 +0530
Subject: [PATCH 1105/1455] drm/msm: Add support for IFPC

Add a new quirk to denote IFPC (Inter-Frame Power Collapse) support
for a gpu. Based on this flag send the feature ctrl hfi message to
GMU to enable IFPC support.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673375/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c   |  5 ++--
 drivers/gpu/drm/msm/adreno/a6xx_hfi.c   | 34 ++++++++++++++++++++-----
 drivers/gpu/drm/msm/adreno/adreno_gpu.h |  1 +
 3 files changed, 32 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index da339459ae4b..5e147be7dc0f 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1959,8 +1959,9 @@ int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node)
 	if (ret)
 		return ret;
 
-	/* Fow now, don't do anything fancy until we get our feet under us */
-	gmu->idle_level = GMU_IDLE_STATE_ACTIVE;
+	/* Set GMU idle level */
+	gmu->idle_level = (adreno_gpu->info->quirks & ADRENO_QUIRK_IFPC) ?
+		GMU_IDLE_STATE_IFPC : GMU_IDLE_STATE_ACTIVE;
 
 	pm_runtime_enable(gmu->dev);
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
index 8e69b1e84657..550de6ad68ef 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c
@@ -21,6 +21,7 @@ static const char * const a6xx_hfi_msg_id[] = {
 	HFI_MSG_ID(HFI_H2F_MSG_PERF_TABLE),
 	HFI_MSG_ID(HFI_H2F_MSG_TEST),
 	HFI_MSG_ID(HFI_H2F_MSG_START),
+	HFI_MSG_ID(HFI_H2F_FEATURE_CTRL),
 	HFI_MSG_ID(HFI_H2F_MSG_CORE_FW_START),
 	HFI_MSG_ID(HFI_H2F_MSG_GX_BW_PERF_VOTE),
 	HFI_MSG_ID(HFI_H2F_MSG_PREPARE_SLUMBER),
@@ -765,23 +766,40 @@ send:
 		NULL, 0);
 }
 
+static int a6xx_hfi_feature_ctrl_msg(struct a6xx_gmu *gmu, u32 feature, u32 enable, u32 data)
+{
+	struct a6xx_hfi_msg_feature_ctrl msg = {
+		.feature = feature,
+		.enable = enable,
+		.data = data,
+	};
+
+	return a6xx_hfi_send_msg(gmu, HFI_H2F_FEATURE_CTRL, &msg, sizeof(msg), NULL, 0);
+}
+
+#define HFI_FEATURE_IFPC 9
+#define IFPC_LONG_HYST 0x1680
+
+static int a6xx_hfi_enable_ifpc(struct a6xx_gmu *gmu)
+{
+	if (gmu->idle_level != GMU_IDLE_STATE_IFPC)
+		return 0;
+
+	return a6xx_hfi_feature_ctrl_msg(gmu, HFI_FEATURE_IFPC, 1, IFPC_LONG_HYST);
+}
+
 #define HFI_FEATURE_ACD 12
 
 static int a6xx_hfi_enable_acd(struct a6xx_gmu *gmu)
 {
 	struct a6xx_hfi_acd_table *acd_table = &gmu->acd_table;
-	struct a6xx_hfi_msg_feature_ctrl msg = {
-		.feature = HFI_FEATURE_ACD,
-		.enable = 1,
-		.data = 0,
-	};
 	int ret;
 
 	if (!acd_table->enable_by_level)
 		return 0;
 
 	/* Enable ACD feature at GMU */
-	ret = a6xx_hfi_send_msg(gmu, HFI_H2F_FEATURE_CTRL, &msg, sizeof(msg), NULL, 0);
+	ret = a6xx_hfi_feature_ctrl_msg(gmu, HFI_FEATURE_ACD, 1, 0);
 	if (ret) {
 		DRM_DEV_ERROR(gmu->dev, "Unable to enable ACD (%d)\n", ret);
 		return ret;
@@ -898,6 +916,10 @@ int a6xx_hfi_start(struct a6xx_gmu *gmu, int boot_state)
 	if (ret)
 		return ret;
 
+	ret = a6xx_hfi_enable_ifpc(gmu);
+	if (ret)
+		return ret;
+
 	ret = a6xx_hfi_send_core_fw_start(gmu);
 	if (ret)
 		return ret;
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 9dc93c247196..390fa6720d9b 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -59,6 +59,7 @@ enum adreno_family {
 #define ADRENO_QUIRK_HAS_CACHED_COHERENT	BIT(4)
 #define ADRENO_QUIRK_PREEMPTION			BIT(5)
 #define ADRENO_QUIRK_4GB_VA			BIT(6)
+#define ADRENO_QUIRK_IFPC			BIT(7)
 
 /* Helper for formating the chip_id in the way that userspace tools like
  * crashdec expect.

From a242ef4a7577bfd3dbf0b4fb7bf92adca145b0c1 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:05 +0530
Subject: [PATCH 1106/1455] drm/msm/a6xx: Fix hangcheck for IFPC

From the hangcheck handler, KMD checks a few registers in GX
domain to see if the GPU made any progress. But it cannot access
those registers when IFPC is enabled. Since HW based hang detection
is pretty decent, lets rely on it instead of these registers when
IFPC is enabled.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673378/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 4cdbad05affc..b6b68b555a0e 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -2420,13 +2420,24 @@ static uint32_t a6xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 
 static bool a6xx_progress(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 {
-	struct msm_cp_state cp_state = {
+	struct msm_cp_state cp_state;
+	bool progress;
+
+	/*
+	 * With IFPC, KMD doesn't know whether GX power domain is collapsed
+	 * or not. So, we can't blindly read the below registers in GX domain.
+	 * Lets trust the hang detection in HW and lie to the caller that
+	 * there was progress.
+	 */
+	if (to_adreno_gpu(gpu)->info->quirks & ADRENO_QUIRK_IFPC)
+		return true;
+
+	cp_state = (struct msm_cp_state) {
 		.ib1_base = gpu_read64(gpu, REG_A6XX_CP_IB1_BASE),
 		.ib2_base = gpu_read64(gpu, REG_A6XX_CP_IB2_BASE),
 		.ib1_rem  = gpu_read(gpu, REG_A6XX_CP_IB1_REM_SIZE),
 		.ib2_rem  = gpu_read(gpu, REG_A6XX_CP_IB2_REM_SIZE),
 	};
-	bool progress;
 
 	/*
 	 * Adjust the remaining data to account for what has already been

From 62cd0fa6990b03d6618e9031d926a2f672a3d801 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:06 +0530
Subject: [PATCH 1107/1455] drm/msm/adreno: Disable IFPC when sysprof is active

Moving to IFPC state clears the 'Perfcounter Select' register setup
by the userspace. So, lets block the IFPC when sysprof is active by
using the perfcounter oob signal to the GMU.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673380/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 37 +++++++++++++++++++++++++++
 drivers/gpu/drm/msm/adreno/a6xx_gmu.h |  2 ++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c |  2 ++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h |  1 +
 drivers/gpu/drm/msm/msm_gpu.h         |  1 +
 drivers/gpu/drm/msm/msm_submitqueue.c |  4 +++
 6 files changed, 47 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 5e147be7dc0f..09611bd4e1c1 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1157,6 +1157,11 @@ int a6xx_gmu_resume(struct a6xx_gpu *a6xx_gpu)
 	/* Set the GPU to the current freq */
 	a6xx_gmu_set_initial_freq(gpu, gmu);
 
+	if (refcount_read(&gpu->sysprof_active) > 1) {
+		ret = a6xx_gmu_set_oob(gmu, GMU_OOB_PERFCOUNTER_SET);
+		if (!ret)
+			set_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status);
+	}
 out:
 	/* On failure, shut down the GMU to leave it in a good state */
 	if (ret) {
@@ -1204,6 +1209,9 @@ static void a6xx_gmu_shutdown(struct a6xx_gmu *gmu)
 		a6xx_gmu_clear_oob(&a6xx_gpu->gmu, GMU_OOB_GPU_SET);
 	}
 
+	if (test_and_clear_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status))
+		a6xx_gmu_clear_oob(gmu, GMU_OOB_PERFCOUNTER_SET);
+
 	ret = a6xx_gmu_wait_for_idle(gmu);
 
 	/* If the GMU isn't responding assume it is hung */
@@ -1817,6 +1825,35 @@ static int a6xx_gmu_get_irq(struct a6xx_gmu *gmu, struct platform_device *pdev,
 	return irq;
 }
 
+void a6xx_gmu_sysprof_setup(struct msm_gpu *gpu)
+{
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
+	struct a6xx_gmu *gmu = &a6xx_gpu->gmu;
+	unsigned int sysprof_active;
+
+	/* Nothing to do if GPU is suspended. We will handle this during GMU resume */
+	if (!pm_runtime_get_if_active(&gpu->pdev->dev))
+		return;
+
+	mutex_lock(&gmu->lock);
+
+	sysprof_active = refcount_read(&gpu->sysprof_active);
+
+	/*
+	 * 'Perfcounter select' register values are lost during IFPC collapse. To avoid that,
+	 * use the currently unused perfcounter oob vote to block IFPC when sysprof is active
+	 */
+	if ((sysprof_active > 1) && !test_and_set_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status))
+		a6xx_gmu_set_oob(gmu, GMU_OOB_PERFCOUNTER_SET);
+	else if ((sysprof_active == 1) && test_and_clear_bit(GMU_STATUS_OOB_PERF_SET, &gmu->status))
+		a6xx_gmu_clear_oob(gmu, GMU_OOB_PERFCOUNTER_SET);
+
+	mutex_unlock(&gmu->lock);
+
+	pm_runtime_put(&gpu->pdev->dev);
+}
+
 void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu)
 {
 	struct adreno_gpu *adreno_gpu = &a6xx_gpu->base;
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
index 40c4c0445b8e..06cfc294016f 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h
@@ -125,6 +125,8 @@ struct a6xx_gmu {
 #define GMU_STATUS_FW_START	0
 /* To track if PDC sleep seq was done */
 #define GMU_STATUS_PDC_SLEEP	1
+/* To track Perfcounter OOB set status */
+#define GMU_STATUS_OOB_PERF_SET 2
 	unsigned long status;
 };
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index b6b68b555a0e..bfffef6fe870 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -2536,6 +2536,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.create_private_vm = a6xx_create_private_vm,
 		.get_rptr = a6xx_get_rptr,
 		.progress = a6xx_progress,
+		.sysprof_setup = a6xx_gmu_sysprof_setup,
 	},
 	.get_timestamp = a6xx_gmu_get_timestamp,
 };
@@ -2596,6 +2597,7 @@ static const struct adreno_gpu_funcs funcs_a7xx = {
 		.create_private_vm = a6xx_create_private_vm,
 		.get_rptr = a6xx_get_rptr,
 		.progress = a6xx_progress,
+		.sysprof_setup = a6xx_gmu_sysprof_setup,
 	},
 	.get_timestamp = a6xx_gmu_get_timestamp,
 };
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index e736c59d566b..124c63c46159 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -254,6 +254,7 @@ void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state);
 int a6xx_gmu_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node);
 int a6xx_gmu_wrapper_init(struct a6xx_gpu *a6xx_gpu, struct device_node *node);
 void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu);
+void a6xx_gmu_sysprof_setup(struct msm_gpu *gpu);
 
 void a6xx_preempt_init(struct msm_gpu *gpu);
 void a6xx_preempt_hw_init(struct msm_gpu *gpu);
diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h
index 866bb090af06..a597f2bee30b 100644
--- a/drivers/gpu/drm/msm/msm_gpu.h
+++ b/drivers/gpu/drm/msm/msm_gpu.h
@@ -92,6 +92,7 @@ struct msm_gpu_funcs {
 	 * for cmdstream that is buffered in this FIFO upstream of the CP fw.
 	 */
 	bool (*progress)(struct msm_gpu *gpu, struct msm_ringbuffer *ring);
+	void (*sysprof_setup)(struct msm_gpu *gpu);
 };
 
 /* Additional state for iommu faults: */
diff --git a/drivers/gpu/drm/msm/msm_submitqueue.c b/drivers/gpu/drm/msm/msm_submitqueue.c
index 8617a82cd6b3..d53dfad16bde 100644
--- a/drivers/gpu/drm/msm/msm_submitqueue.c
+++ b/drivers/gpu/drm/msm/msm_submitqueue.c
@@ -40,6 +40,10 @@ int msm_context_set_sysprof(struct msm_context *ctx, struct msm_gpu *gpu, int sy
 		break;
 	}
 
+	/* Some gpu families require additional setup for sysprof */
+	if (gpu->funcs->sysprof_setup)
+		gpu->funcs->sysprof_setup(gpu);
+
 	ctx->sysprof = sysprof;
 
 	return 0;

From 870d72d000d7d9f788ddc3b079531c313c9d48a8 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:07 +0530
Subject: [PATCH 1108/1455] drm/msm/a6xx: Make crashstate capture IFPC safe

Now with IFPC, GX domain can collapse as soon as GPU becomes IDLE. So
add gx_is_on check before accessing any GX registers during crashstate
capture and recovery.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673383/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c       |  4 +++
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c       | 27 +++++++++++++++------
 drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c | 10 +++++---
 3 files changed, 30 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index 09611bd4e1c1..a719a8d44c1b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -101,6 +101,10 @@ bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu)
 	if (!gmu->initialized)
 		return false;
 
+	/* If GMU is absent, then GX power domain is ON as long as GPU is in active state */
+	if (adreno_has_gmu_wrapper(adreno_gpu))
+		return true;
+
 	val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS);
 
 	if (adreno_is_a7xx(adreno_gpu))
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index bfffef6fe870..cdb6d5f268b3 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1537,21 +1537,25 @@ static void a6xx_recover(struct msm_gpu *gpu)
 
 	adreno_dump_info(gpu);
 
-	for (i = 0; i < 8; i++)
-		DRM_DEV_INFO(&gpu->pdev->dev, "CP_SCRATCH_REG%d: %u\n", i,
-			gpu_read(gpu, REG_A6XX_CP_SCRATCH_REG(i)));
+	if (a6xx_gmu_gx_is_on(&a6xx_gpu->gmu)) {
+		/* Sometimes crashstate capture is skipped, so SQE should be halted here again */
+		gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3);
 
-	if (hang_debug)
-		a6xx_dump(gpu);
+		for (i = 0; i < 8; i++)
+			DRM_DEV_INFO(&gpu->pdev->dev, "CP_SCRATCH_REG%d: %u\n", i,
+				gpu_read(gpu, REG_A6XX_CP_SCRATCH_REG(i)));
+
+		if (hang_debug)
+			a6xx_dump(gpu);
+
+	}
 
 	/*
 	 * To handle recovery specific sequences during the rpm suspend we are
 	 * about to trigger
 	 */
-	a6xx_gpu->hung = true;
 
-	/* Halt SQE first */
-	gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3);
+	a6xx_gpu->hung = true;
 
 	pm_runtime_dont_use_autosuspend(&gpu->pdev->dev);
 
@@ -2415,6 +2419,13 @@ static uint32_t a6xx_get_rptr(struct msm_gpu *gpu, struct msm_ringbuffer *ring)
 	if (adreno_gpu->base.hw_apriv || a6xx_gpu->has_whereami)
 		return a6xx_gpu->shadow[ring->id];
 
+	/*
+	 * This is true only on an A6XX_GEN1 with GMU, has IFPC enabled and a super old SQE firmware
+	 * without 'whereami' support
+	 */
+	WARN_ONCE((to_adreno_gpu(gpu)->info->quirks & ADRENO_QUIRK_IFPC),
+		"Can't read CP_RB_RPTR register reliably\n");
+
 	return ring->memptrs->rptr = gpu_read(gpu, REG_A6XX_CP_RB_RPTR);
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
index d5d1271fce61..4c7f3c642f6a 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c
@@ -1586,8 +1586,7 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu)
 	struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu);
 	struct a6xx_gpu_state *a6xx_state = kzalloc(sizeof(*a6xx_state),
 		GFP_KERNEL);
-	bool stalled = !!(gpu_read(gpu, REG_A6XX_RBBM_STATUS3) &
-			A6XX_RBBM_STATUS3_SMMU_STALLED_ON_FAULT);
+	bool stalled;
 
 	if (!a6xx_state)
 		return ERR_PTR(-ENOMEM);
@@ -1608,15 +1607,20 @@ struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu)
 	}
 
 	/* If GX isn't on the rest of the data isn't going to be accessible */
-	if (!adreno_has_gmu_wrapper(adreno_gpu) && !a6xx_gmu_gx_is_on(&a6xx_gpu->gmu))
+	if (!a6xx_gmu_gx_is_on(&a6xx_gpu->gmu))
 		return &a6xx_state->base;
 
+	/* Halt SQE first */
+	gpu_write(gpu, REG_A6XX_CP_SQE_CNTL, 3);
+
 	/* Get the banks of indexed registers */
 	if (adreno_is_a7xx(adreno_gpu))
 		a7xx_get_indexed_registers(gpu, a6xx_state);
 	else
 		a6xx_get_indexed_registers(gpu, a6xx_state);
 
+	stalled = !!(gpu_read(gpu, REG_A6XX_RBBM_STATUS3) &
+			A6XX_RBBM_STATUS3_SMMU_STALLED_ON_FAULT);
 	/*
 	 * Try to initialize the crashdumper, if we are not dumping state
 	 * with the SMMU stalled.  The crashdumper needs memory access to

From a6a0157cc68ec561392fc81f603b5b9de1a20e4e Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:08 +0530
Subject: [PATCH 1109/1455] drm/msm/a6xx: Enable IFPC on Adreno X1-85

Add the IFPC restore register list and enable IFPC support on Adreno
X1-85 gpu.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673384/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 68 ++++++++++++++++++++++-
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c     | 15 +++--
 drivers/gpu/drm/msm/adreno/a6xx_gpu.h     |  1 +
 3 files changed, 79 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index 4b3f18b2d2b9..7f495212c092 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -1353,6 +1353,69 @@ static const uint32_t a7xx_pwrup_reglist_regs[] = {
 
 DECLARE_ADRENO_REGLIST_LIST(a7xx_pwrup_reglist);
 
+/* Applicable for X185, A750 */
+static const u32 a750_ifpc_reglist_regs[] = {
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_0,
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_1,
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_2,
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_3,
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_4,
+	REG_A6XX_TPL1_NC_MODE_CNTL,
+	REG_A6XX_SP_NC_MODE_CNTL,
+	REG_A6XX_CP_DBG_ECO_CNTL,
+	REG_A6XX_CP_PROTECT_CNTL,
+	REG_A6XX_CP_PROTECT(0),
+	REG_A6XX_CP_PROTECT(1),
+	REG_A6XX_CP_PROTECT(2),
+	REG_A6XX_CP_PROTECT(3),
+	REG_A6XX_CP_PROTECT(4),
+	REG_A6XX_CP_PROTECT(5),
+	REG_A6XX_CP_PROTECT(6),
+	REG_A6XX_CP_PROTECT(7),
+	REG_A6XX_CP_PROTECT(8),
+	REG_A6XX_CP_PROTECT(9),
+	REG_A6XX_CP_PROTECT(10),
+	REG_A6XX_CP_PROTECT(11),
+	REG_A6XX_CP_PROTECT(12),
+	REG_A6XX_CP_PROTECT(13),
+	REG_A6XX_CP_PROTECT(14),
+	REG_A6XX_CP_PROTECT(15),
+	REG_A6XX_CP_PROTECT(16),
+	REG_A6XX_CP_PROTECT(17),
+	REG_A6XX_CP_PROTECT(18),
+	REG_A6XX_CP_PROTECT(19),
+	REG_A6XX_CP_PROTECT(20),
+	REG_A6XX_CP_PROTECT(21),
+	REG_A6XX_CP_PROTECT(22),
+	REG_A6XX_CP_PROTECT(23),
+	REG_A6XX_CP_PROTECT(24),
+	REG_A6XX_CP_PROTECT(25),
+	REG_A6XX_CP_PROTECT(26),
+	REG_A6XX_CP_PROTECT(27),
+	REG_A6XX_CP_PROTECT(28),
+	REG_A6XX_CP_PROTECT(29),
+	REG_A6XX_CP_PROTECT(30),
+	REG_A6XX_CP_PROTECT(31),
+	REG_A6XX_CP_PROTECT(32),
+	REG_A6XX_CP_PROTECT(33),
+	REG_A6XX_CP_PROTECT(34),
+	REG_A6XX_CP_PROTECT(35),
+	REG_A6XX_CP_PROTECT(36),
+	REG_A6XX_CP_PROTECT(37),
+	REG_A6XX_CP_PROTECT(38),
+	REG_A6XX_CP_PROTECT(39),
+	REG_A6XX_CP_PROTECT(40),
+	REG_A6XX_CP_PROTECT(41),
+	REG_A6XX_CP_PROTECT(42),
+	REG_A6XX_CP_PROTECT(43),
+	REG_A6XX_CP_PROTECT(44),
+	REG_A6XX_CP_PROTECT(45),
+	REG_A6XX_CP_PROTECT(46),
+	REG_A6XX_CP_PROTECT(47),
+};
+
+DECLARE_ADRENO_REGLIST_LIST(a750_ifpc_reglist);
+
 static const struct adreno_info a7xx_gpus[] = {
 	{
 		.chip_ids = ADRENO_CHIP_IDS(0x07000200),
@@ -1442,12 +1505,14 @@ static const struct adreno_info a7xx_gpus[] = {
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
 			  ADRENO_QUIRK_HAS_HW_APRIV |
-			  ADRENO_QUIRK_PREEMPTION,
+			  ADRENO_QUIRK_PREEMPTION |
+			  ADRENO_QUIRK_IFPC,
 		.init = a6xx_gpu_init,
 		.a6xx = &(const struct a6xx_info) {
 			.hwcg = a740_hwcg,
 			.protect = &a730_protect,
 			.pwrup_reglist = &a7xx_pwrup_reglist,
+			.ifpc_reglist = &a750_ifpc_reglist,
 			.gmu_chipid = 0x7050001,
 			.gmu_cgc_mode = 0x00020202,
 			.bcms = (const struct a6xx_bcm[]) {
@@ -1487,6 +1552,7 @@ static const struct adreno_info a7xx_gpus[] = {
 		.a6xx = &(const struct a6xx_info) {
 			.protect = &a730_protect,
 			.pwrup_reglist = &a7xx_pwrup_reglist,
+			.ifpc_reglist = &a750_ifpc_reglist,
 			.gmu_chipid = 0x7090100,
 			.gmu_cgc_mode = 0x00020202,
 			.bcms = (const struct a6xx_bcm[]) {
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index cdb6d5f268b3..2f68394d6c3b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -828,11 +828,10 @@ static void a7xx_patch_pwrup_reglist(struct msm_gpu *gpu)
 	u32 *dest = (u32 *)&lock->regs[0];
 	int i;
 
-	reglist = adreno_gpu->info->a6xx->pwrup_reglist;
-
 	lock->gpu_req = lock->cpu_req = lock->turn = 0;
-	lock->ifpc_list_len = 0;
-	lock->preemption_list_len = reglist->count;
+
+	reglist = adreno_gpu->info->a6xx->ifpc_reglist;
+	lock->ifpc_list_len = reglist->count;
 
 	/*
 	 * For each entry in each of the lists, write the offset and the current
@@ -843,6 +842,14 @@ static void a7xx_patch_pwrup_reglist(struct msm_gpu *gpu)
 		*dest++ = gpu_read(gpu, reglist->regs[i]);
 	}
 
+	reglist = adreno_gpu->info->a6xx->pwrup_reglist;
+	lock->preemption_list_len = reglist->count;
+
+	for (i = 0; i < reglist->count; i++) {
+		*dest++ = reglist->regs[i];
+		*dest++ = gpu_read(gpu, reglist->regs[i]);
+	}
+
 	/*
 	 * The overall register list is composed of
 	 * 1. Static IFPC-only registers
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
index 124c63c46159..0b17d36c36a9 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h
@@ -45,6 +45,7 @@ struct a6xx_info {
 	const struct adreno_reglist *hwcg;
 	const struct adreno_protect *protect;
 	const struct adreno_reglist_list *pwrup_reglist;
+	const struct adreno_reglist_list *ifpc_reglist;
 	u32 gmu_chipid;
 	u32 gmu_cgc_mode;
 	u32 prim_fifo_threshold;

From 6341516bc25cef7dff078132905a792790303d59 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 13:57:09 +0530
Subject: [PATCH 1110/1455] drm/msm/a6xx: Enable IFPC on A750 GPU

A750 GPU has similar IFPC related configurations like X1-85. Add
the IFPC QUIRK to enable IFPC feature.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673386/
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_catalog.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index 7f495212c092..a2d587e1a4f5 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -1546,7 +1546,8 @@ static const struct adreno_info a7xx_gpus[] = {
 		.inactive_period = DRM_MSM_INACTIVE_PERIOD,
 		.quirks = ADRENO_QUIRK_HAS_CACHED_COHERENT |
 			  ADRENO_QUIRK_HAS_HW_APRIV |
-			  ADRENO_QUIRK_PREEMPTION,
+			  ADRENO_QUIRK_PREEMPTION |
+			  ADRENO_QUIRK_IFPC,
 		.init = a6xx_gpu_init,
 		.zapfw = "gen70900_zap.mbn",
 		.a6xx = &(const struct a6xx_info) {

From 956f5e5bc80a223cdeaae010bafc17d2ee0c7276 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Fri, 5 Sep 2025 09:22:37 -0700
Subject: [PATCH 1111/1455] drm/xe/configfs: Use config_group_put()

configfs has a config_group_put() helper that was adopted by
commit 88df7939d728 ("drm/xe/configfs: Rename struct xe_config_device").
Another pending work to add psmi later landed in commit
afe902848b41 ("drm/xe/configfs: Allow to enable PSMI") and didn't use
the helper.

Use config_group_put() consistently to hide the inner workings of
configfs. No change in behavior since it does exactly the same thing
as currently being done.

Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250905162236.578117-2-lucas.demarchi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 0337811864cd..e515fa7085fa 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -598,7 +598,7 @@ bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev)
 		return false;
 
 	ret = dev->config.enable_psmi;
-	config_item_put(&dev->group.cg_item);
+	config_group_put(&dev->group);
 
 	return ret;
 }

From 490b30fbaca2abbd6afa8bdc7e2df329b5d82412 Mon Sep 17 00:00:00 2001
From: Zhongtian Wu <wuzhongtian@huaqin.corp-partner.google.com>
Date: Mon, 8 Sep 2025 14:37:32 +0800
Subject: [PATCH 1112/1455] drm/panel-edp: Add 4 more panels needed by mt8189
 Chromebooks

Add a few generic edp panels used by mt8189 chromebooks. For
BOE-NV140WUM-N44 , the enable timing required 80ms. For
CSW-MNE007QB3-1, the hpd_absent timing rquired 80ms, the enable timing
required 50ms, the disable timing required 50ms. For CSW-MNE007QS3-6,
the enable timing required 50ms. For CMN-N140JCA-ELK, the enable timing
required 80ms and disable timing required 50ms.

BOE NV140WUM-N44 V8.2
edid-decode (hex):

00 ff ff ff ff ff ff 00 09 e5 6a 0a 00 00 00 00
2e 20 01 04 a5 1e 13 78 03 fb f5 96 5d 5a 91 29
1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 61 40 80 04 71 b0 3c 40 30 20
36 00 2d bc 10 00 00 1a 81 33 80 04 71 b0 3c 40
30 20 36 00 2d bc 10 00 00 1a 00 00 00 fd 00 28
3c 4c 4c 10 01 0a 20 20 20 20 20 20 00 00 00 fe
00 4e 56 31 34 30 57 55 4d 2d 4e 34 34 0a 01 7c

02 03 0d 00 68 1a 00 00 01 01 28 3c 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 06

CSW MNE007QB3-1:
edid-decode (hex):

00 ff ff ff ff ff ff 00 0e 77 6e 14 00 00 00 00
00 23 01 04 a5 1e 13 78 07 ee 95 a3 54 4c 99 26
0f 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 35 3c 80 a0 70 b0 23 40 30 20
36 00 2d bc 10 00 00 18 2b 30 80 a0 70 b0 23 40
30 20 36 00 2d bc 10 00 00 18 00 00 00 fd 00 28
3c 4a 4a 0f 01 0a 20 20 20 20 20 20 00 00 00 fc
00 4d 4e 45 30 30 37 51 42 33 2d 31 0a 20 01 69

70 20 79 02 00 21 00 1d c8 0b 5d 07 80 07 b0 04
00 3d 8a 54 cd a4 99 66 62 0f 02 45 54 40 5e 40
5e 00 44 12 78 2e 00 06 00 44 40 5e 40 5e 81 00
20 74 1a 00 00 03 01 28 3c 00 00 00 00 00 00 3c
00 00 00 00 8d 00 e3 05 04 00 e6 06 01 00 60 60
ff 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 68 90

CSW MNE007QS3-6:
edid-decode (hex):

00 ff ff ff ff ff ff 00 0e 77 3f 14 00 00 00 00
00 22 01 04 a5 1e 13 78 03 2c c5 94 5c 59 95 29
1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 ea 3d 80 c8 70 b0 2e 40 30 20
36 00 2e bd 10 00 00 1a 88 31 80 c8 70 b0 2e 40
30 20 36 00 2e bd 10 00 00 1a 00 00 00 fd 00 28
3c 4b 4b 10 01 0a 20 20 20 20 20 20 00 00 00 fc
00 4d 4e 45 30 30 37 51 53 33 2d 36 0a 20 01 80

70 20 79 02 00 81 00 14 74 1a 00 00 03 01 28 3c
00 00 00 00 00 00 3c 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 9e 90

CMN N140JCA-ELK:
edid-decode (hex):

00 ff ff ff ff ff ff 00 0d ae 41 14 00 00 00 00
25 21 01 04 a5 1e 13 78 03 28 65 97 59 54 8e 27
1e 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
01 01 01 01 01 01 42 3c 80 a0 70 b0 24 40 30 20
a6 00 2d bc 10 00 00 18 35 30 80 a0 70 b0 24 40
30 20 a6 00 2d bc 10 00 00 18 00 00 00 fd 00 28
3c 4b 4b 10 01 0a 20 20 20 20 20 20 00 00 00 fe
00 4e 31 34 30 4a 43 41 2d 45 4c 4b 0a 20 01 14

02 03 0d 00 68 1a 00 00 01 01 28 3c 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 06

Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Zhongtian Wu <wuzhongtian@huaqin.corp-partner.google.com>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20250908063732.764289-1-wuzhongtian@huaqin.corp-partner.google.com
---
 drivers/gpu/drm/panel/panel-edp.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/panel/panel-edp.c b/drivers/gpu/drm/panel/panel-edp.c
index feba520a0f45..62435e3cd9f4 100644
--- a/drivers/gpu/drm/panel/panel-edp.c
+++ b/drivers/gpu/drm/panel/panel-edp.c
@@ -1843,6 +1843,13 @@ static const struct panel_delay delay_200_500_e50_d100 = {
 	.disable = 100,
 };
 
+static const struct panel_delay delay_80_500_e50_d50 = {
+	.hpd_absent = 80,
+	.unprepare = 500,
+	.enable = 50,
+	.disable = 50,
+};
+
 #define EDP_PANEL_ENTRY(vend_chr_0, vend_chr_1, vend_chr_2, product_id, _delay, _name) \
 { \
 	.ident = { \
@@ -1955,6 +1962,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a36, &delay_200_500_e200, "Unknown"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a3e, &delay_200_500_e80_d50, "NV116WHM-N49"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a5d, &delay_200_500_e50, "NV116WHM-N45"),
+	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0a6a, &delay_200_500_e80, "NV140WUM-N44"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ac5, &delay_200_500_e50, "NV116WHM-N4C"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0ae8, &delay_200_500_e50_p2e80, "NV140WUM-N41"),
 	EDP_PANEL_ENTRY('B', 'O', 'E', 0x0b09, &delay_200_500_e50_po2e200, "NV140FHM-NZ"),
@@ -1996,6 +2004,7 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x124c, &delay_200_500_e80_d50, "N122JCA-ENK"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x142b, &delay_200_500_e80_d50, "N140HCA-EAC"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x142e, &delay_200_500_e80_d50, "N140BGA-EA4"),
+	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1441, &delay_200_500_e80_d50, "N140JCA-ELK"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x144f, &delay_200_500_e80_d50, "N140HGA-EA1"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x1468, &delay_200_500_e80, "N140HGA-EA1"),
 	EDP_PANEL_ENTRY('C', 'M', 'N', 0x14a8, &delay_200_500_e80, "N140JCA-ELP"),
@@ -2011,10 +2020,12 @@ static const struct edp_panel_entry edp_panels[] = {
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1100, &delay_200_500_e80_d50, "MNB601LS1-1"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1103, &delay_200_500_e80_d50, "MNB601LS1-3"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1104, &delay_200_500_e50_d100, "MNB601LS1-4"),
+	EDP_PANEL_ENTRY('C', 'S', 'W', 0x143f, &delay_200_500_e50, "MNE007QS3-6"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1448, &delay_200_500_e50, "MNE007QS3-7"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1457, &delay_80_500_e80_p2e200, "MNE007QS3-8"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1462, &delay_200_500_e50, "MNE007QS5-2"),
 	EDP_PANEL_ENTRY('C', 'S', 'W', 0x1468, &delay_200_500_e50, "MNE007QB2-2"),
+	EDP_PANEL_ENTRY('C', 'S', 'W', 0x146e, &delay_80_500_e50_d50, "MNE007QB3-1"),
 
 	EDP_PANEL_ENTRY('E', 'T', 'C', 0x0000, &delay_50_500_e200_d200_po2e335, "LP079QX1-SP0V"),
 

From 6ea42e9146f7ab8e59ffea8aa3300ad6710399dd Mon Sep 17 00:00:00 2001
From: Lyude Paul <lyude@redhat.com>
Date: Mon, 8 Sep 2025 14:46:36 -0400
Subject: [PATCH 1113/1455] rust: drm: gem: Simplify use of generics

Now that my rust skills have been honed, I noticed that there's a lot of
generics in our gem bindings that don't actually need to be here. Currently
the hierarchy of traits in our gem bindings looks like this:

  * Drivers implement:
    * BaseDriverObject<T: DriverObject> (has the callbacks)
    * DriverObject (has the drm::Driver type)
  * Crate implements:
    * IntoGEMObject for Object<T> where T: DriverObject
      Handles conversion to/from raw object pointers
    * BaseObject for T where T: IntoGEMObject
      Provides methods common to all gem interfaces

  Also of note, this leaves us with two different drm::Driver associated
  types:
    * DriverObject::Driver
    * IntoGEMObject::Driver

I'm not entirely sure of the original intent here unfortunately (if anyone
is, please let me know!), but my guess is that the idea would be that some
objects can implement IntoGEMObject using a different ::Driver than
DriverObject - presumably to enable the usage of gem objects from different
drivers. A reasonable usecase of course.

However - if I'm not mistaken, I don't think that this is actually how
things would go in practice. Driver implementations are of course
implemented by their associated drivers, and generally drivers are not
linked to each-other when building the kernel. Which is to say that even in
a situation where we would theoretically deal with gem objects from another
driver, we still wouldn't have access to its drm::driver::Driver
implementation. It's more likely we would simply want a variant of gem
objects in such a situation that have no association with a
drm::driver::Driver type.

Taking that into consideration, we can assume the following:
* Anything that implements BaseDriverObject will implement DriverObject
  In other words, all BaseDriverObjects indirectly have an associated
  ::Driver type - so the two traits can be combined into one with no
  generics.
* Not everything that implements IntoGEMObject will have an associated
  ::Driver, and that's OK.

And with this, we now can do quite a bit of cleanup with the use of
generics here. As such, this commit:

* Removes the generics on BaseDriverObject
* Moves DriverObject::Driver into BaseDriverObject
* Removes DriverObject
* Removes IntoGEMObject::Driver
* Add AllocImpl::Driver, which we can use as a binding to figure out the
  correct File type for BaseObject

Leaving us with a simpler trait hierarchy that now looks like this:

  * Drivers implement: BaseDriverObject
  * Crate implements:
    * IntoGEMObject for Object<T> where T: DriverObject
    * BaseObject for T where T: IntoGEMObject

Which makes the code a lot easier to understand and build on :).

Signed-off-by: Lyude Paul <lyude@redhat.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250908185239.135849-2-lyude@redhat.com
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
---
 drivers/gpu/drm/nova/gem.rs |  8 ++--
 rust/kernel/drm/driver.rs   |  3 ++
 rust/kernel/drm/gem/mod.rs  | 77 ++++++++++++++++---------------------
 3 files changed, 40 insertions(+), 48 deletions(-)

diff --git a/drivers/gpu/drm/nova/gem.rs b/drivers/gpu/drm/nova/gem.rs
index cd82773dab92..2760ba4f3450 100644
--- a/drivers/gpu/drm/nova/gem.rs
+++ b/drivers/gpu/drm/nova/gem.rs
@@ -16,16 +16,14 @@ use crate::{
 #[pin_data]
 pub(crate) struct NovaObject {}
 
-impl gem::BaseDriverObject<gem::Object<NovaObject>> for NovaObject {
+impl gem::DriverObject for NovaObject {
+    type Driver = NovaDriver;
+
     fn new(_dev: &NovaDevice, _size: usize) -> impl PinInit<Self, Error> {
         try_pin_init!(NovaObject {})
     }
 }
 
-impl gem::DriverObject for NovaObject {
-    type Driver = NovaDriver;
-}
-
 impl NovaObject {
     /// Create a new DRM GEM object.
     pub(crate) fn new(dev: &NovaDevice, size: usize) -> Result<ARef<gem::Object<Self>>> {
diff --git a/rust/kernel/drm/driver.rs b/rust/kernel/drm/driver.rs
index fe7e8d06961a..dae0f4d1bbe3 100644
--- a/rust/kernel/drm/driver.rs
+++ b/rust/kernel/drm/driver.rs
@@ -86,6 +86,9 @@ pub struct AllocOps {
 
 /// Trait for memory manager implementations. Implemented internally.
 pub trait AllocImpl: super::private::Sealed + drm::gem::IntoGEMObject {
+    /// The [`Driver`] implementation for this [`AllocImpl`].
+    type Driver: drm::Driver;
+
     /// The C callback operations for this memory manager.
     const ALLOC_OPS: AllocOps;
 }
diff --git a/rust/kernel/drm/gem/mod.rs b/rust/kernel/drm/gem/mod.rs
index b71821cfb5ea..f5b19865d745 100644
--- a/rust/kernel/drm/gem/mod.rs
+++ b/rust/kernel/drm/gem/mod.rs
@@ -15,31 +15,31 @@ use crate::{
 use core::{mem, ops::Deref, ptr::NonNull};
 
 /// GEM object functions, which must be implemented by drivers.
-pub trait BaseDriverObject<T: BaseObject>: Sync + Send + Sized {
+pub trait DriverObject: Sync + Send + Sized {
+    /// Parent `Driver` for this object.
+    type Driver: drm::Driver;
+
     /// Create a new driver data object for a GEM object of a given size.
-    fn new(dev: &drm::Device<T::Driver>, size: usize) -> impl PinInit<Self, Error>;
+    fn new(dev: &drm::Device<Self::Driver>, size: usize) -> impl PinInit<Self, Error>;
 
     /// Open a new handle to an existing object, associated with a File.
     fn open(
-        _obj: &<<T as IntoGEMObject>::Driver as drm::Driver>::Object,
-        _file: &drm::File<<<T as IntoGEMObject>::Driver as drm::Driver>::File>,
+        _obj: &<Self::Driver as drm::Driver>::Object,
+        _file: &drm::File<<Self::Driver as drm::Driver>::File>,
     ) -> Result {
         Ok(())
     }
 
     /// Close a handle to an existing object, associated with a File.
     fn close(
-        _obj: &<<T as IntoGEMObject>::Driver as drm::Driver>::Object,
-        _file: &drm::File<<<T as IntoGEMObject>::Driver as drm::Driver>::File>,
+        _obj: &<Self::Driver as drm::Driver>::Object,
+        _file: &drm::File<<Self::Driver as drm::Driver>::File>,
     ) {
     }
 }
 
 /// Trait that represents a GEM object subtype
 pub trait IntoGEMObject: Sized + super::private::Sealed + AlwaysRefCounted {
-    /// Owning driver for this type
-    type Driver: drm::Driver;
-
     /// Returns a reference to the raw `drm_gem_object` structure, which must be valid as long as
     /// this owning object is valid.
     fn as_raw(&self) -> *mut bindings::drm_gem_object;
@@ -74,25 +74,15 @@ unsafe impl<T: IntoGEMObject> AlwaysRefCounted for T {
     }
 }
 
-/// Trait which must be implemented by drivers using base GEM objects.
-pub trait DriverObject: BaseDriverObject<Object<Self>> {
-    /// Parent `Driver` for this object.
-    type Driver: drm::Driver;
-}
-
-extern "C" fn open_callback<T: BaseDriverObject<U>, U: BaseObject>(
+extern "C" fn open_callback<T: DriverObject>(
     raw_obj: *mut bindings::drm_gem_object,
     raw_file: *mut bindings::drm_file,
 ) -> core::ffi::c_int {
     // SAFETY: `open_callback` is only ever called with a valid pointer to a `struct drm_file`.
-    let file = unsafe {
-        drm::File::<<<U as IntoGEMObject>::Driver as drm::Driver>::File>::from_raw(raw_file)
-    };
-    // SAFETY: `open_callback` is specified in the AllocOps structure for `Object<T>`, ensuring that
-    // `raw_obj` is indeed contained within a `Object<T>`.
-    let obj = unsafe {
-        <<<U as IntoGEMObject>::Driver as drm::Driver>::Object as IntoGEMObject>::from_raw(raw_obj)
-    };
+    let file = unsafe { drm::File::<<T::Driver as drm::Driver>::File>::from_raw(raw_file) };
+    // SAFETY: `open_callback` is specified in the AllocOps structure for `DriverObject<T>`,
+    // ensuring that `raw_obj` is contained within a `DriverObject<T>`
+    let obj = unsafe { <<T::Driver as drm::Driver>::Object as IntoGEMObject>::from_raw(raw_obj) };
 
     match T::open(obj, file) {
         Err(e) => e.to_errno(),
@@ -100,26 +90,21 @@ extern "C" fn open_callback<T: BaseDriverObject<U>, U: BaseObject>(
     }
 }
 
-extern "C" fn close_callback<T: BaseDriverObject<U>, U: BaseObject>(
+extern "C" fn close_callback<T: DriverObject>(
     raw_obj: *mut bindings::drm_gem_object,
     raw_file: *mut bindings::drm_file,
 ) {
     // SAFETY: `open_callback` is only ever called with a valid pointer to a `struct drm_file`.
-    let file = unsafe {
-        drm::File::<<<U as IntoGEMObject>::Driver as drm::Driver>::File>::from_raw(raw_file)
-    };
+    let file = unsafe { drm::File::<<T::Driver as drm::Driver>::File>::from_raw(raw_file) };
+
     // SAFETY: `close_callback` is specified in the AllocOps structure for `Object<T>`, ensuring
     // that `raw_obj` is indeed contained within a `Object<T>`.
-    let obj = unsafe {
-        <<<U as IntoGEMObject>::Driver as drm::Driver>::Object as IntoGEMObject>::from_raw(raw_obj)
-    };
+    let obj = unsafe { <<T::Driver as drm::Driver>::Object as IntoGEMObject>::from_raw(raw_obj) };
 
     T::close(obj, file);
 }
 
 impl<T: DriverObject> IntoGEMObject for Object<T> {
-    type Driver = T::Driver;
-
     fn as_raw(&self) -> *mut bindings::drm_gem_object {
         self.obj.get()
     }
@@ -141,10 +126,12 @@ pub trait BaseObject: IntoGEMObject {
 
     /// Creates a new handle for the object associated with a given `File`
     /// (or returns an existing one).
-    fn create_handle(
-        &self,
-        file: &drm::File<<<Self as IntoGEMObject>::Driver as drm::Driver>::File>,
-    ) -> Result<u32> {
+    fn create_handle<D, F>(&self, file: &drm::File<F>) -> Result<u32>
+    where
+        Self: AllocImpl<Driver = D>,
+        D: drm::Driver<Object = Self, File = F>,
+        F: drm::file::DriverFile<Driver = D>,
+    {
         let mut handle: u32 = 0;
         // SAFETY: The arguments are all valid per the type invariants.
         to_result(unsafe {
@@ -154,10 +141,12 @@ pub trait BaseObject: IntoGEMObject {
     }
 
     /// Looks up an object by its handle for a given `File`.
-    fn lookup_handle(
-        file: &drm::File<<<Self as IntoGEMObject>::Driver as drm::Driver>::File>,
-        handle: u32,
-    ) -> Result<ARef<Self>> {
+    fn lookup_handle<D, F>(file: &drm::File<F>, handle: u32) -> Result<ARef<Self>>
+    where
+        Self: AllocImpl<Driver = D>,
+        D: drm::Driver<Object = Self, File = F>,
+        F: drm::file::DriverFile<Driver = D>,
+    {
         // SAFETY: The arguments are all valid per the type invariants.
         let ptr = unsafe { bindings::drm_gem_object_lookup(file.as_raw().cast(), handle) };
         if ptr.is_null() {
@@ -212,8 +201,8 @@ impl<T: DriverObject> Object<T> {
 
     const OBJECT_FUNCS: bindings::drm_gem_object_funcs = bindings::drm_gem_object_funcs {
         free: Some(Self::free_callback),
-        open: Some(open_callback::<T, Object<T>>),
-        close: Some(close_callback::<T, Object<T>>),
+        open: Some(open_callback::<T>),
+        close: Some(close_callback::<T>),
         print_info: None,
         export: None,
         pin: None,
@@ -296,6 +285,8 @@ impl<T: DriverObject> Deref for Object<T> {
 }
 
 impl<T: DriverObject> AllocImpl for Object<T> {
+    type Driver = T::Driver;
+
     const ALLOC_OPS: AllocOps = AllocOps {
         gem_create_object: None,
         prime_handle_to_fd: None,

From 1ed10db60f47306c1fd395f87bf6d443926a9488 Mon Sep 17 00:00:00 2001
From: Lyude Paul <lyude@redhat.com>
Date: Mon, 8 Sep 2025 14:46:37 -0400
Subject: [PATCH 1114/1455] rust: drm: gem: Add DriverFile type alias
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Just to reduce the clutter with the File<…> types in gem.rs.

Signed-off-by: Lyude Paul <lyude@redhat.com>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250908185239.135849-3-lyude@redhat.com
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
---
 rust/kernel/drm/gem/mod.rs | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/rust/kernel/drm/gem/mod.rs b/rust/kernel/drm/gem/mod.rs
index f5b19865d745..ead723980b27 100644
--- a/rust/kernel/drm/gem/mod.rs
+++ b/rust/kernel/drm/gem/mod.rs
@@ -14,6 +14,13 @@ use crate::{
 };
 use core::{mem, ops::Deref, ptr::NonNull};
 
+/// A type alias for retrieving a [`Driver`]s [`DriverFile`] implementation from its
+/// [`DriverObject`] implementation.
+///
+/// [`Driver`]: drm::Driver
+/// [`DriverFile`]: drm::file::DriverFile
+pub type DriverFile<T> = drm::File<<<T as DriverObject>::Driver as drm::Driver>::File>;
+
 /// GEM object functions, which must be implemented by drivers.
 pub trait DriverObject: Sync + Send + Sized {
     /// Parent `Driver` for this object.
@@ -23,19 +30,12 @@ pub trait DriverObject: Sync + Send + Sized {
     fn new(dev: &drm::Device<Self::Driver>, size: usize) -> impl PinInit<Self, Error>;
 
     /// Open a new handle to an existing object, associated with a File.
-    fn open(
-        _obj: &<Self::Driver as drm::Driver>::Object,
-        _file: &drm::File<<Self::Driver as drm::Driver>::File>,
-    ) -> Result {
+    fn open(_obj: &<Self::Driver as drm::Driver>::Object, _file: &DriverFile<Self>) -> Result {
         Ok(())
     }
 
     /// Close a handle to an existing object, associated with a File.
-    fn close(
-        _obj: &<Self::Driver as drm::Driver>::Object,
-        _file: &drm::File<<Self::Driver as drm::Driver>::File>,
-    ) {
-    }
+    fn close(_obj: &<Self::Driver as drm::Driver>::Object, _file: &DriverFile<Self>) {}
 }
 
 /// Trait that represents a GEM object subtype
@@ -79,7 +79,8 @@ extern "C" fn open_callback<T: DriverObject>(
     raw_file: *mut bindings::drm_file,
 ) -> core::ffi::c_int {
     // SAFETY: `open_callback` is only ever called with a valid pointer to a `struct drm_file`.
-    let file = unsafe { drm::File::<<T::Driver as drm::Driver>::File>::from_raw(raw_file) };
+    let file = unsafe { DriverFile::<T>::from_raw(raw_file) };
+
     // SAFETY: `open_callback` is specified in the AllocOps structure for `DriverObject<T>`,
     // ensuring that `raw_obj` is contained within a `DriverObject<T>`
     let obj = unsafe { <<T::Driver as drm::Driver>::Object as IntoGEMObject>::from_raw(raw_obj) };
@@ -95,7 +96,7 @@ extern "C" fn close_callback<T: DriverObject>(
     raw_file: *mut bindings::drm_file,
 ) {
     // SAFETY: `open_callback` is only ever called with a valid pointer to a `struct drm_file`.
-    let file = unsafe { drm::File::<<T::Driver as drm::Driver>::File>::from_raw(raw_file) };
+    let file = unsafe { DriverFile::<T>::from_raw(raw_file) };
 
     // SAFETY: `close_callback` is specified in the AllocOps structure for `Object<T>`, ensuring
     // that `raw_obj` is indeed contained within a `Object<T>`.

From 6b35936f058d0cb9171c7be1424b62017b874913 Mon Sep 17 00:00:00 2001
From: Lyude Paul <lyude@redhat.com>
Date: Mon, 8 Sep 2025 14:46:38 -0400
Subject: [PATCH 1115/1455] rust: drm: gem: Drop Object::SIZE

Drive-by fix, it doesn't seem like anything actually uses this constant
anymore.

Signed-off-by: Lyude Paul <lyude@redhat.com>
Reviewed-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Daniel Almeida <daniel.almeida@collabora.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20250908185239.135849-4-lyude@redhat.com
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
---
 rust/kernel/drm/gem/mod.rs | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/rust/kernel/drm/gem/mod.rs b/rust/kernel/drm/gem/mod.rs
index ead723980b27..fd872de3b669 100644
--- a/rust/kernel/drm/gem/mod.rs
+++ b/rust/kernel/drm/gem/mod.rs
@@ -12,7 +12,7 @@ use crate::{
     prelude::*,
     types::{ARef, AlwaysRefCounted, Opaque},
 };
-use core::{mem, ops::Deref, ptr::NonNull};
+use core::{ops::Deref, ptr::NonNull};
 
 /// A type alias for retrieving a [`Driver`]s [`DriverFile`] implementation from its
 /// [`DriverObject`] implementation.
@@ -197,9 +197,6 @@ pub struct Object<T: DriverObject + Send + Sync> {
 }
 
 impl<T: DriverObject> Object<T> {
-    /// The size of this object's structure.
-    pub const SIZE: usize = mem::size_of::<Self>();
-
     const OBJECT_FUNCS: bindings::drm_gem_object_funcs = bindings::drm_gem_object_funcs {
         free: Some(Self::free_callback),
         open: Some(open_callback::<T>),

From 2c4f536c75217476baabbd557a44e8d4c3a2a23a Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Mon, 8 Sep 2025 15:04:18 +0200
Subject: [PATCH 1116/1455] dt-bindings: display: bridge: simple: document the
 Realtek RTD2171 DP-to-HDMI bridge

The Realtek RTD2171 chipset is a transparent DisplayPort 1.4 to
HDMI 2.0 bridge.

This chipset is usually found in USB-C To HDMI Adapters and Docks,
or laptops to provide HDMI display output.

Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250908-topic-x1e80100-hdmi-v3-1-c53b0f2bc2fb@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../devicetree/bindings/display/bridge/simple-bridge.yaml        | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml b/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml
index 421f99ca42d9..9ef587d46506 100644
--- a/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/simple-bridge.yaml
@@ -29,6 +29,7 @@ properties:
           - adi,adv7123
           - dumb-vga-dac
           - radxa,ra620
+          - realtek,rtd2171
           - ti,opa362
           - ti,ths8134
           - ti,ths8135

From 7156602d56e5ad689ae11e03680ab6326238b5e3 Mon Sep 17 00:00:00 2001
From: Neil Armstrong <neil.armstrong@linaro.org>
Date: Mon, 8 Sep 2025 15:04:19 +0200
Subject: [PATCH 1117/1455] drm/bridge: simple: add Realtek RTD2171 DP-to-HDMI
 bridge

Add support for the transparent Realtek RTD2171 DP-to-HDMI bridge.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20250908-topic-x1e80100-hdmi-v3-2-c53b0f2bc2fb@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/simple-bridge.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/bridge/simple-bridge.c b/drivers/gpu/drm/bridge/simple-bridge.c
index 1f16d568bcc4..e4d0bc2200f8 100644
--- a/drivers/gpu/drm/bridge/simple-bridge.c
+++ b/drivers/gpu/drm/bridge/simple-bridge.c
@@ -266,6 +266,11 @@ static const struct of_device_id simple_bridge_match[] = {
 		.data = &(const struct simple_bridge_info) {
 			.connector_type = DRM_MODE_CONNECTOR_HDMIA,
 		},
+	}, {
+		.compatible = "realtek,rtd2171",
+		.data = &(const struct simple_bridge_info) {
+			.connector_type = DRM_MODE_CONNECTOR_HDMIA,
+		},
 	}, {
 		.compatible = "ti,opa362",
 		.data = &(const struct simple_bridge_info) {

From 854c0dc80e25a408ecf7cf27661c28b8c10c291e Mon Sep 17 00:00:00 2001
From: Suraj Kandpal <suraj.kandpal@intel.com>
Date: Thu, 4 Sep 2025 10:18:04 +0530
Subject: [PATCH 1118/1455] drm/i915/backlight: Disable backlight when using
 luminance control

We just return when using luminance control instead we should be
calling the disable helper to get everything cleaned up properly
by default.

Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Reviewed-by: Arun R Murthy <arun.r.murthy@intel.com>
Link: https://lore.kernel.org/r/20250904044804.948391-1-suraj.kandpal@intel.com
---
 drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
index f665d845c809..0cf49219c44d 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
@@ -550,9 +550,6 @@ static void intel_dp_aux_vesa_disable_backlight(const struct drm_connector_state
 	struct intel_panel *panel = &connector->panel;
 	struct intel_dp *intel_dp = enc_to_intel_dp(connector->encoder);
 
-	if (panel->backlight.edp.vesa.luminance_control_support)
-		return;
-
 	drm_edp_backlight_disable(&intel_dp->aux, &panel->backlight.edp.vesa.info);
 
 	if (!panel->backlight.edp.vesa.info.aux_enable)

From f985e105ced10304abd3b2d822994b87c88145f8 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Wed, 3 Sep 2025 12:04:08 +0300
Subject: [PATCH 1119/1455] drm/i915/display: add
 intel_display_device_present()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a proper function for display && HAS_DISPLAY(display) to hide
indirect struct intel_display access via the macro from a number of
places outside of display. This makes struct intel_display * an opaque
pointer in these places. All HAS_DISPLAY() usage is now constrained
within display.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250903090408.3492875-1-jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../drm/i915/display/intel_display_device.c   |  5 +++
 .../drm/i915/display/intel_display_device.h   |  1 +
 .../i915/gem/selftests/i915_gem_client_blt.c  |  4 +--
 drivers/gpu/drm/i915/i915_driver.c            | 18 +++++-----
 drivers/gpu/drm/i915/i915_switcheroo.c        |  7 ++--
 drivers/gpu/drm/i915/soc/intel_dram.c         |  2 +-
 drivers/gpu/drm/xe/display/xe_display.c       | 33 ++++++++-----------
 7 files changed, 36 insertions(+), 34 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display_device.c b/drivers/gpu/drm/i915/display/intel_display_device.c
index 65f0efc35bb7..a002bc6ce7b0 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.c
+++ b/drivers/gpu/drm/i915/display/intel_display_device.c
@@ -1944,6 +1944,11 @@ void intel_display_device_info_print(const struct intel_display_device_info *inf
 	drm_printf(p, "rawclk rate: %u kHz\n", runtime->rawclk_freq);
 }
 
+bool intel_display_device_present(struct intel_display *display)
+{
+	return display && HAS_DISPLAY(display);
+}
+
 /*
  * Assuming the device has display hardware, should it be enabled?
  *
diff --git a/drivers/gpu/drm/i915/display/intel_display_device.h b/drivers/gpu/drm/i915/display/intel_display_device.h
index 6e87b763fe7c..f329f1beafef 100644
--- a/drivers/gpu/drm/i915/display/intel_display_device.h
+++ b/drivers/gpu/drm/i915/display/intel_display_device.h
@@ -306,6 +306,7 @@ struct intel_display_device_info {
 	} color;
 };
 
+bool intel_display_device_present(struct intel_display *display);
 bool intel_display_device_enabled(struct intel_display *display);
 struct intel_display *intel_display_device_probe(struct pci_dev *pdev);
 void intel_display_device_remove(struct intel_display *display);
diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
index e747f5ed195e..539c620364e3 100644
--- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
+++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c
@@ -5,7 +5,7 @@
 
 #include "i915_selftest.h"
 
-#include "display/intel_display_core.h"
+#include "display/intel_display_device.h"
 #include "gt/intel_context.h"
 #include "gt/intel_engine_regs.h"
 #include "gt/intel_engine_user.h"
@@ -122,7 +122,7 @@ static bool fastblit_supports_x_tiling(const struct drm_i915_private *i915)
 	if (GRAPHICS_VER_FULL(i915) < IP_VER(12, 55))
 		return false;
 
-	return HAS_DISPLAY(display);
+	return intel_display_device_present(display);
 }
 
 static bool fast_blit_ok(const struct blit_buffer *buf)
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 70f042ce8705..a28c3710c4d5 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -51,13 +51,15 @@
 #include "display/intel_bw.h"
 #include "display/intel_cdclk.h"
 #include "display/intel_crtc.h"
-#include "display/intel_display_core.h"
+#include "display/intel_display_device.h"
 #include "display/intel_display_driver.h"
+#include "display/intel_display_power.h"
 #include "display/intel_dmc.h"
 #include "display/intel_dp.h"
 #include "display/intel_dpt.h"
 #include "display/intel_encoder.h"
 #include "display/intel_fbdev.h"
+#include "display/intel_gmbus.h"
 #include "display/intel_hotplug.h"
 #include "display/intel_opregion.h"
 #include "display/intel_overlay.h"
@@ -977,7 +979,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915)
 	intel_power_domains_disable(display);
 
 	drm_client_dev_suspend(&i915->drm, false);
-	if (HAS_DISPLAY(display)) {
+	if (intel_display_device_present(display)) {
 		drm_kms_helper_poll_disable(&i915->drm);
 		intel_display_driver_disable_user_access(display);
 
@@ -989,7 +991,7 @@ void i915_driver_shutdown(struct drm_i915_private *i915)
 	intel_irq_suspend(i915);
 	intel_hpd_cancel_work(display);
 
-	if (HAS_DISPLAY(display))
+	if (intel_display_device_present(display))
 		intel_display_driver_suspend_access(display);
 
 	intel_encoder_suspend_all(display);
@@ -1060,7 +1062,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 	 * properly. */
 	intel_power_domains_disable(display);
 	drm_client_dev_suspend(dev, false);
-	if (HAS_DISPLAY(display)) {
+	if (intel_display_device_present(display)) {
 		drm_kms_helper_poll_disable(dev);
 		intel_display_driver_disable_user_access(display);
 	}
@@ -1072,7 +1074,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 	intel_irq_suspend(dev_priv);
 	intel_hpd_cancel_work(display);
 
-	if (HAS_DISPLAY(display))
+	if (intel_display_device_present(display))
 		intel_display_driver_suspend_access(display);
 
 	intel_encoder_suspend_all(display);
@@ -1219,7 +1221,7 @@ static int i915_drm_resume(struct drm_device *dev)
 	 */
 	intel_irq_resume(dev_priv);
 
-	if (HAS_DISPLAY(display))
+	if (intel_display_device_present(display))
 		drm_mode_config_reset(dev);
 
 	i915_gem_resume(dev_priv);
@@ -1228,14 +1230,14 @@ static int i915_drm_resume(struct drm_device *dev)
 
 	intel_clock_gating_init(dev_priv);
 
-	if (HAS_DISPLAY(display))
+	if (intel_display_device_present(display))
 		intel_display_driver_resume_access(display);
 
 	intel_hpd_init(display);
 
 	intel_display_driver_resume(display);
 
-	if (HAS_DISPLAY(display)) {
+	if (intel_display_device_present(display)) {
 		intel_display_driver_enable_user_access(display);
 		drm_kms_helper_poll_enable(dev);
 	}
diff --git a/drivers/gpu/drm/i915/i915_switcheroo.c b/drivers/gpu/drm/i915/i915_switcheroo.c
index 3a95a55b2e87..d5b6d8ab31a2 100644
--- a/drivers/gpu/drm/i915/i915_switcheroo.c
+++ b/drivers/gpu/drm/i915/i915_switcheroo.c
@@ -5,7 +5,7 @@
 
 #include <linux/vga_switcheroo.h>
 
-#include "display/intel_display_core.h"
+#include "display/intel_display_device.h"
 
 #include "i915_driver.h"
 #include "i915_drv.h"
@@ -22,7 +22,7 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev,
 		dev_err(&pdev->dev, "DRM not initialized, aborting switch.\n");
 		return;
 	}
-	if (!HAS_DISPLAY(display)) {
+	if (!intel_display_device_present(display)) {
 		dev_err(&pdev->dev, "Device state not initialized, aborting switch.\n");
 		return;
 	}
@@ -52,7 +52,8 @@ static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
 	 * locking inversion with the driver load path. And the access here is
 	 * completely racy anyway. So don't bother with locking for now.
 	 */
-	return i915 && HAS_DISPLAY(display) && atomic_read(&i915->drm.open_count) == 0;
+	return i915 && intel_display_device_present(display) &&
+		atomic_read(&i915->drm.open_count) == 0;
 }
 
 static const struct vga_switcheroo_client_ops i915_switcheroo_ops = {
diff --git a/drivers/gpu/drm/i915/soc/intel_dram.c b/drivers/gpu/drm/i915/soc/intel_dram.c
index 149527827624..edffaed8f9a7 100644
--- a/drivers/gpu/drm/i915/soc/intel_dram.c
+++ b/drivers/gpu/drm/i915/soc/intel_dram.c
@@ -732,7 +732,7 @@ int intel_dram_detect(struct drm_i915_private *i915)
 	struct dram_info *dram_info;
 	int ret;
 
-	if (IS_DG2(i915) || !HAS_DISPLAY(display))
+	if (IS_DG2(i915) || !intel_display_device_present(display))
 		return 0;
 
 	dram_info = drmm_kzalloc(&i915->drm, sizeof(*dram_info), GFP_KERNEL);
diff --git a/drivers/gpu/drm/xe/display/xe_display.c b/drivers/gpu/drm/xe/display/xe_display.c
index 8b68d70db6c8..19e691fccf8c 100644
--- a/drivers/gpu/drm/xe/display/xe_display.c
+++ b/drivers/gpu/drm/xe/display/xe_display.c
@@ -20,7 +20,7 @@
 #include "intel_audio.h"
 #include "intel_bw.h"
 #include "intel_display.h"
-#include "intel_display_core.h"
+#include "intel_display_device.h"
 #include "intel_display_driver.h"
 #include "intel_display_irq.h"
 #include "intel_display_types.h"
@@ -37,13 +37,6 @@
 
 /* Xe device functions */
 
-static bool has_display(struct xe_device *xe)
-{
-	struct intel_display *display = xe->display;
-
-	return HAS_DISPLAY(display);
-}
-
 /**
  * xe_display_driver_probe_defer - Detect if we need to wait for other drivers
  *				   early on
@@ -290,7 +283,7 @@ static void xe_display_enable_d3cold(struct xe_device *xe)
 
 	intel_dmc_suspend(display);
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		intel_hpd_poll_enable(display);
 }
 
@@ -303,14 +296,14 @@ static void xe_display_disable_d3cold(struct xe_device *xe)
 
 	intel_dmc_resume(display);
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		drm_mode_config_reset(&xe->drm);
 
 	intel_display_driver_init_hw(display);
 
 	intel_hpd_init(display);
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		intel_hpd_poll_disable(display);
 
 	intel_opregion_resume(display);
@@ -333,7 +326,7 @@ void xe_display_pm_suspend(struct xe_device *xe)
 	intel_power_domains_disable(display);
 	drm_client_dev_suspend(&xe->drm, false);
 
-	if (has_display(xe)) {
+	if (intel_display_device_present(display)) {
 		drm_kms_helper_poll_disable(&xe->drm);
 		intel_display_driver_disable_user_access(display);
 		intel_display_driver_suspend(display);
@@ -345,7 +338,7 @@ void xe_display_pm_suspend(struct xe_device *xe)
 
 	intel_hpd_cancel_work(display);
 
-	if (has_display(xe)) {
+	if (intel_display_device_present(display)) {
 		intel_display_driver_suspend_access(display);
 		intel_encoder_suspend_all(display);
 	}
@@ -365,7 +358,7 @@ void xe_display_pm_shutdown(struct xe_device *xe)
 	intel_power_domains_disable(display);
 	drm_client_dev_suspend(&xe->drm, false);
 
-	if (has_display(xe)) {
+	if (intel_display_device_present(display)) {
 		drm_kms_helper_poll_disable(&xe->drm);
 		intel_display_driver_disable_user_access(display);
 		intel_display_driver_suspend(display);
@@ -376,7 +369,7 @@ void xe_display_pm_shutdown(struct xe_device *xe)
 	intel_encoder_block_all_hpds(display);
 	intel_hpd_cancel_work(display);
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		intel_display_driver_suspend_access(display);
 
 	intel_encoder_suspend_all(display);
@@ -465,25 +458,25 @@ void xe_display_pm_resume(struct xe_device *xe)
 
 	intel_dmc_resume(display);
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		drm_mode_config_reset(&xe->drm);
 
 	intel_display_driver_init_hw(display);
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		intel_display_driver_resume_access(display);
 
 	intel_hpd_init(display);
 
 	intel_encoder_unblock_all_hpds(display);
 
-	if (has_display(xe)) {
+	if (intel_display_device_present(display)) {
 		intel_display_driver_resume(display);
 		drm_kms_helper_poll_enable(&xe->drm);
 		intel_display_driver_enable_user_access(display);
 	}
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		intel_hpd_poll_disable(display);
 
 	intel_opregion_resume(display);
@@ -548,7 +541,7 @@ int xe_display_probe(struct xe_device *xe)
 
 	xe->display = display;
 
-	if (has_display(xe))
+	if (intel_display_device_present(display))
 		return 0;
 
 no_display:

From e4aae21c9237ce4f05d40dbdda950f9b92acfd37 Mon Sep 17 00:00:00 2001
From: Thomas Zimmermann <tzimmermann@suse.de>
Date: Thu, 4 Sep 2025 14:09:38 +0200
Subject: [PATCH 1120/1455] drm/gma500: Do not clear framebuffer GEM objects
 during cleanup

Gma500 unnecessarily clears the framebuffer's GEM-object pointer
before calling drm_framebuffer_cleanup(). Remove this code to make
gma500 consistent with the rest of the drivers.

The change is cosmetic, as drm_framebuffer_cleanup() does not
touch the object pointer on gma500.

Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Link: https://lore.kernel.org/r/20250904121157.395128-1-tzimmermann@suse.de
---
 drivers/gpu/drm/gma500/fbdev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/gma500/fbdev.c b/drivers/gpu/drm/gma500/fbdev.c
index 4a37136f90f4..32d31e5f5f1a 100644
--- a/drivers/gpu/drm/gma500/fbdev.c
+++ b/drivers/gpu/drm/gma500/fbdev.c
@@ -120,7 +120,6 @@ static void psb_fbdev_fb_destroy(struct fb_info *info)
 	drm_fb_helper_fini(fb_helper);
 
 	drm_framebuffer_unregister_private(fb);
-	fb->obj[0] = NULL;
 	drm_framebuffer_cleanup(fb);
 	kfree(fb);
 
@@ -245,7 +244,6 @@ int psb_fbdev_driver_fbdev_probe(struct drm_fb_helper *fb_helper,
 
 err_drm_framebuffer_unregister_private:
 	drm_framebuffer_unregister_private(fb);
-	fb->obj[0] = NULL;
 	drm_framebuffer_cleanup(fb);
 	kfree(fb);
 err_drm_gem_object_put:

From dd432009f19671f3d474f0db437333433cb1a27d Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:18 +0200
Subject: [PATCH 1121/1455] drm/xe/guc: Rename xe_guc_register_exec_queue

This function is dedicated for use by the VFs, we shouldn't use
name that might suggests it's general purpose. While there, update
asserts to better reflect intended usage.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-2-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_guc_submit.c   | 14 ++++++++------
 drivers/gpu/drm/xe/xe_guc_submit.h   |  2 +-
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c |  5 ++---
 3 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 897f33ded8ec..2b8da44493ae 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -2528,7 +2528,7 @@ static void guc_exec_queue_print(struct xe_exec_queue *q, struct drm_printer *p)
 }
 
 /**
- * xe_guc_register_exec_queue - Register exec queue for a given context type.
+ * xe_guc_register_vf_exec_queue - Register exec queue for a given context type.
  * @q: Execution queue
  * @ctx_type: Type of the context
  *
@@ -2539,15 +2539,17 @@ static void guc_exec_queue_print(struct xe_exec_queue *q, struct drm_printer *p)
  *
  * Returns - None.
  */
-void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type)
+void xe_guc_register_vf_exec_queue(struct xe_exec_queue *q, int ctx_type)
 {
 	struct xe_guc *guc = exec_queue_to_guc(q);
 	struct xe_device *xe = guc_to_xe(guc);
+	struct xe_gt *gt = guc_to_gt(guc);
 
-	xe_assert(xe, IS_SRIOV_VF(xe));
-	xe_assert(xe, !IS_DGFX(xe));
-	xe_assert(xe, (ctx_type > GUC_CONTEXT_NORMAL &&
-		       ctx_type < GUC_CONTEXT_COUNT));
+	xe_gt_assert(gt, IS_SRIOV_VF(xe));
+	xe_gt_assert(gt, !IS_DGFX(xe));
+	xe_gt_assert(gt, ctx_type == GUC_CONTEXT_COMPRESSION_SAVE ||
+		     ctx_type == GUC_CONTEXT_COMPRESSION_RESTORE);
+	xe_gt_assert(gt, GUC_SUBMIT_VER(guc) >= MAKE_GUC_VER(1, 23, 0));
 
 	register_exec_queue(q, ctx_type);
 	enable_scheduling(q);
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 6b5df5d0956b..9e6f19b03e65 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -46,7 +46,7 @@ xe_guc_exec_queue_snapshot_print(struct xe_guc_submit_exec_queue_snapshot *snaps
 void
 xe_guc_exec_queue_snapshot_free(struct xe_guc_submit_exec_queue_snapshot *snapshot);
 void xe_guc_submit_print(struct xe_guc *guc, struct drm_printer *p);
-void xe_guc_register_exec_queue(struct xe_exec_queue *q, int ctx_type);
+void xe_guc_register_vf_exec_queue(struct xe_exec_queue *q, int ctx_type);
 
 int xe_guc_contexts_hwsp_rebase(struct xe_guc *guc, void *scratch);
 
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 908590fa79d4..19786f64a7da 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -186,7 +186,6 @@ static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
 
 static int register_save_restore_context(struct xe_tile_vf_ccs *ctx)
 {
-	int err = -EINVAL;
 	int ctx_type;
 
 	switch (ctx->ctx_id) {
@@ -197,10 +196,10 @@ static int register_save_restore_context(struct xe_tile_vf_ccs *ctx)
 		ctx_type = GUC_CONTEXT_COMPRESSION_RESTORE;
 		break;
 	default:
-		return err;
+		return -EINVAL;
 	}
 
-	xe_guc_register_exec_queue(ctx->mig_q, ctx_type);
+	xe_guc_register_vf_exec_queue(ctx->mig_q, ctx_type);
 	return 0;
 }
 

From 4e5bc50ad207bc9849396593e7e4a0984299a3a0 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:19 +0200
Subject: [PATCH 1122/1455] drm/xe/guc: Use proper flag definitions when
 registering context

In H2G action context type is specified in flags dword in bits 2:1.
Use generic FIELD_PREP macro instead of misleading BIT logic.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_guc_fwif.h   | 1 +
 drivers/gpu/drm/xe/xe_guc_submit.c | 6 ++----
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index 457d914097c8..a72ff8a68568 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -65,6 +65,7 @@ struct guc_ctxt_registration_info {
 	u32 hwlrca_hi;
 };
 #define CONTEXT_REGISTRATION_FLAG_KMD	BIT(0)
+#define CONTEXT_REGISTRATION_FLAG_TYPE	GENMASK(2, 1)
 
 /* 32-bit KLV structure as used by policy updates and others */
 struct guc_klv_generic_dw_t {
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 2b8da44493ae..5ab242be50dd 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -558,10 +558,8 @@ static void register_exec_queue(struct xe_exec_queue *q, int ctx_type)
 	info.engine_submit_mask = q->logical_mask;
 	info.hwlrca_lo = lower_32_bits(xe_lrc_descriptor(lrc));
 	info.hwlrca_hi = upper_32_bits(xe_lrc_descriptor(lrc));
-	info.flags = CONTEXT_REGISTRATION_FLAG_KMD;
-
-	if (ctx_type != GUC_CONTEXT_NORMAL)
-		info.flags |= BIT(ctx_type);
+	info.flags = CONTEXT_REGISTRATION_FLAG_KMD |
+		FIELD_PREP(CONTEXT_REGISTRATION_FLAG_TYPE, ctx_type);
 
 	if (xe_exec_queue_is_parallel(q)) {
 		u64 ggtt_addr = xe_lrc_parallel_ggtt_addr(lrc);

From aa8d9d75ea9ba6cb827ff96228b0204783f1d706 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:20 +0200
Subject: [PATCH 1123/1455] drm/xe/vf: Drop IS_VF_CCS_INIT_NEEDED macro

We only use this macro once and we can open-code it to explicitly
show relevant conditions and avoid duplications.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-4-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       | 6 +++++-
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h | 6 ------
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 19786f64a7da..b01bb3660fb1 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -15,6 +15,7 @@
 #include "xe_migrate.h"
 #include "xe_sa.h"
 #include "xe_sriov_printk.h"
+#include "xe_sriov_vf.h"
 #include "xe_sriov_vf_ccs.h"
 #include "xe_sriov_vf_ccs_types.h"
 
@@ -263,7 +264,10 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 	u32 flags;
 	int err;
 
-	if (!IS_VF_CCS_INIT_NEEDED(xe))
+	xe_assert(xe, IS_SRIOV_VF(xe));
+	xe_assert(xe, xe_sriov_vf_migration_supported(xe));
+
+	if (IS_DGFX(xe) || !xe_device_has_flat_ccs(xe))
 		return 0;
 
 	for_each_ccs_rw_ctx(ctx_id) {
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
index 93435a6f4cb6..0df6b4130e7c 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -15,12 +15,6 @@
 		___xe->sriov.vf.ccs.initialized; \
 		})
 
-#define IS_VF_CCS_INIT_NEEDED(xe) ({\
-		struct xe_device *___xe = (xe); \
-		IS_SRIOV_VF(___xe) && !IS_DGFX(___xe) && \
-		xe_device_has_flat_ccs(___xe) && GRAPHICS_VER(___xe) >= 20; \
-		})
-
 enum xe_sriov_vf_ccs_rw_ctxs {
 	XE_SRIOV_VF_CCS_READ_CTX,
 	XE_SRIOV_VF_CCS_WRITE_CTX,

From b179dfd0dbedf314f0f1ba5e038282dfe6dd091e Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:21 +0200
Subject: [PATCH 1124/1455] drm/xe/vf: Use single check when calling VF CCS
 functions

All xe_sriov_vf_ccs() functions but init() expect to be called
when initialization was successful and CCS handling is ready.

Update IS_VF_CCS_READY macro and use it as single entry guard.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-5-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_bo.c                 |  8 ++++----
 drivers/gpu/drm/xe/xe_gt_debugfs.c         |  1 +
 drivers/gpu/drm/xe/xe_pm.c                 |  4 ++--
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       | 10 +++++-----
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.h       | 15 +++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h |  6 ------
 6 files changed, 27 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index f03f7432c6fa..4efc1c8d15b8 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -974,11 +974,11 @@ static int xe_bo_move(struct ttm_buffer_object *ttm_bo, bool evict,
 	 * CCS meta data is migrated from TT -> SMEM. So, let us detach the
 	 * BBs from BO as it is no longer needed.
 	 */
-	if (IS_VF_CCS_BB_VALID(xe, bo) && old_mem_type == XE_PL_TT &&
+	if (IS_VF_CCS_READY(xe) && old_mem_type == XE_PL_TT &&
 	    new_mem->mem_type == XE_PL_SYSTEM)
 		xe_sriov_vf_ccs_detach_bo(bo);
 
-	if (IS_SRIOV_VF(xe) &&
+	if (IS_VF_CCS_READY(xe) &&
 	    ((move_lacks_source && new_mem->mem_type == XE_PL_TT) ||
 	     (old_mem_type == XE_PL_SYSTEM && new_mem->mem_type == XE_PL_TT)) &&
 	    handle_system_ccs)
@@ -994,7 +994,7 @@ out:
 		if (timeout < 0)
 			ret = timeout;
 
-		if (IS_VF_CCS_BB_VALID(xe, bo))
+		if (IS_VF_CCS_READY(xe))
 			xe_sriov_vf_ccs_detach_bo(bo);
 
 		xe_tt_unmap_sg(xe, ttm_bo->ttm);
@@ -1518,7 +1518,7 @@ static void xe_ttm_bo_delete_mem_notify(struct ttm_buffer_object *ttm_bo)
 	if (!xe_bo_is_xe_bo(ttm_bo))
 		return;
 
-	if (IS_VF_CCS_BB_VALID(ttm_to_xe_device(ttm_bo->bdev), bo))
+	if (IS_VF_CCS_READY(ttm_to_xe_device(ttm_bo->bdev)))
 		xe_sriov_vf_ccs_detach_bo(bo);
 
 	/*
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index 5aa1eded278d..f6f2c14b642d 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -31,6 +31,7 @@
 #include "xe_reg_whitelist.h"
 #include "xe_sa.h"
 #include "xe_sriov.h"
+#include "xe_sriov_vf_ccs.h"
 #include "xe_tuning.h"
 #include "xe_uc_debugfs.h"
 #include "xe_wa.h"
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index 6eea4190bbd2..a303ed5780b4 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -213,7 +213,7 @@ int xe_pm_resume(struct xe_device *xe)
 
 	xe_pxp_pm_resume(xe->pxp);
 
-	if (IS_SRIOV_VF(xe))
+	if (IS_VF_CCS_READY(xe))
 		xe_sriov_vf_ccs_register_context(xe);
 
 	drm_dbg(&xe->drm, "Device resumed\n");
@@ -598,7 +598,7 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 
 	xe_pxp_pm_resume(xe->pxp);
 
-	if (IS_SRIOV_VF(xe))
+	if (IS_VF_CCS_READY(xe))
 		xe_sriov_vf_ccs_register_context(xe);
 
 out:
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index b01bb3660fb1..30aea958a337 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -220,8 +220,7 @@ int xe_sriov_vf_ccs_register_context(struct xe_device *xe)
 	struct xe_tile_vf_ccs *ctx;
 	int err;
 
-	if (!IS_VF_CCS_READY(xe))
-		return 0;
+	xe_assert(xe, IS_VF_CCS_READY(xe));
 
 	for_each_ccs_rw_ctx(ctx_id) {
 		ctx = &tile->sriov.vf.ccs[ctx_id];
@@ -331,8 +330,7 @@ int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo)
 	struct xe_bb *bb;
 	int err = 0;
 
-	if (!IS_VF_CCS_READY(xe))
-		return 0;
+	xe_assert(xe, IS_VF_CCS_READY(xe));
 
 	tile = xe_device_get_root_tile(xe);
 
@@ -363,7 +361,9 @@ int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo)
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
 	struct xe_bb *bb;
 
-	if (!IS_VF_CCS_READY(xe))
+	xe_assert(xe, IS_VF_CCS_READY(xe));
+
+	if (!IS_VF_CCS_BB_VALID(xe, bo))
 		return 0;
 
 	for_each_ccs_rw_ctx(ctx_id) {
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
index 1f1baf685fec..f0e1189b417a 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
@@ -6,6 +6,10 @@
 #ifndef _XE_SRIOV_VF_CCS_H_
 #define _XE_SRIOV_VF_CCS_H_
 
+#include "xe_device_types.h"
+#include "xe_sriov.h"
+#include "xe_sriov_vf_ccs_types.h"
+
 struct xe_device;
 struct xe_bo;
 
@@ -14,4 +18,15 @@ int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
 int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
 int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
 
+static inline bool xe_sriov_vf_ccs_ready(struct xe_device *xe)
+{
+	xe_assert(xe, IS_SRIOV_VF(xe));
+	return xe->sriov.vf.ccs.initialized;
+}
+
+#define IS_VF_CCS_READY(xe) ({ \
+	struct xe_device *xe__ = (xe); \
+	IS_SRIOV_VF(xe__) && xe_sriov_vf_ccs_ready(xe__); \
+	})
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
index 0df6b4130e7c..79092e386c4a 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -9,12 +9,6 @@
 #define for_each_ccs_rw_ctx(id__) \
 	for ((id__) = 0; (id__) < XE_SRIOV_VF_CCS_CTX_COUNT; (id__)++)
 
-#define IS_VF_CCS_READY(xe) ({ \
-		struct xe_device *___xe = (xe); \
-		xe_assert(___xe, IS_SRIOV_VF(___xe)); \
-		___xe->sriov.vf.ccs.initialized; \
-		})
-
 enum xe_sriov_vf_ccs_rw_ctxs {
 	XE_SRIOV_VF_CCS_READ_CTX,
 	XE_SRIOV_VF_CCS_WRITE_CTX,

From e69970083438d7d59599f6f5502d8a87902e70b1 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:22 +0200
Subject: [PATCH 1125/1455] drm/xe/bo: Add xe_bo_has_valid_ccs_bb helper

This will allow as to drop ugly IS_VF_CCS_BB_VALID macro.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-6-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_bo.h                 | 15 +++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       |  2 +-
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h |  8 --------
 3 files changed, 16 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index cfb1ec266a6d..c7ee9c7605e5 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -315,6 +315,21 @@ static inline size_t xe_bo_ccs_pages_start(struct xe_bo *bo)
 	return PAGE_ALIGN(xe_bo_size(bo));
 }
 
+/**
+ * xe_bo_has_valid_ccs_bb - Check if CCS's BBs were setup for the BO.
+ * @bo: the &xe_bo to check
+ *
+ * The CCS's BBs should only be setup by the driver VF, but it is safe
+ * to call this function also by non-VF driver.
+ *
+ * Return: true iff the CCS's BBs are setup, false otherwise.
+ */
+static inline bool xe_bo_has_valid_ccs_bb(struct xe_bo *bo)
+{
+	return bo->bb_ccs[XE_SRIOV_VF_CCS_READ_CTX] &&
+	       bo->bb_ccs[XE_SRIOV_VF_CCS_WRITE_CTX];
+}
+
 static inline bool xe_bo_has_pages(struct xe_bo *bo)
 {
 	if ((bo->ttm.ttm && ttm_tt_is_populated(bo->ttm.ttm)) ||
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index 30aea958a337..eb8436e44ca4 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -363,7 +363,7 @@ int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo)
 
 	xe_assert(xe, IS_VF_CCS_READY(xe));
 
-	if (!IS_VF_CCS_BB_VALID(xe, bo))
+	if (!xe_bo_has_valid_ccs_bb(bo))
 		return 0;
 
 	for_each_ccs_rw_ctx(ctx_id) {
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
index 79092e386c4a..4d3c10907135 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -15,14 +15,6 @@ enum xe_sriov_vf_ccs_rw_ctxs {
 	XE_SRIOV_VF_CCS_CTX_COUNT
 };
 
-#define IS_VF_CCS_BB_VALID(xe, bo) ({ \
-		struct xe_device *___xe = (xe); \
-		struct xe_bo *___bo = (bo); \
-		IS_SRIOV_VF(___xe) && \
-		___bo->bb_ccs[XE_SRIOV_VF_CCS_READ_CTX] && \
-		___bo->bb_ccs[XE_SRIOV_VF_CCS_WRITE_CTX]; \
-		})
-
 struct xe_migrate;
 struct xe_sa_manager;
 

From 55ddca2a3c0d2808293a7b6ae312a7f7c3a6e089 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:23 +0200
Subject: [PATCH 1126/1455] drm/xe/vf: Move VF CCS data to xe_device

We only need single set of VF CCS contexts, they are not per-tile
as initial implementation might suggest. Move all VF CCS data from
xe_tile.sriov.vf to xe_device.sriov.vf. Also rename some structs to
align with the usage and fix their kernel-doc.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-7-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_bb.c                 |  4 ++--
 drivers/gpu/drm/xe/xe_device_types.h       |  3 ---
 drivers/gpu/drm/xe/xe_gt_debugfs.c         |  4 ++--
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c       | 21 +++++++++----------
 drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h | 24 +++++++++++++++++++---
 drivers/gpu/drm/xe/xe_sriov_vf_types.h     |  7 +++----
 6 files changed, 38 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bb.c b/drivers/gpu/drm/xe/xe_bb.c
index feb6e013dc38..6d20229c11de 100644
--- a/drivers/gpu/drm/xe/xe_bb.c
+++ b/drivers/gpu/drm/xe/xe_bb.c
@@ -64,7 +64,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
 			    enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
 {
 	struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
-	struct xe_tile *tile = gt_to_tile(gt);
+	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_sa_manager *bb_pool;
 	int err;
 
@@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
 	 * So, this extra DW acts as a guard here.
 	 */
 
-	bb_pool = tile->sriov.vf.ccs[ctx_id].mem.ccs_bb_pool;
+	bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
 	bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
 
 	if (IS_ERR(bb->bo)) {
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 9e30dc7d6e58..a6ba880e4181 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -183,9 +183,6 @@ struct xe_tile {
 		struct {
 			/** @sriov.vf.ggtt_balloon: GGTT regions excluded from use. */
 			struct xe_ggtt_node *ggtt_balloon[2];
-
-			/** @sriov.vf.ccs: CCS read and write contexts for VF. */
-			struct xe_tile_vf_ccs ccs[XE_SRIOV_VF_CCS_CTX_COUNT];
 		} vf;
 	} sriov;
 
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index f6f2c14b642d..a9d960de0e5e 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -126,7 +126,7 @@ static int powergate_info(struct xe_gt *gt, struct drm_printer *p)
 
 static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p)
 {
-	struct xe_tile *tile = gt_to_tile(gt);
+	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_sa_manager *bb_pool;
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
 
@@ -136,7 +136,7 @@ static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p)
 	xe_pm_runtime_get(gt_to_xe(gt));
 
 	for_each_ccs_rw_ctx(ctx_id) {
-		bb_pool = tile->sriov.vf.ccs[ctx_id].mem.ccs_bb_pool;
+		bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
 		if (!bb_pool)
 			break;
 
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index eb8436e44ca4..c5c60f05073d 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -136,7 +136,7 @@ static u64 get_ccs_bb_pool_size(struct xe_device *xe)
 	return round_up(bb_pool_size * 2, SZ_1M);
 }
 
-static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx)
+static int alloc_bb_pool(struct xe_tile *tile, struct xe_sriov_vf_ccs_ctx *ctx)
 {
 	struct xe_device *xe = tile_to_xe(tile);
 	struct xe_sa_manager *sa_manager;
@@ -168,7 +168,7 @@ static int alloc_bb_pool(struct xe_tile *tile, struct xe_tile_vf_ccs *ctx)
 	return 0;
 }
 
-static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
+static void ccs_rw_update_ring(struct xe_sriov_vf_ccs_ctx *ctx)
 {
 	u64 addr = xe_sa_manager_gpu_addr(ctx->mem.ccs_bb_pool);
 	struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
@@ -185,7 +185,7 @@ static void ccs_rw_update_ring(struct xe_tile_vf_ccs *ctx)
 	xe_lrc_set_ring_tail(lrc, lrc->ring.tail);
 }
 
-static int register_save_restore_context(struct xe_tile_vf_ccs *ctx)
+static int register_save_restore_context(struct xe_sriov_vf_ccs_ctx *ctx)
 {
 	int ctx_type;
 
@@ -215,15 +215,14 @@ static int register_save_restore_context(struct xe_tile_vf_ccs *ctx)
  */
 int xe_sriov_vf_ccs_register_context(struct xe_device *xe)
 {
-	struct xe_tile *tile = xe_device_get_root_tile(xe);
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-	struct xe_tile_vf_ccs *ctx;
+	struct xe_sriov_vf_ccs_ctx *ctx;
 	int err;
 
 	xe_assert(xe, IS_VF_CCS_READY(xe));
 
 	for_each_ccs_rw_ctx(ctx_id) {
-		ctx = &tile->sriov.vf.ccs[ctx_id];
+		ctx = &xe->sriov.vf.ccs.contexts[ctx_id];
 		err = register_save_restore_context(ctx);
 		if (err)
 			return err;
@@ -234,7 +233,7 @@ int xe_sriov_vf_ccs_register_context(struct xe_device *xe)
 
 static void xe_sriov_vf_ccs_fini(void *arg)
 {
-	struct xe_tile_vf_ccs *ctx = arg;
+	struct xe_sriov_vf_ccs_ctx *ctx = arg;
 	struct xe_lrc *lrc = xe_exec_queue_lrc(ctx->mig_q);
 
 	/*
@@ -258,7 +257,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 {
 	struct xe_tile *tile = xe_device_get_root_tile(xe);
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-	struct xe_tile_vf_ccs *ctx;
+	struct xe_sriov_vf_ccs_ctx *ctx;
 	struct xe_exec_queue *q;
 	u32 flags;
 	int err;
@@ -270,7 +269,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe)
 		return 0;
 
 	for_each_ccs_rw_ctx(ctx_id) {
-		ctx = &tile->sriov.vf.ccs[ctx_id];
+		ctx = &xe->sriov.vf.ccs.contexts[ctx_id];
 		ctx->ctx_id = ctx_id;
 
 		flags = EXEC_QUEUE_FLAG_KERNEL |
@@ -325,7 +324,7 @@ int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo)
 {
 	struct xe_device *xe = xe_bo_device(bo);
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-	struct xe_tile_vf_ccs *ctx;
+	struct xe_sriov_vf_ccs_ctx *ctx;
 	struct xe_tile *tile;
 	struct xe_bb *bb;
 	int err = 0;
@@ -339,7 +338,7 @@ int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo)
 		/* bb should be NULL here. Assert if not NULL */
 		xe_assert(xe, !bb);
 
-		ctx = &tile->sriov.vf.ccs[ctx_id];
+		ctx = &xe->sriov.vf.ccs.contexts[ctx_id];
 		err = xe_migrate_ccs_rw_copy(tile, ctx->mig_q, bo, ctx_id);
 	}
 	return err;
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
index 4d3c10907135..22c499943d2a 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs_types.h
@@ -6,6 +6,8 @@
 #ifndef _XE_SRIOV_VF_CCS_TYPES_H_
 #define _XE_SRIOV_VF_CCS_TYPES_H_
 
+#include <linux/types.h>
+
 #define for_each_ccs_rw_ctx(id__) \
 	for ((id__) = 0; (id__) < XE_SRIOV_VF_CCS_CTX_COUNT; (id__)++)
 
@@ -18,16 +20,32 @@ enum xe_sriov_vf_ccs_rw_ctxs {
 struct xe_migrate;
 struct xe_sa_manager;
 
-struct xe_tile_vf_ccs {
-	/** @id: Id to which context it belongs to */
+/**
+ * struct xe_sriov_vf_ccs_ctx - VF CCS migration context data.
+ */
+struct xe_sriov_vf_ccs_ctx {
+	/** @ctx_id: Id to which context it belongs to */
 	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+
 	/** @mig_q: exec queues used for migration */
 	struct xe_exec_queue *mig_q;
 
+	/** @mem: memory data */
 	struct {
-		/** @ccs_bb_pool: Pool from which batch buffers are allocated. */
+		/** @mem.ccs_bb_pool: Pool from which batch buffers are allocated. */
 		struct xe_sa_manager *ccs_bb_pool;
 	} mem;
 };
 
+/**
+ * struct xe_sriov_vf_ccs - The VF CCS migration support data.
+ */
+struct xe_sriov_vf_ccs {
+	/** @contexts: CCS read and write contexts for VF. */
+	struct xe_sriov_vf_ccs_ctx contexts[XE_SRIOV_VF_CCS_CTX_COUNT];
+
+	/** @initialized: Initialization of VF CCS is completed or not. */
+	bool initialized;
+};
+
 #endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_types.h b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
index 2c94d1f92187..426cc5841958 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_types.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_types.h
@@ -9,6 +9,8 @@
 #include <linux/types.h>
 #include <linux/workqueue_types.h>
 
+#include "xe_sriov_vf_ccs_types.h"
+
 /**
  * struct xe_sriov_vf_relay_version - PF ABI version details.
  */
@@ -43,10 +45,7 @@ struct xe_device_vf {
 	} migration;
 
 	/** @ccs: VF CCS state data */
-	struct {
-		/** @ccs.initialized: Initilalization of VF CCS is completed or not */
-		bool initialized;
-	} ccs;
+	struct xe_sriov_vf_ccs ccs;
 };
 
 #endif

From fd548b77d5ef185138aa6dbc7aafdc44586bd9b5 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 14:30:24 +0200
Subject: [PATCH 1127/1455] drm/xe/vf: Move VF CCS debugfs attribute

The VF CCS handling is per-device so its debugfs file should not
be exposed on per-GT basis. Move it up to the device level.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908123025.747-8-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_debugfs.c      |  3 +++
 drivers/gpu/drm/xe/xe_gt_debugfs.c   | 39 ----------------------------
 drivers/gpu/drm/xe/xe_sriov_vf.c     | 28 ++++++++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf.h     |  2 ++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.c | 32 +++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_sriov_vf_ccs.h |  2 ++
 6 files changed, 67 insertions(+), 39 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 4b71570529a6..544d7d8460d9 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -24,6 +24,7 @@
 #include "xe_pxp_debugfs.h"
 #include "xe_sriov.h"
 #include "xe_sriov_pf.h"
+#include "xe_sriov_vf.h"
 #include "xe_step.h"
 #include "xe_tile_debugfs.h"
 #include "xe_wa.h"
@@ -395,4 +396,6 @@ void xe_debugfs_register(struct xe_device *xe)
 
 	if (IS_SRIOV_PF(xe))
 		xe_sriov_pf_debugfs_register(xe, root);
+	else if (IS_SRIOV_VF(xe))
+		xe_sriov_vf_debugfs_register(xe, root);
 }
diff --git a/drivers/gpu/drm/xe/xe_gt_debugfs.c b/drivers/gpu/drm/xe/xe_gt_debugfs.c
index a9d960de0e5e..f253e2df4907 100644
--- a/drivers/gpu/drm/xe/xe_gt_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_gt_debugfs.c
@@ -124,33 +124,6 @@ static int powergate_info(struct xe_gt *gt, struct drm_printer *p)
 	return ret;
 }
 
-static int sa_info_vf_ccs(struct xe_gt *gt, struct drm_printer *p)
-{
-	struct xe_device *xe = gt_to_xe(gt);
-	struct xe_sa_manager *bb_pool;
-	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
-
-	if (!IS_VF_CCS_READY(gt_to_xe(gt)))
-		return 0;
-
-	xe_pm_runtime_get(gt_to_xe(gt));
-
-	for_each_ccs_rw_ctx(ctx_id) {
-		bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
-		if (!bb_pool)
-			break;
-
-		drm_printf(p, "ccs %s bb suballoc info\n", ctx_id ? "write" : "read");
-		drm_printf(p, "-------------------------\n");
-		drm_suballoc_dump_debug_info(&bb_pool->base, p, xe_sa_manager_gpu_addr(bb_pool));
-		drm_puts(p, "\n");
-	}
-
-	xe_pm_runtime_put(gt_to_xe(gt));
-
-	return 0;
-}
-
 static int topology(struct xe_gt *gt, struct drm_printer *p)
 {
 	xe_pm_runtime_get(gt_to_xe(gt));
@@ -318,13 +291,6 @@ static const struct drm_info_list vf_safe_debugfs_list[] = {
 	{"hwconfig", .show = xe_gt_debugfs_simple_show, .data = hwconfig},
 };
 
-/*
- * only for GT debugfs files which are valid on VF. Not valid on PF.
- */
-static const struct drm_info_list vf_only_debugfs_list[] = {
-	{"sa_info_vf_ccs", .show = xe_gt_debugfs_simple_show, .data = sa_info_vf_ccs},
-};
-
 /* everything else should be added here */
 static const struct drm_info_list pf_only_debugfs_list[] = {
 	{"hw_engines", .show = xe_gt_debugfs_simple_show, .data = hw_engines},
@@ -465,11 +431,6 @@ void xe_gt_debugfs_register(struct xe_gt *gt)
 		drm_debugfs_create_files(pf_only_debugfs_list,
 					 ARRAY_SIZE(pf_only_debugfs_list),
 					 root, minor);
-	else
-		drm_debugfs_create_files(vf_only_debugfs_list,
-					 ARRAY_SIZE(vf_only_debugfs_list),
-					 root, minor);
-
 
 	xe_uc_debugfs_register(&gt->uc, root);
 
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.c b/drivers/gpu/drm/xe/xe_sriov_vf.c
index cb062328551a..cdd9f8e78b2a 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.c
@@ -3,6 +3,7 @@
  * Copyright © 2023-2024 Intel Corporation
  */
 
+#include <drm/drm_debugfs.h>
 #include <drm/drm_managed.h>
 
 #include "xe_assert.h"
@@ -445,3 +446,30 @@ int xe_sriov_vf_init_late(struct xe_device *xe)
 
 	return err;
 }
+
+static int sa_info_vf_ccs(struct seq_file *m, void *data)
+{
+	struct drm_info_node *node = m->private;
+	struct xe_device *xe = to_xe_device(node->minor->dev);
+	struct drm_printer p = drm_seq_file_printer(m);
+
+	xe_sriov_vf_ccs_print(xe, &p);
+	return 0;
+}
+
+static const struct drm_info_list debugfs_list[] = {
+	{ .name = "sa_info_vf_ccs", .show = sa_info_vf_ccs },
+};
+
+/**
+ * xe_sriov_vf_debugfs_register - Register VF debugfs attributes.
+ * @xe: the &xe_device
+ * @root: the root &dentry
+ *
+ * Prepare debugfs attributes exposed by the VF.
+ */
+void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root)
+{
+	drm_debugfs_create_files(debugfs_list, ARRAY_SIZE(debugfs_list),
+				 root, xe->drm.primary);
+}
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf.h b/drivers/gpu/drm/xe/xe_sriov_vf.h
index 3bf3364799ad..9e752105ec2a 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf.h
@@ -8,11 +8,13 @@
 
 #include <linux/types.h>
 
+struct dentry;
 struct xe_device;
 
 void xe_sriov_vf_init_early(struct xe_device *xe);
 int xe_sriov_vf_init_late(struct xe_device *xe);
 void xe_sriov_vf_start_migration_recovery(struct xe_device *xe);
 bool xe_sriov_vf_migration_supported(struct xe_device *xe);
+void xe_sriov_vf_debugfs_register(struct xe_device *xe, struct dentry *root);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
index c5c60f05073d..8dec616c37c9 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.c
@@ -13,6 +13,7 @@
 #include "xe_guc_submit.h"
 #include "xe_lrc.h"
 #include "xe_migrate.h"
+#include "xe_pm.h"
 #include "xe_sa.h"
 #include "xe_sriov_printk.h"
 #include "xe_sriov_vf.h"
@@ -376,3 +377,34 @@ int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo)
 	}
 	return 0;
 }
+
+/**
+ * xe_sriov_vf_ccs_print - Print VF CCS details.
+ * @xe: the &xe_device
+ * @p: the &drm_printer
+ *
+ * This function is for VF use only.
+ */
+void xe_sriov_vf_ccs_print(struct xe_device *xe, struct drm_printer *p)
+{
+	struct xe_sa_manager *bb_pool;
+	enum xe_sriov_vf_ccs_rw_ctxs ctx_id;
+
+	if (!IS_VF_CCS_READY(xe))
+		return;
+
+	xe_pm_runtime_get(xe);
+
+	for_each_ccs_rw_ctx(ctx_id) {
+		bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
+		if (!bb_pool)
+			break;
+
+		drm_printf(p, "ccs %s bb suballoc info\n", ctx_id ? "write" : "read");
+		drm_printf(p, "-------------------------\n");
+		drm_suballoc_dump_debug_info(&bb_pool->base, p, xe_sa_manager_gpu_addr(bb_pool));
+		drm_puts(p, "\n");
+	}
+
+	xe_pm_runtime_put(xe);
+}
diff --git a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
index f0e1189b417a..0745c0ff0228 100644
--- a/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
+++ b/drivers/gpu/drm/xe/xe_sriov_vf_ccs.h
@@ -10,6 +10,7 @@
 #include "xe_sriov.h"
 #include "xe_sriov_vf_ccs_types.h"
 
+struct drm_printer;
 struct xe_device;
 struct xe_bo;
 
@@ -17,6 +18,7 @@ int xe_sriov_vf_ccs_init(struct xe_device *xe);
 int xe_sriov_vf_ccs_attach_bo(struct xe_bo *bo);
 int xe_sriov_vf_ccs_detach_bo(struct xe_bo *bo);
 int xe_sriov_vf_ccs_register_context(struct xe_device *xe);
+void xe_sriov_vf_ccs_print(struct xe_device *xe, struct drm_printer *p);
 
 static inline bool xe_sriov_vf_ccs_ready(struct xe_device *xe)
 {

From fce99326c9cf5a0e57c4283a61c6b622ef5b0de8 Mon Sep 17 00:00:00 2001
From: Raag Jadav <raag.jadav@intel.com>
Date: Mon, 8 Sep 2025 11:23:20 +0530
Subject: [PATCH 1128/1455] drm/xe/i2c: Enable bus mastering

Enable bus mastering for I2C controller to support device initiated
in-band transactions.

Signed-off-by: Raag Jadav <raag.jadav@intel.com>
Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Link: https://lore.kernel.org/r/20250908055320.2549722-1-raag.jadav@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_i2c.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_i2c.c b/drivers/gpu/drm/xe/xe_i2c.c
index 044dda517b7c..48dfcb41fa08 100644
--- a/drivers/gpu/drm/xe/xe_i2c.c
+++ b/drivers/gpu/drm/xe/xe_i2c.c
@@ -259,7 +259,7 @@ void xe_i2c_pm_resume(struct xe_device *xe, bool d3cold)
 		return;
 
 	if (d3cold)
-		xe_mmio_rmw32(mmio, I2C_CONFIG_CMD, 0, PCI_COMMAND_MEMORY);
+		xe_mmio_rmw32(mmio, I2C_CONFIG_CMD, 0, PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER);
 
 	xe_mmio_rmw32(mmio, I2C_CONFIG_PMCSR, PCI_PM_CTRL_STATE_MASK, (__force u32)PCI_D0);
 	drm_dbg(&xe->drm, "pmcsr: 0x%08x\n", xe_mmio_read32(mmio, I2C_CONFIG_PMCSR));

From e57ae80fe01a2b039c689b2d14fef51b1ebbd7f8 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 5 Sep 2025 20:02:25 +0200
Subject: [PATCH 1129/1455] drm/xe/debugfs: Make residencies definitions const

No need to keep them non-const. Also fix declaration of .name
member, as it points to the const string. This translates to:

  add/remove: 1/0 grow/shrink: 0/2 up/down: 80/-248 (-168)
  Function                                     old     new   delta
  residencies                                    -      80     +80
  dgfx_pcie_link_residencies_show              365     263    -102
  dgfx_pkg_residencies_show                    454     308    -146
  Total: Before=2821548, After=2821380, chg -0.01%

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Soham Purkait <soham.purkait@intel.com>
Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
Cc: Karthik Poosa <karthik.poosa@intel.com>
Cc: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250905180225.8434-1-michal.wajdeczko@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_debugfs.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 544d7d8460d9..38169238dbed 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -40,7 +40,7 @@ DECLARE_FAULT_ATTR(gt_reset_failure);
 DECLARE_FAULT_ATTR(inject_csc_hw_error);
 
 static void read_residency_counter(struct xe_device *xe, struct xe_mmio *mmio,
-				   u32 offset, char *name, struct drm_printer *p)
+				   u32 offset, const char *name, struct drm_printer *p)
 {
 	u64 residency = 0;
 	int ret;
@@ -136,9 +136,9 @@ static int dgfx_pkg_residencies_show(struct seq_file *m, void *data)
 	p = drm_seq_file_printer(m);
 	xe_pm_runtime_get(xe);
 	mmio = xe_root_tile_mmio(xe);
-	struct {
+	static const struct {
 		u32 offset;
-		char *name;
+		const char *name;
 	} residencies[] = {
 		{BMG_G2_RESIDENCY_OFFSET, "Package G2"},
 		{BMG_G6_RESIDENCY_OFFSET, "Package G6"},
@@ -165,9 +165,9 @@ static int dgfx_pcie_link_residencies_show(struct seq_file *m, void *data)
 	xe_pm_runtime_get(xe);
 	mmio = xe_root_tile_mmio(xe);
 
-	struct {
+	static const struct {
 		u32 offset;
-		char *name;
+		const char *name;
 	} residencies[] = {
 		{BMG_PCIE_LINK_L0_RESIDENCY_OFFSET, "PCIE LINK L0 RESIDENCY"},
 		{BMG_PCIE_LINK_L1_RESIDENCY_OFFSET, "PCIE LINK L1 RESIDENCY"},

From 7b77941724628e6171a286edbf04d67a1f1c1459 Mon Sep 17 00:00:00 2001
From: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Date: Sun, 7 Sep 2025 15:42:17 +0200
Subject: [PATCH 1130/1455] drm/xe/hwmon: Use devm_mutex_init()

Use devm_mutex_init() instead of hand-writing it.

This saves some LoC, improves readability and saves some space in the
generated .o file.

Before:
======
   text	   data	    bss	    dec	    hex	filename
  36884	  10296	     64	  47244	   b88c	drivers/gpu/drm/xe/xe_hwmon.o

After:
=====
   text	   data	    bss	    dec	    hex	filename
  36651	  10224	     64	  46939	   b75b	drivers/gpu/drm/xe/xe_hwmon.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/989e96369e9e1f8a44b816962917ec76877c912d.1757252520.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_hwmon.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c
index 32a76ae6e9dc..e65382d4426a 100644
--- a/drivers/gpu/drm/xe/xe_hwmon.c
+++ b/drivers/gpu/drm/xe/xe_hwmon.c
@@ -1294,13 +1294,6 @@ xe_hwmon_get_preregistration_info(struct xe_hwmon *hwmon)
 			xe_hwmon_fan_input_read(hwmon, channel, &fan_speed);
 }
 
-static void xe_hwmon_mutex_destroy(void *arg)
-{
-	struct xe_hwmon *hwmon = arg;
-
-	mutex_destroy(&hwmon->hwmon_lock);
-}
-
 int xe_hwmon_register(struct xe_device *xe)
 {
 	struct device *dev = xe->drm.dev;
@@ -1319,8 +1312,7 @@ int xe_hwmon_register(struct xe_device *xe)
 	if (!hwmon)
 		return -ENOMEM;
 
-	mutex_init(&hwmon->hwmon_lock);
-	ret = devm_add_action_or_reset(dev, xe_hwmon_mutex_destroy, hwmon);
+	ret = devm_mutex_init(dev, &hwmon->hwmon_lock);
 	if (ret)
 		return ret;
 

From cb55f39bf7b1e687b88dc9825c21a7d3454b5acb Mon Sep 17 00:00:00 2001
From: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Date: Mon, 8 Sep 2025 11:49:51 +0200
Subject: [PATCH 1131/1455] drm/msm/dsi/phy: Fix reading zero as PLL rates when
 unprepared

Hardware Programming Guide for DSI PHY says that PLL_SHUTDOWNB and
DIGTOP_PWRDN_B have to be asserted for any PLL register access.
Whenever dsi_pll_7nm_vco_recalc_rate() or dsi_pll_7nm_vco_set_rate()
were called on unprepared PLL, driver read values of zero leading to all
sort of further troubles, like failing to set pixel and byte clock
rates.

Asserting the PLL shutdown bit is done by dsi_pll_enable_pll_bias() (and
corresponding dsi_pll_disable_pll_bias()) which are called through the
code, including from PLL .prepare() and .unprepare() callbacks.

The .set_rate() and .recalc_rate() can be called almost anytime from
external users including times when PLL is or is not prepared, thus
driver should not interfere with the prepare status.

Implement simple reference counting for the PLL bias, so
set_rate/recalc_rate will not change the status of prepared PLL.

Issue of reading 0 in .recalc_rate() did not show up on existing
devices, but only after re-ordering the code for SM8750.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673416/
Link: https://lore.kernel.org/r/20250908094950.72877-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/dsi/phy/dsi_phy.h     |  1 +
 drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c | 53 +++++++++++++++++++++++
 2 files changed, 54 insertions(+)

diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
index 3cbf08231492..e391505fdaf0 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.h
@@ -109,6 +109,7 @@ struct msm_dsi_phy {
 	struct msm_dsi_dphy_timing timing;
 	const struct msm_dsi_phy_cfg *cfg;
 	void *tuning_cfg;
+	void *pll_data;
 
 	enum msm_dsi_phy_usecase usecase;
 	bool regulator_ldo_mode;
diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
index 2c2bbda46c78..32f06edd21a9 100644
--- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
+++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_7nm.c
@@ -90,6 +90,13 @@ struct dsi_pll_7nm {
 	/* protects REG_DSI_7nm_PHY_CMN_CLK_CFG1 register */
 	spinlock_t pclk_mux_lock;
 
+	/*
+	 * protects REG_DSI_7nm_PHY_CMN_CTRL_0 register and pll_enable_cnt
+	 * member
+	 */
+	spinlock_t pll_enable_lock;
+	int pll_enable_cnt;
+
 	struct pll_7nm_cached_state cached_state;
 
 	struct dsi_pll_7nm *slave;
@@ -103,6 +110,9 @@ struct dsi_pll_7nm {
  */
 static struct dsi_pll_7nm *pll_7nm_list[DSI_MAX];
 
+static void dsi_pll_enable_pll_bias(struct dsi_pll_7nm *pll);
+static void dsi_pll_disable_pll_bias(struct dsi_pll_7nm *pll);
+
 static void dsi_pll_setup_config(struct dsi_pll_config *config)
 {
 	config->ssc_freq = 31500;
@@ -340,6 +350,7 @@ static int dsi_pll_7nm_vco_set_rate(struct clk_hw *hw, unsigned long rate,
 	struct dsi_pll_7nm *pll_7nm = to_pll_7nm(hw);
 	struct dsi_pll_config config;
 
+	dsi_pll_enable_pll_bias(pll_7nm);
 	DBG("DSI PLL%d rate=%lu, parent's=%lu", pll_7nm->phy->id, rate,
 	    parent_rate);
 
@@ -357,6 +368,7 @@ static int dsi_pll_7nm_vco_set_rate(struct clk_hw *hw, unsigned long rate,
 
 	dsi_pll_ssc_commit(pll_7nm, &config);
 
+	dsi_pll_disable_pll_bias(pll_7nm);
 	/* flush, ensure all register writes are done*/
 	wmb();
 
@@ -385,24 +397,47 @@ static int dsi_pll_7nm_lock_status(struct dsi_pll_7nm *pll)
 
 static void dsi_pll_disable_pll_bias(struct dsi_pll_7nm *pll)
 {
+	unsigned long flags;
 	u32 data;
 
+	spin_lock_irqsave(&pll->pll_enable_lock, flags);
+	--pll->pll_enable_cnt;
+	if (pll->pll_enable_cnt < 0) {
+		spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
+		DRM_DEV_ERROR_RATELIMITED(&pll->phy->pdev->dev,
+					  "bug: imbalance in disabling PLL bias\n");
+		return;
+	} else if (pll->pll_enable_cnt > 0) {
+		spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
+		return;
+	} /* else: == 0 */
+
 	data = readl(pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 	data &= ~DSI_7nm_PHY_CMN_CTRL_0_PLL_SHUTDOWNB;
 	writel(0, pll->phy->pll_base + REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES);
 	writel(data, pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
 	ndelay(250);
 }
 
 static void dsi_pll_enable_pll_bias(struct dsi_pll_7nm *pll)
 {
+	unsigned long flags;
 	u32 data;
 
+	spin_lock_irqsave(&pll->pll_enable_lock, flags);
+	if (pll->pll_enable_cnt++) {
+		spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
+		WARN_ON(pll->pll_enable_cnt == INT_MAX);
+		return;
+	}
+
 	data = readl(pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 	data |= DSI_7nm_PHY_CMN_CTRL_0_PLL_SHUTDOWNB;
 	writel(data, pll->phy->base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 
 	writel(0xc0, pll->phy->pll_base + REG_DSI_7nm_PHY_PLL_SYSTEM_MUXES);
+	spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
 	ndelay(250);
 }
 
@@ -543,6 +578,7 @@ static unsigned long dsi_pll_7nm_vco_recalc_rate(struct clk_hw *hw,
 	u32 dec;
 	u64 pll_freq, tmp64;
 
+	dsi_pll_enable_pll_bias(pll_7nm);
 	dec = readl(base + REG_DSI_7nm_PHY_PLL_DECIMAL_DIV_START_1);
 	dec &= 0xff;
 
@@ -567,6 +603,8 @@ static unsigned long dsi_pll_7nm_vco_recalc_rate(struct clk_hw *hw,
 	DBG("DSI PLL%d returning vco rate = %lu, dec = %x, frac = %x",
 	    pll_7nm->phy->id, (unsigned long)vco_rate, dec, frac);
 
+	dsi_pll_disable_pll_bias(pll_7nm);
+
 	return (unsigned long)vco_rate;
 }
 
@@ -600,6 +638,7 @@ static void dsi_7nm_pll_save_state(struct msm_dsi_phy *phy)
 	void __iomem *phy_base = pll_7nm->phy->base;
 	u32 cmn_clk_cfg0, cmn_clk_cfg1;
 
+	dsi_pll_enable_pll_bias(pll_7nm);
 	cached->pll_out_div = readl(pll_7nm->phy->pll_base +
 			REG_DSI_7nm_PHY_PLL_PLL_OUTDIV_RATE);
 	cached->pll_out_div &= 0x3;
@@ -611,6 +650,7 @@ static void dsi_7nm_pll_save_state(struct msm_dsi_phy *phy)
 	cmn_clk_cfg1 = readl(phy_base + REG_DSI_7nm_PHY_CMN_CLK_CFG1);
 	cached->pll_mux = FIELD_GET(DSI_7nm_PHY_CMN_CLK_CFG1_DSICLK_SEL__MASK, cmn_clk_cfg1);
 
+	dsi_pll_disable_pll_bias(pll_7nm);
 	DBG("DSI PLL%d outdiv %x bit_clk_div %x pix_clk_div %x pll_mux %x",
 	    pll_7nm->phy->id, cached->pll_out_div, cached->bit_clk_div,
 	    cached->pix_clk_div, cached->pll_mux);
@@ -833,8 +873,10 @@ static int dsi_pll_7nm_init(struct msm_dsi_phy *phy)
 
 	spin_lock_init(&pll_7nm->postdiv_lock);
 	spin_lock_init(&pll_7nm->pclk_mux_lock);
+	spin_lock_init(&pll_7nm->pll_enable_lock);
 
 	pll_7nm->phy = phy;
+	phy->pll_data = pll_7nm;
 
 	ret = pll_7nm_register(pll_7nm, phy->provided_clocks->hws);
 	if (ret) {
@@ -923,8 +965,10 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy,
 	u32 const delay_us = 5;
 	u32 const timeout_us = 1000;
 	struct msm_dsi_dphy_timing *timing = &phy->timing;
+	struct dsi_pll_7nm *pll = phy->pll_data;
 	void __iomem *base = phy->base;
 	bool less_than_1500_mhz;
+	unsigned long flags;
 	u32 vreg_ctrl_0, vreg_ctrl_1, lane_ctrl0;
 	u32 glbl_pemph_ctrl_0;
 	u32 glbl_str_swi_cal_sel_ctrl, glbl_hstx_str_ctrl_0;
@@ -1046,10 +1090,13 @@ static int dsi_7nm_phy_enable(struct msm_dsi_phy *phy,
 		glbl_rescode_bot_ctrl = 0x3c;
 	}
 
+	spin_lock_irqsave(&pll->pll_enable_lock, flags);
+	pll->pll_enable_cnt = 1;
 	/* de-assert digital and pll power down */
 	data = DSI_7nm_PHY_CMN_CTRL_0_DIGTOP_PWRDN_B |
 	       DSI_7nm_PHY_CMN_CTRL_0_PLL_SHUTDOWNB;
 	writel(data, base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
 
 	/* Assert PLL core reset */
 	writel(0x00, base + REG_DSI_7nm_PHY_CMN_PLL_CNTRL);
@@ -1162,7 +1209,9 @@ static bool dsi_7nm_set_continuous_clock(struct msm_dsi_phy *phy, bool enable)
 
 static void dsi_7nm_phy_disable(struct msm_dsi_phy *phy)
 {
+	struct dsi_pll_7nm *pll = phy->pll_data;
 	void __iomem *base = phy->base;
+	unsigned long flags;
 	u32 data;
 
 	DBG("");
@@ -1189,8 +1238,12 @@ static void dsi_7nm_phy_disable(struct msm_dsi_phy *phy)
 	writel(data, base + REG_DSI_7nm_PHY_CMN_CTRL_0);
 	writel(0, base + REG_DSI_7nm_PHY_CMN_LANE_CTRL0);
 
+	spin_lock_irqsave(&pll->pll_enable_lock, flags);
+	pll->pll_enable_cnt = 0;
 	/* Turn off all PHY blocks */
 	writel(0x00, base + REG_DSI_7nm_PHY_CMN_CTRL_0);
+	spin_unlock_irqrestore(&pll->pll_enable_lock, flags);
+
 	/* make sure phy is turned off */
 	wmb();
 

From f028bcafb6dfb4c2bb656cbff9e6a66222d3d3d7 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Tue, 2 Sep 2025 17:20:00 +0530
Subject: [PATCH 1132/1455] drm/msm: Fix bootup splat with separate_gpu_drm
 modparam

The drm_gem_for_each_gpuvm_bo() call from lookup_vma() accesses
drm_gem_obj.gpuva.list, which is not initialized when the drm driver
does not support DRIVER_GEM_GPUVA feature. Enable it for msm_kms
drm driver to fix the splat seen when msm.separate_gpu_drm=1 modparam
is set:

[    9.506020] Unable to handle kernel paging request at virtual address fffffffffffffff0
[    9.523160] Mem abort info:
[    9.523161]   ESR = 0x0000000096000006
[    9.523163]   EC = 0x25: DABT (current EL), IL = 32 bits
[    9.523165]   SET = 0, FnV = 0
[    9.523166]   EA = 0, S1PTW = 0
[    9.523167]   FSC = 0x06: level 2 translation fault
[    9.523169] Data abort info:
[    9.523170]   ISV = 0, ISS = 0x00000006, ISS2 = 0x00000000
[    9.523171]   CM = 0, WnR = 0, TnD = 0, TagAccess = 0
[    9.523172]   GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
[    9.523174] swapper pgtable: 4k pages, 48-bit VAs, pgdp=0000000ad370f000
[    9.523176] [fffffffffffffff0] pgd=0000000000000000, p4d=0000000ad4787403, pud=0000000ad4788403, pmd=0000000000000000
[    9.523184] Internal error: Oops: 0000000096000006 [#1]  SMP
[    9.592968] CPU: 9 UID: 0 PID: 448 Comm: (udev-worker) Not tainted 6.17.0-rc4-assorted-fix-00005-g0e9bb53a2282-dirty #3 PREEMPT
[    9.592970] Hardware name: Qualcomm CRD, BIOS 6.0.240718.BOOT.MXF.2.4-00515-HAMOA-1 07/18/2024
[    9.592971] pstate: a1400005 (NzCv daif +PAN -UAO -TCO +DIT -SSBS BTYPE=--)
[    9.592973] pc : lookup_vma+0x28/0xe0 [msm]
[    9.592996] lr : get_vma_locked+0x2c/0x128 [msm]
[    9.763632] sp : ffff800082dab460
[    9.763666] Call trace:
[    9.763668]  lookup_vma+0x28/0xe0 [msm] (P)
[    9.763688]  get_vma_locked+0x2c/0x128 [msm]
[    9.763706]  msm_gem_get_and_pin_iova_range+0x68/0x11c [msm]
[    9.763723]  msm_gem_get_and_pin_iova+0x18/0x24 [msm]
[    9.763740]  msm_fbdev_driver_fbdev_probe+0xd0/0x258 [msm]
[    9.763760]  __drm_fb_helper_initial_config_and_unlock+0x288/0x528 [drm_kms_helper]
[    9.763771]  drm_fb_helper_initial_config+0x44/0x54 [drm_kms_helper]
[    9.763779]  drm_fbdev_client_hotplug+0x84/0xd4 [drm_client_lib]
[    9.763782]  drm_client_register+0x58/0x9c [drm]
[    9.763806]  drm_fbdev_client_setup+0xe8/0xcf0 [drm_client_lib]
[    9.763809]  drm_client_setup+0xb4/0xd8 [drm_client_lib]
[    9.763811]  msm_drm_kms_post_init+0x2c/0x3c [msm]
[    9.763830]  msm_drm_init+0x1a8/0x22c [msm]
[    9.763848]  msm_drm_bind+0x30/0x3c [msm]
[    9.919273]  try_to_bring_up_aggregate_device+0x168/0x1d4
[    9.919283]  __component_add+0xa4/0x170
[    9.919286]  component_add+0x14/0x20
[    9.919288]  msm_dp_display_probe_tail+0x4c/0xac [msm]
[    9.919315]  msm_dp_auxbus_done_probe+0x14/0x20 [msm]
[    9.919335]  dp_aux_ep_probe+0x4c/0xf0 [drm_dp_aux_bus]
[    9.919341]  really_probe+0xbc/0x298
[    9.919345]  __driver_probe_device+0x78/0x12c
[    9.919348]  driver_probe_device+0x40/0x160
[    9.919350]  __driver_attach+0x94/0x19c
[    9.919353]  bus_for_each_dev+0x74/0xd4
[    9.919355]  driver_attach+0x24/0x30
[    9.919358]  bus_add_driver+0xe4/0x208
[    9.919360]  driver_register+0x60/0x128
[    9.919363]  __dp_aux_dp_driver_register+0x24/0x30 [drm_dp_aux_bus]
[    9.919365]  atana33xc20_init+0x20/0x1000 [panel_samsung_atna33xc20]
[    9.919370]  do_one_initcall+0x6c/0x1b0
[    9.919374]  do_init_module+0x58/0x234
[    9.919377]  load_module+0x19cc/0x1bd4
[    9.919380]  init_module_from_file+0x84/0xc4
[    9.919382]  __arm64_sys_finit_module+0x1b8/0x2cc
[    9.919384]  invoke_syscall+0x48/0x110
[    9.919389]  el0_svc_common.constprop.0+0xc8/0xe8
[    9.919393]  do_el0_svc+0x20/0x2c
[    9.919396]  el0_svc+0x34/0xf0
[    9.919401]  el0t_64_sync_handler+0xa0/0xe4
[    9.919403]  el0t_64_sync+0x198/0x19c
[    9.919407] Code: eb0000bf 54000480 d100a003 aa0303e2 (f8418c44)
[    9.919410] ---[ end trace 0000000000000000 ]---

Fixes: 217ed15bd399 ("drm/msm: enable separate binding of GPU and display devices")
Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672257/
Link: https://lore.kernel.org/r/20250902-assorted-sept-1-v1-1-f3ec9baed513@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/msm_drv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 9dcc7a596a11..7e977fec4100 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -826,6 +826,7 @@ static const struct file_operations fops = {
 
 #define DRIVER_FEATURES_KMS ( \
 		DRIVER_GEM | \
+		DRIVER_GEM_GPUVA | \
 		DRIVER_ATOMIC | \
 		DRIVER_MODESET | \
 		0 )

From 3f17991488af62265c8335b42585c36cdee5ad05 Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Tue, 2 Sep 2025 17:20:02 +0530
Subject: [PATCH 1133/1455] drm/msm/adreno: Add a modparam to skip GPU

During bringup of a new GPU support, it is convenient to have knob to
quickly disable GPU, but keep the display support. This helps to
fallback to 'kms_swrast' in case of bootup issues due to GPU. Add a
modparam to support this.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672262/
Link: https://lore.kernel.org/r/20250902-assorted-sept-1-v1-3-f3ec9baed513@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/adreno_device.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 50945bfe9b49..28f744f3caf7 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -24,6 +24,10 @@ bool disable_acd;
 MODULE_PARM_DESC(disable_acd, "Forcefully disable GPU ACD");
 module_param_unsafe(disable_acd, bool, 0400);
 
+static bool skip_gpu;
+MODULE_PARM_DESC(no_gpu, "Disable GPU driver register (0=enable GPU driver register (default), 1=skip GPU driver register");
+module_param(skip_gpu, bool, 0400);
+
 extern const struct adreno_gpulist a2xx_gpulist;
 extern const struct adreno_gpulist a3xx_gpulist;
 extern const struct adreno_gpulist a4xx_gpulist;
@@ -184,6 +188,9 @@ bool adreno_has_gpu(struct device_node *node)
 	uint32_t chip_id;
 	int ret;
 
+	if (skip_gpu)
+		return false;
+
 	ret = find_chipid(node, &chip_id);
 	if (ret)
 		return false;
@@ -404,10 +411,16 @@ static struct platform_driver adreno_driver = {
 
 void __init adreno_register(void)
 {
+	if (skip_gpu)
+		return;
+
 	platform_driver_register(&adreno_driver);
 }
 
 void __exit adreno_unregister(void)
 {
+	if (skip_gpu)
+		return;
+
 	platform_driver_unregister(&adreno_driver);
 }

From 721b7f5f75ff255c231e2869428c4673a124115d Mon Sep 17 00:00:00 2001
From: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Date: Tue, 2 Sep 2025 17:20:03 +0530
Subject: [PATCH 1134/1455] drm/msm/a6xx: Add a comment to acd_probe()

It is not obvious why we can skip error checking of
dev_pm_opp_find_freq_exact() API. Add a comment explaining it.

Signed-off-by: Akhil P Oommen <akhilpo@oss.qualcomm.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/672263/
Link: https://lore.kernel.org/r/20250902-assorted-sept-1-v1-4-f3ec9baed513@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/adreno/a6xx_gmu.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
index a719a8d44c1b..fc62fef2fed8 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c
@@ -1731,6 +1731,7 @@ static int a6xx_gmu_acd_probe(struct a6xx_gmu *gmu)
 		u32 val;
 
 		freq = gmu->gpu_freqs[i];
+		/* This is unlikely to fail because we are passing back a known freq */
 		opp = dev_pm_opp_find_freq_exact(&gpu->pdev->dev, freq, true);
 		np = dev_pm_opp_get_of_node(opp);
 

From 88ec0e01a880e3326794e149efae39e3aa4dbbec Mon Sep 17 00:00:00 2001
From: Qianfeng Rong <rongqianfeng@vivo.com>
Date: Tue, 26 Aug 2025 17:20:45 +0800
Subject: [PATCH 1135/1455] drm/msm/dpu: fix incorrect type for ret

Change 'ret' from unsigned long to int, as storing negative error codes
in an unsigned long makes it never equal to -ETIMEDOUT, causing logical
errors.

Fixes: d7d0e73f7de3 ("drm/msm/dpu: introduce the dpu_encoder_phys_* for writeback")
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/671100/
Link: https://lore.kernel.org/r/20250826092047.224341-1-rongqianfeng@vivo.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
index 56a5b596554d..46f348972a97 100644
--- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
+++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_wb.c
@@ -446,7 +446,7 @@ static void _dpu_encoder_phys_wb_handle_wbdone_timeout(
 static int dpu_encoder_phys_wb_wait_for_commit_done(
 		struct dpu_encoder_phys *phys_enc)
 {
-	unsigned long ret;
+	int ret;
 	struct dpu_encoder_wait_info wait_info;
 	struct dpu_encoder_phys_wb *wb_enc = to_dpu_encoder_phys_wb(phys_enc);
 

From a643abb673790c09a4c188a5ee9b064844f5047b Mon Sep 17 00:00:00 2001
From: Liao Yuanhong <liaoyuanhong@vivo.com>
Date: Wed, 20 Aug 2025 21:12:56 +0800
Subject: [PATCH 1136/1455] drm/msm/mdp4: remove the use of dev_err_probe()

Logging messages that show some type of "out of memory" error are generally
unnecessary as there is a generic message and a stack dump done by the
memory subsystem. These messages generally increase kernel size without
much added value[1].

The dev_err_probe() doesn't do anything when error is '-ENOMEM'. Therefore,
remove the useless call to dev_err_probe(), and just return the value
instead.

[1]: https://lore.kernel.org/lkml/1402419340.30479.18.camel@joe-AO725/

Signed-off-by: Liao Yuanhong <liaoyuanhong@vivo.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/670017/
Link: https://lore.kernel.org/r/20250820131300.499727-1-liaoyuanhong@vivo.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
index 1d2a1e4f6ce7..809ca191e9de 100644
--- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c
@@ -512,7 +512,7 @@ static int mdp4_probe(struct platform_device *pdev)
 
 	mdp4_kms = devm_kzalloc(dev, sizeof(*mdp4_kms), GFP_KERNEL);
 	if (!mdp4_kms)
-		return dev_err_probe(dev, -ENOMEM, "failed to allocate kms\n");
+		return -ENOMEM;
 
 	mdp4_kms->mmio = msm_ioremap(pdev, NULL);
 	if (IS_ERR(mdp4_kms->mmio))

From 86a9fe82e9b1f43e6d2bc867bf96bb40660d8719 Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:09 -0400
Subject: [PATCH 1137/1455] drm/i915: Move struct_mutex to drm_i915_private

Move legacy BKL struct_mutex from drm_device to drm_i915_private, which
is the last remaining user.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-2-luiz.mello@estudante.ufscar.br
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/drm_drv.c                  |  2 --
 drivers/gpu/drm/i915/gt/uc/intel_guc_log.c |  4 ++--
 drivers/gpu/drm/i915/i915_driver.c         |  2 ++
 drivers/gpu/drm/i915/i915_drv.h            | 11 +++++++++++
 drivers/gpu/drm/i915/i915_irq.c            |  4 ++--
 include/drm/drm_device.h                   | 10 ----------
 6 files changed, 17 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 02556363e918..287cf4a09101 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -697,7 +697,6 @@ static void drm_dev_init_release(struct drm_device *dev, void *res)
 	mutex_destroy(&dev->master_mutex);
 	mutex_destroy(&dev->clientlist_mutex);
 	mutex_destroy(&dev->filelist_mutex);
-	mutex_destroy(&dev->struct_mutex);
 }
 
 static int drm_dev_init(struct drm_device *dev,
@@ -738,7 +737,6 @@ static int drm_dev_init(struct drm_device *dev,
 	INIT_LIST_HEAD(&dev->vblank_event_list);
 
 	spin_lock_init(&dev->event_lock);
-	mutex_init(&dev->struct_mutex);
 	mutex_init(&dev->filelist_mutex);
 	mutex_init(&dev->clientlist_mutex);
 	mutex_init(&dev->master_mutex);
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
index e8a04e476c57..7135fdb0ebb4 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
@@ -678,7 +678,7 @@ int intel_guc_log_set_level(struct intel_guc_log *log, u32 level)
 	if (level < GUC_LOG_LEVEL_DISABLED || level > GUC_LOG_LEVEL_MAX)
 		return -EINVAL;
 
-	mutex_lock(&i915->drm.struct_mutex);
+	mutex_lock(&i915->struct_mutex);
 
 	if (log->level == level)
 		goto out_unlock;
@@ -696,7 +696,7 @@ int intel_guc_log_set_level(struct intel_guc_log *log, u32 level)
 	log->level = level;
 
 out_unlock:
-	mutex_unlock(&i915->drm.struct_mutex);
+	mutex_unlock(&i915->struct_mutex);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index a28c3710c4d5..2677c39a3246 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -235,6 +235,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 
 	intel_sbi_init(display);
 	vlv_iosf_sb_init(dev_priv);
+	mutex_init(&dev_priv->struct_mutex);
 	mutex_init(&dev_priv->sb_lock);
 
 	i915_memcpy_init_early(dev_priv);
@@ -293,6 +294,7 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv)
 	i915_workqueues_cleanup(dev_priv);
 
 	mutex_destroy(&dev_priv->sb_lock);
+	mutex_destroy(&dev_priv->struct_mutex);
 	vlv_iosf_sb_fini(dev_priv);
 	intel_sbi_fini(display);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index f76d138df0af..55f13cb2c395 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -224,6 +224,17 @@ struct drm_i915_private {
 
 	bool irqs_enabled;
 
+	/*
+	 * Currently, struct_mutex is only used by the i915 driver as a replacement
+	 * for BKL.
+	 *
+	 * For this reason, it is no longer part of struct drm_device.
+	 */
+	struct mutex struct_mutex;
+
+	/* LPT/WPT IOSF sideband protection */
+	struct mutex sbi_lock;
+
 	/* VLV/CHV IOSF sideband */
 	struct {
 		struct mutex lock; /* protect sideband access */
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index a5fa40ab5de2..5916a648efbb 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -167,7 +167,7 @@ static void ivb_parity_work(struct work_struct *work)
 	 * In order to prevent a get/put style interface, acquire struct mutex
 	 * any time we access those registers.
 	 */
-	mutex_lock(&dev_priv->drm.struct_mutex);
+	mutex_lock(&dev_priv->struct_mutex);
 
 	/* If we've screwed up tracking, just let the interrupt fire again */
 	if (drm_WARN_ON(&dev_priv->drm, !dev_priv->l3_parity.which_slice))
@@ -225,7 +225,7 @@ out:
 	gen5_gt_enable_irq(gt, GT_PARITY_ERROR(dev_priv));
 	spin_unlock_irq(gt->irq_lock);
 
-	mutex_unlock(&dev_priv->drm.struct_mutex);
+	mutex_unlock(&dev_priv->struct_mutex);
 }
 
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)
diff --git a/include/drm/drm_device.h b/include/drm/drm_device.h
index 08b3b2467c4c..cb16fd47e075 100644
--- a/include/drm/drm_device.h
+++ b/include/drm/drm_device.h
@@ -186,16 +186,6 @@ struct drm_device {
 	/** @unique: Unique name of the device */
 	char *unique;
 
-	/**
-	 * @struct_mutex:
-	 *
-	 * Lock for others (not &drm_minor.master and &drm_file.is_master)
-	 *
-	 * TODO: This lock used to be the BKL of the DRM subsystem. Move the
-	 *       lock into i915, which is the only remaining user.
-	 */
-	struct mutex struct_mutex;
-
 	/**
 	 * @master_mutex:
 	 *

From 793d31671bdf583c876ac9fa9948bb231c689036 Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:10 -0400
Subject: [PATCH 1138/1455] drm/i915: Remove struct_mutex in i915_irq.c

Remove struct_mutex from ivb_parity_work() function.

The ivb_parity_work runs in a workqueue so it cannot race with itself.

Also, it is not protecting anything with the other remaining usage of
struct_mutex.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-3-luiz.mello@estudante.ufscar.br
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_irq.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 5916a648efbb..8d5da222a187 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -163,11 +163,6 @@ static void ivb_parity_work(struct work_struct *work)
 	u32 misccpctl;
 	u8 slice = 0;
 
-	/* We must turn off DOP level clock gating to access the L3 registers.
-	 * In order to prevent a get/put style interface, acquire struct mutex
-	 * any time we access those registers.
-	 */
-	mutex_lock(&dev_priv->struct_mutex);
 
 	/* If we've screwed up tracking, just let the interrupt fire again */
 	if (drm_WARN_ON(&dev_priv->drm, !dev_priv->l3_parity.which_slice))
@@ -225,7 +220,6 @@ out:
 	gen5_gt_enable_irq(gt, GT_PARITY_ERROR(dev_priv));
 	spin_unlock_irq(gt->irq_lock);
 
-	mutex_unlock(&dev_priv->struct_mutex);
 }
 
 static irqreturn_t valleyview_irq_handler(int irq, void *arg)

From 723a170811db366b993437b08f1ca707163f7c84 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 19:21:28 +0300
Subject: [PATCH 1139/1455] drm/bridge: adv7511: use update latch for AVI
 infoframes

Instead of disabling and then reenabling AVI infoframe, use the
recommended way of updating it on the fly: latch current values using
the ADV7511_REG_INFOFRAME_UPDATE register.

Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250903-adv7511-audio-infoframe-v1-1-05b24459b9a4@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/adv7511/adv7511_drv.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index 26f8ef482423..26f8310382d8 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -916,22 +916,33 @@ static int adv7511_bridge_hdmi_write_infoframe(struct drm_bridge *bridge,
 {
 	struct adv7511 *adv7511 = bridge_to_adv7511(bridge);
 
-	adv7511_bridge_hdmi_clear_infoframe(bridge, type);
-
 	switch (type) {
 	case HDMI_INFOFRAME_TYPE_AVI:
+		/* send current AVI infoframe values while updating */
+		regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
+				   BIT(6), BIT(6));
+
 		/* The AVI infoframe id is not configurable */
 		regmap_bulk_write(adv7511->regmap, ADV7511_REG_AVI_INFOFRAME_VERSION,
 				  buffer + 1, len - 1);
 
+		regmap_write(adv7511->regmap, ADV7511_REG_AUDIO_INFOFRAME_LENGTH, 0x2);
+		regmap_write(adv7511->regmap, ADV7511_REG_AUDIO_INFOFRAME(1), 0x1);
+
+		/* use AVI infoframe updated info */
+		regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
+				   BIT(6), 0);
+
 		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_AVI_INFOFRAME);
 		break;
 	case HDMI_INFOFRAME_TYPE_SPD:
+		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_SPD);
 		regmap_bulk_write(adv7511->regmap_packet, ADV7511_PACKET_SPD(0),
 				  buffer, len);
 		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_SPD);
 		break;
 	case HDMI_INFOFRAME_TYPE_VENDOR:
+		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_SPARE1);
 		regmap_bulk_write(adv7511->regmap_packet, ADV7511_PACKET_SPARE1(0),
 				  buffer, len);
 		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_SPARE1);

From f0e7f358e72b10b01361787134ebcbd9e9aa72d9 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Date: Wed, 3 Sep 2025 19:21:29 +0300
Subject: [PATCH 1140/1455] drm/bridge: write full Audio InfoFrame

Instead of writing the first byte of the infoframe (and hoping that the
rest is default / zeroes), hook Audio InfoFrame support into the
write_infoframe / clear_infoframes callbacks and use
drm_atomic_helper_connector_hdmi_update_audio_infoframe() to write the
frame.

Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20250903-adv7511-audio-infoframe-v1-2-05b24459b9a4@oss.qualcomm.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 .../gpu/drm/bridge/adv7511/adv7511_audio.c    | 23 +++++--------------
 drivers/gpu/drm/bridge/adv7511/adv7511_drv.c  | 18 +++++++++++++++
 2 files changed, 24 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c
index 766b1c96bc88..87e7e820810a 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_audio.c
@@ -12,6 +12,8 @@
 #include <sound/soc.h>
 #include <linux/of_graph.h>
 
+#include <drm/display/drm_hdmi_state_helper.h>
+
 #include "adv7511.h"
 
 static void adv7511_calc_cts_n(unsigned int f_tmds, unsigned int fs,
@@ -155,17 +157,8 @@ int adv7511_hdmi_audio_prepare(struct drm_bridge *bridge,
 	regmap_update_bits(adv7511->regmap, ADV7511_REG_I2C_FREQ_ID_CFG,
 			   ADV7511_I2C_FREQ_ID_CFG_RATE_MASK, rate << 4);
 
-	/* send current Audio infoframe values while updating */
-	regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
-			   BIT(5), BIT(5));
-
-	regmap_write(adv7511->regmap, ADV7511_REG_AUDIO_INFOFRAME(0), 0x1);
-
-	/* use Audio infoframe updated info */
-	regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
-			   BIT(5), 0);
-
-	return 0;
+	return drm_atomic_helper_connector_hdmi_update_audio_infoframe(connector,
+								       &hparms->cea);
 }
 
 int adv7511_hdmi_audio_startup(struct drm_bridge *bridge,
@@ -188,15 +181,9 @@ int adv7511_hdmi_audio_startup(struct drm_bridge *bridge,
 	/* not copyrighted */
 	regmap_update_bits(adv7511->regmap, ADV7511_REG_AUDIO_CFG1,
 				BIT(5), BIT(5));
-	/* enable audio infoframes */
-	regmap_update_bits(adv7511->regmap, ADV7511_REG_PACKET_ENABLE1,
-				BIT(3), BIT(3));
 	/* AV mute disable */
 	regmap_update_bits(adv7511->regmap, ADV7511_REG_GC(0),
 				BIT(7) | BIT(6), BIT(7));
-	/* use Audio infoframe updated info */
-	regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
-				BIT(5), 0);
 
 	/* enable SPDIF receiver */
 	if (adv7511->audio_source == ADV7511_AUDIO_SOURCE_SPDIF)
@@ -214,4 +201,6 @@ void adv7511_hdmi_audio_shutdown(struct drm_bridge *bridge,
 	if (adv7511->audio_source == ADV7511_AUDIO_SOURCE_SPDIF)
 		regmap_update_bits(adv7511->regmap, ADV7511_REG_AUDIO_CONFIG,
 				   BIT(7), 0);
+
+	drm_atomic_helper_connector_hdmi_clear_audio_infoframe(connector);
 }
diff --git a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
index 26f8310382d8..b9be86541307 100644
--- a/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
+++ b/drivers/gpu/drm/bridge/adv7511/adv7511_drv.c
@@ -893,6 +893,9 @@ static int adv7511_bridge_hdmi_clear_infoframe(struct drm_bridge *bridge,
 	struct adv7511 *adv7511 = bridge_to_adv7511(bridge);
 
 	switch (type) {
+	case HDMI_INFOFRAME_TYPE_AUDIO:
+		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_AUDIO_INFOFRAME);
+		break;
 	case HDMI_INFOFRAME_TYPE_AVI:
 		adv7511_packet_disable(adv7511, ADV7511_PACKET_ENABLE_AVI_INFOFRAME);
 		break;
@@ -917,6 +920,21 @@ static int adv7511_bridge_hdmi_write_infoframe(struct drm_bridge *bridge,
 	struct adv7511 *adv7511 = bridge_to_adv7511(bridge);
 
 	switch (type) {
+	case HDMI_INFOFRAME_TYPE_AUDIO:
+		/* send current Audio infoframe values while updating */
+		regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
+				   BIT(5), BIT(5));
+
+		/* The Audio infoframe id is not configurable */
+		regmap_bulk_write(adv7511->regmap, ADV7511_REG_AUDIO_INFOFRAME_VERSION,
+				  buffer + 1, len - 1);
+
+		/* use Audio infoframe updated info */
+		regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,
+				   BIT(5), 0);
+
+		adv7511_packet_enable(adv7511, ADV7511_PACKET_ENABLE_AUDIO_INFOFRAME);
+		break;
 	case HDMI_INFOFRAME_TYPE_AVI:
 		/* send current AVI infoframe values while updating */
 		regmap_update_bits(adv7511->regmap, ADV7511_REG_INFOFRAME_UPDATE,

From f50b969bafafb2810a07f376387350c4c0d72a21 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Mon, 8 Sep 2025 14:05:48 +0800
Subject: [PATCH 1141/1455] drm/bridge: ite-it6263: Support HDMI vendor
 specific infoframe

IT6263 supports HDMI vendor specific infoframe.  The infoframe header
and payload are configurable via NULL packet registers.  The infoframe
is enabled and disabled via PKT_NULL_CTRL register.  Add the HDMI vendor
specific infoframe support.

Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250908-it6263-vendor-specific-infoframe-v2-1-3f2ebd9135ad@nxp.com
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
---
 drivers/gpu/drm/bridge/ite-it6263.c | 64 ++++++++++++++++++++---------
 1 file changed, 44 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/bridge/ite-it6263.c b/drivers/gpu/drm/bridge/ite-it6263.c
index cf813672b4ff..2eb8fba7016c 100644
--- a/drivers/gpu/drm/bridge/ite-it6263.c
+++ b/drivers/gpu/drm/bridge/ite-it6263.c
@@ -146,6 +146,7 @@
 #define  HDMI_COLOR_DEPTH_24		FIELD_PREP(HDMI_COLOR_DEPTH, 4)
 
 #define HDMI_REG_PKT_GENERAL_CTRL	0xc6
+#define HDMI_REG_PKT_NULL_CTRL		0xc9
 #define HDMI_REG_AVI_INFOFRM_CTRL	0xcd
 #define  ENABLE_PKT			BIT(0)
 #define  REPEAT_PKT			BIT(1)
@@ -154,6 +155,12 @@
  * 3) HDMI register bank1: 0x130 ~ 0x1ff (HDMI packet registers)
  */
 
+/* NULL packet registers */
+/* Header Byte(HB): n = 0 ~ 2 */
+#define HDMI_REG_PKT_HB(n)		(0x138 + (n))
+/* Packet Byte(PB): n = 0 ~ 27(HDMI_MAX_INFOFRAME_SIZE), n = 0 for checksum */
+#define HDMI_REG_PKT_PB(n)		(0x13b + (n))
+
 /* AVI packet registers */
 #define HDMI_REG_AVI_DB1		0x158
 #define HDMI_REG_AVI_DB2		0x159
@@ -224,7 +231,9 @@ static bool it6263_hdmi_writeable_reg(struct device *dev, unsigned int reg)
 	case HDMI_REG_HDMI_MODE:
 	case HDMI_REG_GCP:
 	case HDMI_REG_PKT_GENERAL_CTRL:
+	case HDMI_REG_PKT_NULL_CTRL:
 	case HDMI_REG_AVI_INFOFRM_CTRL:
+	case HDMI_REG_PKT_HB(0) ... HDMI_REG_PKT_PB(HDMI_MAX_INFOFRAME_SIZE):
 	case HDMI_REG_AVI_DB1:
 	case HDMI_REG_AVI_DB2:
 	case HDMI_REG_AVI_DB3:
@@ -755,10 +764,16 @@ static int it6263_hdmi_clear_infoframe(struct drm_bridge *bridge,
 {
 	struct it6263 *it = bridge_to_it6263(bridge);
 
-	if (type == HDMI_INFOFRAME_TYPE_AVI)
+	switch (type) {
+	case HDMI_INFOFRAME_TYPE_AVI:
 		regmap_write(it->hdmi_regmap, HDMI_REG_AVI_INFOFRM_CTRL, 0);
-	else
+		break;
+	case HDMI_INFOFRAME_TYPE_VENDOR:
+		regmap_write(it->hdmi_regmap, HDMI_REG_PKT_NULL_CTRL, 0);
+		break;
+	default:
 		dev_dbg(it->dev, "unsupported HDMI infoframe 0x%x\n", type);
+	}
 
 	return 0;
 }
@@ -770,27 +785,36 @@ static int it6263_hdmi_write_infoframe(struct drm_bridge *bridge,
 	struct it6263 *it = bridge_to_it6263(bridge);
 	struct regmap *regmap = it->hdmi_regmap;
 
-	if (type != HDMI_INFOFRAME_TYPE_AVI) {
+	switch (type) {
+	case HDMI_INFOFRAME_TYPE_AVI:
+		/* write the first AVI infoframe data byte chunk(DB1-DB5) */
+		regmap_bulk_write(regmap, HDMI_REG_AVI_DB1,
+				  &buffer[HDMI_INFOFRAME_HEADER_SIZE],
+				  HDMI_AVI_DB_CHUNK1_SIZE);
+
+		/* write the second AVI infoframe data byte chunk(DB6-DB13) */
+		regmap_bulk_write(regmap, HDMI_REG_AVI_DB6,
+				  &buffer[HDMI_INFOFRAME_HEADER_SIZE +
+					  HDMI_AVI_DB_CHUNK1_SIZE],
+				  HDMI_AVI_DB_CHUNK2_SIZE);
+
+		/* write checksum */
+		regmap_write(regmap, HDMI_REG_AVI_CSUM, buffer[3]);
+
+		regmap_write(regmap, HDMI_REG_AVI_INFOFRM_CTRL,
+			     ENABLE_PKT | REPEAT_PKT);
+		break;
+	case HDMI_INFOFRAME_TYPE_VENDOR:
+		/* write header and payload */
+		regmap_bulk_write(regmap, HDMI_REG_PKT_HB(0), buffer, len);
+
+		regmap_write(regmap, HDMI_REG_PKT_NULL_CTRL,
+			     ENABLE_PKT | REPEAT_PKT);
+		break;
+	default:
 		dev_dbg(it->dev, "unsupported HDMI infoframe 0x%x\n", type);
-		return 0;
 	}
 
-	/* write the first AVI infoframe data byte chunk(DB1-DB5) */
-	regmap_bulk_write(regmap, HDMI_REG_AVI_DB1,
-			  &buffer[HDMI_INFOFRAME_HEADER_SIZE],
-			  HDMI_AVI_DB_CHUNK1_SIZE);
-
-	/* write the second AVI infoframe data byte chunk(DB6-DB13) */
-	regmap_bulk_write(regmap, HDMI_REG_AVI_DB6,
-			  &buffer[HDMI_INFOFRAME_HEADER_SIZE +
-				  HDMI_AVI_DB_CHUNK1_SIZE],
-			  HDMI_AVI_DB_CHUNK2_SIZE);
-
-	/* write checksum */
-	regmap_write(regmap, HDMI_REG_AVI_CSUM, buffer[3]);
-
-	regmap_write(regmap, HDMI_REG_AVI_INFOFRM_CTRL, ENABLE_PKT | REPEAT_PKT);
-
 	return 0;
 }
 

From 1bafff0252e3e46286f558ba67d63bc16e8685cd Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:11 -0400
Subject: [PATCH 1142/1455] drm/i915: Change mutex initialization in
 intel_guc_log

The intel_guc_log->relay.lock is currently initialized in
intel_guc_log_init_early(), but it lacks a corresponding destructor,
which can lead to a memory leak.

This patch replaces the use of mutex_init() with drmm_mutex_init(),
which ensures the lock is properly destroyed when the driver is
unloaded.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-4-luiz.mello@estudante.ufscar.br
Acked-by: Tvrtko Ursulin <tursulin@ursulin.net>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_log.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
index 7135fdb0ebb4..469173791394 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
@@ -6,6 +6,8 @@
 #include <linux/debugfs.h>
 #include <linux/string_helpers.h>
 
+#include <drm/drm_managed.h>
+
 #include "gt/intel_gt.h"
 #include "i915_drv.h"
 #include "i915_irq.h"
@@ -512,7 +514,10 @@ static void guc_log_relay_unmap(struct intel_guc_log *log)
 
 void intel_guc_log_init_early(struct intel_guc_log *log)
 {
-	mutex_init(&log->relay.lock);
+	struct intel_guc *guc = log_to_guc(log);
+	struct drm_i915_private *i915 = guc_to_i915(guc);
+
+	drmm_mutex_init(&i915->drm, &log->relay.lock);
 	INIT_WORK(&log->relay.flush_work, copy_debug_logs_work);
 	log->relay.started = false;
 }

From 1bd3db82e9f39c45ce035ac07e7acd9a4cd2e1d3 Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:12 -0400
Subject: [PATCH 1143/1455] drm/i915: Replace struct_mutex in intel_guc_log
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove the use of struct_mutex from intel_guc_log.c and replace it with
a dedicated lock, guc_lock, defined within the intel_guc_log struct.
    
The struct_mutex was previously used to protect concurrent access and
modification of intel_guc_log->level in intel_guc_log_set_level().
However, it was suggested that the lock should reside within the
intel_guc_log struct itself.
    
Initialize the new guc_lock in intel_guc_log_init_early(), alongside the
existing relay.lock. The lock is initialized using drmm_mutex_init(),
which also ensures it is properly destroyed when the driver is unloaded.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Suggested-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-5-luiz.mello@estudante.ufscar.br
Acked-by: Tvrtko Ursulin <tursulin@ursulin.net>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_log.c | 5 +++--
 drivers/gpu/drm/i915/gt/uc/intel_guc_log.h | 8 ++++++++
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
index 469173791394..0104fffd5852 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.c
@@ -518,6 +518,7 @@ void intel_guc_log_init_early(struct intel_guc_log *log)
 	struct drm_i915_private *i915 = guc_to_i915(guc);
 
 	drmm_mutex_init(&i915->drm, &log->relay.lock);
+	drmm_mutex_init(&i915->drm, &log->guc_lock);
 	INIT_WORK(&log->relay.flush_work, copy_debug_logs_work);
 	log->relay.started = false;
 }
@@ -683,7 +684,7 @@ int intel_guc_log_set_level(struct intel_guc_log *log, u32 level)
 	if (level < GUC_LOG_LEVEL_DISABLED || level > GUC_LOG_LEVEL_MAX)
 		return -EINVAL;
 
-	mutex_lock(&i915->struct_mutex);
+	mutex_lock(&log->guc_lock);
 
 	if (log->level == level)
 		goto out_unlock;
@@ -701,7 +702,7 @@ int intel_guc_log_set_level(struct intel_guc_log *log, u32 level)
 	log->level = level;
 
 out_unlock:
-	mutex_unlock(&i915->struct_mutex);
+	mutex_unlock(&log->guc_lock);
 
 	return ret;
 }
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h
index 02127703be80..13cb93ad0710 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h
@@ -42,6 +42,14 @@ enum {
 struct intel_guc_log {
 	u32 level;
 
+	/*
+	 * Protects concurrent access and modification of intel_guc_log->level.
+	 *
+	 * This lock replaces the legacy struct_mutex usage in
+	 * intel_guc_log system.
+	 */
+	struct mutex guc_lock;
+
 	/* Allocation settings */
 	struct {
 		s32 bytes;	/* Size in bytes */

From 24af523936b81a479d65ff835e27e1a22afa0bf8 Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:13 -0400
Subject: [PATCH 1144/1455] drm/i915/gem: Clean-up outdated struct_mutex
 comments

The struct_mutex will be removed from the DRM subsystem, as it was a
legacy BKL that was only used by i915 driver. After review, it was
concluded that its usage was no longer necessary

This patch updates various comments in the i915/gem and i915/gt
codebase to either remove or clarify references to struct_mutex, in
order to prevent future misunderstandings.

* i915_gem_execbuffer.c: Replace reference to struct_mutex with
  vm->mutex, as noted in the eb_reserve() function, which states that
  vm->mutex handles deadlocks.
* i915_gem_object.c: Change struct_mutex by
  drm_i915_gem_object->vma.lock. i915_gem_object_unbind() in i915_gem.c
  states that this lock is who actually protects the unbind.
* i915_gem_shrinker.c: The correct lock is actually i915->mm.obj, as
  already documented in its declaration.
* i915_gem_wait.c: The existing comment already mentioned that
  struct_mutex was no longer necessary. Updated to refer to a generic
  global lock instead.
* intel_reset_types.h: Cleaned up the comment text. Updated to refer to
  a generic global lock instead.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-6-luiz.mello@estudante.ufscar.br
Acked-by: Tvrtko Ursulin <tursulin@ursulin.net>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 2 +-
 drivers/gpu/drm/i915/gem/i915_gem_object.c     | 4 ++--
 drivers/gpu/drm/i915/gem/i915_gem_shrinker.c   | 4 ++--
 drivers/gpu/drm/i915/gem/i915_gem_wait.c       | 8 ++++----
 drivers/gpu/drm/i915/gt/intel_reset_types.h    | 2 +-
 5 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
index ca7e9216934a..07500fd5b82d 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
@@ -182,7 +182,7 @@ enum {
  * the object. Simple! ... The relocation entries are stored in user memory
  * and so to access them we have to copy them into a local buffer. That copy
  * has to avoid taking any pagefaults as they may lead back to a GEM object
- * requiring the struct_mutex (i.e. recursive deadlock). So once again we split
+ * requiring the vm->mutex (i.e. recursive deadlock). So once again we split
  * the relocation into multiple passes. First we try to do everything within an
  * atomic context (avoid the pagefaults) which requires that we never wait. If
  * we detect that we may wait, or if we need to fault, then we have to fallback
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c
index 1f38e367c60b..478011e5ecb3 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_object.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c
@@ -459,8 +459,8 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj)
 	atomic_inc(&i915->mm.free_count);
 
 	/*
-	 * Since we require blocking on struct_mutex to unbind the freed
-	 * object from the GPU before releasing resources back to the
+	 * Since we require blocking on drm_i915_gem_object->vma.lock to unbind
+	 * the freed object from the GPU before releasing resources back to the
 	 * system, we can not do that directly from the RCU callback (which may
 	 * be a softirq context), but must instead then defer that work onto a
 	 * kthread. We use the RCU callback rather than move the freed object
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
index b81e67504bbe..7a3e74a6676e 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c
@@ -170,7 +170,7 @@ i915_gem_shrink(struct i915_gem_ww_ctx *ww,
 	 * Also note that although these lists do not hold a reference to
 	 * the object we can safely grab one here: The final object
 	 * unreferencing and the bound_list are both protected by the
-	 * dev->struct_mutex and so we won't ever be able to observe an
+	 * i915->mm.obj_lock and so we won't ever be able to observe an
 	 * object on the bound_list with a reference count equals 0.
 	 */
 	for (phase = phases; phase->list; phase++) {
@@ -185,7 +185,7 @@ i915_gem_shrink(struct i915_gem_ww_ctx *ww,
 
 		/*
 		 * We serialize our access to unreferenced objects through
-		 * the use of the struct_mutex. While the objects are not
+		 * the use of the obj_lock. While the objects are not
 		 * yet freed (due to RCU then a workqueue) we still want
 		 * to be able to shrink their pages, so they remain on
 		 * the unbound/bound list until actually freed.
diff --git a/drivers/gpu/drm/i915/gem/i915_gem_wait.c b/drivers/gpu/drm/i915/gem/i915_gem_wait.c
index 7127e90c1a8f..04a18580e7d4 100644
--- a/drivers/gpu/drm/i915/gem/i915_gem_wait.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_wait.c
@@ -222,10 +222,10 @@ static unsigned long to_wait_timeout(s64 timeout_ns)
  *
  * The wait ioctl with a timeout of 0 reimplements the busy ioctl. With any
  * non-zero timeout parameter the wait ioctl will wait for the given number of
- * nanoseconds on an object becoming unbusy. Since the wait itself does so
- * without holding struct_mutex the object may become re-busied before this
- * function completes. A similar but shorter * race condition exists in the busy
- * ioctl
+ * nanoseconds on an object becoming unbusy. Since the wait occurs without
+ * holding a global or exclusive lock the object may become re-busied before
+ * this function completes. A similar but shorter * race condition exists
+ * in the busy ioctl
  */
 int
 i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
diff --git a/drivers/gpu/drm/i915/gt/intel_reset_types.h b/drivers/gpu/drm/i915/gt/intel_reset_types.h
index 4f5fd393af6f..ee4eb574a219 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_reset_types.h
@@ -20,7 +20,7 @@ struct intel_reset {
 	 * FENCE registers).
 	 *
 	 * #I915_RESET_ENGINE[num_engines] - Since the driver doesn't need to
-	 * acquire the struct_mutex to reset an engine, we need an explicit
+	 * acquire a global lock to reset an engine, we need an explicit
 	 * flag to prevent two concurrent reset attempts in the same engine.
 	 * As the number of engines continues to grow, allocate the flags from
 	 * the most significant bits.

From 5be4fa95ec5ce3039a622517d10327639da29e44 Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:14 -0400
Subject: [PATCH 1145/1455] drm/i915/display: Remove outdated struct_mutex
 comments

The struct_mutex will be removed from the DRM subsystem, as it was a
legacy BKL that was only used by i915 driver. After review, it was
concluded that its usage was no longer necessary

This patch update a comment about struct_mutex in i915/display, in order
to prevent future misunderstandings.

* intel_fbc.c: Removed the statement that intel_fbc->lock is the inner
  lock when overlapping with struct_mutex, since struct_mutex is no
  longer used anywhere in the driver.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-7-luiz.mello@estudante.ufscar.br
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/display/intel_fbc.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 06c35c7ab4e2..0d380c825791 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -98,11 +98,7 @@ struct intel_fbc {
 	struct intel_display *display;
 	const struct intel_fbc_funcs *funcs;
 
-	/*
-	 * This is always the inner lock when overlapping with
-	 * struct_mutex and it's the outer lock when overlapping
-	 * with stolen_lock.
-	 */
+	/* This is always the outer lock when overlapping with stolen_lock */
 	struct mutex lock;
 	unsigned int busy_bits;
 

From 9e6651659b8a8686ca16fcefa7739600b047d8c5 Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:15 -0400
Subject: [PATCH 1146/1455] drm/i915: Clean-up outdated struct_mutex comments

The struct_mutex will be removed from the DRM subsystem, as it was a
legacy BKL that was only used by i915 driver. After review, it was
concluded that its usage was no longer necessary

This patch updates various comments in the i915 codebase to
either remove or clarify references to struct_mutex, in order to
prevent future misunderstandings.

* i915_drv.h: Removed the statement that stolen_lock is the inner lock
  when overlaps with struct_mutex, since struct_mutex is no longer used
  in the driver.
* i915_gem.c: Removed parentheses suggesting usage of struct_mutex, which
  which is no longer used.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-8-luiz.mello@estudante.ufscar.br
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_drv.h | 3 +--
 drivers/gpu/drm/i915/i915_gem.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 55f13cb2c395..50fdecb79b49 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -116,8 +116,7 @@ struct i915_gem_mm {
 	struct intel_memory_region *stolen_region;
 	/** Memory allocator for GTT stolen memory */
 	struct drm_mm stolen;
-	/** Protects the usage of the GTT stolen memory allocator. This is
-	 * always the inner lock when overlapping with struct_mutex. */
+	/** Protects the usage of the GTT stolen memory allocator */
 	struct mutex stolen_lock;
 
 	/* Protects bound_list/unbound_list and #drm_i915_gem_object.mm.link */
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 8c8d43451f35..e14a0c3db999 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -847,8 +847,7 @@ void i915_gem_runtime_suspend(struct drm_i915_private *i915)
 	/*
 	 * Only called during RPM suspend. All users of the userfault_list
 	 * must be holding an RPM wakeref to ensure that this can not
-	 * run concurrently with themselves (and use the struct_mutex for
-	 * protection between themselves).
+	 * run concurrently with themselves.
 	 */
 
 	list_for_each_entry_safe(obj, on,

From 34ac58ded89dd81fc2f776b19c44bcc48fe9764d Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:16 -0400
Subject: [PATCH 1147/1455] drm/i915: Drop unused struct_mutex from
 drm_i915_private

The struct_mutex field in drm_i915_private is no longer used anywhere in
the driver. This patch removes it completely to clean up unused code and
avoid confusion.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-9-luiz.mello@estudante.ufscar.br
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/i915_driver.c | 2 --
 drivers/gpu/drm/i915/i915_drv.h    | 8 --------
 2 files changed, 10 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
index 2677c39a3246..a28c3710c4d5 100644
--- a/drivers/gpu/drm/i915/i915_driver.c
+++ b/drivers/gpu/drm/i915/i915_driver.c
@@ -235,7 +235,6 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv)
 
 	intel_sbi_init(display);
 	vlv_iosf_sb_init(dev_priv);
-	mutex_init(&dev_priv->struct_mutex);
 	mutex_init(&dev_priv->sb_lock);
 
 	i915_memcpy_init_early(dev_priv);
@@ -294,7 +293,6 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv)
 	i915_workqueues_cleanup(dev_priv);
 
 	mutex_destroy(&dev_priv->sb_lock);
-	mutex_destroy(&dev_priv->struct_mutex);
 	vlv_iosf_sb_fini(dev_priv);
 	intel_sbi_fini(display);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 50fdecb79b49..7de68ebc2beb 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -223,14 +223,6 @@ struct drm_i915_private {
 
 	bool irqs_enabled;
 
-	/*
-	 * Currently, struct_mutex is only used by the i915 driver as a replacement
-	 * for BKL.
-	 *
-	 * For this reason, it is no longer part of struct drm_device.
-	 */
-	struct mutex struct_mutex;
-
 	/* LPT/WPT IOSF sideband protection */
 	struct mutex sbi_lock;
 

From b69f8c496ea05f80c2bae91a74b48c00c06c524e Mon Sep 17 00:00:00 2001
From: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Date: Mon, 8 Sep 2025 09:15:17 -0400
Subject: [PATCH 1148/1455] drm/i915: Remove todo and comments about
 struct_mutex

This patch completes the removal of struct_mutex from the driver.

Remove the related TODO item, as the transition away from struct_mutex
is now complete.

Also clean up references to struct_mutex in i915.rst to avoid outdated
documentation.

Signed-off-by: Luiz Otavio Mello <luiz.mello@estudante.ufscar.br>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250908131518.36625-10-luiz.mello@estudante.ufscar.br
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 Documentation/gpu/i915.rst |  7 -------
 Documentation/gpu/todo.rst | 25 -------------------------
 2 files changed, 32 deletions(-)

diff --git a/Documentation/gpu/i915.rst b/Documentation/gpu/i915.rst
index 72932fa31b8d..eba09c3ddce4 100644
--- a/Documentation/gpu/i915.rst
+++ b/Documentation/gpu/i915.rst
@@ -358,8 +358,6 @@ Locking Guidelines
 #. All locking rules and interface contracts with cross-driver interfaces
    (dma-buf, dma_fence) need to be followed.
 
-#. No struct_mutex anywhere in the code
-
 #. dma_resv will be the outermost lock (when needed) and ww_acquire_ctx
    is to be hoisted at highest level and passed down within i915_gem_ctx
    in the call chain
@@ -367,11 +365,6 @@ Locking Guidelines
 #. While holding lru/memory manager (buddy, drm_mm, whatever) locks
    system memory allocations are not allowed
 
-	* Enforce this by priming lockdep (with fs_reclaim). If we
-	  allocate memory while holding these looks we get a rehash
-	  of the shrinker vs. struct_mutex saga, and that would be
-	  real bad.
-
 #. Do not nest different lru/memory manager locks within each other.
    Take them in turn to update memory allocations, relying on the object’s
    dma_resv ww_mutex to serialize against other operations.
diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst
index be8637da3fe9..c796331d9acc 100644
--- a/Documentation/gpu/todo.rst
+++ b/Documentation/gpu/todo.rst
@@ -173,31 +173,6 @@ Contact: Simona Vetter
 
 Level: Intermediate
 
-Get rid of dev->struct_mutex from GEM drivers
----------------------------------------------
-
-``dev->struct_mutex`` is the Big DRM Lock from legacy days and infested
-everything. Nowadays in modern drivers the only bit where it's mandatory is
-serializing GEM buffer object destruction. Which unfortunately means drivers
-have to keep track of that lock and either call ``unreference`` or
-``unreference_locked`` depending upon context.
-
-Core GEM doesn't have a need for ``struct_mutex`` any more since kernel 4.8,
-and there's a GEM object ``free`` callback for any drivers which are
-entirely ``struct_mutex`` free.
-
-For drivers that need ``struct_mutex`` it should be replaced with a driver-
-private lock. The tricky part is the BO free functions, since those can't
-reliably take that lock any more. Instead state needs to be protected with
-suitable subordinate locks or some cleanup work pushed to a worker thread. For
-performance-critical drivers it might also be better to go with a more
-fine-grained per-buffer object and per-context lockings scheme. Currently only
-the ``msm`` and `i915` drivers use ``struct_mutex``.
-
-Contact: Simona Vetter, respective driver maintainers
-
-Level: Advanced
-
 Move Buffer Object Locking to dma_resv_lock()
 ---------------------------------------------
 

From f261f5dddec1acdf60198371048b052febf67cf5 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Fri, 5 Sep 2025 19:36:25 +0200
Subject: [PATCH 1149/1455] drm/xe/debugfs: Don't expose dgfx residencies
 attributes on VF

In addition of checking if we are running on the BATTLEMAGE,
we should also check for not being a VF driver, as VFs can't
access necessary registers, and doing so leads to:

 | .. [drm] GT0: VF is trying to read an inaccessible register 0x35b004+0x0
 | RIP: 0010:xe_gt_sriov_vf_read32+0x5e2/0x8a0 [xe]
 | Call Trace:
 |  xe_mmio_read32+0x110/0x280 [xe]
 |  read_residency_counter+0x42/0xd0 [xe]
 |  dgfx_pkg_residencies_show+0x115/0x190 [xe]
 | .. [drm] Package G2 counter failed to read, ret -19

or

 | .. [drm] GT0: VF is trying to read an inaccessible register 0x35b004+0x0
 | RIP: 0010:xe_gt_sriov_vf_read32+0x5e2/0x8a0 [xe]
 | Call Trace:
 |  xe_mmio_read32+0x110/0x280 [xe]
 |  read_residency_counter+0x42/0xd0 [xe]
 |  dgfx_pcie_link_residencies_show+0xe7/0x160 [xe]
 | .. [drm] PCIE LINK L0 RESIDENCY counter failed to read, ret -19

Similarly, there is no point to expose inject_csc_hw_error on VFs,
as HW errors support is already disabled for VFs.

Bspec: 53221
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Soham Purkait <soham.purkait@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Riana Tauro <riana.tauro@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905173625.8398-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_debugfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index 38169238dbed..d67e32ebbe29 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -347,7 +347,7 @@ void xe_debugfs_register(struct xe_device *xe)
 				 ARRAY_SIZE(debugfs_list),
 				 root, minor);
 
-	if (xe->info.platform == XE_BATTLEMAGE) {
+	if (xe->info.platform == XE_BATTLEMAGE && !IS_SRIOV_VF(xe)) {
 		drm_debugfs_create_files(debugfs_residencies,
 					 ARRAY_SIZE(debugfs_residencies),
 					 root, minor);

From f8d2c26df98b487d61e582c01804340b96c44e44 Mon Sep 17 00:00:00 2001
From: Jonathan Cavitt <jonathan.cavitt@intel.com>
Date: Thu, 4 Sep 2025 19:30:42 +0000
Subject: [PATCH 1150/1455] drm/i915/gt: Fix memory leak in hangcheck selftest

In active_engines, if intel_context_create fails, we need to go
backwards through all the created contexts to free/put them.  However,
the way this is currently performed skips the first created context, as
if count == 1, then --count returns 0 and exits the while-loop
prematurely without performing the intel_context_put on context 0.

Fix this by post-decrementing count in the while-loop, rather than
pre-decrementing it.  This change makes the prior guard against count
underflowing unnecessary, as the while-loop exits when count == 0.

Signed-off-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Cc: Krzysztof Karas <krzysztof.karas@intel.com>
Cc: Andi Shyti <andi.shyti@linux.intel.com>
Cc: Sebastian Brzezinka <sebastian.brzezinka@intel.com>
Reviewed-by: Krzysztof Karas <krzysztof.karas@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Andi Shyti <andi.shyti@linux.intel.com>
Link: https://lore.kernel.org/r/20250904193041.12888-2-jonathan.cavitt@intel.com
---
 drivers/gpu/drm/i915/gt/selftest_hangcheck.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c
index 619c70c54ef9..4f252f704975 100644
--- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c
+++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c
@@ -904,9 +904,7 @@ static void active_engine(struct kthread_work *work)
 			arg->result = PTR_ERR(ce[count]);
 			pr_err("[%s] Create context #%ld failed: %d!\n",
 			       engine->name, count, arg->result);
-			if (!count)
-				return;
-			while (--count)
+			while (count--)
 				intel_context_put(ce[count]);
 			return;
 		}

From 30071d58dfcef5da327e2f3b6ce848251814b266 Mon Sep 17 00:00:00 2001
From: Matt Roper <matthew.d.roper@intel.com>
Date: Fri, 5 Sep 2025 14:56:15 -0700
Subject: [PATCH 1151/1455] drm/xe: Never report L3 bank mask for media GT
 going forward

We currently report an L3 bank mask as part of the GT topology on both
GTs (primary and media) because a copy of the L3 bank fuse register
exists on both GTs (e.g., $gsi_offset + 0x9130 on Xe3).  After recent
discussions it's come to light that the only known userspace software
that uses this part of the uapi (the compute UMD and Mesa) only uses the
value reported for the primary GT; the value reported for the media GT
is ignored by both projects, and the media UMDs don't have any use for
L3 information today.  Since we always strive to have our uapi match the
specific needs of userspace and not include additional unused baggage,
let's officially drop L3 bank reporting on the media GT going forward
and only keep it around for the primary GT where it actually gets used.
This change will only apply to future platforms (Xe3 and later); even
though it would probably be safe to remove it from Xe1/Xe2 as well, we
don't want to take any chances with changing existing ABI.

Note that we'd already disabled reading/reporting of the L3 bank for the
media GT on PTL in commit 9ab440a9d042 ("drm/xe/ptl: L3bank mask is not
available on the media GT") because it was discovered that the copy of
the fuse registers on the media GT were just reporting a bogus ~0 value
rather than an accurate mask.  So this is just extending that PTL
behavior forward to WCL and other future platforms.  Note that we're
also free to reinstate this part of the uapi in the future if/when some
new userspace consumer emerges that _does_ have a use for media-specific
L3 bank masks.

Cc: Fei Yang <fei.yang@intel.com>
Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com>
Link: https://lore.kernel.org/r/20250905215614.796247-3-matthew.d.roper@intel.com
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_topology.c | 31 ++++++++++++++++++-----------
 drivers/gpu/drm/xe/xe_gt_topology.h |  2 ++
 drivers/gpu/drm/xe/xe_query.c       |  5 +++--
 drivers/gpu/drm/xe/xe_wa_oob.rules  |  1 -
 4 files changed, 24 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_topology.c b/drivers/gpu/drm/xe/xe_gt_topology.c
index 0ed7dc9044a5..4e61c5e39bcb 100644
--- a/drivers/gpu/drm/xe/xe_gt_topology.c
+++ b/drivers/gpu/drm/xe/xe_gt_topology.c
@@ -123,6 +123,21 @@ gen_l3_mask_from_pattern(struct xe_device *xe, xe_l3_bank_mask_t dst,
 	}
 }
 
+bool xe_gt_topology_report_l3(struct xe_gt *gt)
+{
+	/*
+	 * No known userspace needs/uses the L3 bank mask reported by
+	 * the media GT, and the hardware itself is known to report bogus
+	 * values on several platforms.  Only report L3 bank mask as part
+	 * of the media GT's topology on pre-Xe3 platforms since that's
+	 * already part of our ABI.
+	 */
+	if (xe_gt_is_media_type(gt) && MEDIA_VER(gt_to_xe(gt)) >= 30)
+		return false;
+
+	return true;
+}
+
 static void
 load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask)
 {
@@ -130,16 +145,7 @@ load_l3_bank_mask(struct xe_gt *gt, xe_l3_bank_mask_t l3_bank_mask)
 	struct xe_mmio *mmio = &gt->mmio;
 	u32 fuse3 = xe_mmio_read32(mmio, MIRROR_FUSE3);
 
-	/*
-	 * PTL platforms with media version 30.00 do not provide proper values
-	 * for the media GT's L3 bank registers.  Skip the readout since we
-	 * don't have any way to obtain real values.
-	 *
-	 * This may get re-described as an official workaround in the future,
-	 * but there's no tracking number assigned yet so we use a custom
-	 * OOB workaround descriptor.
-	 */
-	if (XE_GT_WA(gt, no_media_l3))
+	if (!xe_gt_topology_report_l3(gt))
 		return;
 
 	if (GRAPHICS_VER(xe) >= 30) {
@@ -276,8 +282,9 @@ xe_gt_topology_dump(struct xe_gt *gt, struct drm_printer *p)
 	drm_printf(p, "EU type:             %s\n",
 		   eu_type_to_str(gt->fuse_topo.eu_type));
 
-	drm_printf(p, "L3 bank mask:        %*pb\n", XE_MAX_L3_BANK_MASK_BITS,
-		   gt->fuse_topo.l3_bank_mask);
+	if (xe_gt_topology_report_l3(gt))
+		drm_printf(p, "L3 bank mask:        %*pb\n", XE_MAX_L3_BANK_MASK_BITS,
+			   gt->fuse_topo.l3_bank_mask);
 }
 
 /*
diff --git a/drivers/gpu/drm/xe/xe_gt_topology.h b/drivers/gpu/drm/xe/xe_gt_topology.h
index d95cdd6e45be..5e62f5949b7b 100644
--- a/drivers/gpu/drm/xe/xe_gt_topology.h
+++ b/drivers/gpu/drm/xe/xe_gt_topology.h
@@ -49,4 +49,6 @@ bool xe_gt_has_compute_dss(struct xe_gt *gt, unsigned int dss);
 
 bool xe_gt_has_discontiguous_dss_groups(const struct xe_gt *gt);
 
+bool xe_gt_topology_report_l3(struct xe_gt *gt);
+
 #endif /* _XE_GT_TOPOLOGY_H_ */
diff --git a/drivers/gpu/drm/xe/xe_query.c b/drivers/gpu/drm/xe/xe_query.c
index 4dbe5732cb7f..e1b603aba61b 100644
--- a/drivers/gpu/drm/xe/xe_query.c
+++ b/drivers/gpu/drm/xe/xe_query.c
@@ -21,6 +21,7 @@
 #include "xe_force_wake.h"
 #include "xe_ggtt.h"
 #include "xe_gt.h"
+#include "xe_gt_topology.h"
 #include "xe_guc_hwconfig.h"
 #include "xe_macros.h"
 #include "xe_mmio.h"
@@ -477,7 +478,7 @@ static size_t calc_topo_query_size(struct xe_device *xe)
 			sizeof_field(struct xe_gt, fuse_topo.eu_mask_per_dss);
 
 		/* L3bank mask may not be available for some GTs */
-		if (!XE_GT_WA(gt, no_media_l3))
+		if (xe_gt_topology_report_l3(gt))
 			query_size += sizeof(struct drm_xe_query_topology_mask) +
 				sizeof_field(struct xe_gt, fuse_topo.l3_bank_mask);
 	}
@@ -540,7 +541,7 @@ static int query_gt_topology(struct xe_device *xe,
 		 * mask, then it's better to omit L3 from the query rather than
 		 * reporting bogus or zeroed information to userspace.
 		 */
-		if (!XE_GT_WA(gt, no_media_l3)) {
+		if (xe_gt_topology_report_l3(gt)) {
 			topo.type = DRM_XE_TOPO_L3_BANK;
 			err = copy_mask(&query_ptr, &topo, gt->fuse_topo.l3_bank_mask,
 					sizeof(gt->fuse_topo.l3_bank_mask));
diff --git a/drivers/gpu/drm/xe/xe_wa_oob.rules b/drivers/gpu/drm/xe/xe_wa_oob.rules
index 338c344dcd7d..f3a6d5d239ce 100644
--- a/drivers/gpu/drm/xe/xe_wa_oob.rules
+++ b/drivers/gpu/drm/xe/xe_wa_oob.rules
@@ -49,7 +49,6 @@
 16023588340	GRAPHICS_VERSION(2001), FUNC(xe_rtp_match_not_sriov_vf)
 14019789679	GRAPHICS_VERSION(1255)
 		GRAPHICS_VERSION_RANGE(1270, 2004)
-no_media_l3	MEDIA_VERSION_RANGE(3000, 3002)
 14022866841	GRAPHICS_VERSION(3000), GRAPHICS_STEP(A0, B0)
 		MEDIA_VERSION(3000), MEDIA_STEP(A0, B0)
 16021333562	GRAPHICS_VERSION_RANGE(1200, 1274)

From 955f3bc4af440bb950c7a1567197aaf6aa2213ae Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 12:20:51 +0200
Subject: [PATCH 1152/1455] drm/xe/guc: Always add CT disable action during
 second init step

On DGFX, during init_post_hwconfig() step, we are reinitializing
CTB BO in VRAM and we have to replace cleanup action to disable CT
communication prior to release of underlying BO.

But that introduces some discrepancy between DGFX and iGFX, as for
iGFX we keep previously added disable CT action that would be called
during unwind much later.

To keep the same flow on both types of platforms, always replace old
cleanup action and register new one.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908102053.539-2-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_guc_ct.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index f40543b040d6..12372d5b7067 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -300,12 +300,11 @@ int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct)
 
 	xe_assert(xe, !xe_guc_ct_enabled(ct));
 
-	if (!IS_DGFX(xe))
-		return 0;
-
-	ret = xe_managed_bo_reinit_in_vram(xe, tile, &ct->bo);
-	if (ret)
-		return ret;
+	if (IS_DGFX(xe)) {
+		ret = xe_managed_bo_reinit_in_vram(xe, tile, &ct->bo);
+		if (ret)
+			return ret;
+	}
 
 	devm_release_action(xe->drm.dev, guc_action_disable_ct, ct);
 	return devm_add_action_or_reset(xe->drm.dev, guc_action_disable_ct, ct);

From 43fac1b2f015daedc6046863f7664f3178f0bbe2 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Mon, 8 Sep 2025 12:20:52 +0200
Subject: [PATCH 1153/1455] drm/xe/guc: Don't invoke disable_ct action during
 replacement

During second CT initialization step, known as post_hwconfig, we
want to replace previously registered CT disable devm action to
make sure it will be invoked prior to releasing underlying BO.

But to replace this action we don't need to execute it right away
since we know that CT was disabled prior to that late init step
and we already assert that. Use devm_remove_action() instead to
avoid extra message about 'disabling CT' that could be seen now:

 [drm:guc_ct_change_state [xe]] GT0: GuC CT communication channel disabled
 ...
 DEVRES REM ff11000149320940 guc_action_disable_ct (16 bytes)
 [drm:guc_ct_change_state [xe]] GT0: GuC CT communication channel disabled
 DEVRES ADD ff110001664fc040 guc_action_disable_ct (16 bytes)

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Link: https://lore.kernel.org/r/20250908102053.539-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_guc_ct.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index 12372d5b7067..bb8650fdeb45 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -306,7 +306,7 @@ int xe_guc_ct_init_post_hwconfig(struct xe_guc_ct *ct)
 			return ret;
 	}
 
-	devm_release_action(xe->drm.dev, guc_action_disable_ct, ct);
+	devm_remove_action(xe->drm.dev, guc_action_disable_ct, ct);
 	return devm_add_action_or_reset(xe->drm.dev, guc_action_disable_ct, ct);
 }
 

From 0d40ea784304358753bfbc8824d2a7d42687bd1e Mon Sep 17 00:00:00 2001
From: Julia Filipchuk <julia.filipchuk@intel.com>
Date: Fri, 29 Aug 2025 16:46:56 -0700
Subject: [PATCH 1154/1455] drm/xe/guc: Recommend GUC v70.49.4 for PTL, BMG

UAPI compatibility version 1.24.4

Resolves various BMG, SRIOV issues and adds new PTL features.

Signed-off-by: Julia Filipchuk <julia.filipchuk@intel.com>
Reviewed-by: Zhanjun Dong <zhanjun.dong@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20250829235305.672287-7-julia.filipchuk@intel.com
---
 drivers/gpu/drm/xe/xe_uc_fw.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_uc_fw.c b/drivers/gpu/drm/xe/xe_uc_fw.c
index 9bbdde604923..4faf8b0fbcfe 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw.c
+++ b/drivers/gpu/drm/xe/xe_uc_fw.c
@@ -115,8 +115,8 @@ struct fw_blobs_by_type {
 #define XE_GT_TYPE_ANY XE_GT_TYPE_UNINITIALIZED
 
 #define XE_GUC_FIRMWARE_DEFS(fw_def, mmp_ver, major_ver)					\
-	fw_def(PANTHERLAKE,	GT_TYPE_ANY,	major_ver(xe,	guc,	ptl,	70, 47, 0))	\
-	fw_def(BATTLEMAGE,	GT_TYPE_ANY,	major_ver(xe,	guc,	bmg,	70, 45, 2))	\
+	fw_def(PANTHERLAKE,	GT_TYPE_ANY,	major_ver(xe,	guc,	ptl,	70, 49, 4))	\
+	fw_def(BATTLEMAGE,	GT_TYPE_ANY,	major_ver(xe,	guc,	bmg,	70, 49, 4))	\
 	fw_def(LUNARLAKE,	GT_TYPE_ANY,	major_ver(xe,	guc,	lnl,	70, 45, 2))	\
 	fw_def(METEORLAKE,	GT_TYPE_ANY,	major_ver(i915,	guc,	mtl,	70, 44, 1))	\
 	fw_def(DG2,		GT_TYPE_ANY,	major_ver(i915,	guc,	dg2,	70, 45, 2))	\

From a525fa37aac36c4591cc8b07ae8957862415fbd5 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Thu, 4 Sep 2025 12:35:05 -0400
Subject: [PATCH 1155/1455] drm/amdgpu: fix a memory leak in fence cleanup when
 unloading
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Commit b61badd20b44 ("drm/amdgpu: fix usage slab after free")
reordered when amdgpu_fence_driver_sw_fini() was called after
that patch, amdgpu_fence_driver_sw_fini() effectively became
a no-op as the sched entities we never freed because the
ring pointers were already set to NULL.  Remove the NULL
setting.

Reported-by: Lin.Cao <lincao12@amd.com>
Cc: Vitaly Prosyak <vitaly.prosyak@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Fixes: b61badd20b44 ("drm/amdgpu: fix usage slab after free")
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 6379bb25bf5c..486c3646710c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -421,8 +421,6 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 	dma_fence_put(ring->vmid_wait);
 	ring->vmid_wait = NULL;
 	ring->me = 0;
-
-	ring->adev->rings[ring->idx] = NULL;
 }
 
 /**

From 4e89d629dc72965ea05a9b5231862ffd9c9dab55 Mon Sep 17 00:00:00 2001
From: John Olender <john.olender@gmail.com>
Date: Fri, 5 Sep 2025 06:11:28 -0400
Subject: [PATCH 1156/1455] drm/amdgpu: Fix NULL ptr deref in
 amdgpu_device_cache_switch_state()

Kaveri has no upstream bridge, therefore parent is NULL.

$ lspci -PP
...
00:01.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Kaveri [Radeon R7 Graphics] (rev d4)

For comparison, Raphael:

$ lspci -PP
...
00:08.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Raphael/Granite Ridge Internal GPP Bridge to Bus [C:A]
...
00:08.1/0e:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Raphael (rev c5)

Fixes: 1dd2fa0e00f1 ("drm/amdgpu: Save and restore switch state")
Link: https://lore.kernel.org/amd-gfx/38fe6513-f8a9-4669-8e86-89c54c465611@gmail.com/
Reviewed-by: Candice Li <candice.li@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: John Olender <john.olender@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index bdfb80377e6a..b20383021b50 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -7161,7 +7161,7 @@ static void amdgpu_device_cache_switch_state(struct amdgpu_device *adev)
 	struct pci_dev *parent = pci_upstream_bridge(adev->pdev);
 	int r;
 
-	if (parent->vendor != PCI_VENDOR_ID_ATI)
+	if (!parent || parent->vendor != PCI_VENDOR_ID_ATI)
 		return;
 
 	/* If already saved, return */

From 27e4dc2c0543fd1808cc52bd888ee1e0533c4a2e Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 3 Sep 2025 09:11:12 -0400
Subject: [PATCH 1157/1455] drm/amd/display: use udelay rather than fsleep

This function can be called from an atomic context so we can't use
fsleep().

Fixes: 01f60348d8fb ("drm/amd/display: Fix 'failed to blank crtc!'")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4549
Cc: Wen Chen <Wen.Chen3@amd.com>
Cc: Fangzhi Zuo <jerry.zuo@amd.com>
Cc: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Cc: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index 7d24fa1517bf..cc377fcda6ff 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -955,7 +955,7 @@ enum dc_status dcn20_enable_stream_timing(
 		return DC_ERROR_UNEXPECTED;
 	}
 
-	fsleep(stream->timing.v_total * (stream->timing.h_total * 10000u / stream->timing.pix_clk_100hz));
+	udelay(stream->timing.v_total * (stream->timing.h_total * 10000u / stream->timing.pix_clk_100hz));
 
 	params.vertical_total_min = stream->adjust.v_total_min;
 	params.vertical_total_max = stream->adjust.v_total_max;

From d97b74a833eba1f4f69f67198fd98ef036c0e5f9 Mon Sep 17 00:00:00 2001
From: Pratap Nirujogi <pratap.nirujogi@amd.com>
Date: Wed, 3 Sep 2025 16:00:24 -0400
Subject: [PATCH 1158/1455] drm/amd/amdgpu: Declare isp firmware binary file

Declare isp firmware file isp_4_1_1.bin required by isp4.1.1 device.

Suggested-by: Alexey Zagorodnikov <xglooom@gmail.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Pratap Nirujogi <pratap.nirujogi@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c
index a887df520414..4258d3e0b706 100644
--- a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c
@@ -29,6 +29,8 @@
 #include "amdgpu.h"
 #include "isp_v4_1_1.h"
 
+MODULE_FIRMWARE("amdgpu/isp_4_1_1.bin");
+
 #define ISP_PERFORMANCE_STATE_LOW 0
 #define ISP_PERFORMANCE_STATE_HIGH 1
 

From 60df8a5d8f6505974784f7290fdfa94e2aa4e255 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Tue, 26 Aug 2025 15:49:20 +0530
Subject: [PATCH 1159/1455] drm/ttm: Bump TTM_NUM_MEM_TYPES to 9 (Prep for
 AMDGPU_PL_MMIO_REMAP)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Increase TTM_NUM_MEM_TYPES from 8 to 9 to accommodate the upcoming
AMDGPU_PL_MMIO_REMAP placement.

Cc: Alex Deucher <alexander.deucher@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 include/drm/ttm/ttm_resource.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/drm/ttm/ttm_resource.h b/include/drm/ttm/ttm_resource.h
index e52bba15012f..f49daa504c36 100644
--- a/include/drm/ttm/ttm_resource.h
+++ b/include/drm/ttm/ttm_resource.h
@@ -36,7 +36,7 @@
 #include <drm/ttm/ttm_kmap_iter.h>
 
 #define TTM_MAX_BO_PRIORITY	4U
-#define TTM_NUM_MEM_TYPES 8
+#define TTM_NUM_MEM_TYPES 9
 
 struct dmem_cgroup_device;
 struct ttm_device;

From 056132483724a1ba1ff8823914dace71f8e8938c Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 20 Aug 2025 12:33:29 +0530
Subject: [PATCH 1160/1455] drm/amdgpu/uapi: Introduce
 AMDGPU_GEM_DOMAIN_MMIO_REMAP
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a new GEM domain bit AMDGPU_GEM_DOMAIN_MMIO_REMAP to allow
userspace to request the MMIO remap (HDP flush) page via GEM_CREATE.

- include/uapi/drm/amdgpu_drm.h:
  * define AMDGPU_GEM_DOMAIN_MMIO_REMAP
  * include the bit in AMDGPU_GEM_DOMAIN_MASK

v2: Add early reject in amdgpu_gem_create_ioctl() (Alex).

Cc: Christian König <christian.koenig@amd.com>
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 3 +++
 include/uapi/drm/amdgpu_drm.h           | 8 ++++++--
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 630175746780..a8fa09184459 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -457,6 +457,9 @@ int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
 	/* always clear VRAM */
 	flags |= AMDGPU_GEM_CREATE_VRAM_CLEARED;
 
+	if (args->in.domains & AMDGPU_GEM_DOMAIN_MMIO_REMAP)
+		return -EINVAL;
+
 	/* create a gem object to contain this object in */
 	if (args->in.domains & (AMDGPU_GEM_DOMAIN_GDS |
 	    AMDGPU_GEM_DOMAIN_GWS | AMDGPU_GEM_DOMAIN_OA)) {
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 9cebd072a042..85b3ca14f81e 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -105,6 +105,8 @@ extern "C" {
  *
  * %AMDGPU_GEM_DOMAIN_DOORBELL	Doorbell. It is an MMIO region for
  * signalling user mode queues.
+ *
+ * %AMDGPU_GEM_DOMAIN_MMIO_REMAP	MMIO remap page (special mapping for HDP flushing).
  */
 #define AMDGPU_GEM_DOMAIN_CPU		0x1
 #define AMDGPU_GEM_DOMAIN_GTT		0x2
@@ -113,13 +115,15 @@ extern "C" {
 #define AMDGPU_GEM_DOMAIN_GWS		0x10
 #define AMDGPU_GEM_DOMAIN_OA		0x20
 #define AMDGPU_GEM_DOMAIN_DOORBELL	0x40
+#define AMDGPU_GEM_DOMAIN_MMIO_REMAP	0x80
 #define AMDGPU_GEM_DOMAIN_MASK		(AMDGPU_GEM_DOMAIN_CPU | \
 					 AMDGPU_GEM_DOMAIN_GTT | \
 					 AMDGPU_GEM_DOMAIN_VRAM | \
 					 AMDGPU_GEM_DOMAIN_GDS | \
 					 AMDGPU_GEM_DOMAIN_GWS | \
-					 AMDGPU_GEM_DOMAIN_OA | \
-					 AMDGPU_GEM_DOMAIN_DOORBELL)
+					 AMDGPU_GEM_DOMAIN_OA |	\
+					 AMDGPU_GEM_DOMAIN_DOORBELL | \
+					 AMDGPU_GEM_DOMAIN_MMIO_REMAP)
 
 /* Flag that CPU access will be required for the case of VRAM domain */
 #define AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED	(1 << 0)

From 219be4711a1ba788bc2a9fafc117139d133e5fea Mon Sep 17 00:00:00 2001
From: Prike Liang <Prike.Liang@amd.com>
Date: Wed, 14 May 2025 12:43:57 +0800
Subject: [PATCH 1161/1455] drm/amdgpu: validate userq input args

This will help on validating the userq input args, and
rejecting for the invalid userq request at the IOCTLs
first place.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c  | 81 +++++++++++++++-------
 drivers/gpu/drm/amd/amdgpu/mes_userqueue.c |  7 --
 2 files changed, 56 insertions(+), 32 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 467e8fa6cb8b..f8648e169f8b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -404,27 +404,10 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args)
 		(args->in.flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_MASK) >>
 		AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_SHIFT;
 
-	/* Usermode queues are only supported for GFX IP as of now */
-	if (args->in.ip_type != AMDGPU_HW_IP_GFX &&
-	    args->in.ip_type != AMDGPU_HW_IP_DMA &&
-	    args->in.ip_type != AMDGPU_HW_IP_COMPUTE) {
-		drm_file_err(uq_mgr->file, "Usermode queue doesn't support IP type %u\n",
-			     args->in.ip_type);
-		return -EINVAL;
-	}
-
 	r = amdgpu_userq_priority_permit(filp, priority);
 	if (r)
 		return r;
 
-	if ((args->in.flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE) &&
-	    (args->in.ip_type != AMDGPU_HW_IP_GFX) &&
-	    (args->in.ip_type != AMDGPU_HW_IP_COMPUTE) &&
-	    !amdgpu_is_tmz(adev)) {
-		drm_file_err(uq_mgr->file, "Secure only supported on GFX/Compute queues\n");
-		return -EINVAL;
-	}
-
 	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
 	if (r < 0) {
 		drm_file_err(uq_mgr->file, "pm_runtime_get_sync() failed for userqueue create\n");
@@ -543,22 +526,45 @@ unlock:
 	return r;
 }
 
-int amdgpu_userq_ioctl(struct drm_device *dev, void *data,
-		       struct drm_file *filp)
+static int amdgpu_userq_input_args_validate(struct drm_device *dev,
+					union drm_amdgpu_userq *args,
+					struct drm_file *filp)
 {
-	union drm_amdgpu_userq *args = data;
-	int r;
+	struct amdgpu_device *adev = drm_to_adev(dev);
 
 	switch (args->in.op) {
 	case AMDGPU_USERQ_OP_CREATE:
 		if (args->in.flags & ~(AMDGPU_USERQ_CREATE_FLAGS_QUEUE_PRIORITY_MASK |
 				       AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE))
 			return -EINVAL;
-		r = amdgpu_userq_create(filp, args);
-		if (r)
-			drm_file_err(filp, "Failed to create usermode queue\n");
-		break;
+		/* Usermode queues are only supported for GFX IP as of now */
+		if (args->in.ip_type != AMDGPU_HW_IP_GFX &&
+		    args->in.ip_type != AMDGPU_HW_IP_DMA &&
+		    args->in.ip_type != AMDGPU_HW_IP_COMPUTE) {
+			drm_file_err(filp, "Usermode queue doesn't support IP type %u\n",
+				     args->in.ip_type);
+			return -EINVAL;
+		}
 
+		if ((args->in.flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE) &&
+		    (args->in.ip_type != AMDGPU_HW_IP_GFX) &&
+		    (args->in.ip_type != AMDGPU_HW_IP_COMPUTE) &&
+		    !amdgpu_is_tmz(adev)) {
+			drm_file_err(filp, "Secure only supported on GFX/Compute queues\n");
+			return -EINVAL;
+		}
+
+		if (args->in.queue_va == AMDGPU_BO_INVALID_OFFSET ||
+		    args->in.queue_va == 0 ||
+		    args->in.queue_size == 0) {
+			drm_file_err(filp, "invalidate userq queue va or size\n");
+			return -EINVAL;
+		}
+		if (!args->in.wptr_va || !args->in.rptr_va) {
+			drm_file_err(filp, "invalidate userq queue rptr or wptr\n");
+			return -EINVAL;
+		}
+		break;
 	case AMDGPU_USERQ_OP_FREE:
 		if (args->in.ip_type ||
 		    args->in.doorbell_handle ||
@@ -571,6 +577,31 @@ int amdgpu_userq_ioctl(struct drm_device *dev, void *data,
 		    args->in.mqd ||
 		    args->in.mqd_size)
 			return -EINVAL;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int amdgpu_userq_ioctl(struct drm_device *dev, void *data,
+		       struct drm_file *filp)
+{
+	union drm_amdgpu_userq *args = data;
+	int r;
+
+	if (amdgpu_userq_input_args_validate(dev, args, filp) < 0)
+		return -EINVAL;
+
+	switch (args->in.op) {
+	case AMDGPU_USERQ_OP_CREATE:
+		r = amdgpu_userq_create(filp, args);
+		if (r)
+			drm_file_err(filp, "Failed to create usermode queue\n");
+		break;
+
+	case AMDGPU_USERQ_OP_FREE:
 		r = amdgpu_userq_destroy(filp, args->in.queue_id);
 		if (r)
 			drm_file_err(filp, "Failed to destroy usermode queue\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
index aee26f80bd53..66467f41294c 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
@@ -263,13 +263,6 @@ static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 		return -ENOMEM;
 	}
 
-	if (!mqd_user->wptr_va || !mqd_user->rptr_va ||
-	    !mqd_user->queue_va || mqd_user->queue_size == 0) {
-		DRM_ERROR("Invalid MQD parameters for userqueue\n");
-		r = -EINVAL;
-		goto free_props;
-	}
-
 	r = amdgpu_userq_create_object(uq_mgr, &queue->mqd, mqd_hw_default->mqd_size);
 	if (r) {
 		DRM_ERROR("Failed to create MQD object for userqueue\n");

From 1697398555f69b31e939e070b304292513d4c9ff Mon Sep 17 00:00:00 2001
From: Ray Wu <ray.wu@amd.com>
Date: Mon, 8 Sep 2025 17:21:27 +0800
Subject: [PATCH 1162/1455] drm/amd/display: Remove duplicated code

[Why&How]
Remove duplicated code

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c | 3 ---
 .../gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c   | 3 ---
 drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c | 3 ---
 3 files changed, 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
index 8475c6eec547..1d1a002f6d54 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
@@ -1900,9 +1900,6 @@ static bool dcn35_resource_construct(
 	dc->caps.num_of_host_routers = 2;
 	dc->caps.num_of_dpias_per_host_router = 2;
 
-	dc->caps.num_of_host_routers = 2;
-	dc->caps.num_of_dpias_per_host_router = 2;
-
 	/* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order
 	 * to provide some margin.
 	 * It's expected for furture ASIC to have equal or higher value, in order to
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
index 0971c0f74186..47912e0861a2 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
@@ -1872,9 +1872,6 @@ static bool dcn351_resource_construct(
 	dc->caps.num_of_host_routers = 2;
 	dc->caps.num_of_dpias_per_host_router = 2;
 
-	dc->caps.num_of_host_routers = 2;
-	dc->caps.num_of_dpias_per_host_router = 2;
-
 	/* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order
 	 * to provide some margin.
 	 * It's expected for furture ASIC to have equal or higher value, in order to
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
index 8bae7fcedc22..9ba91e214dda 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
@@ -1873,9 +1873,6 @@ static bool dcn36_resource_construct(
 	dc->caps.num_of_host_routers = 2;
 	dc->caps.num_of_dpias_per_host_router = 2;
 
-	dc->caps.num_of_host_routers = 2;
-	dc->caps.num_of_dpias_per_host_router = 2;
-
 	/* max_disp_clock_khz_at_vmin is slightly lower than the STA value in order
 	 * to provide some margin.
 	 * It's expected for furture ASIC to have equal or higher value, in order to

From 28f75f9bcc7da7da12e5dae2ae8d8629a2b2e42e Mon Sep 17 00:00:00 2001
From: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Date: Tue, 19 Aug 2025 16:11:01 +0530
Subject: [PATCH 1163/1455] drm/amdgpu/jpeg: Move parse_cs to amdgpu_jpeg.c

Rename jpeg_v2_dec_ring_parse_cs to amdgpu_jpeg_dec_parse_cs
and move it to amdgpu_jpeg.c as it is shared among jpeg versions.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c | 65 ++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h | 10 ++++
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c   | 58 +--------------------
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h   |  6 ---
 drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c   |  4 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c   |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c   |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c |  2 +-
 drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c |  2 +-
 10 files changed, 83 insertions(+), 70 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
index f0d7e2487237..e7b4b768f7d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c
@@ -539,3 +539,68 @@ void amdgpu_jpeg_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_pri
 			drm_printf(p, "\nInactive Instance:JPEG%d\n", i);
 	}
 }
+
+static inline bool amdgpu_jpeg_reg_valid(u32 reg)
+{
+	if (reg < JPEG_REG_RANGE_START || reg > JPEG_REG_RANGE_END ||
+	    (reg >= JPEG_ATOMIC_RANGE_START && reg <= JPEG_ATOMIC_RANGE_END))
+		return false;
+	else
+		return true;
+}
+
+/**
+ * amdgpu_jpeg_dec_parse_cs - command submission parser
+ *
+ * @parser: Command submission parser context
+ * @job: the job to parse
+ * @ib: the IB to parse
+ *
+ * Parse the command stream, return -EINVAL for invalid packet,
+ * 0 otherwise
+ */
+
+int amdgpu_jpeg_dec_parse_cs(struct amdgpu_cs_parser *parser,
+			      struct amdgpu_job *job,
+			      struct amdgpu_ib *ib)
+{
+	u32 i, reg, res, cond, type;
+	struct amdgpu_device *adev = parser->adev;
+
+	for (i = 0; i < ib->length_dw ; i += 2) {
+		reg  = CP_PACKETJ_GET_REG(ib->ptr[i]);
+		res  = CP_PACKETJ_GET_RES(ib->ptr[i]);
+		cond = CP_PACKETJ_GET_COND(ib->ptr[i]);
+		type = CP_PACKETJ_GET_TYPE(ib->ptr[i]);
+
+		if (res) /* only support 0 at the moment */
+			return -EINVAL;
+
+		switch (type) {
+		case PACKETJ_TYPE0:
+			if (cond != PACKETJ_CONDITION_CHECK0 ||
+			    !amdgpu_jpeg_reg_valid(reg)) {
+				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+				return -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE3:
+			if (cond != PACKETJ_CONDITION_CHECK3 ||
+			    !amdgpu_jpeg_reg_valid(reg)) {
+				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+				return -EINVAL;
+			}
+			break;
+		case PACKETJ_TYPE6:
+			if (ib->ptr[i] == CP_PACKETJ_NOP)
+				continue;
+			dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
+			return -EINVAL;
+		default:
+			dev_err(adev->dev, "Unknown packet type %d !\n", type);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
index 4f0775e39b54..346ae0ab09d3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h
@@ -25,11 +25,18 @@
 #define __AMDGPU_JPEG_H__
 
 #include "amdgpu_ras.h"
+#include "amdgpu_cs.h"
 
 #define AMDGPU_MAX_JPEG_INSTANCES	4
 #define AMDGPU_MAX_JPEG_RINGS           10
 #define AMDGPU_MAX_JPEG_RINGS_4_0_3     8
 
+#define JPEG_REG_RANGE_START            0x4000
+#define JPEG_REG_RANGE_END              0x41c2
+#define JPEG_ATOMIC_RANGE_START         0x4120
+#define JPEG_ATOMIC_RANGE_END           0x412A
+
+
 #define AMDGPU_JPEG_HARVEST_JPEG0 (1 << 0)
 #define AMDGPU_JPEG_HARVEST_JPEG1 (1 << 1)
 
@@ -170,5 +177,8 @@ int amdgpu_jpeg_reg_dump_init(struct amdgpu_device *adev,
 			       const struct amdgpu_hwip_reg_entry *reg, u32 count);
 void amdgpu_jpeg_dump_ip_state(struct amdgpu_ip_block *ip_block);
 void amdgpu_jpeg_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p);
+int amdgpu_jpeg_dec_parse_cs(struct amdgpu_cs_parser *parser,
+			     struct amdgpu_job *job,
+			     struct amdgpu_ib *ib);
 
 #endif /*__AMDGPU_JPEG_H__*/
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
index 58239c405fda..27c76bd424cf 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c
@@ -23,7 +23,6 @@
 
 #include "amdgpu.h"
 #include "amdgpu_jpeg.h"
-#include "amdgpu_cs.h"
 #include "amdgpu_pm.h"
 #include "soc15.h"
 #include "soc15d.h"
@@ -806,7 +805,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v2_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v2_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v2_0_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
@@ -854,58 +853,3 @@ const struct amdgpu_ip_block_version jpeg_v2_0_ip_block = {
 		.rev = 0,
 		.funcs = &jpeg_v2_0_ip_funcs,
 };
-
-/**
- * jpeg_v2_dec_ring_parse_cs - command submission parser
- *
- * @parser: Command submission parser context
- * @job: the job to parse
- * @ib: the IB to parse
- *
- * Parse the command stream, return -EINVAL for invalid packet,
- * 0 otherwise
- */
-int jpeg_v2_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
-			      struct amdgpu_job *job,
-			      struct amdgpu_ib *ib)
-{
-	u32 i, reg, res, cond, type;
-	struct amdgpu_device *adev = parser->adev;
-
-	for (i = 0; i < ib->length_dw ; i += 2) {
-		reg  = CP_PACKETJ_GET_REG(ib->ptr[i]);
-		res  = CP_PACKETJ_GET_RES(ib->ptr[i]);
-		cond = CP_PACKETJ_GET_COND(ib->ptr[i]);
-		type = CP_PACKETJ_GET_TYPE(ib->ptr[i]);
-
-		if (res) /* only support 0 at the moment */
-			return -EINVAL;
-
-		switch (type) {
-		case PACKETJ_TYPE0:
-			if (cond != PACKETJ_CONDITION_CHECK0 || reg < JPEG_REG_RANGE_START ||
-			    reg > JPEG_REG_RANGE_END) {
-				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
-				return -EINVAL;
-			}
-			break;
-		case PACKETJ_TYPE3:
-			if (cond != PACKETJ_CONDITION_CHECK3 || reg < JPEG_REG_RANGE_START ||
-			    reg > JPEG_REG_RANGE_END) {
-				dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
-				return -EINVAL;
-			}
-			break;
-		case PACKETJ_TYPE6:
-			if (ib->ptr[i] == CP_PACKETJ_NOP)
-				continue;
-			dev_err(adev->dev, "Invalid packet [0x%08x]!\n", ib->ptr[i]);
-			return -EINVAL;
-		default:
-			dev_err(adev->dev, "Unknown packet type %d !\n", type);
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h
index 63fadda7a673..654e43e83e2c 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.h
@@ -45,9 +45,6 @@
 
 #define JRBC_DEC_EXTERNAL_REG_WRITE_ADDR				0x18000
 
-#define JPEG_REG_RANGE_START						0x4000
-#define JPEG_REG_RANGE_END						0x41c2
-
 void jpeg_v2_0_dec_ring_insert_start(struct amdgpu_ring *ring);
 void jpeg_v2_0_dec_ring_insert_end(struct amdgpu_ring *ring);
 void jpeg_v2_0_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
@@ -60,9 +57,6 @@ void jpeg_v2_0_dec_ring_emit_vm_flush(struct amdgpu_ring *ring,
 				unsigned vmid, uint64_t pd_addr);
 void jpeg_v2_0_dec_ring_emit_wreg(struct amdgpu_ring *ring, uint32_t reg, uint32_t val);
 void jpeg_v2_0_dec_ring_nop(struct amdgpu_ring *ring, uint32_t count);
-int jpeg_v2_dec_ring_parse_cs(struct amdgpu_cs_parser *parser,
-			      struct amdgpu_job *job,
-			      struct amdgpu_ib *ib);
 
 extern const struct amdgpu_ip_block_version jpeg_v2_0_ip_block;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
index 3e2c389242db..20983f126b49 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c
@@ -696,7 +696,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_5_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v2_5_dec_ring_get_rptr,
 	.get_wptr = jpeg_v2_5_dec_ring_get_wptr,
 	.set_wptr = jpeg_v2_5_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
@@ -727,7 +727,7 @@ static const struct amdgpu_ring_funcs jpeg_v2_6_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v2_5_dec_ring_get_rptr,
 	.get_wptr = jpeg_v2_5_dec_ring_get_wptr,
 	.set_wptr = jpeg_v2_5_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
index a44eb2667664..d1a011c40ba2 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c
@@ -597,7 +597,7 @@ static const struct amdgpu_ring_funcs jpeg_v3_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v3_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v3_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v3_0_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
index da3ee69f1a3b..33db2c1ae6cc 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c
@@ -762,7 +762,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
index a78144773fab..aae7328973d1 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c
@@ -1177,7 +1177,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_3_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_3_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_3_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
index 5d86e1d846eb..54fd9c800c40 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c
@@ -807,7 +807,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_5_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v4_0_5_dec_ring_get_rptr,
 	.get_wptr = jpeg_v4_0_5_dec_ring_get_wptr,
 	.set_wptr = jpeg_v4_0_5_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
index 34c70270ea1d..46bf15dce2bd 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c
@@ -683,7 +683,7 @@ static const struct amdgpu_ring_funcs jpeg_v5_0_0_dec_ring_vm_funcs = {
 	.get_rptr = jpeg_v5_0_0_dec_ring_get_rptr,
 	.get_wptr = jpeg_v5_0_0_dec_ring_get_wptr,
 	.set_wptr = jpeg_v5_0_0_dec_ring_set_wptr,
-	.parse_cs = jpeg_v2_dec_ring_parse_cs,
+	.parse_cs = amdgpu_jpeg_dec_parse_cs,
 	.emit_frame_size =
 		SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 +
 		SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 +

From d426a5b6da2bba04566acfad94b6921141a2bf85 Mon Sep 17 00:00:00 2001
From: Prike Liang <Prike.Liang@amd.com>
Date: Tue, 17 Jun 2025 15:37:31 +0800
Subject: [PATCH 1164/1455] drm/amdgpu: clean up the amdgpu_userq_active()

This is no invocation for amdgpu_userq_active().

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 16 ----------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h |  2 --
 2 files changed, 18 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index f8648e169f8b..02ab788995b0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -112,22 +112,6 @@ amdgpu_userq_cleanup(struct amdgpu_userq_mgr *uq_mgr,
 	kfree(queue);
 }
 
-int
-amdgpu_userq_active(struct amdgpu_userq_mgr *uq_mgr)
-{
-	struct amdgpu_usermode_queue *queue;
-	int queue_id;
-	int ret = 0;
-
-	mutex_lock(&uq_mgr->userq_mutex);
-	/* Resume all the queues for this process */
-	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id)
-		ret += queue->state == AMDGPU_USERQ_STATE_MAPPED;
-
-	mutex_unlock(&uq_mgr->userq_mutex);
-	return ret;
-}
-
 static struct amdgpu_usermode_queue *
 amdgpu_userq_find(struct amdgpu_userq_mgr *uq_mgr, int qid)
 {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
index 1bd84f4cce78..da7e75cda4f8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
@@ -120,8 +120,6 @@ void amdgpu_userq_destroy_object(struct amdgpu_userq_mgr *uq_mgr,
 void amdgpu_userq_evict(struct amdgpu_userq_mgr *uq_mgr,
 			struct amdgpu_eviction_fence *ev_fence);
 
-int amdgpu_userq_active(struct amdgpu_userq_mgr *uq_mgr);
-
 void amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *userq_mgr,
 				  struct amdgpu_eviction_fence_mgr *evf_mgr);
 

From dc8f9f0f45166a6b37864e7a031c726981d6e5fc Mon Sep 17 00:00:00 2001
From: David Rosca <david.rosca@amd.com>
Date: Mon, 18 Aug 2025 09:06:58 +0200
Subject: [PATCH 1165/1455] drm/amdgpu/vcn4: Fix IB parsing with multiple
 engine info packages

There can be multiple engine info packages in one IB and the first one
may be common engine, not decode/encode.
We need to parse the entire IB instead of stopping after finding first
engine info.

Signed-off-by: David Rosca <david.rosca@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 48 +++++++++++----------------
 1 file changed, 19 insertions(+), 29 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index d0d27790b73b..8d20ae5bdb86 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1903,22 +1903,16 @@ out:
 
 #define RADEON_VCN_ENGINE_TYPE_ENCODE			(0x00000002)
 #define RADEON_VCN_ENGINE_TYPE_DECODE			(0x00000003)
-
 #define RADEON_VCN_ENGINE_INFO				(0x30000001)
-#define RADEON_VCN_ENGINE_INFO_MAX_OFFSET		16
-
 #define RENCODE_ENCODE_STANDARD_AV1			2
 #define RENCODE_IB_PARAM_SESSION_INIT			0x00000003
-#define RENCODE_IB_PARAM_SESSION_INIT_MAX_OFFSET	64
 
-/* return the offset in ib if id is found, -1 otherwise
- * to speed up the searching we only search upto max_offset
- */
-static int vcn_v4_0_enc_find_ib_param(struct amdgpu_ib *ib, uint32_t id, int max_offset)
+/* return the offset in ib if id is found, -1 otherwise */
+static int vcn_v4_0_enc_find_ib_param(struct amdgpu_ib *ib, uint32_t id, int start)
 {
 	int i;
 
-	for (i = 0; i < ib->length_dw && i < max_offset && ib->ptr[i] >= 8; i += ib->ptr[i]/4) {
+	for (i = start; i < ib->length_dw && ib->ptr[i] >= 8; i += ib->ptr[i] / 4) {
 		if (ib->ptr[i + 1] == id)
 			return i;
 	}
@@ -1933,33 +1927,29 @@ static int vcn_v4_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p,
 	struct amdgpu_vcn_decode_buffer *decode_buffer;
 	uint64_t addr;
 	uint32_t val;
-	int idx;
+	int idx = 0, sidx;
 
 	/* The first instance can decode anything */
 	if (!ring->me)
 		return 0;
 
-	/* RADEON_VCN_ENGINE_INFO is at the top of ib block */
-	idx = vcn_v4_0_enc_find_ib_param(ib, RADEON_VCN_ENGINE_INFO,
-			RADEON_VCN_ENGINE_INFO_MAX_OFFSET);
-	if (idx < 0) /* engine info is missing */
-		return 0;
+	while ((idx = vcn_v4_0_enc_find_ib_param(ib, RADEON_VCN_ENGINE_INFO, idx)) >= 0) {
+		val = amdgpu_ib_get_value(ib, idx + 2); /* RADEON_VCN_ENGINE_TYPE */
+		if (val == RADEON_VCN_ENGINE_TYPE_DECODE) {
+			decode_buffer = (struct amdgpu_vcn_decode_buffer *)&ib->ptr[idx + 6];
 
-	val = amdgpu_ib_get_value(ib, idx + 2); /* RADEON_VCN_ENGINE_TYPE */
-	if (val == RADEON_VCN_ENGINE_TYPE_DECODE) {
-		decode_buffer = (struct amdgpu_vcn_decode_buffer *)&ib->ptr[idx + 6];
+			if (!(decode_buffer->valid_buf_flag & 0x1))
+				return 0;
 
-		if (!(decode_buffer->valid_buf_flag  & 0x1))
-			return 0;
-
-		addr = ((u64)decode_buffer->msg_buffer_address_hi) << 32 |
-			decode_buffer->msg_buffer_address_lo;
-		return vcn_v4_0_dec_msg(p, job, addr);
-	} else if (val == RADEON_VCN_ENGINE_TYPE_ENCODE) {
-		idx = vcn_v4_0_enc_find_ib_param(ib, RENCODE_IB_PARAM_SESSION_INIT,
-			RENCODE_IB_PARAM_SESSION_INIT_MAX_OFFSET);
-		if (idx >= 0 && ib->ptr[idx + 2] == RENCODE_ENCODE_STANDARD_AV1)
-			return vcn_v4_0_limit_sched(p, job);
+			addr = ((u64)decode_buffer->msg_buffer_address_hi) << 32 |
+				decode_buffer->msg_buffer_address_lo;
+			return vcn_v4_0_dec_msg(p, job, addr);
+		} else if (val == RADEON_VCN_ENGINE_TYPE_ENCODE) {
+			sidx = vcn_v4_0_enc_find_ib_param(ib, RENCODE_IB_PARAM_SESSION_INIT, idx);
+			if (sidx >= 0 && ib->ptr[sidx + 2] == RENCODE_ENCODE_STANDARD_AV1)
+				return vcn_v4_0_limit_sched(p, job);
+		}
+		idx += ib->ptr[idx] / 4;
 	}
 	return 0;
 }

From 8908fdce0634a623404e9923ed2f536101a39db5 Mon Sep 17 00:00:00 2001
From: David Rosca <david.rosca@amd.com>
Date: Mon, 18 Aug 2025 09:18:37 +0200
Subject: [PATCH 1166/1455] drm/amdgpu/vcn: Allow limiting ctx to instance 0
 for AV1 at any time

There is no reason to require this to happen on first submitted IB only.
We need to wait for the queue to be idle, but it can be done at any
time (including when there are multiple video sessions active).

Signed-off-by: David Rosca <david.rosca@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 12 ++++++++----
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c | 12 ++++++++----
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 95173156f956..f3085137ba08 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -1886,15 +1886,19 @@ static int vcn_v3_0_limit_sched(struct amdgpu_cs_parser *p,
 				struct amdgpu_job *job)
 {
 	struct drm_gpu_scheduler **scheds;
-
-	/* The create msg must be in the first IB submitted */
-	if (atomic_read(&job->base.entity->fence_seq))
-		return -EINVAL;
+	struct dma_fence *fence;
 
 	/* if VCN0 is harvested, we can't support AV1 */
 	if (p->adev->vcn.harvest_config & AMDGPU_VCN_HARVEST_VCN0)
 		return -EINVAL;
 
+	/* wait for all jobs to finish before switching to instance 0 */
+	fence = amdgpu_ctx_get_fence(p->ctx, job->base.entity, ~0ull);
+	if (fence) {
+		dma_fence_wait(fence, false);
+		dma_fence_put(fence);
+	}
+
 	scheds = p->adev->gpu_sched[AMDGPU_HW_IP_VCN_DEC]
 		[AMDGPU_RING_PRIO_DEFAULT].sched;
 	drm_sched_entity_modify_sched(job->base.entity, scheds, 1);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 8d20ae5bdb86..bc9dfe5ffea7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1804,15 +1804,19 @@ static int vcn_v4_0_limit_sched(struct amdgpu_cs_parser *p,
 				struct amdgpu_job *job)
 {
 	struct drm_gpu_scheduler **scheds;
-
-	/* The create msg must be in the first IB submitted */
-	if (atomic_read(&job->base.entity->fence_seq))
-		return -EINVAL;
+	struct dma_fence *fence;
 
 	/* if VCN0 is harvested, we can't support AV1 */
 	if (p->adev->vcn.harvest_config & AMDGPU_VCN_HARVEST_VCN0)
 		return -EINVAL;
 
+	/* wait for all jobs to finish before switching to instance 0 */
+	fence = amdgpu_ctx_get_fence(p->ctx, job->base.entity, ~0ull);
+	if (fence) {
+		dma_fence_wait(fence, false);
+		dma_fence_put(fence);
+	}
+
 	scheds = p->adev->gpu_sched[AMDGPU_HW_IP_VCN_ENC]
 		[AMDGPU_RING_PRIO_0].sched;
 	drm_sched_entity_modify_sched(job->base.entity, scheds, 1);

From 357fe94b66f851c8824415d0168a8a72c4778698 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 20 Aug 2025 12:48:50 +0530
Subject: [PATCH 1167/1455] drm/amdgpu/ttm: Add New AMDGPU_PL_MMIO_REMAP
 Placement
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce a kernel-internal TTM placement type AMDGPU_PL_MMIO_REMAP
for the HDP flush MMIO remap page

Plumbing added:
- amdgpu_res_cursor.{first,next}: treat MMIO_REMAP like DOORBELL
- amdgpu_ttm_io_mem_reserve(): return BAR bus address + offset
  for MMIO_REMAP, mark as uncached I/O
- amdgpu_ttm_io_mem_pfn(): PFN from register BAR
- amdgpu_res_cpu_visible(): visible to CPU
- amdgpu_evict_flags()/amdgpu_bo_move(): non-migratable
- amdgpu_ttm_tt_pde_flags(): map as SYSTEM
- amdgpu_bo_mem_stats_placement(): report AMDGPU_PL_MMIO_REMAP
- amdgpu_fdinfo: print “mmioremap” bucket label

Cc: Alex Deucher <alexander.deucher@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 2309df3f68a9..bb17987f0447 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -34,7 +34,8 @@
 #define AMDGPU_PL_OA		(TTM_PL_PRIV + 2)
 #define AMDGPU_PL_PREEMPT	(TTM_PL_PRIV + 3)
 #define AMDGPU_PL_DOORBELL	(TTM_PL_PRIV + 4)
-#define __AMDGPU_PL_NUM	(TTM_PL_PRIV + 5)
+#define AMDGPU_PL_MMIO_REMAP	(TTM_PL_PRIV + 5)
+#define __AMDGPU_PL_NUM	(TTM_PL_PRIV + 6)
 
 #define AMDGPU_GTT_MAX_TRANSFER_SIZE	512
 #define AMDGPU_GTT_NUM_TRANSFER_WINDOWS	2

From 11aaec3566480e6cde0edb8e663c4b5ee32318f5 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 20 Aug 2025 16:17:52 +0530
Subject: [PATCH 1168/1455] drm/amdgpu: Wire up MMIO_REMAP placement and
 User-visible strings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Wire up the conversions and strings for the new MMIO_REMAP placement:

* amdgpu_mem_type_to_domain() maps AMDGPU_PL_MMIO_REMAP -> domain
* amdgpu_bo_placement_from_domain() accepts the new domain
* amdgpu_bo_mem_stats_placement() and amdgpu_bo_print_info() report it
* res cursor supports the new placement
* fdinfo prints "mmioremap" for the new placement

Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c     |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c     | 13 +++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h     |  2 ++
 drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h |  2 ++
 4 files changed, 18 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c
index 91d638098889..b349bb3676d5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c
@@ -70,6 +70,7 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 		[AMDGPU_PL_GWS] = "gws",
 		[AMDGPU_PL_OA] = "oa",
 		[AMDGPU_PL_DOORBELL] = "doorbell",
+		[AMDGPU_PL_MMIO_REMAP] = "mmioremap",
 	};
 	unsigned int hw_ip, i;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 122a88294883..fe486988a738 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -153,6 +153,14 @@ void amdgpu_bo_placement_from_domain(struct amdgpu_bo *abo, u32 domain)
 		c++;
 	}
 
+	if (domain & AMDGPU_GEM_DOMAIN_MMIO_REMAP) {
+		places[c].fpfn = 0;
+		places[c].lpfn = 0;
+		places[c].mem_type = AMDGPU_PL_MMIO_REMAP;
+		places[c].flags = 0;
+		c++;
+	}
+
 	if (domain & AMDGPU_GEM_DOMAIN_GTT) {
 		places[c].fpfn = 0;
 		places[c].lpfn = 0;
@@ -1545,6 +1553,8 @@ uint32_t amdgpu_bo_mem_stats_placement(struct amdgpu_bo *bo)
 		return AMDGPU_PL_OA;
 	case AMDGPU_GEM_DOMAIN_DOORBELL:
 		return AMDGPU_PL_DOORBELL;
+	case AMDGPU_GEM_DOMAIN_MMIO_REMAP:
+		return AMDGPU_PL_MMIO_REMAP;
 	default:
 		return TTM_PL_SYSTEM;
 	}
@@ -1628,6 +1638,9 @@ u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m)
 			case AMDGPU_PL_DOORBELL:
 				placement = "DOORBELL";
 				break;
+			case AMDGPU_PL_MMIO_REMAP:
+				placement = "MMIO REMAP";
+				break;
 			case TTM_PL_SYSTEM:
 			default:
 				placement = "CPU";
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
index 87523fcd4386..656b8a931dae 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h
@@ -167,6 +167,8 @@ static inline unsigned amdgpu_mem_type_to_domain(u32 mem_type)
 		return AMDGPU_GEM_DOMAIN_OA;
 	case AMDGPU_PL_DOORBELL:
 		return AMDGPU_GEM_DOMAIN_DOORBELL;
+	case AMDGPU_PL_MMIO_REMAP:
+		return AMDGPU_GEM_DOMAIN_MMIO_REMAP;
 	default:
 		break;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h
index 50fcd86e1033..be2e56ce1355 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_res_cursor.h
@@ -91,6 +91,7 @@ static inline void amdgpu_res_first(struct ttm_resource *res,
 		break;
 	case TTM_PL_TT:
 	case AMDGPU_PL_DOORBELL:
+	case AMDGPU_PL_MMIO_REMAP:
 		node = to_ttm_range_mgr_node(res)->mm_nodes;
 		while (start >= node->size << PAGE_SHIFT)
 			start -= node++->size << PAGE_SHIFT;
@@ -153,6 +154,7 @@ static inline void amdgpu_res_next(struct amdgpu_res_cursor *cur, uint64_t size)
 		break;
 	case TTM_PL_TT:
 	case AMDGPU_PL_DOORBELL:
+	case AMDGPU_PL_MMIO_REMAP:
 		node = cur->node;
 
 		cur->node = ++node;

From 8b38bf38830a84de70d53ad388268cbbe677118d Mon Sep 17 00:00:00 2001
From: Prike Liang <Prike.Liang@amd.com>
Date: Fri, 16 May 2025 16:20:54 +0800
Subject: [PATCH 1169/1455] drm/amdgpu: validate userq hw unmap status for
 destroying userq

Before destroying the userq buffer object, it requires validating
the userq HW unmap status and ensuring the userq is unmapped from
hardware. If the user HW unmap failed, then it needs to reset the
queue for reusing.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 02ab788995b0..09cea573e5ad 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -307,6 +307,11 @@ amdgpu_userq_destroy(struct drm_file *filp, int queue_id)
 	debugfs_remove_recursive(queue->debugfs_queue);
 #endif
 	r = amdgpu_userq_unmap_helper(uq_mgr, queue);
+	/*TODO: It requires a reset for userq hw unmap error*/
+	if (unlikely(r != AMDGPU_USERQ_STATE_UNMAPPED)) {
+		drm_warn(adev_to_drm(uq_mgr->adev), "trying to destroy a HW mapping userq\n");
+		queue->state = AMDGPU_USERQ_STATE_HUNG;
+	}
 	amdgpu_userq_cleanup(uq_mgr, queue, queue_id);
 	mutex_unlock(&uq_mgr->userq_mutex);
 

From 3d9752f4f9319110896ee497e23b18897189eb14 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Sun, 24 Aug 2025 18:46:40 -0600
Subject: [PATCH 1170/1455] drm/amdgpu/vcn: Document IRQ per-instance irq
 behavior for VCN 4.0.3

When examining the VCN function init, it is common to find a loop that
initializes VCN rings, which uses one IRQ per instance. However, VCN
4.0.3 deviates from this pattern, as it includes a distinct field to
differentiate instances, which results in a slightly different ring
init. This commit makes this difference explicit by using a fixed index
when initializing the ring buffer and also adds a comment.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 7b93a275ec4f..bae1ad4484ed 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -212,7 +212,11 @@ static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block)
 
 		ring->vm_hub = AMDGPU_MMHUB0(adev->vcn.inst[i].aid_id);
 		sprintf(ring->name, "vcn_unified_%d", adev->vcn.inst[i].aid_id);
-		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst->irq, 0,
+
+		/* There are no per-instance irq source IDs on 4.0.3, the IH
+		 * packets use a separate field to differentiate instances.
+		 */
+		r = amdgpu_ring_init(adev, ring, 512, &adev->vcn.inst[0].irq, 0,
 				     AMDGPU_RING_PRIO_DEFAULT,
 				     &adev->vcn.inst[i].sched_score);
 		if (r)

From 0855c764f7e6d58e1f5a502fc26830033ce94dbb Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Sun, 24 Aug 2025 18:46:41 -0600
Subject: [PATCH 1171/1455] drm/amdgpu/vcn: Change amdgpu_vcn_sw_fini return to
 void

The function amdgpu_vcn_sw_fini() returns an integer, but this number is
always 0. This commit changes the amdgpu_vcn_sw_fini() return to void,
and eliminates all checks to this return across different VCNs.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 6 ++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c   | 4 +---
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c   | 4 +---
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c   | 7 ++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 7 ++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 4 +---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 7 ++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 9 +++------
 11 files changed, 19 insertions(+), 39 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index ad415203d245..c2f69369bda6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -257,12 +257,12 @@ int amdgpu_vcn_sw_init(struct amdgpu_device *adev, int i)
 	return 0;
 }
 
-int amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i)
+void amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i)
 {
 	int j;
 
 	if (adev->vcn.harvest_config & (1 << i))
-		return 0;
+		return;
 
 	amdgpu_bo_free_kernel(
 		&adev->vcn.inst[i].dpg_sram_bo,
@@ -292,8 +292,6 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i)
 
 	mutex_destroy(&adev->vcn.inst[i].vcn_pg_lock);
 	mutex_destroy(&adev->vcn.inst[i].vcn1_jpeg1_workaround);
-
-	return 0;
 }
 
 bool amdgpu_vcn_is_disabled_vcn(struct amdgpu_device *adev, enum vcn_ring_type type, uint32_t vcn_instance)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 6d9acd36041d..bebfc2b34afe 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -516,7 +516,7 @@ enum vcn_ring_type {
 
 int amdgpu_vcn_early_init(struct amdgpu_device *adev, int i);
 int amdgpu_vcn_sw_init(struct amdgpu_device *adev, int i);
-int amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i);
+void amdgpu_vcn_sw_fini(struct amdgpu_device *adev, int i);
 int amdgpu_vcn_suspend(struct amdgpu_device *adev, int i);
 int amdgpu_vcn_resume(struct amdgpu_device *adev, int i);
 void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index 1e89ba153d9d..3c1c844c5fde 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -230,11 +230,11 @@ static int vcn_v1_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	jpeg_v1_0_sw_fini(ip_block);
 
-	r = amdgpu_vcn_sw_fini(adev, 0);
+	amdgpu_vcn_sw_fini(adev, 0);
 
 	kfree(adev->vcn.ip_dump);
 
-	return r;
+	return 0;
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index b115137ab2d6..c4e2659f8933 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -267,9 +267,9 @@ static int vcn_v2_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	amdgpu_vcn_sysfs_reset_mask_fini(adev);
 
-	r = amdgpu_vcn_sw_fini(adev, 0);
+	amdgpu_vcn_sw_fini(adev, 0);
 
-	return r;
+	return 0;
 }
 
 /**
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index 904b94bc8693..0505f0955324 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -442,9 +442,7 @@ static int vcn_v2_5_sw_fini(struct amdgpu_ip_block *ip_block)
 		r = amdgpu_vcn_suspend(adev, i);
 		if (r)
 			return r;
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
+		amdgpu_vcn_sw_fini(adev, i);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index f3085137ba08..77a8373444df 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -349,9 +349,7 @@ static int vcn_v3_0_sw_fini(struct amdgpu_ip_block *ip_block)
 		if (r)
 			return r;
 
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
+		amdgpu_vcn_sw_fini(adev, i);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index bc9dfe5ffea7..870c563415af 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -302,11 +302,8 @@ static int vcn_v4_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	amdgpu_vcn_sysfs_reset_mask_fini(adev);
 
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
-	}
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
+		amdgpu_vcn_sw_fini(adev, i);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index bae1ad4484ed..5fbd755fc24c 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -283,11 +283,8 @@ static int vcn_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	amdgpu_vcn_sysfs_reset_mask_fini(adev);
 
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
-	}
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
+		amdgpu_vcn_sw_fini(adev, i);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index 6dbf33b26ee2..732d4a78ab52 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -270,9 +270,7 @@ static int vcn_v4_0_5_sw_fini(struct amdgpu_ip_block *ip_block)
 		if (r)
 			return r;
 
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
+		amdgpu_vcn_sw_fini(adev, i);
 	}
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index 536f06b81706..680df5d48c7a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -232,11 +232,8 @@ static int vcn_v5_0_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	amdgpu_vcn_sysfs_reset_mask_fini(adev);
 
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
-	}
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
+		amdgpu_vcn_sw_fini(adev, i);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 4b01e35ad7ef..09e2e28041c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -245,14 +245,11 @@ static int vcn_v5_0_1_sw_fini(struct amdgpu_ip_block *ip_block)
 			return r;
 	}
 
-	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		r = amdgpu_vcn_sw_fini(adev, i);
-		if (r)
-			return r;
-	}
-
 	amdgpu_vcn_sysfs_reset_mask_fini(adev);
 
+	for (i = 0; i < adev->vcn.num_vcn_inst; i++)
+		amdgpu_vcn_sw_fini(adev, i);
+
 	return 0;
 }
 

From 0131514f97890c4699a4d7ca14b720dafefdbc71 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:34 +0200
Subject: [PATCH 1172/1455] drm/xe: Pass down drm_exec context to validation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We want all validation (potential backing store allocation) to be part
of a drm_exec transaction. Therefore add a drm_exec pointer argument
to xe_bo_validate() and ___xe_bo_create_locked(). Upcoming patches
will deal with making all (or nearly all) calls to these functions
part of a drm_exec transaction. In the meantime, define special values
of the drm_exec pointer:

XE_VALIDATION_UNIMPLEMENTED: Implementation of the drm_exec transaction
has not been done yet.
XE_VALIDATION_UNSUPPORTED: Some Middle-layers (dma-buf) doesn't allow
the drm_exec context to be passed down to map_attachment where
validation takes place.
XE_VALIDATION_OPT_OUT: May be used only for kunit tests where exhaustive
eviction isn't crucial and the ROI of converting those is very
small.

For XE_VALIDATION_UNIMPLEMENTED and XE_VALIDATION_OPT_OUT there is also
a lockdep check that a drm_exec transaction can indeed start at the
location where the macro is expanded. This is to encourage
developers to take this into consideration early in the code
development process.

v2:
- Fix xe_vm_set_validation_exec() imbalance. Add an assert that
  hopefully catches future instances of this (Matt Brost)
v3:
- Extend to psmi_alloc_object

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com> #v3
Link: https://lore.kernel.org/r/20250908101246.65025-2-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/Makefile                   |   1 +
 .../compat-i915-headers/gem/i915_gem_stolen.h |   6 +-
 drivers/gpu/drm/xe/display/xe_fb_pin.c        |   5 +-
 drivers/gpu/drm/xe/tests/xe_bo.c              |  20 +--
 drivers/gpu/drm/xe/tests/xe_dma_buf.c         |  12 +-
 drivers/gpu/drm/xe/tests/xe_migrate.c         |  45 +++---
 drivers/gpu/drm/xe/xe_bo.c                    | 131 +++++++++++++++---
 drivers/gpu/drm/xe/xe_bo.h                    |  20 +--
 drivers/gpu/drm/xe/xe_dma_buf.c               |  19 ++-
 drivers/gpu/drm/xe/xe_exec.c                  |   6 +-
 drivers/gpu/drm/xe/xe_ggtt.c                  |  15 +-
 drivers/gpu/drm/xe/xe_ggtt.h                  |   5 +-
 drivers/gpu/drm/xe/xe_gt_pagefault.c          |   8 +-
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c    |   6 +-
 drivers/gpu/drm/xe/xe_psmi.c                  |   6 +-
 drivers/gpu/drm/xe/xe_svm.c                   |   4 +-
 drivers/gpu/drm/xe/xe_validation.c            |  49 +++++++
 drivers/gpu/drm/xe/xe_validation.h            |  69 +++++++++
 drivers/gpu/drm/xe/xe_vm.c                    |  26 +++-
 drivers/gpu/drm/xe/xe_vm.h                    |  34 ++++-
 drivers/gpu/drm/xe/xe_vm_types.h              |  32 +++--
 21 files changed, 408 insertions(+), 111 deletions(-)
 create mode 100644 drivers/gpu/drm/xe/xe_validation.c
 create mode 100644 drivers/gpu/drm/xe/xe_validation.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index d6bd139c5839..041195ca90e6 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -130,6 +130,7 @@ xe-y += xe_bb.o \
 	xe_tuning.o \
 	xe_uc.o \
 	xe_uc_fw.o \
+	xe_validation.o \
 	xe_vm.o \
 	xe_vm_madvise.o \
 	xe_vram.o \
diff --git a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
index 41d39d67817a..1ce1e9da975b 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
@@ -8,6 +8,7 @@
 
 #include "xe_ttm_stolen_mgr.h"
 #include "xe_res_cursor.h"
+#include "xe_validation.h"
 
 struct xe_bo;
 
@@ -20,6 +21,7 @@ static inline int i915_gem_stolen_insert_node_in_range(struct xe_device *xe,
 						       u32 size, u32 align,
 						       u32 start, u32 end)
 {
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *bo;
 	int err;
 	u32 flags = XE_BO_FLAG_PINNED | XE_BO_FLAG_STOLEN;
@@ -34,13 +36,13 @@ static inline int i915_gem_stolen_insert_node_in_range(struct xe_device *xe,
 
 	bo = xe_bo_create_locked_range(xe, xe_device_get_root_tile(xe),
 				       NULL, size, start, end,
-				       ttm_bo_type_kernel, flags, 0);
+				       ttm_bo_type_kernel, flags, 0, exec);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		bo = NULL;
 		return err;
 	}
-	err = xe_bo_pin(bo);
+	err = xe_bo_pin(bo, exec);
 	xe_bo_unlock_vm_held(bo);
 
 	if (err) {
diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index f2cfba674899..1c5b4e1f1fd8 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -281,6 +281,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 	struct i915_vma *vma = kzalloc(sizeof(*vma), GFP_KERNEL);
 	struct drm_gem_object *obj = intel_fb_bo(&fb->base);
 	struct xe_bo *bo = gem_to_xe_bo(obj);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	int ret;
 
 	if (!vma)
@@ -313,9 +314,9 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 		goto err;
 
 	if (IS_DGFX(xe))
-		ret = xe_bo_migrate(bo, XE_PL_VRAM0);
+		ret = xe_bo_migrate(bo, XE_PL_VRAM0, exec);
 	else
-		ret = xe_bo_validate(bo, NULL, true);
+		ret = xe_bo_validate(bo, NULL, true, exec);
 	if (!ret)
 		ttm_bo_pin(&bo->ttm);
 	ttm_bo_unreserve(&bo->ttm);
diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c
index 7b40cc8be1c9..17366b60ee31 100644
--- a/drivers/gpu/drm/xe/tests/xe_bo.c
+++ b/drivers/gpu/drm/xe/tests/xe_bo.c
@@ -23,7 +23,7 @@
 
 static int ccs_test_migrate(struct xe_tile *tile, struct xe_bo *bo,
 			    bool clear, u64 get_val, u64 assign_val,
-			    struct kunit *test)
+			    struct kunit *test, struct drm_exec *exec)
 {
 	struct dma_fence *fence;
 	struct ttm_tt *ttm;
@@ -35,7 +35,7 @@ static int ccs_test_migrate(struct xe_tile *tile, struct xe_bo *bo,
 	u32 offset;
 
 	/* Move bo to VRAM if not already there. */
-	ret = xe_bo_validate(bo, NULL, false);
+	ret = xe_bo_validate(bo, NULL, false, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to validate bo.\n");
 		return ret;
@@ -60,7 +60,7 @@ static int ccs_test_migrate(struct xe_tile *tile, struct xe_bo *bo,
 	}
 
 	/* Evict to system. CCS data should be copied. */
-	ret = xe_bo_evict(bo);
+	ret = xe_bo_evict(bo, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to evict bo.\n");
 		return ret;
@@ -132,6 +132,7 @@ static void ccs_test_run_tile(struct xe_device *xe, struct xe_tile *tile,
 
 	/* TODO: Sanity check */
 	unsigned int bo_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile);
+	struct drm_exec *exec = XE_VALIDATION_OPT_OUT;
 
 	if (IS_DGFX(xe))
 		kunit_info(test, "Testing vram id %u\n", tile->id);
@@ -149,18 +150,18 @@ static void ccs_test_run_tile(struct xe_device *xe, struct xe_tile *tile,
 
 	kunit_info(test, "Verifying that CCS data is cleared on creation.\n");
 	ret = ccs_test_migrate(tile, bo, false, 0ULL, 0xdeadbeefdeadbeefULL,
-			       test);
+			       test, exec);
 	if (ret)
 		goto out_unlock;
 
 	kunit_info(test, "Verifying that CCS data survives migration.\n");
 	ret = ccs_test_migrate(tile, bo, false, 0xdeadbeefdeadbeefULL,
-			       0xdeadbeefdeadbeefULL, test);
+			       0xdeadbeefdeadbeefULL, test, exec);
 	if (ret)
 		goto out_unlock;
 
 	kunit_info(test, "Verifying that CCS data can be properly cleared.\n");
-	ret = ccs_test_migrate(tile, bo, true, 0ULL, 0ULL, test);
+	ret = ccs_test_migrate(tile, bo, true, 0ULL, 0ULL, test, exec);
 
 out_unlock:
 	xe_bo_unlock(bo);
@@ -210,6 +211,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 	struct xe_bo *bo, *external;
 	unsigned int bo_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile);
 	struct xe_vm *vm = xe_migrate_get_vm(xe_device_get_root_tile(xe)->migrate);
+	struct drm_exec *exec = XE_VALIDATION_OPT_OUT;
 	struct xe_gt *__gt;
 	int err, i, id;
 
@@ -236,7 +238,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 		}
 
 		xe_bo_lock(external, false);
-		err = xe_bo_pin_external(external, false);
+		err = xe_bo_pin_external(external, false, exec);
 		xe_bo_unlock(external);
 		if (err) {
 			KUNIT_FAIL(test, "external bo pin err=%pe\n",
@@ -294,7 +296,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 		if (i) {
 			down_read(&vm->lock);
 			xe_vm_lock(vm, false);
-			err = xe_bo_validate(bo, bo->vm, false);
+			err = xe_bo_validate(bo, bo->vm, false, exec);
 			xe_vm_unlock(vm);
 			up_read(&vm->lock);
 			if (err) {
@@ -303,7 +305,7 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 				goto cleanup_all;
 			}
 			xe_bo_lock(external, false);
-			err = xe_bo_validate(external, NULL, false);
+			err = xe_bo_validate(external, NULL, false, exec);
 			xe_bo_unlock(external);
 			if (err) {
 				KUNIT_FAIL(test, "external bo valid err=%pe\n",
diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
index 5baeab6b6fb7..7117486f0432 100644
--- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
@@ -27,7 +27,8 @@ static bool is_dynamic(struct dma_buf_test_params *params)
 }
 
 static void check_residency(struct kunit *test, struct xe_bo *exported,
-			    struct xe_bo *imported, struct dma_buf *dmabuf)
+			    struct xe_bo *imported, struct dma_buf *dmabuf,
+			    struct drm_exec *exec)
 {
 	struct dma_buf_test_params *params = to_dma_buf_test_params(test->priv);
 	u32 mem_type;
@@ -62,7 +63,7 @@ static void check_residency(struct kunit *test, struct xe_bo *exported,
 	 * importer is on a different device. If they're on the same device,
 	 * the exporter and the importer should be the same bo.
 	 */
-	ret = xe_bo_evict(exported);
+	ret = xe_bo_evict(exported, exec);
 	if (ret) {
 		if (ret != -EINTR && ret != -ERESTARTSYS)
 			KUNIT_FAIL(test, "Evicting exporter failed with err=%d.\n",
@@ -77,7 +78,7 @@ static void check_residency(struct kunit *test, struct xe_bo *exported,
 	}
 
 	/* Re-validate the importer. This should move also exporter in. */
-	ret = xe_bo_validate(imported, NULL, false);
+	ret = xe_bo_validate(imported, NULL, false, exec);
 	if (ret) {
 		if (ret != -EINTR && ret != -ERESTARTSYS)
 			KUNIT_FAIL(test, "Validating importer failed with err=%d.\n",
@@ -142,11 +143,12 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe)
 			KUNIT_FAIL(test,
 				   "xe_gem_prime_import() succeeded when it shouldn't have\n");
 		} else {
+			struct drm_exec *exec = XE_VALIDATION_OPT_OUT;
 			int err;
 
 			/* Is everything where we expect it to be? */
 			xe_bo_lock(import_bo, false);
-			err = xe_bo_validate(import_bo, NULL, false);
+			err = xe_bo_validate(import_bo, NULL, false, exec);
 
 			/* Pinning in VRAM is not allowed. */
 			if (!is_dynamic(params) &&
@@ -159,7 +161,7 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe)
 						  err == -ERESTARTSYS);
 
 			if (!err)
-				check_residency(test, bo, import_bo, dmabuf);
+				check_residency(test, bo, import_bo, dmabuf, exec);
 			xe_bo_unlock(import_bo);
 		}
 		drm_gem_object_put(import);
diff --git a/drivers/gpu/drm/xe/tests/xe_migrate.c b/drivers/gpu/drm/xe/tests/xe_migrate.c
index edd1e701aa1c..dfb445d09759 100644
--- a/drivers/gpu/drm/xe/tests/xe_migrate.c
+++ b/drivers/gpu/drm/xe/tests/xe_migrate.c
@@ -70,7 +70,7 @@ static int run_sanity_job(struct xe_migrate *m, struct xe_device *xe,
 		} } while (0)
 
 static void test_copy(struct xe_migrate *m, struct xe_bo *bo,
-		      struct kunit *test, u32 region)
+		      struct kunit *test, u32 region, struct drm_exec *exec)
 {
 	struct xe_device *xe = tile_to_xe(m->tile);
 	u64 retval, expected = 0;
@@ -84,14 +84,15 @@ static void test_copy(struct xe_migrate *m, struct xe_bo *bo,
 						   ttm_bo_type_kernel,
 						   region |
 						   XE_BO_FLAG_NEEDS_CPU_ACCESS |
-						   XE_BO_FLAG_PINNED);
+						   XE_BO_FLAG_PINNED,
+						   exec);
 	if (IS_ERR(remote)) {
 		KUNIT_FAIL(test, "Failed to allocate remote bo for %s: %pe\n",
 			   str, remote);
 		return;
 	}
 
-	err = xe_bo_validate(remote, NULL, false);
+	err = xe_bo_validate(remote, NULL, false, exec);
 	if (err) {
 		KUNIT_FAIL(test, "Failed to validate system bo for %s: %i\n",
 			   str, err);
@@ -161,13 +162,13 @@ out_unlock:
 }
 
 static void test_copy_sysmem(struct xe_migrate *m, struct xe_bo *bo,
-			     struct kunit *test)
+			     struct drm_exec *exec, struct kunit *test)
 {
-	test_copy(m, bo, test, XE_BO_FLAG_SYSTEM);
+	test_copy(m, bo, test, XE_BO_FLAG_SYSTEM, exec);
 }
 
 static void test_copy_vram(struct xe_migrate *m, struct xe_bo *bo,
-			   struct kunit *test)
+			   struct drm_exec *exec, struct kunit *test)
 {
 	u32 region;
 
@@ -178,10 +179,11 @@ static void test_copy_vram(struct xe_migrate *m, struct xe_bo *bo,
 		region = XE_BO_FLAG_VRAM1;
 	else
 		region = XE_BO_FLAG_VRAM0;
-	test_copy(m, bo, test, region);
+	test_copy(m, bo, test, region, exec);
 }
 
-static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test)
+static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test,
+				   struct drm_exec *exec)
 {
 	struct xe_tile *tile = m->tile;
 	struct xe_device *xe = tile_to_xe(tile);
@@ -290,10 +292,10 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test)
 	check(retval, expected, "Command clear small last value", test);
 
 	kunit_info(test, "Copying small buffer object to system\n");
-	test_copy_sysmem(m, tiny, test);
+	test_copy_sysmem(m, tiny, exec, test);
 	if (xe->info.tile_count > 1) {
 		kunit_info(test, "Copying small buffer object to other vram\n");
-		test_copy_vram(m, tiny, test);
+		test_copy_vram(m, tiny, exec, test);
 	}
 
 	/* Clear a big bo */
@@ -312,10 +314,10 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test)
 	check(retval, expected, "Command clear big last value", test);
 
 	kunit_info(test, "Copying big buffer object to system\n");
-	test_copy_sysmem(m, big, test);
+	test_copy_sysmem(m, big, exec, test);
 	if (xe->info.tile_count > 1) {
 		kunit_info(test, "Copying big buffer object to other vram\n");
-		test_copy_vram(m, big, test);
+		test_copy_vram(m, big, exec, test);
 	}
 
 out:
@@ -343,10 +345,11 @@ static int migrate_test_run_device(struct xe_device *xe)
 
 	for_each_tile(tile, xe, id) {
 		struct xe_migrate *m = tile->migrate;
+		struct drm_exec *exec = XE_VALIDATION_OPT_OUT;
 
 		kunit_info(test, "Testing tile id %d.\n", id);
 		xe_vm_lock(m->q->vm, false);
-		xe_migrate_sanity_test(m, test);
+		xe_migrate_sanity_test(m, test, exec);
 		xe_vm_unlock(m->q->vm);
 	}
 
@@ -490,7 +493,7 @@ err_sync:
 
 static void test_migrate(struct xe_device *xe, struct xe_tile *tile,
 			 struct xe_bo *sys_bo, struct xe_bo *vram_bo, struct xe_bo *ccs_bo,
-			 struct kunit *test)
+			 struct drm_exec *exec, struct kunit *test)
 {
 	struct dma_fence *fence;
 	u64 expected, retval;
@@ -509,7 +512,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile,
 	dma_fence_put(fence);
 
 	kunit_info(test, "Evict vram buffer object\n");
-	ret = xe_bo_evict(vram_bo);
+	ret = xe_bo_evict(vram_bo, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to evict bo.\n");
 		return;
@@ -538,7 +541,7 @@ static void test_migrate(struct xe_device *xe, struct xe_tile *tile,
 	dma_fence_put(fence);
 
 	kunit_info(test, "Restore vram buffer object\n");
-	ret = xe_bo_validate(vram_bo, NULL, false);
+	ret = xe_bo_validate(vram_bo, NULL, false, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to validate vram bo for: %li\n", ret);
 		return;
@@ -636,6 +639,7 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 {
 	struct xe_bo *sys_bo, *vram_bo = NULL, *ccs_bo = NULL;
 	unsigned int bo_flags = XE_BO_FLAG_VRAM_IF_DGFX(tile);
+	struct drm_exec *exec;
 	long ret;
 
 	sys_bo = xe_bo_create_user(xe, NULL, NULL, SZ_4M,
@@ -650,8 +654,9 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 		return;
 	}
 
+	exec = XE_VALIDATION_OPT_OUT;
 	xe_bo_lock(sys_bo, false);
-	ret = xe_bo_validate(sys_bo, NULL, false);
+	ret = xe_bo_validate(sys_bo, NULL, false, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to validate system bo for: %li\n", ret);
 		goto free_sysbo;
@@ -676,7 +681,7 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 	}
 
 	xe_bo_lock(ccs_bo, false);
-	ret = xe_bo_validate(ccs_bo, NULL, false);
+	ret = xe_bo_validate(ccs_bo, NULL, false, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to validate system bo for: %li\n", ret);
 		goto free_ccsbo;
@@ -700,7 +705,7 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 	}
 
 	xe_bo_lock(vram_bo, false);
-	ret = xe_bo_validate(vram_bo, NULL, false);
+	ret = xe_bo_validate(vram_bo, NULL, false, exec);
 	if (ret) {
 		KUNIT_FAIL(test, "Failed to validate vram bo for: %li\n", ret);
 		goto free_vrambo;
@@ -713,7 +718,7 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 	}
 
 	test_clear(xe, tile, sys_bo, vram_bo, test);
-	test_migrate(xe, tile, sys_bo, vram_bo, ccs_bo, test);
+	test_migrate(xe, tile, sys_bo, vram_bo, ccs_bo, exec, test);
 	xe_bo_unlock(vram_bo);
 
 	xe_bo_lock(vram_bo, false);
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 4efc1c8d15b8..36e804dcc38f 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1141,6 +1141,7 @@ out_unref:
 int xe_bo_notifier_prepare_pinned(struct xe_bo *bo)
 {
 	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *backup;
 	int ret = 0;
 
@@ -1165,7 +1166,7 @@ int xe_bo_notifier_prepare_pinned(struct xe_bo *bo)
 	backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
 					DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
 					XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-					XE_BO_FLAG_PINNED);
+					XE_BO_FLAG_PINNED, exec);
 	if (IS_ERR(backup)) {
 		ret = PTR_ERR(backup);
 		goto out_unlock_bo;
@@ -1216,6 +1217,7 @@ int xe_bo_notifier_unprepare_pinned(struct xe_bo *bo)
 int xe_bo_evict_pinned(struct xe_bo *bo)
 {
 	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *backup = bo->backup_obj;
 	bool backup_created = false;
 	bool unmap = false;
@@ -1244,7 +1246,7 @@ int xe_bo_evict_pinned(struct xe_bo *bo)
 						NULL, xe_bo_size(bo),
 						DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
 						XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-						XE_BO_FLAG_PINNED);
+						XE_BO_FLAG_PINNED, exec);
 		if (IS_ERR(backup)) {
 			ret = PTR_ERR(backup);
 			goto out_unlock_bo;
@@ -1721,12 +1723,14 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
 	struct xe_device *xe = to_xe_device(ddev);
 	struct xe_bo *bo = ttm_to_xe_bo(tbo);
 	bool needs_rpm = bo->flags & XE_BO_FLAG_VRAM_MASK;
+	struct drm_exec *exec;
 	vm_fault_t ret;
 	int idx, r = 0;
 
 	if (needs_rpm)
 		xe_pm_runtime_get(xe);
 
+	exec = XE_VALIDATION_UNIMPLEMENTED;
 	ret = ttm_bo_vm_reserve(tbo, vmf);
 	if (ret)
 		goto out;
@@ -1734,10 +1738,11 @@ static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
 	if (drm_dev_enter(ddev, &idx)) {
 		trace_xe_bo_cpu_fault(bo);
 
+		xe_validation_assert_exec(xe, exec, &tbo->base);
 		if (should_migrate_to_smem(bo)) {
 			xe_assert(xe, bo->flags & XE_BO_FLAG_SYSTEM);
 
-			r = xe_bo_migrate(bo, XE_PL_TT);
+			r = xe_bo_migrate(bo, XE_PL_TT, exec);
 			if (r == -EBUSY || r == -ERESTARTSYS || r == -EINTR)
 				ret = VM_FAULT_NOPAGE;
 			else if (r)
@@ -1864,11 +1869,32 @@ void xe_bo_free(struct xe_bo *bo)
 	kfree(bo);
 }
 
+/**
+ * ___xe_bo_create_locked() - Initialize or create an xe_bo.
+ * @xe: The xe device.
+ * @bo: An already allocated buffer object or NULL
+ * if the function should allocate a new one.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @resv: Pointer to a locked shared reservation object to use fo this bo,
+ * or NULL for the xe_bo to use its own.
+ * @bulk: The bulk move to use for LRU bumping, or NULL for external bos.
+ * @size: The storage size to use for the bo.
+ * @cpu_caching: The cpu caching used for system memory backing store.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
+ *
+ * Initialize or create an xe buffer object. On failure, any allocated buffer
+ * object passed in @bo will have been unreferenced.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ */
 struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo,
 				     struct xe_tile *tile, struct dma_resv *resv,
 				     struct ttm_lru_bulk_move *bulk, size_t size,
 				     u16 cpu_caching, enum ttm_bo_type type,
-				     u32 flags)
+				     u32 flags, struct drm_exec *exec)
 {
 	struct ttm_operation_ctx ctx = {
 		.interruptible = true,
@@ -1937,6 +1963,7 @@ struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo,
 		ctx.resv = resv;
 	}
 
+	xe_validation_assert_exec(xe, exec, &bo->ttm.base);
 	if (!(flags & XE_BO_FLAG_FIXED_PLACEMENT)) {
 		err = __xe_bo_placement_for_flags(xe, bo, bo->flags);
 		if (WARN_ON(err)) {
@@ -2038,7 +2065,7 @@ __xe_bo_create_locked(struct xe_device *xe,
 		      struct xe_tile *tile, struct xe_vm *vm,
 		      size_t size, u64 start, u64 end,
 		      u16 cpu_caching, enum ttm_bo_type type, u32 flags,
-		      u64 alignment)
+		      u64 alignment, struct drm_exec *exec)
 {
 	struct xe_bo *bo = NULL;
 	int err;
@@ -2063,7 +2090,7 @@ __xe_bo_create_locked(struct xe_device *xe,
 				    vm && !xe_vm_in_fault_mode(vm) &&
 				    flags & XE_BO_FLAG_USER ?
 				    &vm->lru_bulk_move : NULL, size,
-				    cpu_caching, type, flags);
+				    cpu_caching, type, flags, exec);
 	if (IS_ERR(bo))
 		return bo;
 
@@ -2097,9 +2124,10 @@ __xe_bo_create_locked(struct xe_device *xe,
 
 			if (flags & XE_BO_FLAG_FIXED_PLACEMENT) {
 				err = xe_ggtt_insert_bo_at(t->mem.ggtt, bo,
-							   start + xe_bo_size(bo), U64_MAX);
+							   start + xe_bo_size(bo), U64_MAX,
+							   exec);
 			} else {
-				err = xe_ggtt_insert_bo(t->mem.ggtt, bo);
+				err = xe_ggtt_insert_bo(t->mem.ggtt, bo, exec);
 			}
 			if (err)
 				goto err_unlock_put_bo;
@@ -2116,22 +2144,59 @@ err_unlock_put_bo:
 	return ERR_PTR(err);
 }
 
+/**
+ * xe_bo_create_locked_range() - Create a BO with range- and alignment options
+ * @xe: The xe device.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @vm: The local vm or NULL for external objects.
+ * @size: The storage size to use for the bo.
+ * @start: Start of fixed VRAM range or 0.
+ * @end: End of fixed VRAM range or ~0ULL.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ * @alignment: For GGTT buffer objects, the minimum GGTT alignment.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
+ *
+ * Create an Xe BO with range- and alignment options. If @start and @end indicate
+ * a fixed VRAM range, this must be a ttm_bo_type_kernel bo with VRAM placement
+ * only. The @alignment parameter can be used for GGTT alignment.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ */
 struct xe_bo *
 xe_bo_create_locked_range(struct xe_device *xe,
 			  struct xe_tile *tile, struct xe_vm *vm,
 			  size_t size, u64 start, u64 end,
-			  enum ttm_bo_type type, u32 flags, u64 alignment)
+			  enum ttm_bo_type type, u32 flags, u64 alignment,
+			  struct drm_exec *exec)
 {
 	return __xe_bo_create_locked(xe, tile, vm, size, start, end, 0, type,
-				     flags, alignment);
+				     flags, alignment, exec);
 }
 
+/**
+ * xe_bo_create_locked() - Create a BO
+ * @xe: The xe device.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @vm: The local vm or NULL for external objects.
+ * @size: The storage size to use for the bo.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
+ *
+ * Create a locked xe BO with no range- nor alignment restrictions.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ */
 struct xe_bo *xe_bo_create_locked(struct xe_device *xe, struct xe_tile *tile,
 				  struct xe_vm *vm, size_t size,
-				  enum ttm_bo_type type, u32 flags)
+				  enum ttm_bo_type type, u32 flags,
+				  struct drm_exec *exec)
 {
 	return __xe_bo_create_locked(xe, tile, vm, size, 0, ~0ULL, 0, type,
-				     flags, 0);
+				     flags, 0, exec);
 }
 
 struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_tile *tile,
@@ -2139,9 +2204,10 @@ struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_tile *tile,
 				u16 cpu_caching,
 				u32 flags)
 {
+	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *bo = __xe_bo_create_locked(xe, tile, vm, size, 0, ~0ULL,
 						 cpu_caching, ttm_bo_type_device,
-						 flags | XE_BO_FLAG_USER, 0);
+						 flags | XE_BO_FLAG_USER, 0, exec);
 	if (!IS_ERR(bo))
 		xe_bo_unlock_vm_held(bo);
 
@@ -2152,7 +2218,8 @@ struct xe_bo *xe_bo_create(struct xe_device *xe, struct xe_tile *tile,
 			   struct xe_vm *vm, size_t size,
 			   enum ttm_bo_type type, u32 flags)
 {
-	struct xe_bo *bo = xe_bo_create_locked(xe, tile, vm, size, type, flags);
+	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
+	struct xe_bo *bo = xe_bo_create_locked(xe, tile, vm, size, type, flags, exec);
 
 	if (!IS_ERR(bo))
 		xe_bo_unlock_vm_held(bo);
@@ -2180,6 +2247,7 @@ struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
 	int err;
 	u64 start = offset == ~0ull ? 0 : offset;
 	u64 end = offset == ~0ull ? offset : start + size;
+	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
 
 	if (flags & XE_BO_FLAG_STOLEN &&
 	    xe_ttm_stolen_cpu_access_needs_ggtt(xe))
@@ -2187,11 +2255,11 @@ struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
 
 	bo = xe_bo_create_locked_range(xe, tile, vm, size, start, end, type,
 				       flags | XE_BO_FLAG_NEEDS_CPU_ACCESS | XE_BO_FLAG_PINNED,
-				       alignment);
+				       alignment, exec);
 	if (IS_ERR(bo))
 		return bo;
 
-	err = xe_bo_pin(bo);
+	err = xe_bo_pin(bo, exec);
 	if (err)
 		goto err_put;
 
@@ -2314,6 +2382,7 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res)
  * xe_bo_pin_external - pin an external BO
  * @bo: buffer object to be pinned
  * @in_place: Pin in current placement, don't attempt to migrate.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * Pin an external (not tied to a VM, can be exported via dma-buf / prime FD)
  * BO. Unique call compared to xe_bo_pin as this function has it own set of
@@ -2321,7 +2390,7 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res)
  *
  * Returns 0 for success, negative error code otherwise.
  */
-int xe_bo_pin_external(struct xe_bo *bo, bool in_place)
+int xe_bo_pin_external(struct xe_bo *bo, bool in_place, struct drm_exec *exec)
 {
 	struct xe_device *xe = xe_bo_device(bo);
 	int err;
@@ -2331,7 +2400,7 @@ int xe_bo_pin_external(struct xe_bo *bo, bool in_place)
 
 	if (!xe_bo_is_pinned(bo)) {
 		if (!in_place) {
-			err = xe_bo_validate(bo, NULL, false);
+			err = xe_bo_validate(bo, NULL, false, exec);
 			if (err)
 				return err;
 		}
@@ -2354,7 +2423,17 @@ int xe_bo_pin_external(struct xe_bo *bo, bool in_place)
 	return 0;
 }
 
-int xe_bo_pin(struct xe_bo *bo)
+/**
+ * xe_bo_pin() - Pin a kernel bo after potentially migrating it
+ * @bo: The kernel bo to pin.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
+ *
+ * Attempts to migrate a bo to @bo->placement. If that succeeds,
+ * pins the bo.
+ *
+ * Return: %0 on success, negative error code on migration failure.
+ */
+int xe_bo_pin(struct xe_bo *bo, struct drm_exec *exec)
 {
 	struct ttm_place *place = &bo->placements[0];
 	struct xe_device *xe = xe_bo_device(bo);
@@ -2376,7 +2455,7 @@ int xe_bo_pin(struct xe_bo *bo)
 	/* We only expect at most 1 pin */
 	xe_assert(xe, !xe_bo_is_pinned(bo));
 
-	err = xe_bo_validate(bo, NULL, false);
+	err = xe_bo_validate(bo, NULL, false, exec);
 	if (err)
 		return err;
 
@@ -2469,6 +2548,7 @@ void xe_bo_unpin(struct xe_bo *bo)
  *      NULL. Used together with @allow_res_evict.
  * @allow_res_evict: Whether it's allowed to evict bos sharing @vm's
  *                   reservation object.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * Make sure the bo is in allowed placement, migrating it if necessary. If
  * needed, other bos will be evicted. If bos selected for eviction shares
@@ -2478,7 +2558,8 @@ void xe_bo_unpin(struct xe_bo *bo)
  * Return: 0 on success, negative error code on failure. May return
  * -EINTR or -ERESTARTSYS if internal waits are interrupted by a signal.
  */
-int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict)
+int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict,
+		   struct drm_exec *exec)
 {
 	struct ttm_operation_ctx ctx = {
 		.interruptible = true,
@@ -2500,6 +2581,7 @@ int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict)
 
 	xe_vm_set_validating(vm, allow_res_evict);
 	trace_xe_bo_validate(bo);
+	xe_validation_assert_exec(xe_bo_device(bo), exec, &bo->ttm.base);
 	ret = ttm_bo_validate(&bo->ttm, &bo->placement, &ctx);
 	xe_vm_clear_validating(vm, allow_res_evict);
 
@@ -2937,6 +3019,7 @@ static void xe_place_from_ttm_type(u32 mem_type, struct ttm_place *place)
  * xe_bo_migrate - Migrate an object to the desired region id
  * @bo: The buffer object to migrate.
  * @mem_type: The TTM region type to migrate to.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * Attempt to migrate the buffer object to the desired memory region. The
  * buffer object may not be pinned, and must be locked.
@@ -2948,7 +3031,7 @@ static void xe_place_from_ttm_type(u32 mem_type, struct ttm_place *place)
  * Return: 0 on success. Negative error code on failure. In particular may
  * return -EINTR or -ERESTARTSYS if signal pending.
  */
-int xe_bo_migrate(struct xe_bo *bo, u32 mem_type)
+int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct drm_exec *exec)
 {
 	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
 	struct ttm_operation_ctx ctx = {
@@ -2986,19 +3069,21 @@ int xe_bo_migrate(struct xe_bo *bo, u32 mem_type)
 		add_vram(xe, bo, &requested, bo->flags, mem_type, &c);
 	}
 
+	xe_validation_assert_exec(xe_bo_device(bo), exec, &bo->ttm.base);
 	return ttm_bo_validate(&bo->ttm, &placement, &ctx);
 }
 
 /**
  * xe_bo_evict - Evict an object to evict placement
  * @bo: The buffer object to migrate.
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * On successful completion, the object memory will be moved to evict
  * placement. This function blocks until the object has been fully moved.
  *
  * Return: 0 on success. Negative error code on failure.
  */
-int xe_bo_evict(struct xe_bo *bo)
+int xe_bo_evict(struct xe_bo *bo, struct drm_exec *exec)
 {
 	struct ttm_operation_ctx ctx = {
 		.interruptible = false,
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index c7ee9c7605e5..fa6813a7b4bc 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -10,6 +10,7 @@
 
 #include "xe_bo_types.h"
 #include "xe_macros.h"
+#include "xe_validation.h"
 #include "xe_vm_types.h"
 #include "xe_vm.h"
 #include "xe_vram_types.h"
@@ -92,15 +93,17 @@ struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo,
 				     struct xe_tile *tile, struct dma_resv *resv,
 				     struct ttm_lru_bulk_move *bulk, size_t size,
 				     u16 cpu_caching, enum ttm_bo_type type,
-				     u32 flags);
+				     u32 flags, struct drm_exec *exec);
 struct xe_bo *
 xe_bo_create_locked_range(struct xe_device *xe,
 			  struct xe_tile *tile, struct xe_vm *vm,
 			  size_t size, u64 start, u64 end,
-			  enum ttm_bo_type type, u32 flags, u64 alignment);
+			  enum ttm_bo_type type, u32 flags, u64 alignment,
+			  struct drm_exec *exec);
 struct xe_bo *xe_bo_create_locked(struct xe_device *xe, struct xe_tile *tile,
 				  struct xe_vm *vm, size_t size,
-				  enum ttm_bo_type type, u32 flags);
+				  enum ttm_bo_type type, u32 flags,
+				  struct drm_exec *exec);
 struct xe_bo *xe_bo_create(struct xe_device *xe, struct xe_tile *tile,
 			   struct xe_vm *vm, size_t size,
 			   enum ttm_bo_type type, u32 flags);
@@ -200,11 +203,12 @@ static inline void xe_bo_unlock_vm_held(struct xe_bo *bo)
 	}
 }
 
-int xe_bo_pin_external(struct xe_bo *bo, bool in_place);
-int xe_bo_pin(struct xe_bo *bo);
+int xe_bo_pin_external(struct xe_bo *bo, bool in_place, struct drm_exec *exec);
+int xe_bo_pin(struct xe_bo *bo, struct drm_exec *exec);
 void xe_bo_unpin_external(struct xe_bo *bo);
 void xe_bo_unpin(struct xe_bo *bo);
-int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict);
+int xe_bo_validate(struct xe_bo *bo, struct xe_vm *vm, bool allow_res_evict,
+		   struct drm_exec *exec);
 
 static inline bool xe_bo_is_pinned(struct xe_bo *bo)
 {
@@ -285,8 +289,8 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res);
 
 bool xe_bo_can_migrate(struct xe_bo *bo, u32 mem_type);
 
-int xe_bo_migrate(struct xe_bo *bo, u32 mem_type);
-int xe_bo_evict(struct xe_bo *bo);
+int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct drm_exec *exec);
+int xe_bo_evict(struct xe_bo *bo, struct drm_exec *exec);
 
 int xe_bo_evict_pinned(struct xe_bo *bo);
 int xe_bo_notifier_prepare_pinned(struct xe_bo *bo);
diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index af64baf872ef..199041bdb882 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -51,6 +51,7 @@ static int xe_dma_buf_pin(struct dma_buf_attachment *attach)
 	struct drm_gem_object *obj = attach->dmabuf->priv;
 	struct xe_bo *bo = gem_to_xe_bo(obj);
 	struct xe_device *xe = xe_bo_device(bo);
+	struct drm_exec *exec = XE_VALIDATION_UNSUPPORTED;
 	int ret;
 
 	/*
@@ -63,7 +64,7 @@ static int xe_dma_buf_pin(struct dma_buf_attachment *attach)
 		return -EINVAL;
 	}
 
-	ret = xe_bo_migrate(bo, XE_PL_TT);
+	ret = xe_bo_migrate(bo, XE_PL_TT, exec);
 	if (ret) {
 		if (ret != -EINTR && ret != -ERESTARTSYS)
 			drm_dbg(&xe->drm,
@@ -72,7 +73,7 @@ static int xe_dma_buf_pin(struct dma_buf_attachment *attach)
 		return ret;
 	}
 
-	ret = xe_bo_pin_external(bo, true);
+	ret = xe_bo_pin_external(bo, true, exec);
 	xe_assert(xe, !ret);
 
 	return 0;
@@ -92,6 +93,7 @@ static struct sg_table *xe_dma_buf_map(struct dma_buf_attachment *attach,
 	struct dma_buf *dma_buf = attach->dmabuf;
 	struct drm_gem_object *obj = dma_buf->priv;
 	struct xe_bo *bo = gem_to_xe_bo(obj);
+	struct drm_exec *exec = XE_VALIDATION_UNSUPPORTED;
 	struct sg_table *sgt;
 	int r = 0;
 
@@ -100,9 +102,9 @@ static struct sg_table *xe_dma_buf_map(struct dma_buf_attachment *attach,
 
 	if (!xe_bo_is_pinned(bo)) {
 		if (!attach->peer2peer)
-			r = xe_bo_migrate(bo, XE_PL_TT);
+			r = xe_bo_migrate(bo, XE_PL_TT, exec);
 		else
-			r = xe_bo_validate(bo, NULL, false);
+			r = xe_bo_validate(bo, NULL, false, exec);
 		if (r)
 			return ERR_PTR(r);
 	}
@@ -161,13 +163,14 @@ static int xe_dma_buf_begin_cpu_access(struct dma_buf *dma_buf,
 	struct xe_bo *bo = gem_to_xe_bo(obj);
 	bool reads =  (direction == DMA_BIDIRECTIONAL ||
 		       direction == DMA_FROM_DEVICE);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 
 	if (!reads)
 		return 0;
 
 	/* Can we do interruptible lock here? */
 	xe_bo_lock(bo, false);
-	(void)xe_bo_migrate(bo, XE_PL_TT);
+	(void)xe_bo_migrate(bo, XE_PL_TT, exec);
 	xe_bo_unlock(bo);
 
 	return 0;
@@ -208,13 +211,14 @@ xe_dma_buf_init_obj(struct drm_device *dev, struct xe_bo *storage,
 {
 	struct dma_resv *resv = dma_buf->resv;
 	struct xe_device *xe = to_xe_device(dev);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *bo;
 	int ret;
 
 	dma_resv_lock(resv, NULL);
 	bo = ___xe_bo_create_locked(xe, storage, NULL, resv, NULL, dma_buf->size,
 				    0, /* Will require 1way or 2way for vm_bind */
-				    ttm_bo_type_sg, XE_BO_FLAG_SYSTEM);
+				    ttm_bo_type_sg, XE_BO_FLAG_SYSTEM, exec);
 	if (IS_ERR(bo)) {
 		ret = PTR_ERR(bo);
 		goto error;
@@ -232,8 +236,9 @@ static void xe_dma_buf_move_notify(struct dma_buf_attachment *attach)
 {
 	struct drm_gem_object *obj = attach->importer_priv;
 	struct xe_bo *bo = gem_to_xe_bo(obj);
+	struct drm_exec *exec = XE_VALIDATION_UNSUPPORTED;
 
-	XE_WARN_ON(xe_bo_evict(bo));
+	XE_WARN_ON(xe_bo_evict(bo, exec));
 }
 
 static const struct dma_buf_attach_ops xe_dma_buf_attach_ops = {
diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index 1dc0c54cac5f..56edee596352 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -98,9 +98,13 @@
 static int xe_exec_fn(struct drm_gpuvm_exec *vm_exec)
 {
 	struct xe_vm *vm = container_of(vm_exec->vm, struct xe_vm, gpuvm);
+	int ret;
 
 	/* The fence slot added here is intended for the exec sched job. */
-	return xe_vm_validate_rebind(vm, &vm_exec->exec, 1);
+	xe_vm_set_validation_exec(vm, &vm_exec->exec);
+	ret = xe_vm_validate_rebind(vm, &vm_exec->exec, 1);
+	xe_vm_set_validation_exec(vm, NULL);
+	return ret;
 }
 
 int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index 71c7690a92b3..468364c09d35 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -731,7 +731,7 @@ void xe_ggtt_map_bo_unlocked(struct xe_ggtt *ggtt, struct xe_bo *bo)
 }
 
 static int __xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo,
-				  u64 start, u64 end)
+				  u64 start, u64 end, struct drm_exec *exec)
 {
 	u64 alignment = bo->min_align > 0 ? bo->min_align : XE_PAGE_SIZE;
 	u8 tile_id = ggtt->tile->id;
@@ -746,7 +746,7 @@ static int __xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo,
 		return 0;
 	}
 
-	err = xe_bo_validate(bo, NULL, false);
+	err = xe_bo_validate(bo, NULL, false, exec);
 	if (err)
 		return err;
 
@@ -788,25 +788,28 @@ out:
  * @bo: the &xe_bo to be inserted
  * @start: address where it will be inserted
  * @end: end of the range where it will be inserted
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * Return: 0 on success or a negative error code on failure.
  */
 int xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo,
-			 u64 start, u64 end)
+			 u64 start, u64 end, struct drm_exec *exec)
 {
-	return __xe_ggtt_insert_bo_at(ggtt, bo, start, end);
+	return __xe_ggtt_insert_bo_at(ggtt, bo, start, end, exec);
 }
 
 /**
  * xe_ggtt_insert_bo - Insert BO into GGTT
  * @ggtt: the &xe_ggtt where bo will be inserted
  * @bo: the &xe_bo to be inserted
+ * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * Return: 0 on success or a negative error code on failure.
  */
-int xe_ggtt_insert_bo(struct xe_ggtt *ggtt, struct xe_bo *bo)
+int xe_ggtt_insert_bo(struct xe_ggtt *ggtt, struct xe_bo *bo,
+		      struct drm_exec *exec)
 {
-	return __xe_ggtt_insert_bo_at(ggtt, bo, 0, U64_MAX);
+	return __xe_ggtt_insert_bo_at(ggtt, bo, 0, U64_MAX, exec);
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_ggtt.h b/drivers/gpu/drm/xe/xe_ggtt.h
index fbe1e397d05d..75fc7a1efea7 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.h
+++ b/drivers/gpu/drm/xe/xe_ggtt.h
@@ -10,6 +10,7 @@
 
 struct drm_printer;
 struct xe_tile;
+struct drm_exec;
 
 struct xe_ggtt *xe_ggtt_alloc(struct xe_tile *tile);
 int xe_ggtt_init_early(struct xe_ggtt *ggtt);
@@ -31,9 +32,9 @@ bool xe_ggtt_node_allocated(const struct xe_ggtt_node *node);
 void xe_ggtt_map_bo(struct xe_ggtt *ggtt, struct xe_ggtt_node *node,
 		    struct xe_bo *bo, u16 pat_index);
 void xe_ggtt_map_bo_unlocked(struct xe_ggtt *ggtt, struct xe_bo *bo);
-int xe_ggtt_insert_bo(struct xe_ggtt *ggtt, struct xe_bo *bo);
+int xe_ggtt_insert_bo(struct xe_ggtt *ggtt, struct xe_bo *bo, struct drm_exec *exec);
 int xe_ggtt_insert_bo_at(struct xe_ggtt *ggtt, struct xe_bo *bo,
-			 u64 start, u64 end);
+			 u64 start, u64 end, struct drm_exec *exec);
 void xe_ggtt_remove_bo(struct xe_ggtt *ggtt, struct xe_bo *bo);
 u64 xe_ggtt_largest_hole(struct xe_ggtt *ggtt, u64 alignment, u64 *spare);
 
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index d02d22fb3659..b9653ecbc7d4 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -87,10 +87,8 @@ static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma,
 	if (!bo)
 		return 0;
 
-	err = need_vram_move ? xe_bo_migrate(bo, vram->placement) :
-			       xe_bo_validate(bo, vm, true);
-
-	return err;
+	return need_vram_move ? xe_bo_migrate(bo, vram->placement, exec) :
+		xe_bo_validate(bo, vm, true, exec);
 }
 
 static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
@@ -140,7 +138,9 @@ retry_userptr:
 
 		/* Bind VMA only to the GT that has faulted */
 		trace_xe_vma_pf_bind(vma);
+		xe_vm_set_validation_exec(vm, &exec);
 		fence = xe_vma_rebind(vm, vma, BIT(tile->id));
+		xe_vm_set_validation_exec(vm, NULL);
 		if (IS_ERR(fence)) {
 			err = PTR_ERR(fence);
 			if (xe_vm_validate_should_retry(&exec, err, &end))
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index c8f0320d032f..906011671b60 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -1452,6 +1452,7 @@ static bool pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_confi
 static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size)
 {
 	struct xe_gt_sriov_config *config = pf_pick_vf_config(gt, vfid);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_tile *tile = gt_to_tile(gt);
 	struct xe_bo *bo;
@@ -1484,11 +1485,12 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size)
 				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
 				 XE_BO_FLAG_NEEDS_2M |
 				 XE_BO_FLAG_PINNED |
-				 XE_BO_FLAG_PINNED_LATE_RESTORE);
+				 XE_BO_FLAG_PINNED_LATE_RESTORE,
+				 exec);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
-	err = xe_bo_pin(bo);
+	err = xe_bo_pin(bo, exec);
 	xe_bo_unlock(bo);
 	if (unlikely(err)) {
 		xe_bo_put(bo);
diff --git a/drivers/gpu/drm/xe/xe_psmi.c b/drivers/gpu/drm/xe/xe_psmi.c
index a2c9ff5bfd59..2815987c190d 100644
--- a/drivers/gpu/drm/xe/xe_psmi.c
+++ b/drivers/gpu/drm/xe/xe_psmi.c
@@ -68,6 +68,7 @@ static void psmi_cleanup(struct xe_device *xe)
 static struct xe_bo *psmi_alloc_object(struct xe_device *xe,
 				       unsigned int id, size_t bo_size)
 {
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *bo = NULL;
 	struct xe_tile *tile;
 	int err;
@@ -83,11 +84,12 @@ static struct xe_bo *psmi_alloc_object(struct xe_device *xe,
 				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
 				 XE_BO_FLAG_PINNED |
 				 XE_BO_FLAG_PINNED_LATE_RESTORE |
-				 XE_BO_FLAG_NEEDS_CPU_ACCESS);
+				 XE_BO_FLAG_NEEDS_CPU_ACCESS,
+				 exec);
 
 	if (!IS_ERR(bo)) {
 		/* Buffer written by HW, ensure stays resident */
-		err = xe_bo_pin(bo);
+		err = xe_bo_pin(bo, exec);
 		if (err)
 			bo = ERR_PTR(err);
 		xe_bo_unlock(bo);
diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index c67967cdec79..3a5196b6b72b 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -869,6 +869,7 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 	struct device *dev = xe->drm.dev;
 	struct drm_buddy_block *block;
 	struct list_head *blocks;
+	struct drm_exec *exec;
 	struct xe_bo *bo;
 	ktime_t time_end = 0;
 	int err, idx;
@@ -877,12 +878,13 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 		return -ENODEV;
 
 	xe_pm_runtime_get(xe);
+	exec = XE_VALIDATION_UNIMPLEMENTED;
 
  retry:
 	bo = xe_bo_create_locked(vr->xe, NULL, NULL, end - start,
 				 ttm_bo_type_device,
 				 (IS_DGFX(xe) ? XE_BO_FLAG_VRAM(vr) : XE_BO_FLAG_SYSTEM) |
-				 XE_BO_FLAG_CPU_ADDR_MIRROR);
+				 XE_BO_FLAG_CPU_ADDR_MIRROR, exec);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		if (xe_vm_validate_should_retry(NULL, err, &time_end))
diff --git a/drivers/gpu/drm/xe/xe_validation.c b/drivers/gpu/drm/xe/xe_validation.c
new file mode 100644
index 000000000000..cc0684d24e02
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_validation.c
@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2024 Intel Corporation
+ */
+#include "xe_bo.h"
+#include <drm/drm_exec.h>
+#include <drm/drm_gem.h>
+
+#include "xe_assert.h"
+#include "xe_validation.h"
+
+#ifdef CONFIG_DRM_XE_DEBUG
+/**
+ * xe_validation_assert_exec() - Assert that the drm_exec pointer is suitable
+ * for validation.
+ * @xe: Pointer to the xe device.
+ * @exec: The drm_exec pointer to check.
+ * @obj: Pointer to the object subject to validation.
+ *
+ * NULL exec pointers are not allowed.
+ * For XE_VALIDATION_UNIMPLEMENTED, no checking.
+ * For XE_VLIDATION_OPT_OUT, check that the caller is a kunit test
+ * For XE_VALIDATION_UNSUPPORTED, check that the object subject to
+ * validation is a dma-buf, for which support for ww locking is
+ * not in place in the dma-buf layer.
+ */
+void xe_validation_assert_exec(const struct xe_device *xe,
+			       const struct drm_exec *exec,
+			       const struct drm_gem_object *obj)
+{
+	xe_assert(xe, exec);
+	if (IS_ERR(exec)) {
+		switch (PTR_ERR(exec)) {
+		case __XE_VAL_UNIMPLEMENTED:
+			break;
+		case __XE_VAL_UNSUPPORTED:
+			xe_assert(xe, !!obj->dma_buf);
+			break;
+#if IS_ENABLED(CONFIG_KUNIT)
+		case __XE_VAL_OPT_OUT:
+			xe_assert(xe, current->kunit_test);
+			break;
+#endif
+		default:
+			xe_assert(xe, false);
+		}
+	}
+}
+#endif
diff --git a/drivers/gpu/drm/xe/xe_validation.h b/drivers/gpu/drm/xe/xe_validation.h
new file mode 100644
index 000000000000..db50feacad7a
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_validation.h
@@ -0,0 +1,69 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2024 Intel Corporation
+ */
+#ifndef _XE_VALIDATION_H_
+#define _XE_VALIDATION_H_
+
+#include <linux/dma-resv.h>
+#include <linux/types.h>
+
+struct drm_exec;
+struct drm_gem_object;
+struct xe_device;
+
+#ifdef CONFIG_PROVE_LOCKING
+/**
+ * xe_validation_lockdep() - Assert that a drm_exec locking transaction can
+ * be initialized at this point.
+ */
+static inline void xe_validation_lockdep(void)
+{
+	struct ww_acquire_ctx ticket;
+
+	ww_acquire_init(&ticket, &reservation_ww_class);
+	ww_acquire_fini(&ticket);
+}
+#else
+static inline void xe_validation_lockdep(void)
+{
+}
+#endif
+
+/*
+ * Various values of the drm_exec pointer where we've not (yet)
+ * implemented full ww locking.
+ *
+ * XE_VALIDATION_UNIMPLEMENTED means implementation is pending.
+ * A lockdep check is made to assure that a drm_exec locking
+ * transaction can actually take place where the macro is
+ * used. If this asserts, the exec pointer needs to be assigned
+ * higher up in the callchain and passed down.
+ *
+ * XE_VALIDATION_UNSUPPORTED is for dma-buf code only where
+ * the dma-buf layer doesn't support WW locking.
+ *
+ * XE_VALIDATION_OPT_OUT is for simplification of kunit tests where
+ * exhaustive eviction isn't necessary.
+ */
+#define __XE_VAL_UNIMPLEMENTED -EINVAL
+#define XE_VALIDATION_UNIMPLEMENTED (xe_validation_lockdep(),		\
+				     (struct drm_exec *)ERR_PTR(__XE_VAL_UNIMPLEMENTED))
+
+#define __XE_VAL_UNSUPPORTED -EOPNOTSUPP
+#define XE_VALIDATION_UNSUPPORTED ((struct drm_exec *)ERR_PTR(__XE_VAL_UNSUPPORTED))
+
+#define __XE_VAL_OPT_OUT -ENOMEM
+#define XE_VALIDATION_OPT_OUT (xe_validation_lockdep(), \
+			       (struct drm_exec *)ERR_PTR(__XE_VAL_OPT_OUT))
+#ifdef CONFIG_DRM_XE_DEBUG
+void xe_validation_assert_exec(const struct xe_device *xe, const struct drm_exec *exec,
+			       const struct drm_gem_object *obj);
+#else
+#define xe_validation_assert_exec(_xe, _exec, _obj)	\
+	do {						\
+		(void)_xe; (void)_exec; (void)_obj;	\
+	} while (0)
+#endif
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 346975bc1065..8a6cb92e6a5b 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -346,7 +346,7 @@ static int xe_gpuvm_validate(struct drm_gpuvm_bo *vm_bo, struct drm_exec *exec)
 	if (!try_wait_for_completion(&vm->xe->pm_block))
 		return -EAGAIN;
 
-	ret = xe_bo_validate(gem_to_xe_bo(vm_bo->obj), vm, false);
+	ret = xe_bo_validate(gem_to_xe_bo(vm_bo->obj), vm, false, exec);
 	if (ret)
 		return ret;
 
@@ -513,7 +513,9 @@ retry:
 	if (err)
 		goto out_unlock;
 
+	xe_vm_set_validation_exec(vm, &exec);
 	err = xe_vm_rebind(vm, true);
+	xe_vm_set_validation_exec(vm, NULL);
 	if (err)
 		goto out_unlock;
 
@@ -2822,7 +2824,7 @@ static int vma_lock_and_validate(struct drm_exec *exec, struct xe_vma *vma,
 			err = drm_exec_lock_obj(exec, &bo->ttm.base);
 		if (!err && validate)
 			err = xe_bo_validate(bo, vm,
-					     !xe_vm_in_preempt_fence_mode(vm));
+					     !xe_vm_in_preempt_fence_mode(vm), exec);
 	}
 
 	return err;
@@ -2940,7 +2942,8 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm,
 					    false);
 		if (!err && !xe_vma_has_no_bo(vma))
 			err = xe_bo_migrate(xe_vma_bo(vma),
-					    region_to_mem_type[region]);
+					    region_to_mem_type[region],
+					    exec);
 		break;
 	}
 	default:
@@ -3219,7 +3222,9 @@ static struct dma_fence *vm_bind_ioctl_ops_execute(struct xe_vm *vm,
 			goto unlock;
 		}
 
+		xe_vm_set_validation_exec(vm, &exec);
 		fence = ops_execute(vm, vops);
+		xe_vm_set_validation_exec(vm, NULL);
 		if (IS_ERR(fence)) {
 			if (PTR_ERR(fence) == -ENODATA)
 				vm_bind_ioctl_ops_fini(vm, vops, NULL);
@@ -3784,10 +3789,18 @@ release_vm_lock:
  */
 int xe_vm_lock(struct xe_vm *vm, bool intr)
 {
-	if (intr)
-		return dma_resv_lock_interruptible(xe_vm_resv(vm), NULL);
+	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
+	int ret;
 
-	return dma_resv_lock(xe_vm_resv(vm), NULL);
+	if (intr)
+		ret = dma_resv_lock_interruptible(xe_vm_resv(vm), NULL);
+	else
+		ret = dma_resv_lock(xe_vm_resv(vm), NULL);
+
+	if (!ret)
+		xe_vm_set_validation_exec(vm, exec);
+
+	return ret;
 }
 
 /**
@@ -3798,6 +3811,7 @@ int xe_vm_lock(struct xe_vm *vm, bool intr)
  */
 void xe_vm_unlock(struct xe_vm *vm)
 {
+	xe_vm_set_validation_exec(vm, NULL);
 	dma_resv_unlock(xe_vm_resv(vm));
 }
 
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index f29a4baee05d..2c241f76d5fb 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -321,7 +321,7 @@ static inline void xe_vm_set_validating(struct xe_vm *vm, bool allow_res_evict)
 	if (vm && !allow_res_evict) {
 		xe_vm_assert_held(vm);
 		/* Pairs with READ_ONCE in xe_vm_is_validating() */
-		WRITE_ONCE(vm->validating, current);
+		WRITE_ONCE(vm->validation.validating, current);
 	}
 }
 
@@ -339,7 +339,7 @@ static inline void xe_vm_clear_validating(struct xe_vm *vm, bool allow_res_evict
 {
 	if (vm && !allow_res_evict) {
 		/* Pairs with READ_ONCE in xe_vm_is_validating() */
-		WRITE_ONCE(vm->validating, NULL);
+		WRITE_ONCE(vm->validation.validating, NULL);
 	}
 }
 
@@ -357,13 +357,41 @@ static inline void xe_vm_clear_validating(struct xe_vm *vm, bool allow_res_evict
 static inline bool xe_vm_is_validating(struct xe_vm *vm)
 {
 	/* Pairs with WRITE_ONCE in xe_vm_is_validating() */
-	if (READ_ONCE(vm->validating) == current) {
+	if (READ_ONCE(vm->validation.validating) == current) {
 		xe_vm_assert_held(vm);
 		return true;
 	}
 	return false;
 }
 
+/**
+ * xe_vm_set_validation_exec() - Accessor to set the drm_exec object
+ * @vm: The vm we want to register a drm_exec object with.
+ * @exec: The exec object we want to register.
+ *
+ * Set the drm_exec object used to lock the vm's resv.
+ */
+static inline void xe_vm_set_validation_exec(struct xe_vm *vm, struct drm_exec *exec)
+{
+	xe_vm_assert_held(vm);
+	xe_assert(vm->xe, !!exec ^ !!vm->validation._exec);
+	vm->validation._exec = exec;
+}
+
+/**
+ * xe_vm_set_validation_exec() - Accessor to read the drm_exec object
+ * @vm: The vm we want to register a drm_exec object with.
+ *
+ * Return: The drm_exec object used to lock the vm's resv. The value
+ * is a valid pointer, %NULL, or one of the special values defined in
+ * xe_validation.h.
+ */
+static inline struct drm_exec *xe_vm_validation_exec(struct xe_vm *vm)
+{
+	xe_vm_assert_held(vm);
+	return vm->validation._exec;
+}
+
 /**
  * xe_vm_has_valid_gpu_mapping() - Advisory helper to check if VMA or SVM range has
  * a valid GPU mapping
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 8e188b83ef30..da39940501d8 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -306,19 +306,35 @@ struct xe_vm {
 		bool capture_once;
 	} error_capture;
 
+	/**
+	 * @validation: Validation data only valid with the vm resv held.
+	 * Note: This is really task state of the task holding the vm resv,
+	 * and moving forward we should
+	 * come up with a better way of passing this down the call-
+	 * chain.
+	 */
+	struct {
+		/**
+		 * @validation.validating: The task that is currently making bos resident.
+		 * for this vm.
+		 * Protected by the VM's resv for writing. Opportunistic reading can be done
+		 * using READ_ONCE. Note: This is a workaround for the
+		 * TTM eviction_valuable() callback not being passed a struct
+		 * ttm_operation_context(). Future work might want to address this.
+		 */
+		struct task_struct *validating;
+		/**
+		 *  @validation.exec The drm_exec context used when locking the vm resv.
+		 *  Protected by the vm's resv.
+		 */
+		struct drm_exec *_exec;
+	} validation;
+
 	/**
 	 * @tlb_flush_seqno: Required TLB flush seqno for the next exec.
 	 * protected by the vm resv.
 	 */
 	u64 tlb_flush_seqno;
-	/**
-	 * @validating: The task that is currently making bos resident for this vm.
-	 * Protected by the VM's resv for writing. Opportunistic reading can be done
-	 * using READ_ONCE. Note: This is a workaround for the
-	 * TTM eviction_valuable() callback not being passed a struct
-	 * ttm_operation_context(). Future work might want to address this.
-	 */
-	struct task_struct *validating;
 	/** @batch_invalidate_tlb: Always invalidate TLB before batch start */
 	bool batch_invalidate_tlb;
 	/** @xef: XE file handle for tracking this VM's drm client */

From c460bc2311dfcdf8380199cc913e5e064b5b063c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:35 +0200
Subject: [PATCH 1173/1455] drm/xe: Introduce an xe_validation wrapper around
 drm_exec
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce a validation wrapper xe_validation_guard() as a helper
intended to be used around drm_exec transactions what perform
validations. Once TTM can handle exhaustive eviction we could
remove this wrapper or make it mostly a NO-OP unless other
functionality is added to it.

Currently the wrapper takes a read lock upon entry and if the
transaction hits an OOM, all locks are released and the
transaction is retried with a write-lock. If all other
validations participate in this scheme, the transaction with
the write lock will be the only transaction validating and
should have access to all available non-pinned memory.

There is currently a problem in that TTM converts -EDEADLOCKS to
-ENOMEM, and with ww_mutex slowpath error injections, we can hit
-ENOMEMs without having actually ran out of memory. We abuse
ww_mutex internals to detect such situations until TTM is fixes
to not convert the error code. In the meantime, injecting
ww_mutex slowpath -EDEADLOCKs is a good way to test
the implementation in the absence of real OOMs.

Just introduce the wrapper in this commit. It will be hooked up
to the driver in following commits.

v2:
- Mark class_xe_validation conditional so that the loop is
  skipped on initialization error.
- Argument sanitation (Matt Brost)
- Fix conditional execution of xe_validation_ctx_fini()
  (Matt Brost)
- Add a no_block mode for upcoming use in the CPU fault handler.
v4:
- Update kerneldoc. (Xe CI).

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-3-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_validation.c | 228 +++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_validation.h | 123 ++++++++++++++++
 2 files changed, 351 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_validation.c b/drivers/gpu/drm/xe/xe_validation.c
index cc0684d24e02..b90fda3dd5f4 100644
--- a/drivers/gpu/drm/xe/xe_validation.c
+++ b/drivers/gpu/drm/xe/xe_validation.c
@@ -5,6 +5,7 @@
 #include "xe_bo.h"
 #include <drm/drm_exec.h>
 #include <drm/drm_gem.h>
+#include <drm/drm_gpuvm.h>
 
 #include "xe_assert.h"
 #include "xe_validation.h"
@@ -47,3 +48,230 @@ void xe_validation_assert_exec(const struct xe_device *xe,
 	}
 }
 #endif
+
+static int xe_validation_lock(struct xe_validation_ctx *ctx)
+{
+	struct xe_validation_device *val = ctx->val;
+	int ret = 0;
+
+	if (ctx->val_flags.interruptible) {
+		if (ctx->request_exclusive)
+			ret = down_write_killable(&val->lock);
+		else
+			ret = down_read_interruptible(&val->lock);
+	} else {
+		if (ctx->request_exclusive)
+			down_write(&val->lock);
+		else
+			down_read(&val->lock);
+	}
+
+	if (!ret) {
+		ctx->lock_held = true;
+		ctx->lock_held_exclusive = ctx->request_exclusive;
+	}
+
+	return ret;
+}
+
+static int xe_validation_trylock(struct xe_validation_ctx *ctx)
+{
+	struct xe_validation_device *val = ctx->val;
+	bool locked;
+
+	if (ctx->request_exclusive)
+		locked = down_write_trylock(&val->lock);
+	else
+		locked = down_read_trylock(&val->lock);
+
+	if (locked) {
+		ctx->lock_held = true;
+		ctx->lock_held_exclusive = ctx->request_exclusive;
+	}
+
+	return locked ? 0 : -EWOULDBLOCK;
+}
+
+static void xe_validation_unlock(struct xe_validation_ctx *ctx)
+{
+	if (!ctx->lock_held)
+		return;
+
+	if (ctx->lock_held_exclusive)
+		up_write(&ctx->val->lock);
+	else
+		up_read(&ctx->val->lock);
+
+	ctx->lock_held = false;
+}
+
+/**
+ * xe_validation_ctx_init() - Initialize an xe_validation_ctx
+ * @ctx: The xe_validation_ctx to initialize.
+ * @val: The xe_validation_device representing the validation domain.
+ * @exec: The struct drm_exec to use for the transaction. May be NULL.
+ * @flags: The flags to use for initialization.
+ *
+ * Initialize and lock a an xe_validation transaction using the validation domain
+ * represented by @val. Also initialize the drm_exec object forwarding parts of
+ * @flags to the drm_exec initialization. The @flags.exclusive flag should
+ * typically be set to false to avoid locking out other validators from the
+ * domain until an OOM is hit. For testing- or final attempt purposes it can,
+ * however, be set to true.
+ *
+ * Return: %0 on success, %-EINTR if interruptible initial locking failed with a
+ * signal pending. If @flags.no_block is set to true, a failed trylock
+ * returns %-EWOULDBLOCK.
+ */
+int xe_validation_ctx_init(struct xe_validation_ctx *ctx, struct xe_validation_device *val,
+			   struct drm_exec *exec, const struct xe_val_flags flags)
+{
+	int ret;
+
+	ctx->exec = exec;
+	ctx->val = val;
+	ctx->lock_held = false;
+	ctx->lock_held_exclusive = false;
+	ctx->request_exclusive = flags.exclusive;
+	ctx->val_flags = flags;
+	ctx->exec_flags = 0;
+	ctx->nr = 0;
+
+	if (flags.no_block)
+		ret = xe_validation_trylock(ctx);
+	else
+		ret = xe_validation_lock(ctx);
+	if (ret)
+		return ret;
+
+	if (exec) {
+		if (flags.interruptible)
+			ctx->exec_flags |= DRM_EXEC_INTERRUPTIBLE_WAIT;
+		if (flags.exec_ignore_duplicates)
+			ctx->exec_flags |= DRM_EXEC_IGNORE_DUPLICATES;
+		drm_exec_init(exec, ctx->exec_flags, ctx->nr);
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_DEBUG_WW_MUTEX_SLOWPATH
+/*
+ * This abuses both drm_exec and ww_mutex internals and should be
+ * replaced by checking for -EDEADLK when we can make TTM
+ * stop converting -EDEADLK to -ENOMEM.
+ * An alternative is to not have exhaustive eviction with
+ * CONFIG_DEBUG_WW_MUTEX_SLOWPATH until that happens.
+ */
+static bool xe_validation_contention_injected(struct drm_exec *exec)
+{
+	return !!exec->ticket.contending_lock;
+}
+
+#else
+
+static bool xe_validation_contention_injected(struct drm_exec *exec)
+{
+	return false;
+}
+
+#endif
+
+static bool __xe_validation_should_retry(struct xe_validation_ctx *ctx, int ret)
+{
+	if (ret == -ENOMEM &&
+	    ((ctx->request_exclusive &&
+	      xe_validation_contention_injected(ctx->exec)) ||
+	     !ctx->request_exclusive)) {
+		ctx->request_exclusive = true;
+		return true;
+	}
+
+	return false;
+}
+
+/**
+ * xe_validation_exec_lock() - Perform drm_gpuvm_exec_lock within a validation
+ * transaction.
+ * @ctx: An uninitialized xe_validation_ctx.
+ * @vm_exec: An initialized struct vm_exec.
+ * @val: The validation domain.
+ *
+ * The drm_gpuvm_exec_lock() function internally initializes its drm_exec
+ * transaction and therefore doesn't lend itself very well to be using
+ * xe_validation_ctx_init(). Provide a helper that takes an uninitialized
+ * xe_validation_ctx and calls drm_gpuvm_exec_lock() with OOM retry.
+ *
+ * Return: %0 on success, negative error code on failure.
+ */
+int xe_validation_exec_lock(struct xe_validation_ctx *ctx,
+			    struct drm_gpuvm_exec *vm_exec,
+			    struct xe_validation_device *val)
+{
+	int ret;
+
+	memset(ctx, 0, sizeof(*ctx));
+	ctx->exec = &vm_exec->exec;
+	ctx->exec_flags = vm_exec->flags;
+	ctx->val = val;
+	if (ctx->exec_flags & DRM_EXEC_INTERRUPTIBLE_WAIT)
+		ctx->val_flags.interruptible = 1;
+	if (ctx->exec_flags & DRM_EXEC_IGNORE_DUPLICATES)
+		ctx->val_flags.exec_ignore_duplicates = 1;
+retry:
+	ret = xe_validation_lock(ctx);
+	if (ret)
+		return ret;
+
+	ret = drm_gpuvm_exec_lock(vm_exec);
+	if (ret) {
+		xe_validation_unlock(ctx);
+		if (__xe_validation_should_retry(ctx, ret))
+			goto retry;
+	}
+
+	return ret;
+}
+
+/**
+ * xe_validation_ctx_fini() - Finalize a validation transaction
+ * @ctx: The Validation transaction to finalize.
+ *
+ * Finalize a validation transaction and its related drm_exec transaction.
+ */
+void xe_validation_ctx_fini(struct xe_validation_ctx *ctx)
+{
+	drm_exec_fini(ctx->exec);
+	xe_validation_unlock(ctx);
+}
+
+/**
+ * xe_validation_should_retry() - Determine if a validation transaction should retry
+ * @ctx: The validation transaction.
+ * @ret: Pointer to a return value variable.
+ *
+ * Determines whether a validation transaction should retry based on the
+ * internal transaction state and the return value pointed to by @ret.
+ * If a validation should be retried, the transaction is prepared for that,
+ * and the validation locked might be re-locked in exclusive mode, and *@ret
+ * is set to %0. If the re-locking errors, typically due to interruptible
+ * locking with signal pending, *@ret is instead set to -EINTR and the
+ * function returns %false.
+ *
+ * Return: %true if validation should be retried, %false otherwise.
+ */
+bool xe_validation_should_retry(struct xe_validation_ctx *ctx, int *ret)
+{
+	if (__xe_validation_should_retry(ctx, *ret)) {
+		drm_exec_fini(ctx->exec);
+		*ret = 0;
+		if (ctx->request_exclusive != ctx->lock_held_exclusive) {
+			xe_validation_unlock(ctx);
+			*ret = xe_validation_lock(ctx);
+		}
+		drm_exec_init(ctx->exec, ctx->exec_flags, ctx->nr);
+		return !*ret;
+	}
+
+	return false;
+}
diff --git a/drivers/gpu/drm/xe/xe_validation.h b/drivers/gpu/drm/xe/xe_validation.h
index db50feacad7a..fec331d791e7 100644
--- a/drivers/gpu/drm/xe/xe_validation.h
+++ b/drivers/gpu/drm/xe/xe_validation.h
@@ -7,9 +7,11 @@
 
 #include <linux/dma-resv.h>
 #include <linux/types.h>
+#include <linux/rwsem.h>
 
 struct drm_exec;
 struct drm_gem_object;
+struct drm_gpuvm_exec;
 struct xe_device;
 
 #ifdef CONFIG_PROVE_LOCKING
@@ -66,4 +68,125 @@ void xe_validation_assert_exec(const struct xe_device *xe, const struct drm_exec
 	} while (0)
 #endif
 
+/**
+ * struct xe_validation_device - The domain for exhaustive eviction
+ * @lock: The lock used to exclude other processes from allocating graphics memory
+ *
+ * The struct xe_validation_device represents the domain for which we want to use
+ * exhaustive eviction. The @lock is typically grabbed in read mode for allocations
+ * but when graphics memory allocation fails, it is retried with the write mode held.
+ */
+struct xe_validation_device {
+	struct rw_semaphore lock;
+};
+
+/**
+ * struct xe_val_flags - Flags for xe_validation_ctx_init().
+ * @exclusive: Start the validation transaction by locking out all other validators.
+ * @no_block:  Don't block on initialization.
+ * @interruptible: Block interruptible if blocking. Implies initializing the drm_exec
+ * context with the DRM_EXEC_INTERRUPTIBLE_WAIT flag.
+ * @exec_ignore_duplicates: Initialize the drm_exec context with the
+ * DRM_EXEC_IGNORE_DUPLICATES flag.
+ */
+struct xe_val_flags {
+	u32 exclusive :1;
+	u32 no_block :1;
+	u32 interruptible :1;
+	u32 exec_ignore_duplicates :1;
+};
+
+/**
+ * struct xe_validation_ctx - A struct drm_exec subclass with support for
+ * exhaustive eviction
+ * @exec: The drm_exec object base class. Note that we use a pointer instead of
+ * embedding to avoid diamond inheritance.
+ * @val: The exhaustive eviction domain.
+ * @val_flags: Copy of the struct xe_val_flags passed to xe_validation_ctx_init.
+ * @lock_held: Whether The domain lock is currently held.
+ * @lock_held_exclusive: Whether the domain lock is held in exclusive mode.
+ * @request_exclusive: Whether to lock exclusively (write mode) the next time
+ * the domain lock is locked.
+ * @exec_flags: The drm_exec flags used for drm_exec (re-)initialization.
+ * @nr: The drm_exec nr parameter used for drm_exec (re-)initializaiton.
+ */
+struct xe_validation_ctx {
+	struct drm_exec *exec;
+	struct xe_validation_device *val;
+	struct xe_val_flags val_flags;
+	bool lock_held;
+	bool lock_held_exclusive;
+	bool request_exclusive;
+	u32 exec_flags;
+	unsigned int nr;
+};
+
+int xe_validation_ctx_init(struct xe_validation_ctx *ctx, struct xe_validation_device *val,
+			   struct drm_exec *exec, const struct xe_val_flags flags);
+
+int xe_validation_exec_lock(struct xe_validation_ctx *ctx, struct drm_gpuvm_exec *vm_exec,
+			    struct xe_validation_device *val);
+
+void xe_validation_ctx_fini(struct xe_validation_ctx *ctx);
+
+bool xe_validation_should_retry(struct xe_validation_ctx *ctx, int *ret);
+
+/**
+ * xe_validation_retry_on_oom() - Retry on oom in an xe_validaton transaction
+ * @_ctx: Pointer to the xe_validation_ctx
+ * @_ret: The current error value possibly holding -ENOMEM
+ *
+ * Use this in way similar to drm_exec_retry_on_contention().
+ * If @_ret contains -ENOMEM the tranaction is restarted once in a way that
+ * blocks other transactions and allows exhastive eviction. If the transaction
+ * was already restarted once, Just return the -ENOMEM. May also set
+ * _ret to -EINTR if not retrying and waits are interruptible.
+ * May only be used within a drm_exec_until_all_locked() loop.
+ */
+#define xe_validation_retry_on_oom(_ctx, _ret)				\
+	do {								\
+		if (xe_validation_should_retry(_ctx, _ret))		\
+			goto *__drm_exec_retry_ptr;			\
+	} while (0)
+
+/**
+ * xe_validation_device_init - Initialize a struct xe_validation_device
+ * @val: The xe_validation_device to init.
+ */
+static inline void
+xe_validation_device_init(struct xe_validation_device *val)
+{
+	init_rwsem(&val->lock);
+}
+
+/*
+ * Make guard() and scoped_guard() work with xe_validation_ctx
+ * so that we can exit transactions without caring about the
+ * cleanup.
+ */
+DEFINE_CLASS(xe_validation, struct xe_validation_ctx *,
+	     if (_T) xe_validation_ctx_fini(_T);,
+	     ({_ret = xe_validation_ctx_init(_ctx, _val, _exec, _flags);
+	       _ret ? NULL : _ctx; }),
+	     struct xe_validation_ctx *_ctx, struct xe_validation_device *_val,
+	     struct drm_exec *_exec, const struct xe_val_flags _flags, int _ret);
+static inline void *class_xe_validation_lock_ptr(class_xe_validation_t *_T)
+{return *_T; }
+#define class_xe_validation_is_conditional true
+
+/**
+ * xe_validation_guard() - An auto-cleanup xe_validation_ctx transaction
+ * @_ctx: The xe_validation_ctx.
+ * @_val: The xe_validation_device.
+ * @_exec: The struct drm_exec object
+ * @_flags: Flags for the xe_validation_ctx initialization.
+ * @_ret: Return in / out parameter. May be set by this macro. Typicall 0 when called.
+ *
+ * This macro is will initiate a drm_exec transaction with additional support for
+ * exhaustive eviction.
+ */
+#define xe_validation_guard(_ctx, _val, _exec, _flags, _ret)		\
+	scoped_guard(xe_validation, _ctx, _val, _exec, _flags, _ret) \
+	drm_exec_until_all_locked(_exec)
+
 #endif

From a2f2453c2c2647f16b86a805e9b45c1fcd9b8fc9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:36 +0200
Subject: [PATCH 1174/1455] drm/xe: Convert xe_bo_create_user() for exhaustive
 eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use the xe_validation_guard() to convert xe_bo_create_user()
for exhaustive eviction.

v2:
- Adapt to argument changes of xe_validation_guard()

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com> #v1
Link: https://lore.kernel.org/r/20250908101246.65025-4-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/tests/xe_bo.c      |  16 ++--
 drivers/gpu/drm/xe/tests/xe_dma_buf.c |   4 +-
 drivers/gpu/drm/xe/tests/xe_migrate.c |  12 +--
 drivers/gpu/drm/xe/xe_bo.c            | 114 +++++++++++++++++---------
 drivers/gpu/drm/xe/xe_bo.h            |   9 +-
 drivers/gpu/drm/xe/xe_device.c        |   2 +
 drivers/gpu/drm/xe/xe_device_types.h  |   3 +
 drivers/gpu/drm/xe/xe_vm.c            |  14 ++++
 drivers/gpu/drm/xe/xe_vm.h            |   2 +
 9 files changed, 115 insertions(+), 61 deletions(-)

diff --git a/drivers/gpu/drm/xe/tests/xe_bo.c b/drivers/gpu/drm/xe/tests/xe_bo.c
index 17366b60ee31..2294cf89f3e1 100644
--- a/drivers/gpu/drm/xe/tests/xe_bo.c
+++ b/drivers/gpu/drm/xe/tests/xe_bo.c
@@ -139,8 +139,8 @@ static void ccs_test_run_tile(struct xe_device *xe, struct xe_tile *tile,
 	else
 		kunit_info(test, "Testing system memory\n");
 
-	bo = xe_bo_create_user(xe, NULL, NULL, SZ_1M, DRM_XE_GEM_CPU_CACHING_WC,
-			       bo_flags);
+	bo = xe_bo_create_user(xe, NULL, SZ_1M, DRM_XE_GEM_CPU_CACHING_WC,
+			       bo_flags, exec);
 	if (IS_ERR(bo)) {
 		KUNIT_FAIL(test, "Failed to create bo.\n");
 		return;
@@ -220,18 +220,18 @@ static int evict_test_run_tile(struct xe_device *xe, struct xe_tile *tile, struc
 
 	for (i = 0; i < 2; ++i) {
 		xe_vm_lock(vm, false);
-		bo = xe_bo_create_user(xe, NULL, vm, 0x10000,
+		bo = xe_bo_create_user(xe, vm, 0x10000,
 				       DRM_XE_GEM_CPU_CACHING_WC,
-				       bo_flags);
+				       bo_flags, exec);
 		xe_vm_unlock(vm);
 		if (IS_ERR(bo)) {
 			KUNIT_FAIL(test, "bo create err=%pe\n", bo);
 			break;
 		}
 
-		external = xe_bo_create_user(xe, NULL, NULL, 0x10000,
+		external = xe_bo_create_user(xe, NULL, 0x10000,
 					     DRM_XE_GEM_CPU_CACHING_WC,
-					     bo_flags);
+					     bo_flags, NULL);
 		if (IS_ERR(external)) {
 			KUNIT_FAIL(test, "external bo create err=%pe\n", external);
 			goto cleanup_bo;
@@ -497,9 +497,9 @@ static int shrink_test_run_device(struct xe_device *xe)
 		INIT_LIST_HEAD(&link->link);
 
 		/* We can create bos using WC caching here. But it is slower. */
-		bo = xe_bo_create_user(xe, NULL, NULL, XE_BO_SHRINK_SIZE,
+		bo = xe_bo_create_user(xe, NULL, XE_BO_SHRINK_SIZE,
 				       DRM_XE_GEM_CPU_CACHING_WB,
-				       XE_BO_FLAG_SYSTEM);
+				       XE_BO_FLAG_SYSTEM, NULL);
 		if (IS_ERR(bo)) {
 			if (bo != ERR_PTR(-ENOMEM) && bo != ERR_PTR(-ENOSPC) &&
 			    bo != ERR_PTR(-EINTR) && bo != ERR_PTR(-ERESTARTSYS))
diff --git a/drivers/gpu/drm/xe/tests/xe_dma_buf.c b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
index 7117486f0432..a7e548a2bdfb 100644
--- a/drivers/gpu/drm/xe/tests/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/tests/xe_dma_buf.c
@@ -114,8 +114,8 @@ static void xe_test_dmabuf_import_same_driver(struct xe_device *xe)
 		size = SZ_64K;
 
 	kunit_info(test, "running %s\n", __func__);
-	bo = xe_bo_create_user(xe, NULL, NULL, size, DRM_XE_GEM_CPU_CACHING_WC,
-			       params->mem_mask);
+	bo = xe_bo_create_user(xe, NULL, size, DRM_XE_GEM_CPU_CACHING_WC,
+			       params->mem_mask, NULL);
 	if (IS_ERR(bo)) {
 		KUNIT_FAIL(test, "xe_bo_create() failed with err=%ld\n",
 			   PTR_ERR(bo));
diff --git a/drivers/gpu/drm/xe/tests/xe_migrate.c b/drivers/gpu/drm/xe/tests/xe_migrate.c
index dfb445d09759..afa794e56065 100644
--- a/drivers/gpu/drm/xe/tests/xe_migrate.c
+++ b/drivers/gpu/drm/xe/tests/xe_migrate.c
@@ -642,11 +642,11 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 	struct drm_exec *exec;
 	long ret;
 
-	sys_bo = xe_bo_create_user(xe, NULL, NULL, SZ_4M,
+	sys_bo = xe_bo_create_user(xe, NULL, SZ_4M,
 				   DRM_XE_GEM_CPU_CACHING_WC,
 				   XE_BO_FLAG_SYSTEM |
 				   XE_BO_FLAG_NEEDS_CPU_ACCESS |
-				   XE_BO_FLAG_PINNED);
+				   XE_BO_FLAG_PINNED, NULL);
 
 	if (IS_ERR(sys_bo)) {
 		KUNIT_FAIL(test, "xe_bo_create() failed with err=%ld\n",
@@ -669,10 +669,10 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 	}
 	xe_bo_unlock(sys_bo);
 
-	ccs_bo = xe_bo_create_user(xe, NULL, NULL, SZ_4M,
+	ccs_bo = xe_bo_create_user(xe, NULL, SZ_4M,
 				   DRM_XE_GEM_CPU_CACHING_WC,
 				   bo_flags | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-				   XE_BO_FLAG_PINNED);
+				   XE_BO_FLAG_PINNED, NULL);
 
 	if (IS_ERR(ccs_bo)) {
 		KUNIT_FAIL(test, "xe_bo_create() failed with err=%ld\n",
@@ -694,10 +694,10 @@ static void validate_ccs_test_run_tile(struct xe_device *xe, struct xe_tile *til
 	}
 	xe_bo_unlock(ccs_bo);
 
-	vram_bo = xe_bo_create_user(xe, NULL, NULL, SZ_4M,
+	vram_bo = xe_bo_create_user(xe, NULL, SZ_4M,
 				    DRM_XE_GEM_CPU_CACHING_WC,
 				    bo_flags | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-				    XE_BO_FLAG_PINNED);
+				    XE_BO_FLAG_PINNED, NULL);
 	if (IS_ERR(vram_bo)) {
 		KUNIT_FAIL(test, "xe_bo_create() failed with err=%ld\n",
 			   PTR_ERR(vram_bo));
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 36e804dcc38f..ac330159e339 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -2199,30 +2199,66 @@ struct xe_bo *xe_bo_create_locked(struct xe_device *xe, struct xe_tile *tile,
 				     flags, 0, exec);
 }
 
-struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_tile *tile,
-				struct xe_vm *vm, size_t size,
-				u16 cpu_caching,
-				u32 flags)
+static struct xe_bo *xe_bo_create_novm(struct xe_device *xe, struct xe_tile *tile,
+				       size_t size, u16 cpu_caching,
+				       enum ttm_bo_type type, u32 flags,
+				       u64 alignment, bool intr)
 {
-	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
-	struct xe_bo *bo = __xe_bo_create_locked(xe, tile, vm, size, 0, ~0ULL,
-						 cpu_caching, ttm_bo_type_device,
-						 flags | XE_BO_FLAG_USER, 0, exec);
-	if (!IS_ERR(bo))
-		xe_bo_unlock_vm_held(bo);
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
+	struct xe_bo *bo;
+	int ret = 0;
 
-	return bo;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.interruptible = intr},
+			    ret) {
+		bo = __xe_bo_create_locked(xe, tile, NULL, size, 0, ~0ULL,
+					   cpu_caching, type, flags, alignment, &exec);
+		drm_exec_retry_on_contention(&exec);
+		if (IS_ERR(bo)) {
+			ret = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&ctx, &ret);
+		} else {
+			xe_bo_unlock(bo);
+		}
+	}
+
+	return ret ? ERR_PTR(ret) : bo;
 }
 
-struct xe_bo *xe_bo_create(struct xe_device *xe, struct xe_tile *tile,
-			   struct xe_vm *vm, size_t size,
-			   enum ttm_bo_type type, u32 flags)
+/**
+ * xe_bo_create_user() - Create a user BO
+ * @xe: The xe device.
+ * @vm: The local vm or NULL for external objects.
+ * @size: The storage size to use for the bo.
+ * @cpu_caching: The caching mode to be used for system backing store.
+ * @flags: XE_BO_FLAG_ flags.
+ * @exec: The drm_exec transaction to use for exhaustive eviction, or NULL
+ * if such a transaction should be initiated by the call.
+ *
+ * Create a bo on behalf of user-space.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ */
+struct xe_bo *xe_bo_create_user(struct xe_device *xe,
+				struct xe_vm *vm, size_t size,
+				u16 cpu_caching,
+				u32 flags, struct drm_exec *exec)
 {
-	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
-	struct xe_bo *bo = xe_bo_create_locked(xe, tile, vm, size, type, flags, exec);
+	struct xe_bo *bo;
 
-	if (!IS_ERR(bo))
-		xe_bo_unlock_vm_held(bo);
+	flags |= XE_BO_FLAG_USER;
+
+	if (vm || exec) {
+		xe_assert(xe, exec);
+		bo = __xe_bo_create_locked(xe, NULL, vm, size, 0, ~0ULL,
+					   cpu_caching, ttm_bo_type_device,
+					   flags, 0, exec);
+		if (!IS_ERR(bo))
+			xe_bo_unlock_vm_held(bo);
+	} else {
+		bo = xe_bo_create_novm(xe, NULL, size, cpu_caching,
+				       ttm_bo_type_device, flags, 0, true);
+	}
 
 	return bo;
 }
@@ -2777,8 +2813,9 @@ int xe_gem_create_ioctl(struct drm_device *dev, void *data,
 	struct xe_device *xe = to_xe_device(dev);
 	struct xe_file *xef = to_xe_file(file);
 	struct drm_xe_gem_create *args = data;
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
 	struct xe_vm *vm = NULL;
-	ktime_t end = 0;
 	struct xe_bo *bo;
 	unsigned int bo_flags;
 	u32 handle;
@@ -2852,25 +2889,26 @@ int xe_gem_create_ioctl(struct drm_device *dev, void *data,
 			return -ENOENT;
 	}
 
-retry:
-	if (vm) {
-		err = xe_vm_lock(vm, true);
-		if (err)
-			goto out_vm;
+	err = 0;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.interruptible = true},
+			    err) {
+		if (vm) {
+			err = xe_vm_drm_exec_lock(vm, &exec);
+			drm_exec_retry_on_contention(&exec);
+			if (err)
+				break;
+		}
+		bo = xe_bo_create_user(xe, vm, args->size, args->cpu_caching,
+				       bo_flags, &exec);
+		drm_exec_retry_on_contention(&exec);
+		if (IS_ERR(bo)) {
+			err = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&ctx, &err);
+			break;
+		}
 	}
-
-	bo = xe_bo_create_user(xe, NULL, vm, args->size, args->cpu_caching,
-			       bo_flags);
-
-	if (vm)
-		xe_vm_unlock(vm);
-
-	if (IS_ERR(bo)) {
-		err = PTR_ERR(bo);
-		if (xe_vm_validate_should_retry(NULL, err, &end))
-			goto retry;
+	if (err)
 		goto out_vm;
-	}
 
 	if (args->extensions) {
 		err = gem_create_user_extensions(xe, bo, args->extensions, 0);
@@ -3243,11 +3281,11 @@ int xe_bo_dumb_create(struct drm_file *file_priv,
 	args->size = ALIGN(mul_u32_u32(args->pitch, args->height),
 			   page_size);
 
-	bo = xe_bo_create_user(xe, NULL, NULL, args->size,
+	bo = xe_bo_create_user(xe, NULL, args->size,
 			       DRM_XE_GEM_CPU_CACHING_WC,
 			       XE_BO_FLAG_VRAM_IF_DGFX(xe_device_get_root_tile(xe)) |
 			       XE_BO_FLAG_SCANOUT |
-			       XE_BO_FLAG_NEEDS_CPU_ACCESS);
+			       XE_BO_FLAG_NEEDS_CPU_ACCESS, NULL);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index fa6813a7b4bc..3878df635b03 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -104,13 +104,8 @@ struct xe_bo *xe_bo_create_locked(struct xe_device *xe, struct xe_tile *tile,
 				  struct xe_vm *vm, size_t size,
 				  enum ttm_bo_type type, u32 flags,
 				  struct drm_exec *exec);
-struct xe_bo *xe_bo_create(struct xe_device *xe, struct xe_tile *tile,
-			   struct xe_vm *vm, size_t size,
-			   enum ttm_bo_type type, u32 flags);
-struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_tile *tile,
-				struct xe_vm *vm, size_t size,
-				u16 cpu_caching,
-				u32 flags);
+struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_vm *vm, size_t size,
+				u16 cpu_caching, u32 flags, struct drm_exec *exec);
 struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 				   struct xe_vm *vm, size_t size,
 				   enum ttm_bo_type type, u32 flags);
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index 9e2952c9c06a..c7bba535b145 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -457,6 +457,8 @@ struct xe_device *xe_device_create(struct pci_dev *pdev,
 	if (err)
 		goto err;
 
+	xe_validation_device_init(&xe->val);
+
 	init_waitqueue_head(&xe->ufence_wq);
 
 	init_rwsem(&xe->usm.lock);
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index a6ba880e4181..8233002e1c5b 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -26,6 +26,7 @@
 #include "xe_sriov_vf_ccs_types.h"
 #include "xe_step_types.h"
 #include "xe_survivability_mode_types.h"
+#include "xe_validation.h"
 
 #if IS_ENABLED(CONFIG_DRM_XE_DEBUG)
 #define TEST_VM_OPS_ERROR
@@ -583,6 +584,8 @@ struct xe_device {
 	 */
 	atomic64_t global_total_pages;
 #endif
+	/** @val: The domain for exhaustive eviction, which is currently per device. */
+	struct xe_validation_device val;
 
 	/** @psmi: GPU debugging via additional validation HW */
 	struct {
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 8a6cb92e6a5b..5656da870d6d 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -47,6 +47,20 @@ static struct drm_gem_object *xe_vm_obj(struct xe_vm *vm)
 	return vm->gpuvm.r_obj;
 }
 
+/**
+ * xe_vm_drm_exec_lock() - Lock the vm's resv with a drm_exec transaction
+ * @vm: The vm whose resv is to be locked.
+ * @exec: The drm_exec transaction.
+ *
+ * Helper to lock the vm's resv as part of a drm_exec transaction.
+ *
+ * Return: %0 on success. See drm_exec_lock_obj() for error codes.
+ */
+int xe_vm_drm_exec_lock(struct xe_vm *vm, struct drm_exec *exec)
+{
+	return drm_exec_lock_obj(exec, xe_vm_obj(vm));
+}
+
 static bool preempt_fences_waiting(struct xe_vm *vm)
 {
 	struct xe_exec_queue *q;
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index 2c241f76d5fb..ba6636c1bd58 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -292,6 +292,8 @@ void xe_vm_kill(struct xe_vm *vm, bool unlocked);
  */
 #define xe_vm_assert_held(vm) dma_resv_assert_held(xe_vm_resv(vm))
 
+int xe_vm_drm_exec_lock(struct xe_vm *vm, struct drm_exec *exec);
+
 #if IS_ENABLED(CONFIG_DRM_XE_DEBUG_VM)
 #define vm_dbg drm_dbg
 #else

From 1710cd5c8c1bac8e48eb02cf1123f27a8b90efee Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:37 +0200
Subject: [PATCH 1175/1455] drm/xe: Convert SVM validation for exhaustive
 eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Convert SVM validation to support exhaustive eviction,
using xe_validation_guard().

v2:
- Wrap also xe_vm_range_rebind (Matt Brost)
- Adapt to argument changes of xe_validation_guard().
v5:
- Rebase on SVM stats.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-5-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_svm.c | 109 +++++++++++++++++++-----------------
 1 file changed, 58 insertions(+), 51 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 3a5196b6b72b..8a0fa90a10c0 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -868,51 +868,48 @@ static int xe_drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
 	struct xe_device *xe = vr->xe;
 	struct device *dev = xe->drm.dev;
 	struct drm_buddy_block *block;
+	struct xe_validation_ctx vctx;
 	struct list_head *blocks;
-	struct drm_exec *exec;
+	struct drm_exec exec;
 	struct xe_bo *bo;
-	ktime_t time_end = 0;
-	int err, idx;
+	int err = 0, idx;
 
 	if (!drm_dev_enter(&xe->drm, &idx))
 		return -ENODEV;
 
 	xe_pm_runtime_get(xe);
-	exec = XE_VALIDATION_UNIMPLEMENTED;
 
- retry:
-	bo = xe_bo_create_locked(vr->xe, NULL, NULL, end - start,
-				 ttm_bo_type_device,
-				 (IS_DGFX(xe) ? XE_BO_FLAG_VRAM(vr) : XE_BO_FLAG_SYSTEM) |
-				 XE_BO_FLAG_CPU_ADDR_MIRROR, exec);
-	if (IS_ERR(bo)) {
-		err = PTR_ERR(bo);
-		if (xe_vm_validate_should_retry(NULL, err, &time_end))
-			goto retry;
-		goto out_pm_put;
+	xe_validation_guard(&vctx, &xe->val, &exec, (struct xe_val_flags) {}, err) {
+		bo = xe_bo_create_locked(xe, NULL, NULL, end - start,
+					 ttm_bo_type_device,
+					 (IS_DGFX(xe) ? XE_BO_FLAG_VRAM(vr) : XE_BO_FLAG_SYSTEM) |
+					 XE_BO_FLAG_CPU_ADDR_MIRROR, &exec);
+		drm_exec_retry_on_contention(&exec);
+		if (IS_ERR(bo)) {
+			err = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&vctx, &err);
+			break;
+		}
+
+		drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm,
+					&dpagemap_devmem_ops, dpagemap, end - start);
+
+		blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks;
+		list_for_each_entry(block, blocks, link)
+			block->private = vr;
+
+		xe_bo_get(bo);
+
+		/* Ensure the device has a pm ref while there are device pages active. */
+		xe_pm_runtime_get_noresume(xe);
+		err = drm_pagemap_migrate_to_devmem(&bo->devmem_allocation, mm,
+						    start, end, timeslice_ms,
+						    xe_svm_devm_owner(xe));
+		if (err)
+			xe_svm_devmem_release(&bo->devmem_allocation);
+		xe_bo_unlock(bo);
+		xe_bo_put(bo);
 	}
-
-	drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm,
-				&dpagemap_devmem_ops, dpagemap, end - start);
-
-	blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)->blocks;
-	list_for_each_entry(block, blocks, link)
-		block->private = vr;
-
-	xe_bo_get(bo);
-
-	/* Ensure the device has a pm ref while there are device pages active. */
-	xe_pm_runtime_get_noresume(xe);
-	err = drm_pagemap_migrate_to_devmem(&bo->devmem_allocation, mm,
-					    start, end, timeslice_ms,
-					    xe_svm_devm_owner(xe));
-	if (err)
-		xe_svm_devmem_release(&bo->devmem_allocation);
-
-	xe_bo_unlock(bo);
-	xe_bo_put(bo);
-
-out_pm_put:
 	xe_pm_runtime_put(xe);
 	drm_dev_exit(idx);
 
@@ -1024,12 +1021,13 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ?
 			vm->xe->atomic_svm_timeslice_ms : 0,
 	};
+	struct xe_validation_ctx vctx;
+	struct drm_exec exec;
 	struct xe_svm_range *range;
 	struct dma_fence *fence;
 	struct drm_pagemap *dpagemap;
 	struct xe_tile *tile = gt_to_tile(gt);
 	int migrate_try_count = ctx.devmem_only ? 3 : 1;
-	ktime_t end = 0;
 	ktime_t start = xe_svm_stats_ktime_get(), bind_start, get_pages_start;
 	int err;
 
@@ -1121,22 +1119,23 @@ retry:
 	range_debug(range, "PAGE FAULT - BIND");
 
 	bind_start = xe_svm_stats_ktime_get();
-retry_bind:
-	xe_vm_lock(vm, false);
-	fence = xe_vm_range_rebind(vm, vma, range, BIT(tile->id));
-	if (IS_ERR(fence)) {
-		xe_vm_unlock(vm);
-		err = PTR_ERR(fence);
-		if (err == -EAGAIN) {
-			ctx.timeslice_ms <<= 1;	/* Double timeslice if we have to retry */
-			range_debug(range, "PAGE FAULT - RETRY BIND");
-			goto retry;
+	xe_validation_guard(&vctx, &vm->xe->val, &exec, (struct xe_val_flags) {}, err) {
+		err = xe_vm_drm_exec_lock(vm, &exec);
+		drm_exec_retry_on_contention(&exec);
+
+		xe_vm_set_validation_exec(vm, &exec);
+		fence = xe_vm_range_rebind(vm, vma, range, BIT(tile->id));
+		xe_vm_set_validation_exec(vm, NULL);
+		if (IS_ERR(fence)) {
+			drm_exec_retry_on_contention(&exec);
+			err = PTR_ERR(fence);
+			xe_validation_retry_on_oom(&vctx, &err);
+			xe_svm_range_bind_us_stats_incr(gt, range, bind_start);
+			break;
 		}
-		if (xe_vm_validate_should_retry(NULL, err, &end))
-			goto retry_bind;
-		goto out;
 	}
-	xe_vm_unlock(vm);
+	if (err)
+		goto err_out;
 
 	dma_fence_wait(fence, false);
 	dma_fence_put(fence);
@@ -1144,6 +1143,14 @@ retry_bind:
 
 out:
 	xe_svm_range_fault_us_stats_incr(gt, range, start);
+	return 0;
+
+err_out:
+	if (err == -EAGAIN) {
+		ctx.timeslice_ms <<= 1;	/* Double timeslice if we have to retry */
+		range_debug(range, "PAGE FAULT - RETRY BIND");
+		goto retry;
+	}
 
 	return err;
 }

From 8f25e5abcbfb9cbcd923021774f2e785db3f7f55 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:38 +0200
Subject: [PATCH 1176/1455] drm/xe: Convert existing drm_exec transactions for
 exhaustive eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Convert existing drm_exec transactions, like GT pagefault validation,
non-LR exec() IOCTL and the rebind worker to support
exhaustive eviction using the xe_validation_guard().

v2:
- Adapt to signature change in xe_validation_guard() (Matt Brost)
- Avoid gotos from within xe_validation_guard() (Matt Brost)
- Check error return from xe_validation_guard()

v3:
- Rebase on gpu_madvise()

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com> #v1
Link: https://lore.kernel.org/r/20250908101246.65025-6-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_exec.c         |  20 ++--
 drivers/gpu/drm/xe/xe_gt_pagefault.c |  20 ++--
 drivers/gpu/drm/xe/xe_vm.c           | 139 +++++++++++----------------
 drivers/gpu/drm/xe/xe_vm.h           |   2 -
 4 files changed, 75 insertions(+), 106 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_exec.c b/drivers/gpu/drm/xe/xe_exec.c
index 56edee596352..7715e74bb945 100644
--- a/drivers/gpu/drm/xe/xe_exec.c
+++ b/drivers/gpu/drm/xe/xe_exec.c
@@ -120,10 +120,10 @@ int xe_exec_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
 	struct drm_gpuvm_exec vm_exec = {.extra.fn = xe_exec_fn};
 	struct drm_exec *exec = &vm_exec.exec;
 	u32 i, num_syncs, num_ufence = 0;
+	struct xe_validation_ctx ctx;
 	struct xe_sched_job *job;
 	struct xe_vm *vm;
 	bool write_locked, skip_retry = false;
-	ktime_t end = 0;
 	int err = 0;
 	struct xe_hw_engine_group *group;
 	enum xe_hw_engine_group_execution_mode mode, previous_mode;
@@ -251,17 +251,12 @@ retry:
 	if (err)
 		goto err_unlock_list;
 
-	vm_exec.vm = &vm->gpuvm;
-	vm_exec.flags = DRM_EXEC_INTERRUPTIBLE_WAIT;
-	if (xe_vm_in_lr_mode(vm)) {
-		drm_exec_init(exec, vm_exec.flags, 0);
-	} else {
-		err = drm_gpuvm_exec_lock(&vm_exec);
-		if (err) {
-			if (xe_vm_validate_should_retry(exec, err, &end))
-				err = -EAGAIN;
+	if (!xe_vm_in_lr_mode(vm)) {
+		vm_exec.vm = &vm->gpuvm;
+		vm_exec.flags = DRM_EXEC_INTERRUPTIBLE_WAIT;
+		err = xe_validation_exec_lock(&ctx, &vm_exec, &xe->val);
+		if (err)
 			goto err_unlock_list;
-		}
 	}
 
 	if (xe_vm_is_closed_or_banned(q->vm)) {
@@ -355,7 +350,8 @@ err_put_job:
 	if (err)
 		xe_sched_job_put(job);
 err_exec:
-	drm_exec_fini(exec);
+	if (!xe_vm_in_lr_mode(vm))
+		xe_validation_ctx_fini(&ctx);
 err_unlock_list:
 	up_read(&vm->lock);
 	if (err == -EAGAIN && !skip_retry)
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index b9653ecbc7d4..ec6f6d520a9c 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -96,9 +96,9 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
 {
 	struct xe_vm *vm = xe_vma_vm(vma);
 	struct xe_tile *tile = gt_to_tile(gt);
+	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
 	struct dma_fence *fence;
-	ktime_t end = 0;
 	int err, needs_vram;
 
 	lockdep_assert_held_write(&vm->lock);
@@ -127,12 +127,11 @@ retry_userptr:
 	}
 
 	/* Lock VM and BOs dma-resv */
-	drm_exec_init(&exec, 0, 0);
+	xe_validation_ctx_init(&ctx, &vm->xe->val, &exec, (struct xe_val_flags) {});
 	drm_exec_until_all_locked(&exec) {
 		err = xe_pf_begin(&exec, vma, needs_vram == 1, tile->mem.vram);
 		drm_exec_retry_on_contention(&exec);
-		if (xe_vm_validate_should_retry(&exec, err, &end))
-			err = -EAGAIN;
+		xe_validation_retry_on_oom(&ctx, &err);
 		if (err)
 			goto unlock_dma_resv;
 
@@ -143,8 +142,7 @@ retry_userptr:
 		xe_vm_set_validation_exec(vm, NULL);
 		if (IS_ERR(fence)) {
 			err = PTR_ERR(fence);
-			if (xe_vm_validate_should_retry(&exec, err, &end))
-				err = -EAGAIN;
+			xe_validation_retry_on_oom(&ctx, &err);
 			goto unlock_dma_resv;
 		}
 	}
@@ -153,7 +151,7 @@ retry_userptr:
 	dma_fence_put(fence);
 
 unlock_dma_resv:
-	drm_exec_fini(&exec);
+	xe_validation_ctx_fini(&ctx);
 	if (err == -EAGAIN)
 		goto retry_userptr;
 
@@ -535,6 +533,7 @@ static int handle_acc(struct xe_gt *gt, struct acc *acc)
 {
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_tile *tile = gt_to_tile(gt);
+	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
 	struct xe_vm *vm;
 	struct xe_vma *vma;
@@ -564,15 +563,14 @@ static int handle_acc(struct xe_gt *gt, struct acc *acc)
 		goto unlock_vm;
 
 	/* Lock VM and BOs dma-resv */
-	drm_exec_init(&exec, 0, 0);
+	xe_validation_ctx_init(&ctx, &vm->xe->val, &exec, (struct xe_val_flags) {});
 	drm_exec_until_all_locked(&exec) {
 		ret = xe_pf_begin(&exec, vma, IS_DGFX(vm->xe), tile->mem.vram);
 		drm_exec_retry_on_contention(&exec);
-		if (ret)
-			break;
+		xe_validation_retry_on_oom(&ctx, &ret);
 	}
 
-	drm_exec_fini(&exec);
+	xe_validation_ctx_fini(&ctx);
 unlock_vm:
 	up_read(&vm->lock);
 	xe_vm_put(vm);
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 5656da870d6d..dfe88924fda0 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -210,6 +210,7 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 		.num_fences = 1,
 	};
 	struct drm_exec *exec = &vm_exec.exec;
+	struct xe_validation_ctx ctx;
 	struct dma_fence *pfence;
 	int err;
 	bool wait;
@@ -217,7 +218,7 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 	xe_assert(vm->xe, xe_vm_in_preempt_fence_mode(vm));
 
 	down_write(&vm->lock);
-	err = drm_gpuvm_exec_lock(&vm_exec);
+	err = xe_validation_exec_lock(&ctx, &vm_exec, &vm->xe->val);
 	if (err)
 		goto out_up_write;
 
@@ -249,7 +250,7 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 	xe_svm_notifier_unlock(vm);
 
 out_fini:
-	drm_exec_fini(exec);
+	xe_validation_ctx_fini(&ctx);
 out_up_write:
 	up_write(&vm->lock);
 
@@ -313,39 +314,6 @@ void xe_vm_kill(struct xe_vm *vm, bool unlocked)
 	/* TODO: Inform user the VM is banned */
 }
 
-/**
- * xe_vm_validate_should_retry() - Whether to retry after a validate error.
- * @exec: The drm_exec object used for locking before validation.
- * @err: The error returned from ttm_bo_validate().
- * @end: A ktime_t cookie that should be set to 0 before first use and
- * that should be reused on subsequent calls.
- *
- * With multiple active VMs, under memory pressure, it is possible that
- * ttm_bo_validate() run into -EDEADLK and in such case returns -ENOMEM.
- * Until ttm properly handles locking in such scenarios, best thing the
- * driver can do is retry with a timeout. Check if that is necessary, and
- * if so unlock the drm_exec's objects while keeping the ticket to prepare
- * for a rerun.
- *
- * Return: true if a retry after drm_exec_init() is recommended;
- * false otherwise.
- */
-bool xe_vm_validate_should_retry(struct drm_exec *exec, int err, ktime_t *end)
-{
-	ktime_t cur;
-
-	if (err != -ENOMEM)
-		return false;
-
-	cur = ktime_get();
-	*end = *end ? : ktime_add_ms(cur, XE_VM_REBIND_RETRY_TIMEOUT_MS);
-	if (!ktime_before(cur, *end))
-		return false;
-
-	msleep(20);
-	return true;
-}
-
 static int xe_gpuvm_validate(struct drm_gpuvm_bo *vm_bo, struct drm_exec *exec)
 {
 	struct xe_vm *vm = gpuvm_to_vm(vm_bo->vm);
@@ -476,10 +444,10 @@ void xe_vm_resume_rebind_worker(struct xe_vm *vm)
 static void preempt_rebind_work_func(struct work_struct *w)
 {
 	struct xe_vm *vm = container_of(w, struct xe_vm, preempt.rebind_work);
+	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
 	unsigned int fence_count = 0;
 	LIST_HEAD(preempt_fences);
-	ktime_t end = 0;
 	int err = 0;
 	long wait;
 	int __maybe_unused tries = 0;
@@ -507,18 +475,19 @@ retry:
 			goto out_unlock_outer;
 	}
 
-	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
+	err = xe_validation_ctx_init(&ctx, &vm->xe->val, &exec,
+				     (struct xe_val_flags) {.interruptible = true});
+	if (err)
+		goto out_unlock_outer;
 
 	drm_exec_until_all_locked(&exec) {
 		bool done = false;
 
 		err = xe_preempt_work_begin(&exec, vm, &done);
 		drm_exec_retry_on_contention(&exec);
+		xe_validation_retry_on_oom(&ctx, &err);
 		if (err || done) {
-			drm_exec_fini(&exec);
-			if (err && xe_vm_validate_should_retry(&exec, err, &end))
-				err = -EAGAIN;
-
+			xe_validation_ctx_fini(&ctx);
 			goto out_unlock_outer;
 		}
 	}
@@ -566,7 +535,7 @@ retry:
 	xe_svm_notifier_unlock(vm);
 
 out_unlock:
-	drm_exec_fini(&exec);
+	xe_validation_ctx_fini(&ctx);
 out_unlock_outer:
 	if (err == -EAGAIN) {
 		trace_xe_vm_rebind_worker_retry(vm);
@@ -1164,20 +1133,19 @@ int xe_vm_lock_vma(struct drm_exec *exec, struct xe_vma *vma)
 
 static void xe_vma_destroy_unlocked(struct xe_vma *vma)
 {
+	struct xe_device *xe = xe_vma_vm(vma)->xe;
+	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
-	int err;
+	int err = 0;
 
-	drm_exec_init(&exec, 0, 0);
-	drm_exec_until_all_locked(&exec) {
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {}, err) {
 		err = xe_vm_lock_vma(&exec, vma);
 		drm_exec_retry_on_contention(&exec);
 		if (XE_WARN_ON(err))
 			break;
+		xe_vma_destroy(vma, NULL);
 	}
-
-	xe_vma_destroy(vma, NULL);
-
-	drm_exec_fini(&exec);
+	xe_assert(xe, !err);
 }
 
 struct xe_vma *
@@ -2383,6 +2351,7 @@ static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op,
 			      struct xe_vma_mem_attr *attr, unsigned int flags)
 {
 	struct xe_bo *bo = op->gem.obj ? gem_to_xe_bo(op->gem.obj) : NULL;
+	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
 	struct xe_vma *vma;
 	int err = 0;
@@ -2390,9 +2359,9 @@ static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op,
 	lockdep_assert_held_write(&vm->lock);
 
 	if (bo) {
-		drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT, 0);
-		drm_exec_until_all_locked(&exec) {
-			err = 0;
+		err = 0;
+		xe_validation_guard(&ctx, &vm->xe->val, &exec,
+				    (struct xe_val_flags) {.interruptible = true}, err) {
 			if (!bo->vm) {
 				err = drm_exec_lock_obj(&exec, xe_vm_obj(vm));
 				drm_exec_retry_on_contention(&exec);
@@ -2401,27 +2370,35 @@ static struct xe_vma *new_vma(struct xe_vm *vm, struct drm_gpuva_op_map *op,
 				err = drm_exec_lock_obj(&exec, &bo->ttm.base);
 				drm_exec_retry_on_contention(&exec);
 			}
-			if (err) {
-				drm_exec_fini(&exec);
+			if (err)
 				return ERR_PTR(err);
+
+			vma = xe_vma_create(vm, bo, op->gem.offset,
+					    op->va.addr, op->va.addr +
+					    op->va.range - 1, attr, flags);
+			if (IS_ERR(vma))
+				return vma;
+
+			if (!bo->vm) {
+				err = add_preempt_fences(vm, bo);
+				if (err) {
+					prep_vma_destroy(vm, vma, false);
+					xe_vma_destroy(vma, NULL);
+				}
 			}
 		}
+		if (err)
+			return ERR_PTR(err);
+	} else {
+		vma = xe_vma_create(vm, NULL, op->gem.offset,
+				    op->va.addr, op->va.addr +
+				    op->va.range - 1, attr, flags);
+		if (IS_ERR(vma))
+			return vma;
+
+		if (xe_vma_is_userptr(vma))
+			err = xe_vma_userptr_pin_pages(to_userptr_vma(vma));
 	}
-	vma = xe_vma_create(vm, bo, op->gem.offset,
-			    op->va.addr, op->va.addr +
-			    op->va.range - 1, attr, flags);
-	if (IS_ERR(vma))
-		goto err_unlock;
-
-	if (xe_vma_is_userptr(vma))
-		err = xe_vma_userptr_pin_pages(to_userptr_vma(vma));
-	else if (!xe_vma_has_no_bo(vma) && !bo->vm)
-		err = add_preempt_fences(vm, bo);
-
-err_unlock:
-	if (bo)
-		drm_exec_fini(&exec);
-
 	if (err) {
 		prep_vma_destroy(vm, vma, false);
 		xe_vma_destroy_unlocked(vma);
@@ -3220,21 +3197,23 @@ static void vm_bind_ioctl_ops_fini(struct xe_vm *vm, struct xe_vma_ops *vops,
 static struct dma_fence *vm_bind_ioctl_ops_execute(struct xe_vm *vm,
 						   struct xe_vma_ops *vops)
 {
+	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
 	struct dma_fence *fence;
-	int err;
+	int err = 0;
 
 	lockdep_assert_held_write(&vm->lock);
 
-	drm_exec_init(&exec, DRM_EXEC_INTERRUPTIBLE_WAIT |
-		      DRM_EXEC_IGNORE_DUPLICATES, 0);
-	drm_exec_until_all_locked(&exec) {
+	xe_validation_guard(&ctx, &vm->xe->val, &exec,
+			    ((struct xe_val_flags) {
+				    .interruptible = true,
+				    .exec_ignore_duplicates = true,
+			    }), err) {
 		err = vm_bind_ioctl_ops_lock_and_prep(&exec, vm, vops);
 		drm_exec_retry_on_contention(&exec);
-		if (err) {
-			fence = ERR_PTR(err);
-			goto unlock;
-		}
+		xe_validation_retry_on_oom(&ctx, &err);
+		if (err)
+			return ERR_PTR(err);
 
 		xe_vm_set_validation_exec(vm, &exec);
 		fence = ops_execute(vm, vops);
@@ -3242,15 +3221,13 @@ static struct dma_fence *vm_bind_ioctl_ops_execute(struct xe_vm *vm,
 		if (IS_ERR(fence)) {
 			if (PTR_ERR(fence) == -ENODATA)
 				vm_bind_ioctl_ops_fini(vm, vops, NULL);
-			goto unlock;
+			return fence;
 		}
 
 		vm_bind_ioctl_ops_fini(vm, vops, fence);
 	}
 
-unlock:
-	drm_exec_fini(&exec);
-	return fence;
+	return err ? ERR_PTR(err) : fence;
 }
 ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO);
 
diff --git a/drivers/gpu/drm/xe/xe_vm.h b/drivers/gpu/drm/xe/xe_vm.h
index ba6636c1bd58..ef8a5019574e 100644
--- a/drivers/gpu/drm/xe/xe_vm.h
+++ b/drivers/gpu/drm/xe/xe_vm.h
@@ -260,8 +260,6 @@ static inline void xe_vm_reactivate_rebind(struct xe_vm *vm)
 	}
 }
 
-bool xe_vm_validate_should_retry(struct drm_exec *exec, int err, ktime_t *end);
-
 int xe_vm_lock_vma(struct drm_exec *exec, struct xe_vma *vma);
 
 int xe_vm_validate_rebind(struct xe_vm *vm, struct drm_exec *exec,

From c2ae94cf8cd86cd0a417865d194fb6c4cc81bfb1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:39 +0200
Subject: [PATCH 1177/1455] drm/xe: Convert the CPU fault handler for
 exhaustive eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The CPU fault handler may populate bos and migrate, and in doing
so might interfere with other tasks validating.

Rework the CPU fault handler completely into a fastpath
and a slowpath. The fastpath trylocks only the validation lock
in read-mode. If that fails, there's a fallback to the
slowpath, where we do a full validation transaction.

This mandates open-coding of bo locking, bo idling and
bo populating, but we still call into TTM for fault
finalizing.

v2:
- Rework the CPU fault handler to actually take part in
  the exhaustive eviction scheme (Matthew Brost).
v3:
- Don't return anything but VM_FAULT_RETRY if we've dropped the
  mmap_lock. Not even if a signal is pending.
- Rebase on gpu_madvise() and split out fault migration.
- Wait for idle after migration.
- Check whether the resource manager uses tts to determine
  whether to map the tt or iomem.
- Add a number of asserts.
- Allow passing a ttm_operation_ctx to xe_bo_migrate() so that
  it's possible to try non-blocking migration.
- Don't fall through to TTM on migration / population error
  Instead remove the gfp_retry_mayfail in mode 2 where we
  must succeed. (Matthew Brost)
v5:
- Don't allow faulting in the imported bo case (Matthew Brost)

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthews Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-7-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/display/xe_fb_pin.c |   2 +-
 drivers/gpu/drm/xe/xe_bo.c             | 261 ++++++++++++++++++++-----
 drivers/gpu/drm/xe/xe_bo.h             |   3 +-
 drivers/gpu/drm/xe/xe_dma_buf.c        |   6 +-
 drivers/gpu/drm/xe/xe_gt_pagefault.c   |   2 +-
 drivers/gpu/drm/xe/xe_validation.c     |   3 +-
 drivers/gpu/drm/xe/xe_vm.c             |   1 +
 7 files changed, 226 insertions(+), 52 deletions(-)

diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index 1c5b4e1f1fd8..3406dca285bd 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -314,7 +314,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 		goto err;
 
 	if (IS_DGFX(xe))
-		ret = xe_bo_migrate(bo, XE_PL_VRAM0, exec);
+		ret = xe_bo_migrate(bo, XE_PL_VRAM0, NULL, exec);
 	else
 		ret = xe_bo_validate(bo, NULL, true, exec);
 	if (!ret)
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index ac330159e339..be10673e5c77 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1716,68 +1716,234 @@ static bool should_migrate_to_smem(struct xe_bo *bo)
 	       bo->attr.atomic_access == DRM_XE_ATOMIC_CPU;
 }
 
-static vm_fault_t xe_gem_fault(struct vm_fault *vmf)
+/* Populate the bo if swapped out, or migrate if the access mode requires that. */
+static int xe_bo_fault_migrate(struct xe_bo *bo, struct ttm_operation_ctx *ctx,
+			       struct drm_exec *exec)
+{
+	struct ttm_buffer_object *tbo = &bo->ttm;
+	int err = 0;
+
+	if (ttm_manager_type(tbo->bdev, tbo->resource->mem_type)->use_tt) {
+		xe_assert(xe_bo_device(bo),
+			  dma_resv_test_signaled(tbo->base.resv, DMA_RESV_USAGE_KERNEL) ||
+			  (tbo->ttm && ttm_tt_is_populated(tbo->ttm)));
+		err = ttm_bo_populate(&bo->ttm, ctx);
+	} else if (should_migrate_to_smem(bo)) {
+		xe_assert(xe_bo_device(bo), bo->flags & XE_BO_FLAG_SYSTEM);
+		err = xe_bo_migrate(bo, XE_PL_TT, ctx, exec);
+	}
+
+	return err;
+}
+
+/* Call into TTM to populate PTEs, and register bo for PTE removal on runtime suspend. */
+static vm_fault_t __xe_bo_cpu_fault(struct vm_fault *vmf, struct xe_device *xe, struct xe_bo *bo)
+{
+	vm_fault_t ret;
+
+	trace_xe_bo_cpu_fault(bo);
+
+	ret = ttm_bo_vm_fault_reserved(vmf, vmf->vma->vm_page_prot,
+				       TTM_BO_VM_NUM_PREFAULT);
+	/*
+	 * When TTM is actually called to insert PTEs, ensure no blocking conditions
+	 * remain, in which case TTM may drop locks and return VM_FAULT_RETRY.
+	 */
+	xe_assert(xe, ret != VM_FAULT_RETRY);
+
+	if (ret == VM_FAULT_NOPAGE &&
+	    mem_type_is_vram(bo->ttm.resource->mem_type)) {
+		mutex_lock(&xe->mem_access.vram_userfault.lock);
+		if (list_empty(&bo->vram_userfault_link))
+			list_add(&bo->vram_userfault_link,
+				 &xe->mem_access.vram_userfault.list);
+		mutex_unlock(&xe->mem_access.vram_userfault.lock);
+	}
+
+	return ret;
+}
+
+static vm_fault_t xe_err_to_fault_t(int err)
+{
+	switch (err) {
+	case 0:
+	case -EINTR:
+	case -ERESTARTSYS:
+	case -EAGAIN:
+		return VM_FAULT_NOPAGE;
+	case -ENOMEM:
+	case -ENOSPC:
+		return VM_FAULT_OOM;
+	default:
+		break;
+	}
+	return VM_FAULT_SIGBUS;
+}
+
+static bool xe_ttm_bo_is_imported(struct ttm_buffer_object *tbo)
+{
+	dma_resv_assert_held(tbo->base.resv);
+
+	return tbo->ttm &&
+		(tbo->ttm->page_flags & (TTM_TT_FLAG_EXTERNAL | TTM_TT_FLAG_EXTERNAL_MAPPABLE)) ==
+		TTM_TT_FLAG_EXTERNAL;
+}
+
+static vm_fault_t xe_bo_cpu_fault_fastpath(struct vm_fault *vmf, struct xe_device *xe,
+					   struct xe_bo *bo, bool needs_rpm)
+{
+	struct ttm_buffer_object *tbo = &bo->ttm;
+	vm_fault_t ret = VM_FAULT_RETRY;
+	struct xe_validation_ctx ctx;
+	struct ttm_operation_ctx tctx = {
+		.interruptible = true,
+		.no_wait_gpu = true,
+		.gfp_retry_mayfail = true,
+
+	};
+	int err;
+
+	if (needs_rpm && !xe_pm_runtime_get_if_active(xe))
+		return VM_FAULT_RETRY;
+
+	err = xe_validation_ctx_init(&ctx, &xe->val, NULL,
+				     (struct xe_val_flags) {
+					     .interruptible = true,
+					     .no_block = true
+				     });
+	if (err)
+		goto out_pm;
+
+	if (!dma_resv_trylock(tbo->base.resv))
+		goto out_validation;
+
+	if (xe_ttm_bo_is_imported(tbo)) {
+		ret = VM_FAULT_SIGBUS;
+		drm_dbg(&xe->drm, "CPU trying to access an imported buffer object.\n");
+		goto out_unlock;
+	}
+
+	err = xe_bo_fault_migrate(bo, &tctx, NULL);
+	if (err) {
+		/* Return VM_FAULT_RETRY on these errors. */
+		if (err != -ENOMEM && err != -ENOSPC && err != -EBUSY)
+			ret = xe_err_to_fault_t(err);
+		goto out_unlock;
+	}
+
+	if (dma_resv_test_signaled(bo->ttm.base.resv, DMA_RESV_USAGE_KERNEL))
+		ret = __xe_bo_cpu_fault(vmf, xe, bo);
+
+out_unlock:
+	dma_resv_unlock(tbo->base.resv);
+out_validation:
+	xe_validation_ctx_fini(&ctx);
+out_pm:
+	if (needs_rpm)
+		xe_pm_runtime_put(xe);
+
+	return ret;
+}
+
+static vm_fault_t xe_bo_cpu_fault(struct vm_fault *vmf)
 {
 	struct ttm_buffer_object *tbo = vmf->vma->vm_private_data;
 	struct drm_device *ddev = tbo->base.dev;
 	struct xe_device *xe = to_xe_device(ddev);
 	struct xe_bo *bo = ttm_to_xe_bo(tbo);
 	bool needs_rpm = bo->flags & XE_BO_FLAG_VRAM_MASK;
-	struct drm_exec *exec;
+	bool retry_after_wait = false;
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
 	vm_fault_t ret;
-	int idx, r = 0;
+	int err = 0;
+	int idx;
+
+	if (!drm_dev_enter(&xe->drm, &idx))
+		return ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot);
+
+	ret = xe_bo_cpu_fault_fastpath(vmf, xe, bo, needs_rpm);
+	if (ret != VM_FAULT_RETRY)
+		goto out;
+
+	if (fault_flag_allow_retry_first(vmf->flags)) {
+		if (vmf->flags & FAULT_FLAG_RETRY_NOWAIT)
+			goto out;
+		retry_after_wait = true;
+		xe_bo_get(bo);
+		mmap_read_unlock(vmf->vma->vm_mm);
+	} else {
+		ret = VM_FAULT_NOPAGE;
+	}
+
+	/*
+	 * The fastpath failed and we were not required to return and retry immediately.
+	 * We're now running in one of two modes:
+	 *
+	 * 1) retry_after_wait == true: The mmap_read_lock() is dropped, and we're trying
+	 * to resolve blocking waits. But we can't resolve the fault since the
+	 * mmap_read_lock() is dropped. After retrying the fault, the aim is that the fastpath
+	 * should succeed. But it may fail since we drop the bo lock.
+	 *
+	 * 2) retry_after_wait == false: The fastpath failed, typically even after
+	 * a retry. Do whatever's necessary to resolve the fault.
+	 *
+	 * This construct is recommended to avoid excessive waits under the mmap_lock.
+	 */
 
 	if (needs_rpm)
 		xe_pm_runtime_get(xe);
 
-	exec = XE_VALIDATION_UNIMPLEMENTED;
-	ret = ttm_bo_vm_reserve(tbo, vmf);
-	if (ret)
-		goto out;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.interruptible = true},
+			    err) {
+		struct ttm_operation_ctx tctx = {
+			.interruptible = true,
+			.no_wait_gpu = false,
+			.gfp_retry_mayfail = retry_after_wait,
+		};
+		long lerr;
 
-	if (drm_dev_enter(ddev, &idx)) {
-		trace_xe_bo_cpu_fault(bo);
+		err = drm_exec_lock_obj(&exec, &tbo->base);
+		drm_exec_retry_on_contention(&exec);
+		if (err)
+			break;
 
-		xe_validation_assert_exec(xe, exec, &tbo->base);
-		if (should_migrate_to_smem(bo)) {
-			xe_assert(xe, bo->flags & XE_BO_FLAG_SYSTEM);
-
-			r = xe_bo_migrate(bo, XE_PL_TT, exec);
-			if (r == -EBUSY || r == -ERESTARTSYS || r == -EINTR)
-				ret = VM_FAULT_NOPAGE;
-			else if (r)
-				ret = VM_FAULT_SIGBUS;
+		if (xe_ttm_bo_is_imported(tbo)) {
+			err = -EFAULT;
+			drm_dbg(&xe->drm, "CPU trying to access an imported buffer object.\n");
+			break;
 		}
-		if (!ret)
-			ret = ttm_bo_vm_fault_reserved(vmf,
-						       vmf->vma->vm_page_prot,
-						       TTM_BO_VM_NUM_PREFAULT);
-		drm_dev_exit(idx);
 
-		if (ret == VM_FAULT_RETRY &&
-		    !(vmf->flags & FAULT_FLAG_RETRY_NOWAIT))
-			goto out;
-
-		/*
-		 * ttm_bo_vm_reserve() already has dma_resv_lock.
-		 */
-		if (ret == VM_FAULT_NOPAGE &&
-		    mem_type_is_vram(tbo->resource->mem_type)) {
-			mutex_lock(&xe->mem_access.vram_userfault.lock);
-			if (list_empty(&bo->vram_userfault_link))
-				list_add(&bo->vram_userfault_link,
-					 &xe->mem_access.vram_userfault.list);
-			mutex_unlock(&xe->mem_access.vram_userfault.lock);
+		err = xe_bo_fault_migrate(bo, &tctx, &exec);
+		if (err) {
+			drm_exec_retry_on_contention(&exec);
+			xe_validation_retry_on_oom(&ctx, &err);
+			break;
 		}
-	} else {
-		ret = ttm_bo_vm_dummy_page(vmf, vmf->vma->vm_page_prot);
+
+		lerr = dma_resv_wait_timeout(tbo->base.resv,
+					     DMA_RESV_USAGE_KERNEL, true,
+					     MAX_SCHEDULE_TIMEOUT);
+		if (lerr < 0) {
+			err = lerr;
+			break;
+		}
+
+		if (!retry_after_wait)
+			ret = __xe_bo_cpu_fault(vmf, xe, bo);
 	}
+	/* if retry_after_wait == true, we *must* return VM_FAULT_RETRY. */
+	if (err && !retry_after_wait)
+		ret = xe_err_to_fault_t(err);
 
-	dma_resv_unlock(tbo->base.resv);
-out:
 	if (needs_rpm)
 		xe_pm_runtime_put(xe);
 
+	if (retry_after_wait)
+		xe_bo_put(bo);
+out:
+	drm_dev_exit(idx);
+
 	return ret;
 }
 
@@ -1821,7 +1987,7 @@ int xe_bo_read(struct xe_bo *bo, u64 offset, void *dst, int size)
 }
 
 static const struct vm_operations_struct xe_gem_vm_ops = {
-	.fault = xe_gem_fault,
+	.fault = xe_bo_cpu_fault,
 	.open = ttm_bo_vm_open,
 	.close = ttm_bo_vm_close,
 	.access = xe_bo_vm_access,
@@ -3057,6 +3223,8 @@ static void xe_place_from_ttm_type(u32 mem_type, struct ttm_place *place)
  * xe_bo_migrate - Migrate an object to the desired region id
  * @bo: The buffer object to migrate.
  * @mem_type: The TTM region type to migrate to.
+ * @tctx: A pointer to a struct ttm_operation_ctx or NULL if
+ * a default interruptibe ctx is to be used.
  * @exec: The drm_exec transaction to use for exhaustive eviction.
  *
  * Attempt to migrate the buffer object to the desired memory region. The
@@ -3069,7 +3237,8 @@ static void xe_place_from_ttm_type(u32 mem_type, struct ttm_place *place)
  * Return: 0 on success. Negative error code on failure. In particular may
  * return -EINTR or -ERESTARTSYS if signal pending.
  */
-int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct drm_exec *exec)
+int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct ttm_operation_ctx *tctx,
+		  struct drm_exec *exec)
 {
 	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
 	struct ttm_operation_ctx ctx = {
@@ -3081,6 +3250,7 @@ int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct drm_exec *exec)
 	struct ttm_place requested;
 
 	xe_bo_assert_held(bo);
+	tctx = tctx ? tctx : &ctx;
 
 	if (bo->ttm.resource->mem_type == mem_type)
 		return 0;
@@ -3107,8 +3277,9 @@ int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct drm_exec *exec)
 		add_vram(xe, bo, &requested, bo->flags, mem_type, &c);
 	}
 
-	xe_validation_assert_exec(xe_bo_device(bo), exec, &bo->ttm.base);
-	return ttm_bo_validate(&bo->ttm, &placement, &ctx);
+	if (!tctx->no_wait_gpu)
+		xe_validation_assert_exec(xe_bo_device(bo), exec, &bo->ttm.base);
+	return ttm_bo_validate(&bo->ttm, &placement, tctx);
 }
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 3878df635b03..096b5c71554a 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -284,7 +284,8 @@ uint64_t vram_region_gpu_offset(struct ttm_resource *res);
 
 bool xe_bo_can_migrate(struct xe_bo *bo, u32 mem_type);
 
-int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct drm_exec *exec);
+int xe_bo_migrate(struct xe_bo *bo, u32 mem_type, struct ttm_operation_ctx *ctc,
+		  struct drm_exec *exec);
 int xe_bo_evict(struct xe_bo *bo, struct drm_exec *exec);
 
 int xe_bo_evict_pinned(struct xe_bo *bo);
diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index 199041bdb882..7c2793265561 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -64,7 +64,7 @@ static int xe_dma_buf_pin(struct dma_buf_attachment *attach)
 		return -EINVAL;
 	}
 
-	ret = xe_bo_migrate(bo, XE_PL_TT, exec);
+	ret = xe_bo_migrate(bo, XE_PL_TT, NULL, exec);
 	if (ret) {
 		if (ret != -EINTR && ret != -ERESTARTSYS)
 			drm_dbg(&xe->drm,
@@ -102,7 +102,7 @@ static struct sg_table *xe_dma_buf_map(struct dma_buf_attachment *attach,
 
 	if (!xe_bo_is_pinned(bo)) {
 		if (!attach->peer2peer)
-			r = xe_bo_migrate(bo, XE_PL_TT, exec);
+			r = xe_bo_migrate(bo, XE_PL_TT, NULL, exec);
 		else
 			r = xe_bo_validate(bo, NULL, false, exec);
 		if (r)
@@ -170,7 +170,7 @@ static int xe_dma_buf_begin_cpu_access(struct dma_buf *dma_buf,
 
 	/* Can we do interruptible lock here? */
 	xe_bo_lock(bo, false);
-	(void)xe_bo_migrate(bo, XE_PL_TT, exec);
+	(void)xe_bo_migrate(bo, XE_PL_TT, NULL, exec);
 	xe_bo_unlock(bo);
 
 	return 0;
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index ec6f6d520a9c..a054d6010ae0 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -87,7 +87,7 @@ static int xe_pf_begin(struct drm_exec *exec, struct xe_vma *vma,
 	if (!bo)
 		return 0;
 
-	return need_vram_move ? xe_bo_migrate(bo, vram->placement, exec) :
+	return need_vram_move ? xe_bo_migrate(bo, vram->placement, NULL, exec) :
 		xe_bo_validate(bo, vm, true, exec);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_validation.c b/drivers/gpu/drm/xe/xe_validation.c
index b90fda3dd5f4..826cd09966ef 100644
--- a/drivers/gpu/drm/xe/xe_validation.c
+++ b/drivers/gpu/drm/xe/xe_validation.c
@@ -241,7 +241,8 @@ retry:
  */
 void xe_validation_ctx_fini(struct xe_validation_ctx *ctx)
 {
-	drm_exec_fini(ctx->exec);
+	if (ctx->exec)
+		drm_exec_fini(ctx->exec);
 	xe_validation_unlock(ctx);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index dfe88924fda0..785e81cf023d 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -2934,6 +2934,7 @@ static int op_lock_and_prep(struct drm_exec *exec, struct xe_vm *vm,
 		if (!err && !xe_vma_has_no_bo(vma))
 			err = xe_bo_migrate(xe_vma_bo(vma),
 					    region_to_mem_type[region],
+					    NULL,
 					    exec);
 		break;
 	}

From 7bcb6e38c14d7d2654555a1b05b2a9a5411c693c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:40 +0200
Subject: [PATCH 1178/1455] drm/xe/display: Convert __xe_pin_fb_vma()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Convert __xe_pin_fb_vma() for exhaustive eviction
using xe_validation_guard().

v2:
- Avoid gotos from within xe_validation_guard(). (Matt Brost)
- Adapt to signature change of xe_validation_guard(). (Matt Brost)
- Use interruptible waiting, since xe_bo_migrate() already does that.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-8-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/display/xe_fb_pin.c | 28 ++++++++++++++++----------
 1 file changed, 17 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index 3406dca285bd..a5c35e9ff2ef 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -281,7 +281,8 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 	struct i915_vma *vma = kzalloc(sizeof(*vma), GFP_KERNEL);
 	struct drm_gem_object *obj = intel_fb_bo(&fb->base);
 	struct xe_bo *bo = gem_to_xe_bo(obj);
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
 	int ret;
 
 	if (!vma)
@@ -309,17 +310,22 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 	 * Pin the framebuffer, we can't use xe_bo_(un)pin functions as the
 	 * assumptions are incorrect for framebuffers
 	 */
-	ret = ttm_bo_reserve(&bo->ttm, false, false, NULL);
-	if (ret)
-		goto err;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.interruptible = true},
+			    ret) {
+		ret = drm_exec_lock_obj(&exec, &bo->ttm.base);
+		drm_exec_retry_on_contention(&exec);
+		if (ret)
+			break;
 
-	if (IS_DGFX(xe))
-		ret = xe_bo_migrate(bo, XE_PL_VRAM0, NULL, exec);
-	else
-		ret = xe_bo_validate(bo, NULL, true, exec);
-	if (!ret)
-		ttm_bo_pin(&bo->ttm);
-	ttm_bo_unreserve(&bo->ttm);
+		if (IS_DGFX(xe))
+			ret = xe_bo_migrate(bo, XE_PL_VRAM0, NULL, &exec);
+		else
+			ret = xe_bo_validate(bo, NULL, true, &exec);
+		drm_exec_retry_on_contention(&exec);
+		xe_validation_retry_on_oom(&ctx, &ret);
+		if (!ret)
+			ttm_bo_pin(&bo->ttm);
+	}
 	if (ret)
 		goto err;
 

From eb289a5f6cc668853f9b2ea6aca04afe58ed11c7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:41 +0200
Subject: [PATCH 1179/1455] drm/xe: Convert xe_dma_buf.c for exhaustive
 eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Convert dma-buf migration to XE_PL_TT and dma-buf import to
support exhaustive eviction, using xe_validation_guard().
It seems unlikely that the import would result in an -ENOMEM,
but convert import anyway for completeness.

The dma-buf map_attachment() functionality unfortunately doesn't
support passing a drm_exec, which means that foreign devices
validating a dma-buf that we exported will not, unless they are
xeKMD devices, participate in the exhaustive eviction scheme.

v2:
- Avoid gotos from within xe_validation_guard(). (Matt Brost)
- Adapt to signature change of xe_validation_guard(). (Matt Brost)
- Remove an unneeded (void)ret. (Matt Brost)
- Fix up an error path.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-9-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_dma_buf.c | 59 ++++++++++++++++++++++-----------
 1 file changed, 40 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index 7c2793265561..2e5b4b37118f 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -163,16 +163,26 @@ static int xe_dma_buf_begin_cpu_access(struct dma_buf *dma_buf,
 	struct xe_bo *bo = gem_to_xe_bo(obj);
 	bool reads =  (direction == DMA_BIDIRECTIONAL ||
 		       direction == DMA_FROM_DEVICE);
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
+	int ret = 0;
 
 	if (!reads)
 		return 0;
 
 	/* Can we do interruptible lock here? */
-	xe_bo_lock(bo, false);
-	(void)xe_bo_migrate(bo, XE_PL_TT, NULL, exec);
-	xe_bo_unlock(bo);
+	xe_validation_guard(&ctx, &xe_bo_device(bo)->val, &exec, (struct xe_val_flags) {}, ret) {
+		ret = drm_exec_lock_obj(&exec, &bo->ttm.base);
+		drm_exec_retry_on_contention(&exec);
+		if (ret)
+			break;
 
+		ret = xe_bo_migrate(bo, XE_PL_TT, NULL, &exec);
+		drm_exec_retry_on_contention(&exec);
+		xe_validation_retry_on_oom(&ctx, &ret);
+	}
+
+	/* If we failed, cpu-access takes place in current placement. */
 	return 0;
 }
 
@@ -211,25 +221,36 @@ xe_dma_buf_init_obj(struct drm_device *dev, struct xe_bo *storage,
 {
 	struct dma_resv *resv = dma_buf->resv;
 	struct xe_device *xe = to_xe_device(dev);
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
+	struct xe_validation_ctx ctx;
+	struct drm_gem_object *dummy_obj;
+	struct drm_exec exec;
 	struct xe_bo *bo;
-	int ret;
+	int ret = 0;
 
-	dma_resv_lock(resv, NULL);
-	bo = ___xe_bo_create_locked(xe, storage, NULL, resv, NULL, dma_buf->size,
-				    0, /* Will require 1way or 2way for vm_bind */
-				    ttm_bo_type_sg, XE_BO_FLAG_SYSTEM, exec);
-	if (IS_ERR(bo)) {
-		ret = PTR_ERR(bo);
-		goto error;
+	dummy_obj = drm_gpuvm_resv_object_alloc(&xe->drm);
+	if (!dummy_obj)
+		return ERR_PTR(-ENOMEM);
+
+	dummy_obj->resv = resv;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {}, ret) {
+		ret = drm_exec_lock_obj(&exec, dummy_obj);
+		drm_exec_retry_on_contention(&exec);
+		if (ret)
+			break;
+
+		bo = ___xe_bo_create_locked(xe, storage, NULL, resv, NULL, dma_buf->size,
+					    0, /* Will require 1way or 2way for vm_bind */
+					    ttm_bo_type_sg, XE_BO_FLAG_SYSTEM, &exec);
+		drm_exec_retry_on_contention(&exec);
+		if (IS_ERR(bo)) {
+			ret = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&ctx, &ret);
+			break;
+		}
 	}
-	dma_resv_unlock(resv);
+	drm_gem_object_put(dummy_obj);
 
-	return &bo->ttm.base;
-
-error:
-	dma_resv_unlock(resv);
-	return ERR_PTR(ret);
+	return ret ? ERR_PTR(ret) : &bo->ttm.base;
 }
 
 static void xe_dma_buf_move_notify(struct dma_buf_attachment *attach)

From 550a42a8daee1b056f9b3e6c858e57451a5b315a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:42 +0200
Subject: [PATCH 1180/1455] drm/xe: Rename ___xe_bo_create_locked()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Don't start external function names with underscores.
Rename to xe_bo_init_locked().

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-10-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_bo.c      | 39 ++++++++++++++++-----------------
 drivers/gpu/drm/xe/xe_bo.h      | 10 ++++-----
 drivers/gpu/drm/xe/xe_dma_buf.c |  6 ++---
 3 files changed, 27 insertions(+), 28 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index be10673e5c77..ad2863a7df77 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1163,10 +1163,10 @@ int xe_bo_notifier_prepare_pinned(struct xe_bo *bo)
 	if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
 		goto out_unlock_bo;
 
-	backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
-					DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
-					XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-					XE_BO_FLAG_PINNED, exec);
+	backup = xe_bo_init_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
+				   DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
+				   XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
+				   XE_BO_FLAG_PINNED, exec);
 	if (IS_ERR(backup)) {
 		ret = PTR_ERR(backup);
 		goto out_unlock_bo;
@@ -1242,11 +1242,10 @@ int xe_bo_evict_pinned(struct xe_bo *bo)
 		goto out_unlock_bo;
 
 	if (!backup) {
-		backup = ___xe_bo_create_locked(xe, NULL, NULL, bo->ttm.base.resv,
-						NULL, xe_bo_size(bo),
-						DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
-						XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-						XE_BO_FLAG_PINNED, exec);
+		backup = xe_bo_init_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
+					   DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
+					   XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
+					   XE_BO_FLAG_PINNED, exec);
 		if (IS_ERR(backup)) {
 			ret = PTR_ERR(backup);
 			goto out_unlock_bo;
@@ -2036,7 +2035,7 @@ void xe_bo_free(struct xe_bo *bo)
 }
 
 /**
- * ___xe_bo_create_locked() - Initialize or create an xe_bo.
+ * xe_bo_init_locked() - Initialize or create an xe_bo.
  * @xe: The xe device.
  * @bo: An already allocated buffer object or NULL
  * if the function should allocate a new one.
@@ -2056,11 +2055,11 @@ void xe_bo_free(struct xe_bo *bo)
  *
  * Return: The buffer object on success. Negative error pointer on failure.
  */
-struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo,
-				     struct xe_tile *tile, struct dma_resv *resv,
-				     struct ttm_lru_bulk_move *bulk, size_t size,
-				     u16 cpu_caching, enum ttm_bo_type type,
-				     u32 flags, struct drm_exec *exec)
+struct xe_bo *xe_bo_init_locked(struct xe_device *xe, struct xe_bo *bo,
+				struct xe_tile *tile, struct dma_resv *resv,
+				struct ttm_lru_bulk_move *bulk, size_t size,
+				u16 cpu_caching, enum ttm_bo_type type,
+				u32 flags, struct drm_exec *exec)
 {
 	struct ttm_operation_ctx ctx = {
 		.interruptible = true,
@@ -2252,11 +2251,11 @@ __xe_bo_create_locked(struct xe_device *xe,
 		}
 	}
 
-	bo = ___xe_bo_create_locked(xe, bo, tile, vm ? xe_vm_resv(vm) : NULL,
-				    vm && !xe_vm_in_fault_mode(vm) &&
-				    flags & XE_BO_FLAG_USER ?
-				    &vm->lru_bulk_move : NULL, size,
-				    cpu_caching, type, flags, exec);
+	bo = xe_bo_init_locked(xe, bo, tile, vm ? xe_vm_resv(vm) : NULL,
+			       vm && !xe_vm_in_fault_mode(vm) &&
+			       flags & XE_BO_FLAG_USER ?
+			       &vm->lru_bulk_move : NULL, size,
+			       cpu_caching, type, flags, exec);
 	if (IS_ERR(bo))
 		return bo;
 
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 096b5c71554a..ebca6e9f6959 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -89,11 +89,11 @@ struct sg_table;
 struct xe_bo *xe_bo_alloc(void);
 void xe_bo_free(struct xe_bo *bo);
 
-struct xe_bo *___xe_bo_create_locked(struct xe_device *xe, struct xe_bo *bo,
-				     struct xe_tile *tile, struct dma_resv *resv,
-				     struct ttm_lru_bulk_move *bulk, size_t size,
-				     u16 cpu_caching, enum ttm_bo_type type,
-				     u32 flags, struct drm_exec *exec);
+struct xe_bo *xe_bo_init_locked(struct xe_device *xe, struct xe_bo *bo,
+				struct xe_tile *tile, struct dma_resv *resv,
+				struct ttm_lru_bulk_move *bulk, size_t size,
+				u16 cpu_caching, enum ttm_bo_type type,
+				u32 flags, struct drm_exec *exec);
 struct xe_bo *
 xe_bo_create_locked_range(struct xe_device *xe,
 			  struct xe_tile *tile, struct xe_vm *vm,
diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index 2e5b4b37118f..607c3f4ef3b9 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -238,9 +238,9 @@ xe_dma_buf_init_obj(struct drm_device *dev, struct xe_bo *storage,
 		if (ret)
 			break;
 
-		bo = ___xe_bo_create_locked(xe, storage, NULL, resv, NULL, dma_buf->size,
-					    0, /* Will require 1way or 2way for vm_bind */
-					    ttm_bo_type_sg, XE_BO_FLAG_SYSTEM, &exec);
+		bo = xe_bo_init_locked(xe, storage, NULL, resv, NULL, dma_buf->size,
+				       0, /* Will require 1way or 2way for vm_bind */
+				       ttm_bo_type_sg, XE_BO_FLAG_SYSTEM, &exec);
 		drm_exec_retry_on_contention(&exec);
 		if (IS_ERR(bo)) {
 			ret = PTR_ERR(bo);

From e6108eade1b6d8a5ebe0c872fae646df10b024f8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:43 +0200
Subject: [PATCH 1181/1455] drm/xe: Convert xe_bo_create_pin_map_at() for
 exhaustive eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Most users of xe_bo_create_pin_map_at() and
xe_bo_create_pin_map_at_aligned() are not using the vm parameter,
and that simplifies conversion. Introduce an
xe_bo_create_pin_map_at_novm() function and make the _aligned()
version static. Use xe_validation_guard() for conversion.

v2:
- Adapt to signature change of xe_validation_guard(). (Matt Brost)
- Fix up documentation.
v4:
- Postpone the change to i915_gem_stolen_insert_node_in_range() to
  a later patch.

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-11-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/display/xe_fb_pin.c        | 42 +++++------
 drivers/gpu/drm/xe/display/xe_plane_initial.c |  4 +-
 drivers/gpu/drm/xe/xe_bo.c                    | 72 ++++++++++++++-----
 drivers/gpu/drm/xe/xe_bo.h                    | 13 ++--
 drivers/gpu/drm/xe/xe_eu_stall.c              |  5 +-
 5 files changed, 83 insertions(+), 53 deletions(-)

diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index a5c35e9ff2ef..7e82e49544fa 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -102,29 +102,29 @@ static int __xe_pin_fb_vma_dpt(const struct intel_framebuffer *fb,
 				 XE_PAGE_SIZE);
 
 	if (IS_DGFX(xe))
-		dpt = xe_bo_create_pin_map_at_aligned(xe, tile0, NULL,
-						      dpt_size, ~0ull,
-						      ttm_bo_type_kernel,
-						      XE_BO_FLAG_VRAM0 |
-						      XE_BO_FLAG_GGTT |
-						      XE_BO_FLAG_PAGETABLE,
-						      alignment);
+		dpt = xe_bo_create_pin_map_at_novm(xe, tile0,
+						   dpt_size, ~0ull,
+						   ttm_bo_type_kernel,
+						   XE_BO_FLAG_VRAM0 |
+						   XE_BO_FLAG_GGTT |
+						   XE_BO_FLAG_PAGETABLE,
+						   alignment, false);
 	else
-		dpt = xe_bo_create_pin_map_at_aligned(xe, tile0, NULL,
-						      dpt_size,  ~0ull,
-						      ttm_bo_type_kernel,
-						      XE_BO_FLAG_STOLEN |
-						      XE_BO_FLAG_GGTT |
-						      XE_BO_FLAG_PAGETABLE,
-						      alignment);
+		dpt = xe_bo_create_pin_map_at_novm(xe, tile0,
+						   dpt_size,  ~0ull,
+						   ttm_bo_type_kernel,
+						   XE_BO_FLAG_STOLEN |
+						   XE_BO_FLAG_GGTT |
+						   XE_BO_FLAG_PAGETABLE,
+						   alignment, false);
 	if (IS_ERR(dpt))
-		dpt = xe_bo_create_pin_map_at_aligned(xe, tile0, NULL,
-						      dpt_size,  ~0ull,
-						      ttm_bo_type_kernel,
-						      XE_BO_FLAG_SYSTEM |
-						      XE_BO_FLAG_GGTT |
-						      XE_BO_FLAG_PAGETABLE,
-						      alignment);
+		dpt = xe_bo_create_pin_map_at_novm(xe, tile0,
+						   dpt_size,  ~0ull,
+						   ttm_bo_type_kernel,
+						   XE_BO_FLAG_SYSTEM |
+						   XE_BO_FLAG_GGTT |
+						   XE_BO_FLAG_PAGETABLE,
+						   alignment, false);
 	if (IS_ERR(dpt))
 		return PTR_ERR(dpt);
 
diff --git a/drivers/gpu/drm/xe/display/xe_plane_initial.c b/drivers/gpu/drm/xe/display/xe_plane_initial.c
index 00da1dff0c44..6d52d75cd9bf 100644
--- a/drivers/gpu/drm/xe/display/xe_plane_initial.c
+++ b/drivers/gpu/drm/xe/display/xe_plane_initial.c
@@ -139,8 +139,8 @@ initial_plane_bo(struct xe_device *xe,
 			page_size);
 	size -= base;
 
-	bo = xe_bo_create_pin_map_at(xe, tile0, NULL, size, phys_base,
-				     ttm_bo_type_kernel, flags);
+	bo = xe_bo_create_pin_map_at_novm(xe, tile0, size, phys_base,
+					  ttm_bo_type_kernel, flags, 0, false);
 	if (IS_ERR(bo)) {
 		drm_dbg(&xe->drm,
 			"Failed to create bo phys_base=%pa size %u with flags %x: %li\n",
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index ad2863a7df77..b7d7d1445790 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -2428,27 +2428,17 @@ struct xe_bo *xe_bo_create_user(struct xe_device *xe,
 	return bo;
 }
 
-struct xe_bo *xe_bo_create_pin_map_at(struct xe_device *xe, struct xe_tile *tile,
-				      struct xe_vm *vm,
-				      size_t size, u64 offset,
-				      enum ttm_bo_type type, u32 flags)
-{
-	return xe_bo_create_pin_map_at_aligned(xe, tile, vm, size, offset,
-					       type, flags, 0);
-}
-
-struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
-					      struct xe_tile *tile,
-					      struct xe_vm *vm,
-					      size_t size, u64 offset,
-					      enum ttm_bo_type type, u32 flags,
-					      u64 alignment)
+static struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
+						     struct xe_tile *tile,
+						     struct xe_vm *vm,
+						     size_t size, u64 offset,
+						     enum ttm_bo_type type, u32 flags,
+						     u64 alignment, struct drm_exec *exec)
 {
 	struct xe_bo *bo;
 	int err;
 	u64 start = offset == ~0ull ? 0 : offset;
-	u64 end = offset == ~0ull ? offset : start + size;
-	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
+	u64 end = offset == ~0ull ? ~0ull : start + size;
 
 	if (flags & XE_BO_FLAG_STOLEN &&
 	    xe_ttm_stolen_cpu_access_needs_ggtt(xe))
@@ -2480,11 +2470,57 @@ err_put:
 	return ERR_PTR(err);
 }
 
+/**
+ * xe_bo_create_pin_map_at_novm() - Create pinned and mapped bo at optional VRAM offset
+ * @xe: The xe device.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @size: The storage size to use for the bo.
+ * @offset: Optional VRAM offset or %~0ull for don't care.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ * @alignment: GGTT alignment.
+ * @intr: Whether to execute any waits for backing store interruptible.
+ *
+ * Create a pinned and optionally mapped bo with VRAM offset and GGTT alignment
+ * options. The bo will be external and not associated with a VM.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ * In particular, the function may return ERR_PTR(%-EINTR) if @intr was set
+ * to true on entry.
+ */
+struct xe_bo *
+xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile,
+			     size_t size, u64 offset, enum ttm_bo_type type, u32 flags,
+			     u64 alignment, bool intr)
+{
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
+	struct xe_bo *bo;
+	int ret = 0;
+
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.interruptible = intr},
+			    ret) {
+		bo = xe_bo_create_pin_map_at_aligned(xe, tile, NULL, size, offset,
+						     type, flags, alignment, &exec);
+		if (IS_ERR(bo)) {
+			drm_exec_retry_on_contention(&exec);
+			ret = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&ctx, &ret);
+		}
+	}
+
+	return ret ? ERR_PTR(ret) : bo;
+}
+
 struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 				   struct xe_vm *vm, size_t size,
 				   enum ttm_bo_type type, u32 flags)
 {
-	return xe_bo_create_pin_map_at(xe, tile, vm, size, ~0ull, type, flags);
+	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
+
+	return xe_bo_create_pin_map_at_aligned(xe, tile, vm, size, ~0ull, type, flags,
+					       0, exec);
 }
 
 static void __xe_bo_unpin_map_no_vm(void *arg)
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index ebca6e9f6959..d9aef56ab878 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -109,15 +109,10 @@ struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_vm *vm, size_t s
 struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 				   struct xe_vm *vm, size_t size,
 				   enum ttm_bo_type type, u32 flags);
-struct xe_bo *xe_bo_create_pin_map_at(struct xe_device *xe, struct xe_tile *tile,
-				      struct xe_vm *vm, size_t size, u64 offset,
-				      enum ttm_bo_type type, u32 flags);
-struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
-					      struct xe_tile *tile,
-					      struct xe_vm *vm,
-					      size_t size, u64 offset,
-					      enum ttm_bo_type type, u32 flags,
-					      u64 alignment);
+struct xe_bo *
+xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile,
+			     size_t size, u64 offset, enum ttm_bo_type type,
+			     u32 flags, u64 alignment, bool intr);
 struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 					   size_t size, u32 flags);
 struct xe_bo *xe_managed_bo_create_from_data(struct xe_device *xe, struct xe_tile *tile,
diff --git a/drivers/gpu/drm/xe/xe_eu_stall.c b/drivers/gpu/drm/xe/xe_eu_stall.c
index fdd514fec5ef..f5cfdf29fde3 100644
--- a/drivers/gpu/drm/xe/xe_eu_stall.c
+++ b/drivers/gpu/drm/xe/xe_eu_stall.c
@@ -617,9 +617,8 @@ static int xe_eu_stall_data_buf_alloc(struct xe_eu_stall_data_stream *stream,
 
 	size = stream->per_xecore_buf_size * last_xecore;
 
-	bo = xe_bo_create_pin_map_at_aligned(tile->xe, tile, NULL,
-					     size, ~0ull, ttm_bo_type_kernel,
-					     XE_BO_FLAG_SYSTEM | XE_BO_FLAG_GGTT, SZ_64);
+	bo = xe_bo_create_pin_map_at_novm(tile->xe, tile, size, ~0ull, ttm_bo_type_kernel,
+					  XE_BO_FLAG_SYSTEM | XE_BO_FLAG_GGTT, SZ_64, false);
 	if (IS_ERR(bo)) {
 		kfree(stream->xecore_buf);
 		return PTR_ERR(bo);

From 59eabff2a3524d7f3bf6c24890cd197c4c116fcb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:44 +0200
Subject: [PATCH 1182/1455] drm/xe: Convert xe_bo_create_pin_map() for
 exhaustive eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce an xe_bo_create_pin_map_novm() function that does not
take the drm_exec paramenter to simplify the conversion of many
callsites.
For the rest, ensure that the same drm_exec context that was used
for locking the vm is passed down to validation.

Use xe_validation_guard() where appropriate.

v2:
- Avoid gotos from within xe_validation_guard(). (Matt Brost)
- Break out the change to pf_provision_vf_lmem8 to a separate
  patch.
- Adapt to signature change of xe_validation_guard().

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-12-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/display/intel_fbdev_fb.c   |  18 +--
 drivers/gpu/drm/xe/display/xe_dsb_buffer.c    |  10 +-
 drivers/gpu/drm/xe/display/xe_hdcp_gsc.c      |   8 +-
 drivers/gpu/drm/xe/tests/xe_migrate.c         |   9 +-
 drivers/gpu/drm/xe/xe_bo.c                    |  52 +++++++-
 drivers/gpu/drm/xe/xe_bo.h                    |   6 +-
 drivers/gpu/drm/xe/xe_gsc.c                   |   8 +-
 drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c |  24 ++--
 drivers/gpu/drm/xe/xe_guc_engine_activity.c   |  13 +-
 drivers/gpu/drm/xe/xe_lmtt.c                  |  12 +-
 drivers/gpu/drm/xe/xe_lrc.c                   |   7 +-
 drivers/gpu/drm/xe/xe_migrate.c               |  20 ++-
 drivers/gpu/drm/xe/xe_oa.c                    |   6 +-
 drivers/gpu/drm/xe/xe_pt.c                    |  10 +-
 drivers/gpu/drm/xe/xe_pt.h                    |   3 +-
 drivers/gpu/drm/xe/xe_pxp_submit.c            |  34 +++--
 drivers/gpu/drm/xe/xe_vm.c                    | 123 +++++++++++-------
 17 files changed, 232 insertions(+), 131 deletions(-)

diff --git a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
index d96ba2b51065..8ea9a472113c 100644
--- a/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
+++ b/drivers/gpu/drm/xe/display/intel_fbdev_fb.c
@@ -42,11 +42,11 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 	obj = ERR_PTR(-ENODEV);
 
 	if (!IS_DGFX(xe) && !XE_GT_WA(xe_root_mmio_gt(xe), 22019338487_display)) {
-		obj = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe),
-					   NULL, size,
-					   ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
-					   XE_BO_FLAG_STOLEN |
-					   XE_BO_FLAG_GGTT);
+		obj = xe_bo_create_pin_map_novm(xe, xe_device_get_root_tile(xe),
+						size,
+						ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
+						XE_BO_FLAG_STOLEN |
+						XE_BO_FLAG_GGTT, false);
 		if (!IS_ERR(obj))
 			drm_info(&xe->drm, "Allocated fbdev into stolen\n");
 		else
@@ -54,10 +54,10 @@ struct intel_framebuffer *intel_fbdev_fb_alloc(struct drm_fb_helper *helper,
 	}
 
 	if (IS_ERR(obj)) {
-		obj = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe), NULL, size,
-					   ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
-					   XE_BO_FLAG_VRAM_IF_DGFX(xe_device_get_root_tile(xe)) |
-					   XE_BO_FLAG_GGTT);
+		obj = xe_bo_create_pin_map_novm(xe, xe_device_get_root_tile(xe), size,
+						ttm_bo_type_kernel, XE_BO_FLAG_SCANOUT |
+						XE_BO_FLAG_VRAM_IF_DGFX(xe_device_get_root_tile(xe)) |
+						XE_BO_FLAG_GGTT, false);
 	}
 
 	if (IS_ERR(obj)) {
diff --git a/drivers/gpu/drm/xe/display/xe_dsb_buffer.c b/drivers/gpu/drm/xe/display/xe_dsb_buffer.c
index 9f941fc2e36b..58581d7aaae6 100644
--- a/drivers/gpu/drm/xe/display/xe_dsb_buffer.c
+++ b/drivers/gpu/drm/xe/display/xe_dsb_buffer.c
@@ -43,11 +43,11 @@ bool intel_dsb_buffer_create(struct intel_crtc *crtc, struct intel_dsb_buffer *d
 		return false;
 
 	/* Set scanout flag for WC mapping */
-	obj = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe),
-				   NULL, PAGE_ALIGN(size),
-				   ttm_bo_type_kernel,
-				   XE_BO_FLAG_VRAM_IF_DGFX(xe_device_get_root_tile(xe)) |
-				   XE_BO_FLAG_SCANOUT | XE_BO_FLAG_GGTT);
+	obj = xe_bo_create_pin_map_novm(xe, xe_device_get_root_tile(xe),
+					PAGE_ALIGN(size),
+					ttm_bo_type_kernel,
+					XE_BO_FLAG_VRAM_IF_DGFX(xe_device_get_root_tile(xe)) |
+					XE_BO_FLAG_SCANOUT | XE_BO_FLAG_GGTT, false);
 	if (IS_ERR(obj)) {
 		kfree(vma);
 		return false;
diff --git a/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c b/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c
index 30f1073141fc..4ae847b628e2 100644
--- a/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c
+++ b/drivers/gpu/drm/xe/display/xe_hdcp_gsc.c
@@ -72,10 +72,10 @@ static int intel_hdcp_gsc_initialize_message(struct xe_device *xe,
 	int ret = 0;
 
 	/* allocate object of two page for HDCP command memory and store it */
-	bo = xe_bo_create_pin_map(xe, xe_device_get_root_tile(xe), NULL, PAGE_SIZE * 2,
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM |
-				  XE_BO_FLAG_GGTT);
+	bo = xe_bo_create_pin_map_novm(xe, xe_device_get_root_tile(xe), PAGE_SIZE * 2,
+				       ttm_bo_type_kernel,
+				       XE_BO_FLAG_SYSTEM |
+				       XE_BO_FLAG_GGTT, false);
 
 	if (IS_ERR(bo)) {
 		drm_err(&xe->drm, "Failed to allocate bo for HDCP streaming command!\n");
diff --git a/drivers/gpu/drm/xe/tests/xe_migrate.c b/drivers/gpu/drm/xe/tests/xe_migrate.c
index afa794e56065..5904d658d1f2 100644
--- a/drivers/gpu/drm/xe/tests/xe_migrate.c
+++ b/drivers/gpu/drm/xe/tests/xe_migrate.c
@@ -204,7 +204,8 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test,
 
 	big = xe_bo_create_pin_map(xe, tile, m->q->vm, SZ_4M,
 				   ttm_bo_type_kernel,
-				   XE_BO_FLAG_VRAM_IF_DGFX(tile));
+				   XE_BO_FLAG_VRAM_IF_DGFX(tile),
+				   exec);
 	if (IS_ERR(big)) {
 		KUNIT_FAIL(test, "Failed to allocate bo: %li\n", PTR_ERR(big));
 		goto vunmap;
@@ -212,7 +213,8 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test,
 
 	pt = xe_bo_create_pin_map(xe, tile, m->q->vm, XE_PAGE_SIZE,
 				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_VRAM_IF_DGFX(tile));
+				  XE_BO_FLAG_VRAM_IF_DGFX(tile),
+				  exec);
 	if (IS_ERR(pt)) {
 		KUNIT_FAIL(test, "Failed to allocate fake pt: %li\n",
 			   PTR_ERR(pt));
@@ -222,7 +224,8 @@ static void xe_migrate_sanity_test(struct xe_migrate *m, struct kunit *test,
 	tiny = xe_bo_create_pin_map(xe, tile, m->q->vm,
 				    2 * SZ_4K,
 				    ttm_bo_type_kernel,
-				    XE_BO_FLAG_VRAM_IF_DGFX(tile));
+				    XE_BO_FLAG_VRAM_IF_DGFX(tile),
+				    exec);
 	if (IS_ERR(tiny)) {
 		KUNIT_FAIL(test, "Failed to allocate tiny fake pt: %li\n",
 			   PTR_ERR(tiny));
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index b7d7d1445790..1012655b65fd 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -2513,16 +2513,59 @@ xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile,
 	return ret ? ERR_PTR(ret) : bo;
 }
 
+/**
+ * xe_bo_create_pin_map() - Create pinned and mapped bo
+ * @xe: The xe device.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * @vm: The vm to associate the buffer object with. The vm's resv must be locked
+ * with the transaction represented by @exec.
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @size: The storage size to use for the bo.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ * @exec: The drm_exec transaction to use for exhaustive eviction, and
+ * previously used for locking @vm's resv.
+ *
+ * Create a pinned and mapped bo. The bo will be external and not associated
+ * with a VM.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ * In particular, the function may return ERR_PTR(%-EINTR) if @exec was
+ * configured for interruptible locking.
+ */
 struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 				   struct xe_vm *vm, size_t size,
-				   enum ttm_bo_type type, u32 flags)
+				   enum ttm_bo_type type, u32 flags,
+				   struct drm_exec *exec)
 {
-	struct drm_exec *exec = vm ? xe_vm_validation_exec(vm) : XE_VALIDATION_UNIMPLEMENTED;
-
 	return xe_bo_create_pin_map_at_aligned(xe, tile, vm, size, ~0ull, type, flags,
 					       0, exec);
 }
 
+/**
+ * xe_bo_create_pin_map_novm() - Create pinned and mapped bo
+ * @xe: The xe device.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @size: The storage size to use for the bo.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ * @intr: Whether to execut any waits for backing store interruptible.
+ *
+ * Create a pinned and mapped bo. The bo will be external and not associated
+ * with a VM.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ * In particular, the function may return ERR_PTR(%-EINTR) if @intr was set
+ * to true on entry.
+ */
+struct xe_bo *xe_bo_create_pin_map_novm(struct xe_device *xe, struct xe_tile *tile,
+					size_t size, enum ttm_bo_type type, u32 flags,
+					bool intr)
+{
+	return xe_bo_create_pin_map_at_novm(xe, tile, size, ~0ull, type, flags, 0, intr);
+}
+
 static void __xe_bo_unpin_map_no_vm(void *arg)
 {
 	xe_bo_unpin_map_no_vm(arg);
@@ -2535,8 +2578,7 @@ struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile
 	int ret;
 
 	KUNIT_STATIC_STUB_REDIRECT(xe_managed_bo_create_pin_map, xe, tile, size, flags);
-
-	bo = xe_bo_create_pin_map(xe, tile, NULL, size, ttm_bo_type_kernel, flags);
+	bo = xe_bo_create_pin_map_novm(xe, tile, size, ttm_bo_type_kernel, flags, true);
 	if (IS_ERR(bo))
 		return bo;
 
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index d9aef56ab878..143f59bcfb93 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -108,7 +108,11 @@ struct xe_bo *xe_bo_create_user(struct xe_device *xe, struct xe_vm *vm, size_t s
 				u16 cpu_caching, u32 flags, struct drm_exec *exec);
 struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 				   struct xe_vm *vm, size_t size,
-				   enum ttm_bo_type type, u32 flags);
+				   enum ttm_bo_type type, u32 flags,
+				   struct drm_exec *exec);
+struct xe_bo *xe_bo_create_pin_map_novm(struct xe_device *xe, struct xe_tile *tile,
+					size_t size, enum ttm_bo_type type, u32 flags,
+					bool intr);
 struct xe_bo *
 xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile,
 			     size_t size, u64 offset, enum ttm_bo_type type,
diff --git a/drivers/gpu/drm/xe/xe_gsc.c b/drivers/gpu/drm/xe/xe_gsc.c
index f5ae28af60d4..83d61bf8ec62 100644
--- a/drivers/gpu/drm/xe/xe_gsc.c
+++ b/drivers/gpu/drm/xe/xe_gsc.c
@@ -136,10 +136,10 @@ static int query_compatibility_version(struct xe_gsc *gsc)
 	u64 ggtt_offset;
 	int err;
 
-	bo = xe_bo_create_pin_map(xe, tile, NULL, GSC_VER_PKT_SZ * 2,
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM |
-				  XE_BO_FLAG_GGTT);
+	bo = xe_bo_create_pin_map_novm(xe, tile, GSC_VER_PKT_SZ * 2,
+				       ttm_bo_type_kernel,
+				       XE_BO_FLAG_SYSTEM |
+				       XE_BO_FLAG_GGTT, false);
 	if (IS_ERR(bo)) {
 		xe_gt_err(gt, "failed to allocate bo for GSC version query\n");
 		return PTR_ERR(bo);
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c
index c712111aa30d..44cc612b0a75 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_migration.c
@@ -55,12 +55,12 @@ static int pf_send_guc_save_vf_state(struct xe_gt *gt, unsigned int vfid,
 	xe_gt_assert(gt, size % sizeof(u32) == 0);
 	xe_gt_assert(gt, size == ndwords * sizeof(u32));
 
-	bo = xe_bo_create_pin_map(xe, tile, NULL,
-				  ALIGN(size, PAGE_SIZE),
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM |
-				  XE_BO_FLAG_GGTT |
-				  XE_BO_FLAG_GGTT_INVALIDATE);
+	bo = xe_bo_create_pin_map_novm(xe, tile,
+				       ALIGN(size, PAGE_SIZE),
+				       ttm_bo_type_kernel,
+				       XE_BO_FLAG_SYSTEM |
+				       XE_BO_FLAG_GGTT |
+				       XE_BO_FLAG_GGTT_INVALIDATE, false);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
@@ -91,12 +91,12 @@ static int pf_send_guc_restore_vf_state(struct xe_gt *gt, unsigned int vfid,
 	xe_gt_assert(gt, size % sizeof(u32) == 0);
 	xe_gt_assert(gt, size == ndwords * sizeof(u32));
 
-	bo = xe_bo_create_pin_map(xe, tile, NULL,
-				  ALIGN(size, PAGE_SIZE),
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM |
-				  XE_BO_FLAG_GGTT |
-				  XE_BO_FLAG_GGTT_INVALIDATE);
+	bo = xe_bo_create_pin_map_novm(xe, tile,
+				       ALIGN(size, PAGE_SIZE),
+				       ttm_bo_type_kernel,
+				       XE_BO_FLAG_SYSTEM |
+				       XE_BO_FLAG_GGTT |
+				       XE_BO_FLAG_GGTT_INVALIDATE, false);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/drm/xe/xe_guc_engine_activity.c b/drivers/gpu/drm/xe/xe_guc_engine_activity.c
index 92e1f9f41b8c..2b99c1ebdd58 100644
--- a/drivers/gpu/drm/xe/xe_guc_engine_activity.c
+++ b/drivers/gpu/drm/xe/xe_guc_engine_activity.c
@@ -94,16 +94,17 @@ static int allocate_engine_activity_buffers(struct xe_guc *guc,
 	struct xe_tile *tile = gt_to_tile(gt);
 	struct xe_bo *bo, *metadata_bo;
 
-	metadata_bo = xe_bo_create_pin_map(gt_to_xe(gt), tile, NULL, PAGE_ALIGN(metadata_size),
-					   ttm_bo_type_kernel, XE_BO_FLAG_SYSTEM |
-					   XE_BO_FLAG_GGTT | XE_BO_FLAG_GGTT_INVALIDATE);
+	metadata_bo = xe_bo_create_pin_map_novm(gt_to_xe(gt), tile, PAGE_ALIGN(metadata_size),
+						ttm_bo_type_kernel, XE_BO_FLAG_SYSTEM |
+						XE_BO_FLAG_GGTT | XE_BO_FLAG_GGTT_INVALIDATE,
+						false);
 
 	if (IS_ERR(metadata_bo))
 		return PTR_ERR(metadata_bo);
 
-	bo = xe_bo_create_pin_map(gt_to_xe(gt), tile, NULL, PAGE_ALIGN(size),
-				  ttm_bo_type_kernel, XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				  XE_BO_FLAG_GGTT | XE_BO_FLAG_GGTT_INVALIDATE);
+	bo = xe_bo_create_pin_map_novm(gt_to_xe(gt), tile, PAGE_ALIGN(size),
+				       ttm_bo_type_kernel, XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+				       XE_BO_FLAG_GGTT | XE_BO_FLAG_GGTT_INVALIDATE, false);
 
 	if (IS_ERR(bo)) {
 		xe_bo_unpin_map_no_vm(metadata_bo);
diff --git a/drivers/gpu/drm/xe/xe_lmtt.c b/drivers/gpu/drm/xe/xe_lmtt.c
index f2bfbfa3efa1..62fc5a1a332d 100644
--- a/drivers/gpu/drm/xe/xe_lmtt.c
+++ b/drivers/gpu/drm/xe/xe_lmtt.c
@@ -67,12 +67,12 @@ static struct xe_lmtt_pt *lmtt_pt_alloc(struct xe_lmtt *lmtt, unsigned int level
 		goto out;
 	}
 
-	bo = xe_bo_create_pin_map(lmtt_to_xe(lmtt), lmtt_to_tile(lmtt), NULL,
-				  PAGE_ALIGN(lmtt->ops->lmtt_pte_size(level) *
-					     lmtt->ops->lmtt_pte_num(level)),
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_VRAM_IF_DGFX(lmtt_to_tile(lmtt)) |
-				  XE_BO_FLAG_NEEDS_64K);
+	bo = xe_bo_create_pin_map_novm(lmtt_to_xe(lmtt), lmtt_to_tile(lmtt),
+				       PAGE_ALIGN(lmtt->ops->lmtt_pte_size(level) *
+						  lmtt->ops->lmtt_pte_num(level)),
+				       ttm_bo_type_kernel,
+				       XE_BO_FLAG_VRAM_IF_DGFX(lmtt_to_tile(lmtt)) |
+				       XE_BO_FLAG_NEEDS_64K, false);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		goto out_free_pt;
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 8f6c3ba47882..6d52e0eb97f5 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -1340,9 +1340,10 @@ static int xe_lrc_init(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 	if (vm && vm->xef) /* userspace */
 		bo_flags |= XE_BO_FLAG_PINNED_LATE_RESTORE;
 
-	lrc->bo = xe_bo_create_pin_map(xe, tile, NULL, bo_size,
-				       ttm_bo_type_kernel,
-				       bo_flags);
+	lrc->bo = xe_bo_create_pin_map_novm(xe, tile,
+					    bo_size,
+					    ttm_bo_type_kernel,
+					    bo_flags, false);
 	if (IS_ERR(lrc->bo))
 		return PTR_ERR(lrc->bo);
 
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 3f0c8832120f..6fad5d469629 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -35,6 +35,7 @@
 #include "xe_sched_job.h"
 #include "xe_sync.h"
 #include "xe_trace_bo.h"
+#include "xe_validation.h"
 #include "xe_vm.h"
 #include "xe_vram.h"
 
@@ -173,7 +174,7 @@ static void xe_migrate_program_identity(struct xe_device *xe, struct xe_vm *vm,
 }
 
 static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
-				 struct xe_vm *vm)
+				 struct xe_vm *vm, struct drm_exec *exec)
 {
 	struct xe_device *xe = tile_to_xe(tile);
 	u16 pat_index = xe->pat.idx[XE_CACHE_WB];
@@ -200,7 +201,7 @@ static int xe_migrate_prepare_vm(struct xe_tile *tile, struct xe_migrate *m,
 				  num_entries * XE_PAGE_SIZE,
 				  ttm_bo_type_kernel,
 				  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				  XE_BO_FLAG_PAGETABLE);
+				  XE_BO_FLAG_PAGETABLE, exec);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
@@ -404,6 +405,8 @@ int xe_migrate_init(struct xe_migrate *m)
 	struct xe_tile *tile = m->tile;
 	struct xe_gt *primary_gt = tile->primary_gt;
 	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
 	struct xe_vm *vm;
 	int err;
 
@@ -413,11 +416,16 @@ int xe_migrate_init(struct xe_migrate *m)
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
-	xe_vm_lock(vm, false);
-	err = xe_migrate_prepare_vm(tile, m, vm);
-	xe_vm_unlock(vm);
+	err = 0;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {}, err) {
+		err = xe_vm_drm_exec_lock(vm, &exec);
+		drm_exec_retry_on_contention(&exec);
+		err = xe_migrate_prepare_vm(tile, m, vm, &exec);
+		drm_exec_retry_on_contention(&exec);
+		xe_validation_retry_on_oom(&ctx, &err);
+	}
 	if (err)
-		goto err_out;
+		return err;
 
 	if (xe->info.has_usm) {
 		struct xe_hw_engine *hwe = xe_gt_hw_engine(primary_gt,
diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c
index a188bad172ad..a4894eb0d7f3 100644
--- a/drivers/gpu/drm/xe/xe_oa.c
+++ b/drivers/gpu/drm/xe/xe_oa.c
@@ -883,9 +883,9 @@ static int xe_oa_alloc_oa_buffer(struct xe_oa_stream *stream, size_t size)
 {
 	struct xe_bo *bo;
 
-	bo = xe_bo_create_pin_map(stream->oa->xe, stream->gt->tile, NULL,
-				  size, ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM | XE_BO_FLAG_GGTT);
+	bo = xe_bo_create_pin_map_novm(stream->oa->xe, stream->gt->tile,
+				       size, ttm_bo_type_kernel,
+				       XE_BO_FLAG_SYSTEM | XE_BO_FLAG_GGTT, false);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 07ecf4132d41..01eea8eb1779 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -90,6 +90,7 @@ static void xe_pt_free(struct xe_pt *pt)
  * @vm: The vm to create for.
  * @tile: The tile to create for.
  * @level: The page-table level.
+ * @exec: The drm_exec object used to lock the vm.
  *
  * Allocate and initialize a single struct xe_pt metadata structure. Also
  * create the corresponding page-table bo, but don't initialize it. If the
@@ -101,7 +102,7 @@ static void xe_pt_free(struct xe_pt *pt)
  * error.
  */
 struct xe_pt *xe_pt_create(struct xe_vm *vm, struct xe_tile *tile,
-			   unsigned int level)
+			   unsigned int level, struct drm_exec *exec)
 {
 	struct xe_pt *pt;
 	struct xe_bo *bo;
@@ -125,9 +126,11 @@ struct xe_pt *xe_pt_create(struct xe_vm *vm, struct xe_tile *tile,
 		bo_flags |= XE_BO_FLAG_PINNED_LATE_RESTORE;
 
 	pt->level = level;
+
+	drm_WARN_ON(&vm->xe->drm, IS_ERR_OR_NULL(exec));
 	bo = xe_bo_create_pin_map(vm->xe, tile, vm, SZ_4K,
 				  ttm_bo_type_kernel,
-				  bo_flags);
+				  bo_flags, exec);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		goto err_kfree;
@@ -591,7 +594,8 @@ xe_pt_stage_bind_entry(struct xe_ptw *parent, pgoff_t offset,
 	if (covers || !*child) {
 		u64 flags = 0;
 
-		xe_child = xe_pt_create(xe_walk->vm, xe_walk->tile, level - 1);
+		xe_child = xe_pt_create(xe_walk->vm, xe_walk->tile, level - 1,
+					xe_vm_validation_exec(vm));
 		if (IS_ERR(xe_child))
 			return PTR_ERR(xe_child);
 
diff --git a/drivers/gpu/drm/xe/xe_pt.h b/drivers/gpu/drm/xe/xe_pt.h
index 5ecf003d513c..4daeebaab5a1 100644
--- a/drivers/gpu/drm/xe/xe_pt.h
+++ b/drivers/gpu/drm/xe/xe_pt.h
@@ -10,6 +10,7 @@
 #include "xe_pt_types.h"
 
 struct dma_fence;
+struct drm_exec;
 struct xe_bo;
 struct xe_device;
 struct xe_exec_queue;
@@ -29,7 +30,7 @@ struct xe_vma_ops;
 unsigned int xe_pt_shift(unsigned int level);
 
 struct xe_pt *xe_pt_create(struct xe_vm *vm, struct xe_tile *tile,
-			   unsigned int level);
+			   unsigned int level, struct drm_exec *exec);
 
 void xe_pt_populate_empty(struct xe_tile *tile, struct xe_vm *vm,
 			  struct xe_pt *pt);
diff --git a/drivers/gpu/drm/xe/xe_pxp_submit.c b/drivers/gpu/drm/xe/xe_pxp_submit.c
index ca95f2a4d4ef..e60526e30030 100644
--- a/drivers/gpu/drm/xe/xe_pxp_submit.c
+++ b/drivers/gpu/drm/xe/xe_pxp_submit.c
@@ -54,8 +54,9 @@ static int allocate_vcs_execution_resources(struct xe_pxp *pxp)
 	 * Each termination is 16 DWORDS, so 4K is enough to contain a
 	 * termination for each sessions.
 	 */
-	bo = xe_bo_create_pin_map(xe, tile, NULL, SZ_4K, ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM | XE_BO_FLAG_PINNED | XE_BO_FLAG_GGTT);
+	bo = xe_bo_create_pin_map_novm(xe, tile, SZ_4K, ttm_bo_type_kernel,
+				       XE_BO_FLAG_SYSTEM | XE_BO_FLAG_PINNED | XE_BO_FLAG_GGTT,
+				       false);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		goto out_queue;
@@ -87,7 +88,9 @@ static int allocate_gsc_client_resources(struct xe_gt *gt,
 {
 	struct xe_tile *tile = gt_to_tile(gt);
 	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_validation_ctx ctx;
 	struct xe_hw_engine *hwe;
+	struct drm_exec exec;
 	struct xe_vm *vm;
 	struct xe_bo *bo;
 	struct xe_exec_queue *q;
@@ -106,15 +109,26 @@ static int allocate_gsc_client_resources(struct xe_gt *gt,
 		return PTR_ERR(vm);
 
 	/* We allocate a single object for the batch and the in/out memory */
-	xe_vm_lock(vm, false);
-	bo = xe_bo_create_pin_map(xe, tile, vm, PXP_BB_SIZE + inout_size * 2,
-				  ttm_bo_type_kernel,
-				  XE_BO_FLAG_SYSTEM | XE_BO_FLAG_PINNED | XE_BO_FLAG_NEEDS_UC);
-	xe_vm_unlock(vm);
-	if (IS_ERR(bo)) {
-		err = PTR_ERR(bo);
-		goto vm_out;
+
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags){}, err) {
+		err = xe_vm_drm_exec_lock(vm, &exec);
+		drm_exec_retry_on_contention(&exec);
+		if (err)
+			break;
+
+		bo = xe_bo_create_pin_map(xe, tile, vm, PXP_BB_SIZE + inout_size * 2,
+					  ttm_bo_type_kernel,
+					  XE_BO_FLAG_SYSTEM | XE_BO_FLAG_PINNED |
+					  XE_BO_FLAG_NEEDS_UC, &exec);
+		drm_exec_retry_on_contention(&exec);
+		if (IS_ERR(bo)) {
+			err = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&ctx, &err);
+			break;
+		}
 	}
+	if (err)
+		goto vm_out;
 
 	fence = xe_vm_bind_kernel_bo(vm, bo, NULL, 0, XE_CACHE_WB);
 	if (IS_ERR(fence)) {
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 785e81cf023d..7b04b392b729 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1370,6 +1370,7 @@ static void vm_destroy_work_func(struct work_struct *w);
  * @xe: xe device.
  * @tile: tile to set up for.
  * @vm: vm to set up for.
+ * @exec: The struct drm_exec object used to lock the vm resv.
  *
  * Sets up a pagetable tree with one page-table per level and a single
  * leaf PTE. All pagetable entries point to the single page-table or,
@@ -1379,20 +1380,19 @@ static void vm_destroy_work_func(struct work_struct *w);
  * Return: 0 on success, negative error code on error.
  */
 static int xe_vm_create_scratch(struct xe_device *xe, struct xe_tile *tile,
-				struct xe_vm *vm)
+				struct xe_vm *vm, struct drm_exec *exec)
 {
 	u8 id = tile->id;
 	int i;
 
 	for (i = MAX_HUGEPTE_LEVEL; i < vm->pt_root[id]->level; i++) {
-		vm->scratch_pt[id][i] = xe_pt_create(vm, tile, i);
+		vm->scratch_pt[id][i] = xe_pt_create(vm, tile, i, exec);
 		if (IS_ERR(vm->scratch_pt[id][i])) {
 			int err = PTR_ERR(vm->scratch_pt[id][i]);
 
 			vm->scratch_pt[id][i] = NULL;
 			return err;
 		}
-
 		xe_pt_populate_empty(tile, vm, vm->scratch_pt[id][i]);
 	}
 
@@ -1420,9 +1420,26 @@ static void xe_vm_free_scratch(struct xe_vm *vm)
 	}
 }
 
+static void xe_vm_pt_destroy(struct xe_vm *vm)
+{
+	struct xe_tile *tile;
+	u8 id;
+
+	xe_vm_assert_held(vm);
+
+	for_each_tile(tile, vm->xe, id) {
+		if (vm->pt_root[id]) {
+			xe_pt_destroy(vm->pt_root[id], vm->flags, NULL);
+			vm->pt_root[id] = NULL;
+		}
+	}
+}
+
 struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 {
 	struct drm_gem_object *vm_resv_obj;
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
 	struct xe_vm *vm;
 	int err, number_tiles = 0;
 	struct xe_tile *tile;
@@ -1507,49 +1524,68 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 
 	drm_gem_object_put(vm_resv_obj);
 
-	err = xe_vm_lock(vm, true);
-	if (err)
-		goto err_close;
+	err = 0;
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.interruptible = true},
+			    err) {
+		err = xe_vm_drm_exec_lock(vm, &exec);
+		drm_exec_retry_on_contention(&exec);
 
-	if (IS_DGFX(xe) && xe->info.vram_flags & XE_VRAM_FLAGS_NEED64K)
-		vm->flags |= XE_VM_FLAG_64K;
+		if (IS_DGFX(xe) && xe->info.vram_flags & XE_VRAM_FLAGS_NEED64K)
+			vm->flags |= XE_VM_FLAG_64K;
 
-	for_each_tile(tile, xe, id) {
-		if (flags & XE_VM_FLAG_MIGRATION &&
-		    tile->id != XE_VM_FLAG_TILE_ID(flags))
-			continue;
+		for_each_tile(tile, xe, id) {
+			if (flags & XE_VM_FLAG_MIGRATION &&
+			    tile->id != XE_VM_FLAG_TILE_ID(flags))
+				continue;
 
-		vm->pt_root[id] = xe_pt_create(vm, tile, xe->info.vm_max_level);
-		if (IS_ERR(vm->pt_root[id])) {
-			err = PTR_ERR(vm->pt_root[id]);
-			vm->pt_root[id] = NULL;
-			goto err_unlock_close;
+			vm->pt_root[id] = xe_pt_create(vm, tile, xe->info.vm_max_level,
+						       &exec);
+			if (IS_ERR(vm->pt_root[id])) {
+				err = PTR_ERR(vm->pt_root[id]);
+				vm->pt_root[id] = NULL;
+				xe_vm_pt_destroy(vm);
+				drm_exec_retry_on_contention(&exec);
+				xe_validation_retry_on_oom(&ctx, &err);
+				break;
+			}
 		}
-	}
+		if (err)
+			break;
 
-	if (xe_vm_has_scratch(vm)) {
+		if (xe_vm_has_scratch(vm)) {
+			for_each_tile(tile, xe, id) {
+				if (!vm->pt_root[id])
+					continue;
+
+				err = xe_vm_create_scratch(xe, tile, vm, &exec);
+				if (err) {
+					xe_vm_free_scratch(vm);
+					xe_vm_pt_destroy(vm);
+					drm_exec_retry_on_contention(&exec);
+					xe_validation_retry_on_oom(&ctx, &err);
+					break;
+				}
+			}
+			if (err)
+				break;
+			vm->batch_invalidate_tlb = true;
+		}
+
+		if (vm->flags & XE_VM_FLAG_LR_MODE) {
+			INIT_WORK(&vm->preempt.rebind_work, preempt_rebind_work_func);
+			vm->batch_invalidate_tlb = false;
+		}
+
+		/* Fill pt_root after allocating scratch tables */
 		for_each_tile(tile, xe, id) {
 			if (!vm->pt_root[id])
 				continue;
 
-			err = xe_vm_create_scratch(xe, tile, vm);
-			if (err)
-				goto err_unlock_close;
+			xe_pt_populate_empty(tile, vm, vm->pt_root[id]);
 		}
-		vm->batch_invalidate_tlb = true;
 	}
-
-	if (vm->flags & XE_VM_FLAG_LR_MODE)
-		vm->batch_invalidate_tlb = false;
-
-	/* Fill pt_root after allocating scratch tables */
-	for_each_tile(tile, xe, id) {
-		if (!vm->pt_root[id])
-			continue;
-
-		xe_pt_populate_empty(tile, vm, vm->pt_root[id]);
-	}
-	xe_vm_unlock(vm);
+	if (err)
+		goto err_close;
 
 	/* Kernel migration VM shouldn't have a circular loop.. */
 	if (!(flags & XE_VM_FLAG_MIGRATION)) {
@@ -1582,7 +1618,7 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 				      &xe->usm.next_asid, GFP_KERNEL);
 		up_write(&xe->usm.lock);
 		if (err < 0)
-			goto err_unlock_close;
+			goto err_close;
 
 		vm->usm.asid = asid;
 	}
@@ -1591,8 +1627,6 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags, struct xe_file *xef)
 
 	return vm;
 
-err_unlock_close:
-	xe_vm_unlock(vm);
 err_close:
 	xe_vm_close_and_put(vm);
 	return ERR_PTR(err);
@@ -1725,13 +1759,7 @@ void xe_vm_close_and_put(struct xe_vm *vm)
 	 * destroy the pagetables immediately.
 	 */
 	xe_vm_free_scratch(vm);
-
-	for_each_tile(tile, xe, id) {
-		if (vm->pt_root[id]) {
-			xe_pt_destroy(vm->pt_root[id], vm->flags, NULL);
-			vm->pt_root[id] = NULL;
-		}
-	}
+	xe_vm_pt_destroy(vm);
 	xe_vm_unlock(vm);
 
 	/*
@@ -3781,7 +3809,6 @@ release_vm_lock:
  */
 int xe_vm_lock(struct xe_vm *vm, bool intr)
 {
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	int ret;
 
 	if (intr)
@@ -3789,9 +3816,6 @@ int xe_vm_lock(struct xe_vm *vm, bool intr)
 	else
 		ret = dma_resv_lock(xe_vm_resv(vm), NULL);
 
-	if (!ret)
-		xe_vm_set_validation_exec(vm, exec);
-
 	return ret;
 }
 
@@ -3803,7 +3827,6 @@ int xe_vm_lock(struct xe_vm *vm, bool intr)
  */
 void xe_vm_unlock(struct xe_vm *vm)
 {
-	xe_vm_set_validation_exec(vm, NULL);
 	dma_resv_unlock(xe_vm_resv(vm));
 }
 

From 1f1541720f651727591411e68c16299de0447b43 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:45 +0200
Subject: [PATCH 1183/1455] drm/xe: Rework instances of variants of
 xe_bo_create_locked()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A common pattern is to create a locked bo, pin it without mapping
and then unlock it. Add a function to do that, which internally
uses xe_validation_guard().
With that we can remove xe_bo_create_locked_range() and add
exhaustive eviction to stolen, pf_provision_vf_lmem and
psmi_alloc_object.

v4:
- New patch after reorganization.
v5:
- Replace DRM_XE_GEM_CPU_CACHING_WB with 0. (CI)

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-13-thomas.hellstrom@linux.intel.com
---
 .../compat-i915-headers/gem/i915_gem_stolen.h | 13 +--
 drivers/gpu/drm/xe/xe_bo.c                    | 86 +++++++++++--------
 drivers/gpu/drm/xe/xe_bo.h                    |  9 +-
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c    | 23 ++---
 drivers/gpu/drm/xe/xe_psmi.c                  | 26 ++----
 5 files changed, 70 insertions(+), 87 deletions(-)

diff --git a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
index 1ce1e9da975b..a2d1f684a3a8 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
@@ -21,7 +21,6 @@ static inline int i915_gem_stolen_insert_node_in_range(struct xe_device *xe,
 						       u32 size, u32 align,
 						       u32 start, u32 end)
 {
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_bo *bo;
 	int err;
 	u32 flags = XE_BO_FLAG_PINNED | XE_BO_FLAG_STOLEN;
@@ -34,21 +33,13 @@ static inline int i915_gem_stolen_insert_node_in_range(struct xe_device *xe,
 		start = ALIGN(start, align);
 	}
 
-	bo = xe_bo_create_locked_range(xe, xe_device_get_root_tile(xe),
-				       NULL, size, start, end,
-				       ttm_bo_type_kernel, flags, 0, exec);
+	bo = xe_bo_create_pin_range_novm(xe, xe_device_get_root_tile(xe),
+					 size, start, end, ttm_bo_type_kernel, flags);
 	if (IS_ERR(bo)) {
 		err = PTR_ERR(bo);
 		bo = NULL;
 		return err;
 	}
-	err = xe_bo_pin(bo, exec);
-	xe_bo_unlock_vm_held(bo);
-
-	if (err) {
-		xe_bo_put(fb->bo);
-		bo = NULL;
-	}
 
 	fb->bo = bo;
 
diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 1012655b65fd..21efcb170239 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -2309,37 +2309,6 @@ err_unlock_put_bo:
 	return ERR_PTR(err);
 }
 
-/**
- * xe_bo_create_locked_range() - Create a BO with range- and alignment options
- * @xe: The xe device.
- * @tile: The tile to select for migration of this bo, and the tile used for
- * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
- * @vm: The local vm or NULL for external objects.
- * @size: The storage size to use for the bo.
- * @start: Start of fixed VRAM range or 0.
- * @end: End of fixed VRAM range or ~0ULL.
- * @type: The TTM buffer object type.
- * @flags: XE_BO_FLAG_ flags.
- * @alignment: For GGTT buffer objects, the minimum GGTT alignment.
- * @exec: The drm_exec transaction to use for exhaustive eviction.
- *
- * Create an Xe BO with range- and alignment options. If @start and @end indicate
- * a fixed VRAM range, this must be a ttm_bo_type_kernel bo with VRAM placement
- * only. The @alignment parameter can be used for GGTT alignment.
- *
- * Return: The buffer object on success. Negative error pointer on failure.
- */
-struct xe_bo *
-xe_bo_create_locked_range(struct xe_device *xe,
-			  struct xe_tile *tile, struct xe_vm *vm,
-			  size_t size, u64 start, u64 end,
-			  enum ttm_bo_type type, u32 flags, u64 alignment,
-			  struct drm_exec *exec)
-{
-	return __xe_bo_create_locked(xe, tile, vm, size, start, end, 0, type,
-				     flags, alignment, exec);
-}
-
 /**
  * xe_bo_create_locked() - Create a BO
  * @xe: The xe device.
@@ -2428,6 +2397,55 @@ struct xe_bo *xe_bo_create_user(struct xe_device *xe,
 	return bo;
 }
 
+/**
+ * xe_bo_create_pin_range_novm() - Create and pin a BO with range options.
+ * @xe: The xe device.
+ * @tile: The tile to select for migration of this bo, and the tile used for
+ * GGTT binding if any. Only to be non-NULL for ttm_bo_type_kernel bos.
+ * @size: The storage size to use for the bo.
+ * @start: Start of fixed VRAM range or 0.
+ * @end: End of fixed VRAM range or ~0ULL.
+ * @type: The TTM buffer object type.
+ * @flags: XE_BO_FLAG_ flags.
+ *
+ * Create an Xe BO with range- and options. If @start and @end indicate
+ * a fixed VRAM range, this must be a ttm_bo_type_kernel bo with VRAM placement
+ * only.
+ *
+ * Return: The buffer object on success. Negative error pointer on failure.
+ */
+struct xe_bo *xe_bo_create_pin_range_novm(struct xe_device *xe, struct xe_tile *tile,
+					  size_t size, u64 start, u64 end,
+					  enum ttm_bo_type type, u32 flags)
+{
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
+	struct xe_bo *bo;
+	int err = 0;
+
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {}, err) {
+		bo = __xe_bo_create_locked(xe, tile, NULL, size, start, end,
+					   0, type, flags, 0, &exec);
+		if (IS_ERR(bo)) {
+			drm_exec_retry_on_contention(&exec);
+			err = PTR_ERR(bo);
+			xe_validation_retry_on_oom(&ctx, &err);
+			break;
+		}
+
+		err = xe_bo_pin(bo, &exec);
+		xe_bo_unlock(bo);
+		if (err) {
+			xe_bo_put(bo);
+			drm_exec_retry_on_contention(&exec);
+			xe_validation_retry_on_oom(&ctx, &err);
+			break;
+		}
+	}
+
+	return err ? ERR_PTR(err) : bo;
+}
+
 static struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
 						     struct xe_tile *tile,
 						     struct xe_vm *vm,
@@ -2444,9 +2462,9 @@ static struct xe_bo *xe_bo_create_pin_map_at_aligned(struct xe_device *xe,
 	    xe_ttm_stolen_cpu_access_needs_ggtt(xe))
 		flags |= XE_BO_FLAG_GGTT;
 
-	bo = xe_bo_create_locked_range(xe, tile, vm, size, start, end, type,
-				       flags | XE_BO_FLAG_NEEDS_CPU_ACCESS | XE_BO_FLAG_PINNED,
-				       alignment, exec);
+	bo = __xe_bo_create_locked(xe, tile, vm, size, start, end, 0, type,
+				   flags | XE_BO_FLAG_NEEDS_CPU_ACCESS | XE_BO_FLAG_PINNED,
+				   alignment, exec);
 	if (IS_ERR(bo))
 		return bo;
 
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 143f59bcfb93..33d7c122dcc9 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -94,12 +94,6 @@ struct xe_bo *xe_bo_init_locked(struct xe_device *xe, struct xe_bo *bo,
 				struct ttm_lru_bulk_move *bulk, size_t size,
 				u16 cpu_caching, enum ttm_bo_type type,
 				u32 flags, struct drm_exec *exec);
-struct xe_bo *
-xe_bo_create_locked_range(struct xe_device *xe,
-			  struct xe_tile *tile, struct xe_vm *vm,
-			  size_t size, u64 start, u64 end,
-			  enum ttm_bo_type type, u32 flags, u64 alignment,
-			  struct drm_exec *exec);
 struct xe_bo *xe_bo_create_locked(struct xe_device *xe, struct xe_tile *tile,
 				  struct xe_vm *vm, size_t size,
 				  enum ttm_bo_type type, u32 flags,
@@ -113,6 +107,9 @@ struct xe_bo *xe_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 struct xe_bo *xe_bo_create_pin_map_novm(struct xe_device *xe, struct xe_tile *tile,
 					size_t size, enum ttm_bo_type type, u32 flags,
 					bool intr);
+struct xe_bo *xe_bo_create_pin_range_novm(struct xe_device *xe, struct xe_tile *tile,
+					  size_t size, u64 start, u64 end,
+					  enum ttm_bo_type type, u32 flags);
 struct xe_bo *
 xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile,
 			     size_t size, u64 offset, enum ttm_bo_type type,
diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index 906011671b60..6a9a752fe24a 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -1452,7 +1452,6 @@ static bool pf_release_vf_config_lmem(struct xe_gt *gt, struct xe_gt_sriov_confi
 static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size)
 {
 	struct xe_gt_sriov_config *config = pf_pick_vf_config(gt, vfid);
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
 	struct xe_device *xe = gt_to_xe(gt);
 	struct xe_tile *tile = gt_to_tile(gt);
 	struct xe_bo *bo;
@@ -1479,24 +1478,16 @@ static int pf_provision_vf_lmem(struct xe_gt *gt, unsigned int vfid, u64 size)
 		return 0;
 
 	xe_gt_assert(gt, pf_get_lmem_alignment(gt) == SZ_2M);
-	bo = xe_bo_create_locked(xe, tile, NULL,
-				 ALIGN(size, PAGE_SIZE),
-				 ttm_bo_type_kernel,
-				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				 XE_BO_FLAG_NEEDS_2M |
-				 XE_BO_FLAG_PINNED |
-				 XE_BO_FLAG_PINNED_LATE_RESTORE,
-				 exec);
+	bo = xe_bo_create_pin_range_novm(xe, tile,
+					 ALIGN(size, PAGE_SIZE), 0, ~0ull,
+					 ttm_bo_type_kernel,
+					 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+					 XE_BO_FLAG_NEEDS_2M |
+					 XE_BO_FLAG_PINNED |
+					 XE_BO_FLAG_PINNED_LATE_RESTORE);
 	if (IS_ERR(bo))
 		return PTR_ERR(bo);
 
-	err = xe_bo_pin(bo, exec);
-	xe_bo_unlock(bo);
-	if (unlikely(err)) {
-		xe_bo_put(bo);
-		return err;
-	}
-
 	config->lmem_obj = bo;
 
 	if (xe_device_has_lmtt(xe)) {
diff --git a/drivers/gpu/drm/xe/xe_psmi.c b/drivers/gpu/drm/xe/xe_psmi.c
index 2815987c190d..45d142191d60 100644
--- a/drivers/gpu/drm/xe/xe_psmi.c
+++ b/drivers/gpu/drm/xe/xe_psmi.c
@@ -68,10 +68,7 @@ static void psmi_cleanup(struct xe_device *xe)
 static struct xe_bo *psmi_alloc_object(struct xe_device *xe,
 				       unsigned int id, size_t bo_size)
 {
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
-	struct xe_bo *bo = NULL;
 	struct xe_tile *tile;
-	int err;
 
 	if (!id || !bo_size)
 		return NULL;
@@ -79,23 +76,12 @@ static struct xe_bo *psmi_alloc_object(struct xe_device *xe,
 	tile = &xe->tiles[id - 1];
 
 	/* VRAM: Allocate GEM object for the capture buffer */
-	bo = xe_bo_create_locked(xe, tile, NULL, bo_size,
-				 ttm_bo_type_kernel,
-				 XE_BO_FLAG_VRAM_IF_DGFX(tile) |
-				 XE_BO_FLAG_PINNED |
-				 XE_BO_FLAG_PINNED_LATE_RESTORE |
-				 XE_BO_FLAG_NEEDS_CPU_ACCESS,
-				 exec);
-
-	if (!IS_ERR(bo)) {
-		/* Buffer written by HW, ensure stays resident */
-		err = xe_bo_pin(bo, exec);
-		if (err)
-			bo = ERR_PTR(err);
-		xe_bo_unlock(bo);
-	}
-
-	return bo;
+	return xe_bo_create_pin_range_novm(xe, tile, bo_size, 0, ~0ull,
+					   ttm_bo_type_kernel,
+					   XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+					   XE_BO_FLAG_PINNED |
+					   XE_BO_FLAG_PINNED_LATE_RESTORE |
+					   XE_BO_FLAG_NEEDS_CPU_ACCESS);
 }
 
 /*

From 844150c255c94230aec41f88db8e2875cb97439e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Mon, 8 Sep 2025 12:12:46 +0200
Subject: [PATCH 1184/1455] drm/xe: Convert pinned suspend eviction for
 exhaustive eviction
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Pinned suspend eviction and preparation for eviction validates
system memory for eviction buffers. Do that under a
validation exclusive lock to avoid interfering with other
processes validating system graphics memory.

v2:
- Avoid gotos from within xe_validation_guard().
- Adapt to signature change of xe_validation_guard().

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/20250908101246.65025-14-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_bo.c | 184 +++++++++++++++++++++----------------
 1 file changed, 103 insertions(+), 81 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index 21efcb170239..e6d16b34e5b5 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -1141,43 +1141,47 @@ out_unref:
 int xe_bo_notifier_prepare_pinned(struct xe_bo *bo)
 {
 	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
 	struct xe_bo *backup;
 	int ret = 0;
 
-	xe_bo_lock(bo, false);
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.exclusive = true}, ret) {
+		ret = drm_exec_lock_obj(&exec, &bo->ttm.base);
+		drm_exec_retry_on_contention(&exec);
+		xe_assert(xe, !ret);
+		xe_assert(xe, !bo->backup_obj);
 
-	xe_assert(xe, !bo->backup_obj);
+		/*
+		 * Since this is called from the PM notifier we might have raced with
+		 * someone unpinning this after we dropped the pinned list lock and
+		 * grabbing the above bo lock.
+		 */
+		if (!xe_bo_is_pinned(bo))
+			break;
 
-	/*
-	 * Since this is called from the PM notifier we might have raced with
-	 * someone unpinning this after we dropped the pinned list lock and
-	 * grabbing the above bo lock.
-	 */
-	if (!xe_bo_is_pinned(bo))
-		goto out_unlock_bo;
+		if (!xe_bo_is_vram(bo))
+			break;
 
-	if (!xe_bo_is_vram(bo))
-		goto out_unlock_bo;
+		if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
+			break;
 
-	if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
-		goto out_unlock_bo;
+		backup = xe_bo_init_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
+					   DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
+					   XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
+					   XE_BO_FLAG_PINNED, &exec);
+		if (IS_ERR(backup)) {
+			drm_exec_retry_on_contention(&exec);
+			ret = PTR_ERR(backup);
+			xe_validation_retry_on_oom(&ctx, &ret);
+			break;
+		}
 
-	backup = xe_bo_init_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
-				   DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
-				   XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-				   XE_BO_FLAG_PINNED, exec);
-	if (IS_ERR(backup)) {
-		ret = PTR_ERR(backup);
-		goto out_unlock_bo;
+		backup->parent_obj = xe_bo_get(bo); /* Released by bo_destroy */
+		ttm_bo_pin(&backup->ttm);
+		bo->backup_obj = backup;
 	}
 
-	backup->parent_obj = xe_bo_get(bo); /* Released by bo_destroy */
-	ttm_bo_pin(&backup->ttm);
-	bo->backup_obj = backup;
-
-out_unlock_bo:
-	xe_bo_unlock(bo);
 	return ret;
 }
 
@@ -1203,57 +1207,12 @@ int xe_bo_notifier_unprepare_pinned(struct xe_bo *bo)
 	return 0;
 }
 
-/**
- * xe_bo_evict_pinned() - Evict a pinned VRAM object to system memory
- * @bo: The buffer object to move.
- *
- * On successful completion, the object memory will be moved to system memory.
- *
- * This is needed to for special handling of pinned VRAM object during
- * suspend-resume.
- *
- * Return: 0 on success. Negative error code on failure.
- */
-int xe_bo_evict_pinned(struct xe_bo *bo)
+static int xe_bo_evict_pinned_copy(struct xe_bo *bo, struct xe_bo *backup)
 {
-	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
-	struct drm_exec *exec = XE_VALIDATION_UNIMPLEMENTED;
-	struct xe_bo *backup = bo->backup_obj;
-	bool backup_created = false;
+	struct xe_device *xe = xe_bo_device(bo);
 	bool unmap = false;
 	int ret = 0;
 
-	xe_bo_lock(bo, false);
-
-	if (WARN_ON(!bo->ttm.resource)) {
-		ret = -EINVAL;
-		goto out_unlock_bo;
-	}
-
-	if (WARN_ON(!xe_bo_is_pinned(bo))) {
-		ret = -EINVAL;
-		goto out_unlock_bo;
-	}
-
-	if (!xe_bo_is_vram(bo))
-		goto out_unlock_bo;
-
-	if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
-		goto out_unlock_bo;
-
-	if (!backup) {
-		backup = xe_bo_init_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL, xe_bo_size(bo),
-					   DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
-					   XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
-					   XE_BO_FLAG_PINNED, exec);
-		if (IS_ERR(backup)) {
-			ret = PTR_ERR(backup);
-			goto out_unlock_bo;
-		}
-		backup->parent_obj = xe_bo_get(bo); /* Released by bo_destroy */
-		backup_created = true;
-	}
-
 	if (xe_bo_is_user(bo) || (bo->flags & XE_BO_FLAG_PINNED_LATE_RESTORE)) {
 		struct xe_migrate *migrate;
 		struct dma_fence *fence;
@@ -1286,7 +1245,7 @@ int xe_bo_evict_pinned(struct xe_bo *bo)
 		if (iosys_map_is_null(&bo->vmap)) {
 			ret = xe_bo_vmap(bo);
 			if (ret)
-				goto out_backup;
+				goto out_vunmap;
 			unmap = true;
 		}
 
@@ -1296,15 +1255,78 @@ int xe_bo_evict_pinned(struct xe_bo *bo)
 
 	if (!bo->backup_obj)
 		bo->backup_obj = backup;
-
-out_backup:
+out_vunmap:
 	xe_bo_vunmap(backup);
-	if (ret && backup_created)
-		xe_bo_put(backup);
-out_unlock_bo:
+out_backup:
 	if (unmap)
 		xe_bo_vunmap(bo);
-	xe_bo_unlock(bo);
+
+	return ret;
+}
+
+/**
+ * xe_bo_evict_pinned() - Evict a pinned VRAM object to system memory
+ * @bo: The buffer object to move.
+ *
+ * On successful completion, the object memory will be moved to system memory.
+ *
+ * This is needed to for special handling of pinned VRAM object during
+ * suspend-resume.
+ *
+ * Return: 0 on success. Negative error code on failure.
+ */
+int xe_bo_evict_pinned(struct xe_bo *bo)
+{
+	struct xe_device *xe = ttm_to_xe_device(bo->ttm.bdev);
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
+	struct xe_bo *backup = bo->backup_obj;
+	bool backup_created = false;
+	int ret = 0;
+
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {.exclusive = true}, ret) {
+		ret = drm_exec_lock_obj(&exec, &bo->ttm.base);
+		drm_exec_retry_on_contention(&exec);
+		xe_assert(xe, !ret);
+
+		if (WARN_ON(!bo->ttm.resource)) {
+			ret = -EINVAL;
+			break;
+		}
+
+		if (WARN_ON(!xe_bo_is_pinned(bo))) {
+			ret = -EINVAL;
+			break;
+		}
+
+		if (!xe_bo_is_vram(bo))
+			break;
+
+		if (bo->flags & XE_BO_FLAG_PINNED_NORESTORE)
+			break;
+
+		if (!backup) {
+			backup = xe_bo_init_locked(xe, NULL, NULL, bo->ttm.base.resv, NULL,
+						   xe_bo_size(bo),
+						   DRM_XE_GEM_CPU_CACHING_WB, ttm_bo_type_kernel,
+						   XE_BO_FLAG_SYSTEM | XE_BO_FLAG_NEEDS_CPU_ACCESS |
+						   XE_BO_FLAG_PINNED, &exec);
+			if (IS_ERR(backup)) {
+				drm_exec_retry_on_contention(&exec);
+				ret = PTR_ERR(backup);
+				xe_validation_retry_on_oom(&ctx, &ret);
+				break;
+			}
+			backup->parent_obj = xe_bo_get(bo); /* Released by bo_destroy */
+			backup_created = true;
+		}
+
+		ret = xe_bo_evict_pinned_copy(bo, backup);
+	}
+
+	if (ret && backup_created)
+		xe_bo_put(backup);
+
 	return ret;
 }
 

From 7acbe30813f04cccf7b2e8b571eb7936cfec0a87 Mon Sep 17 00:00:00 2001
From: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Date: Wed, 10 Sep 2025 10:55:25 +0200
Subject: [PATCH 1185/1455] MAINTAINERS: Remove Jacek Lawrynowicz as intel_vpu
 maintainer

Remove myself from the intel_vpu driver maintainer list as I'm
moving to another company. Time to let someone else deal with
the NPU bugs while I pretend to know what I'm doing elsewhere!

Thanks to everyone for the great collaboration (and for putting up
with my creative interpretations of what "minor fixes" means).

Reviewed-by: Maciej Falkowski <maciej.falkowski@linux.intel.com>
Reviewed-by: Karol Wachowski <karol.wachowski@linux.intel.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
Link: https://lore.kernel.org/r/20250910085526.230458-1-jacek.lawrynowicz@linux.intel.com
---
 MAINTAINERS | 1 -
 1 file changed, 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 402fe14091f1..8f8c91144cc2 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7490,7 +7490,6 @@ F:	drivers/soc/ti/smartreflex.c
 F:	include/linux/power/smartreflex.h
 
 DRM ACCEL DRIVERS FOR INTEL VPU
-M:	Jacek Lawrynowicz <jacek.lawrynowicz@linux.intel.com>
 M:	Maciej Falkowski <maciej.falkowski@linux.intel.com>
 M:	Karol Wachowski <karol.wachowski@linux.intel.com>
 L:	dri-devel@lists.freedesktop.org

From 6e490dea61b88aac9762c9f79a54aad4ea2e6cd1 Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <superm1@kernel.org>
Date: Mon, 11 Aug 2025 11:26:03 -0500
Subject: [PATCH 1186/1455] Fix access to video_is_primary_device() when
 compiled without CONFIG_VIDEO

When compiled without CONFIG_VIDEO the architecture specific
implementations of video_is_primary_device() include prototypes and
assume that video-common.c will be linked. Guard against this so that the
fallback inline implementation that returns false will be used when
compiled without CONFIG_VIDEO.

Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202506221312.49Fy1aNA-lkp@intel.com/
Link: https://lore.kernel.org/r/20250811162606.587759-2-superm1@kernel.org
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 arch/parisc/include/asm/video.h | 2 +-
 arch/sparc/include/asm/video.h  | 2 ++
 arch/x86/include/asm/video.h    | 2 ++
 3 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/parisc/include/asm/video.h b/arch/parisc/include/asm/video.h
index c5dff3223194..a9d50ebd6e76 100644
--- a/arch/parisc/include/asm/video.h
+++ b/arch/parisc/include/asm/video.h
@@ -6,7 +6,7 @@
 
 struct device;
 
-#if defined(CONFIG_STI_CORE)
+#if defined(CONFIG_STI_CORE) && defined(CONFIG_VIDEO)
 bool video_is_primary_device(struct device *dev);
 #define video_is_primary_device video_is_primary_device
 #endif
diff --git a/arch/sparc/include/asm/video.h b/arch/sparc/include/asm/video.h
index a6f48f52db58..773717b6d491 100644
--- a/arch/sparc/include/asm/video.h
+++ b/arch/sparc/include/asm/video.h
@@ -19,8 +19,10 @@ static inline pgprot_t pgprot_framebuffer(pgprot_t prot,
 #define pgprot_framebuffer pgprot_framebuffer
 #endif
 
+#ifdef CONFIG_VIDEO
 bool video_is_primary_device(struct device *dev);
 #define video_is_primary_device video_is_primary_device
+#endif
 
 static inline void fb_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
 {
diff --git a/arch/x86/include/asm/video.h b/arch/x86/include/asm/video.h
index 0950c9535fae..08ec328203ef 100644
--- a/arch/x86/include/asm/video.h
+++ b/arch/x86/include/asm/video.h
@@ -13,8 +13,10 @@ pgprot_t pgprot_framebuffer(pgprot_t prot,
 			    unsigned long offset);
 #define pgprot_framebuffer pgprot_framebuffer
 
+#ifdef CONFIG_VIDEO
 bool video_is_primary_device(struct device *dev);
 #define video_is_primary_device video_is_primary_device
+#endif
 
 #include <asm-generic/video.h>
 

From 337bf13aa9ddab8ae47f2677ab75cf69ae33437b Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <superm1@kernel.org>
Date: Mon, 11 Aug 2025 11:26:04 -0500
Subject: [PATCH 1187/1455] PCI/VGA: Replace vga_is_firmware_default() with a
 screen info check

vga_is_firmware_default() checks firmware resources to find the owner
framebuffer resources to find the firmware PCI device.  This is an
open coded implementation of screen_info_pci_dev().  Switch to using
screen_info_pci_dev() instead.

Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Suggested-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250811162606.587759-3-superm1@kernel.org
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 drivers/pci/vgaarb.c | 31 +++++--------------------------
 1 file changed, 5 insertions(+), 26 deletions(-)

diff --git a/drivers/pci/vgaarb.c b/drivers/pci/vgaarb.c
index 78748e8d2dba..b58f94ee4891 100644
--- a/drivers/pci/vgaarb.c
+++ b/drivers/pci/vgaarb.c
@@ -556,34 +556,13 @@ EXPORT_SYMBOL(vga_put);
 
 static bool vga_is_firmware_default(struct pci_dev *pdev)
 {
-#if defined(CONFIG_X86)
-	u64 base = screen_info.lfb_base;
-	u64 size = screen_info.lfb_size;
-	struct resource *r;
-	u64 limit;
+#ifdef CONFIG_SCREEN_INFO
+	struct screen_info *si = &screen_info;
 
-	/* Select the device owning the boot framebuffer if there is one */
-
-	if (screen_info.capabilities & VIDEO_CAPABILITY_64BIT_BASE)
-		base |= (u64)screen_info.ext_lfb_base << 32;
-
-	limit = base + size;
-
-	/* Does firmware framebuffer belong to us? */
-	pci_dev_for_each_resource(pdev, r) {
-		if (resource_type(r) != IORESOURCE_MEM)
-			continue;
-
-		if (!r->start || !r->end)
-			continue;
-
-		if (base < r->start || limit >= r->end)
-			continue;
-
-		return true;
-	}
-#endif
+	return pdev == screen_info_pci_dev(si);
+#else
 	return false;
+#endif
 }
 
 static bool vga_arb_integrated_gpu(struct device *dev)

From ad90860bd10ee3ed387077aed88828b139339976 Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <superm1@kernel.org>
Date: Mon, 11 Aug 2025 11:26:05 -0500
Subject: [PATCH 1188/1455] fbcon: Use screen info to find primary device

On systems with non VGA GPUs fbcon can't find the primary GPU because
video_is_primary_device() only checks the VGA arbiter.

Add a screen info check to video_is_primary_device() so that callers
can get accurate data on such systems.

Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Suggested-by: Thomas Zimmermann <tzimmermann@suse.de>
Suggested-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://lore.kernel.org/r/20250811162606.587759-4-superm1@kernel.org
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 arch/x86/video/video-common.c | 25 ++++++++++++++++++++++++-
 1 file changed, 24 insertions(+), 1 deletion(-)

diff --git a/arch/x86/video/video-common.c b/arch/x86/video/video-common.c
index 81fc97a2a837..e0aeee99bc99 100644
--- a/arch/x86/video/video-common.c
+++ b/arch/x86/video/video-common.c
@@ -9,6 +9,7 @@
 
 #include <linux/module.h>
 #include <linux/pci.h>
+#include <linux/screen_info.h>
 #include <linux/vgaarb.h>
 
 #include <asm/video.h>
@@ -27,6 +28,11 @@ EXPORT_SYMBOL(pgprot_framebuffer);
 
 bool video_is_primary_device(struct device *dev)
 {
+#ifdef CONFIG_SCREEN_INFO
+	struct screen_info *si = &screen_info;
+	struct resource res[SCREEN_INFO_MAX_RESOURCES];
+	ssize_t i, numres;
+#endif
 	struct pci_dev *pdev;
 
 	if (!dev_is_pci(dev))
@@ -34,7 +40,24 @@ bool video_is_primary_device(struct device *dev)
 
 	pdev = to_pci_dev(dev);
 
-	return (pdev == vga_default_device());
+	if (!pci_is_display(pdev))
+		return false;
+
+	if (pdev == vga_default_device())
+		return true;
+
+#ifdef CONFIG_SCREEN_INFO
+	numres = screen_info_resources(si, res, ARRAY_SIZE(res));
+	for (i = 0; i < numres; ++i) {
+		if (!(res[i].flags & IORESOURCE_MEM))
+			continue;
+
+		if (pci_find_resource(pdev, &res[i]))
+			return true;
+	}
+#endif
+
+	return false;
 }
 EXPORT_SYMBOL(video_is_primary_device);
 

From 7df7b728c3cff1d61253c39e38e39c33bb4e7d09 Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <superm1@kernel.org>
Date: Mon, 11 Aug 2025 11:26:06 -0500
Subject: [PATCH 1189/1455] DRM: Add a new 'boot_display' attribute

On systems with multiple GPUs there can be uncertainty which GPU is the
primary one used to drive the display at bootup. In some desktop
environments this can lead to increased power consumption because
secondary GPUs may be used for rendering and never go to a low power
state. In order to disambiguate this add a new sysfs attribute
'boot_display' that uses the output of video_is_primary_device() to
populate whether the PCI device was used for driving the display.

Suggested-by: Manivannan Sadhasivam <mani@kernel.org>
Acked-by: Manivannan Sadhasivam <mani@kernel.org>
Link: https://gitlab.freedesktop.org/xorg/lib/libpciaccess/-/issues/23
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250811162606.587759-5-superm1@kernel.org
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
---
 Documentation/ABI/testing/sysfs-class-drm |  8 +++++
 drivers/gpu/drm/drm_sysfs.c               | 41 +++++++++++++++++++++++
 2 files changed, 49 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-class-drm

diff --git a/Documentation/ABI/testing/sysfs-class-drm b/Documentation/ABI/testing/sysfs-class-drm
new file mode 100644
index 000000000000..d23fed5e29a7
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-class-drm
@@ -0,0 +1,8 @@
+What:		/sys/class/drm/.../boot_display
+Date:		January 2026
+Contact:	Linux DRI developers <dri-devel@vger.kernel.org>
+Description:
+		This file indicates that displays connected to the device were
+		used to display the boot sequence.  If a display connected to
+		the device was used to display the boot sequence the file will
+		be present and contain "1".
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index a455c56dbbeb..b01ffa4d6509 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -18,6 +18,7 @@
 #include <linux/gfp.h>
 #include <linux/i2c.h>
 #include <linux/kdev_t.h>
+#include <linux/pci.h>
 #include <linux/property.h>
 #include <linux/slab.h>
 
@@ -30,6 +31,8 @@
 #include <drm/drm_property.h>
 #include <drm/drm_sysfs.h>
 
+#include <asm/video.h>
+
 #include "drm_internal.h"
 #include "drm_crtc_internal.h"
 
@@ -508,6 +511,43 @@ void drm_sysfs_connector_property_event(struct drm_connector *connector,
 }
 EXPORT_SYMBOL(drm_sysfs_connector_property_event);
 
+static ssize_t boot_display_show(struct device *dev, struct device_attribute *attr,
+				 char *buf)
+{
+	return sysfs_emit(buf, "1\n");
+}
+static DEVICE_ATTR_RO(boot_display);
+
+static struct attribute *display_attrs[] = {
+	&dev_attr_boot_display.attr,
+	NULL
+};
+
+static umode_t boot_display_visible(struct kobject *kobj,
+				    struct attribute *a, int n)
+{
+	struct device *dev = kobj_to_dev(kobj)->parent;
+
+	if (dev_is_pci(dev)) {
+		struct pci_dev *pdev = to_pci_dev(dev);
+
+		if (video_is_primary_device(&pdev->dev))
+			return a->mode;
+	}
+
+	return 0;
+}
+
+static const struct attribute_group display_attr_group = {
+	.attrs = display_attrs,
+	.is_visible = boot_display_visible,
+};
+
+static const struct attribute_group *card_dev_groups[] = {
+	&display_attr_group,
+	NULL
+};
+
 struct device *drm_sysfs_minor_alloc(struct drm_minor *minor)
 {
 	const char *minor_str;
@@ -531,6 +571,7 @@ struct device *drm_sysfs_minor_alloc(struct drm_minor *minor)
 
 		kdev->devt = MKDEV(DRM_MAJOR, minor->index);
 		kdev->class = drm_class;
+		kdev->groups = card_dev_groups;
 		kdev->type = &drm_sysfs_device_minor;
 	}
 

From f03464c638ef10d5a9837d61176463f40e92da42 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 12:30:04 -0700
Subject: [PATCH 1190/1455] drm/msm/registers: Remove license/etc from
 generated headers

Since these generated files are no longer checked in, either in mesa or
in the linux kernel, simplify things by dropping the verbose generated
comment.

These were semi-nerf'd on the kernel side, in the name of build
reproducibility, by commit ba64c6737f86 ("drivers: gpu: drm: msm:
registers: improve reproducibility"), but in a way that was semi-
kernel specific.  We can just reduce the divergence between kernel
and mesa by just dropping all of this.

Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673551/
---
 drivers/gpu/drm/msm/registers/gen_header.py | 37 +--------------------
 1 file changed, 1 insertion(+), 36 deletions(-)

diff --git a/drivers/gpu/drm/msm/registers/gen_header.py b/drivers/gpu/drm/msm/registers/gen_header.py
index a409404627c7..56273a810c1d 100644
--- a/drivers/gpu/drm/msm/registers/gen_header.py
+++ b/drivers/gpu/drm/msm/registers/gen_header.py
@@ -444,9 +444,6 @@ class Parser(object):
 		self.variants = set()
 		self.file = []
 		self.xml_files = []
-		self.copyright_year = None
-		self.authors = []
-		self.license = None
 
 	def error(self, message):
 		parser, filename = self.stack[-1]
@@ -686,10 +683,6 @@ class Parser(object):
 			self.parse_field(attrs["name"], attrs)
 		elif name == "database":
 			self.do_validate(attrs["xsi:schemaLocation"])
-		elif name == "copyright":
-			self.copyright_year = attrs["year"]
-		elif name == "author":
-			self.authors.append(attrs["name"] + " <" + attrs["email"] + "> " + attrs["name"])
 
 	def end_element(self, name):
 		if name == "domain":
@@ -706,8 +699,6 @@ class Parser(object):
 			self.current_array = self.current_array.parent
 		elif name == "enum":
 			self.current_enum = None
-		elif name == "license":
-			self.license = self.cdata
 
 	def character_data(self, data):
 		self.cdata += data
@@ -868,33 +859,7 @@ def dump_c(args, guard, func):
 
 	print("#ifndef %s\n#define %s\n" % (guard, guard))
 
-	print("""/* Autogenerated file, DO NOT EDIT manually!
-
-This file was generated by the rules-ng-ng gen_header.py tool in this git repository:
-http://gitlab.freedesktop.org/mesa/mesa/
-git clone https://gitlab.freedesktop.org/mesa/mesa.git
-
-The rules-ng-ng source files this header was generated from are:
-""")
-	maxlen = 0
-	for filepath in p.xml_files:
-		new_filepath = re.sub("^.+drivers","drivers",filepath)
-		maxlen = max(maxlen, len(new_filepath))
-	for filepath in p.xml_files:
-		pad = " " * (maxlen - len(new_filepath))
-		filesize = str(os.path.getsize(filepath))
-		filesize = " " * (7 - len(filesize)) + filesize
-		filetime = time.ctime(os.path.getmtime(filepath))
-		print("- " + new_filepath + pad + " (" + filesize + " bytes, from <stripped>)")
-	if p.copyright_year:
-		current_year = str(datetime.date.today().year)
-		print()
-		print("Copyright (C) %s-%s by the following authors:" % (p.copyright_year, current_year))
-		for author in p.authors:
-			print("- " + author)
-	if p.license:
-		print(p.license)
-	print("*/")
+	print("/* Autogenerated file, DO NOT EDIT manually! */")
 
 	print()
 	print("#ifdef __KERNEL__")

From 90528176204b7ae003477e0b37a24d540f26531d Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 12:30:05 -0700
Subject: [PATCH 1191/1455] drm/msm/registers: Sync gen_header.py from mesa

Sync from mesa commit 04e2140d8be7 ("freedreno/registers: remove python
3.9 dependency for compiling msm").

Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673556/
---
 drivers/gpu/drm/msm/registers/gen_header.py | 157 +++++++++++++-------
 1 file changed, 107 insertions(+), 50 deletions(-)

diff --git a/drivers/gpu/drm/msm/registers/gen_header.py b/drivers/gpu/drm/msm/registers/gen_header.py
index 56273a810c1d..16239b754804 100644
--- a/drivers/gpu/drm/msm/registers/gen_header.py
+++ b/drivers/gpu/drm/msm/registers/gen_header.py
@@ -11,7 +11,6 @@ import collections
 import argparse
 import time
 import datetime
-import re
 
 class Error(Exception):
 	def __init__(self, message):
@@ -31,7 +30,7 @@ class Enum(object):
 	def names(self):
 		return [n for (n, value) in self.values]
 
-	def dump(self):
+	def dump(self, is_deprecated):
 		use_hex = False
 		for (name, value) in self.values:
 			if value > 0x1000:
@@ -45,7 +44,7 @@ class Enum(object):
 				print("\t%s = %d," % (name, value))
 		print("};\n")
 
-	def dump_pack_struct(self):
+	def dump_pack_struct(self, is_deprecated):
 		pass
 
 class Field(object):
@@ -70,11 +69,11 @@ class Field(object):
 			raise parser.error("booleans should be 1 bit fields")
 		elif self.type == "float" and not (high - low == 31 or high - low == 15):
 			raise parser.error("floats should be 16 or 32 bit fields")
-		elif not self.type in builtin_types and not self.type in parser.enums:
+		elif self.type not in builtin_types and self.type not in parser.enums:
 			raise parser.error("unknown type '%s'" % self.type)
 
 	def ctype(self, var_name):
-		if self.type == None:
+		if self.type is None:
 			type = "uint32_t"
 			val = var_name
 		elif self.type == "boolean":
@@ -124,7 +123,7 @@ def field_name(reg, f):
 		name = f.name.lower()
 	else:
 		# We hit this path when a reg is defined with no bitset fields, ie.
-		# 	<reg32 offset="0x88db" name="RB_BLIT_DST_ARRAY_PITCH" low="0" high="28" shr="6" type="uint"/>
+		# 	<reg32 offset="0x88db" name="RB_RESOLVE_SYSTEM_BUFFER_ARRAY_PITCH" low="0" high="28" shr="6" type="uint"/>
 		name = reg.name.lower()
 
 	if (name in [ "double", "float", "int" ]) or not (name[0].isalpha()):
@@ -146,6 +145,18 @@ def indices_strides(indices):
 					"%s(i%d)" % (offset, idx)
 			for (idx, (ctype, stride, offset)) in  enumerate(indices)])
 
+def is_number(str):
+	try:
+		int(str)
+		return True
+	except ValueError:
+		return False
+
+def sanitize_variant(variant):
+	if variant and "-" in variant:
+		return variant[:variant.index("-")]
+	return variant
+
 class Bitset(object):
 	def __init__(self, name, template):
 		self.name = name
@@ -175,11 +186,7 @@ class Bitset(object):
 		print("#endif\n")
 
 		print("    return (struct fd_reg_pair) {")
-		if reg.array:
-			print("        .reg = REG_%s(__i)," % reg.full_name)
-		else:
-			print("        .reg = REG_%s," % reg.full_name)
-
+		print("        .reg = (uint32_t)%s," % reg.reg_offset())
 		print("        .value =")
 		for f in self.fields:
 			if f.type in [ "address", "waddress" ]:
@@ -204,7 +211,7 @@ class Bitset(object):
 
 		print("    };")
 
-	def dump_pack_struct(self, reg=None):
+	def dump_pack_struct(self, is_deprecated, reg=None):
 		if not reg:
 			return
 
@@ -229,12 +236,15 @@ class Bitset(object):
 			tab_to("    uint32_t", "dword;")
 		print("};\n")
 
+		depcrstr = ""
+		if is_deprecated:
+			depcrstr = " FD_DEPRECATED"
 		if reg.array:
-			print("static inline struct fd_reg_pair\npack_%s(uint32_t __i, struct %s fields)\n{" %
-				  (prefix, prefix))
+			print("static inline%s struct fd_reg_pair\npack_%s(uint32_t __i, struct %s fields)\n{" %
+				  (depcrstr, prefix, prefix))
 		else:
-			print("static inline struct fd_reg_pair\npack_%s(struct %s fields)\n{" %
-				  (prefix, prefix))
+			print("static inline%s struct fd_reg_pair\npack_%s(struct %s fields)\n{" %
+				  (depcrstr, prefix, prefix))
 
 		self.dump_regpair_builder(reg)
 
@@ -253,8 +263,8 @@ class Bitset(object):
 				  (prefix, prefix, prefix, skip))
 
 
-	def dump(self, prefix=None):
-		if prefix == None:
+	def dump(self, is_deprecated, prefix=None):
+		if prefix is None:
 			prefix = self.name
 		for f in self.fields:
 			if f.name:
@@ -262,9 +272,9 @@ class Bitset(object):
 			else:
 				name = prefix
 
-			if not f.name and f.low == 0 and f.shr == 0 and not f.type in ["float", "fixed", "ufixed"]:
+			if not f.name and f.low == 0 and f.shr == 0 and f.type not in ["float", "fixed", "ufixed"]:
 				pass
-			elif f.type == "boolean" or (f.type == None and f.low == f.high):
+			elif f.type == "boolean" or (f.type is None and f.low == f.high):
 				tab_to("#define %s" % name, "0x%08x" % (1 << f.low))
 			else:
 				tab_to("#define %s__MASK" % name, "0x%08x" % mask(f.low, f.high))
@@ -286,6 +296,7 @@ class Array(object):
 		self.domain = domain
 		self.variant = variant
 		self.parent = parent
+		self.children = []
 		if self.parent:
 			self.name = self.parent.name + "_" + self.local_name
 		else:
@@ -337,12 +348,15 @@ class Array(object):
 			offset += self.parent.total_offset()
 		return offset
 
-	def dump(self):
+	def dump(self, is_deprecated):
+		depcrstr = ""
+		if is_deprecated:
+			depcrstr = " FD_DEPRECATED"
 		proto = indices_varlist(self.indices())
 		strides = indices_strides(self.indices())
 		array_offset = self.total_offset()
 		if self.fixed_offsets:
-			print("static inline uint32_t __offset_%s(%s idx)" % (self.local_name, self.index_ctype()))
+			print("static inline%s uint32_t __offset_%s(%s idx)" % (depcrstr, self.local_name, self.index_ctype()))
 			print("{\n\tswitch (idx) {")
 			if self.index_type:
 				for val, offset in zip(self.index_type.names(), self.offsets):
@@ -357,7 +371,7 @@ class Array(object):
 		else:
 			tab_to("#define REG_%s_%s(%s)" % (self.domain, self.name, proto), "(0x%08x + %s )\n" % (array_offset, strides))
 
-	def dump_pack_struct(self):
+	def dump_pack_struct(self, is_deprecated):
 		pass
 
 	def dump_regpair_builder(self):
@@ -373,6 +387,7 @@ class Reg(object):
 		self.bit_size = bit_size
 		if array:
 			self.name = array.name + "_" + self.name
+			array.children.append(self)
 		self.full_name = self.domain + "_" + self.name
 		if "stride" in attrs:
 			self.stride = int(attrs["stride"], 0)
@@ -397,25 +412,34 @@ class Reg(object):
 		else:
 			return self.offset
 
-	def dump(self):
+	def reg_offset(self):
+		if self.array:
+			offset = self.array.offset + self.offset
+			return "(0x%08x + 0x%x*__i)" % (offset, self.array.stride)
+		return "0x%08x" % self.offset
+
+	def dump(self, is_deprecated):
+		depcrstr = ""
+		if is_deprecated:
+			depcrstr = " FD_DEPRECATED "
 		proto = indices_prototype(self.indices())
 		strides = indices_strides(self.indices())
 		offset = self.total_offset()
 		if proto == '':
 			tab_to("#define REG_%s" % self.full_name, "0x%08x" % offset)
 		else:
-			print("static inline uint32_t REG_%s(%s) { return 0x%08x + %s; }" % (self.full_name, proto, offset, strides))
+			print("static inline%s uint32_t REG_%s(%s) { return 0x%08x + %s; }" % (depcrstr, self.full_name, proto, offset, strides))
 
 		if self.bitset.inline:
-			self.bitset.dump(self.full_name)
+			self.bitset.dump(is_deprecated, self.full_name)
+		print("")
 
-	def dump_pack_struct(self):
+	def dump_pack_struct(self, is_deprecated):
 		if self.bitset.inline:
-			self.bitset.dump_pack_struct(self)
+			self.bitset.dump_pack_struct(is_deprecated, self)
 
 	def dump_regpair_builder(self):
-		if self.bitset.inline:
-			self.bitset.dump_regpair_builder(self)
+		self.bitset.dump_regpair_builder(self)
 
 	def dump_py(self):
 		print("\tREG_%s = 0x%08x" % (self.full_name, self.offset))
@@ -451,7 +475,7 @@ class Parser(object):
 
 	def prefix(self, variant=None):
 		if self.current_prefix_type == "variant" and variant:
-			return variant
+			return sanitize_variant(variant)
 		elif self.current_stripe:
 			return self.current_stripe + "_" + self.current_domain
 		elif self.current_prefix:
@@ -497,15 +521,22 @@ class Parser(object):
 		return varset
 
 	def parse_variants(self, attrs):
-		if not "variants" in attrs:
+		if "variants" not in attrs:
 				return None
-		variant = attrs["variants"].split(",")[0]
-		if "-" in variant:
-			variant = variant[:variant.index("-")]
 
+		variant = attrs["variants"].split(",")[0]
 		varset = self.parse_varset(attrs)
 
-		assert varset.has_name(variant)
+		if "-" in variant:
+			# if we have a range, validate that both the start and end
+			# of the range are valid enums:
+			start = variant[:variant.index("-")]
+			end = variant[variant.index("-") + 1:]
+			assert varset.has_name(start)
+			if end != "":
+				assert varset.has_name(end)
+		else:
+			assert varset.has_name(variant)
 
 		return variant
 
@@ -569,9 +600,6 @@ class Parser(object):
 				error_str = str(xmlschema.error_log.filter_from_errors()[0])
 				raise self.error("Schema validation failed for: " + filename + "\n" + error_str)
 		except ImportError as e:
-			if self.validate:
-				raise e
-
 			print("lxml not found, skipping validation", file=sys.stderr)
 
 	def do_parse(self, filename):
@@ -640,7 +668,7 @@ class Parser(object):
 		elif name == "domain":
 			self.current_domain = attrs["name"]
 			if "prefix" in attrs:
-				self.current_prefix = self.parse_variants(attrs)
+				self.current_prefix = sanitize_variant(self.parse_variants(attrs))
 				self.current_prefix_type = attrs["prefix"]
 			else:
 				self.current_prefix = None
@@ -648,7 +676,7 @@ class Parser(object):
 			if "varset" in attrs:
 				self.current_varset = self.enums[attrs["varset"]]
 		elif name == "stripe":
-			self.current_stripe = self.parse_variants(attrs)
+			self.current_stripe = sanitize_variant(self.parse_variants(attrs))
 		elif name == "enum":
 			self.current_enum_value = 0
 			self.current_enum = Enum(attrs["name"])
@@ -696,6 +724,13 @@ class Parser(object):
 		elif name == "reg32":
 			self.current_reg = None
 		elif name == "array":
+			# if the array has no Reg children, push an implicit reg32:
+			if len(self.current_array.children) == 0:
+				attrs = {
+					"name": "REG",
+					"offset": "0",
+				}
+				self.parse_reg(attrs, 32)
 			self.current_array = self.current_array.parent
 		elif name == "enum":
 			self.current_enum = None
@@ -711,10 +746,10 @@ class Parser(object):
 				if variants:
 					for variant, vreg in variants.items():
 						if reg == vreg:
-							d[(usage, variant)].append(reg)
+							d[(usage, sanitize_variant(variant))].append(reg)
 				else:
 					for variant in self.variants:
-						d[(usage, variant)].append(reg)
+						d[(usage, sanitize_variant(variant))].append(reg)
 
 		print("#ifdef __cplusplus")
 
@@ -744,6 +779,9 @@ class Parser(object):
 
 		print("#endif")
 
+	def has_variants(self, reg):
+		return reg.name in self.variant_regs and not is_number(reg.name) and not is_number(reg.name[1:])
+
 	def dump(self):
 		enums = []
 		bitsets = []
@@ -757,7 +795,7 @@ class Parser(object):
 				regs.append(e)
 
 		for e in enums + bitsets + regs:
-			e.dump()
+			e.dump(self.has_variants(e))
 
 		self.dump_reg_usages()
 
@@ -773,8 +811,7 @@ class Parser(object):
 
 
 	def dump_reg_variants(self, regname, variants):
-		# Don't bother for things that only have a single variant:
-		if len(variants) == 1:
+		if is_number(regname) or is_number(regname[1:]):
 			return
 		print("#ifdef __cplusplus")
 		print("struct __%s {" % regname)
@@ -825,11 +862,20 @@ class Parser(object):
 			xtravar = "__i, "
 		print("__%s(%sstruct __%s fields) {" % (regname, xtra, regname))
 		for variant in variants.keys():
-			print("  if (%s == %s) {" % (varenum.upper(), variant))
+			if "-" in variant:
+				start = variant[:variant.index("-")]
+				end = variant[variant.index("-") + 1:]
+				if end != "":
+					print("  if ((%s >= %s) && (%s <= %s)) {" % (varenum.upper(), start, varenum.upper(), end))
+				else:
+					print("  if (%s >= %s) {" % (varenum.upper(), start))
+			else:
+				print("  if (%s == %s) {" % (varenum.upper(), variant))
 			reg = variants[variant]
 			reg.dump_regpair_builder()
 			print("  } else")
 		print("    assert(!\"invalid variant\");")
+		print("  return (struct fd_reg_pair){};")
 		print("}")
 
 		if bit_size == 64:
@@ -842,7 +888,7 @@ class Parser(object):
 
 	def dump_structs(self):
 		for e in self.file:
-			e.dump_pack_struct()
+			e.dump_pack_struct(self.has_variants(e))
 
 		for regname in self.variant_regs:
 			self.dump_reg_variants(regname, self.variant_regs[regname])
@@ -877,9 +923,20 @@ def dump_c(args, guard, func):
 	print("#endif")
 	print()
 
+	print("#ifndef FD_NO_DEPRECATED_PACK")
+	print("#define FD_DEPRECATED __attribute__((deprecated))")
+	print("#else")
+	print("#define FD_DEPRECATED")
+	print("#endif")
+	print()
+
 	func(p)
 
-	print("\n#endif /* %s */" % guard)
+	print()
+	print("#undef FD_DEPRECATED")
+	print()
+
+	print("#endif /* %s */" % guard)
 
 
 def dump_c_defines(args):
@@ -896,7 +953,7 @@ def dump_py_defines(args):
 	p = Parser()
 
 	try:
-		p.parse(args.rnn, args.xml)
+		p.parse(args.rnn, args.xml, args.validate)
 	except Error as e:
 		print(e, file=sys.stderr)
 		exit(1)

From 29e087f31bb104c3ad71396e3a2f94982f2801d2 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 12:30:06 -0700
Subject: [PATCH 1192/1455] drm/msm/registers: Make TPL1_BICUBIC_WEIGHTS_TABLE
 an array

Synced from mesa commit 77c42c1a5752 ("freedreno/registers: Make
TPL1_BICUBIC_WEIGHTS_TABLE an array").

Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673552/
---
 drivers/gpu/drm/msm/adreno/a6xx_catalog.c     | 10 +++++-----
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c         | 10 +++++-----
 drivers/gpu/drm/msm/registers/adreno/a6xx.xml | 16 ++++++----------
 3 files changed, 16 insertions(+), 20 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
index a2d587e1a4f5..44df6410bce1 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_catalog.c
@@ -1355,11 +1355,11 @@ DECLARE_ADRENO_REGLIST_LIST(a7xx_pwrup_reglist);
 
 /* Applicable for X185, A750 */
 static const u32 a750_ifpc_reglist_regs[] = {
-	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_0,
-	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_1,
-	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_2,
-	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_3,
-	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_4,
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(0),
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(1),
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(2),
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(3),
+	REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(4),
 	REG_A6XX_TPL1_NC_MODE_CNTL,
 	REG_A6XX_SP_NC_MODE_CNTL,
 	REG_A6XX_CP_DBG_ECO_CNTL,
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index 2f68394d6c3b..a45c3917ae9b 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -1337,14 +1337,14 @@ static int hw_init(struct msm_gpu *gpu)
 
 	/* Set weights for bicubic filtering */
 	if (adreno_is_a650_family(adreno_gpu) || adreno_is_x185(adreno_gpu)) {
-		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_0, 0);
-		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_1,
+		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(0), 0);
+		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(1),
 			0x3fe05ff4);
-		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_2,
+		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(2),
 			0x3fa0ebee);
-		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_3,
+		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(3),
 			0x3f5193ed);
-		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE_4,
+		gpu_write(gpu, REG_A6XX_TPL1_BICUBIC_WEIGHTS_TABLE(4),
 			0x3f0243f0);
 	}
 
diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx.xml
index 86fab2750ba7..28d4e7149a5c 100644
--- a/drivers/gpu/drm/msm/registers/adreno/a6xx.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/a6xx.xml
@@ -3296,17 +3296,13 @@ by a particular renderpass/blit.
 	</reg32>
 	<reg32 offset="0xb605" name="TPL1_UNKNOWN_B605" low="0" high="7" type="uint" variants="A6XX" usage="cmd"/> <!-- always 0x0 or 0x44 ? -->
 
-	<reg32 offset="0xb608" name="TPL1_BICUBIC_WEIGHTS_TABLE_0" low="0" high="29" variants="A6XX"/>
-	<reg32 offset="0xb609" name="TPL1_BICUBIC_WEIGHTS_TABLE_1" low="0" high="29" variants="A6XX"/>
-	<reg32 offset="0xb60a" name="TPL1_BICUBIC_WEIGHTS_TABLE_2" low="0" high="29" variants="A6XX"/>
-	<reg32 offset="0xb60b" name="TPL1_BICUBIC_WEIGHTS_TABLE_3" low="0" high="29" variants="A6XX"/>
-	<reg32 offset="0xb60c" name="TPL1_BICUBIC_WEIGHTS_TABLE_4" low="0" high="29" variants="A6XX"/>
+	<array offset="0xb608" name="TPL1_BICUBIC_WEIGHTS_TABLE" stride="1" length="5" variants="A6XX">
+		<reg32 offset="0" name="REG" low="0" high="29"/>
+	</array>
 
-	<reg32 offset="0xb608" name="TPL1_BICUBIC_WEIGHTS_TABLE_0" low="0" high="29" variants="A7XX" usage="cmd"/>
-	<reg32 offset="0xb609" name="TPL1_BICUBIC_WEIGHTS_TABLE_1" low="0" high="29" variants="A7XX" usage="cmd"/>
-	<reg32 offset="0xb60a" name="TPL1_BICUBIC_WEIGHTS_TABLE_2" low="0" high="29" variants="A7XX" usage="cmd"/>
-	<reg32 offset="0xb60b" name="TPL1_BICUBIC_WEIGHTS_TABLE_3" low="0" high="29" variants="A7XX" usage="cmd"/>
-	<reg32 offset="0xb60c" name="TPL1_BICUBIC_WEIGHTS_TABLE_4" low="0" high="29" variants="A7XX" usage="cmd"/>
+	<array offset="0xb608" name="TPL1_BICUBIC_WEIGHTS_TABLE" stride="1" length="5" variants="A7XX">
+		<reg32 offset="0" name="REG" low="0" high="29" usage="cmd"/>
+	</array>
 
 	<array offset="0xb610" name="TPL1_PERFCTR_TP_SEL" stride="1" length="12" variants="A6XX"/>
 	<array offset="0xb610" name="TPL1_PERFCTR_TP_SEL" stride="1" length="18" variants="A7XX"/>

From 60e9f776b7932d67c88e8475df7830cb9cdf3154 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 12:30:07 -0700
Subject: [PATCH 1193/1455] drm/msm/registers: Generate _HI/LO builders for
 reg64

The upstream mesa copy of the GPU regs has shifted more things to reg64
instead of seperate 32b HI/LO reg32's.  This works better with the "new-
style" c++ builders that mesa has been migrating to for a6xx+ (to better
handle register shuffling between gens), but it leaves the C builders
with missing _HI/LO builders.

So handle the special case of reg64, automatically generating the
missing _HI/LO builders.

Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673559/
---
 drivers/gpu/drm/msm/registers/gen_header.py | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/msm/registers/gen_header.py b/drivers/gpu/drm/msm/registers/gen_header.py
index 16239b754804..1d603dadfabd 100644
--- a/drivers/gpu/drm/msm/registers/gen_header.py
+++ b/drivers/gpu/drm/msm/registers/gen_header.py
@@ -161,6 +161,7 @@ class Bitset(object):
 	def __init__(self, name, template):
 		self.name = name
 		self.inline = False
+		self.reg = None
 		if template:
 			self.fields = template.fields[:]
 		else:
@@ -266,6 +267,11 @@ class Bitset(object):
 	def dump(self, is_deprecated, prefix=None):
 		if prefix is None:
 			prefix = self.name
+		if self.reg and self.reg.bit_size == 64:
+			print("static inline uint32_t %s_LO(uint32_t val)\n{" % prefix)
+			print("\treturn val;\n}")
+			print("static inline uint32_t %s_HI(uint32_t val)\n{" % prefix)
+			print("\treturn val;\n}")
 		for f in self.fields:
 			if f.name:
 				name = prefix + "_" + f.name
@@ -645,6 +651,7 @@ class Parser(object):
 
 		self.current_reg = Reg(attrs, self.prefix(variant), self.current_array, bit_size)
 		self.current_reg.bitset = self.current_bitset
+		self.current_bitset.reg = self.current_reg
 
 		if len(self.stack) == 1:
 			self.file.append(self.current_reg)

From b5bad77e1e3c7249e4c0c88f98477e1ee7669b63 Mon Sep 17 00:00:00 2001
From: Rob Clark <robin.clark@oss.qualcomm.com>
Date: Mon, 8 Sep 2025 12:30:08 -0700
Subject: [PATCH 1194/1455] drm/msm/registers: Sync GPU registers from mesa

In particular, to pull in a SP_READ_SEL_LOCATION bitfield size fix to
fix a7xx GPU snapshot.

Sync from mesa commit 15ee3873aa4d ("freedreno/registers: Update GMU
register xml").

Cc: Karmjit Mahil <karmjit.mahil@igalia.com>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Patchwork: https://patchwork.freedesktop.org/patch/673558/
---
 drivers/gpu/drm/msm/adreno/a6xx_gpu.c         |   8 +-
 drivers/gpu/drm/msm/adreno/a6xx_preempt.c     |   4 +-
 drivers/gpu/drm/msm/registers/adreno/a6xx.xml | 710 ++++++++++--------
 .../msm/registers/adreno/a6xx_descriptors.xml |  40 -
 .../drm/msm/registers/adreno/a6xx_enums.xml   |  50 +-
 .../drm/msm/registers/adreno/adreno_pm4.xml   | 179 ++---
 6 files changed, 512 insertions(+), 479 deletions(-)

diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
index a45c3917ae9b..b8f8ae940b55 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
@@ -264,8 +264,8 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu,
 	 * Needed for preemption
 	 */
 	OUT_PKT7(ring, CP_MEM_WRITE, 5);
-	OUT_RING(ring, CP_MEM_WRITE_0_ADDR_LO(lower_32_bits(memptr)));
-	OUT_RING(ring, CP_MEM_WRITE_1_ADDR_HI(upper_32_bits(memptr)));
+	OUT_RING(ring, A5XX_CP_MEM_WRITE_ADDR_LO(lower_32_bits(memptr)));
+	OUT_RING(ring, A5XX_CP_MEM_WRITE_ADDR_HI(upper_32_bits(memptr)));
 	OUT_RING(ring, lower_32_bits(ttbr));
 	OUT_RING(ring, upper_32_bits(ttbr));
 	OUT_RING(ring, ctx->seqno);
@@ -295,9 +295,9 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu,
 		 */
 		OUT_PKT7(ring, CP_WAIT_REG_MEM, 6);
 		OUT_RING(ring, CP_WAIT_REG_MEM_0_FUNCTION(WRITE_EQ));
-		OUT_RING(ring, CP_WAIT_REG_MEM_1_POLL_ADDR_LO(
+		OUT_RING(ring, CP_WAIT_REG_MEM_POLL_ADDR_LO(
 				REG_A6XX_RBBM_PERFCTR_SRAM_INIT_STATUS));
-		OUT_RING(ring, CP_WAIT_REG_MEM_2_POLL_ADDR_HI(0));
+		OUT_RING(ring, CP_WAIT_REG_MEM_POLL_ADDR_HI(0));
 		OUT_RING(ring, CP_WAIT_REG_MEM_3_REF(0x1));
 		OUT_RING(ring, CP_WAIT_REG_MEM_4_MASK(0x1));
 		OUT_RING(ring, CP_WAIT_REG_MEM_5_DELAY_LOOP_CYCLES(0));
diff --git a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
index 2ce073d1cf8c..afc5f4aa3b17 100644
--- a/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
+++ b/drivers/gpu/drm/msm/adreno/a6xx_preempt.c
@@ -111,9 +111,9 @@ static void preempt_prepare_postamble(struct a6xx_gpu *a6xx_gpu)
 
 	postamble[count++] = PKT7(CP_WAIT_REG_MEM, 6);
 	postamble[count++] = CP_WAIT_REG_MEM_0_FUNCTION(WRITE_EQ);
-	postamble[count++] = CP_WAIT_REG_MEM_1_POLL_ADDR_LO(
+	postamble[count++] = CP_WAIT_REG_MEM_POLL_ADDR_LO(
 				REG_A6XX_RBBM_PERFCTR_SRAM_INIT_STATUS);
-	postamble[count++] = CP_WAIT_REG_MEM_2_POLL_ADDR_HI(0);
+	postamble[count++] = CP_WAIT_REG_MEM_POLL_ADDR_HI(0);
 	postamble[count++] = CP_WAIT_REG_MEM_3_REF(0x1);
 	postamble[count++] = CP_WAIT_REG_MEM_4_MASK(0x1);
 	postamble[count++] = CP_WAIT_REG_MEM_5_DELAY_LOOP_CYCLES(0);
diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx.xml
index 28d4e7149a5c..9459b6038217 100644
--- a/drivers/gpu/drm/msm/registers/adreno/a6xx.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/a6xx.xml
@@ -814,7 +814,7 @@ by a particular renderpass/blit.
 		<bitfield name="Y" low="16" high="29" type="uint"/>
 	</bitset>
 
-	<reg32 offset="0x8000" name="GRAS_CL_CNTL" usage="rp_blit">
+	<bitset name="a6xx_gras_cl_cntl" inline="yes">
 		<bitfield name="CLIP_DISABLE" pos="0" type="boolean"/>
 		<bitfield name="ZNEAR_CLIP_DISABLE" pos="1" type="boolean"/>
 		<bitfield name="ZFAR_CLIP_DISABLE" pos="2" type="boolean"/>
@@ -826,18 +826,20 @@ by a particular renderpass/blit.
 		<bitfield name="VP_CLIP_CODE_IGNORE" pos="7" type="boolean"/>
 		<bitfield name="VP_XFORM_DISABLE" pos="8" type="boolean"/>
 		<bitfield name="PERSP_DIVISION_DISABLE" pos="9" type="boolean"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x8000" name="GRAS_CL_CNTL" type="a6xx_gras_cl_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<bitset name="a6xx_gras_xs_clip_cull_distance" inline="yes">
 		<bitfield name="CLIP_MASK" low="0" high="7"/>
 		<bitfield name="CULL_MASK" low="8" high="15"/>
 	</bitset>
-	<reg32 offset="0x8001" name="GRAS_CL_VS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit"/>
-	<reg32 offset="0x8002" name="GRAS_CL_DS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit"/>
-	<reg32 offset="0x8003" name="GRAS_CL_GS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit"/>
-	<reg32 offset="0x8004" name="GRAS_CL_ARRAY_SIZE" low="0" high="10" type="uint" usage="rp_blit"/>
+	<reg32 offset="0x8001" name="GRAS_CL_VS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit" variants="A6XX-A7XX" />
+	<reg32 offset="0x8002" name="GRAS_CL_DS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit" variants="A6XX-A7XX" />
+	<reg32 offset="0x8003" name="GRAS_CL_GS_CLIP_CULL_DISTANCE" type="a6xx_gras_xs_clip_cull_distance" usage="rp_blit" variants="A6XX-A7XX" />
+	<reg32 offset="0x8004" name="GRAS_CL_ARRAY_SIZE" low="0" high="10" type="uint" usage="rp_blit" variants="A6XX-A7XX" />
 
-	<reg32 offset="0x8005" name="GRAS_CL_INTERP_CNTL" usage="rp_blit">
+	<bitset name="a6xx_gras_cl_interp_cntl" inline="yes">
 		<!-- see also RB_INTERP_CNTL -->
 		<bitfield name="IJ_PERSP_PIXEL" pos="0" type="boolean"/>
 		<bitfield name="IJ_PERSP_CENTROID" pos="1" type="boolean"/>
@@ -848,26 +850,69 @@ by a particular renderpass/blit.
 		<bitfield name="COORD_MASK" low="6" high="9" type="hex"/>
 		<bitfield name="UNK10" pos="10" type="boolean" variants="A7XX-"/>
 		<bitfield name="UNK11" pos="11" type="boolean" variants="A7XX-"/>
-	</reg32>
-	<reg32 offset="0x8006" name="GRAS_CL_GUARDBAND_CLIP_ADJ" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x8005" name="GRAS_CL_INTERP_CNTL" type="a6xx_gras_cl_interp_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_cl_guardband_clip_adj" inline="true">
 		<bitfield name="HORZ" low="0" high="8" type="uint"/>
 		<bitfield name="VERT" low="10" high="18" type="uint"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x8006" name="GRAS_CL_GUARDBAND_CLIP_ADJ" type="a6xx_gras_cl_guardband_clip_adj" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<!-- Something connected to depth-stencil attachment size -->
 	<reg32 offset="0x8007" name="GRAS_UNKNOWN_8007" variants="A7XX-" usage="rp_blit"/>
 
-	<reg32 offset="0x8008" name="GRAS_UNKNOWN_8008" variants="A7XX-" usage="cmd"/>
+	<!-- the scale/offset is per view, with up to 6 views -->
+	<bitset name="a6xx_gras_bin_foveat" inline="yes">
+		<bitfield name="BINSCALEEN" pos="6" type="boolean"/>
+		<enum name="a7xx_bin_scale">
+			<value value="0" name="NOSCALE"/>
+			<value value="1" name="SCALE2X"/>
+			<value value="2" name="SCALE4X"/>
+		</enum>
+		<bitfield name="XSCALE_0" low="8" high="9" type="a7xx_bin_scale"/>
+		<bitfield name="YSCALE_0" low="10" high="11" type="a7xx_bin_scale"/>
+		<bitfield name="XSCALE_1" low="12" high="13" type="a7xx_bin_scale"/>
+		<bitfield name="YSCALE_1" low="14" high="15" type="a7xx_bin_scale"/>
+		<bitfield name="XSCALE_2" low="16" high="17" type="a7xx_bin_scale"/>
+		<bitfield name="YSCALE_2" low="18" high="19" type="a7xx_bin_scale"/>
+		<bitfield name="XSCALE_3" low="20" high="21" type="a7xx_bin_scale"/>
+		<bitfield name="YSCALE_3" low="22" high="23" type="a7xx_bin_scale"/>
+		<bitfield name="XSCALE_4" low="24" high="25" type="a7xx_bin_scale"/>
+		<bitfield name="YSCALE_4" low="26" high="27" type="a7xx_bin_scale"/>
+		<bitfield name="XSCALE_5" low="28" high="29" type="a7xx_bin_scale"/>
+		<bitfield name="YSCALE_5" low="30" high="31" type="a7xx_bin_scale"/>
+	</bitset>
 
-	<reg32 offset="0x8009" name="GRAS_UNKNOWN_8009" variants="A7XX-" usage="cmd"/>
-	<reg32 offset="0x800a" name="GRAS_UNKNOWN_800A" variants="A7XX-" usage="cmd"/>
-	<reg32 offset="0x800b" name="GRAS_UNKNOWN_800B" variants="A7XX-" usage="cmd"/>
-	<reg32 offset="0x800c" name="GRAS_UNKNOWN_800C" variants="A7XX-" usage="cmd"/>
+	<reg32 offset="0x8008" name="GRAS_BIN_FOVEAT" type="a6xx_gras_bin_foveat" variants="A7XX" usage="cmd"/>
+
+	<reg32 offset="0x8009" name="GRAS_BIN_FOVEAT_OFFSET_0" variants="A7XX-" usage="cmd">
+		<bitfield name="XOFFSET_0" low="0" high="9" shr="2" type="uint"/>
+		<bitfield name="XOFFSET_1" low="10" high="19" shr="2" type="uint"/>
+		<bitfield name="XOFFSET_2" low="20" high="29" shr="2" type="uint"/>
+	</reg32>
+	<reg32 offset="0x800a" name="GRAS_BIN_FOVEAT_OFFSET_1" variants="A7XX-" usage="cmd">
+		<bitfield name="XOFFSET_3" low="0" high="9" shr="2" type="uint"/>
+		<bitfield name="XOFFSET_4" low="10" high="19" shr="2" type="uint"/>
+		<bitfield name="XOFFSET_5" low="20" high="29" shr="2" type="uint"/>
+	</reg32>
+	<reg32 offset="0x800b" name="GRAS_BIN_FOVEAT_OFFSET_2" variants="A7XX-" usage="cmd">
+		<bitfield name="YOFFSET_0" low="0" high="9" shr="2" type="uint"/>
+		<bitfield name="YOFFSET_1" low="10" high="19" shr="2" type="uint"/>
+		<bitfield name="YOFFSET_2" low="20" high="29" shr="2" type="uint"/>
+	</reg32>
+	<reg32 offset="0x800c" name="GRAS_BIN_FOVEAT_OFFSET_3" variants="A7XX-" usage="cmd">
+		<bitfield name="YOFFSET_3" low="0" high="9" shr="2" type="uint"/>
+		<bitfield name="YOFFSET_4" low="10" high="19" shr="2" type="uint"/>
+		<bitfield name="YOFFSET_5" low="20" high="29" shr="2" type="uint"/>
+	</reg32>
 
 	<!-- <reg32 offset="0x80f0" name="GRAS_UNKNOWN_80F0" type="a6xx_reg_xy"/> -->
 
 	<!-- 0x8006-0x800f invalid -->
-	<array offset="0x8010" name="GRAS_CL_VIEWPORT" stride="6" length="16" usage="rp_blit">
+	<array offset="0x8010" name="GRAS_CL_VIEWPORT" stride="6" length="16" variants="A6XX-A7XX" usage="rp_blit">
 		<reg32 offset="0" name="XOFFSET" type="float"/>
 		<reg32 offset="1" name="XSCALE" type="float"/>
 		<reg32 offset="2" name="YOFFSET" type="float"/>
@@ -875,12 +920,13 @@ by a particular renderpass/blit.
 		<reg32 offset="4" name="ZOFFSET" type="float"/>
 		<reg32 offset="5" name="ZSCALE" type="float"/>
 	</array>
-	<array offset="0x8070" name="GRAS_CL_VIEWPORT_ZCLAMP" stride="2" length="16" usage="rp_blit">
+
+	<array offset="0x8070" name="GRAS_CL_VIEWPORT_ZCLAMP" stride="2" length="16" variants="A6XX-A7XX" usage="rp_blit">
 		<reg32 offset="0" name="MIN" type="float"/>
 		<reg32 offset="1" name="MAX" type="float"/>
 	</array>
 
-	<reg32 offset="0x8090" name="GRAS_SU_CNTL" usage="rp_blit">
+	<bitset name="a6xx_gras_su_cntl" varset="chip">
 		<bitfield name="CULL_FRONT" pos="0" type="boolean"/>
 		<bitfield name="CULL_BACK" pos="1" type="boolean"/>
 		<bitfield name="FRONT_CW" pos="2" type="boolean"/>
@@ -890,39 +936,66 @@ by a particular renderpass/blit.
 		<bitfield name="LINE_MODE" pos="13" type="a5xx_line_mode"/>
 		<bitfield name="UNK15" low="15" high="16"/>
 		<!--
-                        On gen1 only MULTIVIEW_ENABLE exists. On gen3 we have
-                        the ability to add the view index to either the RT array
-                        index or the viewport index, and it seems that
-                        MULTIVIEW_ENABLE doesn't do anything, instead we need to
-                        set at least one of RENDERTARGETINDEXINCR or
-                        VIEWPORTINDEXINCR to enable multiview. The blob still
-                        sets MULTIVIEW_ENABLE regardless.
-                        TODO: what about gen2 (a640)?
+			On gen1 only MULTIVIEW_ENABLE exists. On gen3 we have
+			the ability to add the view index to either the RT array
+			index or the viewport index, and it seems that
+			MULTIVIEW_ENABLE doesn't do anything, instead we need to
+			set at least one of RENDERTARGETINDEXINCR or
+			VIEWPORTINDEXINCR to enable multiview. The blob still
+			sets MULTIVIEW_ENABLE regardless.
+			TODO: what about gen2 (a640)?
 		-->
 		<bitfield name="MULTIVIEW_ENABLE" pos="17" type="boolean"/>
-		<bitfield name="RENDERTARGETINDEXINCR" pos="18" type="boolean"/>
-		<bitfield name="VIEWPORTINDEXINCR" pos="19" type="boolean"/>
-		<bitfield name="UNK20" low="20" high="22"/>
-	</reg32>
-	<reg32 offset="0x8091" name="GRAS_SU_POINT_MINMAX" usage="rp_blit">
+		<bitfield name="RENDERTARGETINDEXINCR" pos="18" type="boolean" variants="A6XX-A7XX"/>
+		<bitfield name="VIEWPORTINDEXINCR" pos="19" type="boolean" variants="A6XX-A7XX"/>
+		<bitfield name="UNK20" low="20" high="22" variants="A6XX-A7XX"/>
+	</bitset>
+	<reg32 offset="0x8090" name="GRAS_SU_CNTL" type="a6xx_gras_su_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_su_point_minmax" inline="yes">
 		<bitfield name="MIN" low="0" high="15" type="ufixed" radix="4"/>
 		<bitfield name="MAX" low="16" high="31" type="ufixed" radix="4"/>
-	</reg32>
-	<reg32 offset="0x8092" name="GRAS_SU_POINT_SIZE" low="0" high="15" type="fixed" radix="4" usage="rp_blit"/>
+	</bitset>
+
+	<reg32 offset="0x8091" name="GRAS_SU_POINT_MINMAX" type="a6xx_gras_su_point_minmax" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8092" name="GRAS_SU_POINT_SIZE" low="0" high="15" type="fixed" radix="4" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_su_depth_cntl" inline="yes">
+		<bitfield name="Z_TEST_ENABLE" pos="0" type="boolean"/>
+	</bitset>
+
+	<reg32 offset="0x8114" name="GRAS_SU_DEPTH_CNTL" variants="A6XX-A7XX" type="a6xx_gras_su_depth_cntl" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_su_stencil_cntl" inline="yes">
+		<bitfield name="STENCIL_ENABLE" pos="0" type="boolean"/>
+	</bitset>
+
+	<reg32 offset="0x8115" name="GRAS_SU_STENCIL_CNTL" type="a6xx_gras_su_stencil_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_su_render_cntl" inline="yes">
+		<bitfield name="FS_DISABLE" pos="7" type="boolean"/>
+	</bitset>
+
+	<reg32 offset="0x8116" name="GRAS_SU_RENDER_CNTL" type="a6xx_gras_su_render_cntl" variants="A7XX" usage="rp_blit"/>
+
 	<!-- 0x8093 invalid -->
-	<reg32 offset="0x8094" name="GRAS_SU_DEPTH_PLANE_CNTL" usage="rp_blit">
+	<bitset name="a6xx_depth_plane_cntl" inline="yes">
 		<bitfield name="Z_MODE" low="0" high="1" type="a6xx_ztest_mode"/>
-	</reg32>
-	<reg32 offset="0x8095" name="GRAS_SU_POLY_OFFSET_SCALE" type="float" usage="rp_blit"/>
-	<reg32 offset="0x8096" name="GRAS_SU_POLY_OFFSET_OFFSET" type="float" usage="rp_blit"/>
-	<reg32 offset="0x8097" name="GRAS_SU_POLY_OFFSET_OFFSET_CLAMP" type="float" usage="rp_blit"/>
-	<!-- duplicates RB_DEPTH_BUFFER_INFO: -->
-	<reg32 offset="0x8098" name="GRAS_SU_DEPTH_BUFFER_INFO" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x8094" name="GRAS_SU_DEPTH_PLANE_CNTL" type="a6xx_depth_plane_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8095" name="GRAS_SU_POLY_OFFSET_SCALE" type="float" variants="A6XX-A7XX"  usage="rp_blit"/>
+	<reg32 offset="0x8096" name="GRAS_SU_POLY_OFFSET_OFFSET" type="float" variants="A6XX-A7XX"  usage="rp_blit"/>
+	<reg32 offset="0x8097" name="GRAS_SU_POLY_OFFSET_OFFSET_CLAMP" type="float" variants="A6XX-A7XX" usage="rp_blit"/>
+	<bitset name="a6xx_depth_buffer_info" inline="yes">
 		<bitfield name="DEPTH_FORMAT" low="0" high="2" type="a6xx_depth_format"/>
 		<bitfield name="UNK3" pos="3"/>
-	</reg32>
+	</bitset>
 
-	<reg32 offset="0x8099" name="GRAS_SU_CONSERVATIVE_RAS_CNTL" usage="cmd">
+	<!-- duplicates RB_DEPTH_BUFFER_INFO: -->
+	<reg32 offset="0x8098" name="GRAS_SU_DEPTH_BUFFER_INFO" type="a6xx_depth_buffer_info" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_su_conservative_ras_cntl" inline="yes">
 		<bitfield name="CONSERVATIVERASEN" pos="0" type="boolean"/>
 		<enum name="a6xx_shift_amount">
 			<value value="0" name="NO_SHIFT"/>
@@ -932,7 +1005,10 @@ by a particular renderpass/blit.
 		<bitfield name="SHIFTAMOUNT" low="1" high="2" type="a6xx_shift_amount"/>
 		<bitfield name="INNERCONSERVATIVERASEN" pos="3" type="boolean"/>
 		<bitfield name="UNK4" low="4" high="5"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x8099" name="GRAS_SU_CONSERVATIVE_RAS_CNTL" type="a6xx_gras_su_conservative_ras_cntl" variants="A6XX-A7XX" usage="cmd"/>
+
 	<reg32 offset="0x809a" name="GRAS_SU_PATH_RENDERING_CNTL">
 		<bitfield name="UNK0" pos="0" type="boolean"/>
 		<bitfield name="LINELENGTHEN" pos="1" type="boolean"/>
@@ -942,10 +1018,13 @@ by a particular renderpass/blit.
 		<bitfield name="WRITES_LAYER" pos="0" type="boolean"/>
 		<bitfield name="WRITES_VIEW" pos="1" type="boolean"/>
 	</bitset>
-	<reg32 offset="0x809b" name="GRAS_SU_VS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" usage="rp_blit"/>
-	<reg32 offset="0x809c" name="GRAS_SU_GS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" usage="rp_blit"/>
-	<reg32 offset="0x809d" name="GRAS_SU_DS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" usage="rp_blit"/>
-	<!-- 0x809e/0x809f invalid -->
+	<reg32 offset="0x809b" name="GRAS_SU_VS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x809c" name="GRAS_SU_GS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x809d" name="GRAS_SU_DS_SIV_CNTL" type="a6xx_gras_us_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_rast_cntl" inline="yes">
+		<bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/>
+	</bitset>
 
 	<enum name="a6xx_sequenced_thread_dist">
 		<value value="0x0" name="DIST_SCREEN_COORD"/>
@@ -993,7 +1072,7 @@ by a particular renderpass/blit.
 		<value value="0x3" name="RB_BT"/>
 	</enum>
 
-	<reg32 offset="0x80a0" name="GRAS_SC_CNTL" usage="rp_blit">
+	<bitset name="a6xx_gras_sc_cntl" inline="yes">
 		<bitfield name="CCUSINGLECACHELINESIZE" low="0" high="2"/>
 		<bitfield name="SINGLE_PRIM_MODE" low="3" high="4" type="a6xx_single_prim_mode"/>
 		<bitfield name="RASTER_MODE" pos="5" type="a6xx_raster_mode"/>
@@ -1003,7 +1082,9 @@ by a particular renderpass/blit.
 		<bitfield name="UNK9" pos="9" type="boolean"/>
 		<bitfield name="ROTATION" low="10" high="11" type="uint"/>
 		<bitfield name="EARLYVIZOUTEN" pos="12" type="boolean"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x80a0" name="GRAS_SC_CNTL" type="a6xx_gras_sc_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<enum name="a6xx_render_mode">
 		<value value="0x0" name="RENDERING_PASS"/>
@@ -1024,7 +1105,7 @@ by a particular renderpass/blit.
 		<value value="0x4" name="LRZ_FEEDBACK_LATE_Z"/>
 	</enum>
 
-	<reg32 offset="0x80a1" name="GRAS_SC_BIN_CNTL" usage="rp_blit">
+	<bitset name="a6xx_bin_cntl" inline="yes">
 		<bitfield name="BINW" low="0" high="5" shr="5" type="uint"/>
 		<bitfield name="BINH" low="8" high="14" shr="4" type="uint"/>
 		<bitfield name="RENDER_MODE" low="18" high="20" type="a6xx_render_mode"/>
@@ -1037,18 +1118,25 @@ by a particular renderpass/blit.
 			In sysmem mode GRAS_LRZ_CNTL.LRZ_WRITE is not considered.
 		</doc>
 		<bitfield name="LRZ_FEEDBACK_ZMODE_MASK" low="24" high="26" type="a6xx_lrz_feedback_mask"/>
-		<bitfield name="UNK27" pos="27"/>
-	</reg32>
+		<bitfield name="FORCE_LRZ_DIS" pos="27" type="boolean"/>
+	</bitset>
 
-	<reg32 offset="0x80a2" name="GRAS_SC_RAS_MSAA_CNTL" usage="rp_blit">
+	<reg32 offset="0x80a1" name="GRAS_SC_BIN_CNTL" type="a6xx_bin_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_sc_ras_msaa_cntl" inline="yes">
 		<bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/>
 		<bitfield name="UNK2" pos="2"/>
 		<bitfield name="UNK3" pos="3"/>
-	</reg32>
-	<reg32 offset="0x80a3" name="GRAS_SC_DEST_MSAA_CNTL" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x80a2" name="GRAS_SC_RAS_MSAA_CNTL" type="a6xx_gras_sc_ras_msaa_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_sc_dest_msaa_cntl" inline="yes">
 		<bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/>
 		<bitfield name="MSAA_DISABLE" pos="2" type="boolean"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x80a3" name="GRAS_SC_DEST_MSAA_CNTL" type="a6xx_gras_sc_dest_msaa_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<bitset name="a6xx_msaa_sample_pos_cntl" inline="yes">
 		<bitfield name="UNK0" pos="0"/>
@@ -1066,30 +1154,35 @@ by a particular renderpass/blit.
 		<bitfield name="SAMPLE_3_Y" low="28" high="31" radix="4" type="fixed"/>
 	</bitset>
 
-	<reg32 offset="0x80a4" name="GRAS_SC_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" usage="rp_blit"/>
-	<reg32 offset="0x80a5" name="GRAS_SC_PROGRAMMABLE_MSAA_POS_0" type="a6xx_programmable_msaa_pos" usage="rp_blit"/>
-	<reg32 offset="0x80a6" name="GRAS_SC_PROGRAMMABLE_MSAA_POS_1" type="a6xx_programmable_msaa_pos" usage="rp_blit"/>
+	<reg32 offset="0x80a4" name="GRAS_SC_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x80a5" name="GRAS_SC_PROGRAMMABLE_MSAA_POS_0" type="a6xx_programmable_msaa_pos" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x80a6" name="GRAS_SC_PROGRAMMABLE_MSAA_POS_1" type="a6xx_programmable_msaa_pos" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<reg32 offset="0x80a7" name="GRAS_UNKNOWN_80A7" variants="A7XX-" usage="cmd"/>
+	<reg32 offset="0x80a7" name="GRAS_ROTATION_CNTL" variants="A7XX" usage="cmd"/>
 
-	<!-- 0x80a7-0x80ae invalid -->
-	<reg32 offset="0x80af" name="GRAS_UNKNOWN_80AF" pos="0" usage="cmd"/>
+	<bitset name="a6xx_screen_scissor_cntl" inline="yes">
+		<bitfield name="SCISSOR_DISABLE" pos="0" type="boolean"/>
+	</bitset>
+
+	<reg32 offset="0x80af" name="GRAS_SC_SCREEN_SCISSOR_CNTL" type="a6xx_screen_scissor_cntl" variants="A6XX-A7XX" pos="0" usage="cmd"/>
 
 	<bitset name="a6xx_scissor_xy" inline="yes">
 		<bitfield name="X" low="0" high="15" type="uint"/>
 		<bitfield name="Y" low="16" high="31" type="uint"/>
 	</bitset>
-	<array offset="0x80b0" name="GRAS_SC_SCREEN_SCISSOR" stride="2" length="16" usage="rp_blit">
-		<reg32 offset="0" name="TL" type="a6xx_scissor_xy"/>
-		<reg32 offset="1" name="BR" type="a6xx_scissor_xy"/>
-	</array>
-	<array offset="0x80d0" name="GRAS_SC_VIEWPORT_SCISSOR" stride="2" length="16" usage="rp_blit">
+
+	<array offset="0x80b0" name="GRAS_SC_SCREEN_SCISSOR" stride="2" length="16" variants="A6XX-A7XX" usage="rp_blit">
 		<reg32 offset="0" name="TL" type="a6xx_scissor_xy"/>
 		<reg32 offset="1" name="BR" type="a6xx_scissor_xy"/>
 	</array>
 
-	<reg32 offset="0x80f0" name="GRAS_SC_WINDOW_SCISSOR_TL" type="a6xx_reg_xy" usage="rp_blit"/>
-	<reg32 offset="0x80f1" name="GRAS_SC_WINDOW_SCISSOR_BR" type="a6xx_reg_xy" usage="rp_blit"/>
+	<array offset="0x80d0" name="GRAS_SC_VIEWPORT_SCISSOR" stride="2" length="16" variants="A6XX-A7XX" usage="rp_blit">
+		<reg32 offset="0" name="TL" type="a6xx_scissor_xy"/>
+		<reg32 offset="1" name="BR" type="a6xx_scissor_xy"/>
+	</array>
+
+	<reg32 offset="0x80f0" name="GRAS_SC_WINDOW_SCISSOR_TL" type="a6xx_reg_xy" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x80f1" name="GRAS_SC_WINDOW_SCISSOR_BR" type="a6xx_reg_xy" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<enum name="a6xx_fsr_combiner">
 		<value value="0" name="FSR_COMBINER_OP_KEEP"/>
@@ -1099,7 +1192,7 @@ by a particular renderpass/blit.
 		<value value="4" name="FSR_COMBINER_OP_MUL"/>
 	</enum>
 
-	<reg32 offset="0x80f4" name="GRAS_VRS_CONFIG" variants="A7XX-" usage="rp_blit">
+	<bitset name="a6xx_gras_vrs_config">
 		<bitfield name="PIPELINE_FSR_ENABLE" pos="0" type="boolean"/>
 		<bitfield name="FRAG_SIZE_X" low="1" high="2" type="uint"/>
 		<bitfield name="FRAG_SIZE_Y" low="3" high="4" type="uint"/>
@@ -1107,20 +1200,32 @@ by a particular renderpass/blit.
 		<bitfield name="COMBINER_OP_2" low="8" high="10" type="a6xx_fsr_combiner"/>
 		<bitfield name="ATTACHMENT_FSR_ENABLE" pos="13" type="boolean"/>
 		<bitfield name="PRIMITIVE_FSR_ENABLE" pos="20" type="boolean"/>
-	</reg32>
-	<reg32 offset="0x80f5" name="GRAS_QUALITY_BUFFER_INFO" variants="A7XX-" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x80f4" name="GRAS_VRS_CONFIG" type="a6xx_gras_vrs_config" variants="A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_quality_buffer_info" inline="yes">
 		<bitfield name="LAYERED" pos="0" type="boolean"/>
 		<bitfield name="TILE_MODE" low="1" high="2" type="a6xx_tile_mode"/>
-	</reg32>
-	<reg32 offset="0x80f6" name="GRAS_QUALITY_BUFFER_DIMENSION" variants="A7XX-" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x80f5" name="GRAS_QUALITY_BUFFER_INFO" type="a6xx_gras_quality_buffer_info" variants="A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_quality_buffer_dimension" inline="yes">
 		<bitfield name="WIDTH" low="0" high="15" type="uint"/>
 		<bitfield name="HEIGHT" low="16" high="31" type="uint"/>
-	</reg32>
-	<reg64 offset="0x80f8" name="GRAS_QUALITY_BUFFER_BASE" variants="A7XX-" type="waddress" usage="rp_blit"/>
-	<reg32 offset="0x80fa" name="GRAS_QUALITY_BUFFER_PITCH" variants="A7XX-" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x80f6" name="GRAS_QUALITY_BUFFER_DIMENSION" type="a6xx_gras_quality_buffer_dimension" variants="A7XX" usage="rp_blit"/>
+
+	<reg64 offset="0x80f8" name="GRAS_QUALITY_BUFFER_BASE" variants="A7XX" type="waddress" usage="rp_blit"/>
+
+	<bitset name="a6xx_gras_quality_buffer_pitch" inline="yes">
 		<bitfield name="PITCH" shr="6" low="0" high="7" type="uint"/>
 		<bitfield name="ARRAY_PITCH" shr="6" low="10" high="28" type="uint"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x80fa" name="GRAS_QUALITY_BUFFER_PITCH" type="a6xx_gras_quality_buffer_pitch" variants="A7XX" usage="rp_blit"/>
 
 	<enum name="a6xx_lrz_dir_status">
 		<value value="0x1" name="LRZ_DIR_LE"/>
@@ -1128,7 +1233,7 @@ by a particular renderpass/blit.
 		<value value="0x3" name="LRZ_DIR_INVALID"/>
 	</enum>
 
-	<reg32 offset="0x8100" name="GRAS_LRZ_CNTL" usage="rp_blit">
+	<bitset name="a6xx_gras_lrz_cntl" inline="yes">
 		<bitfield name="ENABLE" pos="0" type="boolean"/>
 		<doc>LRZ write also disabled for blend/etc.</doc>
 		<bitfield name="LRZ_WRITE" pos="1" type="boolean"/>
@@ -1155,26 +1260,36 @@ by a particular renderpass/blit.
 		</doc>
 		<bitfield name="DISABLE_ON_WRONG_DIR" pos="9" type="boolean" variants="A6XX"/>
 		<bitfield name="Z_FUNC" low="11" high="13" type="adreno_compare_func" variants="A7XX-"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x8100" name="GRAS_LRZ_CNTL" type="a6xx_gras_lrz_cntl" usage="rp_blit" variants="A6XX-A7XX"/>
 
 	<enum name="a6xx_fragcoord_sample_mode">
 		<value value="0" name="FRAGCOORD_CENTER"/>
 		<value value="3" name="FRAGCOORD_SAMPLE"/>
 	</enum>
 
-	<reg32 offset="0x8101" name="GRAS_LRZ_PS_INPUT_CNTL" low="0" high="2" usage="rp_blit">
+	<bitset name="a6xx_gras_lrz_ps_input_cntl" inline="yes">
 		<bitfield name="SAMPLEID" pos="0" type="boolean"/>
 		<bitfield name="FRAGCOORDSAMPLEMODE" low="1" high="2" type="a6xx_fragcoord_sample_mode"/>
-	</reg32>
+	</bitset>
 
-	<reg32 offset="0x8102" name="GRAS_LRZ_MRT_BUFFER_INFO_0" usage="rp_blit">
+	<reg32 offset="0x8101" name="GRAS_LRZ_PS_INPUT_CNTL" type="a6xx_gras_lrz_ps_input_cntl" usage="rp_blit" variants="A6XX-A7XX"/>
+
+	<bitset name="a6xx_gras_lrz_mrt_buffer_info_0" inline="yes">
 		<bitfield name="COLOR_FORMAT" low="0" high="7" type="a6xx_format"/>
-	</reg32>
-	<reg64 offset="0x8103" name="GRAS_LRZ_BUFFER_BASE" align="256" type="waddress" usage="rp_blit"/>
-	<reg32 offset="0x8105" name="GRAS_LRZ_BUFFER_PITCH" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x8102" name="GRAS_LRZ_MRT_BUFFER_INFO_0" type="a6xx_gras_lrz_mrt_buffer_info_0" usage="rp_blit" variants="A6XX-A7XX"/>
+
+	<reg64 offset="0x8103" name="GRAS_LRZ_BUFFER_BASE" align="256" type="waddress" usage="rp_blit" variants="A6XX-A7XX"/>
+
+	<bitset name="a6xx_gras_lrz_buffer_pitch" inline="yes">
 		<bitfield name="PITCH" low="0" high="7" shr="5" type="uint"/>
 		<bitfield name="ARRAY_PITCH" low="10" high="28" shr="8" type="uint"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x8105" name="GRAS_LRZ_BUFFER_PITCH" type="a6xx_gras_lrz_buffer_pitch" usage="rp_blit" variants="A6XX-A7XX"/>
 
 	<!--
 	The LRZ "fast clear" buffer is initialized to zero's by blob, and
@@ -1207,7 +1322,6 @@ by a particular renderpass/blit.
 	not.
 	 -->
 	<reg64 offset="0x8106" name="GRAS_LRZ_FAST_CLEAR_BUFFER_BASE" align="64" type="waddress" usage="rp_blit"/>
-	<!-- 0x8108 invalid -->
 	<reg32 offset="0x8109" name="GRAS_LRZ_PS_SAMPLEFREQ_CNTL" usage="rp_blit">
 		<bitfield name="PER_SAMP_MODE" pos="0" type="boolean"/>
 	</reg32>
@@ -1232,19 +1346,20 @@ by a particular renderpass/blit.
 
 	<!-- 0x810c-0x810f invalid -->
 
-	<reg32 offset="0x8110" name="GRAS_UNKNOWN_8110" low="0" high="1" usage="cmd"/>
+	<reg32 offset="0x8110" name="GRAS_MODE_CNTL" low="0" high="1" variants="A6XX-A7XX" usage="cmd"/>
 
 	<!-- A bit tentative but it's a color and it is followed by LRZ_CLEAR -->
-	<reg32 offset="0x8111" name="GRAS_LRZ_DEPTH_CLEAR" type="float" variants="A7XX-"/>
+	<reg32 offset="0x8111" name="GRAS_LRZ_DEPTH_CLEAR" type="float" variants="A7XX"/>
 
-	<reg32 offset="0x8113" name="GRAS_LRZ_DEPTH_BUFFER_INFO" variants="A7XX-" usage="rp_blit">
+	<bitset name="a6xx_gras_lrz_depth_buffer_info" inline="yes">
 		<bitfield name="DEPTH_FORMAT" low="0" high="2" type="a6xx_depth_format"/>
 		<bitfield name="UNK3" pos="3"/>
-	</reg32>
+	</bitset>
 
-	<!-- Always written together and always equal 09510840 00000a62 -->
-	<reg32 offset="0x8120" name="GRAS_UNKNOWN_8120" variants="A7XX-" usage="cmd"/>
-	<reg32 offset="0x8121" name="GRAS_UNKNOWN_8121" variants="A7XX-" usage="cmd"/>
+	<reg32 offset="0x8113" name="GRAS_LRZ_DEPTH_BUFFER_INFO" type="a6xx_gras_lrz_depth_buffer_info" variants="A7XX" usage="rp_blit"/>
+
+	<doc>LUT used to convert quality buffer values to HW shading rate values. An array of 4-bit values.</doc>
+	<array offset="0x8120" name="GRAS_LRZ_QUALITY_LOOKUP_TABLE" variants="A7XX-" stride="1" length="2"/>
 
 	<!-- 0x8112-0x83ff invalid -->
 
@@ -1269,28 +1384,29 @@ by a particular renderpass/blit.
 		<bitfield name="D24S8" pos="19" type="boolean"/>
 		<!-- some sort of channel mask, disabled channels are set to zero ? -->
 		<bitfield name="MASK" low="20" high="23"/>
-		<bitfield name="IFMT" low="24" high="28" type="a6xx_2d_ifmt"/>
+		<bitfield name="IFMT" low="24" high="26" type="a6xx_2d_ifmt"/>
+		<bitfield name="UNK27" pos="27" type="boolean"/>
+		<bitfield name="UNK28" pos="28" type="boolean"/>
 		<bitfield name="RASTER_MODE" pos="29" type="a6xx_raster_mode"/>
-		<bitfield name="UNK30" pos="30" type="boolean" variants="A7XX-"/>
+		<bitfield name="COPY" pos="30" type="boolean" variants="A7XX-"/>
 	</bitset>
 
-	<reg32 offset="0x8400" name="GRAS_A2D_BLT_CNTL" type="a6xx_a2d_bit_cntl" usage="rp_blit"/>
+	<reg32 offset="0x8400" name="GRAS_A2D_BLT_CNTL" type="a6xx_a2d_bit_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 	<!-- note: the low 8 bits for src coords are valid, probably fixed point
 	     it would be a bit weird though, since we subtract 1 from BR coords
 	     apparently signed, gallium driver uses negative coords and it works?
 	 -->
-	<reg32 offset="0x8401" name="GRAS_A2D_SRC_XMIN" low="8" high="24" type="int" usage="rp_blit"/>
-	<reg32 offset="0x8402" name="GRAS_A2D_SRC_XMAX" low="8" high="24" type="int" usage="rp_blit"/>
-	<reg32 offset="0x8403" name="GRAS_A2D_SRC_YMIN" low="8" high="24" type="int" usage="rp_blit"/>
-	<reg32 offset="0x8404" name="GRAS_A2D_SRC_YMAX" low="8" high="24" type="int" usage="rp_blit"/>
-	<reg32 offset="0x8405" name="GRAS_A2D_DEST_TL" type="a6xx_reg_xy" usage="rp_blit"/>
-	<reg32 offset="0x8406" name="GRAS_A2D_DEST_BR" type="a6xx_reg_xy" usage="rp_blit"/>
+	<reg32 offset="0x8401" name="GRAS_A2D_SRC_XMIN" low="8" high="24" type="int" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8402" name="GRAS_A2D_SRC_XMAX" low="8" high="24" type="int" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8403" name="GRAS_A2D_SRC_YMIN" low="8" high="24" type="int" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8404" name="GRAS_A2D_SRC_YMAX" low="8" high="24" type="int" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8405" name="GRAS_A2D_DEST_TL" type="a6xx_reg_xy" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x8406" name="GRAS_A2D_DEST_BR" type="a6xx_reg_xy" variants="A6XX-A7XX" usage="rp_blit"/>
 	<reg32 offset="0x8407" name="GRAS_2D_UNKNOWN_8407" low="0" high="31"/>
 	<reg32 offset="0x8408" name="GRAS_2D_UNKNOWN_8408" low="0" high="31"/>
 	<reg32 offset="0x8409" name="GRAS_2D_UNKNOWN_8409" low="0" high="31"/>
-	<reg32 offset="0x840a" name="GRAS_A2D_SCISSOR_TL" type="a6xx_reg_xy" usage="rp_blit"/>
-	<reg32 offset="0x840b" name="GRAS_A2D_SCISSOR_BR" type="a6xx_reg_xy" usage="rp_blit"/>
-	<!-- 0x840c-0x85ff invalid -->
+	<reg32 offset="0x840a" name="GRAS_A2D_SCISSOR_TL" type="a6xx_reg_xy" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x840b" name="GRAS_A2D_SCISSOR_BR" type="a6xx_reg_xy" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<!-- always 0x880 ? (and 0 in a640/a650 traces?) -->
 	<reg32 offset="0x8600" name="GRAS_DBG_ECO_CNTL" usage="cmd">
@@ -1308,22 +1424,7 @@ by a particular renderpass/blit.
 	-->
 
 	<!-- same as GRAS_BIN_CONTROL, but without bit 27: -->
-	<reg32 offset="0x8800" name="RB_CNTL" variants="A6XX" usage="rp_blit">
-		<bitfield name="BINW" low="0" high="5" shr="5" type="uint"/>
-		<bitfield name="BINH" low="8" high="14" shr="4" type="uint"/>
-		<bitfield name="RENDER_MODE" low="18" high="20" type="a6xx_render_mode"/>
-		<bitfield name="FORCE_LRZ_WRITE_DIS" pos="21" type="boolean"/>
-		<bitfield name="BUFFERS_LOCATION" low="22" high="23" type="a6xx_buffers_location"/>
-		<bitfield name="LRZ_FEEDBACK_ZMODE_MASK" low="24" high="26" type="a6xx_lrz_feedback_mask"/>
-	</reg32>
-
-	<reg32 offset="0x8800" name="RB_CNTL" variants="A7XX-" usage="rp_blit">
-		<bitfield name="BINW" low="0" high="5" shr="5" type="uint"/>
-		<bitfield name="BINH" low="8" high="14" shr="4" type="uint"/>
-		<bitfield name="RENDER_MODE" low="18" high="20" type="a6xx_render_mode"/>
-		<bitfield name="FORCE_LRZ_WRITE_DIS" pos="21" type="boolean"/>
-		<bitfield name="LRZ_FEEDBACK_ZMODE_MASK" low="24" high="26" type="a6xx_lrz_feedback_mask"/>
-	</reg32>
+	<reg32 offset="0x8800" name="RB_CNTL" variants="A6XX-A7XX" type="a6xx_bin_cntl" usage="rp_blit"/>
 
 	<reg32 offset="0x8801" name="RB_RENDER_CNTL" variants="A6XX" usage="rp_blit">
 		<bitfield name="CCUSINGLECACHELINESIZE" low="3" high="5"/>
@@ -1347,9 +1448,6 @@ by a particular renderpass/blit.
 		<bitfield name="CONSERVATIVERASEN" pos="11" type="boolean"/>
 		<bitfield name="INNERCONSERVATIVERASEN" pos="12" type="boolean"/>
 	</reg32>
-	<reg32 offset="0x8116" name="GRAS_SU_RENDER_CNTL" variants="A7XX-" usage="rp_blit">
-		<bitfield name="FS_DISABLE" pos="7" type="boolean"/>
-	</reg32>
 
 	<reg32 offset="0x8802" name="RB_RAS_MSAA_CNTL" usage="rp_blit">
 		<bitfield name="SAMPLES" low="0" high="1" type="a3xx_msaa_samples"/>
@@ -1516,9 +1614,7 @@ by a particular renderpass/blit.
 		<bitfield name="SAMPLE_MASK" low="16" high="31"/>
 	</reg32>
 	<!-- 0x8866-0x886f invalid -->
-	<reg32 offset="0x8870" name="RB_DEPTH_PLANE_CNTL" usage="rp_blit">
-		<bitfield name="Z_MODE" low="0" high="1" type="a6xx_ztest_mode"/>
-	</reg32>
+	<reg32 offset="0x8870" name="RB_DEPTH_PLANE_CNTL" type="a6xx_depth_plane_cntl" usage="rp_blit"/>
 
 	<reg32 offset="0x8871" name="RB_DEPTH_CNTL" usage="rp_blit">
 		<bitfield name="Z_TEST_ENABLE" pos="0" type="boolean"/>
@@ -1532,14 +1628,9 @@ by a particular renderpass/blit.
 		<bitfield name="Z_READ_ENABLE" pos="6" type="boolean"/>
 		<bitfield name="Z_BOUNDS_ENABLE" pos="7" type="boolean"/>
 	</reg32>
-	<reg32 offset="0x8114" name="GRAS_SU_DEPTH_CNTL" usage="rp_blit">
-		<bitfield name="Z_TEST_ENABLE" pos="0" type="boolean"/>
-	</reg32>
+
 	<!-- duplicates GRAS_SU_DEPTH_BUFFER_INFO: -->
-	<reg32 offset="0x8872" name="RB_DEPTH_BUFFER_INFO" variants="A6XX" usage="rp_blit">
-		<bitfield name="DEPTH_FORMAT" low="0" high="2" type="a6xx_depth_format"/>
-		<bitfield name="UNK3" low="3" high="4"/>
-	</reg32>
+	<reg32 offset="0x8872" name="RB_DEPTH_BUFFER_INFO" variants="A6XX" type="a6xx_depth_buffer_info" usage="rp_blit"/>
 	<!-- first 4 bits duplicates GRAS_SU_DEPTH_BUFFER_INFO -->
 	<reg32 offset="0x8872" name="RB_DEPTH_BUFFER_INFO" variants="A7XX-" usage="rp_blit">
 		<bitfield name="DEPTH_FORMAT" low="0" high="2" type="a6xx_depth_format"/>
@@ -1575,9 +1666,7 @@ by a particular renderpass/blit.
 		<bitfield name="ZPASS_BF" low="26" high="28" type="adreno_stencil_op"/>
 		<bitfield name="ZFAIL_BF" low="29" high="31" type="adreno_stencil_op"/>
 	</reg32>
-	<reg32 offset="0x8115" name="GRAS_SU_STENCIL_CNTL" usage="rp_blit">
-		<bitfield name="STENCIL_ENABLE" pos="0" type="boolean"/>
-	</reg32>
+
 	<reg32 offset="0x8881" name="RB_STENCIL_BUFFER_INFO" variants="A6XX" usage="rp_blit">
 		<bitfield name="SEPARATE_STENCIL" pos="0" type="boolean"/>
 		<bitfield name="UNK1" pos="1" type="boolean"/>
@@ -1616,8 +1705,9 @@ by a particular renderpass/blit.
 	<reg32 offset="0x8899" name="RB_UNKNOWN_8899" variants="A7XX-" usage="cmd"/>
 	<!-- 0x8899-0x88bf invalid -->
 	<!-- clamps depth value for depth test/write -->
-	<reg32 offset="0x88c0" name="RB_VIEWPORT_ZCLAMP_MIN" type="float" usage="rp_blit"/>
-	<reg32 offset="0x88c1" name="RB_VIEWPORT_ZCLAMP_MAX" type="float" usage="rp_blit"/>
+	<reg32 offset="0x88c0" name="RB_VIEWPORT_ZCLAMP_MIN" type="float" usage="rp_blit" variants="A6XX-A7XX"/>
+	<reg32 offset="0x88c1" name="RB_VIEWPORT_ZCLAMP_MAX" type="float" usage="rp_blit" variants="A6XX-A7XX"/>
+
 	<!-- 0x88c2-0x88cf invalid-->
 	<reg32 offset="0x88d0" name="RB_RESOLVE_CNTL_0" usage="rp_blit">
 		<bitfield name="UNK0" low="0" high="12"/>
@@ -1626,7 +1716,7 @@ by a particular renderpass/blit.
 	<reg32 offset="0x88d1" name="RB_RESOLVE_CNTL_1" type="a6xx_reg_xy" usage="rp_blit"/>
 	<reg32 offset="0x88d2" name="RB_RESOLVE_CNTL_2" type="a6xx_reg_xy" usage="rp_blit"/>
 	<!-- weird to duplicate other regs from same block?? -->
-	<reg32 offset="0x88d3" name="RB_RESOLVE_CNTL_3" usage="rp_blit">
+	<reg32 offset="0x88d3" name="RB_RESOLVE_CNTL_3" variants="A6XX-A7XX" usage="rp_blit">
 		<bitfield name="BINW" low="0" high="5" shr="5" type="uint"/>
 		<bitfield name="BINH" low="8" high="14" shr="4" type="uint"/>
 	</reg32>
@@ -1650,10 +1740,13 @@ by a particular renderpass/blit.
 	<!-- array-pitch is size of layer -->
 	<reg32 offset="0x88db" name="RB_RESOLVE_SYSTEM_BUFFER_ARRAY_PITCH" low="0" high="28" shr="6" type="uint" usage="rp_blit"/>
 	<reg64 offset="0x88dc" name="RB_RESOLVE_SYSTEM_FLAG_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/>
-	<reg32 offset="0x88de" name="RB_RESOLVE_SYSTEM_FLAG_BUFFER_PITCH" usage="rp_blit">
+
+	<bitset name="a6xx_flag_buffer_pitch" inline="yes">
 		<bitfield name="PITCH" low="0" high="10" shr="6" type="uint"/>
-		<bitfield name="ARRAY_PITCH" low="11" high="27" shr="7" type="uint"/>
-	</reg32>
+		<bitfield name="ARRAY_PITCH" low="11" high="28" shr="7" type="uint"/>
+	</bitset>
+
+	<reg32 offset="0x88de" name="RB_RESOLVE_SYSTEM_FLAG_BUFFER_PITCH" type="a6xx_flag_buffer_pitch" usage="rp_blit"/>
 
 	<reg32 offset="0x88df" name="RB_RESOLVE_CLEAR_COLOR_DW0" usage="rp_blit"/>
 	<reg32 offset="0x88e0" name="RB_RESOLVE_CLEAR_COLOR_DW1" usage="rp_blit"/>
@@ -1726,10 +1819,7 @@ by a particular renderpass/blit.
 	<reg32 offset="0x88f0" name="RB_UNKNOWN_88F0" low="0" high="11" usage="cmd"/>
 	<!-- could be for separate stencil? (or may not be a flag buffer at all) -->
 	<reg64 offset="0x88f1" name="RB_UNK_FLAG_BUFFER_BASE" type="waddress" align="64"/>
-	<reg32 offset="0x88f3" name="RB_UNK_FLAG_BUFFER_PITCH">
-		<bitfield name="PITCH" low="0" high="10" shr="6" type="uint"/>
-		<bitfield name="ARRAY_PITCH" low="11" high="23" shr="7" type="uint"/>
-	</reg32>
+	<reg32 offset="0x88f3" name="RB_UNK_FLAG_BUFFER_PITCH" type="a6xx_flag_buffer_pitch"/>
 
 	<reg32 offset="0x88f4" name="RB_VRS_CONFIG" usage="rp_blit">
 		<bitfield name="UNK2" pos="2" type="boolean"/>
@@ -1737,8 +1827,9 @@ by a particular renderpass/blit.
 		<bitfield name="ATTACHMENT_FSR_ENABLE" pos="5" type="boolean"/>
 		<bitfield name="PRIMITIVE_FSR_ENABLE" pos="18" type="boolean"/>
 	</reg32>
-	<!-- Connected to VK_EXT_fragment_density_map? -->
-	<reg32 offset="0x88f5" name="RB_UNKNOWN_88F5" variants="A7XX-"/>
+	<reg32 offset="0x88f5" name="RB_BIN_FOVEAT" variants="A7XX-" usage="cmd">
+		<bitfield name="BINSCALEEN" pos="6" type="boolean"/>
+	</reg32>
 	<!-- 0x88f6-0x88ff invalid -->
 	<reg64 offset="0x8900" name="RB_DEPTH_FLAG_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/>
 	<reg32 offset="0x8902" name="RB_DEPTH_FLAG_BUFFER_PITCH" usage="rp_blit">
@@ -1747,12 +1838,10 @@ by a particular renderpass/blit.
 		<bitfield name="UNK8" low="8" high="10"/>
 		<bitfield name="ARRAY_PITCH" low="11" high="27" shr="7" type="uint"/>
 	</reg32>
+
 	<array offset="0x8903" name="RB_COLOR_FLAG_BUFFER" stride="3" length="8" usage="rp_blit">
 		<reg64 offset="0" name="ADDR" type="waddress" align="64"/>
-		<reg32 offset="2" name="PITCH">
-			<bitfield name="PITCH" low="0" high="10" shr="6" type="uint"/>
-			<bitfield name="ARRAY_PITCH" low="11" high="28" shr="7" type="uint"/>
-		</reg32>
+		<reg32 offset="2" name="PITCH" type="a6xx_flag_buffer_pitch"/>
 	</array>
 	<!-- 0x891b-0x8926 invalid -->
 	<doc>
@@ -1815,7 +1904,7 @@ by a particular renderpass/blit.
 	<reg64 offset="0x8c1e" name="RB_A2D_DEST_BUFFER_BASE_2" type="waddress" align="64" usage="rp_blit"/>
 
 	<reg64 offset="0x8c20" name="RB_A2D_DEST_FLAG_BUFFER_BASE" type="waddress" align="64" usage="rp_blit"/>
-	<reg32 offset="0x8c22" name="RB_A2D_DEST_FLAG_BUFFER_PITCH" low="0" high="7" shr="6" type="uint" usage="rp_blit"/>
+	<reg32 offset="0x8c22" name="RB_A2D_DEST_FLAG_BUFFER_PITCH" type="a6xx_flag_buffer_pitch" usage="rp_blit"/>
 	<!-- this is a guess but seems likely (for NV12 with UBWC): -->
 	<reg64 offset="0x8c23" name="RB_A2D_DEST_FLAG_BUFFER_BASE_1" type="waddress" align="64" usage="rp_blit"/>
 	<reg32 offset="0x8c25" name="RB_A2D_DEST_FLAG_BUFFER_PITCH_1" low="0" high="7" shr="6" type="uint" usage="rp_blit"/>
@@ -1921,13 +2010,13 @@ by a particular renderpass/blit.
 		<bitfield name="CLIP_DIST_03_LOC" low="8" high="15" type="uint"/>
 		<bitfield name="CLIP_DIST_47_LOC" low="16" high="23" type="uint"/>
 	</bitset>
-	<reg32 offset="0x9101" name="VPC_VS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9102" name="VPC_GS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9103" name="VPC_DS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/>
+	<reg32 offset="0x9101" name="VPC_VS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9102" name="VPC_GS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9103" name="VPC_DS_CLIP_CULL_CNTL" type="a6xx_vpc_xs_clip_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<reg32 offset="0x9311" name="VPC_VS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9312" name="VPC_GS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9313" name="VPC_DS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" usage="rp_blit"/>
+	<reg32 offset="0x9311" name="VPC_VS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9312" name="VPC_GS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9313" name="VPC_DS_CLIP_CULL_CNTL_V2" type="a6xx_vpc_xs_clip_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<bitset name="a6xx_vpc_xs_siv_cntl" inline="yes">
 		<bitfield name="LAYERLOC" low="0" high="7" type="uint"/>
@@ -1935,23 +2024,33 @@ by a particular renderpass/blit.
 		<bitfield name="SHADINGRATELOC" low="16" high="23" type="uint" variants="A7XX-"/>
 	</bitset>
 
-	<reg32 offset="0x9104" name="VPC_VS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9105" name="VPC_GS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9106" name="VPC_DS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/>
+	<reg32 offset="0x9104" name="VPC_VS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9105" name="VPC_GS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9106" name="VPC_DS_SIV_CNTL" type="a6xx_vpc_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<reg32 offset="0x9314" name="VPC_VS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9315" name="VPC_GS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9316" name="VPC_DS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" usage="rp_blit"/>
+
+	<reg32 offset="0x9314" name="VPC_VS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9315" name="VPC_GS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9316" name="VPC_DS_SIV_CNTL_V2" type="a6xx_vpc_xs_siv_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_vpc_rast_stream_cntl" inline="yes">
+		<!-- which stream to send to GRAS -->
+		<bitfield name="STREAM" low="0" high="1" type="uint"/>
+		<!-- discard primitives before rasterization -->
+		<bitfield name="DISCARD" pos="2" type="boolean"/>
+	</bitset>
+
+	<reg32 offset="0x9980" name="VPC_RAST_STREAM_CNTL" type="a6xx_vpc_rast_stream_cntl" variants="A6XX" usage="rp_blit"/>
+	<reg32 offset="0x9107" name="VPC_RAST_STREAM_CNTL" type="a6xx_vpc_rast_stream_cntl" variants="A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9317" name="VPC_RAST_STREAM_CNTL_V2" type="a6xx_vpc_rast_stream_cntl" variants="A7XX" usage="rp_blit"/>
 
 	<reg32 offset="0x9107" name="VPC_UNKNOWN_9107" variants="A6XX" usage="rp_blit">
 		<!-- this mirrors VPC_RAST_STREAM_CNTL::DISCARD, although it seems it's unused -->
 		<bitfield name="RASTER_DISCARD" pos="0" type="boolean"/>
 		<bitfield name="UNK2" pos="2" type="boolean"/>
 	</reg32>
-	<reg32 offset="0x9108" name="VPC_RAST_CNTL" usage="rp_blit">
-		<bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/>
-	</reg32>
 
+	<reg32 offset="0x9108" name="VPC_RAST_CNTL" type="a6xx_rast_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 	<bitset name="a6xx_pc_cntl" inline="yes">
 		<bitfield name="PRIMITIVE_RESTART" pos="0" type="boolean"/>
 		<bitfield name="PROVOKING_VTX_LAST" pos="1" type="boolean"/>
@@ -1991,10 +2090,10 @@ by a particular renderpass/blit.
 		<bitfield name="VIEWS" low="2" high="6" type="uint"/>
 	</bitset>
 
-	<reg32 offset="0x9109" name="VPC_PC_CNTL" type="a6xx_pc_cntl" variants="A7XX-" usage="rp_blit"/>
-	<reg32 offset="0x910a" name="VPC_GS_PARAM_0" type="a6xx_gs_param_0" variants="A7XX-" usage="rp_blit"/>
-	<reg32 offset="0x910b" name="VPC_STEREO_RENDERING_VIEWMASK" type="hex" low="0" high="15" variants="A7XX-" usage="rp_blit"/>
-	<reg32 offset="0x910c" name="VPC_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" variants="A7XX-" usage="rp_blit"/>
+	<reg32 offset="0x9109" name="VPC_PC_CNTL" type="a6xx_pc_cntl" variants="A7XX" usage="rp_blit"/>
+	<reg32 offset="0x910a" name="VPC_GS_PARAM_0" type="a6xx_gs_param_0" variants="A7XX" usage="rp_blit"/>
+	<reg32 offset="0x910b" name="VPC_STEREO_RENDERING_VIEWMASK" type="hex" low="0" high="15" variants="A7XX" usage="rp_blit"/>
+	<reg32 offset="0x910c" name="VPC_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" variants="A7XX" usage="rp_blit"/>
 
 	<enum name="a6xx_varying_interp_mode">
 		<value value="0" name="INTERP_SMOOTH"/>
@@ -2011,11 +2110,11 @@ by a particular renderpass/blit.
 	</enum>
 
 	<!-- 0x9109-0x91ff invalid -->
-	<array offset="0x9200" name="VPC_VARYING_INTERP_MODE" stride="1" length="8" usage="rp_blit">
+	<array offset="0x9200" name="VPC_VARYING_INTERP_MODE" stride="1" length="8" variants="A6XX-A7XX" usage="rp_blit">
 		<doc>Packed array of a6xx_varying_interp_mode</doc>
 		<reg32 offset="0x0" name="MODE"/>
 	</array>
-	<array offset="0x9208" name="VPC_VARYING_REPLACE_MODE_0" stride="1" length="8" usage="rp_blit">
+	<array offset="0x9208" name="VPC_VARYING_REPLACE_MODE" stride="1" length="8" variants="A6XX-A7XX" usage="rp_blit">
 		<doc>Packed array of a6xx_varying_ps_repl_mode</doc>
 		<reg32 offset="0x0" name="MODE"/>
 	</array>
@@ -2024,12 +2123,12 @@ by a particular renderpass/blit.
 	<reg32 offset="0x9210" name="VPC_UNKNOWN_9210" low="0" high="31" variants="A6XX" usage="cmd"/>
 	<reg32 offset="0x9211" name="VPC_UNKNOWN_9211" low="0" high="31" variants="A6XX" usage="cmd"/>
 
-	<array offset="0x9212" name="VPC_VARYING_LM_TRANSFER_CNTL_0" stride="1" length="4" usage="rp_blit">
+	<array offset="0x9212" name="VPC_VARYING_LM_TRANSFER_CNTL" stride="1" length="4" variants="A6XX-A7XX" usage="rp_blit">
 		<!-- one bit per varying component: -->
 		<reg32 offset="0" name="DISABLE"/>
 	</array>
 
-	<reg32 offset="0x9216" name="VPC_SO_MAPPING_WPTR" usage="rp_blit">
+	<bitset name="a6xx_vpc_so_mapping_wptr" inline="yes">
 		<!--
 			Choose which DWORD to write to. There is an array of
 			(4 * 64) DWORD's, dumped in the devcoredump at
@@ -2056,20 +2155,25 @@ by a particular renderpass/blit.
 		<bitfield name="ADDR" low="0" high="7" type="hex"/>
 		<!-- clear all A_EN and B_EN bits for all DWORD's -->
 		<bitfield name="RESET" pos="16" type="boolean"/>
-	</reg32>
-	<!-- special register, write multiple times to load SO program (not readable) -->
-	<reg32 offset="0x9217" name="VPC_SO_MAPPING_PORT" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x9216" name="VPC_SO_MAPPING_WPTR" type="a6xx_vpc_so_mapping_wptr" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_vpc_so_mapping_port" inline="yes">
 		<bitfield name="A_BUF" low="0" high="1" type="uint"/>
 		<bitfield name="A_OFF" low="2" high="10" shr="2" type="uint"/>
 		<bitfield name="A_EN" pos="11" type="boolean"/>
 		<bitfield name="B_BUF" low="12" high="13" type="uint"/>
 		<bitfield name="B_OFF" low="14" high="22" shr="2" type="uint"/>
 		<bitfield name="B_EN" pos="23" type="boolean"/>
-	</reg32>
+	</bitset>
 
-	<reg64 offset="0x9218" name="VPC_SO_QUERY_BASE" type="waddress" align="32" usage="cmd"/>
+	<!-- special register, write multiple times to load SO program (not readable) -->
+	<reg32 offset="0x9217" name="VPC_SO_MAPPING_PORT" type="a6xx_vpc_so_mapping_port" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<array offset="0x921a" name="VPC_SO" stride="7" length="4" usage="cmd">
+	<reg64 offset="0x9218" name="VPC_SO_QUERY_BASE" type="waddress" align="32" variants="A6XX-A7XX" usage="cmd"/>
+
+	<array offset="0x921a" name="VPC_SO" stride="7" length="4" variants="A6XX-A7XX" usage="cmd">
 		<reg64 offset="0" name="BUFFER_BASE" type="waddress" align="32"/>
 		<reg32 offset="2" name="BUFFER_SIZE" low="2" high="31" shr="2"/>
 		<reg32 offset="3" name="BUFFER_STRIDE" low="0" high="9" shr="2"/>
@@ -2077,12 +2181,13 @@ by a particular renderpass/blit.
 		<reg64 offset="5" name="FLUSH_BASE" type="waddress" align="32"/>
 	</array>
 
-	<reg32 offset="0x9236" name="VPC_REPLACE_MODE_CNTL" usage="cmd">
+	<bitset name="a6xx_vpc_replace_mode_cntl" inline="yes">
 		<bitfield name="INVERT" pos="0" type="boolean"/>
-	</reg32>
-	<!-- 0x9237-0x92ff invalid -->
-	<!-- always 0x0 ? -->
-	<reg32 offset="0x9300" name="VPC_UNKNOWN_9300" low="0" high="2" usage="cmd"/>
+	</bitset>
+
+	<reg32 offset="0x9236" name="VPC_REPLACE_MODE_CNTL" type="a6xx_vpc_replace_mode_cntl" variants="A6XX-A7XX" usage="cmd"/>
+
+	<reg32 offset="0x9300" name="VPC_ROTATION_CNTL" low="0" high="2" variants="A6XX-A7XX" usage="cmd"/>
 
 	<bitset name="a6xx_vpc_xs_cntl" inline="yes">
 		<doc>
@@ -2101,11 +2206,12 @@ by a particular renderpass/blit.
 			</doc>
 		</bitfield>
 	</bitset>
-	<reg32 offset="0x9301" name="VPC_VS_CNTL" type="a6xx_vpc_xs_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9302" name="VPC_GS_CNTL" type="a6xx_vpc_xs_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9303" name="VPC_DS_CNTL" type="a6xx_vpc_xs_cntl" usage="rp_blit"/>
 
-	<reg32 offset="0x9304" name="VPC_PS_CNTL" usage="rp_blit">
+	<reg32 offset="0x9301" name="VPC_VS_CNTL" type="a6xx_vpc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9302" name="VPC_GS_CNTL" type="a6xx_vpc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9303" name="VPC_DS_CNTL" type="a6xx_vpc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_vpc_ps_cntl" inline="yes">
 		<bitfield name="NUMNONPOSVAR" low="0" high="7" type="uint"/>
 		<!-- for fixed-function (i.e. no GS) gl_PrimitiveID in FS -->
 		<bitfield name="PRIMIDLOC" low="8" high="15" type="uint"/>
@@ -2122,9 +2228,11 @@ by a particular renderpass/blit.
 				ViewID through the VS.
 			</doc>
 		</bitfield>
-	</reg32>
+	</bitset>
 
-	<reg32 offset="0x9305" name="VPC_SO_CNTL" usage="rp_blit">
+	<reg32 offset="0x9304" name="VPC_PS_CNTL" type="a6xx_vpc_ps_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_vpc_so_cntl" inline="yes">
 		<!--
 		It's offset by 1, and 0 means "disabled"
 		-->
@@ -2133,22 +2241,28 @@ by a particular renderpass/blit.
 		<bitfield name="BUF2_STREAM" low="6" high="8" type="uint"/>
 		<bitfield name="BUF3_STREAM" low="9" high="11" type="uint"/>
 		<bitfield name="STREAM_ENABLE" low="15" high="18" type="hex"/>
-	</reg32>
-	<reg32 offset="0x9306" name="VPC_SO_OVERRIDE" usage="rp_blit">
+	</bitset>
+
+	<reg32 offset="0x9305" name="VPC_SO_CNTL" type="a6xx_vpc_so_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_so_override" inline="yes">
 		<bitfield name="DISABLE" pos="0" type="boolean"/>
-	</reg32>
-	<reg32 offset="0x9307" name="VPC_PS_RAST_CNTL" variants="A6XX-" usage="rp_blit"> <!-- A702 + A7xx -->
-		<bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/>
-	</reg32>
-	<reg32 offset="0x9308" name="VPC_ATTR_BUF_GMEM_SIZE" variants="A7XX-" usage="rp_blit">
-		<bitfield name="SIZE_GMEM" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="0x9309" name="VPC_ATTR_BUF_GMEM_BASE" variants="A7XX-" usage="rp_blit">
-		<bitfield name="BASE_GMEM" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="0x9b09" name="PC_ATTR_BUF_GMEM_SIZE" variants="A7XX-" usage="rp_blit">
-		<bitfield name="SIZE_GMEM" low="0" high="31"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x9306" name="VPC_SO_OVERRIDE" type="a6xx_so_override" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<reg32 offset="0x9807" name="PC_DGEN_SO_OVERRIDE" type="a6xx_so_override" variants="A7XX" usage="rp_blit"/>
+
+	<reg32 offset="0x9307" name="VPC_PS_RAST_CNTL" type="a6xx_rast_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<reg32 offset="0x9308" name="VPC_ATTR_BUF_GMEM_SIZE" variants="A7XX" type="uint" usage="rp_blit"/>
+	<reg32 offset="0x9309" name="VPC_ATTR_BUF_GMEM_BASE" variants="A7XX" type="uint" usage="rp_blit"/>
+
+	<reg32 offset="0x9b09" name="PC_ATTR_BUF_GMEM_SIZE" variants="A7XX" type="uint" usage="rp_blit"/>
+
+	<reg32 offset="0x930a" name="VPC_UNKNOWN_930A" variants="A7XX"/>
+
+	<reg32 offset="0x960a" name="VPC_FLATSHADE_MODE_CNTL" variants="A7XX"/>
 
 	<!-- 0x9307-0x95ff invalid -->
 
@@ -2163,52 +2277,62 @@ by a particular renderpass/blit.
 	<!-- TODO: regs from 0x9624-0x963a -->
 	<!-- 0x963b-0x97ff invalid -->
 
-	<reg32 offset="0x9800" name="PC_HS_PARAM_0" low="0" high="5" type="uint" usage="rp_blit"/>
+	<reg32 offset="0x9800" name="PC_HS_PARAM_0" low="0" high="5" type="uint" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<!-- always 0x0 ? -->
-	<reg32 offset="0x9801" name="PC_HS_PARAM_1" usage="rp_blit">
+	<bitset name="a6xx_pc_hs_param_1" inline="yes">
 		<bitfield name="SIZE" low="0" high="10" type="uint"/>
 		<bitfield name="UNK13" pos="13"/>
-	</reg32>
+	</bitset>
 
-	<reg32 offset="0x9802" name="PC_DS_PARAM" usage="rp_blit">
+	<reg32 offset="0x9801" name="PC_HS_PARAM_1" type="a6xx_pc_hs_param_1" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_pc_ds_param" inline="yes">
 		<bitfield name="SPACING" low="0" high="1" type="a6xx_tess_spacing"/>
 		<bitfield name="OUTPUT" low="2" high="3" type="a6xx_tess_output"/>
-	</reg32>
+	</bitset>
 
-	<reg32 offset="0x9803" name="PC_RESTART_INDEX" low="0" high="31" type="uint" usage="rp_blit"/>
-	<reg32 offset="0x9804" name="PC_MODE_CNTL" low="0" high="7" usage="rp_blit"/>
+	<reg32 offset="0x9802" name="PC_DS_PARAM" type="a6xx_pc_ds_param" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<reg32 offset="0x9803" name="PC_RESTART_INDEX" low="0" high="31" type="uint" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<reg32 offset="0x9804" name="PC_MODE_CNTL" low="0" high="7" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<reg32 offset="0x9805" name="PC_POWER_CNTL" low="0" high="2" usage="rp_blit"/>
 
-	<reg32 offset="0x9806" name="PC_PS_CNTL" usage="rp_blit">
+	<bitset name="a6xx_pc_ps_cntl" inline="yes">
 		<bitfield name="PRIMITIVEIDEN" pos="0" type="boolean"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x9806" name="PC_PS_CNTL" type="a6xx_pc_ps_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+
+	<bitset name="a6xx_pc_dgen_so_cntl" inline="yes">
+		<bitfield name="STREAM_ENABLE" low="15" high="18" type="hex"/>
+	</bitset>
 
 	<!-- New in a6xx gen3+ -->
-	<reg32 offset="0x9808" name="PC_DGEN_SO_CNTL" usage="rp_blit">
-		<bitfield name="STREAM_ENABLE" low="15" high="18" type="hex"/>
-	</reg32>
+	<reg32 offset="0x9808" name="PC_DGEN_SO_CNTL" type="a6xx_pc_dgen_so_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<reg32 offset="0x980a" name="PC_DGEN_SU_CONSERVATIVE_RAS_CNTL">
+	<bitset name="a6xx_pc_dgen_su_conservative_ras_cntl" inline="yes">
 		<bitfield name="CONSERVATIVERASEN" pos="0" type="boolean"/>
-	</reg32>
-	<!-- 0x980b-0x983f invalid -->
+	</bitset>
+
+	<reg32 offset="0x980a" name="PC_DGEN_SU_CONSERVATIVE_RAS_CNTL" type="a6xx_pc_dgen_su_conservative_ras_cntl" variants="A6XX-A7XX"/>
 
 	<!-- 0x9840 - 0x9842 are not readable -->
-	<reg32 offset="0x9840" name="PC_DRAW_INITIATOR">
+	<bitset name="a6xx_draw_initiator" inline="yes">
 		<bitfield name="STATE_ID" low="0" high="7"/>
-	</reg32>
+	</bitset>
 
-	<reg32 offset="0x9841" name="PC_KERNEL_INITIATOR">
-		<bitfield name="STATE_ID" low="0" high="7"/>
-	</reg32>
+	<reg32 offset="0x9840" name="PC_DRAW_INITIATOR" type="a6xx_draw_initiator" variants="A6XX-A7XX"/>
+	<reg32 offset="0x9841" name="PC_KERNEL_INITIATOR" type="a6xx_draw_initiator" variants="A6XX-A7XX"/>
 
-	<reg32 offset="0x9842" name="PC_EVENT_INITIATOR">
+	<bitset name="a6xx_event_initiator" inline="yes">
 		<!-- I think only the low bit is actually used? -->
 		<bitfield name="STATE_ID" low="16" high="23"/>
 		<bitfield name="EVENT" low="0" high="6" type="vgt_event_type"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x9842" name="PC_EVENT_INITIATOR" type="a6xx_event_initiator" variants="A6XX-A7XX"/>
 
 	<!--
 		0x9880 written in a lot of places by SQE, same value gets written
@@ -2219,45 +2343,21 @@ by a particular renderpass/blit.
 
 	<!-- 0x9843-0x997f invalid -->
 
-	<reg32 offset="0x9981" name="PC_DGEN_RAST_CNTL" variants="A6XX" usage="rp_blit">
-		<bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/>
-	</reg32>
-	<reg32 offset="0x9809" name="PC_DGEN_RAST_CNTL" variants="A7XX-" usage="rp_blit">
-		<bitfield name="MODE" low="0" high="1" type="a6xx_polygon_mode"/>
-	</reg32>
-
-	<reg32 offset="0x9980" name="VPC_RAST_STREAM_CNTL" variants="A6XX" usage="rp_blit">
-		<!-- which stream to send to GRAS -->
-		<bitfield name="STREAM" low="0" high="1" type="uint"/>
-		<!-- discard primitives before rasterization -->
-		<bitfield name="DISCARD" pos="2" type="boolean"/>
-	</reg32>
-	<!-- VPC_RAST_STREAM_CNTL -->
-	<reg32 offset="0x9107" name="VPC_RAST_STREAM_CNTL" variants="A7XX-" usage="rp_blit">
-		<!-- which stream to send to GRAS -->
-		<bitfield name="STREAM" low="0" high="1" type="uint"/>
-		<!-- discard primitives before rasterization -->
-		<bitfield name="DISCARD" pos="2" type="boolean"/>
-	</reg32>
-	<reg32 offset="0x9317" name="VPC_RAST_STREAM_CNTL_V2" variants="A7XX-" usage="rp_blit">
-		<!-- which stream to send to GRAS -->
-		<bitfield name="STREAM" low="0" high="1" type="uint"/>
-		<!-- discard primitives before rasterization -->
-		<bitfield name="DISCARD" pos="2" type="boolean"/>
-	</reg32>
+	<reg32 offset="0x9981" name="PC_DGEN_RAST_CNTL" type="a6xx_rast_cntl" variants="A6XX" usage="rp_blit"/>
+	<reg32 offset="0x9809" name="PC_DGEN_RAST_CNTL" type="a6xx_rast_cntl" variants="A7XX" usage="rp_blit"/>
 
 	<!-- Both are a750+.
 	     Probably needed to correctly overlap execution of several draws.
 	-->
-	<reg32 offset="0x9885" name="PC_HS_BUFFER_SIZE" variants="A7XX-" usage="cmd"/>
+	<reg32 offset="0x9885" name="PC_HS_BUFFER_SIZE" variants="A7XX" usage="cmd"/>
 	<!-- Blob adds a bit more space {0x10, 0x20, 0x30, 0x40} bytes, but the meaning of
 	     this additional space is not known.
 	-->
-	<reg32 offset="0x9886" name="PC_TF_BUFFER_SIZE" variants="A7XX-" usage="cmd"/>
+	<reg32 offset="0x9886" name="PC_TF_BUFFER_SIZE" variants="A7XX" usage="cmd"/>
 
 	<!-- 0x9982-0x9aff invalid -->
 
-	<reg32 offset="0x9b00" name="PC_CNTL" type="a6xx_pc_cntl" usage="rp_blit"/>
+	<reg32 offset="0x9b00" name="PC_CNTL" type="a6xx_pc_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<bitset name="a6xx_pc_xs_cntl" inline="yes">
 		<doc>
@@ -2270,18 +2370,18 @@ by a particular renderpass/blit.
 		<bitfield name="LAYER" pos="9" type="boolean"/>
 		<bitfield name="VIEW" pos="10" type="boolean"/>
 		<!-- note: PC_VS_CNTL doesn't have the PRIMITIVE_ID bit -->
+		<!-- since HS can't output anything, only PRIMITIVE_ID is valid -->
 		<bitfield name="PRIMITIVE_ID" pos="11" type="boolean"/>
 		<bitfield name="CLIP_MASK" low="16" high="23" type="uint"/>
 		<bitfield name="SHADINGRATE" pos="24" type="boolean" variants="A7XX-"/>
 	</bitset>
 
-	<reg32 offset="0x9b01" name="PC_VS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9b02" name="PC_GS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/>
-	<!-- since HS can't output anything, only PRIMITIVE_ID is valid -->
-	<reg32 offset="0x9b03" name="PC_HS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/>
-	<reg32 offset="0x9b04" name="PC_DS_CNTL" type="a6xx_pc_xs_cntl" usage="rp_blit"/>
+	<reg32 offset="0x9b01" name="PC_VS_CNTL" type="a6xx_pc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9b02" name="PC_GS_CNTL" type="a6xx_pc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9b03" name="PC_HS_CNTL" type="a6xx_pc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
+	<reg32 offset="0x9b04" name="PC_DS_CNTL" type="a6xx_pc_xs_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 
-	<reg32 offset="0x9b05" name="PC_GS_PARAM_0" type="a6xx_gs_param_0" usage="rp_blit"/>
+	<reg32 offset="0x9b05" name="PC_GS_PARAM_0" type="a6xx_gs_param_0" variants="A6XX-A7XX" usage="rp_blit"/>
 
 	<reg32 offset="0x9b06" name="PC_PRIMITIVE_CNTL_6" variants="A6XX" usage="rp_blit">
 		<doc>
@@ -2290,9 +2390,9 @@ by a particular renderpass/blit.
 		<bitfield name="STRIDE_IN_VPC" low="0" high="10" type="uint"/>
 	</reg32>
 
-	<reg32 offset="0x9b07" name="PC_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" usage="rp_blit"/>
+	<reg32 offset="0x9b07" name="PC_STEREO_RENDERING_CNTL" type="a6xx_stereo_rendering_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 	<!-- mask of enabled views, doesn't exist on A630 -->
-	<reg32 offset="0x9b08" name="PC_STEREO_RENDERING_VIEWMASK" type="hex" low="0" high="15" usage="rp_blit"/>
+	<reg32 offset="0x9b08" name="PC_STEREO_RENDERING_VIEWMASK" type="hex" low="0" high="15" variants="A6XX-A7XX" usage="rp_blit"/>
 	<!-- 0x9b09-0x9bff invalid -->
 	<reg32 offset="0x9c00" name="PC_2D_EVENT_CMD">
 		<!-- special register (but note first 8 bits can be written/read) -->
@@ -2303,34 +2403,39 @@ by a particular renderpass/blit.
 	<!-- TODO: 0x9e00-0xa000 range incomplete -->
 	<reg32 offset="0x9e00" name="PC_DBG_ECO_CNTL"/>
 	<reg32 offset="0x9e01" name="PC_ADDR_MODE_CNTL" type="a5xx_address_mode"/>
-	<reg64 offset="0x9e04" name="PC_DMA_BASE"/>
-	<reg32 offset="0x9e06" name="PC_DMA_OFFSET" type="uint"/>
-	<reg32 offset="0x9e07" name="PC_DMA_SIZE" type="uint"/>
-	<reg64 offset="0x9e08" name="PC_TESS_BASE" variants="A6XX" type="waddress" align="32" usage="cmd"/>
-	<reg64 offset="0x9810" name="PC_TESS_BASE" variants="A7XX-" type="waddress" align="32" usage="cmd"/>
+	<reg64 offset="0x9e04" name="PC_DMA_BASE" type="address" variants="A6XX-A7XX"/>
+	<reg32 offset="0x9e06" name="PC_DMA_OFFSET" type="uint" variants="A6XX-A7XX"/>
+	<reg32 offset="0x9e07" name="PC_DMA_SIZE" type="uint" variants="A6XX-A7XX"/>
 
-	<reg32 offset="0x9e0b" name="PC_DRAWCALL_CNTL" type="vgt_draw_initiator_a4xx">
+	<reg64 offset="0x9e08" name="PC_TESS_BASE" variants="A6XX" type="waddress" align="32" usage="cmd"/>
+	<reg64 offset="0x9810" name="PC_TESS_BASE" variants="A7XX" type="waddress" align="32" usage="cmd"/>
+
+	<reg32 offset="0x9e0b" name="PC_DRAWCALL_CNTL" type="vgt_draw_initiator_a4xx" variants="A6XX-A7XX">
 		<doc>
 			Possibly not really "initiating" the draw but the layout is similar
 			to VGT_DRAW_INITIATOR on older gens
 		</doc>
 	</reg32>
-	<reg32 offset="0x9e0c" name="PC_DRAWCALL_INSTANCE_NUM" type="uint"/>
-	<reg32 offset="0x9e0d" name="PC_DRAWCALL_SIZE" type="uint"/>
+	<reg32 offset="0x9e0c" name="PC_DRAWCALL_INSTANCE_NUM" type="uint" variants="A6XX-A7XX"/>
+	<reg32 offset="0x9e0d" name="PC_DRAWCALL_SIZE" type="uint" variants="A6XX-A7XX"/>
 
 	<!-- These match the contents of CP_SET_BIN_DATA (not written directly) -->
-	<reg32 offset="0x9e11" name="PC_VIS_STREAM_CNTL">
+	<bitset name="a6xx_pc_vis_stream_cntl" inline="yes">
 		<bitfield name="UNK0" low="0" high="15"/>
 		<bitfield name="VSC_SIZE" low="16" high="21" type="uint"/>
 		<bitfield name="VSC_N" low="22" high="26" type="uint"/>
-	</reg32>
-	<reg64 offset="0x9e12" name="PC_PVIS_STREAM_BIN_BASE" type="waddress" align="32"/>
-	<reg64 offset="0x9e14" name="PC_DVIS_STREAM_BIN_BASE" type="waddress" align="32"/>
+	</bitset>
 
-	<reg32 offset="0x9e1c" name="PC_DRAWCALL_CNTL_OVERRIDE">
+	<reg32 offset="0x9e11" name="PC_VIS_STREAM_CNTL" type="a6xx_pc_vis_stream_cntl" variants="A6XX-A7XX"/>
+	<reg64 offset="0x9e12" name="PC_PVIS_STREAM_BIN_BASE" type="waddress" align="32" variants="A6XX-A7XX"/>
+	<reg64 offset="0x9e14" name="PC_DVIS_STREAM_BIN_BASE" type="waddress" align="32" variants="A6XX-A7XX"/>
+
+	<bitset name="a6xx_pc_drawcall_cntl_override" inline="yes">
 		<doc>Written by CP_SET_VISIBILITY_OVERRIDE handler</doc>
 		<bitfield name="OVERRIDE" pos="0" type="boolean"/>
-	</reg32>
+	</bitset>
+
+	<reg32 offset="0x9e1c" name="PC_DRAWCALL_CNTL_OVERRIDE" type="a6xx_pc_drawcall_cntl_override" variants="A6XX-A7XX"/>
 
 	<reg32 offset="0x9e24" name="PC_UNKNOWN_9E24" variants="A7XX-" usage="cmd"/>
 
@@ -2936,7 +3041,7 @@ by a particular renderpass/blit.
 	<reg32 offset="0xa9b3" name="SP_CS_PROGRAM_COUNTER_OFFSET" type="uint" usage="cmd"/>
 	<reg64 offset="0xa9b4" name="SP_CS_BASE" type="address" align="32" usage="cmd"/>
 	<reg32 offset="0xa9b6" name="SP_CS_PVT_MEM_PARAM" type="a6xx_sp_xs_pvt_mem_param" usage="cmd"/>
-	<reg64 offset="0xa9b7" name="SP_CS_PVT_MEM_BASE" align="32" usage="cmd"/>
+	<reg64 offset="0xa9b7" name="SP_CS_PVT_MEM_BASE" type="waddress" align="32" usage="cmd"/>
 	<reg32 offset="0xa9b9" name="SP_CS_PVT_MEM_SIZE" type="a6xx_sp_xs_pvt_mem_size" usage="cmd"/>
 	<reg32 offset="0xa9ba" name="SP_CS_TSIZE" low="0" high="7" type="uint" usage="cmd"/>
 	<reg32 offset="0xa9bb" name="SP_CS_CONFIG" type="a6xx_sp_xs_config" usage="cmd"/>
@@ -3021,7 +3126,7 @@ by a particular renderpass/blit.
 	UAV state for compute shader:
 	 -->
 	<reg64 offset="0xa9f2" name="SP_CS_UAV_BASE" type="address" align="16" variants="A6XX"/>
-	<reg64 offset="0xa9f8" name="SP_CS_UAV_BASE" type="address" align="16" variants="A7XX"/>
+	<reg64 offset="0xa9f8" name="SP_CS_UAV_BASE" type="address" align="16" variants="A7XX-"/>
 	<reg32 offset="0xaa00" name="SP_CS_USIZE" low="0" high="6" type="uint"/>
 
 	<!-- Correlated with avgs/uvgs usage in FS -->
@@ -3104,14 +3209,19 @@ by a particular renderpass/blit.
 	instructions VS/HS/DS/GS/FS.  See SP_CS_UAV_BASE_* for compute shaders.
 	 -->
 	<reg64 offset="0xab1a" name="SP_GFX_UAV_BASE" type="address" align="16" usage="cmd"/>
-	<reg32 offset="0xab20" name="SP_GFX_USIZE" low="0" high="6" type="uint" usage="cmd"/>
+	<reg32 offset="0xab20" name="SP_GFX_USIZE" low="0" high="6" type="uint" variants="A6XX-A7XX" usage="cmd"/>
 
-	<reg32 offset="0xab22" name="SP_UNKNOWN_AB22" variants="A7XX-" usage="cmd"/>
+	<reg32 offset="0xab22" name="SP_UNKNOWN_AB22" variants="A7XX" usage="cmd"/>
+
+	<enum name="a6xx_sp_a2d_output_ifmt_type">
+		<value name="OUTPUT_IFMT_2D_FLOAT" value="0"/>
+		<value name="OUTPUT_IFMT_2D_SINT" value="1"/>
+		<value name="OUTPUT_IFMT_2D_UINT" value="2"/>
+	</enum>
 
 	<bitset name="a6xx_sp_a2d_output_info" inline="yes">
-		<bitfield name="NORM" pos="0" type="boolean"/>
-		<bitfield name="SINT" pos="1" type="boolean"/>
-		<bitfield name="UINT" pos="2" type="boolean"/>
+		<bitfield name="HALF_PRECISION" pos="0" type="boolean"/>
+		<bitfield name="IFMT_TYPE" low="1" high="2" type="a6xx_sp_a2d_output_ifmt_type"/>
 		<!-- looks like HW only cares about the base type of this format,
 		     which matches the ifmt? -->
 		<bitfield name="COLOR_FORMAT" low="3" high="10" type="a6xx_format"/>
@@ -3156,7 +3266,7 @@ by a particular renderpass/blit.
 	<reg32 offset="0xae6b" name="SP_UNKNOWN_AE6B" variants="A7XX-" usage="cmd"/>
 	<reg32 offset="0xae6c" name="SP_HLSQ_DBG_ECO_CNTL" variants="A7XX-" usage="cmd"/>
 	<reg32 offset="0xae6d" name="SP_READ_SEL" variants="A7XX-">
-		<bitfield name="LOCATION" low="18" high="19" type="a7xx_state_location"/>
+		<bitfield name="LOCATION" low="18" high="20" type="a7xx_state_location"/>
 		<bitfield name="PIPE" low="16" high="17" type="a7xx_pipe"/>
 		<bitfield name="STATETYPE" low="8" high="15" type="a7xx_statetype_id"/>
 		<bitfield name="USPTP" low="4" high="7"/>
@@ -3192,7 +3302,7 @@ by a particular renderpass/blit.
 
 	<!-- looks to work in the same way as a5xx: -->
 	<reg64 offset="0xb302" name="TPL1_GFX_BORDER_COLOR_BASE" type="address" align="128" usage="cmd"/>
-	<reg32 offset="0xb304" name="TPL1_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" usage="rp_blit"/>
+	<reg32 offset="0xb304" name="TPL1_MSAA_SAMPLE_POS_CNTL" type="a6xx_msaa_sample_pos_cntl" variants="A6XX-A7XX" usage="rp_blit"/>
 	<reg32 offset="0xb305" name="TPL1_PROGRAMMABLE_MSAA_POS_0" type="a6xx_programmable_msaa_pos" usage="rp_blit"/>
 	<reg32 offset="0xb306" name="TPL1_PROGRAMMABLE_MSAA_POS_1" type="a6xx_programmable_msaa_pos" usage="rp_blit"/>
 	<reg32 offset="0xb307" name="TPL1_WINDOW_OFFSET" type="a6xx_reg_xy" usage="rp_blit"/>
@@ -3232,12 +3342,12 @@ by a particular renderpass/blit.
 	</reg32>
 
 	<reg32 offset="0xb2c0" name="TPL1_A2D_SRC_TEXTURE_INFO" type="a6xx_a2d_src_texture_info" variants="A7XX-" usage="rp_blit"/>
-	<reg32 offset="0xb2c1" name="TPL1_A2D_SRC_TEXTURE_SIZE" variants="A7XX">
+	<reg32 offset="0xb2c1" name="TPL1_A2D_SRC_TEXTURE_SIZE" variants="A7XX-">
 		<bitfield name="WIDTH" low="0" high="14" type="uint"/>
 		<bitfield name="HEIGHT" low="15" high="29" type="uint"/>
 	</reg32>
 	<reg64 offset="0xb2c2" name="TPL1_A2D_SRC_TEXTURE_BASE" type="address" align="16" variants="A7XX-" usage="rp_blit"/>
-	<reg32 offset="0xb2c4" name="TPL1_A2D_SRC_TEXTURE_PITCH" variants="A7XX">
+	<reg32 offset="0xb2c4" name="TPL1_A2D_SRC_TEXTURE_PITCH" variants="A7XX-">
 		<!--
 		Bits from 3..9 must be zero unless 'TPL1_A2D_BLT_CNTL::TYPE'
 		is A6XX_TEX_IMG_BUFFER, which allows for lower alignment.
@@ -3270,13 +3380,13 @@ by a particular renderpass/blit.
 	<reg32 offset="0xb2ce" name="SP_PS_UNKNOWN_B4CE" low="0" high="31" variants="A7XX"/>
 	<reg32 offset="0xb2cf" name="SP_PS_UNKNOWN_B4CF" low="0" high="30" variants="A7XX"/>
 	<reg32 offset="0xb2d0" name="SP_PS_UNKNOWN_B4D0" low="0" high="29" variants="A7XX"/>
-	<reg32 offset="0xb2d1" name="TPL1_A2D_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX"/>
+	<reg32 offset="0xb2d1" name="TPL1_A2D_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX-"/>
 	<reg32 offset="0xb2d2" name="TPL1_A2D_BLT_CNTL" variants="A7XX-" usage="rp_blit">
 		<bitfield name="RAW_COPY" pos="0" type="boolean"/>
 		<bitfield name="START_OFFSET_TEXELS" low="16" high="21"/>
 		<bitfield name="TYPE" low="29" high="31" type="a6xx_tex_type"/>
 	</reg32>
-	<reg32 offset="0xab21" name="SP_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX-" usage="rp_blit"/>
+	<reg32 offset="0xab21" name="SP_WINDOW_OFFSET" type="a6xx_reg_xy" variants="A7XX" usage="rp_blit"/>
 
 	<!-- always 0x100000 or 0x1000000? -->
 	<reg32 offset="0xb600" name="TPL1_DBG_ECO_CNTL" low="0" high="25" usage="cmd"/>
@@ -3634,7 +3744,7 @@ by a particular renderpass/blit.
 	<reg32 offset="0xbb10" name="SP_PS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A6XX" usage="rp_blit"/>
 	<reg32 offset="0xab03" name="SP_PS_CONST_CONFIG" type="a6xx_xs_const_config" variants="A7XX-" usage="rp_blit"/>
 
-	<array offset="0xab40" name="SP_SHARED_CONSTANT_GFX_0" stride="1" length="64" variants="A7XX-"/>
+	<array offset="0xab40" name="SP_SHARED_CONSTANT_GFX" stride="1" length="64" variants="A7XX"/>
 
 	<reg32 offset="0xbb11" name="HLSQ_SHARED_CONSTS" variants="A6XX" usage="cmd">
 		<doc>
@@ -3796,7 +3906,7 @@ by a particular renderpass/blit.
 	<reg32 offset="0x0030" name="CFG_DBGBUS_TRACE_BUF2"/>
 </domain>
 
-<domain name="A7XX_CX_DBGC" width="32">
+<domain name="A7XX_CX_DBGC" width="32" varset="chip">
 	<!-- Bitfields shifted, but otherwise the same: -->
 	<reg32 offset="0x0000" name="CFG_DBGBUS_SEL_A" variants="A7XX-">
 		<bitfield high="7" low="0" name="PING_INDEX"/>
diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml
index 307d43dda8a2..56cfaff614a4 100644
--- a/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/a6xx_descriptors.xml
@@ -9,38 +9,6 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 
 <domain name="A6XX_TEX_SAMP" width="32">
 	<doc>Texture sampler dwords</doc>
-	<enum name="a6xx_tex_filter"> <!-- same as a4xx? -->
-		<value name="A6XX_TEX_NEAREST" value="0"/>
-		<value name="A6XX_TEX_LINEAR" value="1"/>
-		<value name="A6XX_TEX_ANISO" value="2"/>
-		<value name="A6XX_TEX_CUBIC" value="3"/> <!-- a650 only -->
-	</enum>
-	<enum name="a6xx_tex_clamp"> <!-- same as a4xx? -->
-		<value name="A6XX_TEX_REPEAT" value="0"/>
-		<value name="A6XX_TEX_CLAMP_TO_EDGE" value="1"/>
-		<value name="A6XX_TEX_MIRROR_REPEAT" value="2"/>
-		<value name="A6XX_TEX_CLAMP_TO_BORDER" value="3"/>
-		<value name="A6XX_TEX_MIRROR_CLAMP" value="4"/>
-	</enum>
-	<enum name="a6xx_tex_aniso"> <!-- same as a4xx? -->
-		<value name="A6XX_TEX_ANISO_1" value="0"/>
-		<value name="A6XX_TEX_ANISO_2" value="1"/>
-		<value name="A6XX_TEX_ANISO_4" value="2"/>
-		<value name="A6XX_TEX_ANISO_8" value="3"/>
-		<value name="A6XX_TEX_ANISO_16" value="4"/>
-	</enum>
-	<enum name="a6xx_reduction_mode">
-		<value name="A6XX_REDUCTION_MODE_AVERAGE" value="0"/>
-		<value name="A6XX_REDUCTION_MODE_MIN" value="1"/>
-		<value name="A6XX_REDUCTION_MODE_MAX" value="2"/>
-	</enum>
-	<enum name="a6xx_fast_border_color">
-		<!--                           R B G A -->
-		<value name="A6XX_BORDER_COLOR_0_0_0_0" value="0"/>
-		<value name="A6XX_BORDER_COLOR_0_0_0_1" value="1"/>
-		<value name="A6XX_BORDER_COLOR_1_1_1_0" value="2"/>
-		<value name="A6XX_BORDER_COLOR_1_1_1_1" value="3"/>
-	</enum>
 
 	<reg32 offset="0" name="0">
 		<bitfield name="MIPFILTER_LINEAR_NEAR" pos="0" type="boolean"/>
@@ -79,14 +47,6 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 
 <domain name="A6XX_TEX_CONST" width="32" varset="chip">
 	<doc>Texture constant dwords</doc>
-	<enum name="a6xx_tex_swiz"> <!-- same as a4xx? -->
-		<value name="A6XX_TEX_X" value="0"/>
-		<value name="A6XX_TEX_Y" value="1"/>
-		<value name="A6XX_TEX_Z" value="2"/>
-		<value name="A6XX_TEX_W" value="3"/>
-		<value name="A6XX_TEX_ZERO" value="4"/>
-		<value name="A6XX_TEX_ONE" value="5"/>
-	</enum>
 	<reg32 offset="0" name="0">
 		<bitfield name="TILE_MODE" low="0" high="1" type="a6xx_tile_mode"/>
 		<bitfield name="SRGB" pos="2" type="boolean"/>
diff --git a/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml b/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml
index 665539b098c6..4e42f055b85f 100644
--- a/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/a6xx_enums.xml
@@ -320,14 +320,14 @@ to upconvert to 32b float internally?
 16b float:   3
  -->
 <enum name="a6xx_2d_ifmt">
-	<value value="0x10" name="R2D_UNORM8"/>
 	<value value="0x7"  name="R2D_INT32"/>
 	<value value="0x6"  name="R2D_INT16"/>
 	<value value="0x5"  name="R2D_INT8"/>
 	<value value="0x4"  name="R2D_FLOAT32"/>
 	<value value="0x3"  name="R2D_FLOAT16"/>
+	<value value="0x2"  name="R2D_SNORM8"/>
 	<value value="0x1"  name="R2D_UNORM8_SRGB"/>
-	<value value="0x0"  name="R2D_RAW"/>
+	<value value="0x0"  name="R2D_UNORM8"/>
 </enum>
 
 <enum name="a6xx_tex_type">
@@ -380,4 +380,50 @@ to upconvert to 32b float internally?
 	<value value="0x3" name="TESS_CCW_TRIS"/>
 </enum>
 
+<enum name="a6xx_tex_filter"> <!-- same as a4xx? -->
+	<value name="A6XX_TEX_NEAREST" value="0"/>
+	<value name="A6XX_TEX_LINEAR" value="1"/>
+	<value name="A6XX_TEX_ANISO" value="2"/>
+	<value name="A6XX_TEX_CUBIC" value="3"/> <!-- a650 only -->
+</enum>
+
+<enum name="a6xx_tex_clamp"> <!-- same as a4xx? -->
+	<value name="A6XX_TEX_REPEAT" value="0"/>
+	<value name="A6XX_TEX_CLAMP_TO_EDGE" value="1"/>
+	<value name="A6XX_TEX_MIRROR_REPEAT" value="2"/>
+	<value name="A6XX_TEX_CLAMP_TO_BORDER" value="3"/>
+	<value name="A6XX_TEX_MIRROR_CLAMP" value="4"/>
+</enum>
+
+<enum name="a6xx_tex_aniso"> <!-- same as a4xx? -->
+	<value name="A6XX_TEX_ANISO_1" value="0"/>
+	<value name="A6XX_TEX_ANISO_2" value="1"/>
+	<value name="A6XX_TEX_ANISO_4" value="2"/>
+	<value name="A6XX_TEX_ANISO_8" value="3"/>
+	<value name="A6XX_TEX_ANISO_16" value="4"/>
+</enum>
+
+<enum name="a6xx_reduction_mode">
+	<value name="A6XX_REDUCTION_MODE_AVERAGE" value="0"/>
+	<value name="A6XX_REDUCTION_MODE_MIN" value="1"/>
+	<value name="A6XX_REDUCTION_MODE_MAX" value="2"/>
+</enum>
+
+<enum name="a6xx_fast_border_color">
+	<!--                           R B G A -->
+	<value name="A6XX_BORDER_COLOR_0_0_0_0" value="0"/>
+	<value name="A6XX_BORDER_COLOR_0_0_0_1" value="1"/>
+	<value name="A6XX_BORDER_COLOR_1_1_1_0" value="2"/>
+	<value name="A6XX_BORDER_COLOR_1_1_1_1" value="3"/>
+</enum>
+
+<enum name="a6xx_tex_swiz"> <!-- same as a4xx? -->
+	<value name="A6XX_TEX_X" value="0"/>
+	<value name="A6XX_TEX_Y" value="1"/>
+	<value name="A6XX_TEX_Z" value="2"/>
+	<value name="A6XX_TEX_W" value="3"/>
+	<value name="A6XX_TEX_ZERO" value="4"/>
+	<value name="A6XX_TEX_ONE" value="5"/>
+</enum>
+
 </database>
diff --git a/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml b/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml
index 7abc08635495..0e10e1c6d263 100644
--- a/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml
+++ b/drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml
@@ -120,12 +120,12 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 	<value name="LRZ_FLUSH" value="38" variants="A5XX-"/>
 	<value name="BLIT_OP_FILL_2D" value="39" variants="A5XX-"/>
 	<value name="BLIT_OP_COPY_2D" value="40" variants="A5XX-A6XX"/>
-	<value name="UNK_40" value="40" variants="A7XX"/>
+	<value name="LRZ_CACHE_INVALIDATE" value="40" variants="A7XX"/>
 	<value name="LRZ_Q_CACHE_INVALIDATE" value="41" variants="A7XX"/>
 	<value name="BLIT_OP_SCALE_2D" value="42" variants="A5XX-"/>
 	<value name="CONTEXT_DONE_2D" value="43" variants="A5XX-"/>
-	<value name="UNK_2C" value="44" variants="A5XX-"/>
-	<value name="UNK_2D" value="45" variants="A5XX-"/>
+	<value name="VSC_BINNING_START" value="44" variants="A5XX-"/>
+	<value name="VSC_BINNING_END" value="45" variants="A5XX-"/>
 
 	<!-- a6xx events -->
 	<doc>
@@ -523,7 +523,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 	<!--
 	Seems to set the mode flags which control which CP_SET_DRAW_STATE
 	packets are executed, based on their ENABLE_MASK values
-	
+
 	CP_SET_MODE w/ payload of 0x1 seems to cause CP_SET_DRAW_STATE
 	packets w/ ENABLE_MASK & 0x6 to execute immediately
 	 -->
@@ -640,8 +640,7 @@ xsi:schemaLocation="https://gitlab.freedesktop.org/freedreno/ rules-fd.xsd">
 	<value name="CP_BV_BR_COUNT_OPS" value="0x1b" variants="A7XX-"/>
 	<doc> Clears, adds to local, or adds to global timestamp </doc>
 	<value name="CP_MODIFY_TIMESTAMP" value="0x1c" variants="A7XX-"/>
-	<!-- similar to CP_CONTEXT_REG_BUNCH, but discards first two dwords?? -->
-	<value name="CP_CONTEXT_REG_BUNCH2" value="0x5d" variants="A7XX-"/>
+	<value name="CP_NON_CONTEXT_REG_BUNCH" value="0x5d" variants="A7XX-"/>
 	<doc>
 		Write to a scratch memory that is read by CP_REG_TEST with
 		SOURCE_SCRATCH_MEM set. It's not the same scratch as scratch registers.
@@ -918,12 +917,6 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	</reg32>
 
 	<stripe varset="chip" variants="A5XX-">
-		<reg32 offset="4" name="4">
-			<bitfield name="INDX_BASE_LO" low="0" high="31"/>
-		</reg32>
-		<reg32 offset="5" name="5">
-			<bitfield name="INDX_BASE_HI" low="0" high="31"/>
-		</reg32>
 		<reg64 offset="4" name="INDX_BASE" type="address"/>
 		<reg32 offset="6" name="6">
 			<!-- max # of elements in index buffer -->
@@ -1099,8 +1092,10 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 			<bitfield name="BINNING" pos="20" varset="chip" variants="A6XX-" type="boolean"/>
 			<bitfield name="GMEM" pos="21" varset="chip" variants="A6XX-" type="boolean"/>
 			<bitfield name="SYSMEM" pos="22" varset="chip" variants="A6XX-" type="boolean"/>
-			<bitfield name="GROUP_ID" low="24" high="28" type="uint"/>
+			<!-- high bit is 28 until a750: -->
+			<bitfield name="GROUP_ID" low="24" high="29" type="uint"/>
 		</reg32>
+		<reg64 offset="1" name="ADDR" type="address"/>
 		<reg32 offset="1" name="1">
 			<bitfield name="ADDR_LO" low="0" high="31" type="hex"/>
 		</reg32>
@@ -1166,26 +1161,11 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	</reg32>
 	<stripe varset="a7xx_abs_mask_mode" variants="NO_ABS_MASK">
 		<!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS -->
-		<reg32 offset="1" name="1">
-			<bitfield name="BIN_DATA_ADDR_LO" low="0" high="31" type="hex"/>
-		</reg32>
-		<reg32 offset="2" name="2">
-			<bitfield name="BIN_DATA_ADDR_HI" low="0" high="31" type="hex"/>
-		</reg32>
+		<reg64 offset="1" name="BIN_DATA_ADDR" type="address"/>
 		<!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)-->
-		<reg32 offset="3" name="3">
-			<bitfield name="BIN_SIZE_ADDRESS_LO" low="0" high="31"/>
-		</reg32>
-		<reg32 offset="4" name="4">
-			<bitfield name="BIN_SIZE_ADDRESS_HI" low="0" high="31"/>
-		</reg32>
+		<reg64 offset="3" name="BIN_SIZE_ADDR" type="address"/>
 		<!-- new on a6xx, where BIN_DATA_ADDR is the DRAW_STRM: -->
-		<reg32 offset="5" name="5">
-			<bitfield name="BIN_PRIM_STRM_LO" low="0" high="31"/>
-		</reg32>
-		<reg32 offset="6" name="6">
-			<bitfield name="BIN_PRIM_STRM_HI" low="0" high="31"/>
-		</reg32>
+		<reg64 offset="5" name="BIN_PRIM_STRM" type="address"/>
 		<!--
 			a7xx adds a few more addresses to the end of the pkt
 		 -->
@@ -1195,26 +1175,11 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	<stripe varset="a7xx_abs_mask_mode" variants="ABS_MASK">
 		<reg32 offset="1" name="ABS_MASK"/>
 		<!-- BIN_DATA_ADDR -> VSC_PIPE[p].DATA_ADDRESS -->
-		<reg32 offset="2" name="2">
-			<bitfield name="BIN_DATA_ADDR_LO" low="0" high="31" type="hex"/>
-		</reg32>
-		<reg32 offset="3" name="3">
-			<bitfield name="BIN_DATA_ADDR_HI" low="0" high="31" type="hex"/>
-		</reg32>
+		<reg64 offset="2" name="BIN_DATA_ADDR" type="address"/>
 		<!-- BIN_SIZE_ADDRESS -> VSC_SIZE_ADDRESS + (p * 4)-->
-		<reg32 offset="4" name="4">
-			<bitfield name="BIN_SIZE_ADDRESS_LO" low="0" high="31"/>
-		</reg32>
-		<reg32 offset="5" name="5">
-			<bitfield name="BIN_SIZE_ADDRESS_HI" low="0" high="31"/>
-		</reg32>
+		<reg64 offset="4" name="BIN_SIZE_ADDR" type="address"/>
 		<!-- new on a6xx, where BIN_DATA_ADDR is the DRAW_STRM: -->
-		<reg32 offset="6" name="6">
-			<bitfield name="BIN_PRIM_STRM_LO" low="0" high="31"/>
-		</reg32>
-		<reg32 offset="7" name="7">
-			<bitfield name="BIN_PRIM_STRM_HI" low="0" high="31"/>
-		</reg32>
+		<reg64 offset="6" name="BIN_PRIM_STRM" type="address"/>
 		<!--
 			a7xx adds a few more addresses to the end of the pkt
 		 -->
@@ -1300,7 +1265,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	</reg32>
 </domain>
 
-<domain name="CP_REG_TO_MEM" width="32">
+<domain name="CP_REG_TO_MEM" width="32" prefix="chip">
 	<reg32 offset="0" name="0">
 		<bitfield name="REG" low="0" high="17" type="hex"/>
 		<!-- number of registers/dwords copied is max(CNT, 1). -->
@@ -1308,12 +1273,12 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<bitfield name="64B" pos="30" type="boolean"/>
 		<bitfield name="ACCUMULATE" pos="31" type="boolean"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="DEST" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="2" name="2" varset="chip" variants="A5XX-">
-		<bitfield name="DEST_HI" low="0" high="31"/>
-	</reg32>
+	<stripe varset="chip" variants="A2XX-A4XX">
+		<reg32 offset="1" name="DEST" type="address"/>
+	</stripe>
+	<stripe varset="chip" variants="A5XX-">
+		<reg64 offset="1" name="DEST" type="address"/>
+	</stripe>
 </domain>
 
 <domain name="CP_REG_TO_MEM_OFFSET_REG" width="32">
@@ -1329,12 +1294,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<bitfield name="64B" pos="30" type="boolean"/>
 		<bitfield name="ACCUMULATE" pos="31" type="boolean"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="DEST" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="2" name="2" varset="chip" variants="A5XX-">
-		<bitfield name="DEST_HI" low="0" high="31"/>
-	</reg32>
+	<reg64 offset="1" name="DEST" type="waddress"/>
 	<reg32 offset="3" name="3">
 		<bitfield name="OFFSET0" low="0" high="17" type="hex"/>
 		<bitfield name="OFFSET0_SCRATCH" pos="19" type="boolean"/>
@@ -1354,18 +1314,8 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<bitfield name="64B" pos="30" type="boolean"/>
 		<bitfield name="ACCUMULATE" pos="31" type="boolean"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="DEST" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="2" name="2" varset="chip" variants="A5XX-">
-		<bitfield name="DEST_HI" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="3" name="3">
-		<bitfield name="OFFSET_LO" low="0" high="31" type="hex"/>
-	</reg32>
-	<reg32 offset="4" name="4">
-		<bitfield name="OFFSET_HI" low="0" high="31" type="hex"/>
-	</reg32>
+	<reg64 offset="1" name="DEST" type="waddress"/>
+	<reg64 offset="3" name="OFFSET" type="waddress"/>
 </domain>
 
 <domain name="CP_MEM_TO_REG" width="32">
@@ -1378,12 +1328,12 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<!-- does the same thing as CP_MEM_TO_MEM::UNK31 -->
 		<bitfield name="UNK31" pos="31" type="boolean"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="SRC" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="2" name="2" varset="chip" variants="A5XX-">
-		<bitfield name="SRC_HI" low="0" high="31"/>
-	</reg32>
+	<stripe varset="chip" variants="A2XX-A4XX">
+		<reg32 offset="1" name="SRC" type="address"/>
+	</stripe>
+	<stripe varset="chip" variants="A5XX-">
+		<reg64 offset="1" name="SRC" type="address"/>
+	</stripe>
 </domain>
 
 <domain name="CP_MEM_TO_MEM" width="32">
@@ -1403,6 +1353,10 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<!-- some other kind of wait -->
 		<bitfield name="UNK31" pos="31" type="boolean"/>
 	</reg32>
+	<reg64 offset="1" name="DST" type="waddress"/>
+	<reg64 offset="3" name="SRC_A" type="address"/>
+	<reg64 offset="5" name="SRC_B" type="address"/>
+	<reg64 offset="7" name="SRC_C" type="address"/>
 	<!--
 	followed by sequence of addresses.. the first is the
 	destination and the rest are N src addresses which are
@@ -1461,12 +1415,12 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 </domain>
 
 <domain name="CP_MEM_WRITE" width="32">
-	<reg32 offset="0" name="0">
-		<bitfield name="ADDR_LO" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="ADDR_HI" low="0" high="31"/>
-	</reg32>
+	<stripe varset="chip" variants="A2XX-A4XX">
+		<reg32 offset="0" name="ADDR" type="address"/>
+	</stripe>
+	<stripe varset="chip" variants="A5XX-">
+		<reg64 offset="0" name="ADDR" type="address"/>
+	</stripe>
 	<!-- followed by the DWORDs to write -->
 </domain>
 
@@ -1518,24 +1472,14 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<bitfield name="POLL" low="4" high="5" type="poll_memory_type"/>
 		<bitfield name="WRITE_MEMORY" pos="8" type="boolean"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="POLL_ADDR_LO" low="0" high="31" type="hex"/>
-	</reg32>
-	<reg32 offset="2" name="2">
-		<bitfield name="POLL_ADDR_HI" low="0" high="31" type="hex"/>
-	</reg32>
+	<reg64 offset="1" name="POLL_ADDR" type="address"/>
 	<reg32 offset="3" name="3">
 		<bitfield name="REF" low="0" high="31"/>
 	</reg32>
 	<reg32 offset="4" name="4">
 		<bitfield name="MASK" low="0" high="31"/>
 	</reg32>
-	<reg32 offset="5" name="5">
-		<bitfield name="WRITE_ADDR_LO" low="0" high="31" type="hex"/>
-	</reg32>
-	<reg32 offset="6" name="6">
-		<bitfield name="WRITE_ADDR_HI" low="0" high="31" type="hex"/>
-	</reg32>
+	<reg64 offset="5" name="WRITE_ADDR" type="waddress"/>
 	<reg32 offset="7" name="7">
 		<bitfield name="WRITE_DATA" low="0" high="31"/>
 	</reg32>
@@ -1550,12 +1494,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<!-- Reserved for flags, presumably? Unused in FW -->
 		<bitfield name="RESERVED" low="0" high="31" type="hex"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="POLL_ADDR_LO" low="0" high="31" type="hex"/>
-	</reg32>
-	<reg32 offset="2" name="2">
-		<bitfield name="POLL_ADDR_HI" low="0" high="31" type="hex"/>
-	</reg32>
+	<reg64 offset="1" name="POLL_ADDR" type="address"/>
 	<reg32 offset="3" name="3">
 		<bitfield name="REF" low="0" high="31"/>
 	</reg32>
@@ -1573,12 +1512,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		<bitfield name="POLL" low="4" high="5" type="poll_memory_type"/>
 		<bitfield name="WRITE_MEMORY" pos="8" type="boolean"/>
 	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="POLL_ADDR_LO" low="0" high="31" type="hex"/>
-	</reg32>
-	<reg32 offset="2" name="2">
-		<bitfield name="POLL_ADDR_HI" low="0" high="31" type="hex"/>
-	</reg32>
+	<reg64 offset="1" name="POLL_ADDR" type="address"/>
 	<reg32 offset="3" name="3">
 		<bitfield name="REF" low="0" high="31"/>
 	</reg32>
@@ -1712,12 +1646,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	TODO what is gpuaddr for, seems to be all 0's.. maybe needed for
 	context switch?
 	 -->
-	<reg32 offset="1" name="1">
-		<bitfield name="ADDR_0_LO" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="2" name="2">
-		<bitfield name="ADDR_0_HI" low="0" high="31"/>
-	</reg32>
+	<reg64 offset="1" name="ADDR" type="waddress"/>
 	<reg32 offset="3" name="3">
 		<!-- ??? -->
 	</reg32>
@@ -1832,9 +1761,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 	<reg32 offset="0" name="0">
 	</reg32>
 	<stripe varset="chip" variants="A4XX">
-		<reg32 offset="1" name="1">
-			<bitfield name="ADDR" low="0" high="31"/>
-		</reg32>
+		<reg32 offset="1" name="ADDR" type="address"/>
 		<reg32 offset="2" name="2">
 			<!-- localsize is value minus one: -->
 			<bitfield name="LOCALSIZEX" low="2" high="11" type="uint"/>
@@ -1843,12 +1770,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 		</reg32>
 	</stripe>
 	<stripe varset="chip" variants="A5XX-">
-		<reg32 offset="1" name="1">
-			<bitfield name="ADDR_LO" low="0" high="31"/>
-		</reg32>
-		<reg32 offset="2" name="2">
-			<bitfield name="ADDR_HI" low="0" high="31"/>
-		</reg32>
+		<reg64 offset="1" name="ADDR" type="address"/>
 		<reg32 offset="3" name="3">
 			<!-- localsize is value minus one: -->
 			<bitfield name="LOCALSIZEX" low="2" high="11" type="uint"/>
@@ -2161,12 +2083,7 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords)
 			</doc>
 		</value>
 	</enum>
-	<reg32 offset="0" name="0">
-		<bitfield name="ADDR_LO" low="0" high="31"/>
-	</reg32>
-	<reg32 offset="1" name="1">
-		<bitfield name="ADDR_HI" low="0" high="31"/>
-	</reg32>
+	<reg64 offset="0" name="ADDR" type="address"/>
 	<reg32 offset="2" name="2">
 		<bitfield name="DWORDS" low="0" high="19" type="uint"/>
 		<bitfield name="TYPE" low="20" high="21" type="amble_type"/>

From 50243079865ae7c150bc54ea3ed59077cdf3da03 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Thu, 4 Sep 2025 12:16:39 +1000
Subject: [PATCH 1195/1455] ttm/bo: add an API to populate a bo before
 exporting.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

While discussing cgroups we noticed a problem where you could export
a BO to a dma-buf without having it ever being backed or accounted for.

This meant in low memory situations or eventually with cgroups, a
lower privledged process might cause the compositor to try and allocate
a lot of memory on it's behalf and this could fail. At least make
sure the exporter has managed to allocate the RAM at least once
before exporting the object.

This only applies currently to TTM_PL_SYSTEM objects, because
GTT objects get populated on first validate, and VRAM doesn't
use TT.

Reviewed-by: Christian Koenig <christian.koenig@amd.com>
Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Cc: Simona Vetter <simona.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://lore.kernel.org/r/20250904021643.2050497-1-airlied@gmail.com
---
 drivers/gpu/drm/ttm/ttm_bo.c | 15 +++++++++++++++
 include/drm/ttm/ttm_bo.h     |  2 ++
 2 files changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index f4d9e68b21e7..29423ceeec5c 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -1283,3 +1283,18 @@ int ttm_bo_populate(struct ttm_buffer_object *bo,
 	return 0;
 }
 EXPORT_SYMBOL(ttm_bo_populate);
+
+int ttm_bo_setup_export(struct ttm_buffer_object *bo,
+			struct ttm_operation_ctx *ctx)
+{
+	int ret;
+
+	ret = ttm_bo_reserve(bo, false, false, NULL);
+	if (ret != 0)
+		return ret;
+
+	ret = ttm_bo_populate(bo, ctx);
+	ttm_bo_unreserve(bo);
+	return ret;
+}
+EXPORT_SYMBOL(ttm_bo_setup_export);
diff --git a/include/drm/ttm/ttm_bo.h b/include/drm/ttm/ttm_bo.h
index 479b7ed075c0..e664a96540eb 100644
--- a/include/drm/ttm/ttm_bo.h
+++ b/include/drm/ttm/ttm_bo.h
@@ -466,6 +466,8 @@ pgprot_t ttm_io_prot(struct ttm_buffer_object *bo, struct ttm_resource *res,
 void ttm_bo_tt_destroy(struct ttm_buffer_object *bo);
 int ttm_bo_populate(struct ttm_buffer_object *bo,
 		    struct ttm_operation_ctx *ctx);
+int ttm_bo_setup_export(struct ttm_buffer_object *bo,
+			struct ttm_operation_ctx *ctx);
 
 /* Driver LRU walk helpers initially targeted for shrinking. */
 

From 619ddf57cfddf81ffb75905cc5f533e61206b10c Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Thu, 4 Sep 2025 12:16:40 +1000
Subject: [PATCH 1196/1455] amdgpu: populate buffers before exporting them.

Before exporting a buffer, make sure it has been populated with
pages at least once.

While discussing cgroups we noticed a problem where you could export
a BO to a dma-buf without having it ever being backed or accounted for.

This meant in low memory situations or eventually with cgroups, a
lower privledged process might cause the compositor to try and allocate
a lot of memory on it's behalf and this could fail. At least make
sure the exporter has managed to allocate the RAM at least once
before exporting the object.

This only applies currently to TTM_PL_SYSTEM objects, because
GTT objects get populated on first validate, and VRAM doesn't
use TT.

Reviewed-by: Christian Koenig <christian.koenig@amd.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://lore.kernel.org/r/20250904021643.2050497-2-airlied@gmail.com
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index ff98c87b2e0b..878b13efd154 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -313,11 +313,23 @@ struct dma_buf *amdgpu_gem_prime_export(struct drm_gem_object *gobj,
 {
 	struct amdgpu_bo *bo = gem_to_amdgpu_bo(gobj);
 	struct dma_buf *buf;
+	struct ttm_operation_ctx ctx = {
+		.interruptible = true,
+		.no_wait_gpu = true,
+		/* We opt to avoid OOM on system pages allocations */
+		.gfp_retry_mayfail = true,
+		.allow_res_evict = false,
+	};
+	int ret;
 
 	if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm) ||
 	    bo->flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID)
 		return ERR_PTR(-EPERM);
 
+	ret = ttm_bo_setup_export(&bo->tbo, &ctx);
+	if (ret)
+		return ERR_PTR(ret);
+
 	buf = drm_gem_prime_export(gobj, flags);
 	if (!IS_ERR(buf))
 		buf->ops = &amdgpu_dmabuf_ops;

From 3629e1b22e5ba2475fe2531e59c9446c4979e2f9 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Thu, 4 Sep 2025 12:16:41 +1000
Subject: [PATCH 1197/1455] nouveau: populate buffers before exporting them.

Before exporting a buffer, make sure it has been populated with
pages at least once.

While discussing cgroups we noticed a problem where you could export
a BO to a dma-buf without having it ever being backed or accounted for.

This meant in low memory situations or eventually with cgroups, a
lower privledged process might cause the compositor to try and allocate
a lot of memory on it's behalf and this could fail. At least make
sure the exporter has managed to allocate the RAM at least once
before exporting the object.

This only applies currently to TTM_PL_SYSTEM objects, because
GTT objects get populated on first validate, and VRAM doesn't
use TT.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://lore.kernel.org/r/20250904021643.2050497-3-airlied@gmail.com
---
 drivers/gpu/drm/nouveau/nouveau_prime.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/nouveau/nouveau_prime.c b/drivers/gpu/drm/nouveau/nouveau_prime.c
index cd95446d6851..caab60fc62f6 100644
--- a/drivers/gpu/drm/nouveau/nouveau_prime.c
+++ b/drivers/gpu/drm/nouveau/nouveau_prime.c
@@ -108,9 +108,21 @@ struct dma_buf *nouveau_gem_prime_export(struct drm_gem_object *gobj,
 					 int flags)
 {
 	struct nouveau_bo *nvbo = nouveau_gem_object(gobj);
+	struct ttm_operation_ctx ctx = {
+		.interruptible = true,
+		.no_wait_gpu = true,
+		/* We opt to avoid OOM on system pages allocations */
+		.gfp_retry_mayfail = true,
+		.allow_res_evict = false,
+	};
+	int ret;
 
 	if (nvbo->no_share)
 		return ERR_PTR(-EPERM);
 
+	ret = ttm_bo_setup_export(&nvbo->bo, &ctx);
+	if (ret)
+		return ERR_PTR(ret);
+
 	return drm_gem_prime_export(gobj, flags);
 }

From 91494dee1091a14d91da6bcb39e12a907765c793 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@redhat.com>
Date: Thu, 4 Sep 2025 12:16:42 +1000
Subject: [PATCH 1198/1455] xe: populate buffers before exporting them.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Before exporting a buffer, make sure it has been populated with
pages at least once.

While discussing cgroups we noticed a problem where you could export
a BO to a dma-buf without having it ever being backed or accounted for.

This meant in low memory situations or eventually with cgroups, a
lower privledged process might cause the compositor to try and allocate
a lot of memory on it's behalf and this could fail. At least make
sure the exporter has managed to allocate the RAM at least once
before exporting the object.

This only applies currently to TTM_PL_SYSTEM objects, because
GTT objects get populated on first validate, and VRAM doesn't
use TT.

Cc: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://lore.kernel.org/r/20250904021643.2050497-4-airlied@gmail.com
---
 drivers/gpu/drm/xe/xe_dma_buf.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_dma_buf.c b/drivers/gpu/drm/xe/xe_dma_buf.c
index 346f857f3837..71b70e17bddd 100644
--- a/drivers/gpu/drm/xe/xe_dma_buf.c
+++ b/drivers/gpu/drm/xe/xe_dma_buf.c
@@ -191,10 +191,22 @@ struct dma_buf *xe_gem_prime_export(struct drm_gem_object *obj, int flags)
 {
 	struct xe_bo *bo = gem_to_xe_bo(obj);
 	struct dma_buf *buf;
+	struct ttm_operation_ctx ctx = {
+		.interruptible = true,
+		.no_wait_gpu = true,
+		/* We opt to avoid OOM on system pages allocations */
+		.gfp_retry_mayfail = true,
+		.allow_res_evict = false,
+	};
+	int ret;
 
 	if (bo->vm)
 		return ERR_PTR(-EPERM);
 
+	ret = ttm_bo_setup_export(&bo->ttm, &ctx);
+	if (ret)
+		return ERR_PTR(ret);
+
 	buf = drm_gem_prime_export(obj, flags);
 	if (!IS_ERR(buf))
 		buf->ops = &xe_dmabuf_ops;

From b98775bca99511cc22ab459a2de646cd2fa7241f Mon Sep 17 00:00:00 2001
From: Shuicheng Lin <shuicheng.lin@intel.com>
Date: Tue, 19 Aug 2025 15:39:51 +0000
Subject: [PATCH 1199/1455] drm/xe/tile: Release kobject for the failure path

Call kobject_put() for the failure path to release the kobject

v2: remove extra newline. (Matt)

Fixes: e3d0839aa501 ("drm/xe/tile: Abort driver load for sysfs creation failure")
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Link: https://lore.kernel.org/r/20250819153950.2973344-2-shuicheng.lin@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_tile_sysfs.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_tile_sysfs.c b/drivers/gpu/drm/xe/xe_tile_sysfs.c
index b804234a6551..9e1236a9ec67 100644
--- a/drivers/gpu/drm/xe/xe_tile_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_tile_sysfs.c
@@ -44,16 +44,18 @@ int xe_tile_sysfs_init(struct xe_tile *tile)
 	kt->tile = tile;
 
 	err = kobject_add(&kt->base, &dev->kobj, "tile%d", tile->id);
-	if (err) {
-		kobject_put(&kt->base);
-		return err;
-	}
+	if (err)
+		goto err_object;
 
 	tile->sysfs = &kt->base;
 
 	err = xe_vram_freq_sysfs_init(tile);
 	if (err)
-		return err;
+		goto err_object;
 
 	return devm_add_action_or_reset(xe->drm.dev, tile_sysfs_fini, tile);
+
+err_object:
+	kobject_put(&kt->base);
+	return err;
 }

From 692a4802435ba7389810b15b476ed360675e38c1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Wed, 10 Sep 2025 17:11:28 +0200
Subject: [PATCH 1200/1455] drm/xe: Fix uninitialized return values
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

clang warned about two uninitialized variables used as return
values in the exhaustive eviction series.

Fix those.

Fixes: 1f1541720f65 ("drm/xe: Rework instances of variants of xe_bo_create_locked()")
Fixes: 7bcb6e38c14d ("drm/xe/display: Convert __xe_pin_fb_vma()")
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com>
Link: https://lore.kernel.org/r/20250910151128.49693-1-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h | 2 +-
 drivers/gpu/drm/xe/display/xe_fb_pin.c                       | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
index a2d1f684a3a8..f097fc6d5127 100644
--- a/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
+++ b/drivers/gpu/drm/xe/compat-i915-headers/gem/i915_gem_stolen.h
@@ -22,7 +22,7 @@ static inline int i915_gem_stolen_insert_node_in_range(struct xe_device *xe,
 						       u32 start, u32 end)
 {
 	struct xe_bo *bo;
-	int err;
+	int err = 0;
 	u32 flags = XE_BO_FLAG_PINNED | XE_BO_FLAG_STOLEN;
 
 	if (start < SZ_4K)
diff --git a/drivers/gpu/drm/xe/display/xe_fb_pin.c b/drivers/gpu/drm/xe/display/xe_fb_pin.c
index 7e82e49544fa..25fc22e00b58 100644
--- a/drivers/gpu/drm/xe/display/xe_fb_pin.c
+++ b/drivers/gpu/drm/xe/display/xe_fb_pin.c
@@ -283,7 +283,7 @@ static struct i915_vma *__xe_pin_fb_vma(const struct intel_framebuffer *fb,
 	struct xe_bo *bo = gem_to_xe_bo(obj);
 	struct xe_validation_ctx ctx;
 	struct drm_exec exec;
-	int ret;
+	int ret = 0;
 
 	if (!vma)
 		return ERR_PTR(-ENODEV);

From 72136efb875d8438c20b9c8ab72945d474933471 Mon Sep 17 00:00:00 2001
From: Aaron Ma <aaron.ma@canonical.com>
Date: Sat, 23 Aug 2025 20:16:47 +0800
Subject: [PATCH 1201/1455] drm/i915/backlight: Honor VESA eDP backlight
 luminance control capability

The VESA AUX backlight fails to be enable luminance based backlight
mainpulation becaused luminance_set is false by default.
Fix it by using luminance support control capabitliy.

Fixes: e13af5166a359 ("drm/i915/backlight: Use drm helper to initialize edp backlight")
Signed-off-by: Aaron Ma <aaron.ma@canonical.com>
Reviewed-by: Suraj Kandpal <suraj.kandpal@intel.com>
Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com>
Link: https://lore.kernel.org/r/20250823121647.275834-1-aaron.ma@canonical.com
---
 drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
index 0dc45db56b82..eb05ef4bd9f6 100644
--- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
+++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c
@@ -530,7 +530,7 @@ static int intel_dp_aux_vesa_setup_backlight(struct intel_connector *connector,
 				     luminance_range->max_luminance,
 				     panel->vbt.backlight.pwm_freq_hz,
 				     intel_dp->edp_dpcd, &current_level, &current_mode,
-				     false);
+				     panel->backlight.edp.vesa.luminance_control_support);
 	if (ret < 0)
 		return ret;
 

From cf4fd52e323604ccfa8390917593e1fb965653ee Mon Sep 17 00:00:00 2001
From: Daniel Almeida <daniel.almeida@collabora.com>
Date: Wed, 10 Sep 2025 10:51:18 -0300
Subject: [PATCH 1202/1455] rust: drm: Introduce the Tyr driver for Arm Mali
 GPUs

Add a Rust driver for ARM Mali CSF-based GPUs. It is a port of Panthor
and therefore exposes Panthor's uAPI and name to userspace, and the
product of a joint effort between Collabora, Arm and Google engineers.

The aim is to incrementally develop Tyr with the abstractions that are
currently available until it is consider to be in parity with Panthor
feature-wise.

The development of Tyr itself started in January, after a few failed
attempts of converting Panthor piecewise through a mix of Rust and C
code. There is a downstream branch that's much further ahead in terms of
capabilities than this initial patch.

The downstream code is capable of booting the MCU, doing sync VM_BINDS
through the work-in-progress GPUVM abstraction and also doing (trivial)
submits through Asahi's drm_scheduler and dma_fence abstractions. So
basically, most of what one would expect a modern GPU driver to do,
except for power management and some other very important adjacent
pieces. It is not at the point where submits can correctly deal with
dependencies, or at the point where it can rotate access to the GPU
hardware fairly through a software scheduler, but that is simply a
matter of writing more code.

This first patch, however, only implements a subset of the current
features available downstream, as the rest is not implementable without
pulling in even more abstractions. In particular, a lot of things depend
on properly mapping memory on a given VA range, which itself depends on
the GPUVM abstraction that is currently work-in-progress. For this
reason, we still cannot boot the MCU and thus, cannot do much for the
moment.

This constitutes a change in the overall strategy that we have been
using to develop Tyr so far. By submitting small parts of the driver
upstream iteratively, we aim to:

a) evolve together with Nova and rvkms, hopefully reducing regressions
due to upstream changes (that may break us because we were not there, in
the first place)

b) prove any work-in-progress abstractions by having them run on a real
driver and hardware and,

c) provide a reason to work on and review said abstractions by providing
a user, which would be tyr itself.

Despite its limited feature-set, we offer IGT tests. It is only tested
on the rk3588, so any other SoC is probably not going to work at all for
now.

The skeleton is basically taken from Nova and also
rust_platform_driver.rs.

Lastly, the name "Tyr" is inspired by Norse mythology, reflecting ARM's
tradition of naming their GPUs after Nordic mythological figures and
places.

Co-developed-by: Beata Michalska <beata.michalska@arm.com>
Signed-off-by: Beata Michalska <beata.michalska@arm.com>
Co-developed-by: Carsten Haitzler <carsten.haitzler@foss.arm.com>
Signed-off-by: Carsten Haitzler <carsten.haitzler@foss.arm.com>
Co-developed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://www.collabora.com/news-and-blog/news-and-events/introducing-tyr-a-new-rust-drm-driver.html
Signed-off-by: Daniel Almeida <daniel.almeida@collabora.com>
Acked-by: Boris Brezillon <boris.brezillon@collabora.com>
[aliceryhl: minor Kconfig update on apply]
[aliceryhl: s/drm::device::/drm::/]
Link: https://lore.kernel.org/r/20250910-tyr-v3-1-dba3bc2ae623@collabora.com
Co-developed-by: Alice Ryhl <aliceryhl@google.com>
Signed-off-by: Alice Ryhl <aliceryhl@google.com>
---
 MAINTAINERS                   |  13 ++
 drivers/gpu/drm/Kconfig       |   2 +
 drivers/gpu/drm/Makefile      |   1 +
 drivers/gpu/drm/tyr/Kconfig   |  19 +++
 drivers/gpu/drm/tyr/Makefile  |   3 +
 drivers/gpu/drm/tyr/driver.rs | 205 +++++++++++++++++++++++++++++++
 drivers/gpu/drm/tyr/file.rs   |  56 +++++++++
 drivers/gpu/drm/tyr/gem.rs    |  18 +++
 drivers/gpu/drm/tyr/gpu.rs    | 219 ++++++++++++++++++++++++++++++++++
 drivers/gpu/drm/tyr/regs.rs   | 108 +++++++++++++++++
 drivers/gpu/drm/tyr/tyr.rs    |  22 ++++
 rust/uapi/uapi_helper.h       |   1 +
 12 files changed, 667 insertions(+)
 create mode 100644 drivers/gpu/drm/tyr/Kconfig
 create mode 100644 drivers/gpu/drm/tyr/Makefile
 create mode 100644 drivers/gpu/drm/tyr/driver.rs
 create mode 100644 drivers/gpu/drm/tyr/file.rs
 create mode 100644 drivers/gpu/drm/tyr/gem.rs
 create mode 100644 drivers/gpu/drm/tyr/gpu.rs
 create mode 100644 drivers/gpu/drm/tyr/regs.rs
 create mode 100644 drivers/gpu/drm/tyr/tyr.rs

diff --git a/MAINTAINERS b/MAINTAINERS
index 62298a601bbe..8a11e6c5dd80 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2086,6 +2086,19 @@ F:	Documentation/devicetree/bindings/gpu/arm,mali-valhall-csf.yaml
 F:	drivers/gpu/drm/panthor/
 F:	include/uapi/drm/panthor_drm.h
 
+ARM MALI TYR DRM DRIVER
+M:	Daniel Almeida <daniel.almeida@collabora.com>
+M:	Alice Ryhl <aliceryhl@google.com>
+L:	dri-devel@lists.freedesktop.org
+S:	Supported
+W:	https://rust-for-linux.com/tyr-gpu-driver
+W	https://drm.pages.freedesktop.org/maintainer-tools/drm-rust.html
+B:	https://gitlab.freedesktop.org/panfrost/linux/-/issues
+T:	git https://gitlab.freedesktop.org/drm/rust/kernel.git
+F:	Documentation/devicetree/bindings/gpu/arm,mali-valhall-csf.yaml
+F:	drivers/gpu/drm/tyr/
+F:	include/uapi/drm/panthor_drm.h
+
 ARM MALI-DP DRM DRIVER
 M:	Liviu Dudau <liviu.dudau@arm.com>
 S:	Supported
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index f7ea8e895c0c..fda170730468 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -396,6 +396,8 @@ source "drivers/gpu/drm/sprd/Kconfig"
 
 source "drivers/gpu/drm/imagination/Kconfig"
 
+source "drivers/gpu/drm/tyr/Kconfig"
+
 config DRM_HYPERV
 	tristate "DRM Support for Hyper-V synthetic video device"
 	depends on DRM && PCI && HYPERV
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 4dafbdc8f86a..4b2f7d794275 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -220,6 +220,7 @@ obj-$(CONFIG_DRM_VBOXVIDEO) += vboxvideo/
 obj-$(CONFIG_DRM_LIMA)  += lima/
 obj-$(CONFIG_DRM_PANFROST) += panfrost/
 obj-$(CONFIG_DRM_PANTHOR) += panthor/
+obj-$(CONFIG_DRM_TYR) += tyr/
 obj-$(CONFIG_DRM_ASPEED_GFX) += aspeed/
 obj-$(CONFIG_DRM_MCDE) += mcde/
 obj-$(CONFIG_DRM_TIDSS) += tidss/
diff --git a/drivers/gpu/drm/tyr/Kconfig b/drivers/gpu/drm/tyr/Kconfig
new file mode 100644
index 000000000000..4b55308fd2eb
--- /dev/null
+++ b/drivers/gpu/drm/tyr/Kconfig
@@ -0,0 +1,19 @@
+# SPDX-License-Identifier: GPL-2.0 or MIT
+
+config DRM_TYR
+	tristate "Tyr (Rust DRM support for ARM Mali CSF-based GPUs)"
+	depends on DRM=y
+	depends on RUST
+	depends on ARM || ARM64 || COMPILE_TEST
+	depends on !GENERIC_ATOMIC64  # for IOMMU_IO_PGTABLE_LPAE
+	default n
+	help
+	  Rust DRM driver for ARM Mali CSF-based GPUs.
+
+	  This driver is for Mali (or Immortalis) Valhall Gxxx GPUs.
+
+	  Note that the Mali-G68 and Mali-G78, while Valhall architecture, will
+	  be supported with the panfrost driver as they are not CSF GPUs.
+
+	  if M is selected, the module will be called tyr. This driver is work
+	  in progress and may not be functional.
diff --git a/drivers/gpu/drm/tyr/Makefile b/drivers/gpu/drm/tyr/Makefile
new file mode 100644
index 000000000000..ba545f65f2c0
--- /dev/null
+++ b/drivers/gpu/drm/tyr/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0 or MIT
+
+obj-$(CONFIG_DRM_TYR) += tyr.o
diff --git a/drivers/gpu/drm/tyr/driver.rs b/drivers/gpu/drm/tyr/driver.rs
new file mode 100644
index 000000000000..d5625dd1e41c
--- /dev/null
+++ b/drivers/gpu/drm/tyr/driver.rs
@@ -0,0 +1,205 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+
+use kernel::c_str;
+use kernel::clk::Clk;
+use kernel::clk::OptionalClk;
+use kernel::device::Bound;
+use kernel::device::Core;
+use kernel::device::Device;
+use kernel::devres::Devres;
+use kernel::drm;
+use kernel::drm::ioctl;
+use kernel::new_mutex;
+use kernel::of;
+use kernel::platform;
+use kernel::prelude::*;
+use kernel::regulator;
+use kernel::regulator::Regulator;
+use kernel::sizes::SZ_2M;
+use kernel::sync::Arc;
+use kernel::sync::Mutex;
+use kernel::time;
+use kernel::types::ARef;
+
+use crate::file::File;
+use crate::gem::TyrObject;
+use crate::gpu;
+use crate::gpu::GpuInfo;
+use crate::regs;
+
+pub(crate) type IoMem = kernel::io::mem::IoMem<SZ_2M>;
+
+/// Convenience type alias for the DRM device type for this driver.
+pub(crate) type TyrDevice = drm::Device<TyrDriver>;
+
+#[pin_data(PinnedDrop)]
+pub(crate) struct TyrDriver {
+    device: ARef<TyrDevice>,
+}
+
+#[pin_data(PinnedDrop)]
+pub(crate) struct TyrData {
+    pub(crate) pdev: ARef<platform::Device>,
+
+    #[pin]
+    clks: Mutex<Clocks>,
+
+    #[pin]
+    regulators: Mutex<Regulators>,
+
+    /// Some information on the GPU.
+    ///
+    /// This is mainly queried by userspace, i.e.: Mesa.
+    pub(crate) gpu_info: GpuInfo,
+}
+
+// Both `Clk` and `Regulator` do not implement `Send` or `Sync`, but they
+// should. There are patches on the mailing list to address this, but they have
+// not landed yet.
+//
+// For now, add this workaround so that this patch compiles with the promise
+// that it will be removed in a future patch.
+//
+// SAFETY: This will be removed in a future patch.
+unsafe impl Send for TyrData {}
+// SAFETY: This will be removed in a future patch.
+unsafe impl Sync for TyrData {}
+
+fn issue_soft_reset(dev: &Device<Bound>, iomem: &Devres<IoMem>) -> Result {
+    regs::GPU_CMD.write(dev, iomem, regs::GPU_CMD_SOFT_RESET)?;
+
+    // TODO: We cannot poll, as there is no support in Rust currently, so we
+    // sleep. Change this when read_poll_timeout() is implemented in Rust.
+    kernel::time::delay::fsleep(time::Delta::from_millis(100));
+
+    if regs::GPU_IRQ_RAWSTAT.read(dev, iomem)? & regs::GPU_IRQ_RAWSTAT_RESET_COMPLETED == 0 {
+        dev_err!(dev, "GPU reset failed with errno\n");
+        dev_err!(
+            dev,
+            "GPU_INT_RAWSTAT is {}\n",
+            regs::GPU_IRQ_RAWSTAT.read(dev, iomem)?
+        );
+
+        return Err(EIO);
+    }
+
+    Ok(())
+}
+
+kernel::of_device_table!(
+    OF_TABLE,
+    MODULE_OF_TABLE,
+    <TyrDriver as platform::Driver>::IdInfo,
+    [
+        (of::DeviceId::new(c_str!("rockchip,rk3588-mali")), ()),
+        (of::DeviceId::new(c_str!("arm,mali-valhall-csf")), ())
+    ]
+);
+
+impl platform::Driver for TyrDriver {
+    type IdInfo = ();
+    const OF_ID_TABLE: Option<of::IdTable<Self::IdInfo>> = Some(&OF_TABLE);
+
+    fn probe(
+        pdev: &platform::Device<Core>,
+        _info: Option<&Self::IdInfo>,
+    ) -> Result<Pin<KBox<Self>>> {
+        let core_clk = Clk::get(pdev.as_ref(), Some(c_str!("core")))?;
+        let stacks_clk = OptionalClk::get(pdev.as_ref(), Some(c_str!("stacks")))?;
+        let coregroup_clk = OptionalClk::get(pdev.as_ref(), Some(c_str!("coregroup")))?;
+
+        core_clk.prepare_enable()?;
+        stacks_clk.prepare_enable()?;
+        coregroup_clk.prepare_enable()?;
+
+        let mali_regulator = Regulator::<regulator::Enabled>::get(pdev.as_ref(), c_str!("mali"))?;
+        let sram_regulator = Regulator::<regulator::Enabled>::get(pdev.as_ref(), c_str!("sram"))?;
+
+        let request = pdev.io_request_by_index(0).ok_or(ENODEV)?;
+        let iomem = Arc::pin_init(request.iomap_sized::<SZ_2M>(), GFP_KERNEL)?;
+
+        issue_soft_reset(pdev.as_ref(), &iomem)?;
+        gpu::l2_power_on(pdev.as_ref(), &iomem)?;
+
+        let gpu_info = GpuInfo::new(pdev.as_ref(), &iomem)?;
+        gpu_info.log(pdev);
+
+        let platform: ARef<platform::Device> = pdev.into();
+
+        let data = try_pin_init!(TyrData {
+                pdev: platform.clone(),
+                clks <- new_mutex!(Clocks {
+                    core: core_clk,
+                    stacks: stacks_clk,
+                    coregroup: coregroup_clk,
+                }),
+                regulators <- new_mutex!(Regulators {
+                    mali: mali_regulator,
+                    sram: sram_regulator,
+                }),
+                gpu_info,
+        });
+
+        let tdev: ARef<TyrDevice> = drm::Device::new(pdev.as_ref(), data)?;
+        drm::driver::Registration::new_foreign_owned(&tdev, pdev.as_ref(), 0)?;
+
+        let driver = KBox::pin_init(try_pin_init!(TyrDriver { device: tdev }), GFP_KERNEL)?;
+
+        // We need this to be dev_info!() because dev_dbg!() does not work at
+        // all in Rust for now, and we need to see whether probe succeeded.
+        dev_info!(pdev.as_ref(), "Tyr initialized correctly.\n");
+        Ok(driver)
+    }
+}
+
+#[pinned_drop]
+impl PinnedDrop for TyrDriver {
+    fn drop(self: Pin<&mut Self>) {}
+}
+
+#[pinned_drop]
+impl PinnedDrop for TyrData {
+    fn drop(self: Pin<&mut Self>) {
+        // TODO: the type-state pattern for Clks will fix this.
+        let clks = self.clks.lock();
+        clks.core.disable_unprepare();
+        clks.stacks.disable_unprepare();
+        clks.coregroup.disable_unprepare();
+    }
+}
+
+// We need to retain the name "panthor" to achieve drop-in compatibility with
+// the C driver in the userspace stack.
+const INFO: drm::DriverInfo = drm::DriverInfo {
+    major: 1,
+    minor: 5,
+    patchlevel: 0,
+    name: c_str!("panthor"),
+    desc: c_str!("ARM Mali Tyr DRM driver"),
+};
+
+#[vtable]
+impl drm::Driver for TyrDriver {
+    type Data = TyrData;
+    type File = File;
+    type Object = drm::gem::Object<TyrObject>;
+
+    const INFO: drm::DriverInfo = INFO;
+
+    kernel::declare_drm_ioctls! {
+        (PANTHOR_DEV_QUERY, drm_panthor_dev_query, ioctl::RENDER_ALLOW, File::dev_query),
+    }
+}
+
+#[pin_data]
+struct Clocks {
+    core: Clk,
+    stacks: OptionalClk,
+    coregroup: OptionalClk,
+}
+
+#[pin_data]
+struct Regulators {
+    mali: Regulator<regulator::Enabled>,
+    sram: Regulator<regulator::Enabled>,
+}
diff --git a/drivers/gpu/drm/tyr/file.rs b/drivers/gpu/drm/tyr/file.rs
new file mode 100644
index 000000000000..0ef432947b73
--- /dev/null
+++ b/drivers/gpu/drm/tyr/file.rs
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+
+use kernel::drm;
+use kernel::prelude::*;
+use kernel::uaccess::UserSlice;
+use kernel::uapi;
+
+use crate::driver::TyrDevice;
+use crate::TyrDriver;
+
+#[pin_data]
+pub(crate) struct File {}
+
+/// Convenience type alias for our DRM `File` type
+pub(crate) type DrmFile = drm::file::File<File>;
+
+impl drm::file::DriverFile for File {
+    type Driver = TyrDriver;
+
+    fn open(_dev: &drm::Device<Self::Driver>) -> Result<Pin<KBox<Self>>> {
+        KBox::try_pin_init(try_pin_init!(Self {}), GFP_KERNEL)
+    }
+}
+
+impl File {
+    pub(crate) fn dev_query(
+        tdev: &TyrDevice,
+        devquery: &mut uapi::drm_panthor_dev_query,
+        _file: &DrmFile,
+    ) -> Result<u32> {
+        if devquery.pointer == 0 {
+            match devquery.type_ {
+                uapi::drm_panthor_dev_query_type_DRM_PANTHOR_DEV_QUERY_GPU_INFO => {
+                    devquery.size = core::mem::size_of_val(&tdev.gpu_info) as u32;
+                    Ok(0)
+                }
+                _ => Err(EINVAL),
+            }
+        } else {
+            match devquery.type_ {
+                uapi::drm_panthor_dev_query_type_DRM_PANTHOR_DEV_QUERY_GPU_INFO => {
+                    let mut writer = UserSlice::new(
+                        UserPtr::from_addr(devquery.pointer as usize),
+                        devquery.size as usize,
+                    )
+                    .writer();
+
+                    writer.write(&tdev.gpu_info)?;
+
+                    Ok(0)
+                }
+                _ => Err(EINVAL),
+            }
+        }
+    }
+}
diff --git a/drivers/gpu/drm/tyr/gem.rs b/drivers/gpu/drm/tyr/gem.rs
new file mode 100644
index 000000000000..1273bf89dbd5
--- /dev/null
+++ b/drivers/gpu/drm/tyr/gem.rs
@@ -0,0 +1,18 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+
+use crate::driver::TyrDevice;
+use crate::driver::TyrDriver;
+use kernel::drm::gem;
+use kernel::prelude::*;
+
+/// GEM Object inner driver data
+#[pin_data]
+pub(crate) struct TyrObject {}
+
+impl gem::DriverObject for TyrObject {
+    type Driver = TyrDriver;
+
+    fn new(_dev: &TyrDevice, _size: usize) -> impl PinInit<Self, Error> {
+        try_pin_init!(TyrObject {})
+    }
+}
diff --git a/drivers/gpu/drm/tyr/gpu.rs b/drivers/gpu/drm/tyr/gpu.rs
new file mode 100644
index 000000000000..6c582910dd5d
--- /dev/null
+++ b/drivers/gpu/drm/tyr/gpu.rs
@@ -0,0 +1,219 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+
+use kernel::bits::genmask_u32;
+use kernel::device::Bound;
+use kernel::device::Device;
+use kernel::devres::Devres;
+use kernel::platform;
+use kernel::prelude::*;
+use kernel::time;
+use kernel::transmute::AsBytes;
+
+use crate::driver::IoMem;
+use crate::regs;
+
+/// Struct containing information that can be queried by userspace. This is read from
+/// the GPU's registers.
+///
+/// # Invariants
+///
+/// - The layout of this struct identical to the C `struct drm_panthor_gpu_info`.
+#[repr(C)]
+pub(crate) struct GpuInfo {
+    pub(crate) gpu_id: u32,
+    pub(crate) gpu_rev: u32,
+    pub(crate) csf_id: u32,
+    pub(crate) l2_features: u32,
+    pub(crate) tiler_features: u32,
+    pub(crate) mem_features: u32,
+    pub(crate) mmu_features: u32,
+    pub(crate) thread_features: u32,
+    pub(crate) max_threads: u32,
+    pub(crate) thread_max_workgroup_size: u32,
+    pub(crate) thread_max_barrier_size: u32,
+    pub(crate) coherency_features: u32,
+    pub(crate) texture_features: [u32; 4],
+    pub(crate) as_present: u32,
+    pub(crate) pad0: u32,
+    pub(crate) shader_present: u64,
+    pub(crate) l2_present: u64,
+    pub(crate) tiler_present: u64,
+    pub(crate) core_features: u32,
+    pub(crate) pad: u32,
+}
+
+impl GpuInfo {
+    pub(crate) fn new(dev: &Device<Bound>, iomem: &Devres<IoMem>) -> Result<Self> {
+        let gpu_id = regs::GPU_ID.read(dev, iomem)?;
+        let csf_id = regs::GPU_CSF_ID.read(dev, iomem)?;
+        let gpu_rev = regs::GPU_REVID.read(dev, iomem)?;
+        let core_features = regs::GPU_CORE_FEATURES.read(dev, iomem)?;
+        let l2_features = regs::GPU_L2_FEATURES.read(dev, iomem)?;
+        let tiler_features = regs::GPU_TILER_FEATURES.read(dev, iomem)?;
+        let mem_features = regs::GPU_MEM_FEATURES.read(dev, iomem)?;
+        let mmu_features = regs::GPU_MMU_FEATURES.read(dev, iomem)?;
+        let thread_features = regs::GPU_THREAD_FEATURES.read(dev, iomem)?;
+        let max_threads = regs::GPU_THREAD_MAX_THREADS.read(dev, iomem)?;
+        let thread_max_workgroup_size = regs::GPU_THREAD_MAX_WORKGROUP_SIZE.read(dev, iomem)?;
+        let thread_max_barrier_size = regs::GPU_THREAD_MAX_BARRIER_SIZE.read(dev, iomem)?;
+        let coherency_features = regs::GPU_COHERENCY_FEATURES.read(dev, iomem)?;
+
+        let texture_features = regs::GPU_TEXTURE_FEATURES0.read(dev, iomem)?;
+
+        let as_present = regs::GPU_AS_PRESENT.read(dev, iomem)?;
+
+        let shader_present = u64::from(regs::GPU_SHADER_PRESENT_LO.read(dev, iomem)?);
+        let shader_present =
+            shader_present | u64::from(regs::GPU_SHADER_PRESENT_HI.read(dev, iomem)?) << 32;
+
+        let tiler_present = u64::from(regs::GPU_TILER_PRESENT_LO.read(dev, iomem)?);
+        let tiler_present =
+            tiler_present | u64::from(regs::GPU_TILER_PRESENT_HI.read(dev, iomem)?) << 32;
+
+        let l2_present = u64::from(regs::GPU_L2_PRESENT_LO.read(dev, iomem)?);
+        let l2_present = l2_present | u64::from(regs::GPU_L2_PRESENT_HI.read(dev, iomem)?) << 32;
+
+        Ok(Self {
+            gpu_id,
+            gpu_rev,
+            csf_id,
+            l2_features,
+            tiler_features,
+            mem_features,
+            mmu_features,
+            thread_features,
+            max_threads,
+            thread_max_workgroup_size,
+            thread_max_barrier_size,
+            coherency_features,
+            // TODO: Add texture_features_{1,2,3}.
+            texture_features: [texture_features, 0, 0, 0],
+            as_present,
+            pad0: 0,
+            shader_present,
+            l2_present,
+            tiler_present,
+            core_features,
+            pad: 0,
+        })
+    }
+
+    pub(crate) fn log(&self, pdev: &platform::Device) {
+        let major = (self.gpu_id >> 16) & 0xff;
+        let minor = (self.gpu_id >> 8) & 0xff;
+        let status = self.gpu_id & 0xff;
+
+        let model_name = if let Some(model) = GPU_MODELS
+            .iter()
+            .find(|&f| f.major == major && f.minor == minor)
+        {
+            model.name
+        } else {
+            "unknown"
+        };
+
+        dev_info!(
+            pdev.as_ref(),
+            "mali-{} id 0x{:x} major 0x{:x} minor 0x{:x} status 0x{:x}",
+            model_name,
+            self.gpu_id >> 16,
+            major,
+            minor,
+            status
+        );
+
+        dev_info!(
+            pdev.as_ref(),
+            "Features: L2:{:#x} Tiler:{:#x} Mem:{:#x} MMU:{:#x} AS:{:#x}",
+            self.l2_features,
+            self.tiler_features,
+            self.mem_features,
+            self.mmu_features,
+            self.as_present
+        );
+
+        dev_info!(
+            pdev.as_ref(),
+            "shader_present=0x{:016x} l2_present=0x{:016x} tiler_present=0x{:016x}",
+            self.shader_present,
+            self.l2_present,
+            self.tiler_present
+        );
+    }
+
+    /// Returns the number of virtual address bits supported by the GPU.
+    #[expect(dead_code)]
+    pub(crate) fn va_bits(&self) -> u32 {
+        self.mmu_features & genmask_u32(0..=7)
+    }
+
+    /// Returns the number of physical address bits supported by the GPU.
+    #[expect(dead_code)]
+    pub(crate) fn pa_bits(&self) -> u32 {
+        (self.mmu_features >> 8) & genmask_u32(0..=7)
+    }
+}
+
+// SAFETY: `GpuInfo`'s invariant guarantees that it is the same type that is
+// already exposed to userspace by the C driver. This implies that it fulfills
+// the requirements for `AsBytes`.
+//
+// This means:
+//
+// - No implicit padding,
+// - No kernel pointers,
+// - No interior mutability.
+unsafe impl AsBytes for GpuInfo {}
+
+struct GpuModels {
+    name: &'static str,
+    major: u32,
+    minor: u32,
+}
+
+const GPU_MODELS: [GpuModels; 1] = [GpuModels {
+    name: "g610",
+    major: 10,
+    minor: 7,
+}];
+
+#[allow(dead_code)]
+pub(crate) struct GpuId {
+    pub(crate) arch_major: u32,
+    pub(crate) arch_minor: u32,
+    pub(crate) arch_rev: u32,
+    pub(crate) prod_major: u32,
+    pub(crate) ver_major: u32,
+    pub(crate) ver_minor: u32,
+    pub(crate) ver_status: u32,
+}
+
+impl From<u32> for GpuId {
+    fn from(value: u32) -> Self {
+        GpuId {
+            arch_major: (value & genmask_u32(28..=31)) >> 28,
+            arch_minor: (value & genmask_u32(24..=27)) >> 24,
+            arch_rev: (value & genmask_u32(20..=23)) >> 20,
+            prod_major: (value & genmask_u32(16..=19)) >> 16,
+            ver_major: (value & genmask_u32(12..=15)) >> 12,
+            ver_minor: (value & genmask_u32(4..=11)) >> 4,
+            ver_status: value & genmask_u32(0..=3),
+        }
+    }
+}
+
+/// Powers on the l2 block.
+pub(crate) fn l2_power_on(dev: &Device<Bound>, iomem: &Devres<IoMem>) -> Result {
+    regs::L2_PWRON_LO.write(dev, iomem, 1)?;
+
+    // TODO: We cannot poll, as there is no support in Rust currently, so we
+    // sleep. Change this when read_poll_timeout() is implemented in Rust.
+    kernel::time::delay::fsleep(time::Delta::from_millis(100));
+
+    if regs::L2_READY_LO.read(dev, iomem)? != 1 {
+        dev_err!(dev, "Failed to power on the GPU\n");
+        return Err(EIO);
+    }
+
+    Ok(())
+}
diff --git a/drivers/gpu/drm/tyr/regs.rs b/drivers/gpu/drm/tyr/regs.rs
new file mode 100644
index 000000000000..f46933aaa221
--- /dev/null
+++ b/drivers/gpu/drm/tyr/regs.rs
@@ -0,0 +1,108 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+
+// We don't expect that all the registers and fields will be used, even in the
+// future.
+//
+// Nevertheless, it is useful to have most of them defined, like the C driver
+// does.
+#![allow(dead_code)]
+
+use kernel::bits::bit_u32;
+use kernel::device::Bound;
+use kernel::device::Device;
+use kernel::devres::Devres;
+use kernel::prelude::*;
+
+use crate::driver::IoMem;
+
+/// Represents a register in the Register Set
+///
+/// TODO: Replace this with the Nova `register!()` macro when it is available.
+/// In particular, this will automatically give us 64bit register reads and
+/// writes.
+pub(crate) struct Register<const OFFSET: usize>;
+
+impl<const OFFSET: usize> Register<OFFSET> {
+    #[inline]
+    pub(crate) fn read(&self, dev: &Device<Bound>, iomem: &Devres<IoMem>) -> Result<u32> {
+        let value = (*iomem).access(dev)?.read32(OFFSET);
+        Ok(value)
+    }
+
+    #[inline]
+    pub(crate) fn write(&self, dev: &Device<Bound>, iomem: &Devres<IoMem>, value: u32) -> Result {
+        (*iomem).access(dev)?.write32(value, OFFSET);
+        Ok(())
+    }
+}
+
+pub(crate) const GPU_ID: Register<0x0> = Register;
+pub(crate) const GPU_L2_FEATURES: Register<0x4> = Register;
+pub(crate) const GPU_CORE_FEATURES: Register<0x8> = Register;
+pub(crate) const GPU_CSF_ID: Register<0x1c> = Register;
+pub(crate) const GPU_REVID: Register<0x280> = Register;
+pub(crate) const GPU_TILER_FEATURES: Register<0xc> = Register;
+pub(crate) const GPU_MEM_FEATURES: Register<0x10> = Register;
+pub(crate) const GPU_MMU_FEATURES: Register<0x14> = Register;
+pub(crate) const GPU_AS_PRESENT: Register<0x18> = Register;
+pub(crate) const GPU_IRQ_RAWSTAT: Register<0x20> = Register;
+
+pub(crate) const GPU_IRQ_RAWSTAT_FAULT: u32 = bit_u32(0);
+pub(crate) const GPU_IRQ_RAWSTAT_PROTECTED_FAULT: u32 = bit_u32(1);
+pub(crate) const GPU_IRQ_RAWSTAT_RESET_COMPLETED: u32 = bit_u32(8);
+pub(crate) const GPU_IRQ_RAWSTAT_POWER_CHANGED_SINGLE: u32 = bit_u32(9);
+pub(crate) const GPU_IRQ_RAWSTAT_POWER_CHANGED_ALL: u32 = bit_u32(10);
+pub(crate) const GPU_IRQ_RAWSTAT_CLEAN_CACHES_COMPLETED: u32 = bit_u32(17);
+pub(crate) const GPU_IRQ_RAWSTAT_DOORBELL_STATUS: u32 = bit_u32(18);
+pub(crate) const GPU_IRQ_RAWSTAT_MCU_STATUS: u32 = bit_u32(19);
+
+pub(crate) const GPU_IRQ_CLEAR: Register<0x24> = Register;
+pub(crate) const GPU_IRQ_MASK: Register<0x28> = Register;
+pub(crate) const GPU_IRQ_STAT: Register<0x2c> = Register;
+pub(crate) const GPU_CMD: Register<0x30> = Register;
+pub(crate) const GPU_CMD_SOFT_RESET: u32 = 1 | (1 << 8);
+pub(crate) const GPU_CMD_HARD_RESET: u32 = 1 | (2 << 8);
+pub(crate) const GPU_THREAD_FEATURES: Register<0xac> = Register;
+pub(crate) const GPU_THREAD_MAX_THREADS: Register<0xa0> = Register;
+pub(crate) const GPU_THREAD_MAX_WORKGROUP_SIZE: Register<0xa4> = Register;
+pub(crate) const GPU_THREAD_MAX_BARRIER_SIZE: Register<0xa8> = Register;
+pub(crate) const GPU_TEXTURE_FEATURES0: Register<0xb0> = Register;
+pub(crate) const GPU_SHADER_PRESENT_LO: Register<0x100> = Register;
+pub(crate) const GPU_SHADER_PRESENT_HI: Register<0x104> = Register;
+pub(crate) const GPU_TILER_PRESENT_LO: Register<0x110> = Register;
+pub(crate) const GPU_TILER_PRESENT_HI: Register<0x114> = Register;
+pub(crate) const GPU_L2_PRESENT_LO: Register<0x120> = Register;
+pub(crate) const GPU_L2_PRESENT_HI: Register<0x124> = Register;
+pub(crate) const L2_READY_LO: Register<0x160> = Register;
+pub(crate) const L2_READY_HI: Register<0x164> = Register;
+pub(crate) const L2_PWRON_LO: Register<0x1a0> = Register;
+pub(crate) const L2_PWRON_HI: Register<0x1a4> = Register;
+pub(crate) const L2_PWRTRANS_LO: Register<0x220> = Register;
+pub(crate) const L2_PWRTRANS_HI: Register<0x204> = Register;
+pub(crate) const L2_PWRACTIVE_LO: Register<0x260> = Register;
+pub(crate) const L2_PWRACTIVE_HI: Register<0x264> = Register;
+
+pub(crate) const MCU_CONTROL: Register<0x700> = Register;
+pub(crate) const MCU_CONTROL_ENABLE: u32 = 1;
+pub(crate) const MCU_CONTROL_AUTO: u32 = 2;
+pub(crate) const MCU_CONTROL_DISABLE: u32 = 0;
+
+pub(crate) const MCU_STATUS: Register<0x704> = Register;
+pub(crate) const MCU_STATUS_DISABLED: u32 = 0;
+pub(crate) const MCU_STATUS_ENABLED: u32 = 1;
+pub(crate) const MCU_STATUS_HALT: u32 = 2;
+pub(crate) const MCU_STATUS_FATAL: u32 = 3;
+
+pub(crate) const GPU_COHERENCY_FEATURES: Register<0x300> = Register;
+
+pub(crate) const JOB_IRQ_RAWSTAT: Register<0x1000> = Register;
+pub(crate) const JOB_IRQ_CLEAR: Register<0x1004> = Register;
+pub(crate) const JOB_IRQ_MASK: Register<0x1008> = Register;
+pub(crate) const JOB_IRQ_STAT: Register<0x100c> = Register;
+
+pub(crate) const JOB_IRQ_GLOBAL_IF: u32 = bit_u32(31);
+
+pub(crate) const MMU_IRQ_RAWSTAT: Register<0x2000> = Register;
+pub(crate) const MMU_IRQ_CLEAR: Register<0x2004> = Register;
+pub(crate) const MMU_IRQ_MASK: Register<0x2008> = Register;
+pub(crate) const MMU_IRQ_STAT: Register<0x200c> = Register;
diff --git a/drivers/gpu/drm/tyr/tyr.rs b/drivers/gpu/drm/tyr/tyr.rs
new file mode 100644
index 000000000000..861d1db43072
--- /dev/null
+++ b/drivers/gpu/drm/tyr/tyr.rs
@@ -0,0 +1,22 @@
+// SPDX-License-Identifier: GPL-2.0 or MIT
+
+//! Arm Mali Tyr DRM driver.
+//!
+//! The name "Tyr" is inspired by Norse mythology, reflecting Arm's tradition of
+//! naming their GPUs after Nordic mythological figures and places.
+
+use crate::driver::TyrDriver;
+
+mod driver;
+mod file;
+mod gem;
+mod gpu;
+mod regs;
+
+kernel::module_platform_driver! {
+    type: TyrDriver,
+    name: "tyr",
+    authors: ["The Tyr driver authors"],
+    description: "Arm Mali Tyr DRM driver",
+    license: "Dual MIT/GPL",
+}
diff --git a/rust/uapi/uapi_helper.h b/rust/uapi/uapi_helper.h
index 1409441359f5..d4a239cf2a64 100644
--- a/rust/uapi/uapi_helper.h
+++ b/rust/uapi/uapi_helper.h
@@ -9,6 +9,7 @@
 #include <uapi/asm-generic/ioctl.h>
 #include <uapi/drm/drm.h>
 #include <uapi/drm/nova_drm.h>
+#include <uapi/drm/panthor_drm.h>
 #include <uapi/linux/mdio.h>
 #include <uapi/linux/mii.h>
 #include <uapi/linux/ethtool.h>

From 60d2b7899142d56654ba254a454eb282c1d181ca Mon Sep 17 00:00:00 2001
From: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Date: Wed, 3 Sep 2025 16:21:20 -0700
Subject: [PATCH 1203/1455] drm/xe/guc: Add SLPC power profile interface

GuC has an interface to set a power profile for the SLPC algorithm.
Base mode is default and ensures a balanced performance, power_saving
mode has conservative up/down thresholds and is suitable for use with
apps that typically need to be power efficient. This will result in
lower GT frequencies, thus consuming lower power.

Selected power profile will be displayed in this format:

$ cat power_profile

  [base]    power_saving

$ echo power_saving > power_profile
$ cat power_profile

  base    [power_saving]

v2: Address review comments (Rodrigo)

Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250903232120.390190-1-vinay.belgaumkar@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/abi/guc_actions_slpc_abi.h |  5 ++
 drivers/gpu/drm/xe/xe_gt_freq.c               | 26 +++++++
 drivers/gpu/drm/xe/xe_guc_pc.c                | 67 +++++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_pc.h                |  2 +
 drivers/gpu/drm/xe/xe_guc_pc_types.h          |  2 +
 5 files changed, 102 insertions(+)

diff --git a/drivers/gpu/drm/xe/abi/guc_actions_slpc_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_slpc_abi.h
index b28c8fa061f7..ce5c59517528 100644
--- a/drivers/gpu/drm/xe/abi/guc_actions_slpc_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_actions_slpc_abi.h
@@ -210,6 +210,11 @@ struct slpc_shared_data {
 	u8 reserved_mode_definition[4096];
 } __packed;
 
+enum slpc_power_profile {
+	SLPC_POWER_PROFILE_BASE = 0x0,
+	SLPC_POWER_PROFILE_POWER_SAVING = 0x1
+};
+
 /**
  * DOC: SLPC H2G MESSAGE FORMAT
  *
diff --git a/drivers/gpu/drm/xe/xe_gt_freq.c b/drivers/gpu/drm/xe/xe_gt_freq.c
index 60d9354e7dbf..781e4890fb26 100644
--- a/drivers/gpu/drm/xe/xe_gt_freq.c
+++ b/drivers/gpu/drm/xe/xe_gt_freq.c
@@ -227,6 +227,31 @@ static ssize_t max_freq_store(struct kobject *kobj,
 }
 static struct kobj_attribute attr_max_freq = __ATTR_RW(max_freq);
 
+static ssize_t power_profile_show(struct kobject *kobj,
+				  struct kobj_attribute *attr,
+				  char *buff)
+{
+	struct device *dev = kobj_to_dev(kobj);
+
+	xe_guc_pc_get_power_profile(dev_to_pc(dev), buff);
+
+	return strlen(buff);
+}
+
+static ssize_t power_profile_store(struct kobject *kobj,
+				   struct kobj_attribute *attr,
+				   const char *buff, size_t count)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct xe_guc_pc *pc = dev_to_pc(dev);
+	int err;
+
+	err = xe_guc_pc_set_power_profile(pc, buff);
+
+	return err ?: count;
+}
+static struct kobj_attribute attr_power_profile = __ATTR_RW(power_profile);
+
 static const struct attribute *freq_attrs[] = {
 	&attr_act_freq.attr,
 	&attr_cur_freq.attr,
@@ -236,6 +261,7 @@ static const struct attribute *freq_attrs[] = {
 	&attr_rpn_freq.attr,
 	&attr_min_freq.attr,
 	&attr_max_freq.attr,
+	&attr_power_profile.attr,
 	NULL
 };
 
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c
index 88557e86d637..68a5bf8e3946 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.c
+++ b/drivers/gpu/drm/xe/xe_guc_pc.c
@@ -79,6 +79,11 @@
  * Xe driver enables SLPC with all of its defaults features and frequency
  * selection, which varies per platform.
  *
+ * Power profiles add another level of control to SLPC. When power saving
+ * profile is chosen, SLPC will use conservative thresholds to ramp frequency,
+ * thus saving power. Base profile is default and ensures balanced performance
+ * for any workload.
+ *
  * Render-C States:
  * ================
  *
@@ -1171,6 +1176,61 @@ static int pc_action_set_strategy(struct xe_guc_pc *pc, u32 val)
 	return ret;
 }
 
+static const char *power_profile_to_string(struct xe_guc_pc *pc)
+{
+	switch (pc->power_profile) {
+	case SLPC_POWER_PROFILE_BASE:
+		return "base";
+	case SLPC_POWER_PROFILE_POWER_SAVING:
+		return "power_saving";
+	default:
+		return "invalid";
+	}
+}
+
+void xe_guc_pc_get_power_profile(struct xe_guc_pc *pc, char *profile)
+{
+	switch (pc->power_profile) {
+	case SLPC_POWER_PROFILE_BASE:
+		sprintf(profile, "[%s]    %s\n", "base", "power_saving");
+		break;
+	case SLPC_POWER_PROFILE_POWER_SAVING:
+		sprintf(profile, "%s    [%s]\n", "base", "power_saving");
+		break;
+	default:
+		sprintf(profile, "invalid");
+	}
+}
+
+int xe_guc_pc_set_power_profile(struct xe_guc_pc *pc, const char *buf)
+{
+	int ret = 0;
+	u32 val;
+
+	if (strncmp("base", buf, strlen("base")) == 0)
+		val = SLPC_POWER_PROFILE_BASE;
+	else if (strncmp("power_saving", buf, strlen("power_saving")) == 0)
+		val = SLPC_POWER_PROFILE_POWER_SAVING;
+	else
+		return -EINVAL;
+
+	guard(mutex)(&pc->freq_lock);
+	xe_pm_runtime_get(pc_to_xe(pc));
+
+	ret = pc_action_set_param(pc,
+				  SLPC_PARAM_POWER_PROFILE,
+				  val);
+	if (ret)
+		xe_gt_err_once(pc_to_gt(pc), "Failed to set power profile to %d: %pe\n",
+			       val, ERR_PTR(ret));
+	else
+		pc->power_profile = val;
+
+	xe_pm_runtime_put(pc_to_xe(pc));
+
+	return ret;
+}
+
 /**
  * xe_guc_pc_start - Start GuC's Power Conservation component
  * @pc: Xe_GuC_PC instance
@@ -1249,6 +1309,11 @@ int xe_guc_pc_start(struct xe_guc_pc *pc)
 	/* Enable SLPC Optimized Strategy for compute */
 	ret = pc_action_set_strategy(pc, SLPC_OPTIMIZED_STRATEGY_COMPUTE);
 
+	/* Set cached value of power_profile */
+	ret = xe_guc_pc_set_power_profile(pc, power_profile_to_string(pc));
+	if (unlikely(ret))
+		xe_gt_err(gt, "Failed to set SLPC power profile: %pe\n", ERR_PTR(ret));
+
 out:
 	xe_force_wake_put(gt_to_fw(gt), fw_ref);
 	return ret;
@@ -1327,6 +1392,8 @@ int xe_guc_pc_init(struct xe_guc_pc *pc)
 
 	pc->bo = bo;
 
+	pc->power_profile = SLPC_POWER_PROFILE_BASE;
+
 	return devm_add_action_or_reset(xe->drm.dev, xe_guc_pc_fini_hw, pc);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.h b/drivers/gpu/drm/xe/xe_guc_pc.h
index 52ecdd5ddbff..0e31396f103c 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.h
+++ b/drivers/gpu/drm/xe/xe_guc_pc.h
@@ -31,6 +31,8 @@ int xe_guc_pc_get_min_freq(struct xe_guc_pc *pc, u32 *freq);
 int xe_guc_pc_set_min_freq(struct xe_guc_pc *pc, u32 freq);
 int xe_guc_pc_get_max_freq(struct xe_guc_pc *pc, u32 *freq);
 int xe_guc_pc_set_max_freq(struct xe_guc_pc *pc, u32 freq);
+int xe_guc_pc_set_power_profile(struct xe_guc_pc *pc, const char *buf);
+void xe_guc_pc_get_power_profile(struct xe_guc_pc *pc, char *profile);
 
 enum xe_gt_idle_state xe_guc_pc_c_status(struct xe_guc_pc *pc);
 u64 xe_guc_pc_rc6_residency(struct xe_guc_pc *pc);
diff --git a/drivers/gpu/drm/xe/xe_guc_pc_types.h b/drivers/gpu/drm/xe/xe_guc_pc_types.h
index c02053948a57..5e4ea53fbee6 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_pc_types.h
@@ -37,6 +37,8 @@ struct xe_guc_pc {
 	struct mutex freq_lock;
 	/** @freq_ready: Only handle freq changes, if they are really ready */
 	bool freq_ready;
+	/** @power_profile: Base or power_saving profile */
+	u32 power_profile;
 };
 
 #endif	/* _XE_GUC_PC_TYPES_H_ */

From 010629e00d8494d0ca3893fe802ee0cbec92c021 Mon Sep 17 00:00:00 2001
From: Varun Gupta <varun.gupta@intel.com>
Date: Thu, 11 Sep 2025 16:47:12 +0530
Subject: [PATCH 1204/1455] drm/xe: Fix driver reference in FLR comment

Rectify the reference of i915 to Xe in a comment.

v2: Cosmetic changes. (Karthik)
v3: Rephrased the commit message. (Karthik)

Signed-off-by: Varun Gupta <varun.gupta@intel.com>
Reviewed-by: Karthik Poosa <karthik.poosa@intel.com>
Link: https://lore.kernel.org/r/20250911111712.811524-1-varun.gupta@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index c7bba535b145..a4d12ee7d575 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -532,7 +532,7 @@ static bool xe_driver_flr_disabled(struct xe_device *xe)
  * re-init and saving/restoring (or re-populating) the wiped memory. Since we
  * perform the FLR as the very last action before releasing access to the HW
  * during the driver release flow, we don't attempt recovery at all, because
- * if/when a new instance of i915 is bound to the device it will do a full
+ * if/when a new instance of Xe is bound to the device it will do a full
  * re-init anyway.
  */
 static void __xe_driver_flr(struct xe_device *xe)

From 95c1cfa306087142989bff34ea0e05dcd95ddc58 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Thu, 11 Sep 2025 00:24:39 +0200
Subject: [PATCH 1205/1455] drm/xe/pf: Drop rounddown_pow_of_two fair LMEM
 limitation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This effectively reverts commit 4c3fe5eae46b ("drm/xe/pf: Limit
fair VF LMEM provisioning") since we don't need it any more after
non-contig VRAM allocations were fixed. This allows larger LMEM
auto-provisioning for VFs, so instead:

 [ ] GT0: PF: LMEM available(14096M) fair(1 x 8192M)
 [ ] GT0: PF: VF1 provisioned with 8589934592 (8.00 GiB) LMEM
or
 [ ] GT0: PF: LMEM available(14096M) fair(2 x 4096M)
 [ ] GT0: PF: VF1..VF2 provisioned with 4294967296 (4.00 GiB) LMEM

we may get:

 [ ] GT0: PF: LMEM available(14096M) fair(1 x 14096M)
 [ ] GT0: PF: VF1 provisioned with 14780727296 (13.8 GiB) LMEM
and
 [ ] GT0: PF: LMEM available(14096M) fair(2 x 7048M)
 [ ] GT0: PF: VF1..VF2 provisioned with 7390363648 (6.88 GiB) LMEM

Fixes: 1e32ffbc9dc8 ("drm/xe/sriov: support non-contig VRAM provisioning")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Link: https://lore.kernel.org/r/20250910222439.32869-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
index 6a9a752fe24a..6344b5205c08 100644
--- a/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
+++ b/drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c
@@ -1629,7 +1629,6 @@ static u64 pf_estimate_fair_lmem(struct xe_gt *gt, unsigned int num_vfs)
 	u64 fair;
 
 	fair = div_u64(available, num_vfs);
-	fair = rounddown_pow_of_two(fair);	/* XXX: ttm_vram_mgr & drm_buddy limitation */
 	fair = ALIGN_DOWN(fair, alignment);
 #ifdef MAX_FAIR_LMEM
 	fair = min_t(u64, MAX_FAIR_LMEM, fair);

From 88434448438e4302e272b2a2b810b42e05ea024b Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Fri, 5 Sep 2025 16:56:33 -0700
Subject: [PATCH 1206/1455] drm/xe/guc: Set RCS/CCS yield policy

All recent platforms (including all the ones officially supported by the
Xe driver) do not allow concurrent execution of RCS and CCS workloads
from different address spaces, with the HW blocking the context switch
when it detects such a scenario.
The DUAL_QUEUE flag helps with this, by causing the GuC to not submit a
context it knows will not be able to execute. This, however, causes a new
problem: if RCS and CCS queues have pending workloads from different
address spaces, the GuC needs to choose from which of the 2 queues to
pick the next workload to execute. By default, the GuC prioritizes RCS
submissions over CCS ones, which can lead to CCS workloads being
significantly (or completely) starved of execution time.
The driver can tune this by setting a dedicated scheduling policy KLV;
this KLV allows the driver to specify a quantum (in ms) and a ratio
(percentage value between 0 and 100), and the GuC will prioritize the CCS
for that percentage of each quantum.
Given that we want to guarantee enough RCS throughput to avoid missing
frames, we set the yield policy to 20% of each 80ms interval.

v2: updated quantum and ratio, improved comment, use xe_guc_submit_disable
in gt_sanitize

Fixes: d9a1ae0d17bd ("drm/xe/guc: Enable WA_DUAL_QUEUE for newer platforms")
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Tested-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250905235632.3333247-2-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/xe/abi/guc_actions_abi.h |  1 +
 drivers/gpu/drm/xe/abi/guc_klvs_abi.h    | 25 +++++++++
 drivers/gpu/drm/xe/xe_gt.c               |  2 +-
 drivers/gpu/drm/xe/xe_guc.c              |  6 +--
 drivers/gpu/drm/xe/xe_guc_submit.c       | 66 ++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_guc_submit.h       |  2 +
 6 files changed, 97 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
index d8cf68a0516d..1baa969aaa7c 100644
--- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
@@ -117,6 +117,7 @@ enum xe_guc_action {
 	XE_GUC_ACTION_ENTER_S_STATE = 0x501,
 	XE_GUC_ACTION_EXIT_S_STATE = 0x502,
 	XE_GUC_ACTION_GLOBAL_SCHED_POLICY_CHANGE = 0x506,
+	XE_GUC_ACTION_UPDATE_SCHEDULING_POLICIES_KLV = 0x509,
 	XE_GUC_ACTION_SCHED_CONTEXT = 0x1000,
 	XE_GUC_ACTION_SCHED_CONTEXT_MODE_SET = 0x1001,
 	XE_GUC_ACTION_SCHED_CONTEXT_MODE_DONE = 0x1002,
diff --git a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
index 0e78351c6ef5..265a135e7061 100644
--- a/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_klvs_abi.h
@@ -17,6 +17,7 @@
  *  | 0 | 31:16 | **KEY** - KLV key identifier                                 |
  *  |   |       |   - `GuC Self Config KLVs`_                                  |
  *  |   |       |   - `GuC Opt In Feature KLVs`_                               |
+ *  |   |       |   - `GuC Scheduling Policies KLVs`_                          |
  *  |   |       |   - `GuC VGT Policy KLVs`_                                   |
  *  |   |       |   - `GuC VF Configuration KLVs`_                             |
  *  |   |       |                                                              |
@@ -152,6 +153,30 @@ enum  {
 #define GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH_KEY 0x4003
 #define GUC_KLV_OPT_IN_FEATURE_DYNAMIC_INHIBIT_CONTEXT_SWITCH_LEN 0u
 
+/**
+ * DOC: GuC Scheduling Policies KLVs
+ *
+ * `GuC KLV`_ keys available for use with UPDATE_SCHEDULING_POLICIES_KLV.
+ *
+ * _`GUC_KLV_SCHEDULING_POLICIES_RENDER_COMPUTE_YIELD` : 0x1001
+ *      Some platforms do not allow concurrent execution of RCS and CCS
+ *      workloads from different address spaces. By default, the GuC prioritizes
+ *      RCS submissions over CCS ones, which can lead to CCS workloads being
+ *      significantly (or completely) starved of execution time. This KLV allows
+ *      the driver to specify a quantum (in ms) and a ratio (percentage value
+ *      between 0 and 100), and the GuC will prioritize the CCS for that
+ *      percentage of each quantum. For example, specifying 100ms and 30% will
+ *      make the GuC prioritize the CCS for 30ms of every 100ms.
+ *      Note that this does not necessarly mean that RCS and CCS engines will
+ *      only be active for their percentage of the quantum, as the restriction
+ *      only kicks in if both classes are fully busy with non-compatible address
+ *      spaces; i.e., if one engine is idle or running the same address space,
+ *      a pending job on the other engine will still be submitted to the HW no
+ *      matter what the ratio is
+ */
+#define GUC_KLV_SCHEDULING_POLICIES_RENDER_COMPUTE_YIELD_KEY	0x1001
+#define GUC_KLV_SCHEDULING_POLICIES_RENDER_COMPUTE_YIELD_LEN	2u
+
 /**
  * DOC: GuC VGT Policy KLVs
  *
diff --git a/drivers/gpu/drm/xe/xe_gt.c b/drivers/gpu/drm/xe/xe_gt.c
index 34505a6d93ed..3e0ad7e5b5df 100644
--- a/drivers/gpu/drm/xe/xe_gt.c
+++ b/drivers/gpu/drm/xe/xe_gt.c
@@ -98,7 +98,7 @@ void xe_gt_sanitize(struct xe_gt *gt)
 	 * FIXME: if xe_uc_sanitize is called here, on TGL driver will not
 	 * reload
 	 */
-	gt->uc.guc.submission_state.enabled = false;
+	xe_guc_submit_disable(&gt->uc.guc);
 }
 
 static void xe_gt_enable_host_l2_vram(struct xe_gt *gt)
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index e56c9c5c8845..4eb06ffc09bc 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -879,9 +879,7 @@ int xe_guc_post_load_init(struct xe_guc *guc)
 			return ret;
 	}
 
-	guc->submission_state.enabled = true;
-
-	return 0;
+	return xe_guc_submit_enable(guc);
 }
 
 int xe_guc_reset(struct xe_guc *guc)
@@ -1593,7 +1591,7 @@ void xe_guc_sanitize(struct xe_guc *guc)
 {
 	xe_uc_fw_sanitize(&guc->fw);
 	xe_guc_ct_disable(&guc->ct);
-	guc->submission_state.enabled = false;
+	xe_guc_submit_disable(guc);
 }
 
 int xe_guc_reset_prepare(struct xe_guc *guc)
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 5ab242be50dd..a465594b61dc 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -32,6 +32,7 @@
 #include "xe_guc_ct.h"
 #include "xe_guc_exec_queue_types.h"
 #include "xe_guc_id_mgr.h"
+#include "xe_guc_klv_helpers.h"
 #include "xe_guc_submit_types.h"
 #include "xe_hw_engine.h"
 #include "xe_hw_fence.h"
@@ -316,6 +317,71 @@ int xe_guc_submit_init(struct xe_guc *guc, unsigned int num_ids)
 	return drmm_add_action_or_reset(&xe->drm, guc_submit_fini, guc);
 }
 
+/*
+ * Given that we want to guarantee enough RCS throughput to avoid missing
+ * frames, we set the yield policy to 20% of each 80ms interval.
+ */
+#define RC_YIELD_DURATION	80	/* in ms */
+#define RC_YIELD_RATIO		20	/* in percent */
+static u32 *emit_render_compute_yield_klv(u32 *emit)
+{
+	*emit++ = PREP_GUC_KLV_TAG(SCHEDULING_POLICIES_RENDER_COMPUTE_YIELD);
+	*emit++ = RC_YIELD_DURATION;
+	*emit++ = RC_YIELD_RATIO;
+
+	return emit;
+}
+
+#define SCHEDULING_POLICY_MAX_DWORDS 16
+static int guc_init_global_schedule_policy(struct xe_guc *guc)
+{
+	u32 data[SCHEDULING_POLICY_MAX_DWORDS];
+	u32 *emit = data;
+	u32 count = 0;
+	int ret;
+
+	if (GUC_SUBMIT_VER(guc) < MAKE_GUC_VER(1, 1, 0))
+		return 0;
+
+	*emit++ = XE_GUC_ACTION_UPDATE_SCHEDULING_POLICIES_KLV;
+
+	if (CCS_MASK(guc_to_gt(guc)))
+		emit = emit_render_compute_yield_klv(emit);
+
+	count = emit - data;
+	if (count > 1) {
+		xe_assert(guc_to_xe(guc), count <= SCHEDULING_POLICY_MAX_DWORDS);
+
+		ret = xe_guc_ct_send_block(&guc->ct, data, count);
+		if (ret < 0) {
+			xe_gt_err(guc_to_gt(guc),
+				  "failed to enable GuC sheduling policies: %pe\n",
+				  ERR_PTR(ret));
+			return ret;
+		}
+	}
+
+	return 0;
+}
+
+int xe_guc_submit_enable(struct xe_guc *guc)
+{
+	int ret;
+
+	ret = guc_init_global_schedule_policy(guc);
+	if (ret)
+		return ret;
+
+	guc->submission_state.enabled = true;
+
+	return 0;
+}
+
+void xe_guc_submit_disable(struct xe_guc *guc)
+{
+	guc->submission_state.enabled = false;
+}
+
 static void __release_guc_id(struct xe_guc *guc, struct xe_exec_queue *q, u32 xa_count)
 {
 	int i;
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.h b/drivers/gpu/drm/xe/xe_guc_submit.h
index 9e6f19b03e65..78c3f07e31a0 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.h
+++ b/drivers/gpu/drm/xe/xe_guc_submit.h
@@ -13,6 +13,8 @@ struct xe_exec_queue;
 struct xe_guc;
 
 int xe_guc_submit_init(struct xe_guc *guc, unsigned int num_ids);
+int xe_guc_submit_enable(struct xe_guc *guc);
+void xe_guc_submit_disable(struct xe_guc *guc);
 
 int xe_guc_submit_reset_prepare(struct xe_guc *guc);
 void xe_guc_submit_reset_wait(struct xe_guc *guc);

From ba391a102ec11ab7f5b249e46dcf61ef836a57e7 Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Tue, 9 Sep 2025 15:36:22 -0700
Subject: [PATCH 1207/1455] drm/i915/guc: Include the GuC registers in the
 error state

If GuC hangs, the GuC logs might not contain enough information to
understand exactly why the hang occurred. In this case, we need to
look at the GuC HW state to try to understand where the GuC is stuck. It
is therefore useful to include the GuC HW state in the error capture.

The list of registers that are part of the GuC HW state can change based
on platform, but it is the same for all platforms from TGL to MTL so we
only need to support one version for i915.

v2: revised list
v3: remove confusing comment, use sizeof(u32) instead of 4 (John)

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250909223621.3782625-2-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c  |   8 ++
 drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h |   1 +
 drivers/gpu/drm/i915/i915_gpu_error.c      | 102 +++++++++++++++++++++
 drivers/gpu/drm/i915/i915_gpu_error.h      |   1 +
 4 files changed, 112 insertions(+)

diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
index e7ccfa520df3..384d1400134d 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
@@ -46,6 +46,14 @@ static void guc_prepare_xfer(struct intel_gt *gt)
 		/* allows for 5us (in 10ns units) before GT can go to RC6 */
 		intel_uncore_write(uncore, GUC_ARAT_C6DIS, 0x1FF);
 	}
+
+	/*
+	 * Starting from IP 12.50 we need to enable the mirroring of GuC
+	 * internal state to debug registers. This is always enabled on previous
+	 * IPs.
+	 */
+	if (GRAPHICS_VER_FULL(uncore->i915) >= IP_VER(12, 50))
+		intel_uncore_rmw(uncore, GUC_SHIM_CONTROL2, 0, GUC_ENABLE_DEBUG_REG);
 }
 
 static int guc_xfer_rsa_mmio(struct intel_uc_fw *guc_fw,
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h
index 3fd798837502..f73dab527547 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h
@@ -96,6 +96,7 @@
 #define   GUC_GEN10_SHIM_WC_ENABLE		(1<<21)
 
 #define GUC_SHIM_CONTROL2		_MMIO(0xc068)
+#define   GUC_ENABLE_DEBUG_REG		(1<<11)
 #define   GUC_IS_PRIVILEGED		(1<<29)
 #define   GSC_LOADS_HUC			(1<<30)
 
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 4f785cdbd155..01dd77de6d14 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -685,6 +685,74 @@ static void err_print_guc_ctb(struct drm_i915_error_state_buf *m,
 		   ctb->head, ctb->tail, ctb->desc_offset, ctb->cmds_offset, ctb->size);
 }
 
+/* This list includes registers that are useful in debugging GuC hangs. */
+const struct {
+	u32 start;
+	u32 count;
+} guc_hw_reg_state[] = {
+	{ 0xc0b0, 2 },
+	{ 0xc000, 65 },
+	{ 0xc140, 1 },
+	{ 0xc180, 16 },
+	{ 0xc1dc, 10 },
+	{ 0xc300, 79 },
+	{ 0xc4b4, 47 },
+	{ 0xc574, 1 },
+	{ 0xc57c, 1 },
+	{ 0xc584, 11 },
+	{ 0xc5c0, 8 },
+	{ 0xc5e4, 1 },
+	{ 0xc5ec, 103 },
+	{ 0xc7c0, 1 },
+	{ 0xc0b0, 2 }
+};
+
+static u32 print_range_line(struct drm_i915_error_state_buf *m, u32 start, u32 *dump, u32 count)
+{
+	if (count >= 8) {
+		err_printf(m, "[0x%04x] 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x 0x%08x\n",
+			   start, dump[0], dump[1], dump[2], dump[3],
+			   dump[4], dump[5], dump[6], dump[7]);
+		return 8;
+	} else if (count >= 4) {
+		err_printf(m, "[0x%04x] 0x%08x 0x%08x 0x%08x 0x%08x\n",
+			   start, dump[0], dump[1], dump[2], dump[3]);
+		return 4;
+	} else if (count >= 2) {
+		err_printf(m, "[0x%04x] 0x%08x 0x%08x\n", start, dump[0], dump[1]);
+		return 2;
+	}
+
+	err_printf(m, "[0x%04x] 0x%08x\n", start, dump[0]);
+	return 1;
+}
+
+static void err_print_guc_hw_state(struct drm_i915_error_state_buf *m, u32 *hw_state)
+{
+	u32 total = 0;
+	int i;
+
+	if (!hw_state)
+		return;
+
+	err_printf(m, "GuC Register State:\n");
+
+	for (i = 0; i < ARRAY_SIZE(guc_hw_reg_state); i++) {
+		u32 entry = 0;
+
+		while (entry < guc_hw_reg_state[i].count) {
+			u32 start = guc_hw_reg_state[i].start + entry * sizeof(u32);
+			u32 count = guc_hw_reg_state[i].count - entry;
+			u32 *values = hw_state + total + entry;
+
+			entry += print_range_line(m, start, values, count);
+		}
+
+		GEM_BUG_ON(entry != guc_hw_reg_state[i].count);
+		total += entry;
+	}
+}
+
 static void err_print_uc(struct drm_i915_error_state_buf *m,
 			 const struct intel_uc_coredump *error_uc)
 {
@@ -693,6 +761,7 @@ static void err_print_uc(struct drm_i915_error_state_buf *m,
 	intel_uc_fw_dump(&error_uc->guc_fw, &p);
 	intel_uc_fw_dump(&error_uc->huc_fw, &p);
 	err_printf(m, "GuC timestamp: 0x%08x\n", error_uc->guc.timestamp);
+	err_print_guc_hw_state(m, error_uc->guc.hw_state);
 	intel_gpu_error_print_vma(m, NULL, error_uc->guc.vma_log);
 	err_printf(m, "GuC CTB fence: %d\n", error_uc->guc.last_fence);
 	err_print_guc_ctb(m, "Send", error_uc->guc.ctb + 0);
@@ -1025,6 +1094,7 @@ static void cleanup_uc(struct intel_uc_coredump *uc)
 	kfree(uc->huc_fw.file_wanted.path);
 	i915_vma_coredump_free(uc->guc.vma_log);
 	i915_vma_coredump_free(uc->guc.vma_ctb);
+	kfree(uc->guc.hw_state);
 
 	kfree(uc);
 }
@@ -1721,6 +1791,37 @@ static void gt_record_guc_ctb(struct intel_ctb_coredump *saved,
 	saved->cmds_offset = ((void *)ctb->cmds) - blob_ptr;
 }
 
+static u32 read_guc_state_reg(struct intel_uncore *uncore, int range, int count)
+{
+	GEM_BUG_ON(range >= ARRAY_SIZE(guc_hw_reg_state));
+	GEM_BUG_ON(count >= guc_hw_reg_state[range].count);
+
+	return intel_uncore_read(uncore,
+				 _MMIO(guc_hw_reg_state[range].start + count * sizeof(u32)));
+}
+
+static void gt_record_guc_hw_state(struct intel_uncore *uncore,
+				   struct intel_uc_coredump *error_uc)
+{
+	u32 *hw_state;
+	u32 count = 0;
+	int i, j;
+
+	for (i = 0; i < ARRAY_SIZE(guc_hw_reg_state); i++)
+		count += guc_hw_reg_state[i].count;
+
+	hw_state = kcalloc(count, sizeof(u32), ALLOW_FAIL);
+	if (!hw_state)
+		return;
+
+	count = 0;
+	for (i = 0; i < ARRAY_SIZE(guc_hw_reg_state); i++)
+		for (j = 0; j < guc_hw_reg_state[i].count; j++)
+			hw_state[count++] = read_guc_state_reg(uncore, i, j);
+
+	error_uc->guc.hw_state = hw_state;
+}
+
 static struct intel_uc_coredump *
 gt_record_uc(struct intel_gt_coredump *gt,
 	     struct i915_vma_compress *compress)
@@ -1755,6 +1856,7 @@ gt_record_uc(struct intel_gt_coredump *gt,
 			  uc->guc.ct.ctbs.send.desc, (struct intel_guc *)&uc->guc);
 	gt_record_guc_ctb(error_uc->guc.ctb + 1, &uc->guc.ct.ctbs.recv,
 			  uc->guc.ct.ctbs.send.desc, (struct intel_guc *)&uc->guc);
+	gt_record_guc_hw_state(gt->_gt->uncore, error_uc);
 
 	return error_uc;
 }
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h
index 182324979278..91b3df621a49 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.h
+++ b/drivers/gpu/drm/i915/i915_gpu_error.h
@@ -177,6 +177,7 @@ struct intel_gt_coredump {
 			struct intel_ctb_coredump ctb[2];
 			struct i915_vma_coredump *vma_ctb;
 			struct i915_vma_coredump *vma_log;
+			u32 *hw_state;
 			u32 timestamp;
 			u16 last_fence;
 			bool is_guc_capture;

From 3d536277449e4de4609d4f0b8302ae7fd00b04b2 Mon Sep 17 00:00:00 2001
From: Benno Lossin <lossin@kernel.org>
Date: Mon, 18 Aug 2025 15:18:44 +0200
Subject: [PATCH 1208/1455] rust: pin-init: examples: error: use `Error` in `fn
 main()`

When running this example with no cargo features enabled, the compiler
warns on 1.89:

  error: struct `Error` is never constructed
    --> examples/error.rs:11:12
     |
  11 | pub struct Error;
     |            ^^^^^
     |
     = note: `-D dead-code` implied by `-D warnings`
     = help: to override `-D warnings` add `#[allow(dead_code)]`

Thus use the error in the main function to avoid this warning.

Signed-off-by: Benno Lossin <lossin@kernel.org>
---
 rust/pin-init/examples/error.rs | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/rust/pin-init/examples/error.rs b/rust/pin-init/examples/error.rs
index e0cc258746ce..8f4e135eb8ba 100644
--- a/rust/pin-init/examples/error.rs
+++ b/rust/pin-init/examples/error.rs
@@ -24,4 +24,6 @@ impl From<AllocError> for Error {
 }
 
 #[allow(dead_code)]
-fn main() {}
+fn main() {
+    let _ = Error;
+}

From 62a9c709612c1cf0d5684022499f34c56c99b662 Mon Sep 17 00:00:00 2001
From: Benno Lossin <lossin@kernel.org>
Date: Tue, 19 Aug 2025 09:22:00 +0200
Subject: [PATCH 1209/1455] rust: pin-init: README: add information banner on
 the rename to `pin-init`

Signed-off-by: Benno Lossin <lossin@kernel.org>
---
 rust/pin-init/README.md | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/rust/pin-init/README.md b/rust/pin-init/README.md
index a4c01a8d78b2..723e275445d4 100644
--- a/rust/pin-init/README.md
+++ b/rust/pin-init/README.md
@@ -6,6 +6,18 @@
 ![GitHub Workflow Status](https://img.shields.io/github/actions/workflow/status/Rust-for-Linux/pin-init/test.yml)
 # `pin-init`
 
+> [!NOTE]
+> 
+> This crate was originally named [`pinned-init`], but the migration to
+> `pin-init` is not yet complete. The `legcay` branch contains the current
+> version of the `pinned-init` crate & the `main` branch already incorporates
+> the rename to `pin-init`.
+>
+> There are still some changes needed on the kernel side before the migration
+> can be completed.
+
+[`pinned-init`]: https://crates.io/crates/pinned-init
+
 <!-- cargo-rdme start -->
 
 Library to safely and fallibly initialize pinned `struct`s using in-place constructors.

From d49c56368c0cc98cdf02ae21dd275eba92f1c333 Mon Sep 17 00:00:00 2001
From: Benno Lossin <lossin@kernel.org>
Date: Fri, 5 Sep 2025 19:12:06 +0200
Subject: [PATCH 1210/1455] rust: pin-init: rename `project` -> `project_this`
 in doctest

The next commit makes the `#[pin_data]` attribute generate a `project`
function that would collide with any existing ones.

Signed-off-by: Benno Lossin <lossin@kernel.org>
---
 rust/pin-init/src/lib.rs | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/rust/pin-init/src/lib.rs b/rust/pin-init/src/lib.rs
index 62e013a5cc20..2d0d9fd12524 100644
--- a/rust/pin-init/src/lib.rs
+++ b/rust/pin-init/src/lib.rs
@@ -994,7 +994,7 @@ macro_rules! try_init {
 /// }
 ///
 /// impl<T> Foo<T> {
-///     fn project(self: Pin<&mut Self>) -> Pin<&mut T> {
+///     fn project_this(self: Pin<&mut Self>) -> Pin<&mut T> {
 ///         assert_pinned!(Foo<T>, elem, T, inline);
 ///
 ///         // SAFETY: The field is structurally pinned.

From 619db96daf942dad974c6c8157ed06d52f7bb969 Mon Sep 17 00:00:00 2001
From: Benno Lossin <lossin@kernel.org>
Date: Fri, 5 Sep 2025 19:12:07 +0200
Subject: [PATCH 1211/1455] rust: pin-init: add pin projections to
 `#[pin_data]`

Make the `#[pin_data]` macro generate a `*Projection` struct that holds
either `Pin<&mut Field>` or `&mut Field` for every field of the original
struct. Which version is chosen depends on weather there is a `#[pin]`
or not respectively. Access to this projected version is enabled through
generating `fn project(self: Pin<&mut Self>) -> SelfProjection<'_>`.

[ Adapt workqueue to use the new projection instead of its own, custom
  one - Benno ]

Reviewed-by: Gary Guo <gary@garyguo.net>
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Benno Lossin <lossin@kernel.org>
---
 rust/kernel/workqueue.rs    |  9 +-----
 rust/pin-init/src/macros.rs | 61 +++++++++++++++++++++++++++++++++++++
 2 files changed, 62 insertions(+), 8 deletions(-)

diff --git a/rust/kernel/workqueue.rs b/rust/kernel/workqueue.rs
index b9343d5bc00f..706e833e9702 100644
--- a/rust/kernel/workqueue.rs
+++ b/rust/kernel/workqueue.rs
@@ -356,18 +356,11 @@ struct ClosureWork<T> {
     func: Option<T>,
 }
 
-impl<T> ClosureWork<T> {
-    fn project(self: Pin<&mut Self>) -> &mut Option<T> {
-        // SAFETY: The `func` field is not structurally pinned.
-        unsafe { &mut self.get_unchecked_mut().func }
-    }
-}
-
 impl<T: FnOnce()> WorkItem for ClosureWork<T> {
     type Pointer = Pin<KBox<Self>>;
 
     fn run(mut this: Pin<KBox<Self>>) {
-        if let Some(func) = this.as_mut().project().take() {
+        if let Some(func) = this.as_mut().project().func.take() {
             (func)()
         }
     }
diff --git a/rust/pin-init/src/macros.rs b/rust/pin-init/src/macros.rs
index 9ced630737b8..668dcee9b7af 100644
--- a/rust/pin-init/src/macros.rs
+++ b/rust/pin-init/src/macros.rs
@@ -831,6 +831,17 @@ macro_rules! __pin_data {
             $($fields)*
         }
 
+        $crate::__pin_data!(make_pin_projections:
+            @vis($vis),
+            @name($name),
+            @impl_generics($($impl_generics)*),
+            @ty_generics($($ty_generics)*),
+            @decl_generics($($decl_generics)*),
+            @where($($whr)*),
+            @pinned($($pinned)*),
+            @not_pinned($($not_pinned)*),
+        );
+
         // We put the rest into this const item, because it then will not be accessible to anything
         // outside.
         const _: () = {
@@ -980,6 +991,56 @@ macro_rules! __pin_data {
             stringify!($($rest)*),
         );
     };
+    (make_pin_projections:
+        @vis($vis:vis),
+        @name($name:ident),
+        @impl_generics($($impl_generics:tt)*),
+        @ty_generics($($ty_generics:tt)*),
+        @decl_generics($($decl_generics:tt)*),
+        @where($($whr:tt)*),
+        @pinned($($(#[$($p_attr:tt)*])* $pvis:vis $p_field:ident : $p_type:ty),* $(,)?),
+        @not_pinned($($(#[$($attr:tt)*])* $fvis:vis $field:ident : $type:ty),* $(,)?),
+    ) => {
+        $crate::macros::paste! {
+            #[doc(hidden)]
+            $vis struct [< $name Projection >] <'__pin, $($decl_generics)*> {
+                $($(#[$($p_attr)*])* $pvis $p_field : ::core::pin::Pin<&'__pin mut $p_type>,)*
+                $($(#[$($attr)*])* $fvis $field : &'__pin mut $type,)*
+                ___pin_phantom_data: ::core::marker::PhantomData<&'__pin mut ()>,
+            }
+
+            impl<$($impl_generics)*> $name<$($ty_generics)*>
+            where $($whr)*
+            {
+                /// Pin-projects all fields of `Self`.
+                ///
+                /// These fields are structurally pinned:
+                $(#[doc = ::core::concat!(" - `", ::core::stringify!($p_field), "`")])*
+                ///
+                /// These fields are **not** structurally pinned:
+                $(#[doc = ::core::concat!(" - `", ::core::stringify!($field), "`")])*
+                #[inline]
+                $vis fn project<'__pin>(
+                    self: ::core::pin::Pin<&'__pin mut Self>,
+                ) -> [< $name Projection >] <'__pin, $($ty_generics)*> {
+                    // SAFETY: we only give access to `&mut` for fields not structurally pinned.
+                    let this = unsafe { ::core::pin::Pin::get_unchecked_mut(self) };
+                    [< $name Projection >] {
+                        $(
+                            // SAFETY: `$p_field` is structurally pinned.
+                            $(#[$($p_attr)*])*
+                            $p_field : unsafe { ::core::pin::Pin::new_unchecked(&mut this.$p_field) },
+                        )*
+                        $(
+                            $(#[$($attr)*])*
+                            $field : &mut this.$field,
+                        )*
+                        ___pin_phantom_data: ::core::marker::PhantomData,
+                    }
+                }
+            }
+        }
+    };
     (make_pin_data:
         @pin_data($pin_data:ident),
         @impl_generics($($impl_generics:tt)*),

From 1fa516794fdd27b96cee77f8b12ac916b8b6a9a7 Mon Sep 17 00:00:00 2001
From: Benno Lossin <lossin@kernel.org>
Date: Fri, 5 Sep 2025 16:05:31 +0200
Subject: [PATCH 1212/1455] rust: pin-init: add code blocks to
 `[try_][pin_]init!` macros

Allow writing `_: { /* any number of statements */ }` in initializers to
run arbitrary code during initialization.

    try_init!(MyStruct {
        _: {
            if check_something() {
                return Err(MyError);
            }
        },
        foo: Foo::new(val),
        _: {
            println!("successfully initialized `MyStruct`");
        },
    })

Tested-by: Alexandre Courbot <acourbot@nvidia.com>
Reviewed-by: Gary Guo <gary@garyguo.net>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Tested-by: Danilo Krummrich <dakr@kernel.org>
Reviewed-by: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: Benno Lossin <lossin@kernel.org>
---
 rust/pin-init/src/lib.rs    |  2 ++
 rust/pin-init/src/macros.rs | 29 +++++++++++++++++++++++++++++
 2 files changed, 31 insertions(+)

diff --git a/rust/pin-init/src/lib.rs b/rust/pin-init/src/lib.rs
index 2d0d9fd12524..dd553212836e 100644
--- a/rust/pin-init/src/lib.rs
+++ b/rust/pin-init/src/lib.rs
@@ -740,6 +740,8 @@ macro_rules! stack_try_pin_init {
 /// As already mentioned in the examples above, inside of `pin_init!` a `struct` initializer with
 /// the following modifications is expected:
 /// - Fields that you want to initialize in-place have to use `<-` instead of `:`.
+/// - You can use `_: { /* run any user-code here */ },` anywhere where you can place fields in
+///   order to run arbitrary code.
 /// - In front of the initializer you can write `&this in` to have access to a [`NonNull<Self>`]
 ///   pointer named `this` inside of the initializer.
 /// - Using struct update syntax one can place `..Zeroable::init_zeroed()` at the very end of the
diff --git a/rust/pin-init/src/macros.rs b/rust/pin-init/src/macros.rs
index 668dcee9b7af..c3462080b7b6 100644
--- a/rust/pin-init/src/macros.rs
+++ b/rust/pin-init/src/macros.rs
@@ -1263,6 +1263,21 @@ macro_rules! __init_internal {
         // have been initialized. Therefore we can now dismiss the guards by forgetting them.
         $(::core::mem::forget($guards);)*
     };
+    (init_slot($($use_data:ident)?):
+        @data($data:ident),
+        @slot($slot:ident),
+        @guards($($guards:ident,)*),
+        // arbitrary code block
+        @munch_fields(_: { $($code:tt)* }, $($rest:tt)*),
+    ) => {
+        { $($code)* }
+        $crate::__init_internal!(init_slot($($use_data)?):
+            @data($data),
+            @slot($slot),
+            @guards($($guards,)*),
+            @munch_fields($($rest)*),
+        );
+    };
     (init_slot($use_data:ident): // `use_data` is present, so we use the `data` to init fields.
         @data($data:ident),
         @slot($slot:ident),
@@ -1358,6 +1373,20 @@ macro_rules! __init_internal {
             );
         }
     };
+    (make_initializer:
+        @slot($slot:ident),
+        @type_name($t:path),
+        @munch_fields(_: { $($code:tt)* }, $($rest:tt)*),
+        @acc($($acc:tt)*),
+    ) => {
+        // code blocks are ignored for the initializer check
+        $crate::__init_internal!(make_initializer:
+            @slot($slot),
+            @type_name($t),
+            @munch_fields($($rest)*),
+            @acc($($acc)*),
+        );
+    };
     (make_initializer:
         @slot($slot:ident),
         @type_name($t:path),

From 42415d163e5df6db799c7de6262d707e402c2c7e Mon Sep 17 00:00:00 2001
From: Benno Lossin <lossin@kernel.org>
Date: Fri, 5 Sep 2025 16:00:46 +0200
Subject: [PATCH 1213/1455] rust: pin-init: add references to previously
 initialized fields

After initializing a field in an initializer macro, create a variable
holding a reference that points at that field. The type is either
`Pin<&mut T>` or `&mut T` depending on the field's structural pinning
kind.

[ Applied fixes to devres and rust_driver_pci sample - Benno]
Reviewed-by: Danilo Krummrich <dakr@kernel.org>
Signed-off-by: Benno Lossin <lossin@kernel.org>
---
 rust/kernel/devres.rs           |   6 +-
 rust/pin-init/src/macros.rs     | 149 ++++++++++++++++++++++++--------
 samples/rust/rust_driver_pci.rs |   2 +-
 3 files changed, 118 insertions(+), 39 deletions(-)

diff --git a/rust/kernel/devres.rs b/rust/kernel/devres.rs
index da18091143a6..91dbf3f4b166 100644
--- a/rust/kernel/devres.rs
+++ b/rust/kernel/devres.rs
@@ -134,11 +134,9 @@ impl<T: Send> Devres<T> {
         T: 'a,
         Error: From<E>,
     {
-        let callback = Self::devres_callback;
-
         try_pin_init!(&this in Self {
             dev: dev.into(),
-            callback,
+            callback: Self::devres_callback,
             // INVARIANT: `inner` is properly initialized.
             inner <- {
                 // SAFETY: `this` is a valid pointer to uninitialized memory.
@@ -151,7 +149,7 @@ impl<T: Send> Devres<T> {
                 //    properly initialized, because we require `dev` (i.e. the *bound* device) to
                 //    live at least as long as the returned `impl PinInit<Self, Error>`.
                 to_result(unsafe {
-                    bindings::devm_add_action(dev.as_raw(), Some(callback), inner.cast())
+                    bindings::devm_add_action(dev.as_raw(), Some(*callback), inner.cast())
                 })?;
 
                 Opaque::pin_init(try_pin_init!(Inner {
diff --git a/rust/pin-init/src/macros.rs b/rust/pin-init/src/macros.rs
index c3462080b7b6..d6acf2cd291e 100644
--- a/rust/pin-init/src/macros.rs
+++ b/rust/pin-init/src/macros.rs
@@ -1049,38 +1049,56 @@ macro_rules! __pin_data {
         @pinned($($(#[$($p_attr:tt)*])* $pvis:vis $p_field:ident : $p_type:ty),* $(,)?),
         @not_pinned($($(#[$($attr:tt)*])* $fvis:vis $field:ident : $type:ty),* $(,)?),
     ) => {
-        // For every field, we create a projection function according to its projection type. If a
-        // field is structurally pinned, then it must be initialized via `PinInit`, if it is not
-        // structurally pinned, then it can be initialized via `Init`.
-        //
-        // The functions are `unsafe` to prevent accidentally calling them.
-        #[allow(dead_code)]
-        #[expect(clippy::missing_safety_doc)]
-        impl<$($impl_generics)*> $pin_data<$($ty_generics)*>
-        where $($whr)*
-        {
-            $(
-                $(#[$($p_attr)*])*
-                $pvis unsafe fn $p_field<E>(
-                    self,
-                    slot: *mut $p_type,
-                    init: impl $crate::PinInit<$p_type, E>,
-                ) -> ::core::result::Result<(), E> {
-                    // SAFETY: TODO.
-                    unsafe { $crate::PinInit::__pinned_init(init, slot) }
-                }
-            )*
-            $(
-                $(#[$($attr)*])*
-                $fvis unsafe fn $field<E>(
-                    self,
-                    slot: *mut $type,
-                    init: impl $crate::Init<$type, E>,
-                ) -> ::core::result::Result<(), E> {
-                    // SAFETY: TODO.
-                    unsafe { $crate::Init::__init(init, slot) }
-                }
-            )*
+        $crate::macros::paste! {
+            // For every field, we create a projection function according to its projection type. If a
+            // field is structurally pinned, then it must be initialized via `PinInit`, if it is not
+            // structurally pinned, then it can be initialized via `Init`.
+            //
+            // The functions are `unsafe` to prevent accidentally calling them.
+            #[allow(dead_code)]
+            #[expect(clippy::missing_safety_doc)]
+            impl<$($impl_generics)*> $pin_data<$($ty_generics)*>
+            where $($whr)*
+            {
+                $(
+                    $(#[$($p_attr)*])*
+                    $pvis unsafe fn $p_field<E>(
+                        self,
+                        slot: *mut $p_type,
+                        init: impl $crate::PinInit<$p_type, E>,
+                    ) -> ::core::result::Result<(), E> {
+                        // SAFETY: TODO.
+                        unsafe { $crate::PinInit::__pinned_init(init, slot) }
+                    }
+
+                    $(#[$($p_attr)*])*
+                    $pvis unsafe fn [<__project_ $p_field>]<'__slot>(
+                        self,
+                        slot: &'__slot mut $p_type,
+                    ) -> ::core::pin::Pin<&'__slot mut $p_type> {
+                        ::core::pin::Pin::new_unchecked(slot)
+                    }
+                )*
+                $(
+                    $(#[$($attr)*])*
+                    $fvis unsafe fn $field<E>(
+                        self,
+                        slot: *mut $type,
+                        init: impl $crate::Init<$type, E>,
+                    ) -> ::core::result::Result<(), E> {
+                        // SAFETY: TODO.
+                        unsafe { $crate::Init::__init(init, slot) }
+                    }
+
+                    $(#[$($attr)*])*
+                    $fvis unsafe fn [<__project_ $field>]<'__slot>(
+                        self,
+                        slot: &'__slot mut $type,
+                    ) -> &'__slot mut $type {
+                        slot
+                    }
+                )*
+            }
         }
     };
 }
@@ -1292,6 +1310,13 @@ macro_rules! __init_internal {
         // return when an error/panic occurs.
         // We also use the `data` to require the correct trait (`Init` or `PinInit`) for `$field`.
         unsafe { $data.$field(::core::ptr::addr_of_mut!((*$slot).$field), init)? };
+        // SAFETY:
+        // - the project function does the correct field projection,
+        // - the field has been initialized,
+        // - the reference is only valid until the end of the initializer.
+        #[allow(unused_variables)]
+        let $field = $crate::macros::paste!(unsafe { $data.[< __project_ $field >](&mut (*$slot).$field) });
+
         // Create the drop guard:
         //
         // We rely on macro hygiene to make it impossible for users to access this local variable.
@@ -1323,6 +1348,14 @@ macro_rules! __init_internal {
         // SAFETY: `slot` is valid, because we are inside of an initializer closure, we
         // return when an error/panic occurs.
         unsafe { $crate::Init::__init(init, ::core::ptr::addr_of_mut!((*$slot).$field))? };
+
+        // SAFETY:
+        // - the field is not structurally pinned, since the line above must compile,
+        // - the field has been initialized,
+        // - the reference is only valid until the end of the initializer.
+        #[allow(unused_variables)]
+        let $field = unsafe { &mut (*$slot).$field };
+
         // Create the drop guard:
         //
         // We rely on macro hygiene to make it impossible for users to access this local variable.
@@ -1341,7 +1374,7 @@ macro_rules! __init_internal {
             );
         }
     };
-    (init_slot($($use_data:ident)?):
+    (init_slot(): // No `use_data`, so all fields are not structurally pinned
         @data($data:ident),
         @slot($slot:ident),
         @guards($($guards:ident,)*),
@@ -1355,6 +1388,15 @@ macro_rules! __init_internal {
             // SAFETY: The memory at `slot` is uninitialized.
             unsafe { ::core::ptr::write(::core::ptr::addr_of_mut!((*$slot).$field), $field) };
         }
+
+        #[allow(unused_variables)]
+        // SAFETY:
+        // - the field is not structurally pinned, since no `use_data` was required to create this
+        //   initializer,
+        // - the field has been initialized,
+        // - the reference is only valid until the end of the initializer.
+        let $field = unsafe { &mut (*$slot).$field };
+
         // Create the drop guard:
         //
         // We rely on macro hygiene to make it impossible for users to access this local variable.
@@ -1365,7 +1407,46 @@ macro_rules! __init_internal {
                 $crate::__internal::DropGuard::new(::core::ptr::addr_of_mut!((*$slot).$field))
             };
 
-            $crate::__init_internal!(init_slot($($use_data)?):
+            $crate::__init_internal!(init_slot():
+                @data($data),
+                @slot($slot),
+                @guards([< __ $field _guard >], $($guards,)*),
+                @munch_fields($($rest)*),
+            );
+        }
+    };
+    (init_slot($use_data:ident):
+        @data($data:ident),
+        @slot($slot:ident),
+        @guards($($guards:ident,)*),
+        // Init by-value.
+        @munch_fields($field:ident $(: $val:expr)?, $($rest:tt)*),
+    ) => {
+        {
+            $(let $field = $val;)?
+            // Initialize the field.
+            //
+            // SAFETY: The memory at `slot` is uninitialized.
+            unsafe { ::core::ptr::write(::core::ptr::addr_of_mut!((*$slot).$field), $field) };
+        }
+        // SAFETY:
+        // - the project function does the correct field projection,
+        // - the field has been initialized,
+        // - the reference is only valid until the end of the initializer.
+        #[allow(unused_variables)]
+        let $field = $crate::macros::paste!(unsafe { $data.[< __project_ $field >](&mut (*$slot).$field) });
+
+        // Create the drop guard:
+        //
+        // We rely on macro hygiene to make it impossible for users to access this local variable.
+        // We use `paste!` to create new hygiene for `$field`.
+        $crate::macros::paste! {
+            // SAFETY: We forget the guard later when initialization has succeeded.
+            let [< __ $field _guard >] = unsafe {
+                $crate::__internal::DropGuard::new(::core::ptr::addr_of_mut!((*$slot).$field))
+            };
+
+            $crate::__init_internal!(init_slot($use_data):
                 @data($data),
                 @slot($slot),
                 @guards([< __ $field _guard >], $($guards,)*),
diff --git a/samples/rust/rust_driver_pci.rs b/samples/rust/rust_driver_pci.rs
index 606946ff4d7f..1ac0b06fa3b3 100644
--- a/samples/rust/rust_driver_pci.rs
+++ b/samples/rust/rust_driver_pci.rs
@@ -78,8 +78,8 @@ impl pci::Driver for SampleDriver {
 
         let drvdata = KBox::pin_init(
             try_pin_init!(Self {
-                pdev: pdev.into(),
                 bar <- pdev.iomap_region_sized::<{ Regs::END }>(0, c_str!("rust_driver_pci")),
+                pdev: pdev.into(),
                 index: *info,
             }),
             GFP_KERNEL,

From a394f12a4d6d6ec3c72d62107282c6022cd96333 Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 11 Sep 2025 12:17:37 +0300
Subject: [PATCH 1214/1455] drm/i915: split out i915_ptr_util.h

Move pointer related utilities from i915_utils.h to a separate new
i915_ptr_util.h header. Clean up related includes.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/3cd06aa2483e68f19401292e9d4c28bf2977fce5.1757582214.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_context_types.h |  1 -
 drivers/gpu/drm/i915/gt/intel_timeline.h      |  1 +
 drivers/gpu/drm/i915/i915_ptr_util.h          | 66 +++++++++++++++++++
 drivers/gpu/drm/i915/i915_request.h           |  5 +-
 drivers/gpu/drm/i915/i915_utils.h             | 57 ----------------
 drivers/gpu/drm/i915/i915_vma.h               |  6 +-
 6 files changed, 73 insertions(+), 63 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_ptr_util.h

diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h
index 98c7f6052069..10070ee4d74c 100644
--- a/drivers/gpu/drm/i915/gt/intel_context_types.h
+++ b/drivers/gpu/drm/i915/gt/intel_context_types.h
@@ -14,7 +14,6 @@
 
 #include "i915_active_types.h"
 #include "i915_sw_fence.h"
-#include "i915_utils.h"
 #include "intel_engine_types.h"
 #include "intel_sseu.h"
 #include "intel_wakeref.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.h b/drivers/gpu/drm/i915/gt/intel_timeline.h
index 57308c4d664a..5f8c8f79714e 100644
--- a/drivers/gpu/drm/i915/gt/intel_timeline.h
+++ b/drivers/gpu/drm/i915/gt/intel_timeline.h
@@ -10,6 +10,7 @@
 
 #include "i915_active.h"
 #include "i915_syncmap.h"
+#include "i915_utils.h"
 #include "intel_timeline_types.h"
 
 struct drm_printer;
diff --git a/drivers/gpu/drm/i915/i915_ptr_util.h b/drivers/gpu/drm/i915/i915_ptr_util.h
new file mode 100644
index 000000000000..9f8931d7d99b
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_ptr_util.h
@@ -0,0 +1,66 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __I915_PTR_UTIL_H__
+#define __I915_PTR_UTIL_H__
+
+#include <linux/types.h>
+
+#define ptr_mask_bits(ptr, n) ({					\
+	unsigned long __v = (unsigned long)(ptr);			\
+	(typeof(ptr))(__v & -BIT(n));					\
+})
+
+#define ptr_unmask_bits(ptr, n) ((unsigned long)(ptr) & (BIT(n) - 1))
+
+#define ptr_unpack_bits(ptr, bits, n) ({				\
+	unsigned long __v = (unsigned long)(ptr);			\
+	*(bits) = __v & (BIT(n) - 1);					\
+	(typeof(ptr))(__v & -BIT(n));					\
+})
+
+#define ptr_pack_bits(ptr, bits, n) ({					\
+	unsigned long __bits = (bits);					\
+	GEM_BUG_ON(__bits & -BIT(n));					\
+	((typeof(ptr))((unsigned long)(ptr) | __bits));			\
+})
+
+#define ptr_dec(ptr) ({							\
+	unsigned long __v = (unsigned long)(ptr);			\
+	(typeof(ptr))(__v - 1);						\
+})
+
+#define ptr_inc(ptr) ({							\
+	unsigned long __v = (unsigned long)(ptr);			\
+	(typeof(ptr))(__v + 1);						\
+})
+
+#define page_mask_bits(ptr) ptr_mask_bits(ptr, PAGE_SHIFT)
+#define page_unmask_bits(ptr) ptr_unmask_bits(ptr, PAGE_SHIFT)
+#define page_pack_bits(ptr, bits) ptr_pack_bits(ptr, bits, PAGE_SHIFT)
+#define page_unpack_bits(ptr, bits) ptr_unpack_bits(ptr, bits, PAGE_SHIFT)
+
+static __always_inline ptrdiff_t ptrdiff(const void *a, const void *b)
+{
+	return a - b;
+}
+
+#define u64_to_ptr(T, x) ({						\
+	typecheck(u64, x);						\
+	(T *)(uintptr_t)(x);						\
+})
+
+/*
+ * container_of_user: Extract the superclass from a pointer to a member.
+ *
+ * Exactly like container_of() with the exception that it plays nicely
+ * with sparse for __user @ptr.
+ */
+#define container_of_user(ptr, type, member) ({				\
+	void __user *__mptr = (void __user *)(ptr);			\
+	BUILD_BUG_ON_MSG(!__same_type(*(ptr), typeof_member(type, member)) && \
+			 !__same_type(*(ptr), void),			\
+			 "pointer type mismatch in container_of()");	\
+	((type __user *)(__mptr - offsetof(type, member))); })
+
+#endif /* __I915_PTR_UTIL_H__ */
diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h
index 5f7e8138ec14..b09135301f39 100644
--- a/drivers/gpu/drm/i915/i915_request.h
+++ b/drivers/gpu/drm/i915/i915_request.h
@@ -31,19 +31,20 @@
 #include <linux/llist.h>
 #include <linux/lockdep.h>
 
+#include <uapi/drm/i915_drm.h>
+
 #include "gem/i915_gem_context_types.h"
 #include "gt/intel_context_types.h"
 #include "gt/intel_engine_types.h"
 #include "gt/intel_timeline_types.h"
 
 #include "i915_gem.h"
+#include "i915_ptr_util.h"
 #include "i915_scheduler.h"
 #include "i915_selftest.h"
 #include "i915_sw_fence.h"
 #include "i915_vma_resource.h"
 
-#include <uapi/drm/i915_drm.h>
-
 struct drm_file;
 struct drm_i915_gem_object;
 struct drm_printer;
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index eb4d43c40009..dff02a944a57 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -67,64 +67,12 @@ bool i915_error_injected(void);
 		drm_err(&(i915)->drm, fmt, ##__VA_ARGS__); \
 })
 
-#define ptr_mask_bits(ptr, n) ({					\
-	unsigned long __v = (unsigned long)(ptr);			\
-	(typeof(ptr))(__v & -BIT(n));					\
-})
-
-#define ptr_unmask_bits(ptr, n) ((unsigned long)(ptr) & (BIT(n) - 1))
-
-#define ptr_unpack_bits(ptr, bits, n) ({				\
-	unsigned long __v = (unsigned long)(ptr);			\
-	*(bits) = __v & (BIT(n) - 1);					\
-	(typeof(ptr))(__v & -BIT(n));					\
-})
-
-#define ptr_pack_bits(ptr, bits, n) ({					\
-	unsigned long __bits = (bits);					\
-	GEM_BUG_ON(__bits & -BIT(n));					\
-	((typeof(ptr))((unsigned long)(ptr) | __bits));			\
-})
-
-#define ptr_dec(ptr) ({							\
-	unsigned long __v = (unsigned long)(ptr);			\
-	(typeof(ptr))(__v - 1);						\
-})
-
-#define ptr_inc(ptr) ({							\
-	unsigned long __v = (unsigned long)(ptr);			\
-	(typeof(ptr))(__v + 1);						\
-})
-
-#define page_mask_bits(ptr) ptr_mask_bits(ptr, PAGE_SHIFT)
-#define page_unmask_bits(ptr) ptr_unmask_bits(ptr, PAGE_SHIFT)
-#define page_pack_bits(ptr, bits) ptr_pack_bits(ptr, bits, PAGE_SHIFT)
-#define page_unpack_bits(ptr, bits) ptr_unpack_bits(ptr, bits, PAGE_SHIFT)
-
 #define fetch_and_zero(ptr) ({						\
 	typeof(*ptr) __T = *(ptr);					\
 	*(ptr) = (typeof(*ptr))0;					\
 	__T;								\
 })
 
-static __always_inline ptrdiff_t ptrdiff(const void *a, const void *b)
-{
-	return a - b;
-}
-
-/*
- * container_of_user: Extract the superclass from a pointer to a member.
- *
- * Exactly like container_of() with the exception that it plays nicely
- * with sparse for __user @ptr.
- */
-#define container_of_user(ptr, type, member) ({				\
-	void __user *__mptr = (void __user *)(ptr);			\
-	BUILD_BUG_ON_MSG(!__same_type(*(ptr), typeof_member(type, member)) && \
-			 !__same_type(*(ptr), void),			\
-			 "pointer type mismatch in container_of()");	\
-	((type __user *)(__mptr - offsetof(type, member))); })
-
 /*
  * check_user_mbz: Check that a user value exists and is zero
  *
@@ -143,11 +91,6 @@ static __always_inline ptrdiff_t ptrdiff(const void *a, const void *b)
 	get_user(mbz__, (U)) ? -EFAULT : mbz__ ? -EINVAL : 0;		\
 })
 
-#define u64_to_ptr(T, x) ({						\
-	typecheck(u64, x);						\
-	(T *)(uintptr_t)(x);						\
-})
-
 #define __mask_next_bit(mask) ({					\
 	int __idx = ffs(mask) - 1;					\
 	mask &= ~BIT(__idx);						\
diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h
index 0f9eee6d18d2..8054047840aa 100644
--- a/drivers/gpu/drm/i915/i915_vma.h
+++ b/drivers/gpu/drm/i915/i915_vma.h
@@ -30,12 +30,12 @@
 
 #include <drm/drm_mm.h>
 
-#include "gt/intel_ggtt_fencing.h"
 #include "gem/i915_gem_object.h"
-
-#include "i915_gem_gtt.h"
+#include "gt/intel_ggtt_fencing.h"
 
 #include "i915_active.h"
+#include "i915_gem_gtt.h"
+#include "i915_ptr_util.h"
 #include "i915_request.h"
 #include "i915_vma_resource.h"
 #include "i915_vma_types.h"

From df7d085b585016fdda440719a58ee3ac73c5db7c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 11 Sep 2025 12:17:38 +0300
Subject: [PATCH 1215/1455] drm/i915: split out i915_timer_util.[ch]

Move timer related utilities from i915_utils.[ch] to separate new files
i915_timer_util.[ch]. Clean up related includes.

Note: Arguably none of this should exist in i915 in the first place. At
least isolate it better.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/0a83d9489626121dcefcd4c1a05317399b5708f3.1757582214.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 drivers/gpu/drm/i915/Makefile                 |  1 +
 .../drm/i915/gt/intel_execlists_submission.c  |  4 ++-
 drivers/gpu/drm/i915/gt/sysfs_engines.c       |  1 +
 drivers/gpu/drm/i915/i915_timer_util.c        | 36 +++++++++++++++++++
 drivers/gpu/drm/i915/i915_timer_util.h        | 23 ++++++++++++
 drivers/gpu/drm/i915/i915_utils.c             | 30 ----------------
 drivers/gpu/drm/i915/i915_utils.h             | 14 --------
 7 files changed, 64 insertions(+), 45 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_timer_util.c
 create mode 100644 drivers/gpu/drm/i915/i915_timer_util.h

diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
index db150a0c33ce..e58c0c158b3a 100644
--- a/drivers/gpu/drm/i915/Makefile
+++ b/drivers/gpu/drm/i915/Makefile
@@ -32,6 +32,7 @@ i915-y += \
 	i915_scatterlist.o \
 	i915_switcheroo.o \
 	i915_sysfs.o \
+	i915_timer_util.o \
 	i915_utils.o \
 	intel_clock_gating.o \
 	intel_cpu_info.o \
diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
index 03baa7fa0a27..52c8fddedfce 100644
--- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
@@ -106,14 +106,16 @@
  * preemption, but just sampling the new tail pointer).
  *
  */
+
 #include <linux/interrupt.h>
 #include <linux/string_helpers.h>
 
+#include "gen8_engine_cs.h"
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_timer_util.h"
 #include "i915_trace.h"
 #include "i915_vgpu.h"
-#include "gen8_engine_cs.h"
 #include "intel_breadcrumbs.h"
 #include "intel_context.h"
 #include "intel_engine_heartbeat.h"
diff --git a/drivers/gpu/drm/i915/gt/sysfs_engines.c b/drivers/gpu/drm/i915/gt/sysfs_engines.c
index aab2759067d2..4a81bc396b21 100644
--- a/drivers/gpu/drm/i915/gt/sysfs_engines.c
+++ b/drivers/gpu/drm/i915/gt/sysfs_engines.c
@@ -7,6 +7,7 @@
 #include <linux/sysfs.h>
 
 #include "i915_drv.h"
+#include "i915_timer_util.h"
 #include "intel_engine.h"
 #include "intel_engine_heartbeat.h"
 #include "sysfs_engines.h"
diff --git a/drivers/gpu/drm/i915/i915_timer_util.c b/drivers/gpu/drm/i915/i915_timer_util.c
new file mode 100644
index 000000000000..ee4cfd8b3c07
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_timer_util.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: MIT
+/* Copyright © 2025 Intel Corporation */
+
+#include <linux/jiffies.h>
+
+#include "i915_timer_util.h"
+
+void cancel_timer(struct timer_list *t)
+{
+	if (!timer_active(t))
+		return;
+
+	timer_delete(t);
+	WRITE_ONCE(t->expires, 0);
+}
+
+void set_timer_ms(struct timer_list *t, unsigned long timeout)
+{
+	if (!timeout) {
+		cancel_timer(t);
+		return;
+	}
+
+	timeout = msecs_to_jiffies(timeout);
+
+	/*
+	 * Paranoia to make sure the compiler computes the timeout before
+	 * loading 'jiffies' as jiffies is volatile and may be updated in
+	 * the background by a timer tick. All to reduce the complexity
+	 * of the addition and reduce the risk of losing a jiffy.
+	 */
+	barrier();
+
+	/* Keep t->expires = 0 reserved to indicate a canceled timer. */
+	mod_timer(t, jiffies + timeout ?: 1);
+}
diff --git a/drivers/gpu/drm/i915/i915_timer_util.h b/drivers/gpu/drm/i915/i915_timer_util.h
new file mode 100644
index 000000000000..f35ad730820c
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_timer_util.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __I915_TIMER_UTIL_H__
+#define __I915_TIMER_UTIL_H__
+
+#include <linux/timer.h>
+#include <asm/rwonce.h>
+
+void cancel_timer(struct timer_list *t);
+void set_timer_ms(struct timer_list *t, unsigned long timeout);
+
+static inline bool timer_active(const struct timer_list *t)
+{
+	return READ_ONCE(t->expires);
+}
+
+static inline bool timer_expired(const struct timer_list *t)
+{
+	return timer_active(t) && !timer_pending(t);
+}
+
+#endif /* __I915_TIMER_UTIL_H__ */
diff --git a/drivers/gpu/drm/i915/i915_utils.c b/drivers/gpu/drm/i915/i915_utils.c
index b60c28fbd207..49f7ed413132 100644
--- a/drivers/gpu/drm/i915/i915_utils.c
+++ b/drivers/gpu/drm/i915/i915_utils.c
@@ -47,36 +47,6 @@ bool i915_error_injected(void)
 
 #endif
 
-void cancel_timer(struct timer_list *t)
-{
-	if (!timer_active(t))
-		return;
-
-	timer_delete(t);
-	WRITE_ONCE(t->expires, 0);
-}
-
-void set_timer_ms(struct timer_list *t, unsigned long timeout)
-{
-	if (!timeout) {
-		cancel_timer(t);
-		return;
-	}
-
-	timeout = msecs_to_jiffies(timeout);
-
-	/*
-	 * Paranoia to make sure the compiler computes the timeout before
-	 * loading 'jiffies' as jiffies is volatile and may be updated in
-	 * the background by a timer tick. All to reduce the complexity
-	 * of the addition and reduce the risk of losing a jiffy.
-	 */
-	barrier();
-
-	/* Keep t->expires = 0 reserved to indicate a canceled timer. */
-	mod_timer(t, jiffies + timeout ?: 1);
-}
-
 bool i915_vtd_active(struct drm_i915_private *i915)
 {
 	if (device_iommu_mapped(i915->drm.dev))
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index dff02a944a57..6278a74d08c2 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -38,7 +38,6 @@
 #endif
 
 struct drm_i915_private;
-struct timer_list;
 
 #define MISSING_CASE(x) WARN(1, "Missing case (%s == %ld)\n", \
 			     __stringify(x), (long)(x))
@@ -270,19 +269,6 @@ static inline void __add_taint_for_CI(unsigned int taint)
 	add_taint(taint, LOCKDEP_STILL_OK);
 }
 
-void cancel_timer(struct timer_list *t);
-void set_timer_ms(struct timer_list *t, unsigned long timeout);
-
-static inline bool timer_active(const struct timer_list *t)
-{
-	return READ_ONCE(t->expires);
-}
-
-static inline bool timer_expired(const struct timer_list *t)
-{
-	return timer_active(t) && !timer_pending(t);
-}
-
 static inline bool i915_run_as_guest(void)
 {
 #if IS_ENABLED(CONFIG_X86)

From 2f04432fcf0ad7b825d67ee283345e2721c83f5c Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 11 Sep 2025 12:17:39 +0300
Subject: [PATCH 1216/1455] drm/i915: split out i915_list_util.h

Move list related utilities from i915_utils.h to separate new file
i915_list_util.h. Clean up related includes.

Note: Arguably none of this should exist in i915 in the first place. At
least isolate it better.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/d7526809735194137116682f37cfa126a6a87ec9.1757582214.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../drm/i915/gt/intel_execlists_submission.c  |  1 +
 .../gpu/drm/i915/gt/intel_gt_buffer_pool.c    |  1 +
 drivers/gpu/drm/i915/gt/intel_timeline.h      |  2 +-
 drivers/gpu/drm/i915/i915_list_util.h         | 23 +++++++++++++++++++
 drivers/gpu/drm/i915/i915_utils.h             | 14 -----------
 5 files changed, 26 insertions(+), 15 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_list_util.h

diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
index 52c8fddedfce..0c1b2df02d26 100644
--- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
@@ -112,6 +112,7 @@
 
 #include "gen8_engine_cs.h"
 #include "i915_drv.h"
+#include "i915_list_util.h"
 #include "i915_reg.h"
 #include "i915_timer_util.h"
 #include "i915_trace.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
index 86b5a9ba323d..c7befc5c20d0 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c
@@ -7,6 +7,7 @@
 #include "gem/i915_gem_object.h"
 
 #include "i915_drv.h"
+#include "i915_list_util.h"
 #include "intel_engine_pm.h"
 #include "intel_gt_buffer_pool.h"
 
diff --git a/drivers/gpu/drm/i915/gt/intel_timeline.h b/drivers/gpu/drm/i915/gt/intel_timeline.h
index 5f8c8f79714e..85b43f9b9d95 100644
--- a/drivers/gpu/drm/i915/gt/intel_timeline.h
+++ b/drivers/gpu/drm/i915/gt/intel_timeline.h
@@ -9,8 +9,8 @@
 #include <linux/lockdep.h>
 
 #include "i915_active.h"
+#include "i915_list_util.h"
 #include "i915_syncmap.h"
-#include "i915_utils.h"
 #include "intel_timeline_types.h"
 
 struct drm_printer;
diff --git a/drivers/gpu/drm/i915/i915_list_util.h b/drivers/gpu/drm/i915/i915_list_util.h
new file mode 100644
index 000000000000..4e515dc8a3e0
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_list_util.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __I915_LIST_UTIL_H__
+#define __I915_LIST_UTIL_H__
+
+#include <linux/list.h>
+#include <asm/rwonce.h>
+
+static inline void __list_del_many(struct list_head *head,
+				   struct list_head *first)
+{
+	first->prev = head;
+	WRITE_ONCE(head->next, first);
+}
+
+static inline int list_is_last_rcu(const struct list_head *list,
+				   const struct list_head *head)
+{
+	return READ_ONCE(list->next) == head;
+}
+
+#endif /* __I915_LIST_UTIL_H__ */
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 6278a74d08c2..4a6437d6e00e 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -25,7 +25,6 @@
 #ifndef __I915_UTILS_H
 #define __I915_UTILS_H
 
-#include <linux/list.h>
 #include <linux/overflow.h>
 #include <linux/sched.h>
 #include <linux/string_helpers.h>
@@ -101,19 +100,6 @@ static inline bool is_power_of_2_u64(u64 n)
 	return (n != 0 && ((n & (n - 1)) == 0));
 }
 
-static inline void __list_del_many(struct list_head *head,
-				   struct list_head *first)
-{
-	first->prev = head;
-	WRITE_ONCE(head->next, first);
-}
-
-static inline int list_is_last_rcu(const struct list_head *list,
-				   const struct list_head *head)
-{
-	return READ_ONCE(list->next) == head;
-}
-
 static inline unsigned long msecs_to_jiffies_timeout(const unsigned int m)
 {
 	unsigned long j = msecs_to_jiffies(m);

From 65805c2ed74ee47e81246f97f4b3c126a75f0b0e Mon Sep 17 00:00:00 2001
From: Jani Nikula <jani.nikula@intel.com>
Date: Thu, 11 Sep 2025 12:17:40 +0300
Subject: [PATCH 1217/1455] drm/i915: split out i915_wait_util.h

Move waiting related utilities from i915_utils.h to separate new file
i915_wait_util.h. Clean up related includes.

Note: Many of the various wait macro usages could likely be refactored
to use poll_timeout_us().

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/431396ac8cdb3e2f4ff053a8933290289a66ce42.1757582214.git.jani.nikula@intel.com
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
---
 .../drm/i915/gt/intel_execlists_submission.c  |   1 +
 drivers/gpu/drm/i915/gt/intel_gt_mcr.c        |   1 +
 drivers/gpu/drm/i915/gt/intel_reset.c         |   7 +-
 .../gpu/drm/i915/gt/intel_ring_submission.c   |   7 +-
 drivers/gpu/drm/i915/gt/intel_rps.c           |   2 +
 drivers/gpu/drm/i915/gt/selftest_tlb.c        |   6 +-
 drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c  |   6 +-
 .../i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c |   2 +
 drivers/gpu/drm/i915/gt/uc/intel_guc.c        |   8 +-
 drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c     |   3 +-
 drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c     |   4 +-
 drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c   |  15 ++-
 .../gpu/drm/i915/gt/uc/intel_guc_submission.c |  10 +-
 drivers/gpu/drm/i915/gvt/mmio_context.c       |   6 +-
 drivers/gpu/drm/i915/i915_debugfs.c           |   3 +-
 drivers/gpu/drm/i915/i915_utils.h             | 106 ----------------
 drivers/gpu/drm/i915/i915_wait_util.h         | 119 ++++++++++++++++++
 drivers/gpu/drm/i915/intel_pcode.c            |   1 +
 drivers/gpu/drm/i915/intel_uncore.c           |   7 +-
 drivers/gpu/drm/i915/pxp/intel_pxp.c          |   4 +-
 drivers/gpu/drm/i915/selftests/i915_request.c |   5 +-
 .../gpu/drm/i915/selftests/i915_selftest.c    |   2 +-
 drivers/gpu/drm/i915/selftests/igt_spinner.c  |   5 +-
 drivers/gpu/drm/i915/vlv_suspend.c            |   5 +-
 24 files changed, 184 insertions(+), 151 deletions(-)
 create mode 100644 drivers/gpu/drm/i915/i915_wait_util.h

diff --git a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
index 0c1b2df02d26..7f389cb0bde4 100644
--- a/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_execlists_submission.c
@@ -117,6 +117,7 @@
 #include "i915_timer_util.h"
 #include "i915_trace.h"
 #include "i915_vgpu.h"
+#include "i915_wait_util.h"
 #include "intel_breadcrumbs.h"
 #include "intel_context.h"
 #include "intel_engine_heartbeat.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_gt_mcr.c b/drivers/gpu/drm/i915/gt/intel_gt_mcr.c
index a60822e2b5d4..c3afa321fe30 100644
--- a/drivers/gpu/drm/i915/gt/intel_gt_mcr.c
+++ b/drivers/gpu/drm/i915/gt/intel_gt_mcr.c
@@ -4,6 +4,7 @@
  */
 
 #include "i915_drv.h"
+#include "i915_wait_util.h"
 #include "intel_gt.h"
 #include "intel_gt_mcr.h"
 #include "intel_gt_print.h"
diff --git a/drivers/gpu/drm/i915/gt/intel_reset.c b/drivers/gpu/drm/i915/gt/intel_reset.c
index 4a1675dea1c7..41b5036dc538 100644
--- a/drivers/gpu/drm/i915/gt/intel_reset.c
+++ b/drivers/gpu/drm/i915/gt/intel_reset.c
@@ -9,18 +9,17 @@
 
 #include "display/intel_display_reset.h"
 #include "display/intel_overlay.h"
-
 #include "gem/i915_gem_context.h"
-
 #include "gt/intel_gt_regs.h"
-
 #include "gt/uc/intel_gsc_fw.h"
+#include "uc/intel_guc.h"
 
 #include "i915_drv.h"
 #include "i915_file_private.h"
 #include "i915_gpu_error.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "intel_breadcrumbs.h"
 #include "intel_engine_pm.h"
 #include "intel_engine_regs.h"
@@ -32,8 +31,6 @@
 #include "intel_pci_config.h"
 #include "intel_reset.h"
 
-#include "uc/intel_guc.h"
-
 #define RESET_MAX_RETRIES 3
 
 static void client_mark_guilty(struct i915_gem_context *ctx, bool banned)
diff --git a/drivers/gpu/drm/i915/gt/intel_ring_submission.c b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
index 2a6d79abf25b..8314a4b0505e 100644
--- a/drivers/gpu/drm/i915/gt/intel_ring_submission.c
+++ b/drivers/gpu/drm/i915/gt/intel_ring_submission.c
@@ -15,18 +15,19 @@
 #include "i915_irq.h"
 #include "i915_mitigations.h"
 #include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "intel_breadcrumbs.h"
 #include "intel_context.h"
+#include "intel_engine_heartbeat.h"
+#include "intel_engine_pm.h"
 #include "intel_engine_regs.h"
 #include "intel_gt.h"
 #include "intel_gt_irq.h"
+#include "intel_gt_print.h"
 #include "intel_gt_regs.h"
 #include "intel_reset.h"
 #include "intel_ring.h"
 #include "shmem_utils.h"
-#include "intel_engine_heartbeat.h"
-#include "intel_engine_pm.h"
-#include "intel_gt_print.h"
 
 /* Rough estimate of the typical request size, performing a flush,
  * set-context and then emitting the batch.
diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c
index 006042e0b229..4da94098bd3e 100644
--- a/drivers/gpu/drm/i915/gt/intel_rps.c
+++ b/drivers/gpu/drm/i915/gt/intel_rps.c
@@ -10,9 +10,11 @@
 #include "display/intel_display.h"
 #include "display/intel_display_rps.h"
 #include "soc/intel_dram.h"
+
 #include "i915_drv.h"
 #include "i915_irq.h"
 #include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "intel_breadcrumbs.h"
 #include "intel_gt.h"
 #include "intel_gt_clock_utils.h"
diff --git a/drivers/gpu/drm/i915/gt/selftest_tlb.c b/drivers/gpu/drm/i915/gt/selftest_tlb.c
index 69ed946a39e5..a5184f09d1de 100644
--- a/drivers/gpu/drm/i915/gt/selftest_tlb.c
+++ b/drivers/gpu/drm/i915/gt/selftest_tlb.c
@@ -3,17 +3,17 @@
  * Copyright © 2022 Intel Corporation
  */
 
-#include "i915_selftest.h"
-
 #include "gem/i915_gem_internal.h"
 #include "gem/i915_gem_lmem.h"
 #include "gem/i915_gem_region.h"
 
 #include "gen8_engine_cs.h"
 #include "i915_gem_ww.h"
+#include "i915_selftest.h"
+#include "i915_wait_util.h"
+#include "intel_context.h"
 #include "intel_engine_regs.h"
 #include "intel_gpu_commands.h"
-#include "intel_context.h"
 #include "intel_gt.h"
 #include "intel_ring.h"
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
index d8edd7c054c8..e7444ebc373e 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_proxy.c
@@ -10,11 +10,13 @@
 
 #include "gt/intel_gt.h"
 #include "gt/intel_gt_print.h"
+
+#include "i915_drv.h"
+#include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "intel_gsc_proxy.h"
 #include "intel_gsc_uc.h"
 #include "intel_gsc_uc_heci_cmd_submit.h"
-#include "i915_drv.h"
-#include "i915_reg.h"
 
 /*
  * GSC proxy:
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
index 2fde5c360cff..9bd29be7656f 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_gsc_uc_heci_cmd_submit.c
@@ -8,6 +8,8 @@
 #include "gt/intel_gpu_commands.h"
 #include "gt/intel_gt.h"
 #include "gt/intel_ring.h"
+
+#include "i915_wait_util.h"
 #include "intel_gsc_uc_heci_cmd_submit.h"
 
 struct gsc_heci_pkt {
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
index f360f020d8f1..52ec4421a211 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.c
@@ -8,15 +8,17 @@
 #include "gt/intel_gt_irq.h"
 #include "gt/intel_gt_pm_irq.h"
 #include "gt/intel_gt_regs.h"
+
+#include "i915_drv.h"
+#include "i915_irq.h"
+#include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "intel_guc.h"
 #include "intel_guc_ads.h"
 #include "intel_guc_capture.h"
 #include "intel_guc_print.h"
 #include "intel_guc_slpc.h"
 #include "intel_guc_submission.h"
-#include "i915_drv.h"
-#include "i915_irq.h"
-#include "i915_reg.h"
 
 /**
  * DOC: GuC
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
index 380a11c92d63..3e7e5badcc2b 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c
@@ -5,11 +5,12 @@
 
 #include <linux/circ_buf.h>
 #include <linux/ktime.h>
-#include <linux/time64.h>
 #include <linux/string_helpers.h>
+#include <linux/time64.h>
 #include <linux/timekeeping.h>
 
 #include "i915_drv.h"
+#include "i915_wait_util.h"
 #include "intel_guc_ct.h"
 #include "intel_guc_print.h"
 
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
index e7ccfa520df3..1bb1491f90fc 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c
@@ -13,9 +13,11 @@
 #include "gt/intel_gt_mcr.h"
 #include "gt/intel_gt_regs.h"
 #include "gt/intel_rps.h"
+
+#include "i915_drv.h"
+#include "i915_wait_util.h"
 #include "intel_guc_fw.h"
 #include "intel_guc_print.h"
-#include "i915_drv.h"
 
 static void guc_prepare_xfer(struct intel_gt *gt)
 {
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
index d5ee6e5e1443..fa9af08f9708 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c
@@ -3,18 +3,21 @@
  * Copyright © 2021 Intel Corporation
  */
 
-#include <drm/drm_cache.h>
 #include <linux/string_helpers.h>
 
-#include "i915_drv.h"
-#include "i915_reg.h"
-#include "intel_guc_slpc.h"
-#include "intel_guc_print.h"
-#include "intel_mchbar_regs.h"
+#include <drm/drm_cache.h>
+
 #include "gt/intel_gt.h"
 #include "gt/intel_gt_regs.h"
 #include "gt/intel_rps.h"
 
+#include "i915_drv.h"
+#include "i915_reg.h"
+#include "i915_wait_util.h"
+#include "intel_guc_print.h"
+#include "intel_guc_slpc.h"
+#include "intel_mchbar_regs.h"
+
 /**
  * DOC: SLPC - Dynamic Frequency management
  *
diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
index 127316d2c8aa..68f2b8d363ac 100644
--- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
+++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c
@@ -25,16 +25,16 @@
 #include "gt/intel_mocs.h"
 #include "gt/intel_ring.h"
 
+#include "i915_drv.h"
+#include "i915_irq.h"
+#include "i915_reg.h"
+#include "i915_trace.h"
+#include "i915_wait_util.h"
 #include "intel_guc_ads.h"
 #include "intel_guc_capture.h"
 #include "intel_guc_print.h"
 #include "intel_guc_submission.h"
 
-#include "i915_drv.h"
-#include "i915_reg.h"
-#include "i915_irq.h"
-#include "i915_trace.h"
-
 /**
  * DOC: GuC-based command submission
  *
diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c
index 2f7208843367..0b810baad20a 100644
--- a/drivers/gpu/drm/i915/gvt/mmio_context.c
+++ b/drivers/gpu/drm/i915/gvt/mmio_context.c
@@ -33,14 +33,16 @@
  *
  */
 
-#include "i915_drv.h"
-#include "i915_reg.h"
 #include "gt/intel_context.h"
 #include "gt/intel_engine_regs.h"
 #include "gt/intel_gpu_commands.h"
 #include "gt/intel_gt_regs.h"
 #include "gt/intel_ring.h"
+
 #include "gvt.h"
+#include "i915_drv.h"
+#include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "trace.h"
 
 #define GEN9_MOCS_SIZE		64
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index 23fa098c4479..c2e38d4bcd01 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -26,11 +26,11 @@
  *
  */
 
+#include <linux/debugfs.h>
 #include <linux/sched/mm.h>
 #include <linux/sort.h>
 #include <linux/string_helpers.h>
 
-#include <linux/debugfs.h>
 #include <drm/drm_debugfs.h>
 
 #include "gem/i915_gem_context.h"
@@ -54,6 +54,7 @@
 #include "i915_irq.h"
 #include "i915_reg.h"
 #include "i915_scheduler.h"
+#include "i915_wait_util.h"
 #include "intel_mchbar_regs.h"
 
 static inline struct drm_i915_private *node_to_i915(struct drm_info_node *node)
diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h
index 4a6437d6e00e..a0c892e4c40d 100644
--- a/drivers/gpu/drm/i915/i915_utils.h
+++ b/drivers/gpu/drm/i915/i915_utils.h
@@ -134,112 +134,6 @@ wait_remaining_ms_from_jiffies(unsigned long timestamp_jiffies, int to_wait_ms)
 	}
 }
 
-/*
- * __wait_for - magic wait macro
- *
- * Macro to help avoid open coding check/wait/timeout patterns. Note that it's
- * important that we check the condition again after having timed out, since the
- * timeout could be due to preemption or similar and we've never had a chance to
- * check the condition before the timeout.
- */
-#define __wait_for(OP, COND, US, Wmin, Wmax) ({ \
-	const ktime_t end__ = ktime_add_ns(ktime_get_raw(), 1000ll * (US)); \
-	long wait__ = (Wmin); /* recommended min for usleep is 10 us */	\
-	int ret__;							\
-	might_sleep();							\
-	for (;;) {							\
-		const bool expired__ = ktime_after(ktime_get_raw(), end__); \
-		OP;							\
-		/* Guarantee COND check prior to timeout */		\
-		barrier();						\
-		if (COND) {						\
-			ret__ = 0;					\
-			break;						\
-		}							\
-		if (expired__) {					\
-			ret__ = -ETIMEDOUT;				\
-			break;						\
-		}							\
-		usleep_range(wait__, wait__ * 2);			\
-		if (wait__ < (Wmax))					\
-			wait__ <<= 1;					\
-	}								\
-	ret__;								\
-})
-
-#define _wait_for(COND, US, Wmin, Wmax)	__wait_for(, (COND), (US), (Wmin), \
-						   (Wmax))
-#define wait_for(COND, MS)		_wait_for((COND), (MS) * 1000, 10, 1000)
-
-/*
- * If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false.
- * On PREEMPT_RT the context isn't becoming atomic because it is used in an
- * interrupt handler or because a spinlock_t is acquired. This leads to
- * warnings which don't occur otherwise and therefore the check is disabled.
- */
-#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) && IS_ENABLED(CONFIG_PREEMPT_COUNT) && !defined(CONFIG_PREEMPT_RT)
-# define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) WARN_ON_ONCE((ATOMIC) && !in_atomic())
-#else
-# define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) do { } while (0)
-#endif
-
-#define _wait_for_atomic(COND, US, ATOMIC) \
-({ \
-	int cpu, ret, timeout = (US) * 1000; \
-	u64 base; \
-	_WAIT_FOR_ATOMIC_CHECK(ATOMIC); \
-	if (!(ATOMIC)) { \
-		preempt_disable(); \
-		cpu = smp_processor_id(); \
-	} \
-	base = local_clock(); \
-	for (;;) { \
-		u64 now = local_clock(); \
-		if (!(ATOMIC)) \
-			preempt_enable(); \
-		/* Guarantee COND check prior to timeout */ \
-		barrier(); \
-		if (COND) { \
-			ret = 0; \
-			break; \
-		} \
-		if (now - base >= timeout) { \
-			ret = -ETIMEDOUT; \
-			break; \
-		} \
-		cpu_relax(); \
-		if (!(ATOMIC)) { \
-			preempt_disable(); \
-			if (unlikely(cpu != smp_processor_id())) { \
-				timeout -= now - base; \
-				cpu = smp_processor_id(); \
-				base = local_clock(); \
-			} \
-		} \
-	} \
-	ret; \
-})
-
-#define wait_for_us(COND, US) \
-({ \
-	int ret__; \
-	BUILD_BUG_ON(!__builtin_constant_p(US)); \
-	if ((US) > 10) \
-		ret__ = _wait_for((COND), (US), 10, 10); \
-	else \
-		ret__ = _wait_for_atomic((COND), (US), 0); \
-	ret__; \
-})
-
-#define wait_for_atomic_us(COND, US) \
-({ \
-	BUILD_BUG_ON(!__builtin_constant_p(US)); \
-	BUILD_BUG_ON((US) > 50000); \
-	_wait_for_atomic((COND), (US), 1); \
-})
-
-#define wait_for_atomic(COND, MS) wait_for_atomic_us((COND), (MS) * 1000)
-
 #define KHz(x) (1000 * (x))
 #define MHz(x) KHz(1000 * (x))
 
diff --git a/drivers/gpu/drm/i915/i915_wait_util.h b/drivers/gpu/drm/i915/i915_wait_util.h
new file mode 100644
index 000000000000..7376898e3bf8
--- /dev/null
+++ b/drivers/gpu/drm/i915/i915_wait_util.h
@@ -0,0 +1,119 @@
+/* SPDX-License-Identifier: MIT */
+/* Copyright © 2025 Intel Corporation */
+
+#ifndef __I915_WAIT_UTIL_H__
+#define __I915_WAIT_UTIL_H__
+
+#include <linux/compiler.h>
+#include <linux/delay.h>
+#include <linux/ktime.h>
+#include <linux/sched/clock.h>
+#include <linux/smp.h>
+
+/*
+ * __wait_for - magic wait macro
+ *
+ * Macro to help avoid open coding check/wait/timeout patterns. Note that it's
+ * important that we check the condition again after having timed out, since the
+ * timeout could be due to preemption or similar and we've never had a chance to
+ * check the condition before the timeout.
+ */
+#define __wait_for(OP, COND, US, Wmin, Wmax) ({ \
+	const ktime_t end__ = ktime_add_ns(ktime_get_raw(), 1000ll * (US)); \
+	long wait__ = (Wmin); /* recommended min for usleep is 10 us */	\
+	int ret__;							\
+	might_sleep();							\
+	for (;;) {							\
+		const bool expired__ = ktime_after(ktime_get_raw(), end__); \
+		OP;							\
+		/* Guarantee COND check prior to timeout */		\
+		barrier();						\
+		if (COND) {						\
+			ret__ = 0;					\
+			break;						\
+		}							\
+		if (expired__) {					\
+			ret__ = -ETIMEDOUT;				\
+			break;						\
+		}							\
+		usleep_range(wait__, wait__ * 2);			\
+		if (wait__ < (Wmax))					\
+			wait__ <<= 1;					\
+	}								\
+	ret__;								\
+})
+
+#define _wait_for(COND, US, Wmin, Wmax)	__wait_for(, (COND), (US), (Wmin), \
+						   (Wmax))
+#define wait_for(COND, MS)		_wait_for((COND), (MS) * 1000, 10, 1000)
+
+/*
+ * If CONFIG_PREEMPT_COUNT is disabled, in_atomic() always reports false.
+ * On PREEMPT_RT the context isn't becoming atomic because it is used in an
+ * interrupt handler or because a spinlock_t is acquired. This leads to
+ * warnings which don't occur otherwise and therefore the check is disabled.
+ */
+#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) && IS_ENABLED(CONFIG_PREEMPT_COUNT) && !defined(CONFIG_PREEMPT_RT)
+# define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) WARN_ON_ONCE((ATOMIC) && !in_atomic())
+#else
+# define _WAIT_FOR_ATOMIC_CHECK(ATOMIC) do { } while (0)
+#endif
+
+#define _wait_for_atomic(COND, US, ATOMIC) \
+({ \
+	int cpu, ret, timeout = (US) * 1000; \
+	u64 base; \
+	_WAIT_FOR_ATOMIC_CHECK(ATOMIC); \
+	if (!(ATOMIC)) { \
+		preempt_disable(); \
+		cpu = smp_processor_id(); \
+	} \
+	base = local_clock(); \
+	for (;;) { \
+		u64 now = local_clock(); \
+		if (!(ATOMIC)) \
+			preempt_enable(); \
+		/* Guarantee COND check prior to timeout */ \
+		barrier(); \
+		if (COND) { \
+			ret = 0; \
+			break; \
+		} \
+		if (now - base >= timeout) { \
+			ret = -ETIMEDOUT; \
+			break; \
+		} \
+		cpu_relax(); \
+		if (!(ATOMIC)) { \
+			preempt_disable(); \
+			if (unlikely(cpu != smp_processor_id())) { \
+				timeout -= now - base; \
+				cpu = smp_processor_id(); \
+				base = local_clock(); \
+			} \
+		} \
+	} \
+	ret; \
+})
+
+#define wait_for_us(COND, US) \
+({ \
+	int ret__; \
+	BUILD_BUG_ON(!__builtin_constant_p(US)); \
+	if ((US) > 10) \
+		ret__ = _wait_for((COND), (US), 10, 10); \
+	else \
+		ret__ = _wait_for_atomic((COND), (US), 0); \
+	ret__; \
+})
+
+#define wait_for_atomic_us(COND, US) \
+({ \
+	BUILD_BUG_ON(!__builtin_constant_p(US)); \
+	BUILD_BUG_ON((US) > 50000); \
+	_wait_for_atomic((COND), (US), 1); \
+})
+
+#define wait_for_atomic(COND, MS) wait_for_atomic_us((COND), (MS) * 1000)
+
+#endif /* __I915_WAIT_UTIL_H__ */
diff --git a/drivers/gpu/drm/i915/intel_pcode.c b/drivers/gpu/drm/i915/intel_pcode.c
index 81da75108c60..55ffedad2490 100644
--- a/drivers/gpu/drm/i915/intel_pcode.c
+++ b/drivers/gpu/drm/i915/intel_pcode.c
@@ -5,6 +5,7 @@
 
 #include "i915_drv.h"
 #include "i915_reg.h"
+#include "i915_wait_util.h"
 #include "intel_pcode.h"
 
 static int gen6_check_mailbox_status(u32 mbox)
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index 4ccba7c8ffb3..8cb59f8d1f4c 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -21,19 +21,20 @@
  * IN THE SOFTWARE.
  */
 
-#include <drm/drm_managed.h>
 #include <linux/pm_runtime.h>
 
-#include "display/intel_display_core.h"
+#include <drm/drm_managed.h>
 
-#include "gt/intel_gt.h"
+#include "display/intel_display_core.h"
 #include "gt/intel_engine_regs.h"
+#include "gt/intel_gt.h"
 #include "gt/intel_gt_regs.h"
 
 #include "i915_drv.h"
 #include "i915_iosf_mbi.h"
 #include "i915_reg.h"
 #include "i915_vgpu.h"
+#include "i915_wait_util.h"
 #include "intel_uncore_trace.h"
 
 #define FORCEWAKE_ACK_TIMEOUT_MS 50
diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp.c b/drivers/gpu/drm/i915/pxp/intel_pxp.c
index f8da693ad3ce..27d545c4e6a5 100644
--- a/drivers/gpu/drm/i915/pxp/intel_pxp.c
+++ b/drivers/gpu/drm/i915/pxp/intel_pxp.c
@@ -2,15 +2,15 @@
 /*
  * Copyright(c) 2020 Intel Corporation.
  */
+
 #include <linux/workqueue.h>
 
 #include "gem/i915_gem_context.h"
-
 #include "gt/intel_context.h"
 #include "gt/intel_gt.h"
 
 #include "i915_drv.h"
-
+#include "i915_wait_util.h"
 #include "intel_pxp.h"
 #include "intel_pxp_gsccs.h"
 #include "intel_pxp_irq.h"
diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c
index 2fb7a9e7efec..48cd617247d1 100644
--- a/drivers/gpu/drm/i915/selftests/i915_request.c
+++ b/drivers/gpu/drm/i915/selftests/i915_request.c
@@ -22,14 +22,13 @@
  *
  */
 
-#include <linux/prime_numbers.h>
 #include <linux/pm_qos.h>
+#include <linux/prime_numbers.h>
 #include <linux/sort.h>
 
 #include "gem/i915_gem_internal.h"
 #include "gem/i915_gem_pm.h"
 #include "gem/selftests/mock_context.h"
-
 #include "gt/intel_engine_heartbeat.h"
 #include "gt/intel_engine_pm.h"
 #include "gt/intel_engine_user.h"
@@ -40,11 +39,11 @@
 
 #include "i915_random.h"
 #include "i915_selftest.h"
+#include "i915_wait_util.h"
 #include "igt_flush_test.h"
 #include "igt_live_test.h"
 #include "igt_spinner.h"
 #include "lib_sw_fence.h"
-
 #include "mock_drm.h"
 #include "mock_gem_device.h"
 
diff --git a/drivers/gpu/drm/i915/selftests/i915_selftest.c b/drivers/gpu/drm/i915/selftests/i915_selftest.c
index 889281819c5b..9c276c9d0a75 100644
--- a/drivers/gpu/drm/i915/selftests/i915_selftest.c
+++ b/drivers/gpu/drm/i915/selftests/i915_selftest.c
@@ -31,7 +31,7 @@
 #include "i915_driver.h"
 #include "i915_drv.h"
 #include "i915_selftest.h"
-
+#include "i915_wait_util.h"
 #include "igt_flush_test.h"
 
 struct i915_selftest i915_selftest __read_mostly = {
diff --git a/drivers/gpu/drm/i915/selftests/igt_spinner.c b/drivers/gpu/drm/i915/selftests/igt_spinner.c
index 8c3e1f20e5a1..820364171ebe 100644
--- a/drivers/gpu/drm/i915/selftests/igt_spinner.c
+++ b/drivers/gpu/drm/i915/selftests/igt_spinner.c
@@ -3,12 +3,13 @@
  *
  * Copyright © 2018 Intel Corporation
  */
-#include "gt/intel_gpu_commands.h"
-#include "gt/intel_gt.h"
 
 #include "gem/i915_gem_internal.h"
 #include "gem/selftests/igt_gem_utils.h"
+#include "gt/intel_gpu_commands.h"
+#include "gt/intel_gt.h"
 
+#include "i915_wait_util.h"
 #include "igt_spinner.h"
 
 int igt_spinner_init(struct igt_spinner *spin, struct intel_gt *gt)
diff --git a/drivers/gpu/drm/i915/vlv_suspend.c b/drivers/gpu/drm/i915/vlv_suspend.c
index fc9f311ea1db..221e4c0b2c58 100644
--- a/drivers/gpu/drm/i915/vlv_suspend.c
+++ b/drivers/gpu/drm/i915/vlv_suspend.c
@@ -8,16 +8,17 @@
 
 #include <drm/drm_print.h>
 
+#include "gt/intel_gt_regs.h"
+
 #include "i915_drv.h"
 #include "i915_reg.h"
 #include "i915_trace.h"
 #include "i915_utils.h"
+#include "i915_wait_util.h"
 #include "intel_clock_gating.h"
 #include "intel_uncore_trace.h"
 #include "vlv_suspend.h"
 
-#include "gt/intel_gt_regs.h"
-
 struct vlv_s0ix_state {
 	/* GAM */
 	u32 wr_watermark;

From edffa93a93d896463948f966dcf0cf2b9de461d0 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 9 Sep 2025 18:59:35 +0200
Subject: [PATCH 1218/1455] drm/xe: Keep xe_gt_err() macro definitions together

There is no need to keep them separated. No functional changes.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250909165941.31730-2-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_printk.h | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_printk.h b/drivers/gpu/drm/xe/xe_gt_printk.h
index 11da0228cea7..fd9cba14383a 100644
--- a/drivers/gpu/drm/xe/xe_gt_printk.h
+++ b/drivers/gpu/drm/xe/xe_gt_printk.h
@@ -13,11 +13,14 @@
 #define xe_gt_printk(_gt, _level, _fmt, ...) \
 	drm_##_level(&gt_to_xe(_gt)->drm, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
 
+#define xe_gt_err(_gt, _fmt, ...) \
+	xe_gt_printk((_gt), err, _fmt, ##__VA_ARGS__)
+
 #define xe_gt_err_once(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), err_once, _fmt, ##__VA_ARGS__)
 
-#define xe_gt_err(_gt, _fmt, ...) \
-	xe_gt_printk((_gt), err, _fmt, ##__VA_ARGS__)
+#define xe_gt_err_ratelimited(_gt, _fmt, ...) \
+	xe_gt_printk((_gt), err_ratelimited, _fmt, ##__VA_ARGS__)
 
 #define xe_gt_warn(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), warn, _fmt, ##__VA_ARGS__)
@@ -31,9 +34,6 @@
 #define xe_gt_dbg(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), dbg, _fmt, ##__VA_ARGS__)
 
-#define xe_gt_err_ratelimited(_gt, _fmt, ...) \
-	xe_gt_printk((_gt), err_ratelimited, _fmt, ##__VA_ARGS__)
-
 #define xe_gt_WARN(_gt, _condition, _fmt, ...) \
 	drm_WARN(&gt_to_xe(_gt)->drm, _condition, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
 

From a2dc39fb1c0c2a2113ffd72dbcc5888dadba919f Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 9 Sep 2025 18:59:36 +0200
Subject: [PATCH 1219/1455] drm/xe: Drop "gt_" prefix from xe_gt_WARN() macros

Those WARN messages will already include GT-specific "GT%u:" prefix
so there is no point to include additional "gt_" prefix.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250909165941.31730-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_printk.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_printk.h b/drivers/gpu/drm/xe/xe_gt_printk.h
index fd9cba14383a..f2091c734aba 100644
--- a/drivers/gpu/drm/xe/xe_gt_printk.h
+++ b/drivers/gpu/drm/xe/xe_gt_printk.h
@@ -41,10 +41,10 @@
 	drm_WARN_ONCE(&gt_to_xe(_gt)->drm, _condition, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
 
 #define xe_gt_WARN_ON(_gt, _condition) \
-	xe_gt_WARN((_gt), _condition, "%s(%s)", "gt_WARN_ON", __stringify(_condition))
+	xe_gt_WARN((_gt), _condition, "%s(%s)", "WARN_ON", __stringify(_condition))
 
 #define xe_gt_WARN_ON_ONCE(_gt, _condition) \
-	xe_gt_WARN_ONCE((_gt), _condition, "%s(%s)", "gt_WARN_ON_ONCE", __stringify(_condition))
+	xe_gt_WARN_ONCE((_gt), _condition, "%s(%s)", "WARN_ON_ONCE", __stringify(_condition))
 
 static inline void __xe_gt_printfn_err(struct drm_printer *p, struct va_format *vaf)
 {

From efd54b0cff9902638bc48a797c1ef6f57c348ac2 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 9 Sep 2025 18:59:37 +0200
Subject: [PATCH 1220/1455] drm/xe: Prepare format for GT-oriented messages in
 one place

To avoid code duplication (and thus potential mistakes) and to
allow easier changes (if needed) of the prefix format of the
GT-oriented messages, prepare that prefix in dedicated macro.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250909165941.31730-4-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_printk.h | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_printk.h b/drivers/gpu/drm/xe/xe_gt_printk.h
index f2091c734aba..dcc45628cca9 100644
--- a/drivers/gpu/drm/xe/xe_gt_printk.h
+++ b/drivers/gpu/drm/xe/xe_gt_printk.h
@@ -10,8 +10,10 @@
 
 #include "xe_gt_types.h"
 
+#define __XE_GT_PRINTK_FMT(_gt, _fmt, _args...)	"GT%u: " _fmt, (_gt)->info.id, ##_args
+
 #define xe_gt_printk(_gt, _level, _fmt, ...) \
-	drm_##_level(&gt_to_xe(_gt)->drm, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
+	drm_##_level(&gt_to_xe(_gt)->drm, __XE_GT_PRINTK_FMT((_gt), _fmt, ##__VA_ARGS__))
 
 #define xe_gt_err(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), err, _fmt, ##__VA_ARGS__)
@@ -34,11 +36,14 @@
 #define xe_gt_dbg(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), dbg, _fmt, ##__VA_ARGS__)
 
+#define xe_gt_WARN_type(_gt, _type, _condition, _fmt, ...) \
+	drm_WARN##_type(&gt_to_xe(_gt)->drm, _condition, _fmt, ## __VA_ARGS__)
+
 #define xe_gt_WARN(_gt, _condition, _fmt, ...) \
-	drm_WARN(&gt_to_xe(_gt)->drm, _condition, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
+	xe_gt_WARN_type((_gt),, _condition, __XE_GT_PRINTK_FMT((_gt), _fmt, ##__VA_ARGS__))
 
 #define xe_gt_WARN_ONCE(_gt, _condition, _fmt, ...) \
-	drm_WARN_ONCE(&gt_to_xe(_gt)->drm, _condition, "GT%u: " _fmt, (_gt)->info.id, ##__VA_ARGS__)
+	xe_gt_WARN_type((_gt), _ONCE, _condition, __XE_GT_PRINTK_FMT((_gt), _fmt, ##__VA_ARGS__))
 
 #define xe_gt_WARN_ON(_gt, _condition) \
 	xe_gt_WARN((_gt), _condition, "%s(%s)", "WARN_ON", __stringify(_condition))
@@ -72,7 +77,7 @@ static inline void __xe_gt_printfn_dbg(struct drm_printer *p, struct va_format *
 	dbg = drm_dbg_printer(&gt_to_xe(gt)->drm, DRM_UT_DRIVER, NULL);
 	dbg.origin = p->origin;
 
-	drm_printf(&dbg, "GT%u: %pV", gt->info.id, vaf);
+	drm_printf(&dbg, __XE_GT_PRINTK_FMT(gt, "%pV", vaf));
 }
 
 /**

From 48a8659cd5703e539a87f447fb5a0c59628cb742 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 9 Sep 2025 18:59:38 +0200
Subject: [PATCH 1221/1455] drm/xe: Add dedicated printk macros for tile and
 device

We already have dedicated helper macros for printing GT-oriented
messages but we don't have any to print messages that are tile
oriented and we wrongly try to use plain drm or GT-oriented ones.

Add tile-oriented printk messages and to provide similar coverage
as we have with xe_assert() macros. Also add set of simple macros
for the top level xe_device, which we could easily tweak to include
extra device specific info if needed.

Typical output of our printk macros will look like:

 [drm] this is xe_WARN()
 [drm] *ERROR* this is xe_err()
 [drm] *ERROR* this is xe_err_printer()
 [drm] this is xe_info()
 [drm] this is xe_info_printer()
 [drm:printk_demo.cold] this is xe_dbg()
 [drm:printk_demo.cold] this is xe_dbg_printer()

 [drm] Tile0: this is xe_tile_WARN()
 [drm] *ERROR* Tile0: this is xe_tile_err()
 [drm] *ERROR* Tile0: this is xe_tile_err_printer()
 [drm] Tile0: this is xe_tile_info()
 [drm] Tile0: this is xe_tile_info_printer()
 [drm:printk_demo.cold] Tile0: this is xe_tile_dbg()
 [drm:printk_demo.cold] Tile0: this is xe_tile_dbg_printer()

 [drm] Tile0: GT0: this is xe_gt_WARN()
 [drm] *ERROR* Tile0: GT0: this is xe_gt_err()
 [drm] *ERROR* Tile0: GT0: this is xe_gt_err_printer()
 [drm] Tile0: GT0: this is xe_gt_info()
 [drm] Tile0: GT0: this is xe_gt_info_printer()
 [drm:printk_demo.cold] Tile0: GT0: this is xe_gt_dbg()
 [drm:printk_demo.cold] Tile0: GT0: this is xe_gt_dbg_printer()

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250909165941.31730-5-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_gt_printk.h   |  11 ++-
 drivers/gpu/drm/xe/xe_printk.h      | 129 ++++++++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_tile_printk.h | 127 +++++++++++++++++++++++++++
 3 files changed, 261 insertions(+), 6 deletions(-)
 create mode 100644 drivers/gpu/drm/xe/xe_printk.h
 create mode 100644 drivers/gpu/drm/xe/xe_tile_printk.h

diff --git a/drivers/gpu/drm/xe/xe_gt_printk.h b/drivers/gpu/drm/xe/xe_gt_printk.h
index dcc45628cca9..1313d32862db 100644
--- a/drivers/gpu/drm/xe/xe_gt_printk.h
+++ b/drivers/gpu/drm/xe/xe_gt_printk.h
@@ -6,14 +6,13 @@
 #ifndef _XE_GT_PRINTK_H_
 #define _XE_GT_PRINTK_H_
 
-#include <drm/drm_print.h>
-
 #include "xe_gt_types.h"
+#include "xe_tile_printk.h"
 
 #define __XE_GT_PRINTK_FMT(_gt, _fmt, _args...)	"GT%u: " _fmt, (_gt)->info.id, ##_args
 
 #define xe_gt_printk(_gt, _level, _fmt, ...) \
-	drm_##_level(&gt_to_xe(_gt)->drm, __XE_GT_PRINTK_FMT((_gt), _fmt, ##__VA_ARGS__))
+	xe_tile_printk((_gt)->tile, _level, __XE_GT_PRINTK_FMT((_gt), _fmt, ##__VA_ARGS__))
 
 #define xe_gt_err(_gt, _fmt, ...) \
 	xe_gt_printk((_gt), err, _fmt, ##__VA_ARGS__)
@@ -37,7 +36,7 @@
 	xe_gt_printk((_gt), dbg, _fmt, ##__VA_ARGS__)
 
 #define xe_gt_WARN_type(_gt, _type, _condition, _fmt, ...) \
-	drm_WARN##_type(&gt_to_xe(_gt)->drm, _condition, _fmt, ## __VA_ARGS__)
+	xe_tile_WARN##_type((_gt)->tile, _condition, _fmt, ## __VA_ARGS__)
 
 #define xe_gt_WARN(_gt, _condition, _fmt, ...) \
 	xe_gt_WARN_type((_gt),, _condition, __XE_GT_PRINTK_FMT((_gt), _fmt, ##__VA_ARGS__))
@@ -72,9 +71,9 @@ static inline void __xe_gt_printfn_dbg(struct drm_printer *p, struct va_format *
 
 	/*
 	 * The original xe_gt_dbg() callsite annotations are useless here,
-	 * redirect to the tweaked drm_dbg_printer() instead.
+	 * redirect to the tweaked xe_tile_dbg_printer() instead.
 	 */
-	dbg = drm_dbg_printer(&gt_to_xe(gt)->drm, DRM_UT_DRIVER, NULL);
+	dbg = xe_tile_dbg_printer((gt)->tile);
 	dbg.origin = p->origin;
 
 	drm_printf(&dbg, __XE_GT_PRINTK_FMT(gt, "%pV", vaf));
diff --git a/drivers/gpu/drm/xe/xe_printk.h b/drivers/gpu/drm/xe/xe_printk.h
new file mode 100644
index 000000000000..c5be2385aa95
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_printk.h
@@ -0,0 +1,129 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_PRINTK_H_
+#define _XE_PRINTK_H_
+
+#include <drm/drm_print.h>
+
+#include "xe_device_types.h"
+
+#define __XE_PRINTK_FMT(_xe, _fmt, _args...)	_fmt, ##_args
+
+#define xe_printk(_xe, _level, _fmt, ...) \
+	drm_##_level(&(_xe)->drm, __XE_PRINTK_FMT((_xe), _fmt, ## __VA_ARGS__))
+
+#define xe_err(_xe, _fmt, ...) \
+	xe_printk((_xe), err, _fmt, ##__VA_ARGS__)
+
+#define xe_err_once(_xe, _fmt, ...) \
+	xe_printk((_xe), err_once, _fmt, ##__VA_ARGS__)
+
+#define xe_err_ratelimited(_xe, _fmt, ...) \
+	xe_printk((_xe), err_ratelimited, _fmt, ##__VA_ARGS__)
+
+#define xe_warn(_xe, _fmt, ...) \
+	xe_printk((_xe), warn, _fmt, ##__VA_ARGS__)
+
+#define xe_notice(_xe, _fmt, ...) \
+	xe_printk((_xe), notice, _fmt, ##__VA_ARGS__)
+
+#define xe_info(_xe, _fmt, ...) \
+	xe_printk((_xe), info, _fmt, ##__VA_ARGS__)
+
+#define xe_dbg(_xe, _fmt, ...) \
+	xe_printk((_xe), dbg, _fmt, ##__VA_ARGS__)
+
+#define xe_WARN_type(_xe, _type, _condition, _fmt, ...) \
+	drm_WARN##_type(&(_xe)->drm, _condition, _fmt, ## __VA_ARGS__)
+
+#define xe_WARN(_xe, _condition, _fmt, ...) \
+	xe_WARN_type((_xe),, _condition, __XE_PRINTK_FMT((_xe), _fmt, ## __VA_ARGS__))
+
+#define xe_WARN_ONCE(_xe, _condition, _fmt, ...) \
+	xe_WARN_type((_xe), _ONCE, _condition, __XE_PRINTK_FMT((_xe), _fmt, ## __VA_ARGS__))
+
+#define xe_WARN_ON(_xe, _condition) \
+	xe_WARN((_xe), _condition, "%s(%s)", "WARN_ON", __stringify(_condition))
+
+#define xe_WARN_ON_ONCE(_xe, _condition) \
+	xe_WARN_ONCE((_xe), _condition, "%s(%s)", "WARN_ON_ONCE", __stringify(_condition))
+
+static inline void __xe_printfn_err(struct drm_printer *p, struct va_format *vaf)
+{
+	struct xe_device *xe = p->arg;
+
+	xe_err(xe, "%pV", vaf);
+}
+
+static inline void __xe_printfn_info(struct drm_printer *p, struct va_format *vaf)
+{
+	struct xe_device *xe = p->arg;
+
+	xe_info(xe, "%pV", vaf);
+}
+
+static inline void __xe_printfn_dbg(struct drm_printer *p, struct va_format *vaf)
+{
+	struct xe_device *xe = p->arg;
+	struct drm_printer ddp;
+
+	/*
+	 * The original xe_dbg() callsite annotations are useless here,
+	 * redirect to the tweaked drm_dbg_printer() instead.
+	 */
+	ddp = drm_dbg_printer(&xe->drm, DRM_UT_DRIVER, NULL);
+	ddp.origin = p->origin;
+
+	drm_printf(&ddp, __XE_PRINTK_FMT(xe, "%pV", vaf));
+}
+
+/**
+ * xe_err_printer - Construct a &drm_printer that outputs to xe_err()
+ * @xe: the &xe_device pointer to use in xe_err()
+ *
+ * Return: The &drm_printer object.
+ */
+static inline struct drm_printer xe_err_printer(struct xe_device *xe)
+{
+	struct drm_printer p = {
+		.printfn = __xe_printfn_err,
+		.arg = xe,
+	};
+	return p;
+}
+
+/**
+ * xe_info_printer - Construct a &drm_printer that outputs to xe_info()
+ * @xe: the &xe_device pointer to use in xe_info()
+ *
+ * Return: The &drm_printer object.
+ */
+static inline struct drm_printer xe_info_printer(struct xe_device *xe)
+{
+	struct drm_printer p = {
+		.printfn = __xe_printfn_info,
+		.arg = xe,
+	};
+	return p;
+}
+
+/**
+ * xe_dbg_printer - Construct a &drm_printer that outputs like xe_dbg()
+ * @xe: the &xe_device pointer to use in xe_dbg()
+ *
+ * Return: The &drm_printer object.
+ */
+static inline struct drm_printer xe_dbg_printer(struct xe_device *xe)
+{
+	struct drm_printer p = {
+		.printfn = __xe_printfn_dbg,
+		.arg = xe,
+		.origin = (const void *)_THIS_IP_,
+	};
+	return p;
+}
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_tile_printk.h b/drivers/gpu/drm/xe/xe_tile_printk.h
new file mode 100644
index 000000000000..63640a42685d
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_tile_printk.h
@@ -0,0 +1,127 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _xe_tile_printk_H_
+#define _xe_tile_printk_H_
+
+#include "xe_printk.h"
+
+#define __XE_TILE_PRINTK_FMT(_tile, _fmt, _args...)	"Tile%u: " _fmt, (_tile)->id, ##_args
+
+#define xe_tile_printk(_tile, _level, _fmt, ...) \
+	xe_printk((_tile)->xe, _level, __XE_TILE_PRINTK_FMT((_tile), _fmt, ##__VA_ARGS__))
+
+#define xe_tile_err(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), err, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_err_once(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), err_once, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_err_ratelimited(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), err_ratelimited, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_warn(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), warn, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_notice(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), notice, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_info(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), info, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_dbg(_tile, _fmt, ...) \
+	xe_tile_printk((_tile), dbg, _fmt, ##__VA_ARGS__)
+
+#define xe_tile_WARN_type(_tile, _type, _condition, _fmt, ...) \
+	xe_WARN##_type((_tile)->xe, _condition, _fmt, ## __VA_ARGS__)
+
+#define xe_tile_WARN(_tile, _condition, _fmt, ...) \
+	xe_tile_WARN_type((_tile),, _condition, __XE_TILE_PRINTK_FMT((_tile), _fmt, ##__VA_ARGS__))
+
+#define xe_tile_WARN_ONCE(_tile, _condition, _fmt, ...) \
+	xe_tile_WARN_type((_tile), _ONCE, _condition, __XE_TILE_PRINTK_FMT((_tile), _fmt, ##__VA_ARGS__))
+
+#define xe_tile_WARN_ON(_tile, _condition) \
+	xe_tile_WARN((_tile), _condition, "%s(%s)", "WARN_ON", __stringify(_condition))
+
+#define xe_tile_WARN_ON_ONCE(_tile, _condition) \
+	xe_tile_WARN_ONCE((_tile), _condition, "%s(%s)", "WARN_ON_ONCE", __stringify(_condition))
+
+static inline void __xe_tile_printfn_err(struct drm_printer *p, struct va_format *vaf)
+{
+	struct xe_tile *tile = p->arg;
+
+	xe_tile_err(tile, "%pV", vaf);
+}
+
+static inline void __xe_tile_printfn_info(struct drm_printer *p, struct va_format *vaf)
+{
+	struct xe_tile *tile = p->arg;
+
+	xe_tile_info(tile, "%pV", vaf);
+}
+
+static inline void __xe_tile_printfn_dbg(struct drm_printer *p, struct va_format *vaf)
+{
+	struct xe_tile *tile = p->arg;
+	struct drm_printer dbg;
+
+	/*
+	 * The original xe_tile_dbg() callsite annotations are useless here,
+	 * redirect to the tweaked xe_dbg_printer() instead.
+	 */
+	dbg = xe_dbg_printer(tile->xe);
+	dbg.origin = p->origin;
+
+	drm_printf(&dbg, __XE_TILE_PRINTK_FMT(tile, "%pV", vaf));
+}
+
+/**
+ * xe_tile_err_printer - Construct a &drm_printer that outputs to xe_tile_err()
+ * @tile: the &xe_tile pointer to use in xe_tile_err()
+ *
+ * Return: The &drm_printer object.
+ */
+static inline struct drm_printer xe_tile_err_printer(struct xe_tile *tile)
+{
+	struct drm_printer p = {
+		.printfn = __xe_tile_printfn_err,
+		.arg = tile,
+	};
+	return p;
+}
+
+/**
+ * xe_tile_info_printer - Construct a &drm_printer that outputs to xe_tile_info()
+ * @tile: the &xe_tile pointer to use in xe_tile_info()
+ *
+ * Return: The &drm_printer object.
+ */
+static inline struct drm_printer xe_tile_info_printer(struct xe_tile *tile)
+{
+	struct drm_printer p = {
+		.printfn = __xe_tile_printfn_info,
+		.arg = tile,
+	};
+	return p;
+}
+
+/**
+ * xe_tile_dbg_printer - Construct a &drm_printer that outputs like xe_tile_dbg()
+ * @tile: the &xe_tile pointer to use in xe_tile_dbg()
+ *
+ * Return: The &drm_printer object.
+ */
+static inline struct drm_printer xe_tile_dbg_printer(struct xe_tile *tile)
+{
+	struct drm_printer p = {
+		.printfn = __xe_tile_printfn_dbg,
+		.arg = tile,
+		.origin = (const void *)_THIS_IP_,
+	};
+	return p;
+}
+
+#endif

From 01ecf00463975f5fc782b214cf3c34b7d8c4b74c Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 9 Sep 2025 18:59:39 +0200
Subject: [PATCH 1222/1455] drm/xe: Use tile-oriented messages in GGTT code

Use recently added macros to print tile-oriented messages.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250909165941.31730-6-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_ggtt.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c
index 468364c09d35..7fdd0a97a628 100644
--- a/drivers/gpu/drm/xe/xe_ggtt.c
+++ b/drivers/gpu/drm/xe/xe_ggtt.c
@@ -28,6 +28,7 @@
 #include "xe_pm.h"
 #include "xe_res_cursor.h"
 #include "xe_sriov.h"
+#include "xe_tile_printk.h"
 #include "xe_tile_sriov_vf.h"
 #include "xe_tlb_inval.h"
 #include "xe_wa.h"
@@ -269,7 +270,7 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt)
 		gsm_size = probe_gsm_size(pdev);
 
 	if (gsm_size == 0) {
-		drm_err(&xe->drm, "Hardware reported no preallocated GSM\n");
+		xe_tile_err(ggtt->tile, "Hardware reported no preallocated GSM\n");
 		return -ENOMEM;
 	}
 
@@ -466,8 +467,8 @@ static void xe_ggtt_dump_node(struct xe_ggtt *ggtt,
 
 	if (IS_ENABLED(CONFIG_DRM_XE_DEBUG)) {
 		string_get_size(node->size, 1, STRING_UNITS_2, buf, sizeof(buf));
-		xe_gt_dbg(ggtt->tile->primary_gt, "GGTT %#llx-%#llx (%s) %s\n",
-			  node->start, node->start + node->size, buf, description);
+		xe_tile_dbg(ggtt->tile, "GGTT %#llx-%#llx (%s) %s\n",
+			    node->start, node->start + node->size, buf, description);
 	}
 }
 
@@ -499,9 +500,8 @@ int xe_ggtt_node_insert_balloon_locked(struct xe_ggtt_node *node, u64 start, u64
 
 	err = drm_mm_reserve_node(&ggtt->mm, &node->base);
 
-	if (xe_gt_WARN(ggtt->tile->primary_gt, err,
-		       "Failed to balloon GGTT %#llx-%#llx (%pe)\n",
-		       node->base.start, node->base.start + node->base.size, ERR_PTR(err)))
+	if (xe_tile_WARN(ggtt->tile, err, "Failed to balloon GGTT %#llx-%#llx (%pe)\n",
+			 node->base.start, node->base.start + node->base.size, ERR_PTR(err)))
 		return err;
 
 	xe_ggtt_dump_node(ggtt, &node->base, "balloon");

From fed1a9d60fa39faa9b10b927681d85a81bcbe797 Mon Sep 17 00:00:00 2001
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Date: Thu, 11 Sep 2025 17:20:25 -0400
Subject: [PATCH 1223/1455] drm/xe: Fix circular locking dependency

Fix this:

 ======================================================
 WARNING: possible circular locking dependency detected
 6.17.0-rc4-lgci-xe-xe-pw-153723v2+ #1 Tainted: G S   U
 ------------------------------------------------------
 xe_pm/11324 is trying to acquire lock:
 ffff8881085f22a0 (&pc->freq_lock){+.+.}-{3:3}, at:
 		  xe_guc_pc_start+0x39f/0xf70 [xe]

but task is already holding lock:

 ffffffffa1020420 (xe_rpm_nod3cold_map){+.+.}-{0:0}, at:
 		  xe_rpm_lockmap_acquire+0x1a/0x70 [xe]

which lock already depends on the new lock.

Possible unsafe locking scenario:
      CPU0                    CPU1
      ----                    ----
 lock(xe_rpm_nod3cold_map);
                              lock(&pc->freq_lock);
                              lock(xe_rpm_nod3cold_map);
 lock(&pc->freq_lock);

Reported-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6122
Fixes: 60d2b7899142 ("drm/xe/guc: Add SLPC power profile interface")
Cc: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Reviewed-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Tested-by: Vinay Belgaumkar <vinay.belgaumkar@intel.com>
Link: https://lore.kernel.org/r/20250911212024.966757-2-rodrigo.vivi@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_gt_freq.c | 2 ++
 drivers/gpu/drm/xe/xe_guc_pc.c  | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_gt_freq.c b/drivers/gpu/drm/xe/xe_gt_freq.c
index 781e4890fb26..4ff1b6b58d6b 100644
--- a/drivers/gpu/drm/xe/xe_gt_freq.c
+++ b/drivers/gpu/drm/xe/xe_gt_freq.c
@@ -246,7 +246,9 @@ static ssize_t power_profile_store(struct kobject *kobj,
 	struct xe_guc_pc *pc = dev_to_pc(dev);
 	int err;
 
+	xe_pm_runtime_get(dev_to_xe(dev));
 	err = xe_guc_pc_set_power_profile(pc, buff);
+	xe_pm_runtime_put(dev_to_xe(dev));
 
 	return err ?: count;
 }
diff --git a/drivers/gpu/drm/xe/xe_guc_pc.c b/drivers/gpu/drm/xe/xe_guc_pc.c
index 68a5bf8e3946..53fdf59524c4 100644
--- a/drivers/gpu/drm/xe/xe_guc_pc.c
+++ b/drivers/gpu/drm/xe/xe_guc_pc.c
@@ -1215,7 +1215,7 @@ int xe_guc_pc_set_power_profile(struct xe_guc_pc *pc, const char *buf)
 		return -EINVAL;
 
 	guard(mutex)(&pc->freq_lock);
-	xe_pm_runtime_get(pc_to_xe(pc));
+	xe_pm_runtime_get_noresume(pc_to_xe(pc));
 
 	ret = pc_action_set_param(pc,
 				  SLPC_PARAM_POWER_PROFILE,

From 2a810401aa054513f8f740af5e381c9e4f651d0d Mon Sep 17 00:00:00 2001
From: Harish Chegondi <harish.chegondi@intel.com>
Date: Wed, 10 Sep 2025 18:27:41 -0700
Subject: [PATCH 1224/1455] drm/xe/xe3: Extend Wa_18041344222 to graphics IP
 versions 30.00 and 30.01

Apply WA 18041344222 to Xe3 LPG graphics IP versions 30.00 and 30.01 too.

Bspec: 56024
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Matt Atwood <matthew.s.atwood@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/7368f8059013424ac94f4a01c23f9c98a37b06dc.1757552915.git.harish.chegondi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_wa.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index c1fec526bed0..fa69bc7730bb 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -679,6 +679,13 @@ static const struct xe_rtp_entry_sr engine_was[] = {
 	  XE_RTP_RULES(GRAPHICS_VERSION(3003), FUNC(xe_rtp_match_first_render_or_compute)),
 	  XE_RTP_ACTIONS(SET(HALF_SLICE_CHICKEN7, CLEAR_OPTIMIZATION_DISABLE))
 	},
+	{ XE_RTP_NAME("18041344222"),
+	  XE_RTP_RULES(GRAPHICS_VERSION_RANGE(3000, 3001),
+		       FUNC(xe_rtp_match_first_render_or_compute),
+		       FUNC(xe_rtp_match_not_sriov_vf),
+		       FUNC(xe_rtp_match_gt_has_discontiguous_dss_groups)),
+	  XE_RTP_ACTIONS(SET(TDL_CHICKEN, EUSTALL_PERF_SAMPLING_DISABLE))
+	},
 };
 
 static const struct xe_rtp_entry_sr lrc_was[] = {

From 9e0b0fd5311ef68638abcd05306233b367c6b407 Mon Sep 17 00:00:00 2001
From: Colin Ian King <colin.i.king@gmail.com>
Date: Fri, 12 Sep 2025 08:43:30 +0100
Subject: [PATCH 1225/1455] drm/xe/guc: Fix spelling mistake "sheduling" ->
 "scheduling"

There is a spelling mistake in a xe_gt_err error message. Fix it.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Reviewed-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20250912074330.1275279-1-colin.i.king@gmail.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_guc_submit.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index a465594b61dc..8cb8b93a37a5 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -355,7 +355,7 @@ static int guc_init_global_schedule_policy(struct xe_guc *guc)
 		ret = xe_guc_ct_send_block(&guc->ct, data, count);
 		if (ret < 0) {
 			xe_gt_err(guc_to_gt(guc),
-				  "failed to enable GuC sheduling policies: %pe\n",
+				  "failed to enable GuC scheduling policies: %pe\n",
 				  ERR_PTR(ret));
 			return ret;
 		}

From 4e1d3b5e6423dc841acd8691d75626b3d3b2b6a8 Mon Sep 17 00:00:00 2001
From: Mallesh Koujalagi <mallesh.koujalagi@intel.com>
Date: Fri, 12 Sep 2025 17:04:58 +0530
Subject: [PATCH 1226/1455] drm/xe/hwmon: Remove type casting

Refactor: eliminate type casts by using proper u32
declarations.

v2:
- Address review comments. (Karthik)

v3:
- Use the proper u32 type and drop cast. (Lucas De Marchi)
- Modify variable when actually using u64 value.
- Change r value to reg_value with u32 type.

v4:
- Remove newline between trailer and Signed-off-by. (Lucas De Marchi)
- Change reg_val to val for more user-friendly logging.
- Use mul_u32_u32 function since both values are u32.

v5:
- mul_u32_u32 function with shift. (Lucas De Marchi)

Fixes: 7596d839f6228 ("drm/xe/hwmon: Add support to manage power limits though mailbox")
Signed-off-by: Mallesh Koujalagi <mallesh.koujalagi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250912113458.2815172-1-mallesh.koujalagi@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_hwmon.c | 35 +++++++++++++++++++----------------
 1 file changed, 19 insertions(+), 16 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_hwmon.c b/drivers/gpu/drm/xe/xe_hwmon.c
index e65382d4426a..b6790589e623 100644
--- a/drivers/gpu/drm/xe/xe_hwmon.c
+++ b/drivers/gpu/drm/xe/xe_hwmon.c
@@ -286,7 +286,7 @@ static struct xe_reg xe_hwmon_get_reg(struct xe_hwmon *hwmon, enum xe_hwmon_reg
  */
 static void xe_hwmon_power_max_read(struct xe_hwmon *hwmon, u32 attr, int channel, long *value)
 {
-	u64 reg_val = 0, min, max;
+	u32 reg_val = 0;
 	struct xe_device *xe = hwmon->xe;
 	struct xe_reg rapl_limit, pkg_power_sku;
 	struct xe_mmio *mmio = xe_root_tile_mmio(xe);
@@ -294,7 +294,7 @@ static void xe_hwmon_power_max_read(struct xe_hwmon *hwmon, u32 attr, int channe
 	mutex_lock(&hwmon->hwmon_lock);
 
 	if (hwmon->xe->info.has_mbx_power_limits) {
-		xe_hwmon_pcode_read_power_limit(hwmon, attr, channel, (u32 *)&reg_val);
+		xe_hwmon_pcode_read_power_limit(hwmon, attr, channel, &reg_val);
 	} else {
 		rapl_limit = xe_hwmon_get_reg(hwmon, REG_PKG_RAPL_LIMIT, channel);
 		pkg_power_sku = xe_hwmon_get_reg(hwmon, REG_PKG_POWER_SKU, channel);
@@ -304,19 +304,21 @@ static void xe_hwmon_power_max_read(struct xe_hwmon *hwmon, u32 attr, int channe
 	/* Check if PL limits are disabled. */
 	if (!(reg_val & PWR_LIM_EN)) {
 		*value = PL_DISABLE;
-		drm_info(&hwmon->xe->drm, "%s disabled for channel %d, val 0x%016llx\n",
+		drm_info(&hwmon->xe->drm, "%s disabled for channel %d, val 0x%08x\n",
 			 PWR_ATTR_TO_STR(attr), channel, reg_val);
 		goto unlock;
 	}
 
 	reg_val = REG_FIELD_GET(PWR_LIM_VAL, reg_val);
-	*value = mul_u64_u32_shr(reg_val, SF_POWER, hwmon->scl_shift_power);
+	*value = mul_u32_u32(reg_val, SF_POWER) >> hwmon->scl_shift_power;
 
 	/* For platforms with mailbox power limit support clamping would be done by pcode. */
 	if (!hwmon->xe->info.has_mbx_power_limits) {
-		reg_val = xe_mmio_read64_2x32(mmio, pkg_power_sku);
-		min = REG_FIELD_GET(PKG_MIN_PWR, reg_val);
-		max = REG_FIELD_GET(PKG_MAX_PWR, reg_val);
+		u64 pkg_pwr, min, max;
+
+		pkg_pwr = xe_mmio_read64_2x32(mmio, pkg_power_sku);
+		min = REG_FIELD_GET(PKG_MIN_PWR, pkg_pwr);
+		max = REG_FIELD_GET(PKG_MAX_PWR, pkg_pwr);
 		min = mul_u64_u32_shr(min, SF_POWER, hwmon->scl_shift_power);
 		max = mul_u64_u32_shr(max, SF_POWER, hwmon->scl_shift_power);
 		if (min && max)
@@ -493,8 +495,8 @@ xe_hwmon_power_max_interval_show(struct device *dev, struct device_attribute *at
 {
 	struct xe_hwmon *hwmon = dev_get_drvdata(dev);
 	struct xe_mmio *mmio = xe_root_tile_mmio(hwmon->xe);
-	u32 x, y, x_w = 2; /* 2 bits */
-	u64 r, tau4, out;
+	u32 reg_val, x, y, x_w = 2; /* 2 bits */
+	u64 tau4, out;
 	int channel = (to_sensor_dev_attr(attr)->index % 2) ? CHANNEL_PKG : CHANNEL_CARD;
 	u32 power_attr = (to_sensor_dev_attr(attr)->index > 1) ? PL2_HWMON_ATTR : PL1_HWMON_ATTR;
 
@@ -505,23 +507,24 @@ xe_hwmon_power_max_interval_show(struct device *dev, struct device_attribute *at
 	mutex_lock(&hwmon->hwmon_lock);
 
 	if (hwmon->xe->info.has_mbx_power_limits) {
-		ret = xe_hwmon_pcode_read_power_limit(hwmon, power_attr, channel, (u32 *)&r);
+		ret = xe_hwmon_pcode_read_power_limit(hwmon, power_attr, channel, &reg_val);
 		if (ret) {
 			drm_err(&hwmon->xe->drm,
-				"power interval read fail, ch %d, attr %d, r 0%llx, ret %d\n",
-				channel, power_attr, r, ret);
-			r = 0;
+				"power interval read fail, ch %d, attr %d, val 0x%08x, ret %d\n",
+				channel, power_attr, reg_val, ret);
+			reg_val = 0;
 		}
 	} else {
-		r = xe_mmio_read32(mmio, xe_hwmon_get_reg(hwmon, REG_PKG_RAPL_LIMIT, channel));
+		reg_val = xe_mmio_read32(mmio, xe_hwmon_get_reg(hwmon, REG_PKG_RAPL_LIMIT,
+								channel));
 	}
 
 	mutex_unlock(&hwmon->hwmon_lock);
 
 	xe_pm_runtime_put(hwmon->xe);
 
-	x = REG_FIELD_GET(PWR_LIM_TIME_X, r);
-	y = REG_FIELD_GET(PWR_LIM_TIME_Y, r);
+	x = REG_FIELD_GET(PWR_LIM_TIME_X, reg_val);
+	y = REG_FIELD_GET(PWR_LIM_TIME_Y, reg_val);
 
 	/*
 	 * tau = (1 + (x / 4)) * power(2,y), x = bits(23:22), y = bits(21:17)

From c34f9868dfb3888e1475e98b0bf112c2add2ddcd Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 11 Sep 2025 12:36:25 -0700
Subject: [PATCH 1227/1455] drm/xe: Update workaround documentation

Bring it up to reality, better documenting the existing batch buffers,
OOB rules and fixing some typos.

Bspec: 60122
Reviewed-by: Stuart Summers <stuart.summers@intel.com>
Link: https://lore.kernel.org/r/20250911-wa-bb-cmds-v4-1-c8f7e48f7eae@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_wa.c | 45 +++++++++++++++++++++++++++-----------
 1 file changed, 32 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_wa.c b/drivers/gpu/drm/xe/xe_wa.c
index fa69bc7730bb..cd03891654a1 100644
--- a/drivers/gpu/drm/xe/xe_wa.c
+++ b/drivers/gpu/drm/xe/xe_wa.c
@@ -39,7 +39,8 @@
  *   Register Immediate commands) once when initializing the device and saved in
  *   the default context. That default context is then used on every context
  *   creation to have a "primed golden context", i.e. a context image that
- *   already contains the changes needed to all the registers.
+ *   already contains the changes needed to all the registers. See
+ *   drivers/gpu/drm/xe/xe_lrc.c for default context handling.
  *
  * - Engine workarounds: the list of these WAs is applied whenever the specific
  *   engine is reset. It's also possible that a set of engine classes share a
@@ -48,10 +49,10 @@
  *   them need to keeep the workaround programming: the approach taken in the
  *   driver is to tie those workarounds to the first compute/render engine that
  *   is registered.  When executing with GuC submission, engine resets are
- *   outside of kernel driver control, hence the list of registers involved in
+ *   outside of kernel driver control, hence the list of registers involved is
  *   written once, on engine initialization, and then passed to GuC, that
  *   saves/restores their values before/after the reset takes place. See
- *   ``drivers/gpu/drm/xe/xe_guc_ads.c`` for reference.
+ *   drivers/gpu/drm/xe/xe_guc_ads.c for reference.
  *
  * - GT workarounds: the list of these WAs is applied whenever these registers
  *   revert to their default values: on GPU reset, suspend/resume [1]_, etc.
@@ -66,21 +67,39 @@
  *   hardware on every HW context restore. These buffers are created and
  *   programmed in the default context so the hardware always go through those
  *   programming sequences when switching contexts. The support for workaround
- *   batchbuffers is enabled these hardware mechanisms:
+ *   batchbuffers is enabled via these hardware mechanisms:
  *
- *   #. INDIRECT_CTX: A batchbuffer and an offset are provided in the default
- *      context, pointing the hardware to jump to that location when that offset
- *      is reached in the context restore. Workaround batchbuffer in the driver
- *      currently uses this mechanism for all platforms.
+ *   #. INDIRECT_CTX (also known as **mid context restore bb**): A batchbuffer
+ *      and an offset are provided in the default context, pointing the hardware
+ *      to jump to that location when that offset is reached in the context
+ *      restore.  When a context is being restored, this is executed after the
+ *      ring context, in the middle (or beginning) of the engine context image.
  *
- *   #. BB_PER_CTX_PTR: A batchbuffer is provided in the default context,
- *      pointing the hardware to a buffer to continue executing after the
- *      engine registers are restored in a context restore sequence. This is
- *      currently not used in the driver.
+ *   #. BB_PER_CTX_PTR (also known as **post context restore bb**): A
+ *      batchbuffer is provided in the default context, pointing the hardware to
+ *      a buffer to continue executing after the engine registers are restored
+ *      in a context restore sequence.
+ *
+ *   Below is the timeline for a context restore sequence:
+ *
+ *   .. code::
+ *
+ *                        INDIRECT_CTX_OFFSET
+ *                   |----------->|
+ *      .------------.------------.-------------.------------.--------------.-----------.
+ *      |Ring        | Engine     | Mid-context | Engine     | Post-context | Ring      |
+ *      |Restore     | Restore (1)| BB Restore  | Restore (2)| BB Restore   | Execution |
+ *      `------------'------------'-------------'------------'--------------'-----------'
  *
  * - Other/OOB:  There are WAs that, due to their nature, cannot be applied from
  *   a central place. Those are peppered around the rest of the code, as needed.
- *   Workarounds related to the display IP are the main example.
+ *   There's a central place to control which workarounds are enabled:
+ *   drivers/gpu/drm/xe/xe_wa_oob.rules for GT workarounds and
+ *   drivers/gpu/drm/xe/xe_device_wa_oob.rules for device/SoC workarounds.
+ *   These files only record which workarounds are enabled: during early device
+ *   initialization those rules are evaluated and recorded by the driver. Then
+ *   later the driver checks with ``XE_GT_WA()`` and ``XE_DEVICE_WA()`` to
+ *   implement them.
  *
  * .. [1] Technically, some registers are powercontext saved & restored, so they
  *    survive a suspend/resume. In practice, writing them again is not too

From 2ec29456254822f638bd9673a225ea0eb2245382 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Thu, 11 Sep 2025 12:36:26 -0700
Subject: [PATCH 1228/1455] drm/xe/configfs: Fix documentation warning

Fix this warning while building the documentation:

	Documentation/gpu/xe/xe_configfs:9: drivers/gpu/drm/xe/xe_configfs.c:138:
	WARNING: Definition list ends without a blank line; unexpected unindent.

That also makes it better formatted in the output.

While at it, also fix the underline length in "Overview".

Fixes: e2b33fce5eb0 ("drm/xe/configfs: Improve documentation steps")
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250911-wa-bb-cmds-v4-2-c8f7e48f7eae@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index e515fa7085fa..e52808e3199f 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -21,7 +21,7 @@
  * DOC: Xe Configfs
  *
  * Overview
- * =========
+ * ========
  *
  * Configfs is a filesystem-based manager of kernel objects. XE KMD registers a
  * configfs subsystem called ``xe`` that creates a directory in the mounted
@@ -34,7 +34,7 @@
  *
  * To create a device, the ``xe`` module should already be loaded, but some
  * attributes can only be set before binding the device. It can be accomplished
- * by blocking the driver autoprobe:
+ * by blocking the driver autoprobe::
  *
  *	# echo 0 > /sys/bus/pci/drivers_autoprobe
  *	# modprobe xe

From d4c3ed963e41d488695cf91068eabb8eb9f538ec Mon Sep 17 00:00:00 2001
From: Nitin Gote <nitin.r.gote@intel.com>
Date: Thu, 11 Sep 2025 10:58:23 +0530
Subject: [PATCH 1229/1455] drm/xe: defer free of NVM auxiliary container to
 device release callback

Do not kfree the intel_dg_nvm_dev in xe_nvm_fini() right after
auxiliary_device_delete/uninit. The auxiliary_device embeds the
device/kobject (and its name); freeing it too early can race
with asynchronous device_del/udev processing and cause a use-after-free.

Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
Fixes: c28bfb107dac ("drm/xe/nvm: add on-die non-volatile memory device")
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250911052823.226696-1-nitin.r.gote@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_nvm.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_nvm.c b/drivers/gpu/drm/xe/xe_nvm.c
index 9ca4a5ae1693..33f4ac82fc80 100644
--- a/drivers/gpu/drm/xe/xe_nvm.c
+++ b/drivers/gpu/drm/xe/xe_nvm.c
@@ -35,6 +35,10 @@ static const struct intel_dg_nvm_region regions[INTEL_DG_NVM_REGIONS] = {
 
 static void xe_nvm_release_dev(struct device *dev)
 {
+	struct auxiliary_device *aux = container_of(dev, struct auxiliary_device, dev);
+	struct intel_dg_nvm_dev *nvm = container_of(aux, struct intel_dg_nvm_dev, aux_dev);
+
+	kfree(nvm);
 }
 
 static bool xe_nvm_non_posted_erase(struct xe_device *xe)
@@ -162,6 +166,5 @@ void xe_nvm_fini(struct xe_device *xe)
 
 	auxiliary_device_delete(&nvm->aux_dev);
 	auxiliary_device_uninit(&nvm->aux_dev);
-	kfree(nvm);
 	xe->nvm = NULL;
 }

From f0fbbff7e3082b0e1065aef83cc1418ed2cb1b69 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:14 +0900
Subject: [PATCH 1230/1455] gpu: nova-core: require `Send` on `FalconEngine`
 and `FalconHal`

We want to store the GSP and SEC2 falcon instances inside the `Gpu`
structure, but doing so require these types to implement `Send` for
`pci::Driver` to remain implementable on `NovaCore`, which embeds `Gpu`.

All implementors of `FalconEngine` and `FalconHal` satisfy the
requirements of `Send`, and these traits also already required `Sync`,
so this a minor tweak.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-1-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/falcon.rs     | 2 +-
 drivers/gpu/nova-core/falcon/hal.rs | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index 2dbcdf26697b..b16207e7242f 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -286,7 +286,7 @@ pub(crate) struct PFalcon2Base(());
 /// Each engine provides one base for `PFALCON` and `PFALCON2` registers. The `ID` constant is used
 /// to identify a given Falcon instance with register I/O methods.
 pub(crate) trait FalconEngine:
-    Sync + RegisterBase<PFalconBase> + RegisterBase<PFalcon2Base> + Sized
+    Send + Sync + RegisterBase<PFalconBase> + RegisterBase<PFalcon2Base> + Sized
 {
     /// Singleton of the engine, used to identify it with register I/O methods.
     const ID: Self;
diff --git a/drivers/gpu/nova-core/falcon/hal.rs b/drivers/gpu/nova-core/falcon/hal.rs
index b233bc365882..bba288455617 100644
--- a/drivers/gpu/nova-core/falcon/hal.rs
+++ b/drivers/gpu/nova-core/falcon/hal.rs
@@ -13,7 +13,7 @@ mod ga102;
 /// Implements chipset-specific low-level operations. The trait is generic against [`FalconEngine`]
 /// so its `BASE` parameter can be used in order to avoid runtime bound checks when accessing
 /// registers.
-pub(crate) trait FalconHal<E: FalconEngine>: Sync {
+pub(crate) trait FalconHal<E: FalconEngine>: Send + Sync {
     /// Activates the Falcon core if the engine is a risvc/falcon dual engine.
     fn select_core(&self, _falcon: &Falcon<E>, _bar: &Bar0) -> Result {
         Ok(())

From e7c96980ea4d93e79b43b07c5489d700207b0055 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:15 +0900
Subject: [PATCH 1231/1455] gpu: nova-core: move GSP boot code to its own
 module

Right now the GSP boot code is very incomplete and limited to running
FRTS, so having it in `Gpu::new` is not a big constraint.

However, this will change as we add more steps of the GSP boot process,
and not all GPU families follow the same procedure, so having these
steps in a dedicated method is the logical construct.

There is also the fact the GSP will require its own runtime data, and
while it won't immediately need to be pinned, we want to be ready for
the time where it will - most likely when it starts using mutexes.

Thus, add an empty `Gsp` type that is pinned inside `Gpu` and
initialized using a pin initializer. This sets the constraint we need to
observe from the start, and could spare us some costly refactoring down
the road.

Then, move the code related to GSP boot to the `gsp::boot` module, as
part of the `Gsp` implementation.

Doing so allows us to make `Gpu::new` return a fallible `impl PinInit`
instead of a `Result.` This is more idiomatic when working with pinned
objects, and sets up the pinned initialization pattern we want to
preserve as the code grows more complex.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-2-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/driver.rs    |   9 +-
 drivers/gpu/nova-core/gpu.rs       | 159 ++++++++---------------------
 drivers/gpu/nova-core/gsp.rs       |  17 +++
 drivers/gpu/nova-core/gsp/boot.rs  | 118 +++++++++++++++++++++
 drivers/gpu/nova-core/nova_core.rs |   1 +
 5 files changed, 184 insertions(+), 120 deletions(-)
 create mode 100644 drivers/gpu/nova-core/gsp.rs
 create mode 100644 drivers/gpu/nova-core/gsp/boot.rs

diff --git a/drivers/gpu/nova-core/driver.rs b/drivers/gpu/nova-core/driver.rs
index 02b3edd7bbdc..1380b47617f7 100644
--- a/drivers/gpu/nova-core/driver.rs
+++ b/drivers/gpu/nova-core/driver.rs
@@ -34,14 +34,19 @@ impl pci::Driver for NovaCore {
         pdev.enable_device_mem()?;
         pdev.set_master();
 
-        let bar = Arc::pin_init(
+        let devres_bar = Arc::pin_init(
             pdev.iomap_region_sized::<BAR0_SIZE>(0, c_str!("nova-core/bar0")),
             GFP_KERNEL,
         )?;
 
+        // Used to provided a `&Bar0` to `Gpu::new` without tying it to the lifetime of
+        // `devres_bar`.
+        let bar_clone = Arc::clone(&devres_bar);
+        let bar = bar_clone.access(pdev.as_ref())?;
+
         let this = KBox::pin_init(
             try_pin_init!(Self {
-                gpu <- Gpu::new(pdev, bar)?,
+                gpu <- Gpu::new(pdev, devres_bar, bar),
                 _reg: auxiliary::Registration::new(
                     pdev.as_ref(),
                     c_str!("nova-drm"),
diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index a0878ecdc18b..3c019df01d30 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -3,15 +3,13 @@
 use kernel::{device, devres::Devres, error::code::*, pci, prelude::*, sync::Arc};
 
 use crate::driver::Bar0;
-use crate::falcon::{gsp::Gsp, sec2::Sec2, Falcon};
-use crate::fb::FbLayout;
+use crate::falcon::{gsp::Gsp as GspFalcon, sec2::Sec2 as Sec2Falcon, Falcon};
 use crate::fb::SysmemFlush;
-use crate::firmware::fwsec::{FwsecCommand, FwsecFirmware};
 use crate::firmware::{Firmware, FIRMWARE_VERSION};
 use crate::gfw;
+use crate::gsp::Gsp;
 use crate::regs;
 use crate::util;
-use crate::vbios::Vbios;
 use core::fmt;
 
 macro_rules! define_chipset {
@@ -172,133 +170,58 @@ pub(crate) struct Gpu {
     /// System memory page required for flushing all pending GPU-side memory writes done through
     /// PCIE into system memory, via sysmembar (A GPU-initiated HW memory-barrier operation).
     sysmem_flush: SysmemFlush,
+    /// GSP falcon instance, used for GSP boot up and cleanup.
+    gsp_falcon: Falcon<GspFalcon>,
+    /// SEC2 falcon instance, used for GSP boot up and cleanup.
+    sec2_falcon: Falcon<Sec2Falcon>,
+    /// GSP runtime data. Temporarily an empty placeholder.
+    #[pin]
+    gsp: Gsp,
 }
 
 impl Gpu {
-    /// Helper function to load and run the FWSEC-FRTS firmware and confirm that it has properly
-    /// created the WPR2 region.
-    ///
-    /// TODO: this needs to be moved into a larger type responsible for booting the whole GSP
-    /// (`GspBooter`?).
-    fn run_fwsec_frts(
-        dev: &device::Device<device::Bound>,
-        falcon: &Falcon<Gsp>,
-        bar: &Bar0,
-        bios: &Vbios,
-        fb_layout: &FbLayout,
-    ) -> Result<()> {
-        // Check that the WPR2 region does not already exists - if it does, we cannot run
-        // FWSEC-FRTS until the GPU is reset.
-        if regs::NV_PFB_PRI_MMU_WPR2_ADDR_HI::read(bar).higher_bound() != 0 {
-            dev_err!(
-                dev,
-                "WPR2 region already exists - GPU needs to be reset to proceed\n"
-            );
-            return Err(EBUSY);
-        }
-
-        let fwsec_frts = FwsecFirmware::new(
-            dev,
-            falcon,
-            bar,
-            bios,
-            FwsecCommand::Frts {
-                frts_addr: fb_layout.frts.start,
-                frts_size: fb_layout.frts.end - fb_layout.frts.start,
-            },
-        )?;
-
-        // Run FWSEC-FRTS to create the WPR2 region.
-        fwsec_frts.run(dev, falcon, bar)?;
-
-        // SCRATCH_E contains the error code for FWSEC-FRTS.
-        let frts_status = regs::NV_PBUS_SW_SCRATCH_0E_FRTS_ERR::read(bar).frts_err_code();
-        if frts_status != 0 {
-            dev_err!(
-                dev,
-                "FWSEC-FRTS returned with error code {:#x}",
-                frts_status
-            );
-
-            return Err(EIO);
-        }
-
-        // Check that the WPR2 region has been created as we requested.
-        let (wpr2_lo, wpr2_hi) = (
-            regs::NV_PFB_PRI_MMU_WPR2_ADDR_LO::read(bar).lower_bound(),
-            regs::NV_PFB_PRI_MMU_WPR2_ADDR_HI::read(bar).higher_bound(),
-        );
-
-        match (wpr2_lo, wpr2_hi) {
-            (_, 0) => {
-                dev_err!(dev, "WPR2 region not created after running FWSEC-FRTS\n");
-
-                Err(EIO)
-            }
-            (wpr2_lo, _) if wpr2_lo != fb_layout.frts.start => {
-                dev_err!(
-                    dev,
-                    "WPR2 region created at unexpected address {:#x}; expected {:#x}\n",
-                    wpr2_lo,
-                    fb_layout.frts.start,
-                );
-
-                Err(EIO)
-            }
-            (wpr2_lo, wpr2_hi) => {
-                dev_dbg!(dev, "WPR2: {:#x}-{:#x}\n", wpr2_lo, wpr2_hi);
-                dev_dbg!(dev, "GPU instance built\n");
-
-                Ok(())
-            }
-        }
-    }
-
-    pub(crate) fn new(
-        pdev: &pci::Device<device::Bound>,
+    pub(crate) fn new<'a>(
+        pdev: &'a pci::Device<device::Bound>,
         devres_bar: Arc<Devres<Bar0>>,
-    ) -> Result<impl PinInit<Self>> {
-        let bar = devres_bar.access(pdev.as_ref())?;
-        let spec = Spec::new(bar)?;
-        let fw = Firmware::new(pdev.as_ref(), spec.chipset, FIRMWARE_VERSION)?;
+        bar: &'a Bar0,
+    ) -> impl PinInit<Self, Error> + 'a {
+        try_pin_init!(Self {
+            spec: Spec::new(bar).inspect(|spec| {
+                dev_info!(
+                    pdev.as_ref(),
+                    "NVIDIA (Chipset: {}, Architecture: {:?}, Revision: {})\n",
+                    spec.chipset,
+                    spec.chipset.arch(),
+                    spec.revision
+                );
+            })?,
 
-        dev_info!(
-            pdev.as_ref(),
-            "NVIDIA (Chipset: {}, Architecture: {:?}, Revision: {})\n",
-            spec.chipset,
-            spec.chipset.arch(),
-            spec.revision
-        );
+            // We must wait for GFW_BOOT completion before doing any significant setup on the GPU.
+            _: {
+                gfw::wait_gfw_boot_completion(bar)
+                    .inspect_err(|_| dev_err!(pdev.as_ref(), "GFW boot did not complete"))?;
+            },
 
-        // We must wait for GFW_BOOT completion before doing any significant setup on the GPU.
-        gfw::wait_gfw_boot_completion(bar)
-            .inspect_err(|_| dev_err!(pdev.as_ref(), "GFW boot did not complete"))?;
+            fw <- Firmware::new(pdev.as_ref(), spec.chipset, FIRMWARE_VERSION)?,
 
-        let sysmem_flush = SysmemFlush::register(pdev.as_ref(), bar, spec.chipset)?;
+            sysmem_flush: SysmemFlush::register(pdev.as_ref(), bar, spec.chipset)?,
 
-        let gsp_falcon = Falcon::<Gsp>::new(
-            pdev.as_ref(),
-            spec.chipset,
-            bar,
-            spec.chipset > Chipset::GA100,
-        )?;
-        gsp_falcon.clear_swgen0_intr(bar);
+            gsp_falcon: Falcon::new(
+                pdev.as_ref(),
+                spec.chipset,
+                bar,
+                spec.chipset > Chipset::GA100,
+            )
+            .inspect(|falcon| falcon.clear_swgen0_intr(bar))?,
 
-        let _sec2_falcon = Falcon::<Sec2>::new(pdev.as_ref(), spec.chipset, bar, true)?;
+            sec2_falcon: Falcon::new(pdev.as_ref(), spec.chipset, bar, true)?,
 
-        let fb_layout = FbLayout::new(spec.chipset, bar)?;
-        dev_dbg!(pdev.as_ref(), "{:#x?}\n", fb_layout);
+            gsp <- Gsp::new(),
 
-        let bios = Vbios::new(pdev.as_ref(), bar)?;
+            _: { gsp.boot(pdev, bar, spec.chipset, gsp_falcon, sec2_falcon)? },
 
-        Self::run_fwsec_frts(pdev.as_ref(), &gsp_falcon, bar, &bios, &fb_layout)?;
-
-        Ok(pin_init!(Self {
-            spec,
             bar: devres_bar,
-            fw,
-            sysmem_flush,
-        }))
+        })
     }
 
     /// Called when the corresponding [`Device`](device::Device) is unbound.
diff --git a/drivers/gpu/nova-core/gsp.rs b/drivers/gpu/nova-core/gsp.rs
new file mode 100644
index 000000000000..33cc4f2011c1
--- /dev/null
+++ b/drivers/gpu/nova-core/gsp.rs
@@ -0,0 +1,17 @@
+// SPDX-License-Identifier: GPL-2.0
+
+mod boot;
+
+use kernel::prelude::*;
+
+/// GSP runtime data.
+///
+/// This is an empty pinned placeholder for now.
+#[pin_data]
+pub(crate) struct Gsp {}
+
+impl Gsp {
+    pub(crate) fn new() -> impl PinInit<Self> {
+        pin_init!(Self {})
+    }
+}
diff --git a/drivers/gpu/nova-core/gsp/boot.rs b/drivers/gpu/nova-core/gsp/boot.rs
new file mode 100644
index 000000000000..6625a11ccc88
--- /dev/null
+++ b/drivers/gpu/nova-core/gsp/boot.rs
@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0
+
+use kernel::device;
+use kernel::pci;
+use kernel::prelude::*;
+
+use crate::driver::Bar0;
+use crate::falcon::{gsp::Gsp, sec2::Sec2, Falcon};
+use crate::fb::FbLayout;
+use crate::firmware::fwsec::{FwsecCommand, FwsecFirmware};
+use crate::gpu::Chipset;
+use crate::regs;
+use crate::vbios::Vbios;
+
+impl super::Gsp {
+    /// Helper function to load and run the FWSEC-FRTS firmware and confirm that it has properly
+    /// created the WPR2 region.
+    fn run_fwsec_frts(
+        dev: &device::Device<device::Bound>,
+        falcon: &Falcon<Gsp>,
+        bar: &Bar0,
+        bios: &Vbios,
+        fb_layout: &FbLayout,
+    ) -> Result<()> {
+        // Check that the WPR2 region does not already exists - if it does, we cannot run
+        // FWSEC-FRTS until the GPU is reset.
+        if regs::NV_PFB_PRI_MMU_WPR2_ADDR_HI::read(bar).higher_bound() != 0 {
+            dev_err!(
+                dev,
+                "WPR2 region already exists - GPU needs to be reset to proceed\n"
+            );
+            return Err(EBUSY);
+        }
+
+        let fwsec_frts = FwsecFirmware::new(
+            dev,
+            falcon,
+            bar,
+            bios,
+            FwsecCommand::Frts {
+                frts_addr: fb_layout.frts.start,
+                frts_size: fb_layout.frts.end - fb_layout.frts.start,
+            },
+        )?;
+
+        // Run FWSEC-FRTS to create the WPR2 region.
+        fwsec_frts.run(dev, falcon, bar)?;
+
+        // SCRATCH_E contains the error code for FWSEC-FRTS.
+        let frts_status = regs::NV_PBUS_SW_SCRATCH_0E_FRTS_ERR::read(bar).frts_err_code();
+        if frts_status != 0 {
+            dev_err!(
+                dev,
+                "FWSEC-FRTS returned with error code {:#x}",
+                frts_status
+            );
+
+            return Err(EIO);
+        }
+
+        // Check that the WPR2 region has been created as we requested.
+        let (wpr2_lo, wpr2_hi) = (
+            regs::NV_PFB_PRI_MMU_WPR2_ADDR_LO::read(bar).lower_bound(),
+            regs::NV_PFB_PRI_MMU_WPR2_ADDR_HI::read(bar).higher_bound(),
+        );
+
+        match (wpr2_lo, wpr2_hi) {
+            (_, 0) => {
+                dev_err!(dev, "WPR2 region not created after running FWSEC-FRTS\n");
+
+                Err(EIO)
+            }
+            (wpr2_lo, _) if wpr2_lo != fb_layout.frts.start => {
+                dev_err!(
+                    dev,
+                    "WPR2 region created at unexpected address {:#x}; expected {:#x}\n",
+                    wpr2_lo,
+                    fb_layout.frts.start,
+                );
+
+                Err(EIO)
+            }
+            (wpr2_lo, wpr2_hi) => {
+                dev_dbg!(dev, "WPR2: {:#x}-{:#x}\n", wpr2_lo, wpr2_hi);
+                dev_dbg!(dev, "GPU instance built\n");
+
+                Ok(())
+            }
+        }
+    }
+
+    /// Attempt to boot the GSP.
+    ///
+    /// This is a GPU-dependent and complex procedure that involves loading firmware files from
+    /// user-space, patching them with signatures, and building firmware-specific intricate data
+    /// structures that the GSP will use at runtime.
+    ///
+    /// Upon return, the GSP is up and running, and its runtime object given as return value.
+    pub(crate) fn boot(
+        self: Pin<&mut Self>,
+        pdev: &pci::Device<device::Bound>,
+        bar: &Bar0,
+        chipset: Chipset,
+        gsp_falcon: &Falcon<Gsp>,
+        _sec2_falcon: &Falcon<Sec2>,
+    ) -> Result {
+        let dev = pdev.as_ref();
+
+        let bios = Vbios::new(dev, bar)?;
+
+        let fb_layout = FbLayout::new(chipset, bar)?;
+        dev_dbg!(dev, "{:#x?}\n", fb_layout);
+
+        Self::run_fwsec_frts(dev, gsp_falcon, bar, &bios, &fb_layout)?;
+
+        Ok(())
+    }
+}
diff --git a/drivers/gpu/nova-core/nova_core.rs b/drivers/gpu/nova-core/nova_core.rs
index cb2bbb30cba1..fffcaee2249f 100644
--- a/drivers/gpu/nova-core/nova_core.rs
+++ b/drivers/gpu/nova-core/nova_core.rs
@@ -9,6 +9,7 @@ mod fb;
 mod firmware;
 mod gfw;
 mod gpu;
+mod gsp;
 mod regs;
 mod util;
 mod vbios;

From b345c917d7c1e3841d286560afca476ab7caff74 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:16 +0900
Subject: [PATCH 1232/1455] gpu: nova-core: add Chipset::name() method

There are a few cases where we need the lowercase name of a given
chipset, notably to resolve firmware files paths for dynamic loading or
to build the module information.

So far, we relied on a static `NAMES` array for the latter, and some
CString hackery for the former.

Replace both with a new `name` const method that returns the lowercase
name of a chipset instance. We can generate it using the `paste!` macro.

Using this method removes the need to create a `CString` when loading
firmware, and lets us remove a couple of utility functions that now have
no user.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-3-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/firmware.rs |  8 +++-----
 drivers/gpu/nova-core/gpu.rs      | 25 +++++++++++++++++--------
 drivers/gpu/nova-core/util.rs     | 20 --------------------
 3 files changed, 20 insertions(+), 33 deletions(-)

diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index 2931912ddba0..213d4506a53f 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -30,9 +30,7 @@ pub(crate) struct Firmware {
 
 impl Firmware {
     pub(crate) fn new(dev: &device::Device, chipset: Chipset, ver: &str) -> Result<Firmware> {
-        let mut chip_name = CString::try_from_fmt(fmt!("{chipset}"))?;
-        chip_name.make_ascii_lowercase();
-        let chip_name = &*chip_name;
+        let chip_name = chipset.name();
 
         let request = |name_| {
             CString::try_from_fmt(fmt!("nvidia/{chip_name}/gsp/{name_}-{ver}.bin"))
@@ -180,8 +178,8 @@ impl<const N: usize> ModInfoBuilder<N> {
         let mut this = Self(firmware::ModInfoBuilder::new(module_name));
         let mut i = 0;
 
-        while i < gpu::Chipset::NAMES.len() {
-            this = this.make_entry_chipset(gpu::Chipset::NAMES[i]);
+        while i < gpu::Chipset::ALL.len() {
+            this = this.make_entry_chipset(gpu::Chipset::ALL[i].name());
             i += 1;
         }
 
diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index 3c019df01d30..d2395335727a 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -9,7 +9,6 @@ use crate::firmware::{Firmware, FIRMWARE_VERSION};
 use crate::gfw;
 use crate::gsp::Gsp;
 use crate::regs;
-use crate::util;
 use core::fmt;
 
 macro_rules! define_chipset {
@@ -26,13 +25,23 @@ macro_rules! define_chipset {
                 $( Chipset::$variant, )*
             ];
 
-            pub(crate) const NAMES: [&'static str; Self::ALL.len()] = [
-                $( util::const_bytes_to_str(
-                        util::to_lowercase_bytes::<{ stringify!($variant).len() }>(
-                            stringify!($variant)
-                        ).as_slice()
-                ), )*
-            ];
+            ::kernel::macros::paste!(
+            /// Returns the name of this chipset, in lowercase.
+            ///
+            /// # Examples
+            ///
+            /// ```
+            /// let chipset = Chipset::GA102;
+            /// assert_eq!(chipset.name(), "ga102");
+            /// ```
+            pub(crate) const fn name(&self) -> &'static str {
+                match *self {
+                $(
+                    Chipset::$variant => stringify!([<$variant:lower>]),
+                )*
+                }
+            }
+            );
         }
 
         // TODO[FPRI]: replace with something like derive(FromPrimitive)
diff --git a/drivers/gpu/nova-core/util.rs b/drivers/gpu/nova-core/util.rs
index 76cedf3710d7..bf35f00cb732 100644
--- a/drivers/gpu/nova-core/util.rs
+++ b/drivers/gpu/nova-core/util.rs
@@ -3,26 +3,6 @@
 use kernel::prelude::*;
 use kernel::time::{Delta, Instant, Monotonic};
 
-pub(crate) const fn to_lowercase_bytes<const N: usize>(s: &str) -> [u8; N] {
-    let src = s.as_bytes();
-    let mut dst = [0; N];
-    let mut i = 0;
-
-    while i < src.len() && i < N {
-        dst[i] = (src[i] as char).to_ascii_lowercase() as u8;
-        i += 1;
-    }
-
-    dst
-}
-
-pub(crate) const fn const_bytes_to_str(bytes: &[u8]) -> &str {
-    match core::str::from_utf8(bytes) {
-        Ok(string) => string,
-        Err(_) => kernel::build_error!("Bytes are not valid UTF-8."),
-    }
-}
-
 /// Wait until `cond` is true or `timeout` elapsed.
 ///
 /// When `cond` evaluates to `Some`, its return value is returned.

From ebe658732c6325aa3f4063726af00c0d12449d3c Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:17 +0900
Subject: [PATCH 1233/1455] gpu: nova-core: firmware: move firmware request
 code into a function

When all the firmware files are loaded from `Firmware::new`, it makes
sense to have the firmware request code as a closure. However, since we
eventually want each individual firmware constructor to request its own
file (and get rid of `Firmware` altogether), move this code into a
dedicated function that can be called by individual firmware types.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-4-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/firmware.rs | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index 213d4506a53f..8e40a671eca8 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -19,6 +19,19 @@ pub(crate) mod fwsec;
 
 pub(crate) const FIRMWARE_VERSION: &str = "535.113.01";
 
+/// Requests the GPU firmware `name` suitable for `chipset`, with version `ver`.
+fn request_firmware(
+    dev: &device::Device,
+    chipset: gpu::Chipset,
+    name: &str,
+    ver: &str,
+) -> Result<firmware::Firmware> {
+    let chip_name = chipset.name();
+
+    CString::try_from_fmt(fmt!("nvidia/{chip_name}/gsp/{name}-{ver}.bin"))
+        .and_then(|path| firmware::Firmware::request(&path, dev))
+}
+
 /// Structure encapsulating the firmware blobs required for the GPU to operate.
 #[expect(dead_code)]
 pub(crate) struct Firmware {
@@ -30,12 +43,7 @@ pub(crate) struct Firmware {
 
 impl Firmware {
     pub(crate) fn new(dev: &device::Device, chipset: Chipset, ver: &str) -> Result<Firmware> {
-        let chip_name = chipset.name();
-
-        let request = |name_| {
-            CString::try_from_fmt(fmt!("nvidia/{chip_name}/gsp/{name_}-{ver}.bin"))
-                .and_then(|path| firmware::Firmware::request(&path, dev))
-        };
+        let request = |name| request_firmware(dev, chipset, name, ver);
 
         Ok(Firmware {
             booter_load: request("booter_load")?,

From d6cb7319e64eecbdaa63e54a38dec03b30a4fdfd Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:18 +0900
Subject: [PATCH 1234/1455] gpu: nova-core: firmware: add support for common
 firmware header

Several firmware files loaded from userspace feature a common header
that describes their payload. Add basic support for it so subsequent
patches can leverage it.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-5-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/firmware.rs | 62 +++++++++++++++++++++++++++++++
 1 file changed, 62 insertions(+)

diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index 8e40a671eca8..dbc62c00d95b 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -4,11 +4,13 @@
 //! to be loaded into a given execution unit.
 
 use core::marker::PhantomData;
+use core::mem::size_of;
 
 use kernel::device;
 use kernel::firmware;
 use kernel::prelude::*;
 use kernel::str::CString;
+use kernel::transmute::FromBytes;
 
 use crate::dma::DmaObject;
 use crate::falcon::FalconFirmware;
@@ -156,6 +158,66 @@ impl<F: FalconFirmware> FirmwareDmaObject<F, Unsigned> {
     }
 }
 
+/// Header common to most firmware files.
+#[repr(C)]
+#[derive(Debug, Clone)]
+struct BinHdr {
+    /// Magic number, must be `0x10de`.
+    bin_magic: u32,
+    /// Version of the header.
+    bin_ver: u32,
+    /// Size in bytes of the binary (to be ignored).
+    bin_size: u32,
+    /// Offset of the start of the application-specific header.
+    header_offset: u32,
+    /// Offset of the start of the data payload.
+    data_offset: u32,
+    /// Size in bytes of the data payload.
+    data_size: u32,
+}
+
+// SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+unsafe impl FromBytes for BinHdr {}
+
+// A firmware blob starting with a `BinHdr`.
+struct BinFirmware<'a> {
+    hdr: BinHdr,
+    fw: &'a [u8],
+}
+
+#[expect(dead_code)]
+impl<'a> BinFirmware<'a> {
+    /// Interpret `fw` as a firmware image starting with a [`BinHdr`], and returns the
+    /// corresponding [`BinFirmware`] that can be used to extract its payload.
+    fn new(fw: &'a firmware::Firmware) -> Result<Self> {
+        const BIN_MAGIC: u32 = 0x10de;
+        let fw = fw.data();
+
+        fw.get(0..size_of::<BinHdr>())
+            // Extract header.
+            .and_then(BinHdr::from_bytes_copy)
+            // Validate header.
+            .and_then(|hdr| {
+                if hdr.bin_magic == BIN_MAGIC {
+                    Some(hdr)
+                } else {
+                    None
+                }
+            })
+            .map(|hdr| Self { hdr, fw })
+            .ok_or(EINVAL)
+    }
+
+    /// Returns the data payload of the firmware, or `None` if the data range is out of bounds of
+    /// the firmware image.
+    fn data(&self) -> Option<&[u8]> {
+        let fw_start = self.hdr.data_offset as usize;
+        let fw_size = self.hdr.data_size as usize;
+
+        self.fw.get(fw_start..fw_start + fw_size)
+    }
+}
+
 pub(crate) struct ModInfoBuilder<const N: usize>(firmware::ModInfoBuilder<N>);
 
 impl<const N: usize> ModInfoBuilder<N> {

From 3e5c9681bf86461d26e9db85eeb9b8cbaa256e6a Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:19 +0900
Subject: [PATCH 1235/1455] gpu: nova-core: firmware: process Booter and patch
 its signature

The Booter signed firmware is an essential part of bringing up the GSP
on Turing and Ampere. It is loaded on the sec2 falcon core and is
responsible for loading and running the RISC-V GSP bootloader into the
GSP core.

Add support for parsing the Booter firmware loaded from userspace, patch
its signatures, and store it into a form that is ready to be loaded and
executed on the sec2 falcon.

Then, move the Booter instance from the `Firmware` struct to the
`start_gsp` method since it doesn't need to be kept after the GSP is
booted.

We do not run Booter yet, as its own payload (the GSP bootloader and
firmware image) still need to be prepared.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-6-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/falcon.rs          |   4 +-
 drivers/gpu/nova-core/firmware.rs        |   6 +-
 drivers/gpu/nova-core/firmware/booter.rs | 375 +++++++++++++++++++++++
 drivers/gpu/nova-core/gsp/boot.rs        |  17 +-
 4 files changed, 393 insertions(+), 9 deletions(-)
 create mode 100644 drivers/gpu/nova-core/firmware/booter.rs

diff --git a/drivers/gpu/nova-core/falcon.rs b/drivers/gpu/nova-core/falcon.rs
index b16207e7242f..37e6298195e4 100644
--- a/drivers/gpu/nova-core/falcon.rs
+++ b/drivers/gpu/nova-core/falcon.rs
@@ -293,7 +293,7 @@ pub(crate) trait FalconEngine:
 }
 
 /// Represents a portion of the firmware to be loaded into a particular memory (e.g. IMEM or DMEM).
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub(crate) struct FalconLoadTarget {
     /// Offset from the start of the source object to copy from.
     pub(crate) src_start: u32,
@@ -304,7 +304,7 @@ pub(crate) struct FalconLoadTarget {
 }
 
 /// Parameters for the falcon boot ROM.
-#[derive(Debug)]
+#[derive(Debug, Clone)]
 pub(crate) struct FalconBromParams {
     /// Offset in `DMEM`` of the firmware's signature.
     pub(crate) pkc_data_offset: u32,
diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index dbc62c00d95b..d821744d9961 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -17,6 +17,7 @@ use crate::falcon::FalconFirmware;
 use crate::gpu;
 use crate::gpu::Chipset;
 
+pub(crate) mod booter;
 pub(crate) mod fwsec;
 
 pub(crate) const FIRMWARE_VERSION: &str = "535.113.01";
@@ -37,8 +38,6 @@ fn request_firmware(
 /// Structure encapsulating the firmware blobs required for the GPU to operate.
 #[expect(dead_code)]
 pub(crate) struct Firmware {
-    booter_load: firmware::Firmware,
-    booter_unload: firmware::Firmware,
     bootloader: firmware::Firmware,
     gsp: firmware::Firmware,
 }
@@ -48,8 +47,6 @@ impl Firmware {
         let request = |name| request_firmware(dev, chipset, name, ver);
 
         Ok(Firmware {
-            booter_load: request("booter_load")?,
-            booter_unload: request("booter_unload")?,
             bootloader: request("bootloader")?,
             gsp: request("gsp")?,
         })
@@ -185,7 +182,6 @@ struct BinFirmware<'a> {
     fw: &'a [u8],
 }
 
-#[expect(dead_code)]
 impl<'a> BinFirmware<'a> {
     /// Interpret `fw` as a firmware image starting with a [`BinHdr`], and returns the
     /// corresponding [`BinFirmware`] that can be used to extract its payload.
diff --git a/drivers/gpu/nova-core/firmware/booter.rs b/drivers/gpu/nova-core/firmware/booter.rs
new file mode 100644
index 000000000000..b4ff1b17e4a0
--- /dev/null
+++ b/drivers/gpu/nova-core/firmware/booter.rs
@@ -0,0 +1,375 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! Support for loading and patching the `Booter` firmware. `Booter` is a Heavy Secured firmware
+//! running on [`Sec2`], that is used on Turing/Ampere to load the GSP firmware into the GSP falcon
+//! (and optionally unload it through a separate firmware image).
+
+use core::marker::PhantomData;
+use core::mem::size_of;
+use core::ops::Deref;
+
+use kernel::device;
+use kernel::prelude::*;
+use kernel::transmute::FromBytes;
+
+use crate::dma::DmaObject;
+use crate::driver::Bar0;
+use crate::falcon::sec2::Sec2;
+use crate::falcon::{Falcon, FalconBromParams, FalconFirmware, FalconLoadParams, FalconLoadTarget};
+use crate::firmware::{BinFirmware, FirmwareDmaObject, FirmwareSignature, Signed, Unsigned};
+use crate::gpu::Chipset;
+
+/// Local convenience function to return a copy of `S` by reinterpreting the bytes starting at
+/// `offset` in `slice`.
+fn frombytes_at<S: FromBytes + Sized>(slice: &[u8], offset: usize) -> Result<S> {
+    slice
+        .get(offset..offset + size_of::<S>())
+        .and_then(S::from_bytes_copy)
+        .ok_or(EINVAL)
+}
+
+/// Heavy-Secured firmware header.
+///
+/// Such firmwares have an application-specific payload that needs to be patched with a given
+/// signature.
+#[repr(C)]
+#[derive(Debug, Clone)]
+struct HsHeaderV2 {
+    /// Offset to the start of the signatures.
+    sig_prod_offset: u32,
+    /// Size in bytes of the signatures.
+    sig_prod_size: u32,
+    /// Offset to a `u32` containing the location at which to patch the signature in the microcode
+    /// image.
+    patch_loc_offset: u32,
+    /// Offset to a `u32` containing the index of the signature to patch.
+    patch_sig_offset: u32,
+    /// Start offset to the signature metadata.
+    meta_data_offset: u32,
+    /// Size in bytes of the signature metadata.
+    meta_data_size: u32,
+    /// Offset to a `u32` containing the number of signatures in the signatures section.
+    num_sig_offset: u32,
+    /// Offset of the application-specific header.
+    header_offset: u32,
+    /// Size in bytes of the application-specific header.
+    header_size: u32,
+}
+
+// SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+unsafe impl FromBytes for HsHeaderV2 {}
+
+/// Heavy-Secured Firmware image container.
+///
+/// This provides convenient access to the fields of [`HsHeaderV2`] that are actually indices to
+/// read from in the firmware data.
+struct HsFirmwareV2<'a> {
+    hdr: HsHeaderV2,
+    fw: &'a [u8],
+}
+
+impl<'a> HsFirmwareV2<'a> {
+    /// Interprets the header of `bin_fw` as a [`HsHeaderV2`] and returns an instance of
+    /// `HsFirmwareV2` for further parsing.
+    ///
+    /// Fails if the header pointed at by `bin_fw` is not within the bounds of the firmware image.
+    fn new(bin_fw: &BinFirmware<'a>) -> Result<Self> {
+        frombytes_at::<HsHeaderV2>(bin_fw.fw, bin_fw.hdr.header_offset as usize)
+            .map(|hdr| Self { hdr, fw: bin_fw.fw })
+    }
+
+    /// Returns the location at which the signatures should be patched in the microcode image.
+    ///
+    /// Fails if the offset of the patch location is outside the bounds of the firmware
+    /// image.
+    fn patch_location(&self) -> Result<u32> {
+        frombytes_at::<u32>(self.fw, self.hdr.patch_loc_offset as usize)
+    }
+
+    /// Returns an iterator to the signatures of the firmware. The iterator can be empty if the
+    /// firmware is unsigned.
+    ///
+    /// Fails if the pointed signatures are outside the bounds of the firmware image.
+    fn signatures_iter(&'a self) -> Result<impl Iterator<Item = BooterSignature<'a>>> {
+        let num_sig = frombytes_at::<u32>(self.fw, self.hdr.num_sig_offset as usize)?;
+        let iter = match self.hdr.sig_prod_size.checked_div(num_sig) {
+            // If there are no signatures, return an iterator that will yield zero elements.
+            None => (&[] as &[u8]).chunks_exact(1),
+            Some(sig_size) => {
+                let patch_sig = frombytes_at::<u32>(self.fw, self.hdr.patch_sig_offset as usize)?;
+                let signatures_start = (self.hdr.sig_prod_offset + patch_sig) as usize;
+
+                self.fw
+                    // Get signatures range.
+                    .get(signatures_start..signatures_start + self.hdr.sig_prod_size as usize)
+                    .ok_or(EINVAL)?
+                    .chunks_exact(sig_size as usize)
+            }
+        };
+
+        // Map the byte slices into signatures.
+        Ok(iter.map(BooterSignature))
+    }
+}
+
+/// Signature parameters, as defined in the firmware.
+#[repr(C)]
+struct HsSignatureParams {
+    /// Fuse version to use.
+    fuse_ver: u32,
+    /// Mask of engine IDs this firmware applies to.
+    engine_id_mask: u32,
+    /// ID of the microcode.
+    ucode_id: u32,
+}
+
+// SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+unsafe impl FromBytes for HsSignatureParams {}
+
+impl HsSignatureParams {
+    /// Returns the signature parameters contained in `hs_fw`.
+    ///
+    /// Fails if the meta data parameter of `hs_fw` is outside the bounds of the firmware image, or
+    /// if its size doesn't match that of [`HsSignatureParams`].
+    fn new(hs_fw: &HsFirmwareV2<'_>) -> Result<Self> {
+        let start = hs_fw.hdr.meta_data_offset as usize;
+        let end = start
+            .checked_add(hs_fw.hdr.meta_data_size as usize)
+            .ok_or(EINVAL)?;
+
+        hs_fw
+            .fw
+            .get(start..end)
+            .and_then(Self::from_bytes_copy)
+            .ok_or(EINVAL)
+    }
+}
+
+/// Header for code and data load offsets.
+#[repr(C)]
+#[derive(Debug, Clone)]
+struct HsLoadHeaderV2 {
+    // Offset at which the code starts.
+    os_code_offset: u32,
+    // Total size of the code, for all apps.
+    os_code_size: u32,
+    // Offset at which the data starts.
+    os_data_offset: u32,
+    // Size of the data.
+    os_data_size: u32,
+    // Number of apps following this header. Each app is described by a [`HsLoadHeaderV2App`].
+    num_apps: u32,
+}
+
+// SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+unsafe impl FromBytes for HsLoadHeaderV2 {}
+
+impl HsLoadHeaderV2 {
+    /// Returns the load header contained in `hs_fw`.
+    ///
+    /// Fails if the header pointed at by `hs_fw` is not within the bounds of the firmware image.
+    fn new(hs_fw: &HsFirmwareV2<'_>) -> Result<Self> {
+        frombytes_at::<Self>(hs_fw.fw, hs_fw.hdr.header_offset as usize)
+    }
+}
+
+/// Header for app code loader.
+#[repr(C)]
+#[derive(Debug, Clone)]
+struct HsLoadHeaderV2App {
+    /// Offset at which to load the app code.
+    offset: u32,
+    /// Length in bytes of the app code.
+    len: u32,
+}
+
+// SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+unsafe impl FromBytes for HsLoadHeaderV2App {}
+
+impl HsLoadHeaderV2App {
+    /// Returns the [`HsLoadHeaderV2App`] for app `idx` of `hs_fw`.
+    ///
+    /// Fails if `idx` is larger than the number of apps declared in `hs_fw`, or if the header is
+    /// not within the bounds of the firmware image.
+    fn new(hs_fw: &HsFirmwareV2<'_>, idx: u32) -> Result<Self> {
+        let load_hdr = HsLoadHeaderV2::new(hs_fw)?;
+        if idx >= load_hdr.num_apps {
+            Err(EINVAL)
+        } else {
+            frombytes_at::<Self>(
+                hs_fw.fw,
+                (hs_fw.hdr.header_offset as usize)
+                    // Skip the load header...
+                    .checked_add(size_of::<HsLoadHeaderV2>())
+                    // ... and jump to app header `idx`.
+                    .and_then(|offset| {
+                        offset.checked_add((idx as usize).checked_mul(size_of::<Self>())?)
+                    })
+                    .ok_or(EINVAL)?,
+            )
+        }
+    }
+}
+
+/// Signature for Booter firmware. Their size is encoded into the header and not known a compile
+/// time, so we just wrap a byte slices on which we can implement [`FirmwareSignature`].
+struct BooterSignature<'a>(&'a [u8]);
+
+impl<'a> AsRef<[u8]> for BooterSignature<'a> {
+    fn as_ref(&self) -> &[u8] {
+        self.0
+    }
+}
+
+impl<'a> FirmwareSignature<BooterFirmware> for BooterSignature<'a> {}
+
+/// The `Booter` loader firmware, responsible for loading the GSP.
+pub(crate) struct BooterFirmware {
+    // Load parameters for `IMEM` falcon memory.
+    imem_load_target: FalconLoadTarget,
+    // Load parameters for `DMEM` falcon memory.
+    dmem_load_target: FalconLoadTarget,
+    // BROM falcon parameters.
+    brom_params: FalconBromParams,
+    // Device-mapped firmware image.
+    ucode: FirmwareDmaObject<Self, Signed>,
+}
+
+impl FirmwareDmaObject<BooterFirmware, Unsigned> {
+    fn new_booter(dev: &device::Device<device::Bound>, data: &[u8]) -> Result<Self> {
+        DmaObject::from_data(dev, data).map(|ucode| Self(ucode, PhantomData))
+    }
+}
+
+#[derive(Copy, Clone, Debug, PartialEq)]
+pub(crate) enum BooterKind {
+    Loader,
+    #[expect(unused)]
+    Unloader,
+}
+
+impl BooterFirmware {
+    /// Parses the Booter firmware contained in `fw`, and patches the correct signature so it is
+    /// ready to be loaded and run on `falcon`.
+    pub(crate) fn new(
+        dev: &device::Device<device::Bound>,
+        kind: BooterKind,
+        chipset: Chipset,
+        ver: &str,
+        falcon: &Falcon<<Self as FalconFirmware>::Target>,
+        bar: &Bar0,
+    ) -> Result<Self> {
+        let fw_name = match kind {
+            BooterKind::Loader => "booter_load",
+            BooterKind::Unloader => "booter_unload",
+        };
+        let fw = super::request_firmware(dev, chipset, fw_name, ver)?;
+        let bin_fw = BinFirmware::new(&fw)?;
+
+        // The binary firmware embeds a Heavy-Secured firmware.
+        let hs_fw = HsFirmwareV2::new(&bin_fw)?;
+
+        // The Heavy-Secured firmware embeds a firmware load descriptor.
+        let load_hdr = HsLoadHeaderV2::new(&hs_fw)?;
+
+        // Offset in `ucode` where to patch the signature.
+        let patch_loc = hs_fw.patch_location()?;
+
+        let sig_params = HsSignatureParams::new(&hs_fw)?;
+        let brom_params = FalconBromParams {
+            // `load_hdr.os_data_offset` is an absolute index, but `pkc_data_offset` is from the
+            // signature patch location.
+            pkc_data_offset: patch_loc
+                .checked_sub(load_hdr.os_data_offset)
+                .ok_or(EINVAL)?,
+            engine_id_mask: u16::try_from(sig_params.engine_id_mask).map_err(|_| EINVAL)?,
+            ucode_id: u8::try_from(sig_params.ucode_id).map_err(|_| EINVAL)?,
+        };
+        let app0 = HsLoadHeaderV2App::new(&hs_fw, 0)?;
+
+        // Object containing the firmware microcode to be signature-patched.
+        let ucode = bin_fw
+            .data()
+            .ok_or(EINVAL)
+            .and_then(|data| FirmwareDmaObject::<Self, _>::new_booter(dev, data))?;
+
+        let ucode_signed = {
+            let mut signatures = hs_fw.signatures_iter()?.peekable();
+
+            if signatures.peek().is_none() {
+                // If there are no signatures, then the firmware is unsigned.
+                ucode.no_patch_signature()
+            } else {
+                // Obtain the version from the fuse register, and extract the corresponding
+                // signature.
+                let reg_fuse_version = falcon.signature_reg_fuse_version(
+                    bar,
+                    brom_params.engine_id_mask,
+                    brom_params.ucode_id,
+                )?;
+
+                // `0` means the last signature should be used.
+                const FUSE_VERSION_USE_LAST_SIG: u32 = 0;
+                let signature = match reg_fuse_version {
+                    FUSE_VERSION_USE_LAST_SIG => signatures.last(),
+                    // Otherwise hardware fuse version needs to be subtracted to obtain the index.
+                    reg_fuse_version => {
+                        let Some(idx) = sig_params.fuse_ver.checked_sub(reg_fuse_version) else {
+                            dev_err!(dev, "invalid fuse version for Booter firmware\n");
+                            return Err(EINVAL);
+                        };
+                        signatures.nth(idx as usize)
+                    }
+                }
+                .ok_or(EINVAL)?;
+
+                ucode.patch_signature(&signature, patch_loc as usize)?
+            }
+        };
+
+        Ok(Self {
+            imem_load_target: FalconLoadTarget {
+                src_start: app0.offset,
+                dst_start: 0,
+                len: app0.len,
+            },
+            dmem_load_target: FalconLoadTarget {
+                src_start: load_hdr.os_data_offset,
+                dst_start: 0,
+                len: load_hdr.os_data_size,
+            },
+            brom_params,
+            ucode: ucode_signed,
+        })
+    }
+}
+
+impl FalconLoadParams for BooterFirmware {
+    fn imem_load_params(&self) -> FalconLoadTarget {
+        self.imem_load_target.clone()
+    }
+
+    fn dmem_load_params(&self) -> FalconLoadTarget {
+        self.dmem_load_target.clone()
+    }
+
+    fn brom_params(&self) -> FalconBromParams {
+        self.brom_params.clone()
+    }
+
+    fn boot_addr(&self) -> u32 {
+        self.imem_load_target.src_start
+    }
+}
+
+impl Deref for BooterFirmware {
+    type Target = DmaObject;
+
+    fn deref(&self) -> &Self::Target {
+        &self.ucode.0
+    }
+}
+
+impl FalconFirmware for BooterFirmware {
+    type Target = Sec2;
+}
diff --git a/drivers/gpu/nova-core/gsp/boot.rs b/drivers/gpu/nova-core/gsp/boot.rs
index 6625a11ccc88..faa553635b7c 100644
--- a/drivers/gpu/nova-core/gsp/boot.rs
+++ b/drivers/gpu/nova-core/gsp/boot.rs
@@ -7,7 +7,11 @@ use kernel::prelude::*;
 use crate::driver::Bar0;
 use crate::falcon::{gsp::Gsp, sec2::Sec2, Falcon};
 use crate::fb::FbLayout;
-use crate::firmware::fwsec::{FwsecCommand, FwsecFirmware};
+use crate::firmware::{
+    booter::{BooterFirmware, BooterKind},
+    fwsec::{FwsecCommand, FwsecFirmware},
+    FIRMWARE_VERSION,
+};
 use crate::gpu::Chipset;
 use crate::regs;
 use crate::vbios::Vbios;
@@ -102,7 +106,7 @@ impl super::Gsp {
         bar: &Bar0,
         chipset: Chipset,
         gsp_falcon: &Falcon<Gsp>,
-        _sec2_falcon: &Falcon<Sec2>,
+        sec2_falcon: &Falcon<Sec2>,
     ) -> Result {
         let dev = pdev.as_ref();
 
@@ -113,6 +117,15 @@ impl super::Gsp {
 
         Self::run_fwsec_frts(dev, gsp_falcon, bar, &bios, &fb_layout)?;
 
+        let _booter_loader = BooterFirmware::new(
+            dev,
+            BooterKind::Loader,
+            chipset,
+            FIRMWARE_VERSION,
+            sec2_falcon,
+            bar,
+        )?;
+
         Ok(())
     }
 }

From a841614e607c9e232dd56ec726ba63d2750025a2 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:20 +0900
Subject: [PATCH 1236/1455] gpu: nova-core: firmware: process and prepare the
 GSP firmware

The GSP firmware is a binary blob that is verified, loaded, and run by
the GSP bootloader. Its presentation is a bit peculiar as the GSP
bootloader expects to be given a DMA address to a 3-levels page table
mapping the GSP firmware at address 0 of its own address space.

Prepare such a structure containing the DMA-mapped firmware as well as
the DMA-mapped page tables, and a way to obtain the DMA handle of the
level 0 page table.

Then, move the GSP firmware instance from the `Firmware` struct to the
`start_gsp` method since it doesn't need to be kept after the GSP is
booted.

As we are performing the required ELF section parsing and radix3 page
table building, remove these items from the TODO file.

Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-7-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 Documentation/gpu/nova/core/todo.rst  |  17 --
 drivers/gpu/nova-core/firmware.rs     |   3 +-
 drivers/gpu/nova-core/firmware/gsp.rs | 236 ++++++++++++++++++++++++++
 drivers/gpu/nova-core/gsp.rs          |   3 +
 drivers/gpu/nova-core/gsp/boot.rs     |   6 +
 5 files changed, 246 insertions(+), 19 deletions(-)
 create mode 100644 drivers/gpu/nova-core/firmware/gsp.rs

diff --git a/Documentation/gpu/nova/core/todo.rst b/Documentation/gpu/nova/core/todo.rst
index 48b20656dcb1..091a2fb78c6b 100644
--- a/Documentation/gpu/nova/core/todo.rst
+++ b/Documentation/gpu/nova/core/todo.rst
@@ -230,23 +230,6 @@ Rust abstraction for debugfs APIs.
 GPU (general)
 =============
 
-Parse firmware headers
-----------------------
-
-Parse ELF headers from the firmware files loaded from the filesystem.
-
-| Reference: ELF utils
-| Complexity: Beginner
-| Contact: Abdiel Janulgue
-
-Build radix3 page table
------------------------
-
-Build the radix3 page table to map the firmware.
-
-| Complexity: Intermediate
-| Contact: Abdiel Janulgue
-
 Initial Devinit support
 -----------------------
 
diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index d821744d9961..10200d11bbbf 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -19,6 +19,7 @@ use crate::gpu::Chipset;
 
 pub(crate) mod booter;
 pub(crate) mod fwsec;
+pub(crate) mod gsp;
 
 pub(crate) const FIRMWARE_VERSION: &str = "535.113.01";
 
@@ -39,7 +40,6 @@ fn request_firmware(
 #[expect(dead_code)]
 pub(crate) struct Firmware {
     bootloader: firmware::Firmware,
-    gsp: firmware::Firmware,
 }
 
 impl Firmware {
@@ -48,7 +48,6 @@ impl Firmware {
 
         Ok(Firmware {
             bootloader: request("bootloader")?,
-            gsp: request("gsp")?,
         })
     }
 }
diff --git a/drivers/gpu/nova-core/firmware/gsp.rs b/drivers/gpu/nova-core/firmware/gsp.rs
new file mode 100644
index 000000000000..a7eda9ca0d21
--- /dev/null
+++ b/drivers/gpu/nova-core/firmware/gsp.rs
@@ -0,0 +1,236 @@
+// SPDX-License-Identifier: GPL-2.0
+
+use core::mem::size_of_val;
+
+use kernel::device;
+use kernel::dma::{DataDirection, DmaAddress};
+use kernel::kvec;
+use kernel::prelude::*;
+use kernel::scatterlist::{Owned, SGTable};
+
+use crate::dma::DmaObject;
+use crate::gpu::{Architecture, Chipset};
+use crate::gsp::GSP_PAGE_SIZE;
+
+/// Ad-hoc and temporary module to extract sections from ELF images.
+///
+/// Some firmware images are currently packaged as ELF files, where sections names are used as keys
+/// to specific and related bits of data. Future firmware versions are scheduled to move away from
+/// that scheme before nova-core becomes stable, which means this module will eventually be
+/// removed.
+mod elf {
+    use core::mem::size_of;
+
+    use kernel::bindings;
+    use kernel::str::CStr;
+    use kernel::transmute::FromBytes;
+
+    /// Newtype to provide a [`FromBytes`] implementation.
+    #[repr(transparent)]
+    struct Elf64Hdr(bindings::elf64_hdr);
+    // SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+    unsafe impl FromBytes for Elf64Hdr {}
+
+    #[repr(transparent)]
+    struct Elf64SHdr(bindings::elf64_shdr);
+    // SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+    unsafe impl FromBytes for Elf64SHdr {}
+
+    /// Tries to extract section with name `name` from the ELF64 image `elf`, and returns it.
+    pub(super) fn elf64_section<'a, 'b>(elf: &'a [u8], name: &'b str) -> Option<&'a [u8]> {
+        let hdr = &elf
+            .get(0..size_of::<bindings::elf64_hdr>())
+            .and_then(Elf64Hdr::from_bytes)?
+            .0;
+
+        // Get all the section headers.
+        let mut shdr = {
+            let shdr_num = usize::from(hdr.e_shnum);
+            let shdr_start = usize::try_from(hdr.e_shoff).ok()?;
+            let shdr_end = shdr_num
+                .checked_mul(size_of::<Elf64SHdr>())
+                .and_then(|v| v.checked_add(shdr_start))?;
+
+            elf.get(shdr_start..shdr_end)
+                .map(|slice| slice.chunks_exact(size_of::<Elf64SHdr>()))?
+        };
+
+        // Get the strings table.
+        let strhdr = shdr
+            .clone()
+            .nth(usize::from(hdr.e_shstrndx))
+            .and_then(Elf64SHdr::from_bytes)?;
+
+        // Find the section which name matches `name` and return it.
+        shdr.find(|&sh| {
+            let Some(hdr) = Elf64SHdr::from_bytes(sh) else {
+                return false;
+            };
+
+            let Some(name_idx) = strhdr
+                .0
+                .sh_offset
+                .checked_add(u64::from(hdr.0.sh_name))
+                .and_then(|idx| usize::try_from(idx).ok())
+            else {
+                return false;
+            };
+
+            // Get the start of the name.
+            elf.get(name_idx..)
+                // Stop at the first `0`.
+                .and_then(|nstr| nstr.get(0..=nstr.iter().position(|b| *b == 0)?))
+                // Convert into CStr. This should never fail because of the line above.
+                .and_then(|nstr| CStr::from_bytes_with_nul(nstr).ok())
+                // Convert into str.
+                .and_then(|c_str| c_str.to_str().ok())
+                // Check that the name matches.
+                .map(|str| str == name)
+                .unwrap_or(false)
+        })
+        // Return the slice containing the section.
+        .and_then(|sh| {
+            let hdr = Elf64SHdr::from_bytes(sh)?;
+            let start = usize::try_from(hdr.0.sh_offset).ok()?;
+            let end = usize::try_from(hdr.0.sh_size)
+                .ok()
+                .and_then(|sh_size| start.checked_add(sh_size))?;
+
+            elf.get(start..end)
+        })
+    }
+}
+
+/// GSP firmware with 3-level radix page tables for the GSP bootloader.
+///
+/// The bootloader expects firmware to be mapped starting at address 0 in GSP's virtual address
+/// space:
+///
+/// ```text
+/// Level 0:  1 page, 1 entry         -> points to first level 1 page
+/// Level 1:  Multiple pages/entries  -> each entry points to a level 2 page
+/// Level 2:  Multiple pages/entries  -> each entry points to a firmware page
+/// ```
+///
+/// Each page is 4KB, each entry is 8 bytes (64-bit DMA address).
+/// Also known as "Radix3" firmware.
+#[pin_data]
+pub(crate) struct GspFirmware {
+    /// The GSP firmware inside a [`VVec`], device-mapped via a SG table.
+    #[pin]
+    fw: SGTable<Owned<VVec<u8>>>,
+    /// Level 2 page table whose entries contain DMA addresses of firmware pages.
+    #[pin]
+    level2: SGTable<Owned<VVec<u8>>>,
+    /// Level 1 page table whose entries contain DMA addresses of level 2 pages.
+    #[pin]
+    level1: SGTable<Owned<VVec<u8>>>,
+    /// Level 0 page table (single 4KB page) with one entry: DMA address of first level 1 page.
+    level0: DmaObject,
+    /// Size in bytes of the firmware contained in [`Self::fw`].
+    size: usize,
+    /// Device-mapped GSP signatures matching the GPU's [`Chipset`].
+    signatures: DmaObject,
+}
+
+impl GspFirmware {
+    /// Loads the GSP firmware binaries, map them into `dev`'s address-space, and creates the page
+    /// tables expected by the GSP bootloader to load it.
+    pub(crate) fn new<'a, 'b>(
+        dev: &'a device::Device<device::Bound>,
+        chipset: Chipset,
+        ver: &'b str,
+    ) -> Result<impl PinInit<Self, Error> + 'a> {
+        let fw = super::request_firmware(dev, chipset, "gsp", ver)?;
+
+        let fw_section = elf::elf64_section(fw.data(), ".fwimage").ok_or(EINVAL)?;
+
+        let sigs_section = match chipset.arch() {
+            Architecture::Ampere => ".fwsignature_ga10x",
+            _ => return Err(ENOTSUPP),
+        };
+        let signatures = elf::elf64_section(fw.data(), sigs_section)
+            .ok_or(EINVAL)
+            .and_then(|data| DmaObject::from_data(dev, data))?;
+
+        let size = fw_section.len();
+
+        // Move the firmware into a vmalloc'd vector and map it into the device address
+        // space.
+        let fw_vvec = VVec::with_capacity(fw_section.len(), GFP_KERNEL)
+            .and_then(|mut v| {
+                v.extend_from_slice(fw_section, GFP_KERNEL)?;
+                Ok(v)
+            })
+            .map_err(|_| ENOMEM)?;
+
+        Ok(try_pin_init!(Self {
+            fw <- SGTable::new(dev, fw_vvec, DataDirection::ToDevice, GFP_KERNEL),
+            level2 <- {
+                // Allocate the level 2 page table, map the firmware onto it, and map it into the
+                // device address space.
+                VVec::<u8>::with_capacity(
+                    fw.iter().count() * core::mem::size_of::<u64>(),
+                    GFP_KERNEL,
+                )
+                .map_err(|_| ENOMEM)
+                .and_then(|level2| map_into_lvl(&fw, level2))
+                .map(|level2| SGTable::new(dev, level2, DataDirection::ToDevice, GFP_KERNEL))?
+            },
+            level1 <- {
+                // Allocate the level 1 page table, map the level 2 page table onto it, and map it
+                // into the device address space.
+                VVec::<u8>::with_capacity(
+                    level2.iter().count() * core::mem::size_of::<u64>(),
+                    GFP_KERNEL,
+                )
+                .map_err(|_| ENOMEM)
+                .and_then(|level1| map_into_lvl(&level2, level1))
+                .map(|level1| SGTable::new(dev, level1, DataDirection::ToDevice, GFP_KERNEL))?
+            },
+            level0: {
+                // Allocate the level 0 page table as a device-visible DMA object, and map the
+                // level 1 page table onto it.
+
+                // Level 0 page table data.
+                let mut level0_data = kvec![0u8; GSP_PAGE_SIZE]?;
+
+                // Fill level 1 page entry.
+                #[allow(clippy::useless_conversion)]
+                let level1_entry = u64::from(level1.iter().next().unwrap().dma_address());
+                let dst = &mut level0_data[..size_of_val(&level1_entry)];
+                dst.copy_from_slice(&level1_entry.to_le_bytes());
+
+                // Turn the level0 page table into a [`DmaObject`].
+                DmaObject::from_data(dev, &level0_data)?
+            },
+            size,
+            signatures,
+        }))
+    }
+
+    #[expect(unused)]
+    /// Returns the DMA handle of the radix3 level 0 page table.
+    pub(crate) fn radix3_dma_handle(&self) -> DmaAddress {
+        self.level0.dma_handle()
+    }
+}
+
+/// Build a page table from a scatter-gather list.
+///
+/// Takes each DMA-mapped region from `sg_table` and writes page table entries
+/// for all 4KB pages within that region. For example, a 16KB SG entry becomes
+/// 4 consecutive page table entries.
+fn map_into_lvl(sg_table: &SGTable<Owned<VVec<u8>>>, mut dst: VVec<u8>) -> Result<VVec<u8>> {
+    for sg_entry in sg_table.iter() {
+        // Number of pages we need to map.
+        let num_pages = (sg_entry.dma_len() as usize).div_ceil(GSP_PAGE_SIZE);
+
+        for i in 0..num_pages {
+            let entry = sg_entry.dma_address() + (i as u64 * GSP_PAGE_SIZE as u64);
+            dst.extend_from_slice(&entry.to_le_bytes(), GFP_KERNEL)?;
+        }
+    }
+
+    Ok(dst)
+}
diff --git a/drivers/gpu/nova-core/gsp.rs b/drivers/gpu/nova-core/gsp.rs
index 33cc4f2011c1..917e8a83d293 100644
--- a/drivers/gpu/nova-core/gsp.rs
+++ b/drivers/gpu/nova-core/gsp.rs
@@ -4,6 +4,9 @@ mod boot;
 
 use kernel::prelude::*;
 
+pub(crate) const GSP_PAGE_SHIFT: usize = 12;
+pub(crate) const GSP_PAGE_SIZE: usize = 1 << GSP_PAGE_SHIFT;
+
 /// GSP runtime data.
 ///
 /// This is an empty pinned placeholder for now.
diff --git a/drivers/gpu/nova-core/gsp/boot.rs b/drivers/gpu/nova-core/gsp/boot.rs
index faa553635b7c..2800f3aee37d 100644
--- a/drivers/gpu/nova-core/gsp/boot.rs
+++ b/drivers/gpu/nova-core/gsp/boot.rs
@@ -10,6 +10,7 @@ use crate::fb::FbLayout;
 use crate::firmware::{
     booter::{BooterFirmware, BooterKind},
     fwsec::{FwsecCommand, FwsecFirmware},
+    gsp::GspFirmware,
     FIRMWARE_VERSION,
 };
 use crate::gpu::Chipset;
@@ -112,6 +113,11 @@ impl super::Gsp {
 
         let bios = Vbios::new(dev, bar)?;
 
+        let _gsp_fw = KBox::pin_init(
+            GspFirmware::new(dev, chipset, FIRMWARE_VERSION)?,
+            GFP_KERNEL,
+        )?;
+
         let fb_layout = FbLayout::new(chipset, bar)?;
         dev_dbg!(dev, "{:#x?}\n", fb_layout);
 

From 015b1d36505a1ad6830dc2ead65db2d36a209c23 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:21 +0900
Subject: [PATCH 1237/1455] gpu: nova-core: firmware: process the GSP
 bootloader

The GSP bootloader is a small RISC-V firmware that is loaded by Booter
onto the GSP core and is in charge of loading, validating, and starting
the actual GSP firmware.

It is a regular binary firmware file containing a specific header.
Create a type holding the DMA-mapped firmware as well as useful
information extracted from the header, and hook it into our firmware
structure for later use.

The GSP bootloader is stored into the `GspFirmware` structure, since it
is part of the GSP firmware package. This makes the `Firmware` structure
empty, so remove it.

Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-8-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/firmware.rs       | 18 +----
 drivers/gpu/nova-core/firmware/gsp.rs   |  7 ++
 drivers/gpu/nova-core/firmware/riscv.rs | 91 +++++++++++++++++++++++++
 drivers/gpu/nova-core/gpu.rs            |  4 --
 4 files changed, 99 insertions(+), 21 deletions(-)
 create mode 100644 drivers/gpu/nova-core/firmware/riscv.rs

diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index 10200d11bbbf..dc9645a789b0 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -15,11 +15,11 @@ use kernel::transmute::FromBytes;
 use crate::dma::DmaObject;
 use crate::falcon::FalconFirmware;
 use crate::gpu;
-use crate::gpu::Chipset;
 
 pub(crate) mod booter;
 pub(crate) mod fwsec;
 pub(crate) mod gsp;
+pub(crate) mod riscv;
 
 pub(crate) const FIRMWARE_VERSION: &str = "535.113.01";
 
@@ -36,22 +36,6 @@ fn request_firmware(
         .and_then(|path| firmware::Firmware::request(&path, dev))
 }
 
-/// Structure encapsulating the firmware blobs required for the GPU to operate.
-#[expect(dead_code)]
-pub(crate) struct Firmware {
-    bootloader: firmware::Firmware,
-}
-
-impl Firmware {
-    pub(crate) fn new(dev: &device::Device, chipset: Chipset, ver: &str) -> Result<Firmware> {
-        let request = |name| request_firmware(dev, chipset, name, ver);
-
-        Ok(Firmware {
-            bootloader: request("bootloader")?,
-        })
-    }
-}
-
 /// Structure used to describe some firmwares, notably FWSEC-FRTS.
 #[repr(C)]
 #[derive(Debug, Clone)]
diff --git a/drivers/gpu/nova-core/firmware/gsp.rs b/drivers/gpu/nova-core/firmware/gsp.rs
index a7eda9ca0d21..9b70095434c6 100644
--- a/drivers/gpu/nova-core/firmware/gsp.rs
+++ b/drivers/gpu/nova-core/firmware/gsp.rs
@@ -9,6 +9,7 @@ use kernel::prelude::*;
 use kernel::scatterlist::{Owned, SGTable};
 
 use crate::dma::DmaObject;
+use crate::firmware::riscv::RiscvFirmware;
 use crate::gpu::{Architecture, Chipset};
 use crate::gsp::GSP_PAGE_SIZE;
 
@@ -131,6 +132,8 @@ pub(crate) struct GspFirmware {
     size: usize,
     /// Device-mapped GSP signatures matching the GPU's [`Chipset`].
     signatures: DmaObject,
+    /// GSP bootloader, verifies the GSP firmware before loading and running it.
+    bootloader: RiscvFirmware,
 }
 
 impl GspFirmware {
@@ -164,6 +167,9 @@ impl GspFirmware {
             })
             .map_err(|_| ENOMEM)?;
 
+        let bl = super::request_firmware(dev, chipset, "bootloader", ver)?;
+        let bootloader = RiscvFirmware::new(dev, &bl)?;
+
         Ok(try_pin_init!(Self {
             fw <- SGTable::new(dev, fw_vvec, DataDirection::ToDevice, GFP_KERNEL),
             level2 <- {
@@ -206,6 +212,7 @@ impl GspFirmware {
             },
             size,
             signatures,
+            bootloader,
         }))
     }
 
diff --git a/drivers/gpu/nova-core/firmware/riscv.rs b/drivers/gpu/nova-core/firmware/riscv.rs
new file mode 100644
index 000000000000..afb08f5bc4ba
--- /dev/null
+++ b/drivers/gpu/nova-core/firmware/riscv.rs
@@ -0,0 +1,91 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! Support for firmware binaries designed to run on a RISC-V core. Such firmwares files have a
+//! dedicated header.
+
+use core::mem::size_of;
+
+use kernel::device;
+use kernel::firmware::Firmware;
+use kernel::prelude::*;
+use kernel::transmute::FromBytes;
+
+use crate::dma::DmaObject;
+use crate::firmware::BinFirmware;
+
+/// Descriptor for microcode running on a RISC-V core.
+#[repr(C)]
+#[derive(Debug)]
+struct RmRiscvUCodeDesc {
+    version: u32,
+    bootloader_offset: u32,
+    bootloader_size: u32,
+    bootloader_param_offset: u32,
+    bootloader_param_size: u32,
+    riscv_elf_offset: u32,
+    riscv_elf_size: u32,
+    app_version: u32,
+    manifest_offset: u32,
+    manifest_size: u32,
+    monitor_data_offset: u32,
+    monitor_data_size: u32,
+    monitor_code_offset: u32,
+    monitor_code_size: u32,
+}
+
+// SAFETY: all bit patterns are valid for this type, and it doesn't use interior mutability.
+unsafe impl FromBytes for RmRiscvUCodeDesc {}
+
+impl RmRiscvUCodeDesc {
+    /// Interprets the header of `bin_fw` as a [`RmRiscvUCodeDesc`] and returns it.
+    ///
+    /// Fails if the header pointed at by `bin_fw` is not within the bounds of the firmware image.
+    fn new(bin_fw: &BinFirmware<'_>) -> Result<Self> {
+        let offset = bin_fw.hdr.header_offset as usize;
+
+        bin_fw
+            .fw
+            .get(offset..offset + size_of::<Self>())
+            .and_then(Self::from_bytes_copy)
+            .ok_or(EINVAL)
+    }
+}
+
+/// A parsed firmware for a RISC-V core, ready to be loaded and run.
+#[expect(unused)]
+pub(crate) struct RiscvFirmware {
+    /// Offset at which the code starts in the firmware image.
+    code_offset: u32,
+    /// Offset at which the data starts in the firmware image.
+    data_offset: u32,
+    /// Offset at which the manifest starts in the firmware image.
+    manifest_offset: u32,
+    /// Application version.
+    app_version: u32,
+    /// Device-mapped firmware image.
+    ucode: DmaObject,
+}
+
+impl RiscvFirmware {
+    /// Parses the RISC-V firmware image contained in `fw`.
+    pub(crate) fn new(dev: &device::Device<device::Bound>, fw: &Firmware) -> Result<Self> {
+        let bin_fw = BinFirmware::new(fw)?;
+
+        let riscv_desc = RmRiscvUCodeDesc::new(&bin_fw)?;
+
+        let ucode = {
+            let start = bin_fw.hdr.data_offset as usize;
+            let len = bin_fw.hdr.data_size as usize;
+
+            DmaObject::from_data(dev, fw.data().get(start..start + len).ok_or(EINVAL)?)?
+        };
+
+        Ok(Self {
+            ucode,
+            code_offset: riscv_desc.monitor_code_offset,
+            data_offset: riscv_desc.monitor_data_offset,
+            manifest_offset: riscv_desc.manifest_offset,
+            app_version: riscv_desc.app_version,
+        })
+    }
+}
diff --git a/drivers/gpu/nova-core/gpu.rs b/drivers/gpu/nova-core/gpu.rs
index d2395335727a..5da9ad726483 100644
--- a/drivers/gpu/nova-core/gpu.rs
+++ b/drivers/gpu/nova-core/gpu.rs
@@ -5,7 +5,6 @@ use kernel::{device, devres::Devres, error::code::*, pci, prelude::*, sync::Arc}
 use crate::driver::Bar0;
 use crate::falcon::{gsp::Gsp as GspFalcon, sec2::Sec2 as Sec2Falcon, Falcon};
 use crate::fb::SysmemFlush;
-use crate::firmware::{Firmware, FIRMWARE_VERSION};
 use crate::gfw;
 use crate::gsp::Gsp;
 use crate::regs;
@@ -175,7 +174,6 @@ pub(crate) struct Gpu {
     spec: Spec,
     /// MMIO mapping of PCI BAR 0
     bar: Arc<Devres<Bar0>>,
-    fw: Firmware,
     /// System memory page required for flushing all pending GPU-side memory writes done through
     /// PCIE into system memory, via sysmembar (A GPU-initiated HW memory-barrier operation).
     sysmem_flush: SysmemFlush,
@@ -211,8 +209,6 @@ impl Gpu {
                     .inspect_err(|_| dev_err!(pdev.as_ref(), "GFW boot did not complete"))?;
             },
 
-            fw <- Firmware::new(pdev.as_ref(), spec.chipset, FIRMWARE_VERSION)?,
-
             sysmem_flush: SysmemFlush::register(pdev.as_ref(), bar, spec.chipset)?,
 
             gsp_falcon: Falcon::new(

From cb05748b4295d811e47f3adc324b021a919cd4d5 Mon Sep 17 00:00:00 2001
From: Alexandre Courbot <acourbot@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:22 +0900
Subject: [PATCH 1238/1455] gpu: nova-core: firmware: use 570.144 firmware

570.144 is the latest available into linux-firmware as of this commit,
and the one we will use to start development of nova-core. It should
eventually be dropped for a newer version before the driver becomes able
to do anything useful. The newer firmware is expected to iron out some
of the inelegances of 570.144, notably related to packaging.

Reviewed-by: John Hubbard <jhubbard@nvidia.com>
Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-9-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/firmware.rs | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/nova-core/firmware.rs b/drivers/gpu/nova-core/firmware.rs
index dc9645a789b0..4179a74a2342 100644
--- a/drivers/gpu/nova-core/firmware.rs
+++ b/drivers/gpu/nova-core/firmware.rs
@@ -21,7 +21,7 @@ pub(crate) mod fwsec;
 pub(crate) mod gsp;
 pub(crate) mod riscv;
 
-pub(crate) const FIRMWARE_VERSION: &str = "535.113.01";
+pub(crate) const FIRMWARE_VERSION: &str = "570.144";
 
 /// Requests the GPU firmware `name` suitable for `chipset`, with version `ver`.
 fn request_firmware(

From 299eb32863e584cfff7c6b667c3e92ae7d4d2bf9 Mon Sep 17 00:00:00 2001
From: Alistair Popple <apopple@nvidia.com>
Date: Sat, 13 Sep 2025 23:12:23 +0900
Subject: [PATCH 1239/1455] gpu: nova-core: Add base files for r570.144
 firmware bindings

Interacting with the GSP currently requires using definitions from C
header files. Rust definitions for the types needed for Nova core will
be generated using the Rust bindgen tool. This patch adds the base
module to allow inclusion of the generated bindings. The generated
bindings themselves are added by subsequent patches when they are first
used.

Currently we only intend to support a single firmware version, 570.144,
with these bindings. Longer term we intend to move to a more stable GSP
interface that isn't tied to specific firmware versions.

Signed-off-by: Alistair Popple <apopple@nvidia.com>
Reviewed-by: John Hubbard <jhubbard@nvidia.com>
[acourbot@nvidia.com: adapt the bindings module comment a bit]
Acked-by: Danilo Krummrich <dakr@kernel.org>
Link: https://lore.kernel.org/r/20250913-nova_firmware-v6-10-9007079548b0@nvidia.com
Signed-off-by: Alexandre Courbot <acourbot@nvidia.com>
---
 drivers/gpu/nova-core/gsp.rs                  |  2 ++
 drivers/gpu/nova-core/gsp/fw.rs               |  7 +++++
 drivers/gpu/nova-core/gsp/fw/r570_144.rs      | 29 +++++++++++++++++++
 .../gpu/nova-core/gsp/fw/r570_144/bindings.rs |  1 +
 4 files changed, 39 insertions(+)
 create mode 100644 drivers/gpu/nova-core/gsp/fw.rs
 create mode 100644 drivers/gpu/nova-core/gsp/fw/r570_144.rs
 create mode 100644 drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs

diff --git a/drivers/gpu/nova-core/gsp.rs b/drivers/gpu/nova-core/gsp.rs
index 917e8a83d293..64e472e7a9d3 100644
--- a/drivers/gpu/nova-core/gsp.rs
+++ b/drivers/gpu/nova-core/gsp.rs
@@ -4,6 +4,8 @@ mod boot;
 
 use kernel::prelude::*;
 
+mod fw;
+
 pub(crate) const GSP_PAGE_SHIFT: usize = 12;
 pub(crate) const GSP_PAGE_SIZE: usize = 1 << GSP_PAGE_SHIFT;
 
diff --git a/drivers/gpu/nova-core/gsp/fw.rs b/drivers/gpu/nova-core/gsp/fw.rs
new file mode 100644
index 000000000000..34226dd00982
--- /dev/null
+++ b/drivers/gpu/nova-core/gsp/fw.rs
@@ -0,0 +1,7 @@
+// SPDX-License-Identifier: GPL-2.0
+
+mod r570_144;
+
+// Alias to avoid repeating the version number with every use.
+#[expect(unused)]
+use r570_144 as bindings;
diff --git a/drivers/gpu/nova-core/gsp/fw/r570_144.rs b/drivers/gpu/nova-core/gsp/fw/r570_144.rs
new file mode 100644
index 000000000000..35cb0370a7c9
--- /dev/null
+++ b/drivers/gpu/nova-core/gsp/fw/r570_144.rs
@@ -0,0 +1,29 @@
+// SPDX-License-Identifier: GPL-2.0
+
+//! Firmware bindings.
+//!
+//! Imports the generated bindings by `bindgen`.
+//!
+//! This module may not be directly used. Please abstract or re-export the needed symbols in the
+//! parent module instead.
+
+#![cfg_attr(test, allow(deref_nullptr))]
+#![cfg_attr(test, allow(unaligned_references))]
+#![cfg_attr(test, allow(unsafe_op_in_unsafe_fn))]
+#![allow(
+    dead_code,
+    unused_imports,
+    clippy::all,
+    clippy::undocumented_unsafe_blocks,
+    clippy::ptr_as_ptr,
+    clippy::ref_as_ptr,
+    missing_docs,
+    non_camel_case_types,
+    non_upper_case_globals,
+    non_snake_case,
+    improper_ctypes,
+    unreachable_pub,
+    unsafe_op_in_unsafe_fn
+)]
+use kernel::ffi;
+include!("r570_144/bindings.rs");
diff --git a/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs b/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs
new file mode 100644
index 000000000000..cec594032515
--- /dev/null
+++ b/drivers/gpu/nova-core/gsp/fw/r570_144/bindings.rs
@@ -0,0 +1 @@
+// SPDX-License-Identifier: GPL-2.0

From 0265d0ebb409a25d3bb3a19494e0173d06d9d132 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <nathan@kernel.org>
Date: Mon, 8 Sep 2025 13:24:59 -0700
Subject: [PATCH 1240/1455] drm/pixpaper: Fix return type of
 pixpaper_mode_valid()

When building with -Wincompatible-function-pointer-types-strict, a
warning designed to catch kernel control flow integrity (kCFI) issues at
build time, there is an instance in the new tiny DRM pixpaper driver:

  drivers/gpu/drm/tiny/pixpaper.c:982:16: error: incompatible function pointer types initializing 'enum drm_mode_status (*)(struct drm_crtc *, const struct drm_display_mode *)' with an expression of type 'int (struct drm_crtc *, const struct drm_display_mode *)' [-Werror,-Wincompatible-function-pointer-types-strict]
    982 |         .mode_valid = pixpaper_mode_valid,
        |                       ^~~~~~~~~~~~~~~~~~~

While 'int' and 'enum drm_mode_status' are ABI compatible, hence no
regular warning from -Wincompatible-function-pointer-types, the mismatch
will trigger a kCFI violation when pixpaper_mode_valid() is called
indirectly.

Update the return type of pixpaper_mode_valid() to be
'enum drm_mode_status' to clear up the warning and kCFI violation.

Fixes: c9e70639f591 ("drm: tiny: Add support for Mayqueen Pixpaper e-ink panel")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250908-drm-pixpaper-fix-mode_valid-return-type-v1-1-705ceaf03757@kernel.org
---
 drivers/gpu/drm/tiny/pixpaper.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/tiny/pixpaper.c b/drivers/gpu/drm/tiny/pixpaper.c
index b1379cb5f030..32598fb2fee7 100644
--- a/drivers/gpu/drm/tiny/pixpaper.c
+++ b/drivers/gpu/drm/tiny/pixpaper.c
@@ -968,8 +968,8 @@ static const struct drm_crtc_funcs pixpaper_crtc_funcs = {
 	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
 };
 
-static int pixpaper_mode_valid(struct drm_crtc *crtc,
-			       const struct drm_display_mode *mode)
+static enum drm_mode_status
+pixpaper_mode_valid(struct drm_crtc *crtc, const struct drm_display_mode *mode)
 {
 	if (mode->hdisplay == PIXPAPER_WIDTH &&
 	    mode->vdisplay == PIXPAPER_HEIGHT) {

From 4e445729dc103ff7780ed03da9f8310759ea5dae Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:35 +0530
Subject: [PATCH 1241/1455] drm/bridge: samsung-dsim: support separate LINK and
 DPHY status registers

Exynos7870's DSIM has separate registers for LINK and DPHY status. This
is in contrast to older variants in the driver which use a single
register for both.

Add a driver data flag which indicates that the device variant supports
the legacy status register. Change the register read calls
appropriately.

Suggested-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 22 ++++++++++++++++++----
 include/drm/bridge/samsung-dsim.h     |  1 +
 2 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index b5dd71f6a990..f41650ff7e13 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -31,7 +31,7 @@
 /* returns true iff both arguments logically differs */
 #define NEQV(a, b) (!(a) ^ !(b))
 
-/* DSIM_STATUS */
+/* DSIM_STATUS or DSIM_DPHY_STATUS */
 #define DSIM_STOP_STATE_DAT(x)		(((x) & 0xf) << 0)
 #define DSIM_STOP_STATE_CLK		BIT(8)
 #define DSIM_TX_READY_HS_CLK		BIT(10)
@@ -240,7 +240,9 @@ enum samsung_dsim_transfer_type {
 };
 
 enum reg_idx {
-	DSIM_STATUS_REG,	/* Status register */
+	DSIM_STATUS_REG,	/* Status register (legacy) */
+	DSIM_LINK_STATUS_REG,	/* Link status register */
+	DSIM_DPHY_STATUS_REG,	/* D-PHY status register */
 	DSIM_SWRST_REG,		/* Software reset register */
 	DSIM_CLKCTRL_REG,	/* Clock control register */
 	DSIM_TIMEOUT_REG,	/* Time out register */
@@ -405,6 +407,7 @@ static const unsigned int imx8mm_dsim_reg_values[] = {
 static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x50,
+	.has_legacy_status_reg = 1,
 	.has_freqband = 1,
 	.has_clklane_stop = 1,
 	.num_clks = 2,
@@ -424,6 +427,7 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x50,
+	.has_legacy_status_reg = 1,
 	.has_freqband = 1,
 	.has_clklane_stop = 1,
 	.num_clks = 2,
@@ -443,6 +447,7 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x58,
+	.has_legacy_status_reg = 1,
 	.num_clks = 2,
 	.max_freq = 1000,
 	.wait_for_reset = 1,
@@ -459,6 +464,7 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.reg_ofs = exynos5433_reg_ofs,
 	.plltmr_reg = 0xa0,
+	.has_legacy_status_reg = 1,
 	.has_clklane_stop = 1,
 	.num_clks = 5,
 	.max_freq = 1500,
@@ -476,6 +482,7 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.reg_ofs = exynos5433_reg_ofs,
 	.plltmr_reg = 0xa0,
+	.has_legacy_status_reg = 1,
 	.has_clklane_stop = 1,
 	.num_clks = 2,
 	.max_freq = 1500,
@@ -493,6 +500,7 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.reg_ofs = exynos5433_reg_ofs,
 	.plltmr_reg = 0xa0,
+	.has_legacy_status_reg = 1,
 	.has_clklane_stop = 1,
 	.num_clks = 2,
 	.max_freq = 2100,
@@ -688,7 +696,10 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 			dev_err(dsi->dev, "PLL failed to stabilize\n");
 			return 0;
 		}
-		reg = samsung_dsim_read(dsi, DSIM_STATUS_REG);
+		if (driver_data->has_legacy_status_reg)
+			reg = samsung_dsim_read(dsi, DSIM_STATUS_REG);
+		else
+			reg = samsung_dsim_read(dsi, DSIM_LINK_STATUS_REG);
 	} while ((reg & DSIM_PLL_STABLE) == 0);
 
 	dsi->hs_clock = fout;
@@ -962,7 +973,10 @@ static int samsung_dsim_init_link(struct samsung_dsim *dsi)
 			return -EFAULT;
 		}
 
-		reg = samsung_dsim_read(dsi, DSIM_STATUS_REG);
+		if (driver_data->has_legacy_status_reg)
+			reg = samsung_dsim_read(dsi, DSIM_STATUS_REG);
+		else
+			reg = samsung_dsim_read(dsi, DSIM_DPHY_STATUS_REG);
 		if ((reg & DSIM_STOP_STATE_DAT(lanes_mask))
 		    != DSIM_STOP_STATE_DAT(lanes_mask))
 			continue;
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 9764d6eb5beb..d7877191bad1 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -53,6 +53,7 @@ struct samsung_dsim_transfer {
 struct samsung_dsim_driver_data {
 	const unsigned int *reg_ofs;
 	unsigned int plltmr_reg;
+	unsigned int has_legacy_status_reg:1;
 	unsigned int has_freqband:1;
 	unsigned int has_clklane_stop:1;
 	unsigned int has_broken_fifoctrl_emptyhdr:1;

From 7c9b998947f19457e32496ab9edeea798373c426 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:36 +0530
Subject: [PATCH 1242/1455] drm/bridge: samsung-dsim: add SFRCTRL register

On Exynos7870 devices, enabling the display requires disabling
standby by writing to the SFRCTRL register. Add the register and related
bit values. Since this behavior isn't available on other SoCs, implement
a flag in the driver data struct indicating the availability of this
feature.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 16 ++++++++++++++++
 include/drm/bridge/samsung-dsim.h     |  1 +
 2 files changed, 17 insertions(+)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index f41650ff7e13..49c4456ebd23 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -157,6 +157,11 @@
 #define DSIM_INT_RX_ECC_ERR		BIT(15)
 #define DSIM_INT_RX_CRC_ERR		BIT(14)
 
+/* DSIM_SFRCTRL */
+#define DSIM_SFR_CTRL_STAND_BY		BIT(4)
+#define DSIM_SFR_CTRL_SHADOW_UPDATE	BIT(1)
+#define DSIM_SFR_CTRL_SHADOW_EN		BIT(0)
+
 /* DSIM_FIFOCTRL */
 #define DSIM_RX_DATA_FULL		BIT(25)
 #define DSIM_RX_DATA_EMPTY		BIT(24)
@@ -257,6 +262,7 @@ enum reg_idx {
 	DSIM_PKTHDR_REG,	/* Packet Header FIFO register */
 	DSIM_PAYLOAD_REG,	/* Payload FIFO register */
 	DSIM_RXFIFO_REG,	/* Read FIFO register */
+	DSIM_SFRCTRL_REG,	/* SFR standby and shadow control register */
 	DSIM_FIFOCTRL_REG,	/* FIFO status and control register */
 	DSIM_PLLCTRL_REG,	/* PLL control register */
 	DSIM_PHYCTRL_REG,
@@ -1037,6 +1043,7 @@ static void samsung_dsim_set_display_mode(struct samsung_dsim *dsi)
 
 static void samsung_dsim_set_display_enable(struct samsung_dsim *dsi, bool enable)
 {
+	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
 	u32 reg;
 
 	reg = samsung_dsim_read(dsi, DSIM_MDRESOL_REG);
@@ -1045,6 +1052,15 @@ static void samsung_dsim_set_display_enable(struct samsung_dsim *dsi, bool enabl
 	else
 		reg &= ~DSIM_MAIN_STAND_BY;
 	samsung_dsim_write(dsi, DSIM_MDRESOL_REG, reg);
+
+	if (driver_data->has_sfrctrl) {
+		reg = samsung_dsim_read(dsi, DSIM_SFRCTRL_REG);
+		if (enable)
+			reg |= DSIM_SFR_CTRL_STAND_BY;
+		else
+			reg &= ~DSIM_SFR_CTRL_STAND_BY;
+		samsung_dsim_write(dsi, DSIM_SFRCTRL_REG, reg);
+	}
 }
 
 static int samsung_dsim_wait_for_hdr_fifo(struct samsung_dsim *dsi)
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index d7877191bad1..f0c1e5c5ed49 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -57,6 +57,7 @@ struct samsung_dsim_driver_data {
 	unsigned int has_freqband:1;
 	unsigned int has_clklane_stop:1;
 	unsigned int has_broken_fifoctrl_emptyhdr:1;
+	unsigned int has_sfrctrl:1;
 	unsigned int num_clks;
 	unsigned int min_freq;
 	unsigned int max_freq;

From 92beab1a397d80d04d90f511c6d0af696da67a33 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:37 +0530
Subject: [PATCH 1243/1455] drm/bridge: samsung-dsim: add flag to control
 header FIFO wait

Exynos7870's DSIM device doesn't require waiting for the header FIFO
during a MIPI DSI transfer. Add a flag in the driver data in order to
control said behavior.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 15 ++++++++++++---
 include/drm/bridge/samsung-dsim.h     |  1 +
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 49c4456ebd23..dd6c99dada89 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -418,6 +418,7 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.has_clklane_stop = 1,
 	.num_clks = 2,
 	.max_freq = 1000,
+	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
 	.pll_p_offset = 13,
@@ -438,6 +439,7 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.has_clklane_stop = 1,
 	.num_clks = 2,
 	.max_freq = 1000,
+	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
 	.pll_p_offset = 13,
@@ -456,6 +458,7 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.has_legacy_status_reg = 1,
 	.num_clks = 2,
 	.max_freq = 1000,
+	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
 	.pll_p_offset = 13,
@@ -474,6 +477,7 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.has_clklane_stop = 1,
 	.num_clks = 5,
 	.max_freq = 1500,
+	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
 	.pll_p_offset = 13,
@@ -492,6 +496,7 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.has_clklane_stop = 1,
 	.num_clks = 2,
 	.max_freq = 1500,
+	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 12,
 	.pll_p_offset = 13,
@@ -510,6 +515,7 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.has_clklane_stop = 1,
 	.num_clks = 2,
 	.max_freq = 2100,
+	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
 	/*
@@ -1117,6 +1123,7 @@ static void samsung_dsim_send_to_fifo(struct samsung_dsim *dsi,
 {
 	struct device *dev = dsi->dev;
 	struct mipi_dsi_packet *pkt = &xfer->packet;
+	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
 	const u8 *payload = pkt->payload + xfer->tx_done;
 	u16 length = pkt->payload_length - xfer->tx_done;
 	bool first = !xfer->tx_done;
@@ -1157,9 +1164,11 @@ static void samsung_dsim_send_to_fifo(struct samsung_dsim *dsi,
 		return;
 
 	reg = get_unaligned_le32(pkt->header);
-	if (samsung_dsim_wait_for_hdr_fifo(dsi)) {
-		dev_err(dev, "waiting for header FIFO timed out\n");
-		return;
+	if (driver_data->wait_for_hdr_fifo) {
+		if (samsung_dsim_wait_for_hdr_fifo(dsi)) {
+			dev_err(dev, "waiting for header FIFO timed out\n");
+			return;
+		}
 	}
 
 	if (NEQV(xfer->flags & MIPI_DSI_MSG_USE_LPM,
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index f0c1e5c5ed49..62c07952bd00 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -61,6 +61,7 @@ struct samsung_dsim_driver_data {
 	unsigned int num_clks;
 	unsigned int min_freq;
 	unsigned int max_freq;
+	unsigned int wait_for_hdr_fifo;
 	unsigned int wait_for_reset;
 	unsigned int num_bits_resol;
 	unsigned int pll_p_offset;

From f6ba4c1577a8184105133a79b5d1b6d465158362 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:38 +0530
Subject: [PATCH 1244/1455] drm/bridge: samsung-dsim: allow configuring bits
 and offsets of CLKCTRL register

DSIM_CLKCTRL bit and offset values hardcoded in the driver:

name                      | bit/offset value
--------------------------+-----------------
DSIM_LANE_ESC_CLK_EN_CLK  | 19
DSIM_LANE_ESC_CLK_EN_DATA | 20
DSIM_BYTE_CLKEN           | 24
DSIM_ESC_CLKEN            | 28
DSIM_TX_REQUEST_HSCLK     | 31

DSIM_CLKCTRL bit and offset values in Exynos7870 DSIM as per downstream
kernel sources:

name                      | bit/offset value
--------------------------+-----------------
DSIM_LANE_ESC_CLK_EN_CLK  | 8
DSIM_LANE_ESC_CLK_EN_DATA | 9
DSIM_BYTE_CLKEN           | 17
DSIM_ESC_CLKEN            | 16
DSIM_TX_REQUEST_HSCLK     | 20

In order to support both, move all values to the driver data struct and
define it for every driver compatible. Reference the values from there
instead, in functions wherever required.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 76 +++++++++++++++++++--------
 include/drm/bridge/samsung-dsim.h     |  5 ++
 2 files changed, 59 insertions(+), 22 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index dd6c99dada89..72190787ff13 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -46,17 +46,13 @@
 #define DSIM_BTA_TIMEOUT(x)		((x) << 16)
 
 /* DSIM_CLKCTRL */
-#define DSIM_ESC_PRESCALER(x)		(((x) & 0xffff) << 0)
-#define DSIM_ESC_PRESCALER_MASK		(0xffff << 0)
-#define DSIM_LANE_ESC_CLK_EN_CLK	BIT(19)
-#define DSIM_LANE_ESC_CLK_EN_DATA(x)	(((x) & 0xf) << 20)
-#define DSIM_LANE_ESC_CLK_EN_DATA_MASK	(0xf << 20)
-#define DSIM_BYTE_CLKEN			BIT(24)
-#define DSIM_BYTE_CLK_SRC(x)		(((x) & 0x3) << 25)
-#define DSIM_BYTE_CLK_SRC_MASK		(0x3 << 25)
-#define DSIM_PLL_BYPASS			BIT(27)
-#define DSIM_ESC_CLKEN			BIT(28)
-#define DSIM_TX_REQUEST_HSCLK		BIT(31)
+#define DSIM_ESC_PRESCALER(x)			(((x) & 0xffff) << 0)
+#define DSIM_ESC_PRESCALER_MASK			(0xffff << 0)
+#define DSIM_LANE_ESC_CLK_EN_DATA(x, offset)	(((x) & 0xf) << offset)
+#define DSIM_LANE_ESC_CLK_EN_DATA_MASK(offset)	(0xf << offset)
+#define DSIM_BYTE_CLK_SRC(x)			(((x) & 0x3) << 25)
+#define DSIM_BYTE_CLK_SRC_MASK			(0x3 << 25)
+#define DSIM_PLL_BYPASS				BIT(27)
 
 /* DSIM_CONFIG */
 #define DSIM_LANE_EN_CLK		BIT(0)
@@ -421,6 +417,11 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
+	.esc_clken_bit = 28,
+	.byte_clken_bit = 24,
+	.tx_req_hsclk_bit = 31,
+	.lane_esc_clk_bit = 19,
+	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
@@ -442,6 +443,11 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
+	.esc_clken_bit = 28,
+	.byte_clken_bit = 24,
+	.tx_req_hsclk_bit = 31,
+	.lane_esc_clk_bit = 19,
+	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
@@ -461,6 +467,11 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
+	.esc_clken_bit = 28,
+	.byte_clken_bit = 24,
+	.tx_req_hsclk_bit = 31,
+	.lane_esc_clk_bit = 19,
+	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
@@ -480,6 +491,11 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
+	.esc_clken_bit = 28,
+	.byte_clken_bit = 24,
+	.tx_req_hsclk_bit = 31,
+	.lane_esc_clk_bit = 19,
+	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
 	.reg_values = exynos5433_reg_values,
 	.pll_fin_min = 6,
@@ -499,6 +515,11 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 12,
+	.esc_clken_bit = 28,
+	.byte_clken_bit = 24,
+	.tx_req_hsclk_bit = 31,
+	.lane_esc_clk_bit = 19,
+	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
 	.reg_values = exynos5422_reg_values,
 	.pll_fin_min = 6,
@@ -518,6 +539,11 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
+	.esc_clken_bit = 28,
+	.byte_clken_bit = 24,
+	.tx_req_hsclk_bit = 31,
+	.lane_esc_clk_bit = 19,
+	.lane_esc_data_offset = 20,
 	/*
 	 * Unlike Exynos, PLL_P(PMS_P) offset 14 is used in i.MX8M Mini/Nano/Plus
 	 * downstream driver - drivers/gpu/drm/bridge/sec-dsim.c
@@ -721,6 +747,7 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 
 static int samsung_dsim_enable_clock(struct samsung_dsim *dsi)
 {
+	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
 	unsigned long hs_clk, byte_clk, esc_clk, pix_clk;
 	unsigned long esc_div;
 	u32 reg;
@@ -754,15 +781,17 @@ static int samsung_dsim_enable_clock(struct samsung_dsim *dsi)
 		hs_clk, byte_clk, esc_clk);
 
 	reg = samsung_dsim_read(dsi, DSIM_CLKCTRL_REG);
-	reg &= ~(DSIM_ESC_PRESCALER_MASK | DSIM_LANE_ESC_CLK_EN_CLK
-			| DSIM_LANE_ESC_CLK_EN_DATA_MASK | DSIM_PLL_BYPASS
-			| DSIM_BYTE_CLK_SRC_MASK);
-	reg |= DSIM_ESC_CLKEN | DSIM_BYTE_CLKEN
-			| DSIM_ESC_PRESCALER(esc_div)
-			| DSIM_LANE_ESC_CLK_EN_CLK
-			| DSIM_LANE_ESC_CLK_EN_DATA(BIT(dsi->lanes) - 1)
-			| DSIM_BYTE_CLK_SRC(0)
-			| DSIM_TX_REQUEST_HSCLK;
+	reg &= ~(DSIM_ESC_PRESCALER_MASK | BIT(driver_data->lane_esc_clk_bit)
+		| DSIM_LANE_ESC_CLK_EN_DATA_MASK(driver_data->lane_esc_data_offset)
+		| DSIM_PLL_BYPASS
+		| DSIM_BYTE_CLK_SRC_MASK);
+	reg |= BIT(driver_data->esc_clken_bit) | BIT(driver_data->byte_clken_bit)
+		| DSIM_ESC_PRESCALER(esc_div)
+		| BIT(driver_data->lane_esc_clk_bit)
+		| DSIM_LANE_ESC_CLK_EN_DATA(BIT(dsi->lanes) - 1,
+					    driver_data->lane_esc_data_offset)
+		| DSIM_BYTE_CLK_SRC(0)
+		| BIT(driver_data->tx_req_hsclk_bit);
 	samsung_dsim_write(dsi, DSIM_CLKCTRL_REG, reg);
 
 	return 0;
@@ -866,11 +895,14 @@ static void samsung_dsim_set_phy_ctrl(struct samsung_dsim *dsi)
 
 static void samsung_dsim_disable_clock(struct samsung_dsim *dsi)
 {
+	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
 	u32 reg;
 
 	reg = samsung_dsim_read(dsi, DSIM_CLKCTRL_REG);
-	reg &= ~(DSIM_LANE_ESC_CLK_EN_CLK | DSIM_LANE_ESC_CLK_EN_DATA_MASK
-			| DSIM_ESC_CLKEN | DSIM_BYTE_CLKEN);
+	reg &= ~(BIT(driver_data->lane_esc_clk_bit)
+		| DSIM_LANE_ESC_CLK_EN_DATA_MASK(driver_data->lane_esc_data_offset)
+		| BIT(driver_data->esc_clken_bit)
+		| BIT(driver_data->byte_clken_bit));
 	samsung_dsim_write(dsi, DSIM_CLKCTRL_REG, reg);
 
 	reg = samsung_dsim_read(dsi, DSIM_PLLCTRL_REG);
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 62c07952bd00..b1e64c7f9931 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -64,6 +64,11 @@ struct samsung_dsim_driver_data {
 	unsigned int wait_for_hdr_fifo;
 	unsigned int wait_for_reset;
 	unsigned int num_bits_resol;
+	unsigned int esc_clken_bit;
+	unsigned int byte_clken_bit;
+	unsigned int tx_req_hsclk_bit;
+	unsigned int lane_esc_clk_bit;
+	unsigned int lane_esc_data_offset;
 	unsigned int pll_p_offset;
 	const unsigned int *reg_values;
 	unsigned int pll_fin_min;

From 4d244122dd90c72f6c3f10eb7a53678d78d3b857 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:39 +0530
Subject: [PATCH 1245/1455] drm/bridge: samsung-dsim: allow configuring the
 MAIN_VSA offset

The MAIN_VSA offset of DSIM_MSYNC is hardcoded to a 22-bit offset, but
Exynos7870's DSIM has it in a 16-bit offset as per the downstream kernel
sources.

In order to support both, move this offset value to the driver data
struct and define it for every driver compatible. Reference the value
from there instead, in functions wherever required.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 13 ++++++++++---
 include/drm/bridge/samsung-dsim.h     |  1 +
 2 files changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 72190787ff13..325acad1f577 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -125,9 +125,9 @@
 #define DSIM_MAIN_HBP_MASK		((0xffff) << 0)
 
 /* DSIM_MSYNC */
-#define DSIM_MAIN_VSA(x)		((x) << 22)
+#define DSIM_MAIN_VSA(x, offset)	((x) << offset)
 #define DSIM_MAIN_HSA(x)		((x) << 0)
-#define DSIM_MAIN_VSA_MASK		((0x3ff) << 22)
+#define DSIM_MAIN_VSA_MASK(offset)	((0x3ff) << offset)
 #define DSIM_MAIN_HSA_MASK		((0xffff) << 0)
 
 /* DSIM_SDRESOL */
@@ -423,6 +423,7 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.main_vsa_offset = 22,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
 	.pll_fin_max = 12,
@@ -449,6 +450,7 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.main_vsa_offset = 22,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
 	.pll_fin_max = 12,
@@ -473,6 +475,7 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.main_vsa_offset = 22,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
 	.pll_fin_max = 12,
@@ -497,6 +500,7 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.main_vsa_offset = 22,
 	.reg_values = exynos5433_reg_values,
 	.pll_fin_min = 6,
 	.pll_fin_max = 12,
@@ -521,6 +525,7 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.main_vsa_offset = 22,
 	.reg_values = exynos5422_reg_values,
 	.pll_fin_min = 6,
 	.pll_fin_max = 12,
@@ -549,6 +554,7 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	 * downstream driver - drivers/gpu/drm/bridge/sec-dsim.c
 	 */
 	.pll_p_offset = 14,
+	.main_vsa_offset = 22,
 	.reg_values = imx8mm_dsim_reg_values,
 	.pll_fin_min = 2,
 	.pll_fin_max = 30,
@@ -1041,6 +1047,7 @@ static void samsung_dsim_set_display_mode(struct samsung_dsim *dsi)
 {
 	struct drm_display_mode *m = &dsi->mode;
 	unsigned int num_bits_resol = dsi->driver_data->num_bits_resol;
+	unsigned int main_vsa_offset = dsi->driver_data->main_vsa_offset;
 	u32 reg;
 
 	if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
@@ -1067,7 +1074,7 @@ static void samsung_dsim_set_display_mode(struct samsung_dsim *dsi)
 		reg = DSIM_MAIN_HFP(hfp) | DSIM_MAIN_HBP(hbp);
 		samsung_dsim_write(dsi, DSIM_MHPORCH_REG, reg);
 
-		reg = DSIM_MAIN_VSA(m->vsync_end - m->vsync_start)
+		reg = DSIM_MAIN_VSA(m->vsync_end - m->vsync_start, main_vsa_offset)
 			| DSIM_MAIN_HSA(hsa);
 		samsung_dsim_write(dsi, DSIM_MSYNC_REG, reg);
 	}
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index b1e64c7f9931..7f6d353f34af 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -70,6 +70,7 @@ struct samsung_dsim_driver_data {
 	unsigned int lane_esc_clk_bit;
 	unsigned int lane_esc_data_offset;
 	unsigned int pll_p_offset;
+	unsigned int main_vsa_offset;
 	const unsigned int *reg_values;
 	unsigned int pll_fin_min;
 	unsigned int pll_fin_max;

From d6dbefb2fed7d7f333c4241965296d84c202b6bf Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:40 +0530
Subject: [PATCH 1246/1455] drm/bridge: samsung-dsim: allow configuring the
 VIDEO_MODE bit

The VIDEO_MODE bit of DSIM_CONFIG is hardcoded to BIT(25), but
Exynos7870's DSIM has it in BIT(18) as per downstream kernel sources.

In order to support both, move this bit value to the driver data struct
and define it for every driver compatible. Reference the value from
there instead, in functions wherever required.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 9 +++++++--
 include/drm/bridge/samsung-dsim.h     | 1 +
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 325acad1f577..db154418e446 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -87,7 +87,6 @@
  */
 #define DSIM_HSE_DISABLE_MODE		BIT(23)
 #define DSIM_AUTO_MODE			BIT(24)
-#define DSIM_VIDEO_MODE			BIT(25)
 #define DSIM_BURST_MODE			BIT(26)
 #define DSIM_SYNC_INFORM		BIT(27)
 #define DSIM_EOT_DISABLE		BIT(28)
@@ -417,6 +416,7 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
+	.video_mode_bit = 25,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -444,6 +444,7 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
+	.video_mode_bit = 25,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -469,6 +470,7 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
+	.video_mode_bit = 25,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -494,6 +496,7 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
+	.video_mode_bit = 25,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -519,6 +522,7 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
 	.num_bits_resol = 12,
+	.video_mode_bit = 25,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -544,6 +548,7 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
+	.video_mode_bit = 25,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -952,7 +957,7 @@ static int samsung_dsim_init_link(struct samsung_dsim *dsi)
 	 * mode, otherwise it will support command mode.
 	 */
 	if (dsi->mode_flags & MIPI_DSI_MODE_VIDEO) {
-		reg |= DSIM_VIDEO_MODE;
+		reg |= BIT(driver_data->video_mode_bit);
 
 		/*
 		 * The user manual describes that following bits are ignored in
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 7f6d353f34af..9d11c3e39fe5 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -64,6 +64,7 @@ struct samsung_dsim_driver_data {
 	unsigned int wait_for_hdr_fifo;
 	unsigned int wait_for_reset;
 	unsigned int num_bits_resol;
+	unsigned int video_mode_bit;
 	unsigned int esc_clken_bit;
 	unsigned int byte_clken_bit;
 	unsigned int tx_req_hsclk_bit;

From 9aa49c21aac071383353315036520ba753484c93 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:41 +0530
Subject: [PATCH 1247/1455] drm/bridge: samsung-dsim: allow configuring PLL_M
 and PLL_S offsets

Currently, PLL_P offset of DSIM_PLLCTRL is configurable in the driver
data, while PLL_M and PLL_S offsets are hardcoded as 4-bit and 1-bit
offsets respectively, but Exynos7870's DSIM have them at 3-bit and 0-bit
offsets as per downstream kernel sources.

In order to support both, move both offset values to the driver data
struct and define it for every driver compatible. Reference the values
from there instead, in functions wherever required.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 21 ++++++++++++++++-----
 include/drm/bridge/samsung-dsim.h     |  2 ++
 2 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index db154418e446..2884b3b7dc73 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -191,9 +191,7 @@
 #define DSIM_PLL_DPDNSWAP_DAT		(1 << 24)
 #define DSIM_FREQ_BAND(x)		((x) << 24)
 #define DSIM_PLL_EN			BIT(23)
-#define DSIM_PLL_P(x, offset)		((x) << (offset))
-#define DSIM_PLL_M(x)			((x) << 4)
-#define DSIM_PLL_S(x)			((x) << 1)
+#define DSIM_PLL(x, offset)		((x) << (offset))
 
 /* DSIM_PHYCTRL */
 #define DSIM_PHYCTRL_ULPS_EXIT(x)	(((x) & 0x1ff) << 0)
@@ -423,6 +421,8 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.pll_m_offset = 4,
+	.pll_s_offset = 1,
 	.main_vsa_offset = 22,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
@@ -451,6 +451,8 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.pll_m_offset = 4,
+	.pll_s_offset = 1,
 	.main_vsa_offset = 22,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
@@ -477,6 +479,8 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.pll_m_offset = 4,
+	.pll_s_offset = 1,
 	.main_vsa_offset = 22,
 	.reg_values = reg_values,
 	.pll_fin_min = 6,
@@ -503,6 +507,8 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.pll_m_offset = 4,
+	.pll_s_offset = 1,
 	.main_vsa_offset = 22,
 	.reg_values = exynos5433_reg_values,
 	.pll_fin_min = 6,
@@ -529,6 +535,8 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.lane_esc_clk_bit = 19,
 	.lane_esc_data_offset = 20,
 	.pll_p_offset = 13,
+	.pll_m_offset = 4,
+	.pll_s_offset = 1,
 	.main_vsa_offset = 22,
 	.reg_values = exynos5422_reg_values,
 	.pll_fin_min = 6,
@@ -559,6 +567,8 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	 * downstream driver - drivers/gpu/drm/bridge/sec-dsim.c
 	 */
 	.pll_p_offset = 14,
+	.pll_m_offset = 4,
+	.pll_s_offset = 1,
 	.main_vsa_offset = 22,
 	.reg_values = imx8mm_dsim_reg_values,
 	.pll_fin_min = 2,
@@ -710,8 +720,9 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 	writel(driver_data->reg_values[PLL_TIMER],
 	       dsi->reg_base + driver_data->plltmr_reg);
 
-	reg = DSIM_PLL_EN | DSIM_PLL_P(p, driver_data->pll_p_offset) |
-	      DSIM_PLL_M(m) | DSIM_PLL_S(s);
+	reg = DSIM_PLL_EN | DSIM_PLL(p, driver_data->pll_p_offset)
+			  | DSIM_PLL(m, driver_data->pll_m_offset)
+			  | DSIM_PLL(s, driver_data->pll_s_offset);
 
 	if (driver_data->has_freqband) {
 		static const unsigned long freq_bands[] = {
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 9d11c3e39fe5..000ada3ece4d 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -71,6 +71,8 @@ struct samsung_dsim_driver_data {
 	unsigned int lane_esc_clk_bit;
 	unsigned int lane_esc_data_offset;
 	unsigned int pll_p_offset;
+	unsigned int pll_m_offset;
+	unsigned int pll_s_offset;
 	unsigned int main_vsa_offset;
 	const unsigned int *reg_values;
 	unsigned int pll_fin_min;

From f7754d843a05c685ba453be176a29ae157f88b0c Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:42 +0530
Subject: [PATCH 1248/1455] drm/bridge: samsung-dsim: allow configuring the
 PLL_STABLE bit

The PLL_STABLE bit of DSIM_DPHY_STATUS is hardcoded to BIT(31), but
Exynos7870's DSIM has it in BIT(24) as per downstream kernel sources.

In order to support both, move this bit value to the driver data struct
and define it for every driver compatible. Reference the value from
there instead, in functions wherever required.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 9 +++++++--
 include/drm/bridge/samsung-dsim.h     | 1 +
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 2884b3b7dc73..456d961abc05 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -35,7 +35,6 @@
 #define DSIM_STOP_STATE_DAT(x)		(((x) & 0xf) << 0)
 #define DSIM_STOP_STATE_CLK		BIT(8)
 #define DSIM_TX_READY_HS_CLK		BIT(10)
-#define DSIM_PLL_STABLE			BIT(31)
 
 /* DSIM_SWRST */
 #define DSIM_FUNCRST			BIT(16)
@@ -415,6 +414,7 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
 	.video_mode_bit = 25,
+	.pll_stable_bit = 31,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -445,6 +445,7 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
 	.video_mode_bit = 25,
+	.pll_stable_bit = 31,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -473,6 +474,7 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.wait_for_reset = 1,
 	.num_bits_resol = 11,
 	.video_mode_bit = 25,
+	.pll_stable_bit = 31,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -501,6 +503,7 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
 	.video_mode_bit = 25,
+	.pll_stable_bit = 31,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -529,6 +532,7 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.wait_for_reset = 1,
 	.num_bits_resol = 12,
 	.video_mode_bit = 25,
+	.pll_stable_bit = 31,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -557,6 +561,7 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.wait_for_reset = 0,
 	.num_bits_resol = 12,
 	.video_mode_bit = 25,
+	.pll_stable_bit = 31,
 	.esc_clken_bit = 28,
 	.byte_clken_bit = 24,
 	.tx_req_hsclk_bit = 31,
@@ -760,7 +765,7 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 			reg = samsung_dsim_read(dsi, DSIM_STATUS_REG);
 		else
 			reg = samsung_dsim_read(dsi, DSIM_LINK_STATUS_REG);
-	} while ((reg & DSIM_PLL_STABLE) == 0);
+	} while ((reg & BIT(driver_data->pll_stable_bit)) == 0);
 
 	dsi->hs_clock = fout;
 
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 000ada3ece4d..04ed11787bbd 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -65,6 +65,7 @@ struct samsung_dsim_driver_data {
 	unsigned int wait_for_reset;
 	unsigned int num_bits_resol;
 	unsigned int video_mode_bit;
+	unsigned int pll_stable_bit;
 	unsigned int esc_clken_bit;
 	unsigned int byte_clken_bit;
 	unsigned int tx_req_hsclk_bit;

From 7ef93667ab19370a2a05b8dbcad36d76b887ff38 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:43 +0530
Subject: [PATCH 1249/1455] drm/bridge: samsung-dsim: increase timeout value
 for PLL_STABLE

Exynos7870's DSIM requires more time to stabilize its PLL. The current
timeout value, 1000, doesn't suffice. Increase the value to 3000, which
is just about enough as observed experimentally.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 456d961abc05..2e422f3ec103 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -755,7 +755,7 @@ static unsigned long samsung_dsim_set_pll(struct samsung_dsim *dsi,
 
 	samsung_dsim_write(dsi, DSIM_PLLCTRL_REG, reg);
 
-	timeout = 1000;
+	timeout = 3000;
 	do {
 		if (timeout-- == 0) {
 			dev_err(dsi->dev, "PLL failed to stabilize\n");

From f08051a4158fec363e1f33b75dd48131f524fa5f Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:44 +0530
Subject: [PATCH 1250/1455] drm/bridge: samsung-dsim: add ability to define
 clock names for every variant

Presently, all devices refer to clock names from a single array. The
only controlling parameter is the number of clocks (num_clks field of
samsung_dsim_driver_data) which uses the first n clocks of that array.
As new devices are added, this approach turns out to be cumbersome.

Separate the clock names in individual arrays required by each variant,
in a struct clk_bulk_data. Add a pointer field to the driver data struct
which points to their respective clock names, and rework the clock usage
code to use the clk_bulk_* API instead.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 88 +++++++++++++--------------
 include/drm/bridge/samsung-dsim.h     |  2 +-
 2 files changed, 44 insertions(+), 46 deletions(-)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 2e422f3ec103..802164e6adff 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -219,23 +219,31 @@
 #define DSI_XFER_TIMEOUT_MS		100
 #define DSI_RX_FIFO_EMPTY		0x30800002
 
-#define OLD_SCLK_MIPI_CLK_NAME		"pll_clk"
-
 #define PS_TO_CYCLE(ps, hz) DIV64_U64_ROUND_CLOSEST(((ps) * (hz)), 1000000000000ULL)
 
-static const char *const clk_names[5] = {
-	"bus_clk",
-	"sclk_mipi",
-	"phyclk_mipidphy0_bitclkdiv8",
-	"phyclk_mipidphy0_rxclkesc0",
-	"sclk_rgb_vclk_to_dsim0"
-};
-
 enum samsung_dsim_transfer_type {
 	EXYNOS_DSI_TX,
 	EXYNOS_DSI_RX,
 };
 
+static struct clk_bulk_data exynos3_clk_bulk_data[] = {
+	{ .id = "bus_clk" },
+	{ .id = "pll_clk" },
+};
+
+static struct clk_bulk_data exynos4_clk_bulk_data[] = {
+	{ .id = "bus_clk" },
+	{ .id = "sclk_mipi" },
+};
+
+static struct clk_bulk_data exynos5433_clk_bulk_data[] = {
+	{ .id = "bus_clk" },
+	{ .id = "sclk_mipi" },
+	{ .id = "phyclk_mipidphy0_bitclkdiv8" },
+	{ .id = "phyclk_mipidphy0_rxclkesc0" },
+	{ .id = "sclk_rgb_vclk_to_dsim0" },
+};
+
 enum reg_idx {
 	DSIM_STATUS_REG,	/* Status register (legacy) */
 	DSIM_LINK_STATUS_REG,	/* Link status register */
@@ -408,7 +416,8 @@ static const struct samsung_dsim_driver_data exynos3_dsi_driver_data = {
 	.has_legacy_status_reg = 1,
 	.has_freqband = 1,
 	.has_clklane_stop = 1,
-	.num_clks = 2,
+	.clk_data = exynos3_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos3_clk_bulk_data),
 	.max_freq = 1000,
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
@@ -439,7 +448,8 @@ static const struct samsung_dsim_driver_data exynos4_dsi_driver_data = {
 	.has_legacy_status_reg = 1,
 	.has_freqband = 1,
 	.has_clklane_stop = 1,
-	.num_clks = 2,
+	.clk_data = exynos4_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos4_clk_bulk_data),
 	.max_freq = 1000,
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
@@ -468,7 +478,8 @@ static const struct samsung_dsim_driver_data exynos5_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x58,
 	.has_legacy_status_reg = 1,
-	.num_clks = 2,
+	.clk_data = exynos3_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos3_clk_bulk_data),
 	.max_freq = 1000,
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
@@ -497,7 +508,8 @@ static const struct samsung_dsim_driver_data exynos5433_dsi_driver_data = {
 	.plltmr_reg = 0xa0,
 	.has_legacy_status_reg = 1,
 	.has_clklane_stop = 1,
-	.num_clks = 5,
+	.clk_data = exynos5433_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos5433_clk_bulk_data),
 	.max_freq = 1500,
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
@@ -526,7 +538,8 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.plltmr_reg = 0xa0,
 	.has_legacy_status_reg = 1,
 	.has_clklane_stop = 1,
-	.num_clks = 2,
+	.clk_data = exynos3_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos3_clk_bulk_data),
 	.max_freq = 1500,
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 1,
@@ -555,7 +568,8 @@ static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.plltmr_reg = 0xa0,
 	.has_legacy_status_reg = 1,
 	.has_clklane_stop = 1,
-	.num_clks = 2,
+	.clk_data = exynos4_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos4_clk_bulk_data),
 	.max_freq = 2100,
 	.wait_for_hdr_fifo = 1,
 	.wait_for_reset = 0,
@@ -2021,7 +2035,7 @@ int samsung_dsim_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct samsung_dsim *dsi;
-	int ret, i;
+	int ret;
 
 	dsi = devm_drm_bridge_alloc(dev, struct samsung_dsim, bridge, &samsung_dsim_bridge_funcs);
 	if (IS_ERR(dsi))
@@ -2045,23 +2059,11 @@ int samsung_dsim_probe(struct platform_device *pdev)
 	if (ret)
 		return dev_err_probe(dev, ret, "failed to get regulators\n");
 
-	dsi->clks = devm_kcalloc(dev, dsi->driver_data->num_clks,
-				 sizeof(*dsi->clks), GFP_KERNEL);
-	if (!dsi->clks)
-		return -ENOMEM;
-
-	for (i = 0; i < dsi->driver_data->num_clks; i++) {
-		dsi->clks[i] = devm_clk_get(dev, clk_names[i]);
-		if (IS_ERR(dsi->clks[i])) {
-			if (strcmp(clk_names[i], "sclk_mipi") == 0) {
-				dsi->clks[i] = devm_clk_get(dev, OLD_SCLK_MIPI_CLK_NAME);
-				if (!IS_ERR(dsi->clks[i]))
-					continue;
-			}
-
-			dev_info(dev, "failed to get the clock: %s\n", clk_names[i]);
-			return PTR_ERR(dsi->clks[i]);
-		}
+	ret = devm_clk_bulk_get(dev, dsi->driver_data->num_clks,
+				dsi->driver_data->clk_data);
+	if (ret) {
+		dev_err(dev, "failed to get clocks in bulk (%d)\n", ret);
+		return ret;
 	}
 
 	dsi->reg_base = devm_platform_ioremap_resource(pdev, 0);
@@ -2134,7 +2136,7 @@ static int samsung_dsim_suspend(struct device *dev)
 {
 	struct samsung_dsim *dsi = dev_get_drvdata(dev);
 	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
-	int ret, i;
+	int ret;
 
 	usleep_range(10000, 20000);
 
@@ -2150,8 +2152,7 @@ static int samsung_dsim_suspend(struct device *dev)
 
 	phy_power_off(dsi->phy);
 
-	for (i = driver_data->num_clks - 1; i > -1; i--)
-		clk_disable_unprepare(dsi->clks[i]);
+	clk_bulk_disable_unprepare(driver_data->num_clks, driver_data->clk_data);
 
 	ret = regulator_bulk_disable(ARRAY_SIZE(dsi->supplies), dsi->supplies);
 	if (ret < 0)
@@ -2164,7 +2165,7 @@ static int samsung_dsim_resume(struct device *dev)
 {
 	struct samsung_dsim *dsi = dev_get_drvdata(dev);
 	const struct samsung_dsim_driver_data *driver_data = dsi->driver_data;
-	int ret, i;
+	int ret;
 
 	ret = regulator_bulk_enable(ARRAY_SIZE(dsi->supplies), dsi->supplies);
 	if (ret < 0) {
@@ -2172,11 +2173,9 @@ static int samsung_dsim_resume(struct device *dev)
 		return ret;
 	}
 
-	for (i = 0; i < driver_data->num_clks; i++) {
-		ret = clk_prepare_enable(dsi->clks[i]);
-		if (ret < 0)
-			goto err_clk;
-	}
+	ret = clk_bulk_prepare_enable(driver_data->num_clks, driver_data->clk_data);
+	if (ret < 0)
+		goto err_clk;
 
 	ret = phy_power_on(dsi->phy);
 	if (ret < 0) {
@@ -2187,8 +2186,7 @@ static int samsung_dsim_resume(struct device *dev)
 	return 0;
 
 err_clk:
-	while (--i > -1)
-		clk_disable_unprepare(dsi->clks[i]);
+	clk_bulk_disable_unprepare(driver_data->num_clks, driver_data->clk_data);
 	regulator_bulk_disable(ARRAY_SIZE(dsi->supplies), dsi->supplies);
 
 	return ret;
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index 04ed11787bbd..eb9fdbab1b34 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -58,6 +58,7 @@ struct samsung_dsim_driver_data {
 	unsigned int has_clklane_stop:1;
 	unsigned int has_broken_fifoctrl_emptyhdr:1;
 	unsigned int has_sfrctrl:1;
+	struct clk_bulk_data *clk_data;
 	unsigned int num_clks;
 	unsigned int min_freq;
 	unsigned int max_freq;
@@ -104,7 +105,6 @@ struct samsung_dsim {
 
 	void __iomem *reg_base;
 	struct phy *phy;
-	struct clk **clks;
 	struct clk *pll_clk;
 	struct regulator_bulk_data supplies[2];
 	int irq;

From 7f8d35b589c9f6acda0e36b184c193d38f092c36 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:45 +0530
Subject: [PATCH 1251/1455] dt-bindings: samsung,mipi-dsim: document exynos7870
 DSIM compatible

Add compatible string for Exynos7870 DSIM bridge controller. The
device requires four clock sources, in schema they're named as "bus",
"pll", "byte", and "esc".

Suggested-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 .../display/bridge/samsung,mipi-dsim.yaml     | 27 +++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
index 1acad99f3965..ad279f0993fa 100644
--- a/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
+++ b/Documentation/devicetree/bindings/display/bridge/samsung,mipi-dsim.yaml
@@ -24,6 +24,7 @@ properties:
           - samsung,exynos5410-mipi-dsi
           - samsung,exynos5422-mipi-dsi
           - samsung,exynos5433-mipi-dsi
+          - samsung,exynos7870-mipi-dsi
           - fsl,imx8mm-mipi-dsim
           - fsl,imx8mp-mipi-dsim
       - items:
@@ -144,6 +145,32 @@ required:
 
 allOf:
   - $ref: ../dsi-controller.yaml#
+  - if:
+      properties:
+        compatible:
+          contains:
+            const: samsung,exynos7870-mipi-dsi
+
+    then:
+      properties:
+        clocks:
+          minItems: 4
+          maxItems: 4
+
+        clock-names:
+          items:
+            - const: bus
+            - const: pll
+            - const: byte
+            - const: esc
+
+        ports:
+          required:
+            - port@0
+
+      required:
+        - ports
+
   - if:
       properties:
         compatible:

From 77169a11d4e9916f6c22587df396d6128505dbfb Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:46 +0530
Subject: [PATCH 1252/1455] drm/bridge: samsung-dsim: add driver support for
 exynos7870 DSIM bridge

Add support for Exynos7870's DSIM IP block in the bridge driver.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/bridge/samsung-dsim.c | 82 +++++++++++++++++++++++++++
 include/drm/bridge/samsung-dsim.h     |  1 +
 2 files changed, 83 insertions(+)

diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c b/drivers/gpu/drm/bridge/samsung-dsim.c
index 802164e6adff..eabc4c32f6ab 100644
--- a/drivers/gpu/drm/bridge/samsung-dsim.c
+++ b/drivers/gpu/drm/bridge/samsung-dsim.c
@@ -244,6 +244,13 @@ static struct clk_bulk_data exynos5433_clk_bulk_data[] = {
 	{ .id = "sclk_rgb_vclk_to_dsim0" },
 };
 
+static struct clk_bulk_data exynos7870_clk_bulk_data[] = {
+	{ .id = "bus" },
+	{ .id = "pll" },
+	{ .id = "byte" },
+	{ .id = "esc" },
+};
+
 enum reg_idx {
 	DSIM_STATUS_REG,	/* Status register (legacy) */
 	DSIM_LINK_STATUS_REG,	/* Link status register */
@@ -320,6 +327,32 @@ static const unsigned int exynos5433_reg_ofs[] = {
 	[DSIM_PHYTIMING2_REG] = 0xBC,
 };
 
+static const unsigned int exynos7870_reg_ofs[] = {
+	[DSIM_LINK_STATUS_REG] = 0x04,
+	[DSIM_DPHY_STATUS_REG] = 0x08,
+	[DSIM_SWRST_REG] = 0x0C,
+	[DSIM_CLKCTRL_REG] = 0x10,
+	[DSIM_TIMEOUT_REG] = 0x14,
+	[DSIM_ESCMODE_REG] = 0x1C,
+	[DSIM_MDRESOL_REG] = 0x20,
+	[DSIM_MVPORCH_REG] = 0x24,
+	[DSIM_MHPORCH_REG] = 0x28,
+	[DSIM_MSYNC_REG] = 0x2C,
+	[DSIM_CONFIG_REG] = 0x30,
+	[DSIM_INTSRC_REG] = 0x34,
+	[DSIM_INTMSK_REG] = 0x38,
+	[DSIM_PKTHDR_REG] = 0x3C,
+	[DSIM_PAYLOAD_REG] = 0x40,
+	[DSIM_RXFIFO_REG] = 0x44,
+	[DSIM_SFRCTRL_REG] = 0x48,
+	[DSIM_FIFOCTRL_REG] = 0x4C,
+	[DSIM_PLLCTRL_REG] = 0x94,
+	[DSIM_PHYCTRL_REG] = 0xA4,
+	[DSIM_PHYTIMING_REG] = 0xB4,
+	[DSIM_PHYTIMING1_REG] = 0xB8,
+	[DSIM_PHYTIMING2_REG] = 0xBC,
+};
+
 enum reg_value_idx {
 	RESET_TYPE,
 	PLL_TIMER,
@@ -392,6 +425,24 @@ static const unsigned int exynos5433_reg_values[] = {
 	[PHYTIMING_HS_TRAIL] = DSIM_PHYTIMING2_HS_TRAIL(0x0c),
 };
 
+static const unsigned int exynos7870_reg_values[] = {
+	[RESET_TYPE] = DSIM_SWRST,
+	[PLL_TIMER] = 80000,
+	[STOP_STATE_CNT] = 0xa,
+	[PHYCTRL_ULPS_EXIT] = DSIM_PHYCTRL_ULPS_EXIT(0x177),
+	[PHYCTRL_VREG_LP] = 0,
+	[PHYCTRL_SLEW_UP] = 0,
+	[PHYTIMING_LPX] = DSIM_PHYTIMING_LPX(0x07),
+	[PHYTIMING_HS_EXIT] = DSIM_PHYTIMING_HS_EXIT(0x0c),
+	[PHYTIMING_CLK_PREPARE] = DSIM_PHYTIMING1_CLK_PREPARE(0x08),
+	[PHYTIMING_CLK_ZERO] = DSIM_PHYTIMING1_CLK_ZERO(0x2b),
+	[PHYTIMING_CLK_POST] = DSIM_PHYTIMING1_CLK_POST(0x0d),
+	[PHYTIMING_CLK_TRAIL] = DSIM_PHYTIMING1_CLK_TRAIL(0x09),
+	[PHYTIMING_HS_PREPARE] = DSIM_PHYTIMING2_HS_PREPARE(0x09),
+	[PHYTIMING_HS_ZERO] = DSIM_PHYTIMING2_HS_ZERO(0x0f),
+	[PHYTIMING_HS_TRAIL] = DSIM_PHYTIMING2_HS_TRAIL(0x0c),
+};
+
 static const unsigned int imx8mm_dsim_reg_values[] = {
 	[RESET_TYPE] = DSIM_SWRST,
 	[PLL_TIMER] = 500,
@@ -563,6 +614,36 @@ static const struct samsung_dsim_driver_data exynos5422_dsi_driver_data = {
 	.min_freq = 500,
 };
 
+static const struct samsung_dsim_driver_data exynos7870_dsi_driver_data = {
+	.reg_ofs = exynos7870_reg_ofs,
+	.plltmr_reg = 0xa0,
+	.has_clklane_stop = 1,
+	.has_sfrctrl = 1,
+	.clk_data = exynos7870_clk_bulk_data,
+	.num_clks = ARRAY_SIZE(exynos7870_clk_bulk_data),
+	.max_freq = 1500,
+	.wait_for_hdr_fifo = 0,
+	.wait_for_reset = 1,
+	.num_bits_resol = 12,
+	.video_mode_bit = 18,
+	.pll_stable_bit = 24,
+	.esc_clken_bit = 16,
+	.byte_clken_bit = 17,
+	.tx_req_hsclk_bit = 20,
+	.lane_esc_clk_bit = 8,
+	.lane_esc_data_offset = 9,
+	.pll_p_offset = 13,
+	.pll_m_offset = 3,
+	.pll_s_offset = 0,
+	.main_vsa_offset = 16,
+	.reg_values = exynos7870_reg_values,
+	.pll_fin_min = 6,
+	.pll_fin_max = 12,
+	.m_min = 41,
+	.m_max = 125,
+	.min_freq = 500,
+};
+
 static const struct samsung_dsim_driver_data imx8mm_dsi_driver_data = {
 	.reg_ofs = exynos5433_reg_ofs,
 	.plltmr_reg = 0xa0,
@@ -604,6 +685,7 @@ samsung_dsim_types[DSIM_TYPE_COUNT] = {
 	[DSIM_TYPE_EXYNOS5410] = &exynos5_dsi_driver_data,
 	[DSIM_TYPE_EXYNOS5422] = &exynos5422_dsi_driver_data,
 	[DSIM_TYPE_EXYNOS5433] = &exynos5433_dsi_driver_data,
+	[DSIM_TYPE_EXYNOS7870] = &exynos7870_dsi_driver_data,
 	[DSIM_TYPE_IMX8MM] = &imx8mm_dsi_driver_data,
 	[DSIM_TYPE_IMX8MP] = &imx8mm_dsi_driver_data,
 };
diff --git a/include/drm/bridge/samsung-dsim.h b/include/drm/bridge/samsung-dsim.h
index eb9fdbab1b34..31d7ed589233 100644
--- a/include/drm/bridge/samsung-dsim.h
+++ b/include/drm/bridge/samsung-dsim.h
@@ -29,6 +29,7 @@ enum samsung_dsim_type {
 	DSIM_TYPE_EXYNOS5410,
 	DSIM_TYPE_EXYNOS5422,
 	DSIM_TYPE_EXYNOS5433,
+	DSIM_TYPE_EXYNOS7870,
 	DSIM_TYPE_IMX8MM,
 	DSIM_TYPE_IMX8MP,
 	DSIM_TYPE_COUNT,

From 0e6ee83c4b68448063e88e14c13c17d83e7d9735 Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 22:59:45 +0530
Subject: [PATCH 1253/1455] dt-bindings: display: samsung,exynos7-decon:
 document iommus, memory-region, and ports

Similar to FIMD and Exynos5433's DECON, the Exynos7 DECON hardware:
- May optionally require an IOMMU to initialize a display region.
- Outputs image data to another block, say an MIC or a DSI master.

If an IOMMU is present, it may also require to access the reserved
framebuffer region.

Document these bindings in the devicetree schema.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 .../samsung/samsung,exynos7-decon.yaml        | 21 +++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/Documentation/devicetree/bindings/display/samsung/samsung,exynos7-decon.yaml b/Documentation/devicetree/bindings/display/samsung/samsung,exynos7-decon.yaml
index 53916e4c95d8..14b954718008 100644
--- a/Documentation/devicetree/bindings/display/samsung/samsung,exynos7-decon.yaml
+++ b/Documentation/devicetree/bindings/display/samsung/samsung,exynos7-decon.yaml
@@ -80,6 +80,21 @@ properties:
       - const: vsync
       - const: lcd_sys
 
+  iommus:
+    maxItems: 1
+
+  memory-region:
+    maxItems: 1
+    description:
+      A phandle to a node describing a reserved framebuffer memory region.
+      For example, the splash memory region set up by the bootloader.
+
+  port:
+    $ref: /schemas/graph.yaml#/properties/port
+    description:
+      Output port which is connected to either a Mobile Image Compressor
+      (MIC) or a DSI Master device.
+
   power-domains:
     maxItems: 1
 
@@ -92,6 +107,7 @@ required:
   - clock-names
   - interrupts
   - interrupt-names
+  - port
   - reg
 
 additionalProperties: false
@@ -118,4 +134,9 @@ examples:
                       "decon0_vclk";
         pinctrl-0 = <&lcd_clk &pwm1_out>;
         pinctrl-names = "default";
+        port {
+            decon_to_dsi: endpoint {
+                remote-endpoint = <&dsi_to_decon>;
+            };
+        };
     };

From e1361a4f1be9cb69a662c6d7b5ce218007d6e82b Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 22:59:46 +0530
Subject: [PATCH 1254/1455] drm/exynos: exynos7_drm_decon: remove
 ctx->suspended

Condition guards are found to be redundant, as the call flow is properly
managed now, as also observed in the Exynos5433 DECON driver. Since
state checking is no longer necessary, remove it.

This also fixes an issue which prevented decon_commit() from
decon_atomic_enable() due to an incorrect state change setting.

Fixes: 96976c3d9aff ("drm/exynos: Add DECON driver")
Cc: stable@vger.kernel.org
Suggested-by: Inki Dae <inki.dae@samsung.com>
Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/exynos/exynos7_drm_decon.c | 36 ----------------------
 1 file changed, 36 deletions(-)

diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index 805aa28c1723..b8d9b7251319 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -69,7 +69,6 @@ struct decon_context {
 	void __iomem			*regs;
 	unsigned long			irq_flags;
 	bool				i80_if;
-	bool				suspended;
 	wait_queue_head_t		wait_vsync_queue;
 	atomic_t			wait_vsync_event;
 
@@ -132,9 +131,6 @@ static void decon_shadow_protect_win(struct decon_context *ctx,
 
 static void decon_wait_for_vblank(struct decon_context *ctx)
 {
-	if (ctx->suspended)
-		return;
-
 	atomic_set(&ctx->wait_vsync_event, 1);
 
 	/*
@@ -210,9 +206,6 @@ static void decon_commit(struct exynos_drm_crtc *crtc)
 	struct drm_display_mode *mode = &crtc->base.state->adjusted_mode;
 	u32 val, clkdiv;
 
-	if (ctx->suspended)
-		return;
-
 	/* nothing to do if we haven't set the mode yet */
 	if (mode->htotal == 0 || mode->vtotal == 0)
 		return;
@@ -274,9 +267,6 @@ static int decon_enable_vblank(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	u32 val;
 
-	if (ctx->suspended)
-		return -EPERM;
-
 	if (!test_and_set_bit(0, &ctx->irq_flags)) {
 		val = readl(ctx->regs + VIDINTCON0);
 
@@ -299,9 +289,6 @@ static void decon_disable_vblank(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	u32 val;
 
-	if (ctx->suspended)
-		return;
-
 	if (test_and_clear_bit(0, &ctx->irq_flags)) {
 		val = readl(ctx->regs + VIDINTCON0);
 
@@ -404,9 +391,6 @@ static void decon_atomic_begin(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	int i;
 
-	if (ctx->suspended)
-		return;
-
 	for (i = 0; i < WINDOWS_NR; i++)
 		decon_shadow_protect_win(ctx, i, true);
 }
@@ -427,9 +411,6 @@ static void decon_update_plane(struct exynos_drm_crtc *crtc,
 	unsigned int pitch = fb->pitches[0];
 	unsigned int vidw_addr0_base = ctx->data->vidw_buf_start_base;
 
-	if (ctx->suspended)
-		return;
-
 	/*
 	 * SHADOWCON/PRTCON register is used for enabling timing.
 	 *
@@ -517,9 +498,6 @@ static void decon_disable_plane(struct exynos_drm_crtc *crtc,
 	unsigned int win = plane->index;
 	u32 val;
 
-	if (ctx->suspended)
-		return;
-
 	/* protect windows */
 	decon_shadow_protect_win(ctx, win, true);
 
@@ -538,9 +516,6 @@ static void decon_atomic_flush(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	int i;
 
-	if (ctx->suspended)
-		return;
-
 	for (i = 0; i < WINDOWS_NR; i++)
 		decon_shadow_protect_win(ctx, i, false);
 	exynos_crtc_handle_event(crtc);
@@ -568,9 +543,6 @@ static void decon_atomic_enable(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	int ret;
 
-	if (!ctx->suspended)
-		return;
-
 	ret = pm_runtime_resume_and_get(ctx->dev);
 	if (ret < 0) {
 		DRM_DEV_ERROR(ctx->dev, "failed to enable DECON device.\n");
@@ -584,8 +556,6 @@ static void decon_atomic_enable(struct exynos_drm_crtc *crtc)
 		decon_enable_vblank(ctx->crtc);
 
 	decon_commit(ctx->crtc);
-
-	ctx->suspended = false;
 }
 
 static void decon_atomic_disable(struct exynos_drm_crtc *crtc)
@@ -593,9 +563,6 @@ static void decon_atomic_disable(struct exynos_drm_crtc *crtc)
 	struct decon_context *ctx = crtc->ctx;
 	int i;
 
-	if (ctx->suspended)
-		return;
-
 	/*
 	 * We need to make sure that all windows are disabled before we
 	 * suspend that connector. Otherwise we might try to scan from
@@ -605,8 +572,6 @@ static void decon_atomic_disable(struct exynos_drm_crtc *crtc)
 		decon_disable_plane(crtc, &ctx->planes[i]);
 
 	pm_runtime_put_sync(ctx->dev);
-
-	ctx->suspended = true;
 }
 
 static const struct exynos_drm_crtc_ops decon_crtc_ops = {
@@ -727,7 +692,6 @@ static int decon_probe(struct platform_device *pdev)
 		return -ENOMEM;
 
 	ctx->dev = dev;
-	ctx->suspended = true;
 	ctx->data = of_device_get_match_data(dev);
 
 	i80_if_timings = of_get_child_by_name(dev->of_node, "i80-if-timings");

From 472df181d7c4c732d5c5b98d506d1902e7580acb Mon Sep 17 00:00:00 2001
From: Kaustabh Chakraborty <kauschluss@disroot.org>
Date: Sun, 6 Jul 2025 23:55:47 +0530
Subject: [PATCH 1255/1455] drm/exynos: dsi: add support for exynos7870

Add glue layer support for Exynos7870's DSIM IP bridge driver.

Signed-off-by: Kaustabh Chakraborty <kauschluss@disroot.org>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
---
 drivers/gpu/drm/exynos/exynos_drm_dsi.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 896a03639e2d..c4d098ab7863 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -154,6 +154,11 @@ static const struct samsung_dsim_plat_data exynos5433_dsi_pdata = {
 	.host_ops = &exynos_dsi_exynos_host_ops,
 };
 
+static const struct samsung_dsim_plat_data exynos7870_dsi_pdata = {
+	.hw_type = DSIM_TYPE_EXYNOS7870,
+	.host_ops = &exynos_dsi_exynos_host_ops,
+};
+
 static const struct of_device_id exynos_dsi_of_match[] = {
 	{
 		.compatible = "samsung,exynos3250-mipi-dsi",
@@ -175,6 +180,10 @@ static const struct of_device_id exynos_dsi_of_match[] = {
 		.compatible = "samsung,exynos5433-mipi-dsi",
 		.data = &exynos5433_dsi_pdata,
 	},
+	{
+		.compatible = "samsung,exynos7870-mipi-dsi",
+		.data = &exynos7870_dsi_pdata,
+	},
 	{ /* sentinel. */ }
 };
 MODULE_DEVICE_TABLE(of, exynos_dsi_of_match);

From 75cc23ffe5b422bc3cbd5cf0956b8b86e4b0e162 Mon Sep 17 00:00:00 2001
From: Dan Carpenter <dan.carpenter@linaro.org>
Date: Thu, 7 Aug 2025 18:53:41 +0300
Subject: [PATCH 1256/1455] drm/xe: Fix a NULL vs IS_ERR() in
 xe_vm_add_compute_exec_queue()

The xe_preempt_fence_create() function returns error pointers.  It
never returns NULL.  Update the error checking to match.

Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs")
Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Link: https://lore.kernel.org/r/aJTMBdX97cof_009@stanley.mountain
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/xe/xe_vm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 7b04b392b729..0cacab20ff85 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -224,8 +224,8 @@ int xe_vm_add_compute_exec_queue(struct xe_vm *vm, struct xe_exec_queue *q)
 
 	pfence = xe_preempt_fence_create(q, q->lr.context,
 					 ++q->lr.seqno);
-	if (!pfence) {
-		err = -ENOMEM;
+	if (IS_ERR(pfence)) {
+		err = PTR_ERR(pfence);
 		goto out_fini;
 	}
 

From 19baa830fbb60d712137dfbdf97a538f7057d100 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Fri, 12 Sep 2025 15:05:34 -0700
Subject: [PATCH 1257/1455] drm/xe: Use ARRAY_SIZE in guc_waklv_init()

Prefer using ARRAY_SIZE where needed and just passing 1 instead of
calculating the size of one element.

Cc: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202508130158.eogeBZQT-lkp@intel.com/
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20250912-guc-ads-array-size-v1-1-a6555392a1f8@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_guc_ads.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c
index 5631722f34f5..58e0b0294a5b 100644
--- a/drivers/gpu/drm/xe/xe_guc_ads.c
+++ b/drivers/gpu/drm/xe/xe_guc_ads.c
@@ -339,7 +339,7 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 	if (XE_GT_WA(gt, 13011645652)) {
 		u32 data = 0xC40;
 
-		guc_waklv_enable(ads, &data, sizeof(data) / sizeof(u32), &offset, &remain,
+		guc_waklv_enable(ads, &data, 1, &offset, &remain,
 				 GUC_WA_KLV_NP_RD_WRITE_TO_CLEAR_RCSM_AT_CGP_LATE_RESTORE);
 	}
 
@@ -355,7 +355,7 @@ static void guc_waklv_init(struct xe_guc_ads *ads)
 			0x0,
 			0xF,
 		};
-		guc_waklv_enable(ads, data, sizeof(data) / sizeof(u32), &offset, &remain,
+		guc_waklv_enable(ads, data, ARRAY_SIZE(data), &offset, &remain,
 				 GUC_WA_KLV_RESTORE_UNSAVED_MEDIA_CONTROL_REG);
 	}
 

From 84afb84bcc28b08e6d8a4e36865959d770e2643d Mon Sep 17 00:00:00 2001
From: Fushuai Wang <wangfushuai@baidu.com>
Date: Sun, 14 Sep 2025 18:16:30 +0800
Subject: [PATCH 1258/1455] drm/xe: Use ERR_CAST instead of
 ERR_PTR(PTR_ERR(...))

Use ERR_CAST inline function instead of ERR_PTR(PTR_ERR(...)).

Signed-off-by: Fushuai Wang <wangfushuai@baidu.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250914101630.17719-1-wangfushuai@baidu.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_svm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 8a0fa90a10c0..86e7c72c89f1 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -1278,7 +1278,7 @@ struct xe_svm_range *xe_svm_range_find_or_insert(struct xe_vm *vm, u64 addr,
 	r = drm_gpusvm_range_find_or_insert(&vm->svm.gpusvm, max(addr, xe_vma_start(vma)),
 					    xe_vma_start(vma), xe_vma_end(vma), ctx);
 	if (IS_ERR(r))
-		return ERR_PTR(PTR_ERR(r));
+		return ERR_CAST(r);
 
 	return to_xe_range(r);
 }

From 7d0ca56e9176578babd527736840e7c32c3ee5fc Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Wed, 10 Sep 2025 14:02:32 -0700
Subject: [PATCH 1259/1455] drm/xe/guc: Update CSS header structures

Rework the CSS header structure according to recent updates to the GuC
API spec. Also include more field definitions.

v2: Also pass the new GuC specific structure to a GuC specific
function instead of the higher level, generic structure (review
feedback from Daniele).
Also correct naming of CSS_TIME_* fields.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://lore.kernel.org/r/20250910210237.603576-2-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/xe_uc_fw.c     | 24 ++++++------
 drivers/gpu/drm/xe/xe_uc_fw_abi.h | 64 ++++++++++++++++++++-----------
 2 files changed, 53 insertions(+), 35 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_uc_fw.c b/drivers/gpu/drm/xe/xe_uc_fw.c
index 4faf8b0fbcfe..57b8230cec58 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw.c
+++ b/drivers/gpu/drm/xe/xe_uc_fw.c
@@ -328,7 +328,7 @@ static void uc_fw_fini(struct drm_device *drm, void *arg)
 	xe_uc_fw_change_status(uc_fw, XE_UC_FIRMWARE_SELECTED);
 }
 
-static int guc_read_css_info(struct xe_uc_fw *uc_fw, struct uc_css_header *css)
+static int guc_read_css_info(struct xe_uc_fw *uc_fw, struct uc_css_guc_info *guc_info)
 {
 	struct xe_gt *gt = uc_fw_to_gt(uc_fw);
 	struct xe_uc_fw_version *release = &uc_fw->versions.found[XE_UC_FW_VER_RELEASE];
@@ -343,11 +343,11 @@ static int guc_read_css_info(struct xe_uc_fw *uc_fw, struct uc_css_header *css)
 		return -EINVAL;
 	}
 
-	compatibility->major = FIELD_GET(CSS_SW_VERSION_UC_MAJOR, css->submission_version);
-	compatibility->minor = FIELD_GET(CSS_SW_VERSION_UC_MINOR, css->submission_version);
-	compatibility->patch = FIELD_GET(CSS_SW_VERSION_UC_PATCH, css->submission_version);
+	compatibility->major = FIELD_GET(CSS_SW_VERSION_UC_MAJOR, guc_info->submission_version);
+	compatibility->minor = FIELD_GET(CSS_SW_VERSION_UC_MINOR, guc_info->submission_version);
+	compatibility->patch = FIELD_GET(CSS_SW_VERSION_UC_PATCH, guc_info->submission_version);
 
-	uc_fw->private_data_size = css->private_data_size;
+	uc_fw->private_data_size = guc_info->private_data_size;
 
 	return 0;
 }
@@ -416,8 +416,8 @@ static int parse_css_header(struct xe_uc_fw *uc_fw, const void *fw_data, size_t
 	css = (struct uc_css_header *)fw_data;
 
 	/* Check integrity of size values inside CSS header */
-	size = (css->header_size_dw - css->key_size_dw - css->modulus_size_dw -
-		css->exponent_size_dw) * sizeof(u32);
+	size = (css->header_size_dw - css->rsa_info.key_size_dw - css->rsa_info.modulus_size_dw -
+		css->rsa_info.exponent_size_dw) * sizeof(u32);
 	if (unlikely(size != sizeof(struct uc_css_header))) {
 		drm_warn(&xe->drm,
 			 "%s firmware %s: unexpected header size: %zu != %zu\n",
@@ -430,7 +430,7 @@ static int parse_css_header(struct xe_uc_fw *uc_fw, const void *fw_data, size_t
 	uc_fw->ucode_size = (css->size_dw - css->header_size_dw) * sizeof(u32);
 
 	/* now RSA */
-	uc_fw->rsa_size = css->key_size_dw * sizeof(u32);
+	uc_fw->rsa_size = css->rsa_info.key_size_dw * sizeof(u32);
 
 	/* At least, it should have header, uCode and RSA. Size of all three. */
 	size = sizeof(struct uc_css_header) + uc_fw->ucode_size +
@@ -443,12 +443,12 @@ static int parse_css_header(struct xe_uc_fw *uc_fw, const void *fw_data, size_t
 	}
 
 	/* Get version numbers from the CSS header */
-	release->major = FIELD_GET(CSS_SW_VERSION_UC_MAJOR, css->sw_version);
-	release->minor = FIELD_GET(CSS_SW_VERSION_UC_MINOR, css->sw_version);
-	release->patch = FIELD_GET(CSS_SW_VERSION_UC_PATCH, css->sw_version);
+	release->major = FIELD_GET(CSS_SW_VERSION_UC_MAJOR, css->guc_info.sw_version);
+	release->minor = FIELD_GET(CSS_SW_VERSION_UC_MINOR, css->guc_info.sw_version);
+	release->patch = FIELD_GET(CSS_SW_VERSION_UC_PATCH, css->guc_info.sw_version);
 
 	if (uc_fw->type == XE_UC_FW_TYPE_GUC)
-		return guc_read_css_info(uc_fw, css);
+		return guc_read_css_info(uc_fw, &css->guc_info);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/xe/xe_uc_fw_abi.h b/drivers/gpu/drm/xe/xe_uc_fw_abi.h
index 87ade41209d0..faceb437fd2f 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw_abi.h
+++ b/drivers/gpu/drm/xe/xe_uc_fw_abi.h
@@ -44,6 +44,39 @@
  *    in fw. So driver will load a truncated firmware in this case.
  */
 
+struct uc_css_rsa_info {
+	u32 key_size_dw;
+	u32 modulus_size_dw;
+	u32 exponent_size_dw;
+} __packed;
+
+struct uc_css_guc_info {
+	u32 time;
+#define CSS_TIME_HOUR				(0xFF << 0)
+#define CSS_TIME_MIN				(0xFF << 8)
+#define CSS_TIME_SEC				(0xFFFF << 16)
+	u32 reserved0[5];
+	u32 sw_version;
+#define CSS_SW_VERSION_UC_MAJOR			(0xFF << 16)
+#define CSS_SW_VERSION_UC_MINOR			(0xFF << 8)
+#define CSS_SW_VERSION_UC_PATCH			(0xFF << 0)
+	u32 submission_version;
+	u32 reserved1[11];
+	u32 header_info;
+#define CSS_HEADER_INFO_SVN			(0xFF)
+#define CSS_HEADER_INFO_COPY_VALID		(0x1 << 31)
+	u32 private_data_size;
+	u32 ukernel_info;
+#define CSS_UKERNEL_INFO_DEVICEID		(0xFFFF << 16)
+#define CSS_UKERNEL_INFO_PRODKEY		(0xFF << 8)
+#define CSS_UKERNEL_INFO_BUILDTYPE		(0x3 << 2)
+#define CSS_UKERNEL_INFO_BUILDTYPE_PROD		0
+#define CSS_UKERNEL_INFO_BUILDTYPE_PREPROD	1
+#define CSS_UKERNEL_INFO_BUILDTYPE_DEBUG	2
+#define CSS_UKERNEL_INFO_ENCSTATUS		(0x1 << 1)
+#define CSS_UKERNEL_INFO_COPY_VALID		(0x1 << 0)
+} __packed;
+
 struct uc_css_header {
 	u32 module_type;
 	/*
@@ -52,36 +85,21 @@ struct uc_css_header {
 	 */
 	u32 header_size_dw;
 	u32 header_version;
-	u32 module_id;
+	u32 reserved0;
 	u32 module_vendor;
 	u32 date;
-#define CSS_DATE_DAY			(0xFF << 0)
-#define CSS_DATE_MONTH			(0xFF << 8)
-#define CSS_DATE_YEAR			(0xFFFF << 16)
+#define CSS_DATE_DAY				(0xFF << 0)
+#define CSS_DATE_MONTH				(0xFF << 8)
+#define CSS_DATE_YEAR				(0xFFFF << 16)
 	u32 size_dw; /* uCode plus header_size_dw */
-	u32 key_size_dw;
-	u32 modulus_size_dw;
-	u32 exponent_size_dw;
-	u32 time;
-#define CSS_TIME_HOUR			(0xFF << 0)
-#define CSS_DATE_MIN			(0xFF << 8)
-#define CSS_DATE_SEC			(0xFFFF << 16)
-	char username[8];
-	char buildnumber[12];
-	u32 sw_version;
-#define CSS_SW_VERSION_UC_MAJOR		(0xFF << 16)
-#define CSS_SW_VERSION_UC_MINOR		(0xFF << 8)
-#define CSS_SW_VERSION_UC_PATCH		(0xFF << 0)
 	union {
-		u32 submission_version; /* only applies to GuC */
-		u32 reserved2;
+		u32 reserved1[3];
+		struct uc_css_rsa_info rsa_info;
 	};
-	u32 reserved0[12];
 	union {
-		u32 private_data_size; /* only applies to GuC */
-		u32 reserved1;
+		u32 reserved2[22];
+		struct uc_css_guc_info guc_info;
 	};
-	u32 header_info;
 } __packed;
 static_assert(sizeof(struct uc_css_header) == 128);
 

From acf01c79f0684e3cbaa8e6326b970d7a7381c2ee Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Wed, 10 Sep 2025 14:02:33 -0700
Subject: [PATCH 1260/1455] drm/xe/guc: Add firmware build type to available
 info

Some test features are not available in production builds of the GuC
firmware. So add the build type field to the available information
that tests can inspect to decide if they should skip or run.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://lore.kernel.org/r/20250910210237.603576-3-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/xe_uc_fw.c       | 1 +
 drivers/gpu/drm/xe/xe_uc_fw_types.h | 3 +++
 2 files changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_uc_fw.c b/drivers/gpu/drm/xe/xe_uc_fw.c
index 57b8230cec58..622b76078567 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw.c
+++ b/drivers/gpu/drm/xe/xe_uc_fw.c
@@ -347,6 +347,7 @@ static int guc_read_css_info(struct xe_uc_fw *uc_fw, struct uc_css_guc_info *guc
 	compatibility->minor = FIELD_GET(CSS_SW_VERSION_UC_MINOR, guc_info->submission_version);
 	compatibility->patch = FIELD_GET(CSS_SW_VERSION_UC_PATCH, guc_info->submission_version);
 
+	uc_fw->build_type = FIELD_GET(CSS_UKERNEL_INFO_BUILDTYPE, guc_info->ukernel_info);
 	uc_fw->private_data_size = guc_info->private_data_size;
 
 	return 0;
diff --git a/drivers/gpu/drm/xe/xe_uc_fw_types.h b/drivers/gpu/drm/xe/xe_uc_fw_types.h
index 914026015019..77a1dcf8b4ed 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw_types.h
+++ b/drivers/gpu/drm/xe/xe_uc_fw_types.h
@@ -147,6 +147,9 @@ struct xe_uc_fw {
 
 	/** @private_data_size: size of private data found in uC css header */
 	u32 private_data_size;
+
+	/** @build_type: Firmware build type (see CSS_UKERNEL_INFO_BUILDTYPE for definitions) */
+	u32 build_type;
 };
 
 #endif

From 537773db91c2765898eee0521d5f03acae5a545f Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Wed, 10 Sep 2025 14:02:34 -0700
Subject: [PATCH 1261/1455] drm/xe: Allow freeing of a managed bo

If a bo is created via xe_managed_bo_create_pin_map() then it cannot be
freed by the driver using xe_bo_unpin_map_no_vm(), or indeed any other
existing function. The DRM layer will still have a pointer stashed
away for later freeing, causing a invalid memory access on driver
unload. So add a helper for releasing the DRM action as well.

v2: Drop 'xe' parameter (review feedbak from Michal W)

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://lore.kernel.org/r/20250910210237.603576-4-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/xe_bo.c | 5 +++++
 drivers/gpu/drm/xe/xe_bo.h | 1 +
 2 files changed, 6 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
index e6d16b34e5b5..8422f3cab113 100644
--- a/drivers/gpu/drm/xe/xe_bo.c
+++ b/drivers/gpu/drm/xe/xe_bo.c
@@ -2629,6 +2629,11 @@ struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile
 	return bo;
 }
 
+void xe_managed_bo_unpin_map_no_vm(struct xe_bo *bo)
+{
+	devm_release_action(xe_bo_device(bo)->drm.dev, __xe_bo_unpin_map_no_vm, bo);
+}
+
 struct xe_bo *xe_managed_bo_create_from_data(struct xe_device *xe, struct xe_tile *tile,
 					     const void *data, size_t size, u32 flags)
 {
diff --git a/drivers/gpu/drm/xe/xe_bo.h b/drivers/gpu/drm/xe/xe_bo.h
index 33d7c122dcc9..a77af42b5f9e 100644
--- a/drivers/gpu/drm/xe/xe_bo.h
+++ b/drivers/gpu/drm/xe/xe_bo.h
@@ -116,6 +116,7 @@ xe_bo_create_pin_map_at_novm(struct xe_device *xe, struct xe_tile *tile,
 			     u32 flags, u64 alignment, bool intr);
 struct xe_bo *xe_managed_bo_create_pin_map(struct xe_device *xe, struct xe_tile *tile,
 					   size_t size, u32 flags);
+void xe_managed_bo_unpin_map_no_vm(struct xe_bo *bo);
 struct xe_bo *xe_managed_bo_create_from_data(struct xe_device *xe, struct xe_tile *tile,
 					     const void *data, size_t size, u32 flags);
 int xe_managed_bo_reinit_in_vram(struct xe_device *xe, struct xe_tile *tile, struct xe_bo **src);

From 456b32c9c1bc6c99310163281edb53eefd51330c Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Wed, 10 Sep 2025 14:02:35 -0700
Subject: [PATCH 1262/1455] drm/xe/guc: Add test for G2G communications

Add a test for sending messages from every GuC to every other GuC to
test G2G communications.

Note that, being a debug only feature, the test interface only exists
in pre-production builds of the GuC firmware.

v2: Fix 'default' case to actually use the driver's registration code
as well as allocation. Add comments explaining the different test
types. Fix (C) date and an assert. Review feedback from Daniele.

Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Link: https://lore.kernel.org/r/20250910210237.603576-5-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/abi/guc_actions_abi.h    |   2 +
 drivers/gpu/drm/xe/tests/xe_guc_g2g_test.c  | 776 ++++++++++++++++++++
 drivers/gpu/drm/xe/tests/xe_live_test_mod.c |   2 +
 drivers/gpu/drm/xe/xe_device_types.h        |   7 +
 drivers/gpu/drm/xe/xe_guc.c                 |   4 +
 drivers/gpu/drm/xe/xe_guc.h                 |   4 +
 drivers/gpu/drm/xe/xe_guc_ct.c              |   5 +
 drivers/gpu/drm/xe/xe_guc_fwif.h            |   1 +
 8 files changed, 801 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/tests/xe_guc_g2g_test.c

diff --git a/drivers/gpu/drm/xe/abi/guc_actions_abi.h b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
index 1baa969aaa7c..31090c69dfbe 100644
--- a/drivers/gpu/drm/xe/abi/guc_actions_abi.h
+++ b/drivers/gpu/drm/xe/abi/guc_actions_abi.h
@@ -155,6 +155,8 @@ enum xe_guc_action {
 	XE_GUC_ACTION_NOTIFY_FLUSH_LOG_BUFFER_TO_FILE = 0x8003,
 	XE_GUC_ACTION_NOTIFY_CRASH_DUMP_POSTED = 0x8004,
 	XE_GUC_ACTION_NOTIFY_EXCEPTION = 0x8005,
+	XE_GUC_ACTION_TEST_G2G_SEND = 0xF001,
+	XE_GUC_ACTION_TEST_G2G_RECV = 0xF002,
 	XE_GUC_ACTION_LIMIT
 };
 
diff --git a/drivers/gpu/drm/xe/tests/xe_guc_g2g_test.c b/drivers/gpu/drm/xe/tests/xe_guc_g2g_test.c
new file mode 100644
index 000000000000..3b213fcae916
--- /dev/null
+++ b/drivers/gpu/drm/xe/tests/xe_guc_g2g_test.c
@@ -0,0 +1,776 @@
+// SPDX-License-Identifier: GPL-2.0 AND MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/delay.h>
+
+#include <kunit/test.h>
+#include <kunit/visibility.h>
+
+#include "tests/xe_kunit_helpers.h"
+#include "tests/xe_pci_test.h"
+#include "tests/xe_test.h"
+
+#include "xe_bo.h"
+#include "xe_device.h"
+#include "xe_pm.h"
+
+/*
+ * There are different ways to allocate the G2G buffers. The plan for this test
+ * is to make sure that all the possible options work. The particular option
+ * chosen by the driver may vary from one platform to another, it may also change
+ * with time. So to ensure consistency of testing, the relevant driver code is
+ * replicated here to guarantee it won't change without the test being updated
+ * to keep testing the other options.
+ *
+ * In order to test the actual code being used by the driver, there is also the
+ * 'default' scheme. That will use the official driver routines to test whatever
+ * method the driver is using on the current platform at the current time.
+ */
+enum {
+	/* Driver defined allocation scheme */
+	G2G_CTB_TYPE_DEFAULT,
+	/* Single buffer in host memory */
+	G2G_CTB_TYPE_HOST,
+	/* Single buffer in a specific tile, loops across all tiles */
+	G2G_CTB_TYPE_TILE,
+};
+
+/*
+ * Payload is opaque to GuC. So KMD can define any structure or size it wants.
+ */
+struct g2g_test_payload  {
+	u32 tx_dev;
+	u32 tx_tile;
+	u32 rx_dev;
+	u32 rx_tile;
+	u32 seqno;
+};
+
+static void g2g_test_send(struct kunit *test, struct xe_guc *guc,
+			  u32 far_tile, u32 far_dev,
+			  struct g2g_test_payload *payload)
+{
+	struct xe_device *xe = guc_to_xe(guc);
+	struct xe_gt *gt = guc_to_gt(guc);
+	u32 *action, total;
+	size_t payload_len;
+	int ret;
+
+	static_assert(IS_ALIGNED(sizeof(*payload), sizeof(u32)));
+	payload_len = sizeof(*payload) / sizeof(u32);
+
+	total = 4 + payload_len;
+	action = kunit_kmalloc_array(test, total, sizeof(*action), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, action);
+
+	action[0] = XE_GUC_ACTION_TEST_G2G_SEND;
+	action[1] = far_tile;
+	action[2] = far_dev;
+	action[3] = payload_len;
+	memcpy(action + 4, payload, payload_len * sizeof(u32));
+
+	atomic_inc(&xe->g2g_test_count);
+
+	/*
+	 * Should specify the expected response notification here. Problem is that
+	 * the response will be coming from a different GuC. By the end, it should
+	 * all add up as long as an equal number of messages are sent from each GuC
+	 * and to each GuC. However, in the middle negative reservation space errors
+	 * and such like can occur. Rather than add intrusive changes to the CT layer
+	 * it is simpler to just not bother counting it at all. The system should be
+	 * idle when running the selftest, and the selftest's notification total size
+	 * is well within the G2H allocation size. So there should be no issues with
+	 * needing to block for space, which is all the tracking code is really for.
+	 */
+	ret = xe_guc_ct_send(&guc->ct, action, total, 0, 0);
+	kunit_kfree(test, action);
+	KUNIT_ASSERT_EQ_MSG(test, 0, ret, "G2G send failed: %d [%d:%d -> %d:%d]\n", ret,
+			    gt_to_tile(gt)->id, G2G_DEV(gt), far_tile, far_dev);
+}
+
+/*
+ * NB: Can't use KUNIT_ASSERT and friends in here as this is called asynchronously
+ * from the G2H notification handler. Need that to actually complete rather than
+ * thread-abort in order to keep the rest of the driver alive!
+ */
+int xe_guc_g2g_test_notification(struct xe_guc *guc, u32 *msg, u32 len)
+{
+	struct xe_device *xe = guc_to_xe(guc);
+	struct xe_gt *rx_gt = guc_to_gt(guc), *test_gt, *tx_gt = NULL;
+	u32 tx_tile, tx_dev, rx_tile, rx_dev, idx, got_len;
+	struct g2g_test_payload *payload;
+	size_t payload_len;
+	int ret = 0, i;
+
+	payload_len = sizeof(*payload) / sizeof(u32);
+
+	if (unlikely(len != (G2H_LEN_DW_G2G_NOTIFY_MIN + payload_len))) {
+		xe_gt_err(rx_gt, "G2G test notification invalid length %u", len);
+		ret = -EPROTO;
+		goto done;
+	}
+
+	tx_tile = msg[0];
+	tx_dev = msg[1];
+	got_len = msg[2];
+	payload = (struct g2g_test_payload *)(msg + 3);
+
+	rx_tile = gt_to_tile(rx_gt)->id;
+	rx_dev = G2G_DEV(rx_gt);
+
+	if (got_len != payload_len) {
+		xe_gt_err(rx_gt, "G2G: Invalid payload length: %u vs %zu\n", got_len, payload_len);
+		ret = -EPROTO;
+		goto done;
+	}
+
+	if (payload->tx_dev != tx_dev || payload->tx_tile != tx_tile ||
+	    payload->rx_dev != rx_dev || payload->rx_tile != rx_tile) {
+		xe_gt_err(rx_gt, "G2G: Invalid payload: %d:%d -> %d:%d vs %d:%d -> %d:%d! [%d]\n",
+			  payload->tx_tile, payload->tx_dev, payload->rx_tile, payload->rx_dev,
+			  tx_tile, tx_dev, rx_tile, rx_dev, payload->seqno);
+		ret = -EPROTO;
+		goto done;
+	}
+
+	if (!xe->g2g_test_array) {
+		xe_gt_err(rx_gt, "G2G: Missing test array!\n");
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	for_each_gt(test_gt, xe, i) {
+		if (gt_to_tile(test_gt)->id != tx_tile)
+			continue;
+
+		if (G2G_DEV(test_gt) != tx_dev)
+			continue;
+
+		if (tx_gt) {
+			xe_gt_err(rx_gt, "G2G: Got duplicate TX GTs: %d vs %d for %d:%d!\n",
+				  tx_gt->info.id, test_gt->info.id, tx_tile, tx_dev);
+			ret = -EINVAL;
+			goto done;
+		}
+
+		tx_gt = test_gt;
+	}
+	if (!tx_gt) {
+		xe_gt_err(rx_gt, "G2G: Failed to find a TX GT for %d:%d!\n", tx_tile, tx_dev);
+		ret = -EINVAL;
+		goto done;
+	}
+
+	idx = (tx_gt->info.id * xe->info.gt_count) + rx_gt->info.id;
+
+	if (xe->g2g_test_array[idx] != payload->seqno - 1) {
+		xe_gt_err(rx_gt, "G2G: Seqno mismatch %d vs %d for %d:%d -> %d:%d!\n",
+			  xe->g2g_test_array[idx], payload->seqno - 1,
+			  tx_tile, tx_dev, rx_tile, rx_dev);
+		ret = -EINVAL;
+		goto done;
+	}
+
+	xe->g2g_test_array[idx] = payload->seqno;
+
+done:
+	atomic_dec(&xe->g2g_test_count);
+	return ret;
+}
+
+/*
+ * Send the given seqno from all GuCs to all other GuCs in tile/GT order
+ */
+static void g2g_test_in_order(struct kunit *test, struct xe_device *xe, u32 seqno)
+{
+	struct xe_gt *near_gt, *far_gt;
+	int i, j;
+
+	for_each_gt(near_gt, xe, i) {
+		u32 near_tile = gt_to_tile(near_gt)->id;
+		u32 near_dev = G2G_DEV(near_gt);
+
+		for_each_gt(far_gt, xe, j) {
+			u32 far_tile = gt_to_tile(far_gt)->id;
+			u32 far_dev = G2G_DEV(far_gt);
+			struct g2g_test_payload payload;
+
+			if (far_gt->info.id == near_gt->info.id)
+				continue;
+
+			payload.tx_dev = near_dev;
+			payload.tx_tile = near_tile;
+			payload.rx_dev = far_dev;
+			payload.rx_tile = far_tile;
+			payload.seqno = seqno;
+			g2g_test_send(test, &near_gt->uc.guc, far_tile, far_dev, &payload);
+		}
+	}
+}
+
+#define WAIT_TIME_MS	100
+#define WAIT_COUNT	(1000 / WAIT_TIME_MS)
+
+static void g2g_wait_for_complete(void *_xe)
+{
+	struct xe_device *xe = (struct xe_device *)_xe;
+	struct kunit *test = kunit_get_current_test();
+	int wait = 0;
+
+	/* Wait for all G2H messages to be received */
+	while (atomic_read(&xe->g2g_test_count)) {
+		if (++wait > WAIT_COUNT)
+			break;
+
+		msleep(WAIT_TIME_MS);
+	}
+
+	KUNIT_ASSERT_EQ_MSG(test, 0, atomic_read(&xe->g2g_test_count),
+			    "Timed out waiting for notifications\n");
+	kunit_info(test, "Got all notifications back\n");
+}
+
+#undef WAIT_TIME_MS
+#undef WAIT_COUNT
+
+static void g2g_clean_array(void *_xe)
+{
+	struct xe_device *xe = (struct xe_device *)_xe;
+
+	xe->g2g_test_array = NULL;
+}
+
+#define NUM_LOOPS	16
+
+static void g2g_run_test(struct kunit *test, struct xe_device *xe)
+{
+	u32 seqno, max_array;
+	int ret, i, j;
+
+	max_array = xe->info.gt_count * xe->info.gt_count;
+	xe->g2g_test_array = kunit_kcalloc(test, max_array, sizeof(u32), GFP_KERNEL);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, xe->g2g_test_array);
+
+	ret = kunit_add_action_or_reset(test, g2g_clean_array, xe);
+	KUNIT_ASSERT_EQ_MSG(test, 0, ret, "Failed to register clean up action\n");
+
+	/*
+	 * Send incrementing seqnos from all GuCs to all other GuCs in tile/GT order.
+	 * Tile/GT order doesn't really mean anything to the hardware but it is going
+	 * to be a fixed sequence every time.
+	 *
+	 * Verify that each one comes back having taken the correct route.
+	 */
+	ret = kunit_add_action(test, g2g_wait_for_complete, xe);
+	KUNIT_ASSERT_EQ_MSG(test, 0, ret, "Failed to register clean up action\n");
+	for (seqno = 1; seqno < NUM_LOOPS; seqno++)
+		g2g_test_in_order(test, xe, seqno);
+	seqno--;
+
+	kunit_release_action(test, &g2g_wait_for_complete, xe);
+
+	/* Check for the final seqno in each slot */
+	for (i = 0; i < xe->info.gt_count; i++) {
+		for (j = 0; j < xe->info.gt_count; j++) {
+			u32 idx = (j * xe->info.gt_count) + i;
+
+			if (i == j)
+				KUNIT_ASSERT_EQ_MSG(test, 0, xe->g2g_test_array[idx],
+						    "identity seqno modified: %d for %dx%d!\n",
+						    xe->g2g_test_array[idx], i, j);
+			else
+				KUNIT_ASSERT_EQ_MSG(test, seqno, xe->g2g_test_array[idx],
+						    "invalid seqno: %d vs %d for %dx%d!\n",
+						    xe->g2g_test_array[idx], seqno, i, j);
+		}
+	}
+
+	kunit_kfree(test, xe->g2g_test_array);
+	kunit_release_action(test, &g2g_clean_array, xe);
+
+	kunit_info(test, "Test passed\n");
+}
+
+#undef NUM_LOOPS
+
+static void g2g_ct_stop(struct xe_guc *guc)
+{
+	struct xe_gt *remote_gt, *gt = guc_to_gt(guc);
+	struct xe_device *xe = gt_to_xe(gt);
+	int i, t;
+
+	for_each_gt(remote_gt, xe, i) {
+		u32 tile, dev;
+
+		if (remote_gt->info.id == gt->info.id)
+			continue;
+
+		tile = gt_to_tile(remote_gt)->id;
+		dev = G2G_DEV(remote_gt);
+
+		for (t = 0; t < XE_G2G_TYPE_LIMIT; t++)
+			guc_g2g_deregister(guc, tile, dev, t);
+	}
+}
+
+/* Size of a single allocation that contains all G2G CTBs across all GTs */
+static u32 g2g_ctb_size(struct kunit *test, struct xe_device *xe)
+{
+	unsigned int count = xe->info.gt_count;
+	u32 num_channels = (count * (count - 1)) / 2;
+
+	kunit_info(test, "Size: (%d * %d / 2) * %d * 0x%08X + 0x%08X => 0x%08X [%d]\n",
+		   count, count - 1, XE_G2G_TYPE_LIMIT, G2G_BUFFER_SIZE, G2G_DESC_AREA_SIZE,
+		   num_channels * XE_G2G_TYPE_LIMIT * G2G_BUFFER_SIZE + G2G_DESC_AREA_SIZE,
+		   num_channels * XE_G2G_TYPE_LIMIT);
+
+	return num_channels * XE_G2G_TYPE_LIMIT * G2G_BUFFER_SIZE + G2G_DESC_AREA_SIZE;
+}
+
+/*
+ * Use the driver's regular CTB allocation scheme.
+ */
+static void g2g_alloc_default(struct kunit *test, struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	int i;
+
+	kunit_info(test, "Default [tiles = %d, GTs = %d]\n",
+		   xe->info.tile_count, xe->info.gt_count);
+
+	for_each_gt(gt, xe, i) {
+		struct xe_guc *guc = &gt->uc.guc;
+		int ret;
+
+		ret = guc_g2g_alloc(guc);
+		KUNIT_ASSERT_EQ_MSG(test, 0, ret, "G2G alloc failed: %pe", ERR_PTR(ret));
+		continue;
+	}
+}
+
+static void g2g_distribute(struct kunit *test, struct xe_device *xe, struct xe_bo *bo)
+{
+	struct xe_gt *root_gt, *gt;
+	int i;
+
+	root_gt = xe_device_get_gt(xe, 0);
+	root_gt->uc.guc.g2g.bo = bo;
+	root_gt->uc.guc.g2g.owned = true;
+	kunit_info(test, "[%d.%d] Assigned 0x%p\n", gt_to_tile(root_gt)->id, root_gt->info.id, bo);
+
+	for_each_gt(gt, xe, i) {
+		if (gt->info.id != 0) {
+			gt->uc.guc.g2g.owned = false;
+			gt->uc.guc.g2g.bo = xe_bo_get(bo);
+			kunit_info(test, "[%d.%d] Pinned 0x%p\n",
+				   gt_to_tile(gt)->id, gt->info.id, gt->uc.guc.g2g.bo);
+		}
+
+		KUNIT_ASSERT_NOT_ERR_OR_NULL(test, gt->uc.guc.g2g.bo);
+	}
+}
+
+/*
+ * Allocate a single blob on the host and split between all G2G CTBs.
+ */
+static void g2g_alloc_host(struct kunit *test, struct xe_device *xe)
+{
+	struct xe_bo *bo;
+	u32 g2g_size;
+
+	kunit_info(test, "Host [tiles = %d, GTs = %d]\n", xe->info.tile_count, xe->info.gt_count);
+
+	g2g_size = g2g_ctb_size(test, xe);
+	bo = xe_managed_bo_create_pin_map(xe, xe_device_get_root_tile(xe), g2g_size,
+					  XE_BO_FLAG_SYSTEM |
+					  XE_BO_FLAG_GGTT |
+					  XE_BO_FLAG_GGTT_ALL |
+					  XE_BO_FLAG_GGTT_INVALIDATE);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, bo);
+	kunit_info(test, "[HST] G2G buffer create: 0x%p\n", bo);
+
+	xe_map_memset(xe, &bo->vmap, 0, 0, g2g_size);
+
+	g2g_distribute(test, xe, bo);
+}
+
+/*
+ * Allocate a single blob on the given tile and split between all G2G CTBs.
+ */
+static void g2g_alloc_tile(struct kunit *test, struct xe_device *xe, struct xe_tile *tile)
+{
+	struct xe_bo *bo;
+	u32 g2g_size;
+
+	KUNIT_ASSERT_TRUE(test, IS_DGFX(xe));
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, tile);
+
+	kunit_info(test, "Tile %d [tiles = %d, GTs = %d]\n",
+		   tile->id, xe->info.tile_count, xe->info.gt_count);
+
+	g2g_size = g2g_ctb_size(test, xe);
+	bo = xe_managed_bo_create_pin_map(xe, tile, g2g_size,
+					  XE_BO_FLAG_VRAM_IF_DGFX(tile) |
+					  XE_BO_FLAG_GGTT |
+					  XE_BO_FLAG_GGTT_ALL |
+					  XE_BO_FLAG_GGTT_INVALIDATE);
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, bo);
+	kunit_info(test, "[%d.*] G2G buffer create: 0x%p\n", tile->id, bo);
+
+	xe_map_memset(xe, &bo->vmap, 0, 0, g2g_size);
+
+	g2g_distribute(test, xe, bo);
+}
+
+static void g2g_free(struct kunit *test, struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	struct xe_bo *bo;
+	int i;
+
+	for_each_gt(gt, xe, i) {
+		bo = gt->uc.guc.g2g.bo;
+		if (!bo)
+			continue;
+
+		if (gt->uc.guc.g2g.owned) {
+			xe_managed_bo_unpin_map_no_vm(bo);
+			kunit_info(test, "[%d.%d] Unmapped 0x%p\n",
+				   gt_to_tile(gt)->id, gt->info.id, bo);
+		} else {
+			xe_bo_put(bo);
+			kunit_info(test, "[%d.%d] Unpinned 0x%p\n",
+				   gt_to_tile(gt)->id, gt->info.id, bo);
+		}
+
+		gt->uc.guc.g2g.bo = NULL;
+	}
+}
+
+static void g2g_stop(struct kunit *test, struct xe_device *xe)
+{
+	struct xe_gt *gt;
+	int i;
+
+	for_each_gt(gt, xe, i) {
+		struct xe_guc *guc = &gt->uc.guc;
+
+		if (!guc->g2g.bo)
+			continue;
+
+		g2g_ct_stop(guc);
+	}
+
+	g2g_free(test, xe);
+}
+
+/*
+ * Generate a unique id for each bi-directional CTB for each pair of
+ * near and far tiles/devices. The id can then be used as an index into
+ * a single allocation that is sub-divided into multiple CTBs.
+ *
+ * For example, with two devices per tile and two tiles, the table should
+ * look like:
+ *           Far <tile>.<dev>
+ *         0.0   0.1   1.0   1.1
+ * N 0.0  --/-- 00/01 02/03 04/05
+ * e 0.1  01/00 --/-- 06/07 08/09
+ * a 1.0  03/02 07/06 --/-- 10/11
+ * r 1.1  05/04 09/08 11/10 --/--
+ *
+ * Where each entry is Rx/Tx channel id.
+ *
+ * So GuC #3 (tile 1, dev 1) talking to GuC #2 (tile 1, dev 0) would
+ * be reading from channel #11 and writing to channel #10. Whereas,
+ * GuC #2 talking to GuC #3 would be read on #10 and write to #11.
+ */
+static int g2g_slot_flat(u32 near_tile, u32 near_dev, u32 far_tile, u32 far_dev,
+			 u32 type, u32 max_inst, bool have_dev)
+{
+	u32 near = near_tile, far = far_tile;
+	u32 idx = 0, x, y, direction;
+	int i;
+
+	if (have_dev) {
+		near = (near << 1) | near_dev;
+		far = (far << 1) | far_dev;
+	}
+
+	/* No need to send to one's self */
+	if (far == near)
+		return -1;
+
+	if (far > near) {
+		/* Top right table half */
+		x = far;
+		y = near;
+
+		/* T/R is 'forwards' direction */
+		direction = type;
+	} else {
+		/* Bottom left table half */
+		x = near;
+		y = far;
+
+		/* B/L is 'backwards' direction */
+		direction = (1 - type);
+	}
+
+	/* Count the rows prior to the target */
+	for (i = y; i > 0; i--)
+		idx += max_inst - i;
+
+	/* Count this row up to the target */
+	idx += (x - 1 - y);
+
+	/* Slots are in Rx/Tx pairs */
+	idx *= 2;
+
+	/* Pick Rx/Tx direction */
+	idx += direction;
+
+	return idx;
+}
+
+static int g2g_register_flat(struct xe_guc *guc, u32 far_tile, u32 far_dev, u32 type, bool have_dev)
+{
+	struct xe_gt *gt = guc_to_gt(guc);
+	struct xe_device *xe = gt_to_xe(gt);
+	u32 near_tile = gt_to_tile(gt)->id;
+	u32 near_dev = G2G_DEV(gt);
+	u32 max = xe->info.gt_count;
+	int idx;
+	u32 base, desc, buf;
+
+	if (!guc->g2g.bo)
+		return -ENODEV;
+
+	idx = g2g_slot_flat(near_tile, near_dev, far_tile, far_dev, type, max, have_dev);
+	xe_assert(xe, idx >= 0);
+
+	base = guc_bo_ggtt_addr(guc, guc->g2g.bo);
+	desc = base + idx * G2G_DESC_SIZE;
+	buf = base + idx * G2G_BUFFER_SIZE + G2G_DESC_AREA_SIZE;
+
+	xe_assert(xe, (desc - base + G2G_DESC_SIZE) <= G2G_DESC_AREA_SIZE);
+	xe_assert(xe, (buf - base + G2G_BUFFER_SIZE) <= xe_bo_size(guc->g2g.bo));
+
+	return guc_action_register_g2g_buffer(guc, type, far_tile, far_dev,
+					      desc, buf, G2G_BUFFER_SIZE);
+}
+
+static void g2g_start(struct kunit *test, struct xe_guc *guc)
+{
+	struct xe_gt *remote_gt, *gt = guc_to_gt(guc);
+	struct xe_device *xe = gt_to_xe(gt);
+	unsigned int i;
+	int t, ret;
+	bool have_dev;
+
+	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, guc->g2g.bo);
+
+	/* GuC interface will need extending if more GT device types are ever created. */
+	KUNIT_ASSERT_TRUE(test,
+			  (gt->info.type == XE_GT_TYPE_MAIN) ||
+			  (gt->info.type == XE_GT_TYPE_MEDIA));
+
+	/* Channel numbering depends on whether there are multiple GTs per tile */
+	have_dev = xe->info.gt_count > xe->info.tile_count;
+
+	for_each_gt(remote_gt, xe, i) {
+		u32 tile, dev;
+
+		if (remote_gt->info.id == gt->info.id)
+			continue;
+
+		tile = gt_to_tile(remote_gt)->id;
+		dev = G2G_DEV(remote_gt);
+
+		for (t = 0; t < XE_G2G_TYPE_LIMIT; t++) {
+			ret = g2g_register_flat(guc, tile, dev, t, have_dev);
+			KUNIT_ASSERT_EQ_MSG(test, 0, ret, "G2G register failed: %pe", ERR_PTR(ret));
+		}
+	}
+}
+
+static void g2g_reinit(struct kunit *test, struct xe_device *xe, int ctb_type, struct xe_tile *tile)
+{
+	struct xe_gt *gt;
+	int i, found = 0;
+
+	g2g_stop(test, xe);
+
+	for_each_gt(gt, xe, i) {
+		struct xe_guc *guc = &gt->uc.guc;
+
+		KUNIT_ASSERT_NULL(test, guc->g2g.bo);
+	}
+
+	switch (ctb_type) {
+	case G2G_CTB_TYPE_DEFAULT:
+		g2g_alloc_default(test, xe);
+		break;
+
+	case G2G_CTB_TYPE_HOST:
+		g2g_alloc_host(test, xe);
+		break;
+
+	case G2G_CTB_TYPE_TILE:
+		g2g_alloc_tile(test, xe, tile);
+		break;
+
+	default:
+		KUNIT_ASSERT_TRUE(test, false);
+	}
+
+	for_each_gt(gt, xe, i) {
+		struct xe_guc *guc = &gt->uc.guc;
+
+		if (!guc->g2g.bo)
+			continue;
+
+		if (ctb_type == G2G_CTB_TYPE_DEFAULT)
+			guc_g2g_start(guc);
+		else
+			g2g_start(test, guc);
+		found++;
+	}
+
+	KUNIT_ASSERT_GT_MSG(test, found, 1, "insufficient G2G channels running: %d", found);
+
+	kunit_info(test, "Testing across %d GTs\n", found);
+}
+
+static void g2g_recreate_ctb(void *_xe)
+{
+	struct xe_device *xe = (struct xe_device *)_xe;
+	struct kunit *test = kunit_get_current_test();
+
+	g2g_stop(test, xe);
+
+	if (xe_guc_g2g_wanted(xe))
+		g2g_reinit(test, xe, G2G_CTB_TYPE_DEFAULT, NULL);
+}
+
+static void g2g_pm_runtime_put(void *_xe)
+{
+	struct xe_device *xe = (struct xe_device *)_xe;
+
+	xe_pm_runtime_put(xe);
+}
+
+static void g2g_pm_runtime_get(struct kunit *test)
+{
+	struct xe_device *xe = test->priv;
+	int ret;
+
+	xe_pm_runtime_get(xe);
+	ret = kunit_add_action_or_reset(test, g2g_pm_runtime_put, xe);
+	KUNIT_ASSERT_EQ_MSG(test, 0, ret, "Failed to register runtime PM action\n");
+}
+
+static void g2g_check_skip(struct kunit *test)
+{
+	struct xe_device *xe = test->priv;
+	struct xe_gt *gt;
+	int i;
+
+	if (IS_SRIOV_VF(xe))
+		kunit_skip(test, "not supported from a VF");
+
+	if (xe->info.gt_count <= 1)
+		kunit_skip(test, "not enough GTs");
+
+	for_each_gt(gt, xe, i) {
+		struct xe_guc *guc = &gt->uc.guc;
+
+		if (guc->fw.build_type == CSS_UKERNEL_INFO_BUILDTYPE_PROD)
+			kunit_skip(test,
+				   "G2G test interface not available in production firmware builds\n");
+	}
+}
+
+/*
+ * Simple test that does not try to recreate the CTBs.
+ * Requires that the platform already enables G2G comms
+ * but has no risk of leaving the system in a broken state
+ * afterwards.
+ */
+static void xe_live_guc_g2g_kunit_default(struct kunit *test)
+{
+	struct xe_device *xe = test->priv;
+
+	if (!xe_guc_g2g_wanted(xe))
+		kunit_skip(test, "G2G not enabled");
+
+	g2g_check_skip(test);
+
+	g2g_pm_runtime_get(test);
+
+	kunit_info(test, "Testing default CTBs\n");
+	g2g_run_test(test, xe);
+
+	kunit_release_action(test, &g2g_pm_runtime_put, xe);
+}
+
+/*
+ * More complex test that re-creates the CTBs in various location to
+ * test access to each location from each GuC. Can be run even on
+ * systems that do not enable G2G by default. On the other hand,
+ * because it recreates the CTBs, if something goes wrong it could
+ * leave the system with broken G2G comms.
+ */
+static void xe_live_guc_g2g_kunit_allmem(struct kunit *test)
+{
+	struct xe_device *xe = test->priv;
+	int ret;
+
+	g2g_check_skip(test);
+
+	g2g_pm_runtime_get(test);
+
+	/* Make sure to leave the system as we found it */
+	ret = kunit_add_action_or_reset(test, g2g_recreate_ctb, xe);
+	KUNIT_ASSERT_EQ_MSG(test, 0, ret, "Failed to register CTB re-creation action\n");
+
+	kunit_info(test, "Testing CTB type 'default'...\n");
+	g2g_reinit(test, xe, G2G_CTB_TYPE_DEFAULT, NULL);
+	g2g_run_test(test, xe);
+
+	kunit_info(test, "Testing CTB type 'host'...\n");
+	g2g_reinit(test, xe, G2G_CTB_TYPE_HOST, NULL);
+	g2g_run_test(test, xe);
+
+	if (IS_DGFX(xe)) {
+		struct xe_tile *tile;
+		int id;
+
+		for_each_tile(tile, xe, id) {
+			kunit_info(test, "Testing CTB type 'tile: #%d'...\n", id);
+
+			g2g_reinit(test, xe, G2G_CTB_TYPE_TILE, tile);
+			g2g_run_test(test, xe);
+		}
+	} else {
+		kunit_info(test, "Skipping local memory on integrated platform\n");
+	}
+
+	kunit_release_action(test, g2g_recreate_ctb, xe);
+	kunit_release_action(test, g2g_pm_runtime_put, xe);
+}
+
+static struct kunit_case xe_guc_g2g_tests[] = {
+	KUNIT_CASE_PARAM(xe_live_guc_g2g_kunit_default, xe_pci_live_device_gen_param),
+	KUNIT_CASE_PARAM(xe_live_guc_g2g_kunit_allmem, xe_pci_live_device_gen_param),
+	{}
+};
+
+VISIBLE_IF_KUNIT
+struct kunit_suite xe_guc_g2g_test_suite = {
+	.name = "xe_guc_g2g",
+	.test_cases = xe_guc_g2g_tests,
+	.init = xe_kunit_helper_xe_device_live_test_init,
+};
+EXPORT_SYMBOL_IF_KUNIT(xe_guc_g2g_test_suite);
diff --git a/drivers/gpu/drm/xe/tests/xe_live_test_mod.c b/drivers/gpu/drm/xe/tests/xe_live_test_mod.c
index 81277c77016d..c55e46f1ae92 100644
--- a/drivers/gpu/drm/xe/tests/xe_live_test_mod.c
+++ b/drivers/gpu/drm/xe/tests/xe_live_test_mod.c
@@ -10,12 +10,14 @@ extern struct kunit_suite xe_bo_shrink_test_suite;
 extern struct kunit_suite xe_dma_buf_test_suite;
 extern struct kunit_suite xe_migrate_test_suite;
 extern struct kunit_suite xe_mocs_test_suite;
+extern struct kunit_suite xe_guc_g2g_test_suite;
 
 kunit_test_suite(xe_bo_test_suite);
 kunit_test_suite(xe_bo_shrink_test_suite);
 kunit_test_suite(xe_dma_buf_test_suite);
 kunit_test_suite(xe_migrate_test_suite);
 kunit_test_suite(xe_mocs_test_suite);
+kunit_test_suite(xe_guc_g2g_test_suite);
 
 MODULE_AUTHOR("Intel Corporation");
 MODULE_LICENSE("GPL");
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 8233002e1c5b..02e313cf0ae6 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -595,6 +595,13 @@ struct xe_device {
 		u8 region_mask;
 	} psmi;
 
+#if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST)
+	/** @g2g_test_array: for testing G2G communications */
+	u32 *g2g_test_array;
+	/** @g2g_test_count: for testing G2G communications */
+	atomic_t g2g_test_count;
+#endif
+
 	/* private: */
 
 #if IS_ENABLED(CONFIG_DRM_XE_DISPLAY)
diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 4eb06ffc09bc..171b49817adf 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -1684,3 +1684,7 @@ void xe_guc_declare_wedged(struct xe_guc *guc)
 	xe_guc_ct_stop(&guc->ct);
 	xe_guc_submit_wedge(guc);
 }
+
+#if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST)
+#include "tests/xe_guc_g2g_test.c"
+#endif
diff --git a/drivers/gpu/drm/xe/xe_guc.h b/drivers/gpu/drm/xe/xe_guc.h
index 22cf019a11bf..1cca05967e62 100644
--- a/drivers/gpu/drm/xe/xe_guc.h
+++ b/drivers/gpu/drm/xe/xe_guc.h
@@ -53,6 +53,10 @@ void xe_guc_stop(struct xe_guc *guc);
 int xe_guc_start(struct xe_guc *guc);
 void xe_guc_declare_wedged(struct xe_guc *guc);
 
+#if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST)
+int xe_guc_g2g_test_notification(struct xe_guc *guc, u32 *payload, u32 len);
+#endif
+
 static inline u16 xe_engine_class_to_guc_class(enum xe_engine_class class)
 {
 	switch (class) {
diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c
index bb8650fdeb45..18f6327bf552 100644
--- a/drivers/gpu/drm/xe/xe_guc_ct.c
+++ b/drivers/gpu/drm/xe/xe_guc_ct.c
@@ -1486,6 +1486,11 @@ static int process_g2h_msg(struct xe_guc_ct *ct, u32 *msg, u32 len)
 	case XE_GUC_ACTION_NOTIFY_EXCEPTION:
 		ret = guc_crash_process_msg(ct, action);
 		break;
+#if IS_ENABLED(CONFIG_DRM_XE_KUNIT_TEST)
+	case XE_GUC_ACTION_TEST_G2G_RECV:
+		ret = xe_guc_g2g_test_notification(guc, payload, adj_len);
+		break;
+#endif
 	default:
 		xe_gt_err(gt, "unexpected G2H action 0x%04x\n", action);
 	}
diff --git a/drivers/gpu/drm/xe/xe_guc_fwif.h b/drivers/gpu/drm/xe/xe_guc_fwif.h
index a72ff8a68568..50c4c2406132 100644
--- a/drivers/gpu/drm/xe/xe_guc_fwif.h
+++ b/drivers/gpu/drm/xe/xe_guc_fwif.h
@@ -15,6 +15,7 @@
 #define G2H_LEN_DW_SCHED_CONTEXT_MODE_SET	4
 #define G2H_LEN_DW_DEREGISTER_CONTEXT		3
 #define G2H_LEN_DW_TLB_INVALIDATE		3
+#define G2H_LEN_DW_G2G_NOTIFY_MIN		3
 
 #define GUC_ID_MAX			65535
 #define GUC_ID_UNKNOWN			0xffffffff

From d8586afe38984f9eb98d6753c9b1134295d55ceb Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Mon, 8 Sep 2025 17:15:36 -0600
Subject: [PATCH 1263/1455] drm/amdgpu: Remove volatile from CSB functions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The CSB buffer manipulation occurs in memory where the BO is mapped
during initialization, and some references to this buffer are handled
with volatile, which is incorrect in this scenario. There are a few
cases where the use of volatile is accepted, but none of them align with
CSB operations. Therefore, this commit removes all the volatile
variables associated with the CSB code.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c | 6 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h | 6 +++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h | 4 ++--
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c  | 3 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c  | 3 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c  | 3 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c   | 7 +++----
 drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c   | 5 ++---
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c   | 3 +--
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c   | 3 +--
 10 files changed, 18 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index c80c8f543532..89fc1015d3a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -2279,7 +2279,7 @@ void amdgpu_gfx_profile_ring_end_use(struct amdgpu_ring *ring)
  * Return:
  * return the latest index.
  */
-u32 amdgpu_gfx_csb_preamble_start(volatile u32 *buffer)
+u32 amdgpu_gfx_csb_preamble_start(u32 *buffer)
 {
 	u32 count = 0;
 
@@ -2303,7 +2303,7 @@ u32 amdgpu_gfx_csb_preamble_start(volatile u32 *buffer)
  * Return:
  * return the latest index.
  */
-u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, volatile u32 *buffer, u32 count)
+u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, u32 *buffer, u32 count)
 {
 	const struct cs_section_def *sect = NULL;
 	const struct cs_extent_def *ext = NULL;
@@ -2330,7 +2330,7 @@ u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, volatile u32 *buffer,
  * @buffer: This is an output variable that gets the PACKET3 preamble end.
  * @count: Index to start set the preemble end.
  */
-void amdgpu_gfx_csb_preamble_end(volatile u32 *buffer, u32 count)
+void amdgpu_gfx_csb_preamble_end(u32 *buffer, u32 count)
 {
 	buffer[count++] = cpu_to_le32(PACKET3(PACKET3_PREAMBLE_CNTL, 0));
 	buffer[count++] = cpu_to_le32(PACKET3_PREAMBLE_END_CLEAR_STATE);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index 08f268dab8f5..fb5f7a0ee029 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -642,9 +642,9 @@ void amdgpu_gfx_enforce_isolation_ring_end_use(struct amdgpu_ring *ring);
 void amdgpu_gfx_profile_idle_work_handler(struct work_struct *work);
 void amdgpu_gfx_profile_ring_begin_use(struct amdgpu_ring *ring);
 void amdgpu_gfx_profile_ring_end_use(struct amdgpu_ring *ring);
-u32 amdgpu_gfx_csb_preamble_start(volatile u32 *buffer);
-u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, volatile u32 *buffer, u32 count);
-void amdgpu_gfx_csb_preamble_end(volatile u32 *buffer, u32 count);
+u32 amdgpu_gfx_csb_preamble_start(u32 *buffer);
+u32 amdgpu_gfx_csb_data_parser(struct amdgpu_device *adev, u32 *buffer, u32 count);
+void amdgpu_gfx_csb_preamble_end(u32 *buffer, u32 count);
 
 void amdgpu_debugfs_gfx_sched_mask_init(struct amdgpu_device *adev);
 void amdgpu_debugfs_compute_sched_mask_init(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
index c210625be220..0f2647d099b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
@@ -251,7 +251,7 @@ struct amdgpu_rlc_funcs {
 	 * and it also provides a pointer to it which is used by the firmware
 	 * to load the clear state in some cases.
 	 */
-	void (*get_csb_buffer)(struct amdgpu_device *adev, volatile u32 *buffer);
+	void (*get_csb_buffer)(struct amdgpu_device *adev, u32 *buffer);
 	int  (*get_cp_table_num)(struct amdgpu_device *adev);
 	int  (*resume)(struct amdgpu_device *adev);
 	void (*stop)(struct amdgpu_device *adev);
@@ -281,7 +281,7 @@ struct amdgpu_rlc {
 	/* for clear state */
 	struct amdgpu_bo        *clear_state_obj;
 	uint64_t                clear_state_gpu_addr;
-	volatile uint32_t       *cs_ptr;
+	uint32_t		*cs_ptr;
 	const struct cs_section_def   *cs_data;
 	u32                     clear_state_size;
 	/* for cp tables */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index 264183ab24ec..f6ac6a36bc44 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4322,8 +4322,7 @@ static u32 gfx_v10_0_get_csb_size(struct amdgpu_device *adev)
 	return count;
 }
 
-static void gfx_v10_0_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
+static void gfx_v10_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0;
 	int ctx_reg_offset;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 3d9c045a8a64..ff600a6c80ae 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -850,8 +850,7 @@ static u32 gfx_v11_0_get_csb_size(struct amdgpu_device *adev)
 	return count;
 }
 
-static void gfx_v11_0_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
+static void gfx_v11_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0;
 	int ctx_reg_offset;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index 5dbc5dbc694a..a14fd94af90d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -685,8 +685,7 @@ static u32 gfx_v12_0_get_csb_size(struct amdgpu_device *adev)
 	return count;
 }
 
-static void gfx_v12_0_get_csb_buffer(struct amdgpu_device *adev,
-				     volatile u32 *buffer)
+static void gfx_v12_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0, clustercount = 0, i;
 	const struct cs_section_def *sect = NULL;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index 70d7a1f434c4..7693b7953426 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -86,7 +86,7 @@ MODULE_FIRMWARE("amdgpu/hainan_ce.bin");
 MODULE_FIRMWARE("amdgpu/hainan_rlc.bin");
 
 static u32 gfx_v6_0_get_csb_size(struct amdgpu_device *adev);
-static void gfx_v6_0_get_csb_buffer(struct amdgpu_device *adev, volatile u32 *buffer);
+static void gfx_v6_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer);
 //static void gfx_v6_0_init_cp_pg_table(struct amdgpu_device *adev);
 static void gfx_v6_0_init_pg(struct amdgpu_device *adev);
 
@@ -2354,7 +2354,7 @@ static void gfx_v6_0_ring_emit_wreg(struct amdgpu_ring *ring,
 static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
 {
 	const u32 *src_ptr;
-	volatile u32 *dst_ptr;
+	u32 *dst_ptr;
 	u32 dws;
 	u64 reg_list_mc_addr;
 	const struct cs_section_def *cs_data;
@@ -2855,8 +2855,7 @@ static u32 gfx_v6_0_get_csb_size(struct amdgpu_device *adev)
 	return count;
 }
 
-static void gfx_v6_0_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
+static void gfx_v6_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 2aa323dab34e..5976ed55d9db 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -883,7 +883,7 @@ static const u32 kalindi_rlc_save_restore_register_list[] = {
 };
 
 static u32 gfx_v7_0_get_csb_size(struct amdgpu_device *adev);
-static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev, volatile u32 *buffer);
+static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer);
 static void gfx_v7_0_init_pg(struct amdgpu_device *adev);
 static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev);
 
@@ -3882,8 +3882,7 @@ static u32 gfx_v7_0_get_csb_size(struct amdgpu_device *adev)
 	return count;
 }
 
-static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
+static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 367449d8061b..0856ff65288c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -1220,8 +1220,7 @@ out:
 	return err;
 }
 
-static void gfx_v8_0_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
+static void gfx_v8_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index a6ff9a137a83..dd19a97436db 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1648,8 +1648,7 @@ static u32 gfx_v9_0_get_csb_size(struct amdgpu_device *adev)
 	return count;
 }
 
-static void gfx_v9_0_get_csb_buffer(struct amdgpu_device *adev,
-				    volatile u32 *buffer)
+static void gfx_v9_0_get_csb_buffer(struct amdgpu_device *adev, u32 *buffer)
 {
 	u32 count = 0;
 

From b8fc541089582ed77cc567bb28a83abe834e8902 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Mon, 8 Sep 2025 17:15:37 -0600
Subject: [PATCH 1264/1455] drm/amdgpu: Remove volatile from RLC files
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The RLC uses volatile with some pointers that are not directly related
to any of the situations where volatile is advised to be used [1]. For
this reason, this commit removes all the volatile occurrences associated
with RLC.

1. https://docs.kernel.org/process/volatile-considered-harmful.html

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
index db5791e1a7ce..5aa830a02d80 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
@@ -89,7 +89,7 @@ void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev, int xcc_id)
 int amdgpu_gfx_rlc_init_sr(struct amdgpu_device *adev, u32 dws)
 {
 	const u32 *src_ptr;
-	volatile u32 *dst_ptr;
+	u32 *dst_ptr;
 	u32 i;
 	int r;
 
@@ -189,7 +189,7 @@ int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev)
 void amdgpu_gfx_rlc_setup_cp_table(struct amdgpu_device *adev)
 {
 	const __le32 *fw_data;
-	volatile u32 *dst_ptr;
+	u32 *dst_ptr;
 	int me, i, max_me;
 	u32 bo_offset = 0;
 	u32 table_offset, table_size;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
index 0f2647d099b2..2ce310b31942 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
@@ -275,7 +275,7 @@ struct amdgpu_rlc {
 	/* for power gating */
 	struct amdgpu_bo        *save_restore_obj;
 	uint64_t                save_restore_gpu_addr;
-	volatile uint32_t       *sr_ptr;
+	uint32_t		*sr_ptr;
 	const u32               *reg_list;
 	u32                     reg_list_size;
 	/* for clear state */
@@ -287,7 +287,7 @@ struct amdgpu_rlc {
 	/* for cp tables */
 	struct amdgpu_bo        *cp_table_obj;
 	uint64_t                cp_table_gpu_addr;
-	volatile uint32_t       *cp_table_ptr;
+	uint32_t		*cp_table_ptr;
 	u32                     cp_table_size;
 
 	/* safe mode for updating CG/PG state */

From f307cfb91734b36bde2597552971553d91838048 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Mon, 8 Sep 2025 17:15:38 -0600
Subject: [PATCH 1265/1455] drm/amdgpu: Remove volatile from ring manipulation
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

None of the pointer operations handled by the ring file requires
volatile, for this reason, this commit removes all occurrences of
volatile associated with rings.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h      |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 12 ++++++------
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c   |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c   |  2 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c   |  2 +-
 5 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 17848ce65d1f..6d123172bb0a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -545,7 +545,7 @@ struct amdgpu_wb {
 	 * this value can be accessed directly by using the offset as an index.
 	 * For the GPU address, it is necessary to use gpu_addr and the offset.
 	 */
-	volatile uint32_t	*wb;
+	uint32_t		*wb;
 
 	/**
 	 * @gpu_addr:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 7670f5d82b9e..80b85547c810 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -114,7 +114,7 @@ struct amdgpu_sched {
  */
 struct amdgpu_fence_driver {
 	uint64_t			gpu_addr;
-	volatile uint32_t		*cpu_addr;
+	uint32_t			*cpu_addr;
 	/* sync_seq is protected by ring emission lock */
 	uint32_t			sync_seq;
 	atomic_t			last_seq;
@@ -298,7 +298,7 @@ struct amdgpu_ring {
 	unsigned int		ring_backup_entries_to_copy;
 	unsigned		rptr_offs;
 	u64			rptr_gpu_addr;
-	volatile u32		*rptr_cpu_addr;
+	u32			*rptr_cpu_addr;
 
 	/**
 	 * @wptr:
@@ -378,19 +378,19 @@ struct amdgpu_ring {
 	 * This is the CPU address pointer in the writeback slot. This is used
 	 * to commit changes to the GPU.
 	 */
-	volatile u32		*wptr_cpu_addr;
+	u32			*wptr_cpu_addr;
 	unsigned		fence_offs;
 	u64			fence_gpu_addr;
-	volatile u32		*fence_cpu_addr;
+	u32			*fence_cpu_addr;
 	uint64_t		current_ctx;
 	char			name[16];
 	u32                     trail_seq;
 	unsigned		trail_fence_offs;
 	u64			trail_fence_gpu_addr;
-	volatile u32		*trail_fence_cpu_addr;
+	u32			*trail_fence_cpu_addr;
 	unsigned		cond_exe_offs;
 	u64			cond_exe_gpu_addr;
-	volatile u32		*cond_exe_cpu_addr;
+	u32			*cond_exe_cpu_addr;
 	unsigned int		set_q_mode_offs;
 	u32			*set_q_mode_ptr;
 	u64			set_q_mode_token;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
index f6ac6a36bc44..8841d7213de4 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -4075,7 +4075,7 @@ static int gfx_v10_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	struct dma_fence *f = NULL;
 	unsigned int index;
 	uint64_t gpu_addr;
-	volatile uint32_t *cpu_ptr;
+	uint32_t *cpu_ptr;
 	long r;
 
 	memset(&ib, 0, sizeof(ib));
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index ff600a6c80ae..86b3fcab5772 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -603,7 +603,7 @@ static int gfx_v11_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	struct dma_fence *f = NULL;
 	unsigned index;
 	uint64_t gpu_addr;
-	volatile uint32_t *cpu_ptr;
+	uint32_t *cpu_ptr;
 	long r;
 
 	/* MES KIQ fw hasn't indirect buffer support for now */
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
index a14fd94af90d..710ec9c34e43 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c
@@ -497,7 +497,7 @@ static int gfx_v12_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
 	struct dma_fence *f = NULL;
 	unsigned index;
 	uint64_t gpu_addr;
-	volatile uint32_t *cpu_ptr;
+	uint32_t *cpu_ptr;
 	long r;
 
 	/* MES KIQ fw hasn't indirect buffer support for now */

From 897025300721eb9b585c9b90ae531a5f3ee29022 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Mon, 8 Sep 2025 17:15:39 -0600
Subject: [PATCH 1266/1455] drm/amdgpu: Remove volatile from amdgpu and
 amdgpu_ih headers
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remove the unnecessary use of volatile in some of the amdgpu.h and
amdgpu_ih.h headers.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h    | 2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 6d123172bb0a..0e6a09651bbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -721,7 +721,7 @@ int amdgpu_cs_wait_fences_ioctl(struct drm_device *dev, void *data,
 /* VRAM scratch page for HDP bug, default vram page */
 struct amdgpu_mem_scratch {
 	struct amdgpu_bo		*robj;
-	volatile uint32_t		*ptr;
+	uint32_t			*ptr;
 	u64				gpu_addr;
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
index 7f7ea046e209..f58b6be7fccc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h
@@ -56,14 +56,14 @@ struct amdgpu_ih_ring {
 	bool			use_bus_addr;
 
 	struct amdgpu_bo	*ring_obj;
-	volatile uint32_t	*ring;
+	uint32_t		*ring;
 	uint64_t		gpu_addr;
 
 	uint64_t		wptr_addr;
-	volatile uint32_t	*wptr_cpu;
+	uint32_t		*wptr_cpu;
 
 	uint64_t		rptr_addr;
-	volatile uint32_t	*rptr_cpu;
+	uint32_t		*rptr_cpu;
 
 	bool                    enabled;
 	unsigned		rptr;

From 65307484851980aaa965e35d84ab7cb195308f2c Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Mon, 8 Sep 2025 17:15:40 -0600
Subject: [PATCH 1267/1455] drm/amdgpu: Remove volatile references from VCN
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c   |  2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c   | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c   | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c   | 12 ++++++------
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c   | 14 +++++++-------
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c | 12 ++++++------
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c | 10 +++++-----
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 12 ++++++------
 10 files changed, 51 insertions(+), 51 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index c2f69369bda6..5ae7cc0d5f57 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -1157,7 +1157,7 @@ static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf,
 {
 	struct amdgpu_vcn_inst *vcn;
 	void *log_buf;
-	volatile struct amdgpu_vcn_fwlog *plog;
+	struct amdgpu_vcn_fwlog *plog;
 	unsigned int read_pos, write_pos, available, i, read_bytes = 0;
 	unsigned int read_num[2] = {0};
 
@@ -1170,7 +1170,7 @@ static ssize_t amdgpu_debugfs_vcn_fwlog_read(struct file *f, char __user *buf,
 
 	log_buf = vcn->fw_shared.cpu_addr + vcn->fw_shared.mem_size;
 
-	plog = (volatile struct amdgpu_vcn_fwlog *)log_buf;
+	plog = (struct amdgpu_vcn_fwlog *)log_buf;
 	read_pos = plog->rptr;
 	write_pos = plog->wptr;
 
@@ -1237,11 +1237,11 @@ void amdgpu_debugfs_vcn_fwlog_init(struct amdgpu_device *adev, uint8_t i,
 void amdgpu_vcn_fwlog_init(struct amdgpu_vcn_inst *vcn)
 {
 #if defined(CONFIG_DEBUG_FS)
-	volatile uint32_t *flag = vcn->fw_shared.cpu_addr;
+	uint32_t *flag = vcn->fw_shared.cpu_addr;
 	void *fw_log_cpu_addr = vcn->fw_shared.cpu_addr + vcn->fw_shared.mem_size;
 	uint64_t fw_log_gpu_addr = vcn->fw_shared.gpu_addr + vcn->fw_shared.mem_size;
-	volatile struct amdgpu_vcn_fwlog *log_buf = fw_log_cpu_addr;
-	volatile struct amdgpu_fw_shared_fw_logging *fw_log = vcn->fw_shared.cpu_addr
+	struct amdgpu_vcn_fwlog *log_buf = fw_log_cpu_addr;
+	struct amdgpu_fw_shared_fw_logging *fw_log = vcn->fw_shared.cpu_addr
 							 + vcn->fw_shared.log_offset;
 	*flag |= cpu_to_le32(AMDGPU_VCN_FW_LOGGING_FLAG);
 	fw_log->is_enabled = 1;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index 3c1c844c5fde..a316797875a8 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -193,7 +193,7 @@ static int vcn_v1_0_sw_init(struct amdgpu_ip_block *ip_block)
 	adev->vcn.inst[0].pause_dpg_mode = vcn_v1_0_pause_dpg_mode;
 
 	if (amdgpu_vcnfw_log) {
-		volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+		struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 
 		fw_shared->present_flag_0 = 0;
 		amdgpu_vcn_fwlog_init(adev->vcn.inst);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
index c4e2659f8933..8897dcc9c1a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c
@@ -137,7 +137,7 @@ static int vcn_v2_0_sw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_ring *ring;
 	int i, r;
 	struct amdgpu_device *adev = ip_block->adev;
-	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_fw_shared *fw_shared;
 
 	/* VCN DEC TRAP */
 	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN,
@@ -252,7 +252,7 @@ static int vcn_v2_0_sw_fini(struct amdgpu_ip_block *ip_block)
 {
 	int r, idx;
 	struct amdgpu_device *adev = ip_block->adev;
-	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		fw_shared->present_flag_0 = 0;
@@ -853,7 +853,7 @@ static void vcn_v2_0_enable_static_power_gating(struct amdgpu_vcn_inst *vinst)
 static int vcn_v2_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 {
 	struct amdgpu_device *adev = vinst->adev;
-	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
 	uint32_t rb_bufsz, tmp;
 	int ret;
@@ -1001,7 +1001,7 @@ static int vcn_v2_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 static int vcn_v2_0_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
-	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+	struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 	struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec;
 	uint32_t rb_bufsz, tmp;
 	uint32_t lmi_swap_cntl;
@@ -1308,7 +1308,7 @@ static int vcn_v2_0_pause_dpg_mode(struct amdgpu_vcn_inst *vinst,
 				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 			if (!ret_code) {
-				volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
+				struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr;
 				/* pause DPG */
 				reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK;
 				WREG32_SOC15(UVD, 0, mmUVD_DPG_PAUSE, reg_data);
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index 0505f0955324..cebee453871c 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -277,7 +277,7 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block)
 	struct amdgpu_device *adev = ip_block->adev;
 
 	for (j = 0; j < adev->vcn.num_vcn_inst; j++) {
-		volatile struct amdgpu_fw_shared *fw_shared;
+		struct amdgpu_fw_shared *fw_shared;
 
 		if (adev->vcn.harvest_config & (1 << j))
 			continue;
@@ -420,7 +420,7 @@ static int vcn_v2_5_sw_fini(struct amdgpu_ip_block *ip_block)
 {
 	int i, r, idx;
 	struct amdgpu_device *adev = ip_block->adev;
-	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_fw_shared *fw_shared;
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
@@ -998,7 +998,7 @@ static int vcn_v2_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
+	struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t rb_bufsz, tmp;
 	int ret;
@@ -1155,7 +1155,7 @@ static int vcn_v2_5_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_fw_shared *fw_shared =
+	struct amdgpu_fw_shared *fw_shared =
 		adev->vcn.inst[i].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t rb_bufsz, tmp;
@@ -1667,7 +1667,7 @@ static int vcn_v2_5_pause_dpg_mode(struct amdgpu_vcn_inst *vinst,
 				UVD_POWER_STATUS__UVD_POWER_STATUS_MASK);
 
 			if (!ret_code) {
-				volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
+				struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 
 				/* pause DPG */
 				reg_data |= UVD_DPG_PAUSE__NJ_PAUSE_DPG_REQ_MASK;
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
index 77a8373444df..d9cf8f0feeb3 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c
@@ -191,7 +191,7 @@ static int vcn_v3_0_sw_init(struct amdgpu_ip_block *ip_block)
 	}
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		volatile struct amdgpu_fw_shared *fw_shared;
+		struct amdgpu_fw_shared *fw_shared;
 
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
@@ -327,7 +327,7 @@ static int vcn_v3_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-			volatile struct amdgpu_fw_shared *fw_shared;
+			struct amdgpu_fw_shared *fw_shared;
 
 			if (adev->vcn.harvest_config & (1 << i))
 				continue;
@@ -1029,7 +1029,7 @@ static int vcn_v3_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
+	struct amdgpu_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t rb_bufsz, tmp;
 	int ret;
@@ -1194,7 +1194,7 @@ static int vcn_v3_0_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t rb_bufsz, tmp;
 	int j, k, r;
@@ -1715,7 +1715,7 @@ static int vcn_v3_0_pause_dpg_mode(struct amdgpu_vcn_inst *vinst,
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t reg_data = 0;
 	int ret_code;
@@ -1834,7 +1834,7 @@ static uint64_t vcn_v3_0_dec_ring_get_wptr(struct amdgpu_ring *ring)
 static void vcn_v3_0_dec_ring_set_wptr(struct amdgpu_ring *ring)
 {
 	struct amdgpu_device *adev = ring->adev;
-	volatile struct amdgpu_fw_shared *fw_shared;
+	struct amdgpu_fw_shared *fw_shared;
 
 	if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) {
 		/*whenever update RBC_RB_WPTR, we save the wptr in shared rb.wptr and scratch2 */
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 870c563415af..8d3f18b55fc2 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -148,7 +148,7 @@ static int vcn_v4_0_early_init(struct amdgpu_ip_block *ip_block)
 
 static int vcn_v4_0_fw_shared_init(struct amdgpu_device *adev, int inst_idx)
 {
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 
 	fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	fw_shared->present_flag_0 = cpu_to_le32(AMDGPU_FW_SHARED_FLAG_0_UNIFIED_QUEUE);
@@ -278,7 +278,7 @@ static int vcn_v4_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-			volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+			struct amdgpu_vcn4_fw_shared *fw_shared;
 
 			if (adev->vcn.harvest_config & (1 << i))
 				continue;
@@ -997,7 +997,7 @@ static int vcn_v4_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst, bool indirect)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
+	struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int ret;
@@ -1137,7 +1137,7 @@ static int vcn_v4_0_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int j, k, r;
@@ -1354,8 +1354,8 @@ static int vcn_v4_0_start_sriov(struct amdgpu_device *adev)
 	struct mmsch_v4_0_cmd_end end = { {0} };
 	struct mmsch_v4_0_init_header header;
 
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
-	volatile struct amdgpu_fw_shared_rb_setup *rb_setup;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_fw_shared_rb_setup *rb_setup;
 
 	direct_wt.cmd_header.command_type =
 		MMSCH_COMMAND__DIRECT_REG_WRITE;
@@ -1606,7 +1606,7 @@ static int vcn_v4_0_stop(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 	uint32_t tmp;
 	int r = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
index 5fbd755fc24c..eacf4e93ba2f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c
@@ -263,7 +263,7 @@ static int vcn_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	if (drm_dev_enter(&adev->ddev, &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-			volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+			struct amdgpu_vcn4_fw_shared *fw_shared;
 
 			fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 			fw_shared->present_flag_0 = 0;
@@ -845,7 +845,7 @@ static int vcn_v4_0_3_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared =
+	struct amdgpu_vcn4_fw_shared *fw_shared =
 						adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	int vcn_inst, ret;
@@ -1012,8 +1012,8 @@ static int vcn_v4_0_3_start_sriov(struct amdgpu_device *adev)
 	struct mmsch_v4_0_cmd_end end = { {0} };
 	struct mmsch_v4_0_3_init_header header;
 
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
-	volatile struct amdgpu_fw_shared_rb_setup *rb_setup;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_fw_shared_rb_setup *rb_setup;
 
 	direct_wt.cmd_header.command_type =
 		MMSCH_COMMAND__DIRECT_REG_WRITE;
@@ -1187,7 +1187,7 @@ static int vcn_v4_0_3_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	int j, k, r, vcn_inst;
 	uint32_t tmp;
@@ -1397,7 +1397,7 @@ static int vcn_v4_0_3_stop(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 	int r = 0, vcn_inst;
 	uint32_t tmp;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
index 732d4a78ab52..b107ee80e472 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c
@@ -149,7 +149,7 @@ static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block)
 	int i, r;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+		struct amdgpu_vcn4_fw_shared *fw_shared;
 
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
@@ -249,7 +249,7 @@ static int vcn_v4_0_5_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-			volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+			struct amdgpu_vcn4_fw_shared *fw_shared;
 
 			if (adev->vcn.harvest_config & (1 << i))
 				continue;
@@ -910,7 +910,7 @@ static int vcn_v4_0_5_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
+	struct amdgpu_vcn4_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int ret;
@@ -1047,7 +1047,7 @@ static int vcn_v4_0_5_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int j, k, r;
@@ -1266,7 +1266,7 @@ static int vcn_v4_0_5_stop(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn4_fw_shared *fw_shared;
+	struct amdgpu_vcn4_fw_shared *fw_shared;
 	uint32_t tmp;
 	int r = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
index 680df5d48c7a..0202df5db1e1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c
@@ -129,7 +129,7 @@ static int vcn_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block)
 	int i, r;
 
 	for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-		volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+		struct amdgpu_vcn5_fw_shared *fw_shared;
 
 		if (adev->vcn.harvest_config & (1 << i))
 			continue;
@@ -211,7 +211,7 @@ static int vcn_v5_0_0_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-			volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+			struct amdgpu_vcn5_fw_shared *fw_shared;
 
 			if (adev->vcn.harvest_config & (1 << i))
 				continue;
@@ -692,7 +692,7 @@ static int vcn_v5_0_0_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
+	struct amdgpu_vcn5_fw_shared *fw_shared = adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int ret;
@@ -802,7 +802,7 @@ static int vcn_v5_0_0_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+	struct amdgpu_vcn5_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int j, k, r;
@@ -995,7 +995,7 @@ static int vcn_v5_0_0_stop(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+	struct amdgpu_vcn5_fw_shared *fw_shared;
 	uint32_t tmp;
 	int r = 0;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 09e2e28041c0..9c281ba6bced 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -226,7 +226,7 @@ static int vcn_v5_0_1_sw_fini(struct amdgpu_ip_block *ip_block)
 
 	if (drm_dev_enter(adev_to_drm(adev), &idx)) {
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++) {
-			volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+			struct amdgpu_vcn5_fw_shared *fw_shared;
 
 			fw_shared = adev->vcn.inst[i].fw_shared.cpu_addr;
 			fw_shared->present_flag_0 = 0;
@@ -640,7 +640,7 @@ static int vcn_v5_0_1_start_dpg_mode(struct amdgpu_vcn_inst *vinst,
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int inst_idx = vinst->inst;
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared =
+	struct amdgpu_vcn5_fw_shared *fw_shared =
 		adev->vcn.inst[inst_idx].fw_shared.cpu_addr;
 	struct amdgpu_ring *ring;
 	struct dpg_pause_state state = {.fw_based = VCN_DPG_STATE__PAUSE};
@@ -776,8 +776,8 @@ static int vcn_v5_0_1_start_sriov(struct amdgpu_device *adev)
 	struct mmsch_v5_0_cmd_end end = { {0} };
 	struct mmsch_v5_0_init_header header;
 
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared;
-	volatile struct amdgpu_fw_shared_rb_setup *rb_setup;
+	struct amdgpu_vcn5_fw_shared *fw_shared;
+	struct amdgpu_fw_shared_rb_setup *rb_setup;
 
 	direct_wt.cmd_header.command_type =
 		MMSCH_COMMAND__DIRECT_REG_WRITE;
@@ -951,7 +951,7 @@ static int vcn_v5_0_1_start(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+	struct amdgpu_vcn5_fw_shared *fw_shared;
 	struct amdgpu_ring *ring;
 	uint32_t tmp;
 	int j, k, r, vcn_inst;
@@ -1143,7 +1143,7 @@ static int vcn_v5_0_1_stop(struct amdgpu_vcn_inst *vinst)
 {
 	struct amdgpu_device *adev = vinst->adev;
 	int i = vinst->inst;
-	volatile struct amdgpu_vcn5_fw_shared *fw_shared;
+	struct amdgpu_vcn5_fw_shared *fw_shared;
 	uint32_t tmp;
 	int r = 0, vcn_inst;
 

From c4ac100e9ae252b09986766ad23b1f83ca3a369d Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Thu, 12 Jun 2025 11:44:26 +0100
Subject: [PATCH 1268/1455] drm/amdgpu: Use vmemdup_array_user in
 amdgpu_bo_create_list_entry_array

Replace kvmalloc_array() + copy_from_user() with vmemdup_array_user() on
the fast path.

This shrinks the source code and improves separation between the kernel
and userspace slabs.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 41 +++++++++------------
 1 file changed, 17 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 702f6610d024..66fb37b64388 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -184,43 +184,36 @@ void amdgpu_bo_list_put(struct amdgpu_bo_list *list)
 int amdgpu_bo_create_list_entry_array(struct drm_amdgpu_bo_list_in *in,
 				      struct drm_amdgpu_bo_list_entry **info_param)
 {
-	const void __user *uptr = u64_to_user_ptr(in->bo_info_ptr);
 	const uint32_t info_size = sizeof(struct drm_amdgpu_bo_list_entry);
+	const void __user *uptr = u64_to_user_ptr(in->bo_info_ptr);
+	const uint32_t bo_info_size = in->bo_info_size;
+	const uint32_t bo_number = in->bo_number;
 	struct drm_amdgpu_bo_list_entry *info;
-	int r;
-
-	info = kvmalloc_array(in->bo_number, info_size, GFP_KERNEL);
-	if (!info)
-		return -ENOMEM;
 
 	/* copy the handle array from userspace to a kernel buffer */
-	r = -EFAULT;
-	if (likely(info_size == in->bo_info_size)) {
-		unsigned long bytes = in->bo_number *
-			in->bo_info_size;
-
-		if (copy_from_user(info, uptr, bytes))
-			goto error_free;
-
+	if (likely(info_size == bo_info_size)) {
+		info = vmemdup_array_user(uptr, bo_number, info_size);
+		if (IS_ERR(info))
+			return PTR_ERR(info);
 	} else {
-		unsigned long bytes = min(in->bo_info_size, info_size);
+		const uint32_t bytes = min(bo_info_size, info_size);
 		unsigned i;
 
-		memset(info, 0, in->bo_number * info_size);
-		for (i = 0; i < in->bo_number; ++i) {
-			if (copy_from_user(&info[i], uptr, bytes))
-				goto error_free;
+		info = kvmalloc_array(bo_number, info_size, GFP_KERNEL);
+		if (!info)
+			return -ENOMEM;
 
-			uptr += in->bo_info_size;
+		memset(info, 0, bo_number * info_size);
+		for (i = 0; i < bo_number; ++i, uptr += bo_info_size) {
+			if (copy_from_user(&info[i], uptr, bytes)) {
+				kvfree(info);
+				return -EFAULT;
+			}
 		}
 	}
 
 	*info_param = info;
 	return 0;
-
-error_free:
-	kvfree(info);
-	return r;
 }
 
 int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,

From dea75df7afe14d6217576dbc28cc3ec1d1f712fb Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Thu, 12 Jun 2025 11:44:27 +0100
Subject: [PATCH 1269/1455] drm/amdgpu: Use memdup_array_user in
 amdgpu_cs_wait_fences_ioctl

Replace kmalloc_array() + copy_from_user() with memdup_array_user().

This shrinks the source code and improves separation between the kernel
and userspace slabs.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 19 +++++--------------
 1 file changed, 5 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 2ac9729e4c86..0e41589a7f18 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1767,30 +1767,21 @@ int amdgpu_cs_wait_fences_ioctl(struct drm_device *dev, void *data,
 {
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	union drm_amdgpu_wait_fences *wait = data;
-	uint32_t fence_count = wait->in.fence_count;
-	struct drm_amdgpu_fence *fences_user;
 	struct drm_amdgpu_fence *fences;
 	int r;
 
 	/* Get the fences from userspace */
-	fences = kmalloc_array(fence_count, sizeof(struct drm_amdgpu_fence),
-			GFP_KERNEL);
-	if (fences == NULL)
-		return -ENOMEM;
-
-	fences_user = u64_to_user_ptr(wait->in.fences);
-	if (copy_from_user(fences, fences_user,
-		sizeof(struct drm_amdgpu_fence) * fence_count)) {
-		r = -EFAULT;
-		goto err_free_fences;
-	}
+	fences = memdup_array_user(u64_to_user_ptr(wait->in.fences),
+				   wait->in.fence_count,
+				   sizeof(struct drm_amdgpu_fence));
+	if (IS_ERR(fences))
+		return PTR_ERR(fences);
 
 	if (wait->in.wait_all)
 		r = amdgpu_cs_wait_all_fences(adev, filp, wait, fences);
 	else
 		r = amdgpu_cs_wait_any_fence(adev, filp, wait, fences);
 
-err_free_fences:
 	kfree(fences);
 
 	return r;

From d4b6274cbf0b0ee92294d699f138b63aa4796ad0 Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Thu, 12 Jun 2025 11:44:28 +0100
Subject: [PATCH 1270/1455] drm/amdgpu: Use (v)memdup_array_user in
 amdgpu_cs_pass1

Replace k(v)malloc_array() + copy_from_user() with (v)memdup_array_user().

This shrinks the source code and improves separation between the kernel
and userspace slabs.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 33 ++++++++------------------
 1 file changed, 10 insertions(+), 23 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 0e41589a7f18..2e93d570153c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -178,25 +178,17 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 	struct amdgpu_fpriv *fpriv = p->filp->driver_priv;
 	unsigned int num_ibs[AMDGPU_CS_GANG_SIZE] = { };
 	struct amdgpu_vm *vm = &fpriv->vm;
-	uint64_t *chunk_array_user;
 	uint64_t *chunk_array;
 	uint32_t uf_offset = 0;
 	size_t size;
 	int ret;
 	int i;
 
-	chunk_array = kvmalloc_array(cs->in.num_chunks, sizeof(uint64_t),
-				     GFP_KERNEL);
-	if (!chunk_array)
-		return -ENOMEM;
-
-	/* get chunks */
-	chunk_array_user = u64_to_user_ptr(cs->in.chunks);
-	if (copy_from_user(chunk_array, chunk_array_user,
-			   sizeof(uint64_t)*cs->in.num_chunks)) {
-		ret = -EFAULT;
-		goto free_chunk;
-	}
+	chunk_array = memdup_array_user(u64_to_user_ptr(cs->in.chunks),
+					cs->in.num_chunks,
+					sizeof(uint64_t));
+	if (IS_ERR(chunk_array))
+		return PTR_ERR(chunk_array);
 
 	p->nchunks = cs->in.num_chunks;
 	p->chunks = kvmalloc_array(p->nchunks, sizeof(struct amdgpu_cs_chunk),
@@ -209,7 +201,6 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 	for (i = 0; i < p->nchunks; i++) {
 		struct drm_amdgpu_cs_chunk __user *chunk_ptr = NULL;
 		struct drm_amdgpu_cs_chunk user_chunk;
-		uint32_t __user *cdata;
 
 		chunk_ptr = u64_to_user_ptr(chunk_array[i]);
 		if (copy_from_user(&user_chunk, chunk_ptr,
@@ -222,20 +213,16 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 		p->chunks[i].length_dw = user_chunk.length_dw;
 
 		size = p->chunks[i].length_dw;
-		cdata = u64_to_user_ptr(user_chunk.chunk_data);
 
-		p->chunks[i].kdata = kvmalloc_array(size, sizeof(uint32_t),
-						    GFP_KERNEL);
-		if (p->chunks[i].kdata == NULL) {
-			ret = -ENOMEM;
+		p->chunks[i].kdata = vmemdup_array_user(u64_to_user_ptr(user_chunk.chunk_data),
+							size,
+							sizeof(uint32_t));
+		if (IS_ERR(p->chunks[i].kdata)) {
+			ret = PTR_ERR(p->chunks[i].kdata);
 			i--;
 			goto free_partial_kdata;
 		}
 		size *= sizeof(uint32_t);
-		if (copy_from_user(p->chunks[i].kdata, cdata, size)) {
-			ret = -EFAULT;
-			goto free_partial_kdata;
-		}
 
 		/* Assume the worst on the following checks */
 		ret = -EINVAL;

From 6156c101e5f082ee8911fbf0e112319d67403794 Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@linux.dev>
Date: Mon, 8 Sep 2025 23:31:56 +0200
Subject: [PATCH 1271/1455] drm/amdkfd: Replace kzalloc + copy_from_user with
 memdup_user

Replace kzalloc() followed by copy_from_user() with memdup_user() to
improve and simplify kfd_ioctl_set_cu_mask().

Return early if an error occurs and remove the obsolete 'out' label.

No functional changes intended.

Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 8535a52a62ca..0f0719528bcc 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -521,15 +521,10 @@ static int kfd_ioctl_set_cu_mask(struct file *filp, struct kfd_process *p,
 		cu_mask_size = sizeof(uint32_t) * (max_num_cus/32);
 	}
 
-	minfo.cu_mask.ptr = kzalloc(cu_mask_size, GFP_KERNEL);
-	if (!minfo.cu_mask.ptr)
-		return -ENOMEM;
-
-	retval = copy_from_user(minfo.cu_mask.ptr, cu_mask_ptr, cu_mask_size);
-	if (retval) {
+	minfo.cu_mask.ptr = memdup_user(cu_mask_ptr, cu_mask_size);
+	if (IS_ERR(minfo.cu_mask.ptr)) {
 		pr_debug("Could not copy CU mask from userspace");
-		retval = -EFAULT;
-		goto out;
+		return PTR_ERR(minfo.cu_mask.ptr);
 	}
 
 	mutex_lock(&p->mutex);
@@ -538,7 +533,6 @@ static int kfd_ioctl_set_cu_mask(struct file *filp, struct kfd_process *p,
 
 	mutex_unlock(&p->mutex);
 
-out:
 	kfree(minfo.cu_mask.ptr);
 	return retval;
 }

From 99eeb8358e6cdb7050bf2956370c15dcceda8c7e Mon Sep 17 00:00:00 2001
From: Thorsten Blum <thorsten.blum@linux.dev>
Date: Mon, 8 Sep 2025 23:15:54 +0200
Subject: [PATCH 1272/1455] drm/amdgpu: Replace kzalloc + copy_from_user with
 memdup_user

Replace kzalloc() followed by copy_from_user() with memdup_user() to
improve and simplify ta_if_load_debugfs_write() and
ta_if_invoke_debugfs_write().

No functional changes intended.

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Thorsten Blum <thorsten.blum@linux.dev>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c | 20 ++++++--------------
 1 file changed, 6 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c
index 38face981c3e..6e8aad91bcd3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp_ta.c
@@ -171,13 +171,9 @@ static ssize_t ta_if_load_debugfs_write(struct file *fp, const char *buf, size_t
 
 	copy_pos += sizeof(uint32_t);
 
-	ta_bin = kzalloc(ta_bin_len, GFP_KERNEL);
-	if (!ta_bin)
-		return -ENOMEM;
-	if (copy_from_user((void *)ta_bin, &buf[copy_pos], ta_bin_len)) {
-		ret = -EFAULT;
-		goto err_free_bin;
-	}
+	ta_bin = memdup_user(&buf[copy_pos], ta_bin_len);
+	if (IS_ERR(ta_bin))
+		return PTR_ERR(ta_bin);
 
 	/* Set TA context and functions */
 	set_ta_context_funcs(psp, ta_type, &context);
@@ -327,13 +323,9 @@ static ssize_t ta_if_invoke_debugfs_write(struct file *fp, const char *buf, size
 		return -EFAULT;
 	copy_pos += sizeof(uint32_t);
 
-	shared_buf = kzalloc(shared_buf_len, GFP_KERNEL);
-	if (!shared_buf)
-		return -ENOMEM;
-	if (copy_from_user((void *)shared_buf, &buf[copy_pos], shared_buf_len)) {
-		ret = -EFAULT;
-		goto err_free_shared_buf;
-	}
+	shared_buf = memdup_user(&buf[copy_pos], shared_buf_len);
+	if (IS_ERR(shared_buf))
+		return PTR_ERR(shared_buf);
 
 	set_ta_context_funcs(psp, ta_type, &context);
 

From fb5a52dbe9fecde432b4da7ea2d2724a947e9a79 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 20 Aug 2025 16:21:16 +0530
Subject: [PATCH 1273/1455] drm/amdgpu: Implement TTM handling for MMIO_REMAP
 placement
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Implement TTM-level behavior for AMDGPU_PL_MMIO_REMAP so it behaves as a
CPU-visible IO page:

* amdgpu_evict_flags(): mark as unmovable
* amdgpu_res_cpu_visible(): consider CPU-visible
* amdgpu_bo_move(): use null move when src/dst is MMIO_REMAP
* amdgpu_ttm_io_mem_reserve(): program base/is_iomem/caching using
  the device's mmio_remap_* metadata
* amdgpu_ttm_io_mem_pfn(): return PFN for the remapped HDP page
* amdgpu_ttm_tt_pde_flags(): set AMDGPU_PTE_SYSTEM for this mem type

v2:
- Drop HDP-specific comment; keep generic remap (Alex).

v3:
- Fix indentation in amdgpu_res_cpu_visible (Christian).
- Use adev->rmmio_remap.bus_addr for MMIO_REMAP bus/PFN calculations
  (Alex).

v4:
- Drop unnecessary (resource_size_t) casts in MMIO_REMAP io-mem paths
  (Alex)

Cc: Christian König <christian.koenig@amd.com>
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 27ab4e754b2a..42f97d568059 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -123,6 +123,7 @@ static void amdgpu_evict_flags(struct ttm_buffer_object *bo,
 	case AMDGPU_PL_GWS:
 	case AMDGPU_PL_OA:
 	case AMDGPU_PL_DOORBELL:
+	case AMDGPU_PL_MMIO_REMAP:
 		placement->num_placement = 0;
 		return;
 
@@ -447,7 +448,8 @@ bool amdgpu_res_cpu_visible(struct amdgpu_device *adev,
 		return false;
 
 	if (res->mem_type == TTM_PL_SYSTEM || res->mem_type == TTM_PL_TT ||
-	    res->mem_type == AMDGPU_PL_PREEMPT || res->mem_type == AMDGPU_PL_DOORBELL)
+	    res->mem_type == AMDGPU_PL_PREEMPT || res->mem_type == AMDGPU_PL_DOORBELL ||
+	    res->mem_type == AMDGPU_PL_MMIO_REMAP)
 		return true;
 
 	if (res->mem_type != TTM_PL_VRAM)
@@ -538,10 +540,12 @@ static int amdgpu_bo_move(struct ttm_buffer_object *bo, bool evict,
 	    old_mem->mem_type == AMDGPU_PL_GWS ||
 	    old_mem->mem_type == AMDGPU_PL_OA ||
 	    old_mem->mem_type == AMDGPU_PL_DOORBELL ||
+	    old_mem->mem_type == AMDGPU_PL_MMIO_REMAP ||
 	    new_mem->mem_type == AMDGPU_PL_GDS ||
 	    new_mem->mem_type == AMDGPU_PL_GWS ||
 	    new_mem->mem_type == AMDGPU_PL_OA ||
-	    new_mem->mem_type == AMDGPU_PL_DOORBELL) {
+	    new_mem->mem_type == AMDGPU_PL_DOORBELL ||
+	    new_mem->mem_type == AMDGPU_PL_MMIO_REMAP) {
 		/* Nothing to save here */
 		amdgpu_bo_move_notify(bo, evict, new_mem);
 		ttm_bo_move_null(bo, new_mem);
@@ -629,6 +633,12 @@ static int amdgpu_ttm_io_mem_reserve(struct ttm_device *bdev,
 		mem->bus.is_iomem = true;
 		mem->bus.caching = ttm_uncached;
 		break;
+	case AMDGPU_PL_MMIO_REMAP:
+		mem->bus.offset = mem->start << PAGE_SHIFT;
+		mem->bus.offset += adev->rmmio_remap.bus_addr;
+		mem->bus.is_iomem = true;
+		mem->bus.caching = ttm_uncached;
+		break;
 	default:
 		return -EINVAL;
 	}
@@ -646,6 +656,8 @@ static unsigned long amdgpu_ttm_io_mem_pfn(struct ttm_buffer_object *bo,
 
 	if (bo->resource->mem_type == AMDGPU_PL_DOORBELL)
 		return ((uint64_t)(adev->doorbell.base + cursor.start)) >> PAGE_SHIFT;
+	else if (bo->resource->mem_type == AMDGPU_PL_MMIO_REMAP)
+		return ((uint64_t)(adev->rmmio_remap.bus_addr + cursor.start)) >> PAGE_SHIFT;
 
 	return (adev->gmc.aper_base + cursor.start) >> PAGE_SHIFT;
 }
@@ -1355,7 +1367,8 @@ uint64_t amdgpu_ttm_tt_pde_flags(struct ttm_tt *ttm, struct ttm_resource *mem)
 
 	if (mem && (mem->mem_type == TTM_PL_TT ||
 		    mem->mem_type == AMDGPU_PL_DOORBELL ||
-		    mem->mem_type == AMDGPU_PL_PREEMPT)) {
+		    mem->mem_type == AMDGPU_PL_PREEMPT ||
+		    mem->mem_type == AMDGPU_PL_MMIO_REMAP)) {
 		flags |= AMDGPU_PTE_SYSTEM;
 
 		if (ttm->caching == ttm_cached)

From a3ac30ffea5de2f511696ae46f5a7f793e6d4bfb Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 20 Aug 2025 16:39:38 +0530
Subject: [PATCH 1274/1455] drm/amdgpu/ttm: Initialize AMDGPU_PL_MMIO_REMAP
 Heap
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a one-page TTM range manager for AMDGPU_PL_MMIO_REMAP via
amdgpu_ttm_init_on_chip(). This only registers the placement with TTM;
no BO is allocated in this patch.

The singleton 4K remap BO is created and freed in the following patch.

This split follows to separate heap bring-up from BO allocation.

Cc: Christian König <christian.koenig@amd.com>
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 42f97d568059..1a68ba17a62d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -2021,6 +2021,13 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		return r;
 	}
 
+	/* Initialize MMIO-remap pool (single page 4K) */
+	r = amdgpu_ttm_init_on_chip(adev, AMDGPU_PL_MMIO_REMAP, 1);
+	if (r) {
+		dev_err(adev->dev, "Failed initializing MMIO-remap heap.\n");
+		return r;
+	}
+
 	/* Initialize preemptible memory pool */
 	r = amdgpu_preempt_mgr_init(adev);
 	if (r) {
@@ -2083,6 +2090,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	}
 	amdgpu_bo_free_kernel(&adev->mman.sdma_access_bo, NULL,
 					&adev->mman.sdma_access_ptr);
+
 	amdgpu_ttm_fw_reserve_vram_fini(adev);
 	amdgpu_ttm_drv_reserve_vram_fini(adev);
 
@@ -2104,6 +2112,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_GWS);
 	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_OA);
 	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_DOORBELL);
+	ttm_range_man_fini(&adev->mman.bdev, AMDGPU_PL_MMIO_REMAP);
 	ttm_device_fini(&adev->mman.bdev);
 	adev->mman.initialized = false;
 	dev_info(adev->dev, "amdgpu: ttm finalized\n");

From 4ba48fc3dac4c1dd4b54269afc40e43297f9e073 Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Tue, 9 Sep 2025 10:10:40 +0800
Subject: [PATCH 1275/1455] drm/amd/pm: make smu_set_temp_funcs() smu specific
 for smu v13.0.6

move smu_set_temp_funcs() into smu_v13.0.6 ppt.c file to keep same code
layer in amdgpu_smu.c. (only set_ppt func in amdgpu_smu.c)

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c            |  1 -
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 12 +++++++-----
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h |  1 -
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index c5965924e7c6..f6022c2ae361 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -766,7 +766,6 @@ static int smu_set_funcs(struct amdgpu_device *adev)
 	case IP_VERSION(13, 0, 14):
 	case IP_VERSION(13, 0, 12):
 		smu_v13_0_6_set_ppt_funcs(smu);
-		smu_v13_0_6_set_temp_funcs(smu);
 		/* Enable pp_od_clk_voltage node */
 		smu->od_enabled = true;
 		break;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index ebee659f8a1c..7b087251f891 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3839,6 +3839,12 @@ static const struct aca_smu_funcs smu_v13_0_6_aca_smu_funcs = {
 	.parse_error_code = aca_smu_parse_error_code,
 };
 
+static void smu_v13_0_6_set_temp_funcs(struct smu_context *smu)
+{
+	smu->smu_temp.temp_funcs = (amdgpu_ip_version(smu->adev, MP1_HWIP, 0)
+			== IP_VERSION(13, 0, 12)) ? &smu_v13_0_12_temp_funcs : NULL;
+}
+
 static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	/* init dpm */
 	.get_allowed_feature_mask = smu_v13_0_6_get_allowed_feature_mask,
@@ -3913,12 +3919,8 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
 	smu->smc_driver_if_version = SMU13_0_6_DRIVER_IF_VERSION;
 	smu->smc_fw_caps |= SMU_FW_CAP_RAS_PRI;
 	smu_v13_0_set_smu_mailbox_registers(smu);
+	smu_v13_0_6_set_temp_funcs(smu);
 	amdgpu_mca_smu_init_funcs(smu->adev, &smu_v13_0_6_mca_smu_funcs);
 	amdgpu_aca_set_smu_funcs(smu->adev, &smu_v13_0_6_aca_smu_funcs);
 }
 
-void smu_v13_0_6_set_temp_funcs(struct smu_context *smu)
-{
-	smu->smu_temp.temp_funcs = (amdgpu_ip_version(smu->adev, MP1_HWIP, 0)
-			== IP_VERSION(13, 0, 12)) ? &smu_v13_0_12_temp_funcs : NULL;
-}
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index aae9a546a67e..4652fcd5e068 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -74,7 +74,6 @@ enum smu_v13_0_6_caps {
 };
 
 extern void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu);
-extern void smu_v13_0_6_set_temp_funcs(struct smu_context *smu);
 bool smu_v13_0_6_cap_supported(struct smu_context *smu, enum smu_v13_0_6_caps cap);
 int smu_v13_0_6_get_static_metrics_table(struct smu_context *smu);
 int smu_v13_0_6_get_metrics_table(struct smu_context *smu, void *metrics_table,

From 9e46b8bb0539d7bc9a9e7b3072fa4f6082490392 Mon Sep 17 00:00:00 2001
From: Prike Liang <Prike.Liang@amd.com>
Date: Mon, 23 Jun 2025 16:29:38 +0800
Subject: [PATCH 1276/1455] drm/amdgpu: validate userq buffer virtual address
 and size

It needs to validate the userq object virtual address to
determine whether it is residented in a valid vm mapping.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c  | 40 ++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h  |  2 ++
 drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 16 +++++++++
 3 files changed, 58 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 09cea573e5ad..c3c1a714b06a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -44,6 +44,38 @@ u32 amdgpu_userq_get_supported_ip_mask(struct amdgpu_device *adev)
 	return userq_ip_mask;
 }
 
+int amdgpu_userq_input_va_validate(struct amdgpu_vm *vm, u64 addr,
+				   u64 expected_size)
+{
+	struct amdgpu_bo_va_mapping *va_map;
+	u64 user_addr;
+	u64 size;
+	int r = 0;
+
+	user_addr = (addr & AMDGPU_GMC_HOLE_MASK) >> AMDGPU_GPU_PAGE_SHIFT;
+	size = expected_size >> AMDGPU_GPU_PAGE_SHIFT;
+
+	r = amdgpu_bo_reserve(vm->root.bo, false);
+	if (r)
+		return r;
+
+	va_map = amdgpu_vm_bo_lookup_mapping(vm, user_addr);
+	if (!va_map) {
+		r = -EINVAL;
+		goto out_err;
+	}
+	/* Only validate the userq whether resident in the VM mapping range */
+	if (user_addr >= va_map->start  &&
+	    va_map->last - user_addr + 1 >= size) {
+		amdgpu_bo_unreserve(vm->root.bo);
+		return 0;
+	}
+
+out_err:
+	amdgpu_bo_unreserve(vm->root.bo);
+	return r;
+}
+
 static int
 amdgpu_userq_unmap_helper(struct amdgpu_userq_mgr *uq_mgr,
 			  struct amdgpu_usermode_queue *queue)
@@ -428,6 +460,14 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args)
 		r = -ENOMEM;
 		goto unlock;
 	}
+
+	/* Validate the userq virtual address.*/
+	if (amdgpu_userq_input_va_validate(&fpriv->vm, args->in.queue_va, args->in.queue_size) ||
+	    amdgpu_userq_input_va_validate(&fpriv->vm, args->in.rptr_va, AMDGPU_GPU_PAGE_SIZE) ||
+	    amdgpu_userq_input_va_validate(&fpriv->vm, args->in.wptr_va, AMDGPU_GPU_PAGE_SIZE)) {
+		kfree(queue);
+		goto unlock;
+	}
 	queue->doorbell_handle = args->in.doorbell_handle;
 	queue->queue_type = args->in.ip_type;
 	queue->vm = &fpriv->vm;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
index da7e75cda4f8..c027dd916672 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h
@@ -137,4 +137,6 @@ int amdgpu_userq_stop_sched_for_enforce_isolation(struct amdgpu_device *adev,
 int amdgpu_userq_start_sched_for_enforce_isolation(struct amdgpu_device *adev,
 						   u32 idx);
 
+int amdgpu_userq_input_va_validate(struct amdgpu_vm *vm, u64 addr,
+				   u64 expected_size);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
index 66467f41294c..3a4fd6de08ce 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
@@ -254,6 +254,7 @@ static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 	struct amdgpu_mqd *mqd_hw_default = &adev->mqds[queue->queue_type];
 	struct drm_amdgpu_userq_in *mqd_user = args_in;
 	struct amdgpu_mqd_prop *userq_props;
+	struct amdgpu_gfx_shadow_info shadow_info;
 	int r;
 
 	/* Structure to initialize MQD for userqueue using generic MQD init function */
@@ -279,6 +280,8 @@ static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 	userq_props->doorbell_index = queue->doorbell_index;
 	userq_props->fence_address = queue->fence_drv->gpu_addr;
 
+	if (adev->gfx.funcs->get_gfx_shadow_info)
+		adev->gfx.funcs->get_gfx_shadow_info(adev, &shadow_info, true);
 	if (queue->queue_type == AMDGPU_HW_IP_COMPUTE) {
 		struct drm_amdgpu_userq_mqd_compute_gfx11 *compute_mqd;
 
@@ -295,6 +298,10 @@ static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 			goto free_mqd;
 		}
 
+		if (amdgpu_userq_input_va_validate(queue->vm, compute_mqd->eop_va,
+		    max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE)))
+			goto free_mqd;
+
 		userq_props->eop_gpu_addr = compute_mqd->eop_va;
 		userq_props->hqd_pipe_priority = AMDGPU_GFX_PIPE_PRIO_NORMAL;
 		userq_props->hqd_queue_priority = AMDGPU_GFX_QUEUE_PRIORITY_MINIMUM;
@@ -322,6 +329,11 @@ static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 		userq_props->csa_addr = mqd_gfx_v11->csa_va;
 		userq_props->tmz_queue =
 			mqd_user->flags & AMDGPU_USERQ_CREATE_FLAGS_QUEUE_SECURE;
+
+		if (amdgpu_userq_input_va_validate(queue->vm, mqd_gfx_v11->shadow_va,
+		    shadow_info.shadow_size))
+			goto free_mqd;
+
 		kfree(mqd_gfx_v11);
 	} else if (queue->queue_type == AMDGPU_HW_IP_DMA) {
 		struct drm_amdgpu_userq_mqd_sdma_gfx11 *mqd_sdma_v11;
@@ -339,6 +351,10 @@ static int mes_userq_mqd_create(struct amdgpu_userq_mgr *uq_mgr,
 			goto free_mqd;
 		}
 
+		if (amdgpu_userq_input_va_validate(queue->vm, mqd_sdma_v11->csa_va,
+		    shadow_info.csa_size))
+			goto free_mqd;
+
 		userq_props->csa_addr = mqd_sdma_v11->csa_va;
 		kfree(mqd_sdma_v11);
 	}

From 2a7a794eb82c37014ea95f7f58ebeff73ef72858 Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Thu, 28 Aug 2025 19:18:49 +0530
Subject: [PATCH 1277/1455] drm/amdgpu/ttm: Allocate/Free 4K MMIO_REMAP
 Singleton
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add mmio_remap bookkeeping to amdgpu_device and introduce
amdgpu_ttm_mmio_remap_bo_init()/fini() to manage a kernel-owned,
one-page (4K) BO in AMDGPU_GEM_DOMAIN_MMIO_REMAP.

Bookkeeping:
  - adev->rmmio_remap.bo : kernel-owned singleton BO

The BO is allocated during TTM init when a remap bus address is available
(adev->rmmio_remap.bus_addr) and PAGE_SIZE <= AMDGPU_GPU_PAGE_SIZE (4K),
and freed during TTM fini.

v2:
 - Check mmio_remap bus address (adev->rmmio_remap.bus_addr) instead of
   rmmio_base. (Alex)
 - Skip quietly if PAGE_SIZE > AMDGPU_GPU_PAGE_SIZE or no bus address
   (no warn). (Alex)
 - Use `amdgpu_bo_create()` (not *_kernel) - Only with this The object
   is stored in adev->mmio_remap.bo and will later be exposed to
   userspace via a GEM handle. (Christian)

v3:
 - Remove obvious comment before amdgpu_ttm_mmio_remap_bo_fini() call.
   (Alex)

v4:
 - Squash bookkeeping into this patch (Christian)

Suggested-by: Christian König <christian.koenig@amd.com>
Suggested-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h     |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 59 +++++++++++++++++++++++++
 2 files changed, 60 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 0e6a09651bbc..c56839528843 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -752,6 +752,7 @@ typedef void (*amdgpu_block_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t, u
 struct amdgpu_mmio_remap {
 	u32 reg_offset;
 	resource_size_t bus_addr;
+	struct amdgpu_bo *bo;
 };
 
 /* Define the HW IP blocks will be used in driver , add more if necessary */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 1a68ba17a62d..f38bc9542cd6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1854,6 +1854,59 @@ static void amdgpu_ttm_pools_fini(struct amdgpu_device *adev)
 	adev->mman.ttm_pools = NULL;
 }
 
+/**
+ * amdgpu_ttm_mmio_remap_bo_init - Allocate the singleton 4K MMIO_REMAP BO
+ * @adev: amdgpu device
+ *
+ * Allocates a one-page (4K) GEM BO in AMDGPU_GEM_DOMAIN_MMIO_REMAP when the
+ * hardware exposes a remap base (adev->rmmio_remap.bus_addr) and the host
+ * PAGE_SIZE is <= AMDGPU_GPU_PAGE_SIZE (4K). The BO is created as a regular
+ * GEM object (amdgpu_bo_create).
+ *
+ * Return:
+ *  * 0 on success or intentional skip (feature not present/unsupported)
+ *  * negative errno on allocation failure
+ */
+static int amdgpu_ttm_mmio_remap_bo_init(struct amdgpu_device *adev)
+{
+	struct amdgpu_bo_param bp;
+	int r;
+
+	/* Skip if HW doesn't expose remap, or if PAGE_SIZE > AMDGPU_GPU_PAGE_SIZE (4K). */
+	if (!adev->rmmio_remap.bus_addr || PAGE_SIZE > AMDGPU_GPU_PAGE_SIZE)
+		return 0;
+
+	memset(&bp, 0, sizeof(bp));
+
+	/* Create exactly one GEM BO in the MMIO_REMAP domain. */
+	bp.type        = ttm_bo_type_device;          /* userspace-mappable GEM */
+	bp.size        = AMDGPU_GPU_PAGE_SIZE;        /* 4K */
+	bp.byte_align  = AMDGPU_GPU_PAGE_SIZE;
+	bp.domain      = AMDGPU_GEM_DOMAIN_MMIO_REMAP;
+	bp.flags       = 0;
+	bp.resv        = NULL;
+	bp.bo_ptr_size = sizeof(struct amdgpu_bo);
+
+	r = amdgpu_bo_create(adev, &bp, &adev->rmmio_remap.bo);
+	if (r)
+		return r;
+
+	return 0;
+}
+
+/**
+ * amdgpu_ttm_mmio_remap_bo_fini - Free the singleton MMIO_REMAP BO
+ * @adev: amdgpu device
+ *
+ * Frees the kernel-owned MMIO_REMAP BO if it was allocated by
+ * amdgpu_ttm_mmio_remap_bo_init().
+ */
+static void amdgpu_ttm_mmio_remap_bo_fini(struct amdgpu_device *adev)
+{
+	amdgpu_bo_unref(&adev->rmmio_remap.bo);
+	adev->rmmio_remap.bo = NULL;
+}
+
 /*
  * amdgpu_ttm_init - Init the memory management (ttm) as well as various
  * gtt/vram related fields.
@@ -2028,6 +2081,11 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 		return r;
 	}
 
+	/* Allocate the singleton MMIO_REMAP BO (4K) if supported */
+	r = amdgpu_ttm_mmio_remap_bo_init(adev);
+	if (r)
+		return r;
+
 	/* Initialize preemptible memory pool */
 	r = amdgpu_preempt_mgr_init(adev);
 	if (r) {
@@ -2091,6 +2149,7 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 	amdgpu_bo_free_kernel(&adev->mman.sdma_access_bo, NULL,
 					&adev->mman.sdma_access_ptr);
 
+	amdgpu_ttm_mmio_remap_bo_fini(adev);
 	amdgpu_ttm_fw_reserve_vram_fini(adev);
 	amdgpu_ttm_drv_reserve_vram_fini(adev);
 

From c1456fadce0c99175f97e66c2b982dd051e01aa2 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Tue, 2 Sep 2025 11:37:37 +0530
Subject: [PATCH 1278/1455] drm/amdgpu: Release hive reference properly

xgmi hive reference is taken on function entry, but not released
correctly for all paths. Use __free() to release reference properly.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Ce Sun <cesun102@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 7 +++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h   | 4 ++++
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index b20383021b50..81a48b3a1931 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -6937,7 +6937,8 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 {
 	struct drm_device *dev = pci_get_drvdata(pdev);
 	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
+	struct amdgpu_hive_info *hive __free(xgmi_put_hive) =
+		amdgpu_get_xgmi_hive(adev);
 	struct amdgpu_reset_context reset_context;
 	struct list_head device_list;
 
@@ -6976,10 +6977,8 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
 		amdgpu_device_recovery_get_reset_lock(adev, &device_list);
 		amdgpu_device_halt_activities(adev, NULL, &reset_context, &device_list,
 					      hive, false);
-		if (hive) {
+		if (hive)
 			mutex_unlock(&hive->hive_lock);
-			amdgpu_put_xgmi_hive(hive);
-		}
 		return PCI_ERS_RESULT_NEED_RESET;
 	case pci_channel_io_perm_failure:
 		/* Permanent error, prepare for device removal */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
index bba0b26fee8f..5f36aff17e79 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h
@@ -126,4 +126,8 @@ uint32_t amdgpu_xgmi_get_max_bandwidth(struct amdgpu_device *adev);
 
 void amgpu_xgmi_set_max_speed_width(struct amdgpu_device *adev,
 				    uint16_t max_speed, uint8_t max_width);
+
+/* Cleanup macro for use with __free(xgmi_put_hive) */
+DEFINE_FREE(xgmi_put_hive, struct amdgpu_hive_info *, if (_T) amdgpu_put_xgmi_hive(_T))
+
 #endif

From 92d6295a29dba56148406a8452c69ab49787741b Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Tue, 9 Sep 2025 23:22:54 -0500
Subject: [PATCH 1279/1455] Documentation/amdgpu: Add Ryzen AI 330 series
 processor

This has been announced so add to the table.

Link: https://www.amd.com/en/products/processors/laptop/ryzen/ai-300-series/amd-ryzen-ai-5-330.html
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/apu-asic-info-table.csv | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/gpu/amdgpu/apu-asic-info-table.csv b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
index b479c5629146..dee5f663a47f 100644
--- a/Documentation/gpu/amdgpu/apu-asic-info-table.csv
+++ b/Documentation/gpu/amdgpu/apu-asic-info-table.csv
@@ -13,5 +13,6 @@ Ryzen 7x20 series, Mendocino, 3.1.6, 10.3.7, 3.1.1, 5.2.7, 13.0.8, 13.0.8
 Ryzen 7x40 series, Phoenix, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11, 13.0.4 / 13.0.11
 Ryzen 8x40 series, Hawk Point, 3.1.4, 11.0.1 / 11.0.4, 4.0.2, 6.0.1, 13.0.4 / 13.0.11, 13.0.4 / 13.0.11
 Ryzen AI 300 series, Strix Point, 3.5.0, 11.5.0, 4.0.5, 6.1.0, 14.0.0, 14.0.0
+Ryzen AI 330 series, Krackan Point, 3.6.0, 11.5.3, 4.0.5, 6.1.3, 14.0.5, 14.0.5
 Ryzen AI 350 series, Krackan Point, 3.5.0, 11.5.2, 4.0.5, 6.1.2, 14.0.4, 14.0.4
 Ryzen AI Max 300 series, Strix Halo, 3.5.1, 11.5.1, 4.0.6, 6.1.1, 14.0.1, 14.0.1

From b515dcb0dc4e85d8254f5459cfb32fce88dacbfb Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Tue, 9 Sep 2025 16:17:50 +0200
Subject: [PATCH 1280/1455] drm/amd/display: Add pixel_clock to
 amd_pp_display_configuration
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit adds the pixel_clock field to the display config
struct so that power management (DPM) can use it.

We currently don't have a proper bandwidth calculation on old
GPUs with DCE 6-10 because dce_calcs only supports DCE 11+.
So the power management (DPM) on these GPUs may need to make
ad-hoc decisions for display based on the pixel clock.

Also rename sym_clock to pixel_clock in dm_pp_single_disp_config
to avoid confusion with other code where the sym_clock refers to
the DisplayPort symbol clock.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c       | 1 +
 drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c | 2 +-
 drivers/gpu/drm/amd/display/dc/dm_services_types.h             | 2 +-
 drivers/gpu/drm/amd/include/dm_pp_interface.h                  | 1 +
 4 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
index e5771f490f2e..11b2ea6edf95 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
@@ -98,6 +98,7 @@ bool dm_pp_apply_display_requirements(
 			const struct dm_pp_single_disp_config *dc_cfg =
 						&pp_display_cfg->disp_configs[i];
 			adev->pm.pm_display_cfg.displays[i].controller_id = dc_cfg->pipe_idx + 1;
+			adev->pm.pm_display_cfg.displays[i].pixel_clock = dc_cfg->pixel_clock;
 		}
 
 		amdgpu_dpm_display_configuration_change(adev, &adev->pm.pm_display_cfg);
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
index 13cf415e38e5..d50b9440210e 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce110/dce110_clk_mgr.c
@@ -164,7 +164,7 @@ void dce110_fill_display_configs(
 			stream->link->cur_link_settings.link_rate;
 		cfg->link_settings.link_spread =
 			stream->link->cur_link_settings.link_spread;
-		cfg->sym_clock = stream->phy_pix_clk;
+		cfg->pixel_clock = stream->phy_pix_clk;
 		/* Round v_refresh*/
 		cfg->v_refresh = stream->timing.pix_clk_100hz * 100;
 		cfg->v_refresh /= stream->timing.h_total;
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
index bf63da266a18..3b093b8699ab 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
@@ -127,7 +127,7 @@ struct dm_pp_single_disp_config {
 	uint32_t src_height;
 	uint32_t src_width;
 	uint32_t v_refresh;
-	uint32_t sym_clock; /* HDMI only */
+	uint32_t pixel_clock; /* Pixel clock in KHz (for HDMI only: normalized) */
 	struct dc_link_settings link_settings; /* DP only */
 };
 
diff --git a/drivers/gpu/drm/amd/include/dm_pp_interface.h b/drivers/gpu/drm/amd/include/dm_pp_interface.h
index acd1cef61b7c..349544504c93 100644
--- a/drivers/gpu/drm/amd/include/dm_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/dm_pp_interface.h
@@ -65,6 +65,7 @@ struct single_display_configuration {
 	uint32_t view_resolution_cy;
 	enum amd_pp_display_config_type displayconfigtype;
 	uint32_t vertical_refresh; /* for active display */
+	uint32_t pixel_clock; /* Pixel clock in KHz (for HDMI only: normalized) */
 };
 
 #define MAX_NUM_DISPLAY 32

From 9d73b107a61b73e7101d4b728ddac3d2c77db111 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Tue, 9 Sep 2025 16:17:51 +0200
Subject: [PATCH 1281/1455] drm/amd/pm: Use pm_display_cfg in legacy DPM (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit is necessary for DC to function well with chips
that use the legacy power management code, ie. SI and KV.
Communicate display information from DC to the legacy PM code.

Currently DC uses pm_display_cfg to communicate power management
requirements from the display code to the DPM code.
However, the legacy (non-DC) code path used different fields
and therefore could not take into account anything from DC.

Change the legacy display code to fill the same pm_display_cfg
struct as DC and use the same in the legacy DPM code.

To ease review and reduce churn, this commit does not yet
delete the now unneeded code, that is done in the next commit.

v2:
Rebase.
Fix single_display in amdgpu_dpm_pick_power_state.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c  | 67 +++++++++++++++++++
 .../gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h  |  2 +
 drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c    |  4 +-
 .../gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c    |  5 +-
 drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c    | 65 ++++++------------
 .../gpu/drm/amd/pm/powerplay/amd_powerplay.c  | 11 +--
 6 files changed, 97 insertions(+), 57 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
index 2d2d2d5e6763..9ef965e4a92e 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
@@ -100,3 +100,70 @@ u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev)
 
 	return vrefresh;
 }
+
+void amdgpu_dpm_get_display_cfg(struct amdgpu_device *adev)
+{
+	struct drm_device *ddev = adev_to_drm(adev);
+	struct amd_pp_display_configuration *cfg = &adev->pm.pm_display_cfg;
+	struct single_display_configuration *display_cfg;
+	struct drm_crtc *crtc;
+	struct amdgpu_crtc *amdgpu_crtc;
+	struct amdgpu_connector *conn;
+	int num_crtcs = 0;
+	int vrefresh;
+	u32 vblank_in_pixels, vblank_time_us;
+
+	cfg->min_vblank_time = 0xffffffff; /* if the displays are off, vblank time is max */
+
+	if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) {
+		list_for_each_entry(crtc, &ddev->mode_config.crtc_list, head) {
+			amdgpu_crtc = to_amdgpu_crtc(crtc);
+
+			/* The array should only contain active displays. */
+			if (!amdgpu_crtc->enabled)
+				continue;
+
+			conn = to_amdgpu_connector(amdgpu_crtc->connector);
+			display_cfg = &adev->pm.pm_display_cfg.displays[num_crtcs++];
+
+			if (amdgpu_crtc->hw_mode.clock) {
+				vrefresh = drm_mode_vrefresh(&amdgpu_crtc->hw_mode);
+
+				vblank_in_pixels =
+					amdgpu_crtc->hw_mode.crtc_htotal *
+					(amdgpu_crtc->hw_mode.crtc_vblank_end -
+					amdgpu_crtc->hw_mode.crtc_vdisplay +
+					(amdgpu_crtc->v_border * 2));
+
+				vblank_time_us =
+					vblank_in_pixels * 1000 / amdgpu_crtc->hw_mode.clock;
+
+				/* The legacy (non-DC) code has issues with mclk switching
+				 * with refresh rates over 120 Hz. Disable mclk switching.
+				 */
+				if (vrefresh > 120)
+					vblank_time_us = 0;
+
+				/* Find minimum vblank time. */
+				if (vblank_time_us < cfg->min_vblank_time)
+					cfg->min_vblank_time = vblank_time_us;
+
+				/* Find vertical refresh rate of first active display. */
+				if (!cfg->vrefresh)
+					cfg->vrefresh = vrefresh;
+			}
+
+			if (amdgpu_crtc->crtc_id < cfg->crtc_index) {
+				/* Find first active CRTC and its line time. */
+				cfg->crtc_index = amdgpu_crtc->crtc_id;
+				cfg->line_time_in_us = amdgpu_crtc->line_time;
+			}
+
+			display_cfg->controller_id = amdgpu_crtc->crtc_id;
+			display_cfg->pixel_clock = conn->pixelclock_for_modeset;
+		}
+	}
+
+	cfg->display_clk = adev->clock.default_dispclk;
+	cfg->num_display = num_crtcs;
+}
diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h
index 5c2a89f0d5d5..8be11510cd92 100644
--- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h
@@ -29,4 +29,6 @@ u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev);
 
 u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev);
 
+void amdgpu_dpm_get_display_cfg(struct amdgpu_device *adev);
+
 #endif
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c
index 307ebf7e3226..33eb85dd68e9 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c
@@ -2299,7 +2299,7 @@ static void kv_apply_state_adjust_rules(struct amdgpu_device *adev,
 
 		if (pi->sys_info.nb_dpm_enable) {
 			force_high = (mclk >= pi->sys_info.nbp_memory_clock[3]) ||
-				pi->video_start || (adev->pm.dpm.new_active_crtc_count >= 3) ||
+				pi->video_start || (adev->pm.pm_display_cfg.num_display >= 3) ||
 				pi->disable_nb_ps3_in_battery;
 			ps->dpm0_pg_nb_ps_lo = force_high ? 0x2 : 0x3;
 			ps->dpm0_pg_nb_ps_hi = 0x2;
@@ -2358,7 +2358,7 @@ static int kv_calculate_nbps_level_settings(struct amdgpu_device *adev)
 			return 0;
 
 		force_high = ((mclk >= pi->sys_info.nbp_memory_clock[3]) ||
-			      (adev->pm.dpm.new_active_crtc_count >= 3) || pi->video_start);
+			      (adev->pm.pm_display_cfg.num_display >= 3) || pi->video_start);
 
 		if (force_high) {
 			for (i = pi->lowest_valid; i <= pi->highest_valid; i++)
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
index 52dbf6d0469d..6ebe3d0f5b87 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
@@ -771,7 +771,7 @@ static struct amdgpu_ps *amdgpu_dpm_pick_power_state(struct amdgpu_device *adev,
 	int i;
 	struct amdgpu_ps *ps;
 	u32 ui_class;
-	bool single_display = adev->pm.dpm.new_active_crtc_count < 2;
+	bool single_display = adev->pm.pm_display_cfg.num_display < 2;
 
 	/* check if the vblank period is too short to adjust the mclk */
 	if (single_display && adev->powerplay.pp_funcs->vblank_too_short) {
@@ -967,7 +967,8 @@ void amdgpu_legacy_dpm_compute_clocks(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	amdgpu_dpm_get_active_displays(adev);
+	if (!adev->dc_enabled)
+		amdgpu_dpm_get_display_cfg(adev);
 
 	amdgpu_dpm_change_power_state_locked(adev);
 }
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
index 6595a611ce6e..cf9932e68055 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c
@@ -3081,7 +3081,7 @@ static int si_get_vce_clock_voltage(struct amdgpu_device *adev,
 static bool si_dpm_vblank_too_short(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
-	u32 vblank_time = amdgpu_dpm_get_vblank_time(adev);
+	u32 vblank_time = adev->pm.pm_display_cfg.min_vblank_time;
 	/* we never hit the non-gddr5 limit so disable it */
 	u32 switch_limit = adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5 ? 450 : 0;
 
@@ -3447,9 +3447,10 @@ static void rv770_get_engine_memory_ss(struct amdgpu_device *adev)
 static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 					struct amdgpu_ps *rps)
 {
+	const struct amd_pp_display_configuration *display_cfg =
+		&adev->pm.pm_display_cfg;
 	struct  si_ps *ps = si_get_ps(rps);
 	struct amdgpu_clock_and_voltage_limits *max_limits;
-	struct amdgpu_connector *conn;
 	bool disable_mclk_switching = false;
 	bool disable_sclk_switching = false;
 	u32 mclk, sclk;
@@ -3488,14 +3489,9 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 	 * For example, 4K 60Hz and 1080p 144Hz fall into this category.
 	 * Find number of such displays connected.
 	 */
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		if (!(adev->pm.dpm.new_active_crtcs & (1 << i)) ||
-			!adev->mode_info.crtcs[i]->enabled)
-			continue;
-
-		conn = to_amdgpu_connector(adev->mode_info.crtcs[i]->connector);
-
-		if (conn->pixelclock_for_modeset > 297000)
+	for (i = 0; i < display_cfg->num_display; i++) {
+		/* The array only contains active displays. */
+		if (display_cfg->displays[i].pixel_clock > 297000)
 			high_pixelclock_count++;
 	}
 
@@ -3523,7 +3519,7 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 		rps->ecclk = 0;
 	}
 
-	if ((adev->pm.dpm.new_active_crtc_count > 1) ||
+	if ((adev->pm.pm_display_cfg.num_display > 1) ||
 	    si_dpm_vblank_too_short(adev))
 		disable_mclk_switching = true;
 
@@ -3671,7 +3667,7 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev,
 						   ps->performance_levels[i].mclk,
 						   max_limits->vddc,  &ps->performance_levels[i].vddc);
 		btc_apply_voltage_dependency_rules(&adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk,
-						   adev->clock.current_dispclk,
+						   display_cfg->display_clk,
 						   max_limits->vddc,  &ps->performance_levels[i].vddc);
 	}
 
@@ -4196,16 +4192,16 @@ static void si_program_ds_registers(struct amdgpu_device *adev)
 
 static void si_program_display_gap(struct amdgpu_device *adev)
 {
+	const struct amd_pp_display_configuration *cfg = &adev->pm.pm_display_cfg;
 	u32 tmp, pipe;
-	int i;
 
 	tmp = RREG32(mmCG_DISPLAY_GAP_CNTL) & ~(CG_DISPLAY_GAP_CNTL__DISP1_GAP_MASK | CG_DISPLAY_GAP_CNTL__DISP2_GAP_MASK);
-	if (adev->pm.dpm.new_active_crtc_count > 0)
+	if (cfg->num_display > 0)
 		tmp |= R600_PM_DISPLAY_GAP_VBLANK_OR_WM << CG_DISPLAY_GAP_CNTL__DISP1_GAP__SHIFT;
 	else
 		tmp |= R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP1_GAP__SHIFT;
 
-	if (adev->pm.dpm.new_active_crtc_count > 1)
+	if (cfg->num_display > 1)
 		tmp |= R600_PM_DISPLAY_GAP_VBLANK_OR_WM << CG_DISPLAY_GAP_CNTL__DISP2_GAP__SHIFT;
 	else
 		tmp |= R600_PM_DISPLAY_GAP_IGNORE << CG_DISPLAY_GAP_CNTL__DISP2_GAP__SHIFT;
@@ -4215,17 +4211,8 @@ static void si_program_display_gap(struct amdgpu_device *adev)
 	tmp = RREG32(DCCG_DISP_SLOW_SELECT_REG);
 	pipe = (tmp & DCCG_DISP1_SLOW_SELECT_MASK) >> DCCG_DISP1_SLOW_SELECT_SHIFT;
 
-	if ((adev->pm.dpm.new_active_crtc_count > 0) &&
-	    (!(adev->pm.dpm.new_active_crtcs & (1 << pipe)))) {
-		/* find the first active crtc */
-		for (i = 0; i < adev->mode_info.num_crtc; i++) {
-			if (adev->pm.dpm.new_active_crtcs & (1 << i))
-				break;
-		}
-		if (i == adev->mode_info.num_crtc)
-			pipe = 0;
-		else
-			pipe = i;
+	if (cfg->num_display > 0 && pipe != cfg->crtc_index) {
+		pipe = cfg->crtc_index;
 
 		tmp &= ~DCCG_DISP1_SLOW_SELECT_MASK;
 		tmp |= DCCG_DISP1_SLOW_SELECT(pipe);
@@ -4236,7 +4223,7 @@ static void si_program_display_gap(struct amdgpu_device *adev)
 	 * This can be a problem on PowerXpress systems or if you want to use the card
 	 * for offscreen rendering or compute if there are no crtcs enabled.
 	 */
-	si_notify_smc_display_change(adev, adev->pm.dpm.new_active_crtc_count > 0);
+	si_notify_smc_display_change(adev, cfg->num_display > 0);
 }
 
 static void si_enable_spread_spectrum(struct amdgpu_device *adev, bool enable)
@@ -5545,7 +5532,7 @@ static int si_convert_power_level_to_smc(struct amdgpu_device *adev,
 	    (pl->mclk <= pi->mclk_stutter_mode_threshold) &&
 	    !eg_pi->uvd_enabled &&
 	    (RREG32(mmDPG_PIPE_STUTTER_CONTROL) & DPG_PIPE_STUTTER_CONTROL__STUTTER_ENABLE_MASK) &&
-	    (adev->pm.dpm.new_active_crtc_count <= 2)) {
+	    (adev->pm.pm_display_cfg.num_display <= 2)) {
 		level->mcFlags |= SISLANDS_SMC_MC_STUTTER_EN;
 	}
 
@@ -5694,7 +5681,7 @@ static bool si_is_state_ulv_compatible(struct amdgpu_device *adev,
 	/* XXX validate against display requirements! */
 
 	for (i = 0; i < adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.count; i++) {
-		if (adev->clock.current_dispclk <=
+		if (adev->pm.pm_display_cfg.display_clk <=
 		    adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[i].clk) {
 			if (ulv->pl.vddc <
 			    adev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries[i].v)
@@ -5848,30 +5835,22 @@ static int si_upload_ulv_state(struct amdgpu_device *adev)
 
 static int si_upload_smc_data(struct amdgpu_device *adev)
 {
-	struct amdgpu_crtc *amdgpu_crtc = NULL;
-	int i;
+	const struct amd_pp_display_configuration *cfg = &adev->pm.pm_display_cfg;
 	u32 crtc_index = 0;
 	u32 mclk_change_block_cp_min = 0;
 	u32 mclk_change_block_cp_max = 0;
 
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		if (adev->pm.dpm.new_active_crtcs & (1 << i)) {
-			amdgpu_crtc = adev->mode_info.crtcs[i];
-			break;
-		}
-	}
-
 	/* When a display is plugged in, program these so that the SMC
 	 * performs MCLK switching when it doesn't cause flickering.
 	 * When no display is plugged in, there is no need to restrict
 	 * MCLK switching, so program them to zero.
 	 */
-	if (adev->pm.dpm.new_active_crtc_count && amdgpu_crtc) {
-		crtc_index = amdgpu_crtc->crtc_id;
+	if (cfg->num_display) {
+		crtc_index = cfg->crtc_index;
 
-		if (amdgpu_crtc->line_time) {
-			mclk_change_block_cp_min = 200 / amdgpu_crtc->line_time;
-			mclk_change_block_cp_max = 100 / amdgpu_crtc->line_time;
+		if (cfg->line_time_in_us) {
+			mclk_change_block_cp_min = 200 / cfg->line_time_in_us;
+			mclk_change_block_cp_max = 100 / cfg->line_time_in_us;
 		}
 	}
 
diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
index b48a031cbba0..554492dfa3c0 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
@@ -1554,16 +1554,7 @@ static void pp_pm_compute_clocks(void *handle)
 	struct amdgpu_device *adev = hwmgr->adev;
 
 	if (!adev->dc_enabled) {
-		amdgpu_dpm_get_active_displays(adev);
-		adev->pm.pm_display_cfg.num_display = adev->pm.dpm.new_active_crtc_count;
-		adev->pm.pm_display_cfg.vrefresh = amdgpu_dpm_get_vrefresh(adev);
-		adev->pm.pm_display_cfg.min_vblank_time = amdgpu_dpm_get_vblank_time(adev);
-		/* we have issues with mclk switching with
-		 * refresh rates over 120 hz on the non-DC code.
-		 */
-		if (adev->pm.pm_display_cfg.vrefresh > 120)
-			adev->pm.pm_display_cfg.min_vblank_time = 0;
-
+		amdgpu_dpm_get_display_cfg(adev);
 		pp_display_configuration_change(handle,
 						&adev->pm.pm_display_cfg);
 	}

From 156db7d353286b7bfd7f43f548a4046a97fe67c6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Tue, 9 Sep 2025 16:17:52 +0200
Subject: [PATCH 1282/1455] drm/amd/pm: Remove unneeded legacy DPM related
 code.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This code isn't needed anymore as we collect the same information
into pm_display_cfg instead.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h           |  1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c  |  1 -
 drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c  | 74 -------------------
 drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h       |  4 -
 .../gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h  |  6 --
 .../gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c    |  3 -
 6 files changed, 89 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index c56839528843..adb4ea792c38 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -434,7 +434,6 @@ struct amdgpu_clock {
 	uint32_t default_mclk;
 	uint32_t default_sclk;
 	uint32_t default_dispclk;
-	uint32_t current_dispclk;
 	uint32_t dp_extclk;
 	uint32_t max_pixel_clock;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
index 9dfdc08cc887..763f2b8dcf13 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c
@@ -706,7 +706,6 @@ int amdgpu_atombios_get_clock_info(struct amdgpu_device *adev)
 		}
 		adev->clock.dp_extclk =
 			le16_to_cpu(firmware_info->info_21.usUniphyDPModeExtClkFreq);
-		adev->clock.current_dispclk = adev->clock.default_dispclk;
 
 		adev->clock.max_pixel_clock = le16_to_cpu(firmware_info->info.usMaxPixelClock);
 		if (adev->clock.max_pixel_clock == 0)
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
index 9ef965e4a92e..b5e9c3ecf703 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_dpm_internal.c
@@ -27,80 +27,6 @@
 #include "amdgpu_smu.h"
 #include "amdgpu_dpm_internal.h"
 
-void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev)
-{
-	struct drm_device *ddev = adev_to_drm(adev);
-	struct drm_crtc *crtc;
-	struct amdgpu_crtc *amdgpu_crtc;
-
-	adev->pm.dpm.new_active_crtcs = 0;
-	adev->pm.dpm.new_active_crtc_count = 0;
-	if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) {
-		list_for_each_entry(crtc,
-				    &ddev->mode_config.crtc_list, head) {
-			amdgpu_crtc = to_amdgpu_crtc(crtc);
-			if (amdgpu_crtc->enabled) {
-				adev->pm.dpm.new_active_crtcs |= (1 << amdgpu_crtc->crtc_id);
-				adev->pm.dpm.new_active_crtc_count++;
-			}
-		}
-	}
-}
-
-u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev)
-{
-	struct drm_device *dev = adev_to_drm(adev);
-	struct drm_crtc *crtc;
-	struct amdgpu_crtc *amdgpu_crtc;
-	u32 vblank_in_pixels;
-	u32 vblank_time_us = 0xffffffff; /* if the displays are off, vblank time is max */
-
-	if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) {
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-			amdgpu_crtc = to_amdgpu_crtc(crtc);
-			if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) {
-				vblank_in_pixels =
-					amdgpu_crtc->hw_mode.crtc_htotal *
-					(amdgpu_crtc->hw_mode.crtc_vblank_end -
-					amdgpu_crtc->hw_mode.crtc_vdisplay +
-					(amdgpu_crtc->v_border * 2));
-
-				vblank_time_us = vblank_in_pixels * 1000 / amdgpu_crtc->hw_mode.clock;
-
-				/* we have issues with mclk switching with
-				 * refresh rates over 120 hz on the non-DC code.
-				 */
-				if (drm_mode_vrefresh(&amdgpu_crtc->hw_mode) > 120)
-					vblank_time_us = 0;
-
-				break;
-			}
-		}
-	}
-
-	return vblank_time_us;
-}
-
-u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev)
-{
-	struct drm_device *dev = adev_to_drm(adev);
-	struct drm_crtc *crtc;
-	struct amdgpu_crtc *amdgpu_crtc;
-	u32 vrefresh = 0;
-
-	if (adev->mode_info.num_crtc && adev->mode_info.mode_config_initialized) {
-		list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
-			amdgpu_crtc = to_amdgpu_crtc(crtc);
-			if (crtc->enabled && amdgpu_crtc->enabled && amdgpu_crtc->hw_mode.clock) {
-				vrefresh = drm_mode_vrefresh(&amdgpu_crtc->hw_mode);
-				break;
-			}
-		}
-	}
-
-	return vrefresh;
-}
-
 void amdgpu_dpm_get_display_cfg(struct amdgpu_device *adev)
 {
 	struct drm_device *ddev = adev_to_drm(adev);
diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
index 9748744133d9..65c1d98af26c 100644
--- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h
@@ -263,10 +263,6 @@ struct amdgpu_dpm {
 	u32                     voltage_response_time;
 	u32                     backbias_response_time;
 	void                    *priv;
-	u32			new_active_crtcs;
-	int			new_active_crtc_count;
-	u32			current_active_crtcs;
-	int			current_active_crtc_count;
 	struct amdgpu_dpm_dynamic_state dyn_state;
 	struct amdgpu_dpm_fan fan;
 	u32 tdp_limit;
diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h
index 8be11510cd92..cc6d7ba040e9 100644
--- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h
+++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm_internal.h
@@ -23,12 +23,6 @@
 #ifndef __AMDGPU_DPM_INTERNAL_H__
 #define __AMDGPU_DPM_INTERNAL_H__
 
-void amdgpu_dpm_get_active_displays(struct amdgpu_device *adev);
-
-u32 amdgpu_dpm_get_vblank_time(struct amdgpu_device *adev);
-
-u32 amdgpu_dpm_get_vrefresh(struct amdgpu_device *adev);
-
 void amdgpu_dpm_get_display_cfg(struct amdgpu_device *adev);
 
 #endif
diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
index 6ebe3d0f5b87..c7ed0b457129 100644
--- a/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
+++ b/drivers/gpu/drm/amd/pm/legacy-dpm/legacy_dpm.c
@@ -944,9 +944,6 @@ static int amdgpu_dpm_change_power_state_locked(struct amdgpu_device *adev)
 
 	amdgpu_dpm_post_set_power_state(adev);
 
-	adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs;
-	adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count;
-
 	if (pp_funcs->force_performance_level) {
 		if (adev->pm.dpm.thermal_active) {
 			enum amd_dpm_forced_level level = adev->pm.dpm.forced_level;

From ae5c2bee1680436d9bf8bfaca7416496adff0ee0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Tue, 9 Sep 2025 16:49:35 +0200
Subject: [PATCH 1283/1455] drm/amdgpu: Fix allocating extra dwords for rings
 (v2)
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Rename extra_dw to extra_bytes and document what it's for.

The value is already used as if it were bytes in vcn_v4_0.c
and in amdgpu_ring_init. Just adjust the dword count in
jpeg_v1_0.c so that it becomes a byte count.

v2:
Rename extra_dw to extra_bytes as discussed during review.

Fixes: c8c1a1d2ef04 ("drm/amdgpu: define and add extra dword for jpeg ring")
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c |  3 ++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 13 ++++++++++++-
 drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c   |  2 +-
 drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c    |  2 +-
 4 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 486c3646710c..8f6ce948c684 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -364,7 +364,8 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 
 	/* Allocate ring buffer */
 	if (ring->ring_obj == NULL) {
-		r = amdgpu_bo_create_kernel(adev, ring->ring_size + ring->funcs->extra_dw, PAGE_SIZE,
+		r = amdgpu_bo_create_kernel(adev, ring->ring_size + ring->funcs->extra_bytes,
+					    PAGE_SIZE,
 					    AMDGPU_GEM_DOMAIN_GTT,
 					    &ring->ring_obj,
 					    &ring->gpu_addr,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 80b85547c810..4116dfb03499 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -211,7 +211,18 @@ struct amdgpu_ring_funcs {
 	bool			support_64bit_ptrs;
 	bool			no_user_fence;
 	bool			secure_submission_supported;
-	unsigned		extra_dw;
+
+	/**
+	 * @extra_bytes:
+	 *
+	 * Optional extra space in bytes that is added to the ring size
+	 * when allocating the BO that holds the contents of the ring.
+	 * This space isn't used for command submission to the ring,
+	 * but is just there to satisfy some hardware requirements or
+	 * implement workarounds. It's up to the implementation of each
+	 * specific ring to initialize this space.
+	 */
+	unsigned		extra_bytes;
 
 	/* ring read/write ptr handling */
 	u64 (*get_rptr)(struct amdgpu_ring *ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
index 9e428e669ada..b5bb7f4d607c 100644
--- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c
@@ -557,7 +557,7 @@ static const struct amdgpu_ring_funcs jpeg_v1_0_decode_ring_vm_funcs = {
 	.nop = PACKET0(0x81ff, 0),
 	.support_64bit_ptrs = false,
 	.no_user_fence = true,
-	.extra_dw = 64,
+	.extra_bytes = 256,
 	.get_rptr = jpeg_v1_0_decode_ring_get_rptr,
 	.get_wptr = jpeg_v1_0_decode_ring_get_wptr,
 	.set_wptr = jpeg_v1_0_decode_ring_set_wptr,
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
index 8d3f18b55fc2..3ae666522d57 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c
@@ -1977,7 +1977,7 @@ static struct amdgpu_ring_funcs vcn_v4_0_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
 	.nop = VCN_ENC_CMD_NO_OP,
-	.extra_dw = sizeof(struct amdgpu_vcn_rb_metadata),
+	.extra_bytes = sizeof(struct amdgpu_vcn_rb_metadata),
 	.get_rptr = vcn_v4_0_unified_ring_get_rptr,
 	.get_wptr = vcn_v4_0_unified_ring_get_wptr,
 	.set_wptr = vcn_v4_0_unified_ring_set_wptr,

From e2ee0f1b1ab1cfdab1ddc832e336fc85da286dfe Mon Sep 17 00:00:00 2001
From: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Date: Tue, 9 Sep 2025 16:49:37 +0200
Subject: [PATCH 1284/1455] drm/amdgpu: Use memset32 for ring clearing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Use memset32 instead of open coding it, just because it is
a tiny bit nicer.

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 4116dfb03499..b6b649179776 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -481,10 +481,7 @@ static inline void amdgpu_ring_set_preempt_cond_exec(struct amdgpu_ring *ring,
 
 static inline void amdgpu_ring_clear_ring(struct amdgpu_ring *ring)
 {
-	int i = 0;
-	while (i <= ring->buf_mask)
-		ring->ring[i++] = ring->funcs->nop;
-
+	memset32(ring->ring, ring->funcs->nop, ring->buf_mask + 1);
 }
 
 static inline void amdgpu_ring_write(struct amdgpu_ring *ring, uint32_t v)

From e09b081d8ae0e60022a66b26e9d51909c6f2e383 Mon Sep 17 00:00:00 2001
From: "Stanley.Yang" <Stanley.Yang@amd.com>
Date: Thu, 21 Aug 2025 16:05:42 +0800
Subject: [PATCH 1285/1455] drm/amdgpu: wait pmfw polling mca bank info done

wait 500ms to ensure pmfw polling mca bank info done.

Signed-off-by: Stanley.Yang <Stanley.Yang@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 7fe5b1940df8..663b96583bfd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -2702,6 +2702,7 @@ static void amdgpu_ras_do_recovery(struct work_struct *work)
 	struct amdgpu_device *adev = ras->adev;
 	struct list_head device_list, *device_list_handle =  NULL;
 	struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev);
+	unsigned int error_query_mode;
 	enum ras_event_type type;
 
 	if (hive) {
@@ -2730,6 +2731,13 @@ static void amdgpu_ras_do_recovery(struct work_struct *work)
 			device_list_handle = &device_list;
 		}
 
+		if (amdgpu_ras_get_error_query_mode(adev, &error_query_mode)) {
+			if (error_query_mode == AMDGPU_RAS_FIRMWARE_ERROR_QUERY) {
+				/* wait 500ms to ensure pmfw polling mca bank info done */
+				msleep(500);
+			}
+		}
+
 		type = amdgpu_ras_get_fatal_error_event(adev);
 		list_for_each_entry(remote_adev,
 				device_list_handle, gmc.xgmi.head) {

From c3cdc3bebd1fc610d6051f8e0678c05091fcab4c Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Wed, 10 Sep 2025 11:38:08 +0530
Subject: [PATCH 1286/1455] drm/amdgpu: Read memory vendor information

Read VRAM vendor information from scratch register for GC v9.4.3 and
GC v9.5.0 SOCs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 8404695eb13f..0d1dd587db5f 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1834,11 +1834,19 @@ static void gmc_v9_0_save_registers(struct amdgpu_device *adev)
 
 static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev)
 {
+	static const u32 regBIF_BIOS_SCRATCH_4 = 0x50;
+	u32 vram_info;
+
 	adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM;
 	adev->gmc.vram_width = 128 * 64;
 
 	if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 5, 0))
 		adev->gmc.vram_type = AMDGPU_VRAM_TYPE_HBM3E;
+
+	if (!(adev->flags & AMD_IS_APU) && !amdgpu_sriov_vf(adev)) {
+		vram_info = RREG32(regBIF_BIOS_SCRATCH_4);
+		adev->gmc.vram_vendor = vram_info & 0xF;
+	}
 }
 
 static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block)

From 1f9ba8ea04efa0e4366eac2d2ebf2deff167bf14 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 4 Sep 2025 17:43:44 +0530
Subject: [PATCH 1287/1455] drm/amdgpu: Add generic capability class

Define a utility macro for defining capabilities and their attributes.
Capability attributes are read-only, write-only, read-write.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu.h       |  1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_utils.h | 91 +++++++++++++++++++++++
 2 files changed, 92 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/amdgpu/amdgpu_utils.h

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index adb4ea792c38..2a0df4cabb99 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -63,6 +63,7 @@
 #include "kgd_pp_interface.h"
 
 #include "amd_shared.h"
+#include "amdgpu_utils.h"
 #include "amdgpu_mode.h"
 #include "amdgpu_ih.h"
 #include "amdgpu_irq.h"
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_utils.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_utils.h
new file mode 100644
index 000000000000..1e40ca3b1584
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_utils.h
@@ -0,0 +1,91 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef AMDGPU_UTILS_H_
+#define AMDGPU_UTILS_H_
+
+/* ---------- Generic 2‑bit capability attribute encoding ----------
+ * 00 INVALID, 01 RO, 10 WO, 11 RW
+ */
+enum amdgpu_cap_attr {
+	AMDGPU_CAP_ATTR_INVALID = 0,
+	AMDGPU_CAP_ATTR_RO      = 1 << 0,
+	AMDGPU_CAP_ATTR_WO      = 1 << 1,
+	AMDGPU_CAP_ATTR_RW      = (AMDGPU_CAP_ATTR_RO | AMDGPU_CAP_ATTR_WO),
+};
+
+#define AMDGPU_CAP_ATTR_BITS 2
+#define AMDGPU_CAP_ATTR_MAX  ((1U << AMDGPU_CAP_ATTR_BITS) - 1)
+
+/* Internal helper to build helpers for a given enum NAME */
+#define DECLARE_ATTR_CAP_CLASS_HELPERS(NAME)							\
+enum { NAME##_BITMAP_BITS = NAME##_COUNT * AMDGPU_CAP_ATTR_BITS };				\
+struct NAME##_caps {										\
+	DECLARE_BITMAP(bmap, NAME##_BITMAP_BITS);						\
+};												\
+static inline unsigned int NAME##_ATTR_START(enum NAME##_cap_id cap)				\
+{ return (unsigned int)cap * AMDGPU_CAP_ATTR_BITS; }						\
+static inline void NAME##_attr_init(struct NAME##_caps *c)					\
+{ if (c) bitmap_zero(c->bmap, NAME##_BITMAP_BITS); }						\
+static inline int NAME##_attr_set(struct NAME##_caps *c,					\
+				  enum NAME##_cap_id cap, enum amdgpu_cap_attr attr)		\
+{												\
+	if (!c)											\
+		return -EINVAL;									\
+	if (cap >= NAME##_COUNT)								\
+		return -EINVAL;									\
+	if ((unsigned int)attr > AMDGPU_CAP_ATTR_MAX)						\
+		return -EINVAL;									\
+	bitmap_write(c->bmap, (unsigned long)attr,						\
+			NAME##_ATTR_START(cap), AMDGPU_CAP_ATTR_BITS);				\
+	return 0;										\
+}												\
+static inline int NAME##_attr_get(const struct NAME##_caps *c,					\
+				  enum NAME##_cap_id cap, enum amdgpu_cap_attr *out)		\
+{												\
+	unsigned long v;									\
+	if (!c || !out)										\
+		return -EINVAL;									\
+	if (cap >= NAME##_COUNT)								\
+		return -EINVAL;									\
+	v = bitmap_read(c->bmap, NAME##_ATTR_START(cap), AMDGPU_CAP_ATTR_BITS);			\
+	*out = (enum amdgpu_cap_attr)v;								\
+	return 0;										\
+}												\
+static inline bool NAME##_cap_is_ro(const struct NAME##_caps *c, enum NAME##_cap_id id)		\
+{ enum amdgpu_cap_attr a; return !NAME##_attr_get(c, id, &a) && a == AMDGPU_CAP_ATTR_RO; }	\
+static inline bool NAME##_cap_is_wo(const struct NAME##_caps *c, enum NAME##_cap_id id)		\
+{ enum amdgpu_cap_attr a; return !NAME##_attr_get(c, id, &a) && a == AMDGPU_CAP_ATTR_WO; }	\
+static inline bool NAME##_cap_is_rw(const struct NAME##_caps *c, enum NAME##_cap_id id)		\
+{ enum amdgpu_cap_attr a; return !NAME##_attr_get(c, id, &a) && a == AMDGPU_CAP_ATTR_RW; }
+
+/* Element expander for enum creation */
+#define _CAP_ENUM_ELEM(x) x,
+
+/* Public macro: declare enum + helpers from an X‑macro list */
+#define DECLARE_ATTR_CAP_CLASS(NAME, LIST_MACRO)						\
+	enum NAME##_cap_id { LIST_MACRO(_CAP_ENUM_ELEM) NAME##_COUNT };				\
+	DECLARE_ATTR_CAP_CLASS_HELPERS(NAME)
+
+#endif /* AMDGPU_UTILS_H_ */

From 780f7a45e56e9fd2bbae146687c054fcf6a08743 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 4 Sep 2025 17:47:20 +0530
Subject: [PATCH 1288/1455] drm/amdgpu: Add virtual device capabilities

Add a member to define the capabilities of virtual device.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 58accf2259b3..3afbdf7b487a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -256,6 +256,10 @@ struct amdgpu_virt_ras {
 	uint64_t cper_rptr;
 };
 
+#define AMDGPU_VIRT_CAPS_LIST(X) X(AMDGPU_VIRT_CAP_POWER_LIMIT)
+
+DECLARE_ATTR_CAP_CLASS(amdgpu_virt, AMDGPU_VIRT_CAPS_LIST);
+
 /* GPU virtualization */
 struct amdgpu_virt {
 	uint32_t			caps;
@@ -274,6 +278,7 @@ struct amdgpu_virt {
 	const struct amdgpu_virt_ops	*ops;
 	struct amdgpu_vf_error_buffer	vf_errors;
 	struct amdgpu_virt_fw_reserve	fw_reserve;
+	struct amdgpu_virt_caps virt_caps;
 	uint32_t gim_feature;
 	uint32_t reg_access_mode;
 	int req_init_data_ver;

From 10a9c09a7c3b905b761d4e5df24d958314d6cf3d Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Tue, 26 Aug 2025 00:33:53 +0800
Subject: [PATCH 1289/1455] drm/amd/pm: Allow to set power cap in vf mode

Allow setting power cap for smu_v13_0_6 in 1vf mode

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c                   | 7 ++++---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 6 +++++-
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 96590c1da553..3e935a89d279 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -3254,9 +3254,6 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev,
 	int err;
 	u32 value;
 
-	if (amdgpu_sriov_vf(adev))
-		return -EINVAL;
-
 	err = kstrtou32(buf, 10, &value);
 	if (err)
 		return err;
@@ -3598,6 +3595,10 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 			return 0;
 	}
 
+	if (attr == &sensor_dev_attr_power1_cap.dev_attr.attr &&
+	    amdgpu_virt_cap_is_rw(&adev->virt.virt_caps, AMDGPU_VIRT_CAP_POWER_LIMIT))
+		effective_mode |= S_IWUSR;
+
 	/* not implemented yet for APUs having < GC 9.3.0 (Renoir) */
 	if (((adev->family == AMDGPU_FAMILY_SI) ||
 	     ((adev->flags & AMD_IS_APU) && (gc_ver < IP_VERSION(9, 3, 0)))) &&
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 7b087251f891..cbe1f314917a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -143,7 +143,7 @@ static const struct cmn2asic_msg_mapping smu_v13_0_6_message_map[SMU_MSG_MAX_COU
 	MSG_MAP(GetMinDpmFreq,			     PPSMC_MSG_GetMinDpmFreq,			1),
 	MSG_MAP(GetMaxDpmFreq,			     PPSMC_MSG_GetMaxDpmFreq,			1),
 	MSG_MAP(GetDpmFreqByIndex,		     PPSMC_MSG_GetDpmFreqByIndex,		1),
-	MSG_MAP(SetPptLimit,			     PPSMC_MSG_SetPptLimit,			0),
+	MSG_MAP(SetPptLimit,			     PPSMC_MSG_SetPptLimit,			1),
 	MSG_MAP(GetPptLimit,			     PPSMC_MSG_GetPptLimit,			1),
 	MSG_MAP(GfxDeviceDriverReset,		     PPSMC_MSG_GfxDriverReset,			SMU_MSG_RAS_PRI | SMU_MSG_NO_PRECHECK),
 	MSG_MAP(DramLogSetDramAddrHigh,		     PPSMC_MSG_DramLogSetDramAddrHigh,		0),
@@ -413,6 +413,10 @@ static void smu_v13_0_6_init_caps(struct smu_context *smu)
 			smu_v13_0_6_cap_set(smu, SMU_CAP(HST_LIMIT_METRICS));
 
 		if (amdgpu_sriov_vf(adev)) {
+			if (fw_ver >= 0x00558200)
+				amdgpu_virt_attr_set(&adev->virt.virt_caps,
+						     AMDGPU_VIRT_CAP_POWER_LIMIT,
+						     AMDGPU_CAP_ATTR_RW);
 			if ((pgm == 0 && fw_ver >= 0x00558000) ||
 			    (pgm == 7 && fw_ver >= 0x7551000)) {
 				smu_v13_0_6_cap_set(smu,

From efbacd8aeace6e0ab161e3254eb5e73fc6a85bf2 Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Thu, 4 Sep 2025 08:52:53 +0800
Subject: [PATCH 1290/1455] drm/amd/pm: unified smu feature cap interface

add a unified interface to provide smu feature cap set.

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c     | 29 +++++++++++++++++++
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h | 12 ++++++++
 2 files changed, 41 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index f6022c2ae361..c9ba9625e85e 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -1315,6 +1315,33 @@ static void smu_init_power_profile(struct smu_context *smu)
 	smu_power_profile_mode_get(smu, smu->power_profile_mode);
 }
 
+void smu_feature_cap_set(struct smu_context *smu, enum smu_feature_cap_id fea_id)
+{
+	struct smu_feature_cap *fea_cap = &smu->fea_cap;
+
+	if (fea_id >= SMU_FEATURE_CAP_ID__COUNT)
+		return;
+
+	set_bit(fea_id, fea_cap->cap_map);
+}
+
+bool smu_feature_cap_test(struct smu_context *smu, enum smu_feature_cap_id fea_id)
+{
+	struct smu_feature_cap *fea_cap = &smu->fea_cap;
+
+	if (fea_id >= SMU_FEATURE_CAP_ID__COUNT)
+		return false;
+
+	return test_bit(fea_id, fea_cap->cap_map);
+}
+
+static void smu_feature_cap_init(struct smu_context *smu)
+{
+	struct smu_feature_cap *fea_cap = &smu->fea_cap;
+
+	bitmap_zero(fea_cap->cap_map, SMU_FEATURE_CAP_ID__COUNT);
+}
+
 static int smu_sw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
@@ -1347,6 +1374,8 @@ static int smu_sw_init(struct amdgpu_ip_block *ip_block)
 	INIT_DELAYED_WORK(&smu->swctf_delayed_work,
 			  smu_swctf_delayed_work_handler);
 
+	smu_feature_cap_init(smu);
+
 	ret = smu_smc_table_sw_init(smu);
 	if (ret) {
 		dev_err(adev->dev, "Failed to sw init smc table!\n");
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 5976eda80035..ba1bab4b7036 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -528,6 +528,14 @@ enum smu_fw_status {
  */
 #define SMU_WBRF_EVENT_HANDLING_PACE	10
 
+enum smu_feature_cap_id {
+	SMU_FEATURE_CAP_ID__COUNT,
+};
+
+struct smu_feature_cap {
+	DECLARE_BITMAP(cap_map, SMU_FEATURE_CAP_ID__COUNT);
+};
+
 struct smu_context {
 	struct amdgpu_device            *adev;
 	struct amdgpu_irq_src		irq_source;
@@ -550,6 +558,7 @@ struct smu_context {
 	struct amd_pp_display_configuration  *display_config;
 	struct smu_baco_context		smu_baco;
 	struct smu_temperature_range	thermal_range;
+	struct smu_feature_cap		fea_cap;
 	void *od_settings;
 
 	struct smu_umd_pstate_table	pstate_table;
@@ -1788,4 +1797,7 @@ ssize_t smu_get_pm_policy_info(struct smu_context *smu,
 			       enum pp_pm_policy p_type, char *sysbuf);
 
 #endif
+
+void smu_feature_cap_set(struct smu_context *smu, enum smu_feature_cap_id fea_id);
+bool smu_feature_cap_test(struct smu_context *smu, enum smu_feature_cap_id fea_id);
 #endif

From e0dd9b8e9e669c5f7de0bbf2e1d2706a7c7d63f7 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 10 Sep 2025 13:40:30 -0500
Subject: [PATCH 1291/1455] drm/amd: Duplicate DC_FEATURE_MASK and
 DC_DEBUG_MASK enum values into kdoc

[Why]
When kernel documentation is generated the enum values themselves don't
end up in the documentation.  This makes browsing them in HTML a lot
less useful.

[How]
Copy DC_DEBUG_MASK and DC_FEATURE_MASK enum values into matching kdoc
comments.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/driver-core.rst |  2 +-
 drivers/gpu/drm/amd/include/amd_shared.h | 98 ++++++++++++++++--------
 2 files changed, 67 insertions(+), 33 deletions(-)

diff --git a/Documentation/gpu/amdgpu/driver-core.rst b/Documentation/gpu/amdgpu/driver-core.rst
index bd4be32f2725..3ce276272171 100644
--- a/Documentation/gpu/amdgpu/driver-core.rst
+++ b/Documentation/gpu/amdgpu/driver-core.rst
@@ -210,4 +210,4 @@ IP Blocks
    :doc: IP Blocks
 
 .. kernel-doc:: drivers/gpu/drm/amd/include/amd_shared.h
-   :identifiers: amd_ip_block_type amd_ip_funcs DC_DEBUG_MASK
+   :identifiers: amd_ip_block_type amd_ip_funcs DC_FEATURE_MASK DC_DEBUG_MASK
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index bfb446736ca8..75efda2969cf 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -239,18 +239,51 @@ enum amd_harvest_ip_mask {
     AMD_HARVEST_IP_DMU_MASK = 0x4,
 };
 
+/**
+ * enum DC_FEATURE_MASK - Bits that control DC feature defaults
+ */
 enum DC_FEATURE_MASK {
 	//Default value can be found at "uint amdgpu_dc_feature_mask"
-	DC_FBC_MASK = (1 << 0), //0x1, disabled by default
-	DC_MULTI_MON_PP_MCLK_SWITCH_MASK = (1 << 1), //0x2, enabled by default
-	DC_DISABLE_FRACTIONAL_PWM_MASK = (1 << 2), //0x4, disabled by default
-	DC_PSR_MASK = (1 << 3), //0x8, disabled by default for dcn < 3.1
-	DC_EDP_NO_POWER_SEQUENCING = (1 << 4), //0x10, disabled by default
-	DC_DISABLE_LTTPR_DP1_4A = (1 << 5), //0x20, disabled by default
-	DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default
-	DC_PSR_ALLOW_SMU_OPT = (1 << 7), //0x80, disabled by default
-	DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8), //0x100, disabled by default
-	DC_REPLAY_MASK = (1 << 9), //0x200, disabled by default for dcn < 3.1.4
+	/**
+	 * @DC_FBC_MASK: (0x1) disabled by default
+	 */
+	DC_FBC_MASK = (1 << 0),
+	/**
+	 * @DC_MULTI_MON_PP_MCLK_SWITCH_MASK: (0x2) enabled by default
+	 */
+	DC_MULTI_MON_PP_MCLK_SWITCH_MASK = (1 << 1),
+	/**
+	 * @DC_DISABLE_FRACTIONAL_PWM_MASK: (0x4) disabled by default
+	 */
+	DC_DISABLE_FRACTIONAL_PWM_MASK = (1 << 2),
+	/**
+	 * @DC_PSR_MASK: (0x8) disabled by default for DCN < 3.1
+	 */
+	DC_PSR_MASK = (1 << 3),
+	/**
+	 * @DC_EDP_NO_POWER_SEQUENCING: (0x10) disabled by default
+	 */
+	DC_EDP_NO_POWER_SEQUENCING = (1 << 4),
+	/**
+	 * @DC_DISABLE_LTTPR_DP1_4A: (0x20) disabled by default
+	 */
+	DC_DISABLE_LTTPR_DP1_4A = (1 << 5),
+	/**
+	 * @DC_DISABLE_LTTPR_DP2_0: (0x40) disabled by default
+	 */
+	DC_DISABLE_LTTPR_DP2_0 = (1 << 6),
+	/**
+	 * @DC_PSR_ALLOW_SMU_OPT: (0x80) disabled by default
+	 */
+	DC_PSR_ALLOW_SMU_OPT = (1 << 7),
+	/**
+	 * @DC_PSR_ALLOW_MULTI_DISP_OPT: (0x100) disabled by default
+	 */
+	DC_PSR_ALLOW_MULTI_DISP_OPT = (1 << 8),
+	/**
+	 * @DC_REPLAY_MASK: (0x200) disabled by default for DCN < 3.1.4
+	 */
+	DC_REPLAY_MASK = (1 << 9),
 };
 
 /**
@@ -258,64 +291,64 @@ enum DC_FEATURE_MASK {
  */
 enum DC_DEBUG_MASK {
 	/**
-	 * @DC_DISABLE_PIPE_SPLIT: If set, disable pipe-splitting
+	 * @DC_DISABLE_PIPE_SPLIT: (0x1) If set, disable pipe-splitting
 	 */
 	DC_DISABLE_PIPE_SPLIT = 0x1,
 
 	/**
-	 * @DC_DISABLE_STUTTER: If set, disable memory stutter mode
+	 * @DC_DISABLE_STUTTER: (0x2) If set, disable memory stutter mode
 	 */
 	DC_DISABLE_STUTTER = 0x2,
 
 	/**
-	 * @DC_DISABLE_DSC: If set, disable display stream compression
+	 * @DC_DISABLE_DSC: (0x4) If set, disable display stream compression
 	 */
 	DC_DISABLE_DSC = 0x4,
 
 	/**
-	 * @DC_DISABLE_CLOCK_GATING: If set, disable clock gating optimizations
+	 * @DC_DISABLE_CLOCK_GATING: (0x8) If set, disable clock gating optimizations
 	 */
 	DC_DISABLE_CLOCK_GATING = 0x8,
 
 	/**
-	 * @DC_DISABLE_PSR: If set, disable Panel self refresh v1 and PSR-SU
+	 * @DC_DISABLE_PSR: (0x10) If set, disable Panel self refresh v1 and PSR-SU
 	 */
 	DC_DISABLE_PSR = 0x10,
 
 	/**
-	 * @DC_FORCE_SUBVP_MCLK_SWITCH: If set, force mclk switch in subvp, even
+	 * @DC_FORCE_SUBVP_MCLK_SWITCH: (0x20) If set, force mclk switch in subvp, even
 	 * if mclk switch in vblank is possible
 	 */
 	DC_FORCE_SUBVP_MCLK_SWITCH = 0x20,
 
 	/**
-	 * @DC_DISABLE_MPO: If set, disable multi-plane offloading
+	 * @DC_DISABLE_MPO: (0x40) If set, disable multi-plane offloading
 	 */
 	DC_DISABLE_MPO = 0x40,
 
 	/**
-	 * @DC_ENABLE_DPIA_TRACE: If set, enable trace logging for DPIA
+	 * @DC_ENABLE_DPIA_TRACE: (0x80) If set, enable trace logging for DPIA
 	 */
 	DC_ENABLE_DPIA_TRACE = 0x80,
 
 	/**
-	 * @DC_ENABLE_DML2: If set, force usage of DML2, even if the DCN version
+	 * @DC_ENABLE_DML2: (0x100) If set, force usage of DML2, even if the DCN version
 	 * does not default to it.
 	 */
 	DC_ENABLE_DML2 = 0x100,
 
 	/**
-	 * @DC_DISABLE_PSR_SU: If set, disable PSR SU
+	 * @DC_DISABLE_PSR_SU: (0x200) If set, disable PSR SU
 	 */
 	DC_DISABLE_PSR_SU = 0x200,
 
 	/**
-	 * @DC_DISABLE_REPLAY: If set, disable Panel Replay
+	 * @DC_DISABLE_REPLAY: (0x400) If set, disable Panel Replay
 	 */
 	DC_DISABLE_REPLAY = 0x400,
 
 	/**
-	 * @DC_DISABLE_IPS: If set, disable all Idle Power States, all the time.
+	 * @DC_DISABLE_IPS: (0x800) If set, disable all Idle Power States, all the time.
 	 * If more than one IPS debug bit is set, the lowest bit takes
 	 * precedence. For example, if DC_FORCE_IPS_ENABLE and
 	 * DC_DISABLE_IPS_DYNAMIC are set, then DC_DISABLE_IPS_DYNAMIC takes
@@ -324,56 +357,57 @@ enum DC_DEBUG_MASK {
 	DC_DISABLE_IPS = 0x800,
 
 	/**
-	 * @DC_DISABLE_IPS_DYNAMIC: If set, disable all IPS, all the time,
+	 * @DC_DISABLE_IPS_DYNAMIC: (0x1000) If set, disable all IPS, all the time,
 	 * *except* when driver goes into suspend.
 	 */
 	DC_DISABLE_IPS_DYNAMIC = 0x1000,
 
 	/**
-	 * @DC_DISABLE_IPS2_DYNAMIC: If set, disable IPS2 (IPS1 allowed) if
+	 * @DC_DISABLE_IPS2_DYNAMIC: (0x2000) If set, disable IPS2 (IPS1 allowed) if
 	 * there is an enabled display. Otherwise, enable all IPS.
 	 */
 	DC_DISABLE_IPS2_DYNAMIC = 0x2000,
 
 	/**
-	 * @DC_FORCE_IPS_ENABLE: If set, force enable all IPS, all the time.
+	 * @DC_FORCE_IPS_ENABLE: (0x4000) If set, force enable all IPS, all the time.
 	 */
 	DC_FORCE_IPS_ENABLE = 0x4000,
 	/**
-	 * @DC_DISABLE_ACPI_EDID: If set, don't attempt to fetch EDID for
+	 * @DC_DISABLE_ACPI_EDID: (0x8000) If set, don't attempt to fetch EDID for
 	 * eDP display from ACPI _DDC method.
 	 */
 	DC_DISABLE_ACPI_EDID = 0x8000,
 
 	/**
-	 * @DC_DISABLE_HDMI_CEC: If set, disable HDMI-CEC feature in amdgpu driver.
+	 * @DC_DISABLE_HDMI_CEC: (0x10000) If set, disable HDMI-CEC feature in amdgpu driver.
 	 */
 	DC_DISABLE_HDMI_CEC = 0x10000,
 
 	/**
-	 * @DC_DISABLE_SUBVP_FAMS: If set, disable DCN Sub-Viewport & Firmware Assisted
+	 * @DC_DISABLE_SUBVP_FAMS: (0x20000) If set, disable DCN Sub-Viewport & Firmware Assisted
 	 * Memory Clock Switching (FAMS) feature in amdgpu driver.
 	 */
 	DC_DISABLE_SUBVP_FAMS = 0x20000,
 	/**
-	 * @DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE: If set, disable support for custom brightness curves
+	 * @DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE: (0x40000) If set, disable support for custom
+	 * brightness curves
 	 */
 	DC_DISABLE_CUSTOM_BRIGHTNESS_CURVE = 0x40000,
 
 	/**
-	 * @DC_HDCP_LC_FORCE_FW_ENABLE: If set, use HDCP Locality Check FW
+	 * @DC_HDCP_LC_FORCE_FW_ENABLE: (0x80000) If set, use HDCP Locality Check FW
 	 * path regardless of reported HW capabilities.
 	 */
 	DC_HDCP_LC_FORCE_FW_ENABLE = 0x80000,
 
 	/**
-	 * @DC_HDCP_LC_ENABLE_SW_FALLBACK: If set, upon HDCP Locality Check FW
+	 * @DC_HDCP_LC_ENABLE_SW_FALLBACK: (0x100000) If set, upon HDCP Locality Check FW
 	 * path failure, retry using legacy SW path.
 	 */
 	DC_HDCP_LC_ENABLE_SW_FALLBACK = 0x100000,
 
 	/**
-	 * @DC_SKIP_DETECTION_LT: If set, skip detection link training
+	 * @DC_SKIP_DETECTION_LT: (0x200000) If set, skip detection link training
 	 */
 	DC_SKIP_DETECTION_LT = 0x200000,
 };

From 0a71ceb27f88a944c2de2808b67b2f46ac75076b Mon Sep 17 00:00:00 2001
From: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Date: Wed, 10 Sep 2025 12:27:05 +0530
Subject: [PATCH 1292/1455] drm/amdgpu/gfx11: Add Cleaner Shader Support for
 GFX11.0.1/11.0.4 GPUs
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Enable the cleaner shader for additional GFX11.0.1/11.0.4 series GPUs to
ensure data isolation among GPU tasks. The cleaner shader is tasked with
clearing the Local Data Store (LDS), Vector General Purpose Registers
(VGPRs), and Scalar General Purpose Registers (SGPRs), which helps avoid
data leakage and guarantees the accuracy of computational results.

This update extends cleaner shader support to GFX11.0.1/11.0.4 GPUs,
previously available for GFX11.0.3. It enhances security by clearing GPU
memory between processes and maintains a consistent GPU state across KGD
and KFD workloads.

Cc: Wasee Alam <wasee.alam@amd.com>
Cc: Mario Sopena-Novales <mario.novales@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 86b3fcab5772..66c47c466532 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1653,6 +1653,21 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
 			}
 		}
 		break;
+	case IP_VERSION(11, 0, 1):
+	case IP_VERSION(11, 0, 4):
+		adev->gfx.cleaner_shader_ptr = gfx_11_0_3_cleaner_shader_hex;
+		adev->gfx.cleaner_shader_size = sizeof(gfx_11_0_3_cleaner_shader_hex);
+		if (adev->gfx.pfp_fw_version >= 102 &&
+		    adev->gfx.mec_fw_version >= 66 &&
+		    adev->mes.fw_version[0] >= 128) {
+			adev->gfx.enable_cleaner_shader = true;
+			r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size);
+			if (r) {
+				adev->gfx.enable_cleaner_shader = false;
+				dev_err(adev->dev, "Failed to initialize cleaner shader\n");
+			}
+		}
+		break;
 	case IP_VERSION(11, 5, 0):
 	case IP_VERSION(11, 5, 1):
 		adev->gfx.cleaner_shader_ptr = gfx_11_0_3_cleaner_shader_hex;

From 46fbe1e349b89eff48d6d8e87cfb6ffe955f75e9 Mon Sep 17 00:00:00 2001
From: Hawking Zhang <Hawking.Zhang@amd.com>
Date: Thu, 11 Sep 2025 23:22:04 +0800
Subject: [PATCH 1293/1455] Revert "drm/amdgpu: Allocate psp fw private buffer
 in vram"

This reverts commit 22dcb283d63d5677a5875d0002d04d2c61720f78.
Need to certain APU platforms and will proceed to rework
the patch accordingly

Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Le Ma <Le.Ma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 3696f48c233b..1578e4e2bf84 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -506,7 +506,8 @@ static int psp_sw_init(struct amdgpu_ip_block *ip_block)
 	}
 
 	ret = amdgpu_bo_create_kernel(adev, PSP_1_MEG, PSP_1_MEG,
-				      AMDGPU_GEM_DOMAIN_VRAM,
+				      (amdgpu_sriov_vf(adev) || adev->debug_use_vram_fw_buf) ?
+				      AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT,
 				      &psp->fw_pri_bo,
 				      &psp->fw_pri_mc_addr,
 				      &psp->fw_pri_buf);

From 0bfcb56e8b074c9f013f357e799c446a9dbc1fad Mon Sep 17 00:00:00 2001
From: Zhikai Zhai <zhikai.zhai@amd.com>
Date: Tue, 29 Jul 2025 17:39:54 +0800
Subject: [PATCH 1294/1455] drm/amd/display: Modify the link training policy

[Why&How]
Currently fallback to low link rate if the link training
fails once on USB4. It may cause the bandwidth couldn't
satisfy the requirement of streams. Modify the policy
to do training retry in the previous few times, only
do fallback at the last time.

Reviewed-by: Meenakshikumar Somasundaram <meenakshikumar.somasundaram@amd.com>
Signed-off-by: Zhikai Zhai <zhikai.zhai@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/link/protocols/link_dp_training.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
index 134093ce5a8e..08e2b572e0ff 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c
@@ -1729,6 +1729,15 @@ bool perform_link_training_with_retries(
 				break;
 		}
 
+		if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA &&
+				stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST &&
+				!link->dc->config.enable_dpia_pre_training) {
+			if (j == (attempts - 1))
+				do_fallback = true;
+			else
+				do_fallback = false;
+		}
+
 		if (j == (attempts - 1)) {
 			DC_LOG_WARNING(
 				"%s: Link(%d) training attempt %u of %d failed @ rate(%d) x lane(%d) @ spread = %x : fail reason:(%d)\n",

From 44343e8b250abb2f6bfd615493ca07a7f11f3cc2 Mon Sep 17 00:00:00 2001
From: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Date: Wed, 11 Jun 2025 16:51:21 -0400
Subject: [PATCH 1295/1455] drm/amd/display: prepare dml 2.1 for new asic

[Why&How]
prepare dml 2.1 for new asic

Acked-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc_state.c              | 2 +-
 .../gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h   | 6 +++---
 drivers/gpu/drm/amd/display/dc/inc/resource.h               | 1 +
 3 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_state.c b/drivers/gpu/drm/amd/display/dc/core/dc_state.c
index 883054bb18e7..c61300a7cb1c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_state.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_state.c
@@ -211,7 +211,7 @@ struct dc_state *dc_state_create(struct dc *dc, struct dc_state_create_params *p
 			return NULL;
 		}
 
-		if (!dml2_create(dc, &dc->dml2_dc_power_options, &state->bw_ctx.dml2_dc_power_source)) {
+		if (dc->caps.dcmode_power_limits_present && !dml2_create(dc, &dc->dml2_dc_power_options, &state->bw_ctx.dml2_dc_power_source)) {
 			dc_state_release(state);
 			return NULL;
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h
index 7de10a95cfdb..41adb1104d0f 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/inc/dml_top_types.h
@@ -16,9 +16,9 @@ struct dml2_instance;
 
 enum dml2_project_id {
 	dml2_project_invalid = 0,
-	dml2_project_dcn4x_stage1 = 1,
-	dml2_project_dcn4x_stage2 = 2,
-	dml2_project_dcn4x_stage2_auto_drr_svp = 3,
+	dml2_project_dcn4x_stage1,
+	dml2_project_dcn4x_stage2,
+	dml2_project_dcn4x_stage2_auto_drr_svp,
 };
 
 enum dml2_pstate_change_support {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h
index a890f581f4e8..4e26a16a8743 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/resource.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h
@@ -45,6 +45,7 @@ enum dce_version resource_parse_asic_id(
 struct resource_caps {
 	int num_timing_generator;
 	int num_opp;
+	int num_dpp;
 	int num_video_plane;
 	int num_audio;
 	int num_stream_encoder;

From 18e755155caa57a6e6c4aa4a40b0db0fba015289 Mon Sep 17 00:00:00 2001
From: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Date: Fri, 8 Aug 2025 10:26:22 -0400
Subject: [PATCH 1296/1455] drm/amd/display: Fix DMCUB loading sequence for
 DCN3.2

[Why]
New sequence from HW for reset and firmware reloading has been
provided that aims to stabilize the reload sequence in the case the
firmware is hung or has outstanding requests.

[How]
Update the sequence to remove the DMUIF reset and the redundant
writes in the release.

Reviewed-by: Sreeja Golui <sreeja.golui@amd.com>
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dmub/src/dmub_dcn32.c | 53 ++++++++++---------
 .../gpu/drm/amd/display/dmub/src/dmub_dcn32.h |  8 ++-
 2 files changed, 35 insertions(+), 26 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
index e7056205b050..ce041f6239dc 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.c
@@ -89,44 +89,50 @@ static inline void dmub_dcn32_translate_addr(const union dmub_addr *addr_in,
 void dmub_dcn32_reset(struct dmub_srv *dmub)
 {
 	union dmub_gpint_data_register cmd;
-	const uint32_t timeout = 30;
-	uint32_t in_reset, scratch, i;
+	const uint32_t timeout = 100000;
+	uint32_t in_reset, is_enabled, scratch, i, pwait_mode;
 
 	REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &in_reset);
+	REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_enabled);
 
-	if (in_reset == 0) {
+	if (in_reset == 0 && is_enabled != 0) {
 		cmd.bits.status = 1;
 		cmd.bits.command_code = DMUB_GPINT__STOP_FW;
 		cmd.bits.param = 0;
 
 		dmub->hw_funcs.set_gpint(dmub, cmd);
 
-		/**
-		 * Timeout covers both the ACK and the wait
-		 * for remaining work to finish.
-		 *
-		 * This is mostly bound by the PHY disable sequence.
-		 * Each register check will be greater than 1us, so
-		 * don't bother using udelay.
-		 */
-
 		for (i = 0; i < timeout; ++i) {
 			if (dmub->hw_funcs.is_gpint_acked(dmub, cmd))
 				break;
+
+			udelay(1);
 		}
 
 		for (i = 0; i < timeout; ++i) {
-			scratch = dmub->hw_funcs.get_gpint_response(dmub);
+			scratch = REG_READ(DMCUB_SCRATCH7);
 			if (scratch == DMUB_GPINT__STOP_FW_RESPONSE)
 				break;
+
+			udelay(1);
 		}
 
+		for (i = 0; i < timeout; ++i) {
+			REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &pwait_mode);
+			if (pwait_mode & (1 << 0))
+				break;
+
+			udelay(1);
+		}
 		/* Force reset in case we timed out, DMCUB is likely hung. */
 	}
 
-	REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1);
-	REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0);
-	REG_UPDATE(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET, 1);
+	if (is_enabled) {
+		REG_UPDATE(DMCUB_CNTL2, DMCUB_SOFT_RESET, 1);
+		udelay(1);
+		REG_UPDATE(DMCUB_CNTL, DMCUB_ENABLE, 0);
+	}
+
 	REG_WRITE(DMCUB_INBOX1_RPTR, 0);
 	REG_WRITE(DMCUB_INBOX1_WPTR, 0);
 	REG_WRITE(DMCUB_OUTBOX1_RPTR, 0);
@@ -135,7 +141,7 @@ void dmub_dcn32_reset(struct dmub_srv *dmub)
 	REG_WRITE(DMCUB_OUTBOX0_WPTR, 0);
 	REG_WRITE(DMCUB_SCRATCH0, 0);
 
-	/* Clear the GPINT command manually so we don't reset again. */
+	/* Clear the GPINT command manually so we don't send anything during boot. */
 	cmd.all = 0;
 	dmub->hw_funcs.set_gpint(dmub, cmd);
 }
@@ -419,8 +425,8 @@ uint32_t dmub_dcn32_get_current_time(struct dmub_srv *dmub)
 
 void dmub_dcn32_get_diagnostic_data(struct dmub_srv *dmub)
 {
-	uint32_t is_dmub_enabled, is_soft_reset, is_sec_reset;
-	uint32_t is_traceport_enabled, is_cw0_enabled, is_cw6_enabled;
+	uint32_t is_dmub_enabled, is_soft_reset, is_pwait;
+	uint32_t is_traceport_enabled, is_cw6_enabled;
 	struct dmub_timeout_info timeout = {0};
 
 	if (!dmub)
@@ -470,18 +476,15 @@ void dmub_dcn32_get_diagnostic_data(struct dmub_srv *dmub)
 	REG_GET(DMCUB_CNTL, DMCUB_ENABLE, &is_dmub_enabled);
 	dmub->debug.is_dmcub_enabled = is_dmub_enabled;
 
+	REG_GET(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS, &is_pwait);
+	dmub->debug.is_pwait = is_pwait;
+
 	REG_GET(DMCUB_CNTL2, DMCUB_SOFT_RESET, &is_soft_reset);
 	dmub->debug.is_dmcub_soft_reset = is_soft_reset;
 
-	REG_GET(DMCUB_SEC_CNTL, DMCUB_SEC_RESET_STATUS, &is_sec_reset);
-	dmub->debug.is_dmcub_secure_reset = is_sec_reset;
-
 	REG_GET(DMCUB_CNTL, DMCUB_TRACEPORT_EN, &is_traceport_enabled);
 	dmub->debug.is_traceport_en  = is_traceport_enabled;
 
-	REG_GET(DMCUB_REGION3_CW0_TOP_ADDRESS, DMCUB_REGION3_CW0_ENABLE, &is_cw0_enabled);
-	dmub->debug.is_cw0_enabled = is_cw0_enabled;
-
 	REG_GET(DMCUB_REGION3_CW6_TOP_ADDRESS, DMCUB_REGION3_CW6_ENABLE, &is_cw6_enabled);
 	dmub->debug.is_cw6_enabled = is_cw6_enabled;
 
diff --git a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
index 1a229450c53d..daf81027d663 100644
--- a/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
+++ b/drivers/gpu/drm/amd/display/dmub/src/dmub_dcn32.h
@@ -89,6 +89,9 @@ struct dmub_srv;
 	DMUB_SR(DMCUB_REGION5_OFFSET) \
 	DMUB_SR(DMCUB_REGION5_OFFSET_HIGH) \
 	DMUB_SR(DMCUB_REGION5_TOP_ADDRESS) \
+	DMUB_SR(DMCUB_REGION6_OFFSET) \
+	DMUB_SR(DMCUB_REGION6_OFFSET_HIGH) \
+	DMUB_SR(DMCUB_REGION6_TOP_ADDRESS) \
 	DMUB_SR(DMCUB_SCRATCH0) \
 	DMUB_SR(DMCUB_SCRATCH1) \
 	DMUB_SR(DMCUB_SCRATCH2) \
@@ -155,6 +158,8 @@ struct dmub_srv;
 	DMUB_SF(DMCUB_REGION4_TOP_ADDRESS, DMCUB_REGION4_ENABLE) \
 	DMUB_SF(DMCUB_REGION5_TOP_ADDRESS, DMCUB_REGION5_TOP_ADDRESS) \
 	DMUB_SF(DMCUB_REGION5_TOP_ADDRESS, DMCUB_REGION5_ENABLE) \
+	DMUB_SF(DMCUB_REGION6_TOP_ADDRESS, DMCUB_REGION6_TOP_ADDRESS) \
+	DMUB_SF(DMCUB_REGION6_TOP_ADDRESS, DMCUB_REGION6_ENABLE) \
 	DMUB_SF(CC_DC_PIPE_DIS, DC_DMCUB_ENABLE) \
 	DMUB_SF(MMHUBBUB_SOFT_RESET, DMUIF_SOFT_RESET) \
 	DMUB_SF(DCN_VM_FB_LOCATION_BASE, FB_BASE) \
@@ -162,7 +167,8 @@ struct dmub_srv;
 	DMUB_SF(DMCUB_INBOX0_WPTR, DMCUB_INBOX0_WPTR) \
 	DMUB_SF(DMCUB_REGION3_TMR_AXI_SPACE, DMCUB_REGION3_TMR_AXI_SPACE) \
 	DMUB_SF(DMCUB_INTERRUPT_ENABLE, DMCUB_GPINT_IH_INT_EN) \
-	DMUB_SF(DMCUB_INTERRUPT_ACK, DMCUB_GPINT_IH_INT_ACK)
+	DMUB_SF(DMCUB_INTERRUPT_ACK, DMCUB_GPINT_IH_INT_ACK) \
+	DMUB_SF(DMCUB_CNTL, DMCUB_PWAIT_MODE_STATUS)
 
 struct dmub_srv_dcn32_reg_offset {
 #define DMUB_SR(reg) uint32_t reg;

From 782f0bb5a129f588dc16ad54a2e88c93ef2e079a Mon Sep 17 00:00:00 2001
From: Wesley Chalmers <Wesley.Chalmers@amd.com>
Date: Tue, 2 Sep 2025 21:47:39 -0400
Subject: [PATCH 1297/1455] drm/amd/display: Rename header file link.h to
 link_service.h

[WHY]
Header file name "link.h" collides with system header when dc is
compiled as a user-mode library

[WHAT]
Rename link.h to link_service.h to avoid name collision

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c    | 2 +-
 drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c            | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c    | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c  | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c  | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c  | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c    | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c    | 2 +-
 .../gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c  | 2 +-
 drivers/gpu/drm/amd/display/dc/core/dc.c                    | 2 +-
 drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c       | 2 +-
 drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c       | 3 ++-
 drivers/gpu/drm/amd/display/dc/core/dc_resource.c           | 2 +-
 drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c            | 2 +-
 .../gpu/drm/amd/display/dc/dio/dcn10/dcn10_stream_encoder.c | 2 +-
 .../gpu/drm/amd/display/dc/dio/dcn20/dcn20_stream_encoder.c | 2 +-
 .../drm/amd/display/dc/dio/dcn31/dcn31_dio_link_encoder.c   | 2 +-
 .../amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c   | 2 +-
 .../drm/amd/display/dc/dio/dcn32/dcn32_dio_stream_encoder.c | 2 +-
 .../drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c | 2 +-
 .../amd/display/dc/dio/dcn401/dcn401_dio_stream_encoder.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c        | 3 +--
 drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c        | 2 +-
 drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c        | 2 +-
 drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c      | 2 +-
 drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c              | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c     | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c   | 2 +-
 .../gpu/drm/amd/display/dc/inc/{link.h => link_service.h}   | 6 +++---
 .../gpu/drm/amd/display/dc/link/accessories/link_dp_cts.h   | 2 +-
 .../gpu/drm/amd/display/dc/link/accessories/link_dp_trace.h | 2 +-
 drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h    | 2 +-
 .../dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h        | 2 +-
 drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h | 2 +-
 .../dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h     | 2 +-
 drivers/gpu/drm/amd/display/dc/link/link_detection.h        | 2 +-
 drivers/gpu/drm/amd/display/dc/link/link_dpms.h             | 2 +-
 drivers/gpu/drm/amd/display/dc/link/link_factory.h          | 2 +-
 drivers/gpu/drm/amd/display/dc/link/link_resource.h         | 2 +-
 drivers/gpu/drm/amd/display/dc/link/link_validation.h       | 2 +-
 drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h    | 2 +-
 .../drm/amd/display/dc/link/protocols/link_dp_capability.h  | 2 +-
 .../gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h    | 2 +-
 .../gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h | 2 +-
 .../drm/amd/display/dc/link/protocols/link_dp_irq_handler.h | 2 +-
 drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_phy.h | 2 +-
 .../drm/amd/display/dc/link/protocols/link_dp_training.h    | 2 +-
 drivers/gpu/drm/amd/display/dc/link/protocols/link_dpcd.h   | 2 +-
 .../amd/display/dc/link/protocols/link_edp_panel_control.h  | 2 +-
 drivers/gpu/drm/amd/display/dc/link/protocols/link_hpd.h    | 2 +-
 .../drm/amd/display/dc/resource/dce120/dce120_resource.c    | 2 +-
 .../gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c  | 2 +-
 .../gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c  | 2 +-
 .../drm/amd/display/dc/resource/dcn302/dcn302_resource.c    | 3 ++-
 .../drm/amd/display/dc/resource/dcn303/dcn303_resource.c    | 2 +-
 .../gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c  | 2 +-
 .../drm/amd/display/dc/resource/dcn321/dcn321_resource.c    | 2 +-
 .../gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c  | 2 +-
 .../drm/amd/display/dc/resource/dcn351/dcn351_resource.c    | 2 +-
 .../gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c  | 2 +-
 .../drm/amd/display/dc/resource/dcn401/dcn401_resource.c    | 2 +-
 69 files changed, 73 insertions(+), 72 deletions(-)
 rename drivers/gpu/drm/amd/display/dc/inc/{link.h => link_service.h} (98%)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c
index 82ea3fe5e764..80704d709e44 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_replay.c
@@ -31,7 +31,7 @@
 #include "amdgpu_dm.h"
 #include "modules/power/power_helpers.h"
 #include "dmub/inc/dmub_cmd.h"
-#include "dc/inc/link.h"
+#include "dc/inc/link_service.h"
 
 /*
  * amdgpu_dm_link_supports_replay() - check if the link supports replay
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
index 4071851f9e86..15cf13ec5302 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c
@@ -28,7 +28,7 @@
 #include "dccg.h"
 #include "clk_mgr_internal.h"
 #include "dc_state_priv.h"
-#include "link.h"
+#include "link_service.h"
 
 #include "dce100/dce_clk_mgr.h"
 #include "dce110/dce110_clk_mgr.h"
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
index bc123f1884da..051052bd10c9 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c
@@ -47,7 +47,7 @@
 #include "dcn30/dcn30_clk_mgr.h"
 
 #include "dc_dmub_srv.h"
-#include "link.h"
+#include "link_service.h"
 
 #include "logger_types.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 91d872d6d392..790bbd8235b1 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -48,7 +48,7 @@
 #include "dcn31/dcn31_clk_mgr.h"
 
 #include "dc_dmub_srv.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn314_smu.h"
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
index e4d22f74f986..b315ed91e010 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn315/dcn315_clk_mgr.c
@@ -46,7 +46,7 @@
 #define DC_LOGGER \
 	clk_mgr->base.base.ctx->logger
 
-#include "link.h"
+#include "link_service.h"
 
 #define TO_CLK_MGR_DCN315(clk_mgr)\
 	container_of(clk_mgr, struct clk_mgr_dcn315, base)
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
index 49efea0c8fcf..1769b1f26e75 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn316/dcn316_clk_mgr.c
@@ -39,7 +39,7 @@
 #include "dcn316_smu.h"
 #include "dm_helpers.h"
 #include "dc_dmub_srv.h"
-#include "link.h"
+#include "link_service.h"
 
 // DCN316 this is CLK1 instance
 #define MAX_INSTANCE                                        7
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
index 8376e2b0e73d..7da7b41bd092 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr.c
@@ -33,7 +33,7 @@
 #include "reg_helper.h"
 #include "core_types.h"
 #include "dm_helpers.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_state_priv.h"
 #include "atomfirmware.h"
 #include "dcn32_smu13_driver_if.h"
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
index bb1ac12a2b09..86edf11b8c5a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
@@ -44,7 +44,7 @@
 #include "dcn31/dcn31_clk_mgr.h"
 
 #include "dc_dmub_srv.h"
-#include "link.h"
+#include "link_service.h"
 #include "logger_types.h"
 
 #undef DC_LOGGER
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
index 47ff4c965d76..47461f249e83 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
@@ -13,7 +13,7 @@
 #include "reg_helper.h"
 #include "core_types.h"
 #include "dm_helpers.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_state_priv.h"
 #include "atomfirmware.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 5963019d1e74..269206ebee60 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -60,7 +60,7 @@
 #include "link_encoder.h"
 #include "link_enc_cfg.h"
 
-#include "link.h"
+#include "link_service.h"
 #include "dm_helpers.h"
 #include "mem_input.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c
index 814f68d76257..a180f68f711c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_enc_cfg.c
@@ -24,7 +24,7 @@
 
 #include "link_enc_cfg.h"
 #include "resource.h"
-#include "link.h"
+#include "link_service.h"
 
 #define DC_LOGGER dc->ctx->logger
 
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
index b7a5de4ecb61..9acd30019717 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_exports.c
@@ -33,8 +33,9 @@
  * dc.h with detail interface documentation, then add function implementation
  * in this file which calls link functions.
  */
-#include "link.h"
+#include "link_service.h"
 #include "dce/dce_i2c.h"
+
 struct dc_link *dc_get_link_at_index(struct dc *dc, uint32_t link_index)
 {
 	if (link_index >= MAX_LINKS)
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index d712548b1927..89c805457ed8 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -40,7 +40,7 @@
 #include "virtual/virtual_stream_encoder.h"
 #include "dpcd_defs.h"
 #include "link_enc_cfg.h"
-#include "link.h"
+#include "link_service.h"
 #include "clk_mgr.h"
 #include "dc_state_priv.h"
 #include "dc_stream_priv.h"
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
index 65b979617b0c..9e2a473a8852 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
@@ -3,7 +3,7 @@
 // Copyright 2024 Advanced Micro Devices, Inc.
 
 #include "dc.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_dmub_srv.h"
 #include "dmub/dmub_srv.h"
 #include "core_types.h"
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_stream_encoder.c
index 22e66b375a7f..d928b4dcf6b8 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn10/dcn10_stream_encoder.c
@@ -28,7 +28,7 @@
 #include "dcn10_stream_encoder.h"
 #include "reg_helper.h"
 #include "hw_shared.h"
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 #include "dcn30/dcn30_afmt.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn20/dcn20_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn20/dcn20_stream_encoder.c
index 0b47aeb60e79..bec0b4aaeb2b 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn20/dcn20_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn20/dcn20_stream_encoder.c
@@ -29,7 +29,7 @@
 #include "dcn20_stream_encoder.h"
 #include "reg_helper.h"
 #include "hw_shared.h"
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn31/dcn31_dio_link_encoder.c
index 9a92f73d5b7f..84cc2ddc52fe 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn31/dcn31_dio_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn31/dcn31_dio_link_encoder.c
@@ -37,7 +37,7 @@
 #include "link_enc_cfg.h"
 #include "dc_dmub_srv.h"
 #include "dal_asic_id.h"
-#include "link.h"
+#include "link_service.h"
 
 #define CTX \
 	enc10->base.ctx
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c
index ae81451a3a72..3e85e9c3d2cb 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c
@@ -30,7 +30,7 @@
 #include "dcn314_dio_stream_encoder.h"
 #include "reg_helper.h"
 #include "hw_shared.h"
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn32/dcn32_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn32/dcn32_dio_stream_encoder.c
index 1a9bb614c41e..3523d1cdc1a3 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn32/dcn32_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn32/dcn32_dio_stream_encoder.c
@@ -29,7 +29,7 @@
 #include "dcn32_dio_stream_encoder.h"
 #include "reg_helper.h"
 #include "hw_shared.h"
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c
index 6f30b6cc3c76..fd5d1dbf9dc6 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn35/dcn35_dio_stream_encoder.c
@@ -29,7 +29,7 @@
 #include "dcn35_dio_stream_encoder.h"
 #include "reg_helper.h"
 #include "hw_shared.h"
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn401/dcn401_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn401/dcn401_dio_stream_encoder.c
index d5fa551dd3c9..99aab70ef3e1 100644
--- a/drivers/gpu/drm/amd/display/dc/dio/dcn401/dcn401_dio_stream_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dio/dcn401/dcn401_dio_stream_encoder.c
@@ -32,7 +32,7 @@
 #include "dcn401_dio_stream_encoder.h"
 #include "reg_helper.h"
 #include "hw_shared.h"
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
index 2a2eaf6adf26..7aaf13bbd4e4 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c
@@ -30,8 +30,7 @@
 #include "dcn20/dcn20_resource.h"
 #include "dcn21/dcn21_resource.h"
 #include "clk_mgr/dcn21/rn_clk_mgr.h"
-
-#include "link.h"
+#include "link_service.h"
 #include "dcn20_fpu.h"
 #include "dc_state_priv.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 18388fb00be8..8a0f128722b0 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -31,7 +31,7 @@
 // We need this includes for WATERMARKS_* defines
 #include "clk_mgr/dcn32/dcn32_smu13_driver_if.h"
 #include "dcn30/dcn30_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_state_priv.h"
 
 #define DC_LOGGER_INIT(logger)
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c
index 5d73efa2f0c9..c9dd920744c9 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn35/dcn35_fpu.c
@@ -31,7 +31,7 @@
 #include "dml/dcn31/dcn31_fpu.h"
 #include "dml/dml_inline_defs.h"
 
-#include "link.h"
+#include "link_service.h"
 
 #define DC_LOGGER_INIT(logger)
 
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c
index 6f516af82956..8cda18ce1a76 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c
@@ -10,7 +10,7 @@
 #include "dml/dcn35/dcn35_fpu.h"
 #include "dml/dml_inline_defs.h"
 
-#include "link.h"
+#include "link_service.h"
 
 #define DC_LOGGER_INIT(logger)
 
diff --git a/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c b/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c
index 1313a7c5d87b..73a1e6a03719 100644
--- a/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c
+++ b/drivers/gpu/drm/amd/display/dc/hdcp/hdcp_msg.c
@@ -28,7 +28,7 @@
 #include "include/hdcp_msg_types.h"
 #include "include/signal_types.h"
 #include "core_types.h"
-#include "link.h"
+#include "link_service.h"
 #include "link_hwss.h"
 #include "link/protocols/link_dpcd.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index 153d68375fa3..ea904d52af20 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -48,7 +48,7 @@
 #include "link_encoder.h"
 #include "link_enc_cfg.h"
 #include "link_hwss.h"
-#include "link.h"
+#include "link_service.h"
 #include "dccg.h"
 #include "clock_source.h"
 #include "clk_mgr.h"
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index 506c3bbbf221..74f5e05f9cb4 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -55,7 +55,7 @@
 #include "dce/dmub_hw_lock_mgr.h"
 #include "dc_trace.h"
 #include "dce/dmub_outbox.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_state_priv.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index cc377fcda6ff..417f2679723e 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -54,7 +54,7 @@
 #include "dpcd_defs.h"
 #include "inc/link_enc_cfg.h"
 #include "link_hwss.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_state_priv.h"
 
 #define DC_LOGGER \
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c
index 61efb15572ff..e2269211553c 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c
@@ -35,7 +35,7 @@
 #include "hw/clk_mgr.h"
 #include "dc_dmub_srv.h"
 #include "abm.h"
-#include "link.h"
+#include "link_service.h"
 
 #define DC_LOGGER_INIT(logger)
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c
index 139a63101488..e47ed5571dfd 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c
@@ -50,7 +50,7 @@
 #include "dpcd_defs.h"
 #include "dcn20/dcn20_hwseq.h"
 #include "dcn30/dcn30_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "dc_state_priv.h"
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c
index 8ba934b83957..b822f2dffff0 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c
@@ -45,7 +45,7 @@
 #include "link_hwss.h"
 #include "dpcd_defs.h"
 #include "dce/dmub_outbox.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn10/dcn10_hwseq.h"
 #include "dcn21/dcn21_hwseq.h"
 #include "inc/link_enc_cfg.h"
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
index 560984533950..f925f669f2a4 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c
@@ -46,7 +46,7 @@
 #include "link_hwss.h"
 #include "dpcd_defs.h"
 #include "dce/dmub_outbox.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn10/dcn10_hwseq.h"
 #include "inc/link_enc_cfg.h"
 #include "dcn30/dcn30_vpg.h"
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
index 416b1dca3dac..3a6f946643ef 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
@@ -49,7 +49,7 @@
 #include "dcn20/dcn20_optc.h"
 #include "dce/dmub_hw_lock_mgr.h"
 #include "dcn32/dcn32_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "../dcn20/dcn20_hwseq.h"
 #include "dc_state_priv.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
index 764eff6a4ec6..05011061822c 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_hwseq.c
@@ -46,7 +46,7 @@
 #include "link_hwss.h"
 #include "dpcd_defs.h"
 #include "dce/dmub_outbox.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn10/dcn10_hwseq.h"
 #include "inc/link_enc_cfg.h"
 #include "dcn30/dcn30_vpg.h"
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index d5b5e2ce6ff6..9956145e3d18 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -25,7 +25,7 @@
 #include "dpcd_defs.h"
 #include "clk_mgr.h"
 #include "dsc.h"
-#include "link.h"
+#include "link_service.h"
 
 #include "dce/dmub_hw_lock_mgr.h"
 #include "dcn10/dcn10_cm_common.h"
diff --git a/drivers/gpu/drm/amd/display/dc/inc/link.h b/drivers/gpu/drm/amd/display/dc/inc/link_service.h
similarity index 98%
rename from drivers/gpu/drm/amd/display/dc/inc/link.h
rename to drivers/gpu/drm/amd/display/dc/inc/link_service.h
index 0cce49d95e26..1e34e84160aa 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/link.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/link_service.h
@@ -42,8 +42,8 @@
  * dc_link_exports.c or other dc files implement dc.h
  *
  * DC to Link:
- * dc_link_exports.c or other dc files include link.h
- * link_factory.c implements link.h
+ * dc_link_exports.c or other dc files include link_service.h
+ * link_factory.c implements link_service.h
  *
  * Link sub-component to Link sub-component:
  * link_factory.c includes --> link_xxx.h
@@ -73,7 +73,7 @@
  * 2. Implement your function in the suitable link_xxx.c file.
  * 3. Assign the function to link_service in link_factory.c
  * 4. NEVER include link_xxx.h headers outside link component.
- * 5. NEVER include link.h on DM side.
+ * 5. NEVER include link_service.h on DM side.
  */
 #include "core_types.h"
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.h b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.h
index eae23ea7f6ec..033650cdb811 100644
--- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.h
+++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.h
@@ -24,7 +24,7 @@
  */
 #ifndef __LINK_DP_CTS_H__
 #define __LINK_DP_CTS_H__
-#include "link.h"
+#include "link_service.h"
 void dp_handle_automated_test(struct dc_link *link);
 bool dp_set_test_pattern(
 		struct dc_link *link,
diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_trace.h b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_trace.h
index ab437a0c9101..9ff4a6c46a2b 100644
--- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_trace.h
+++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_trace.h
@@ -24,7 +24,7 @@
  */
 #ifndef __LINK_DP_TRACE_H__
 #define __LINK_DP_TRACE_H__
-#include "link.h"
+#include "link_service.h"
 
 void dp_trace_init(struct dc_link *link);
 void dp_trace_reset(struct dc_link *link);
diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h
index 45f0e091fcb0..4a25210a344f 100644
--- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h
+++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio.h
@@ -27,7 +27,7 @@
 #define __LINK_HWSS_DIO_H__
 
 #include "link_hwss.h"
-#include "link.h"
+#include "link_service.h"
 
 const struct link_hwss *get_dio_link_hwss(void);
 bool can_use_dio_link_hwss(const struct dc_link *link,
diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h
index 9ac08a332540..cf578a8662a4 100644
--- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h
+++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_dio_fixed_vs_pe_retimer.h
@@ -25,7 +25,7 @@
 #ifndef __LINK_HWSS_DIO_FIXED_VS_PE_RETIMER_H__
 #define __LINK_HWSS_DIO_FIXED_VS_PE_RETIMER_H__
 
-#include "link.h"
+#include "link_service.h"
 
 uint32_t dp_dio_fixed_vs_pe_retimer_get_lttpr_write_address(struct dc_link *link);
 uint8_t dp_dio_fixed_vs_pe_retimer_lane_cfg_to_hw_cfg(struct dc_link *link);
diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h
index 1d3ed8ca83b5..7c9005bc2587 100644
--- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h
+++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_dp.h
@@ -26,7 +26,7 @@
 #define __LINK_HWSS_HPO_DP_H__
 
 #include "link_hwss.h"
-#include "link.h"
+#include "link_service.h"
 
 void set_hpo_dp_throttled_vcp_size(struct pipe_ctx *pipe_ctx,
 		struct fixed31_32 throttled_vcp_size);
diff --git a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h
index 82301187bc7c..8bf36827ecfb 100644
--- a/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h
+++ b/drivers/gpu/drm/amd/display/dc/link/hwss/link_hwss_hpo_fixed_vs_pe_retimer_dp.h
@@ -25,7 +25,7 @@
 #ifndef __LINK_HWSS_HPO_FIXED_VS_PE_RETIMER_DP_H__
 #define __LINK_HWSS_HPO_FIXED_VS_PE_RETIMER_DP_H__
 
-#include "link.h"
+#include "link_service.h"
 
 bool requires_fixed_vs_pe_retimer_hpo_link_hwss(const struct dc_link *link);
 const struct link_hwss *get_hpo_fixed_vs_pe_retimer_dp_link_hwss(void);
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.h b/drivers/gpu/drm/amd/display/dc/link/link_detection.h
index 7da05078721e..1ab29476060b 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.h
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.h
@@ -25,7 +25,7 @@
 
 #ifndef __DC_LINK_DETECTION_H__
 #define __DC_LINK_DETECTION_H__
-#include "link.h"
+#include "link_service.h"
 bool link_detect(struct dc_link *link, enum dc_detect_reason reason);
 bool link_detect_connection_type(struct dc_link *link,
 		enum dc_connection_type *type);
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.h b/drivers/gpu/drm/amd/display/dc/link/link_dpms.h
index 9398f9c1666a..bd6fc63064a3 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.h
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.h
@@ -26,7 +26,7 @@
 #ifndef __DC_LINK_DPMS_H__
 #define __DC_LINK_DPMS_H__
 
-#include "link.h"
+#include "link_service.h"
 void link_set_dpms_on(
 		struct dc_state *state,
 		struct pipe_ctx *pipe_ctx);
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_factory.h b/drivers/gpu/drm/amd/display/dc/link/link_factory.h
index e96220d48d03..aad36ca1a31c 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_factory.h
+++ b/drivers/gpu/drm/amd/display/dc/link/link_factory.h
@@ -24,7 +24,7 @@
  */
 #ifndef __LINK_FACTORY_H__
 #define __LINK_FACTORY_H__
-#include "link.h"
+#include "link_service.h"
 struct dc_link *link_create(const struct link_init_data *init_params);
 void link_destroy(struct dc_link **link);
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_resource.h b/drivers/gpu/drm/amd/display/dc/link/link_resource.h
index 1907bda3cb6e..f7aa3bc3a93a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/link/link_resource.h
@@ -24,7 +24,7 @@
  */
 #ifndef __LINK_RESOURCE_H__
 #define __LINK_RESOURCE_H__
-#include "link.h"
+#include "link_service.h"
 void link_get_cur_res_map(const struct dc *dc, uint32_t *map);
 void link_restore_res_map(const struct dc *dc, uint32_t *map);
 void link_get_cur_link_res(const struct dc_link *link,
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_validation.h b/drivers/gpu/drm/amd/display/dc/link/link_validation.h
index 9553c81053fe..595774e76453 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_validation.h
+++ b/drivers/gpu/drm/amd/display/dc/link/link_validation.h
@@ -24,7 +24,7 @@
  */
 #ifndef __LINK_VALIDATION_H__
 #define __LINK_VALIDATION_H__
-#include "link.h"
+#include "link_service.h"
 
 enum dc_status link_validate_mode_timing(
 		const struct dc_stream_state *stream,
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h
index a3e25e55bed6..d3e6f01a6a90 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_ddc.h
@@ -26,7 +26,7 @@
 #ifndef __DAL_DDC_SERVICE_H__
 #define __DAL_DDC_SERVICE_H__
 
-#include "link.h"
+#include "link_service.h"
 
 #define AUX_POWER_UP_WA_DELAY 500
 #define I2C_OVER_AUX_DEFER_WA_DELAY 70
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
index 7170db5a1c13..6e17f72a752f 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.h
@@ -26,7 +26,7 @@
 #ifndef __DC_LINK_DP_CAPABILITY_H__
 #define __DC_LINK_DP_CAPABILITY_H__
 
-#include "link.h"
+#include "link_service.h"
 
 bool detect_dp_sink_caps(struct dc_link *link);
 
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h
index a61edfc9ca7a..7cd03fa4892b 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.h
@@ -27,7 +27,7 @@
 #ifndef __DC_LINK_DPIA_H__
 #define __DC_LINK_DPIA_H__
 
-#include "link.h"
+#include "link_service.h"
 
 /* Read tunneling device capability from DPCD and update link capability
  * accordingly.
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h
index 41efcb3e44e2..30cd8e2b9d35 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia_bw.h
@@ -26,7 +26,7 @@
 #ifndef DC_INC_LINK_DP_DPIA_BW_H_
 #define DC_INC_LINK_DP_DPIA_BW_H_
 
-#include "link.h"
+#include "link_service.h"
 
 
 /*
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.h
index ac33730fedd4..87516fb3b45a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_irq_handler.h
@@ -26,7 +26,7 @@
 #ifndef __DC_LINK_DP_IRQ_HANDLER_H__
 #define __DC_LINK_DP_IRQ_HANDLER_H__
 
-#include "link.h"
+#include "link_service.h"
 bool dp_parse_link_loss_status(
 	struct dc_link *link,
 	union hpd_irq_data *hpd_irq_dpcd_data);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_phy.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_phy.h
index ab1c1f8f1f8b..58e154494582 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_phy.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_phy.h
@@ -26,7 +26,7 @@
 #ifndef __DC_LINK_DP_PHY_H__
 #define __DC_LINK_DP_PHY_H__
 
-#include "link.h"
+#include "link_service.h"
 void dp_enable_link_phy(
 	struct dc_link *link,
 	const struct link_resource *link_res,
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h
index 574b083e0936..ce52de22ab7a 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.h
@@ -26,7 +26,7 @@
 
 #ifndef __DC_LINK_DP_TRAINING_H__
 #define __DC_LINK_DP_TRAINING_H__
-#include "link.h"
+#include "link_service.h"
 
 bool perform_link_training_with_retries(
 	const struct dc_link_settings *link_setting,
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dpcd.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dpcd.h
index 08d787a1e451..c2717c678c72 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dpcd.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dpcd.h
@@ -25,7 +25,7 @@
 
 #ifndef __LINK_DPCD_H__
 #define __LINK_DPCD_H__
-#include "link.h"
+#include "link_service.h"
 #include "dpcd_defs.h"
 
 enum dc_status core_link_read_dpcd(
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h
index 4a475d5b9dde..62a6344e613e 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.h
@@ -25,7 +25,7 @@
 
 #ifndef __DC_LINK_EDP_PANEL_CONTROL_H__
 #define __DC_LINK_EDP_PANEL_CONTROL_H__
-#include "link.h"
+#include "link_service.h"
 
 enum dp_panel_mode dp_get_panel_mode(struct dc_link *link);
 void dp_set_panel_mode(struct dc_link *link, enum dp_panel_mode panel_mode);
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_hpd.h b/drivers/gpu/drm/amd/display/dc/link/protocols/link_hpd.h
index 4fb526b264f9..af529328ba17 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_hpd.h
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_hpd.h
@@ -26,7 +26,7 @@
 
 #ifndef __DC_LINK_HPD_H__
 #define __DC_LINK_HPD_H__
-#include "link.h"
+#include "link_service.h"
 
 enum hpd_source_id get_hpd_line(struct dc_link *link);
 /*
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
index 2f23cc6df571..540e04ec1e2d 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c
@@ -67,7 +67,7 @@
 #include "reg_helper.h"
 
 #include "dce100/dce100_resource.h"
-#include "link.h"
+#include "link_service.h"
 
 #ifndef mmDP0_DP_DPHY_INTERNAL_CTRL
 	#define mmDP0_DP_DPHY_INTERNAL_CTRL		0x210f
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
index f9cbdad3ef37..84b38d2d6967 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c
@@ -85,7 +85,7 @@
 #include "vm_helper.h"
 
 #include "link_enc_cfg.h"
-#include "link.h"
+#include "link_service.h"
 
 #define DC_LOGGER_INIT(logger)
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
index 201ed863b69e..ff63f59ff928 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c
@@ -60,7 +60,7 @@
 #include "dml/display_mode_vba.h"
 #include "dcn30/dcn30_dccg.h"
 #include "dcn10/dcn10_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "dce/dce_panel_cntl.h"
 
 #include "dcn30/dcn30_dwb.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c
index 3345068a878c..61623cb518d9 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c
@@ -47,7 +47,8 @@
 
 #include "dcn10/dcn10_resource.h"
 
-#include "link.h"
+#include "link_service.h"
+
 #include "dce/dce_abm.h"
 #include "dce/dce_audio.h"
 #include "dce/dce_aux.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c
index 3479e1eab4cd..02b9a84f2db3 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c
@@ -47,7 +47,7 @@
 
 #include "dcn10/dcn10_resource.h"
 
-#include "link.h"
+#include "link_service.h"
 
 #include "dce/dce_abm.h"
 #include "dce/dce_audio.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
index 9917b366f00c..4acd97b3e4c9 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
@@ -69,7 +69,7 @@
 #include "dml/display_mode_vba.h"
 #include "dcn32/dcn32_dccg.h"
 #include "dcn10/dcn10_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn31/dcn31_panel_cntl.h"
 
 #include "dcn30/dcn30_dwb.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
index 061c0907d802..ad214986f7ac 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c
@@ -72,7 +72,7 @@
 #include "dml/display_mode_vba.h"
 #include "dcn32/dcn32_dccg.h"
 #include "dcn10/dcn10_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn31/dcn31_panel_cntl.h"
 
 #include "dcn30/dcn30_dwb.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
index 1d1a002f6d54..07552445e424 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c
@@ -61,7 +61,7 @@
 #include "dcn31/dcn31_hpo_dp_stream_encoder.h"
 #include "dcn31/dcn31_hpo_dp_link_encoder.h"
 #include "dcn32/dcn32_hpo_dp_link_encoder.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn31/dcn31_apg.h"
 #include "dcn32/dcn32_dio_link_encoder.h"
 #include "dcn31/dcn31_vpg.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
index 47912e0861a2..cb0478a9a34d 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c
@@ -40,7 +40,7 @@
 #include "dcn31/dcn31_hpo_dp_stream_encoder.h"
 #include "dcn31/dcn31_hpo_dp_link_encoder.h"
 #include "dcn32/dcn32_hpo_dp_link_encoder.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn31/dcn31_apg.h"
 #include "dcn32/dcn32_dio_link_encoder.h"
 #include "dcn31/dcn31_vpg.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
index 9ba91e214dda..126090c9bb8a 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn36/dcn36_resource.c
@@ -40,7 +40,7 @@
 #include "dcn31/dcn31_hpo_dp_stream_encoder.h"
 #include "dcn31/dcn31_hpo_dp_link_encoder.h"
 #include "dcn32/dcn32_hpo_dp_link_encoder.h"
-#include "link.h"
+#include "link_service.h"
 #include "dcn31/dcn31_apg.h"
 #include "dcn32/dcn32_dio_link_encoder.h"
 #include "dcn31/dcn31_vpg.h"
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
index 068c123ea8a8..d498c0983d74 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
@@ -50,7 +50,7 @@
 #include "dml/display_mode_vba.h"
 #include "dcn401/dcn401_dccg.h"
 #include "dcn10/dcn10_resource.h"
-#include "link.h"
+#include "link_service.h"
 #include "link_enc_cfg.h"
 #include "dcn31/dcn31_panel_cntl.h"
 

From 95d168b367aa28a59f94fc690ff76ebf69312c6d Mon Sep 17 00:00:00 2001
From: Ivan Lipski <ivan.lipski@amd.com>
Date: Tue, 2 Sep 2025 16:20:09 -0400
Subject: [PATCH 1298/1455] drm/amd/display: Allow RX6xxx & RX7700 to invoke
 amdgpu_irq_get/put

[Why&How]
As reported on https://gitlab.freedesktop.org/drm/amd/-/issues/3936,
SMU hang can occur if the interrupts are not enabled appropriately,
causing a vblank timeout.

This patch reverts commit 5009628d8509 ("drm/amd/display: Remove unnecessary
amdgpu_irq_get/put"), but only for RX6xxx & RX7700 GPUs, on which the
issue was observed.

This will re-enable interrupts regardless of whether the user space needed
it or not.

Fixes: 5009628d8509 ("drm/amd/display: Remove unnecessary amdgpu_irq_get/put")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3936
Suggested-by: Sun peng Li <sunpeng.li@amd.com>
Reviewed-by: Sun peng Li <sunpeng.li@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 39 ++++++++++++++++++-
 1 file changed, 38 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index fadc6098eaee..b088cb8ae780 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8796,7 +8796,16 @@ static int amdgpu_dm_encoder_init(struct drm_device *dev,
 static void manage_dm_interrupts(struct amdgpu_device *adev,
 				 struct amdgpu_crtc *acrtc,
 				 struct dm_crtc_state *acrtc_state)
-{
+{	/*
+	 * We cannot be sure that the frontend index maps to the same
+	 * backend index - some even map to more than one.
+	 * So we have to go through the CRTC to find the right IRQ.
+	 */
+	int irq_type = amdgpu_display_crtc_idx_to_irq_type(
+			adev,
+			acrtc->crtc_id);
+	struct drm_device *dev = adev_to_drm(adev);
+
 	struct drm_vblank_crtc_config config = {0};
 	struct dc_crtc_timing *timing;
 	int offdelay;
@@ -8849,7 +8858,35 @@ static void manage_dm_interrupts(struct amdgpu_device *adev,
 
 		drm_crtc_vblank_on_config(&acrtc->base,
 					  &config);
+		/* Allow RX6xxx, RX7700, RX7800 GPUs to call amdgpu_irq_get.*/
+		switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) {
+		case IP_VERSION(3, 0, 0):
+		case IP_VERSION(3, 0, 2):
+		case IP_VERSION(3, 0, 3):
+		case IP_VERSION(3, 2, 0):
+			if (amdgpu_irq_get(adev, &adev->pageflip_irq, irq_type))
+				drm_err(dev, "DM_IRQ: Cannot get pageflip irq!\n");
+#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
+			if (amdgpu_irq_get(adev, &adev->vline0_irq, irq_type))
+				drm_err(dev, "DM_IRQ: Cannot get vline0 irq!\n");
+#endif
+		}
+
 	} else {
+		/* Allow RX6xxx, RX7700, RX7800 GPUs to call amdgpu_irq_put.*/
+		switch (amdgpu_ip_version(adev, DCE_HWIP, 0)) {
+		case IP_VERSION(3, 0, 0):
+		case IP_VERSION(3, 0, 2):
+		case IP_VERSION(3, 0, 3):
+		case IP_VERSION(3, 2, 0):
+#if defined(CONFIG_DRM_AMD_SECURE_DISPLAY)
+			if (amdgpu_irq_put(adev, &adev->vline0_irq, irq_type))
+				drm_err(dev, "DM_IRQ: Cannot put vline0 irq!\n");
+#endif
+			if (amdgpu_irq_put(adev, &adev->pageflip_irq, irq_type))
+				drm_err(dev, "DM_IRQ: Cannot put pageflip irq!\n");
+		}
+
 		drm_crtc_vblank_off(&acrtc->base);
 	}
 }

From 29591e6fd422c154ef14abbffe02c29644149a35 Mon Sep 17 00:00:00 2001
From: Charlene Liu <Charlene.Liu@amd.com>
Date: Wed, 3 Sep 2025 15:00:22 -0400
Subject: [PATCH 1299/1455] drm/amd/display: limit one non-related log to dGPU

[Why&How]
some log are for dGPU only.
Added check to limit log.

Reviewed-by: Chris Park <chris.park@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 269206ebee60..c4dd52ed377d 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -5622,8 +5622,8 @@ void dc_allow_idle_optimizations_internal(struct dc *dc, bool allow, char const
 			subvp_pipe_type[i] = dc_state_get_pipe_subvp_type(context, pipe);
 		}
 	}
-
-	DC_LOG_DC("%s: allow_idle=%d\n HardMinUClk_Khz=%d HardMinDramclk_Khz=%d\n Pipe_0=%d Pipe_1=%d Pipe_2=%d Pipe_3=%d Pipe_4=%d Pipe_5=%d (caller=%s)\n",
+	if (!dc->caps.is_apu)
+		DC_LOG_DC("%s: allow_idle=%d\n HardMinUClk_Khz=%d HardMinDramclk_Khz=%d\n Pipe_0=%d Pipe_1=%d Pipe_2=%d Pipe_3=%d Pipe_4=%d Pipe_5=%d (caller=%s)\n",
 			__func__, allow, idle_fclk_khz, idle_dramclk_khz, subvp_pipe_type[0], subvp_pipe_type[1], subvp_pipe_type[2],
 			subvp_pipe_type[3], subvp_pipe_type[4], subvp_pipe_type[5], caller_name);
 

From 02a6c2e4b28ff31f7a904c196a99fb2efe81e2cf Mon Sep 17 00:00:00 2001
From: Ausef Yousof <Ausef.Yousof@amd.com>
Date: Tue, 2 Sep 2025 12:10:18 -0400
Subject: [PATCH 1300/1455] drm/amd/display: fix dml ms order of operations

[why&how]
small error in order of operations in immediateflipbytes
calculation on dml ms side that can result in dml ms
and mp mismatch immediateflip support for a given pipe
and thus an invalid hw state, correct the order to align
with mp.

Reviewed-by: Leo Chen <leo.chen@amd.com>
Signed-off-by: Ausef Yousof <Ausef.Yousof@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c
index 715f9019a33e..4b9b2e84d381 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/display_mode_core.c
@@ -6529,7 +6529,7 @@ static noinline_for_stack void dml_prefetch_check(struct display_mode_lib_st *mo
 				mode_lib->ms.TotImmediateFlipBytes = 0;
 				for (k = 0; k <= mode_lib->ms.num_active_planes - 1; k++) {
 					if (!(mode_lib->ms.policy.ImmediateFlipRequirement[k] == dml_immediate_flip_not_required)) {
-						mode_lib->ms.TotImmediateFlipBytes = mode_lib->ms.TotImmediateFlipBytes + mode_lib->ms.NoOfDPP[j][k] * mode_lib->ms.PDEAndMetaPTEBytesPerFrame[j][k] + mode_lib->ms.MetaRowBytes[j][k];
+						mode_lib->ms.TotImmediateFlipBytes = mode_lib->ms.TotImmediateFlipBytes + mode_lib->ms.NoOfDPP[j][k] * (mode_lib->ms.PDEAndMetaPTEBytesPerFrame[j][k] + mode_lib->ms.MetaRowBytes[j][k]);
 						if (mode_lib->ms.use_one_row_for_frame_flip[j][k]) {
 							mode_lib->ms.TotImmediateFlipBytes = mode_lib->ms.TotImmediateFlipBytes + mode_lib->ms.NoOfDPP[j][k] * (2 * mode_lib->ms.DPTEBytesPerRow[j][k]);
 						} else {

From 5e76bc677cb7c92b37d8bc66bb67a18922895be2 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <Mario.Limonciello@amd.com>
Date: Tue, 26 Aug 2025 20:18:22 -0500
Subject: [PATCH 1301/1455] drm/amd/display: Set up pixel encoding for YCBCR422

[Why]
fill_stream_properties_from_drm_display_mode() will not configure pixel
encoding to YCBCR422 when the DRM color format supports YCBCR422 but not
YCBCR420 or YCBCR4444.  Instead it will fallback to RGB.

[How]
Add support for YCBCR422 in pixel encoding mapping.

Suggested-by: Mauri Carvalho <mcarvalho3@lenovo.com>
Reviewed-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Mario Limonciello <Mario.Limonciello@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index b088cb8ae780..cb22cad279be 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6406,6 +6406,9 @@ static void fill_stream_properties_from_drm_display_mode(
 			&& aconnector
 			&& aconnector->force_yuv420_output)
 		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420;
+	else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCBCR422)
+			&& stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
+		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR422;
 	else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCBCR444)
 			&& stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
 		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR444;

From db291ed1732e02e79dca431838713bbf602bda1c Mon Sep 17 00:00:00 2001
From: Mario Limonciello <Mario.Limonciello@amd.com>
Date: Wed, 27 Aug 2025 12:17:19 -0500
Subject: [PATCH 1302/1455] drm/amd/display: Add fallback path for YCBCR422

[Why]
DP validation may fail with multiple displays and higher color depths.
The sink may support others though.

[How]
When DP bandwidth validation fails, progressively fallback through:
- YUV422 8bpc (bandwidth efficient)
- YUV422 6bpc (reduced color depth)
- YUV420 (last resort)

This resolves cases where displays would show no image due to insufficient
DP link bandwidth for the requested RGB mode.

Suggested-by: Mauri Carvalho <mcarvalho3@lenovo.com>
Reviewed-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Mario Limonciello <Mario.Limonciello@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 45 +++++++++++++++----
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  1 +
 2 files changed, 37 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index cb22cad279be..7495122358bd 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6407,7 +6407,8 @@ static void fill_stream_properties_from_drm_display_mode(
 			&& aconnector->force_yuv420_output)
 		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR420;
 	else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCBCR422)
-			&& stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
+			&& aconnector
+			&& aconnector->force_yuv422_output)
 		timing_out->pixel_encoding = PIXEL_ENCODING_YCBCR422;
 	else if ((connector->display_info.color_formats & DRM_COLOR_FORMAT_YCBCR444)
 			&& stream->signal == SIGNAL_TYPE_HDMI_TYPE_A)
@@ -7669,6 +7670,7 @@ create_validate_stream_for_sink(struct drm_connector *connector,
 		bpc_limit = 8;
 
 	do {
+		drm_dbg_kms(connector->dev, "Trying with %d bpc\n", requested_bpc);
 		stream = create_stream_for_sink(connector, drm_mode,
 						dm_state, old_stream,
 						requested_bpc);
@@ -7704,16 +7706,41 @@ create_validate_stream_for_sink(struct drm_connector *connector,
 
 	} while (stream == NULL && requested_bpc >= bpc_limit);
 
-	if ((dc_result == DC_FAIL_ENC_VALIDATE ||
-	     dc_result == DC_EXCEED_DONGLE_CAP) &&
-	     !aconnector->force_yuv420_output) {
-		DRM_DEBUG_KMS("%s:%d Retry forcing yuv420 encoding\n",
-				     __func__, __LINE__);
-
-		aconnector->force_yuv420_output = true;
+	switch (dc_result) {
+	/*
+	 * If we failed to validate DP bandwidth stream with the requested RGB color depth,
+	 * we try to fallback and configure in order:
+	 * YUV422 (8bpc, 6bpc)
+	 * YUV420 (8bpc, 6bpc)
+	 */
+	case DC_FAIL_ENC_VALIDATE:
+	case DC_EXCEED_DONGLE_CAP:
+	case DC_NO_DP_LINK_BANDWIDTH:
+		/* recursively entered twice and already tried both YUV422 and YUV420 */
+		if (aconnector->force_yuv422_output && aconnector->force_yuv420_output)
+			break;
+		/* first failure; try YUV422 */
+		if (!aconnector->force_yuv422_output) {
+			drm_dbg_kms(connector->dev, "%s:%d Validation failed with %d, retrying w/ YUV422\n",
+				    __func__, __LINE__, dc_result);
+			aconnector->force_yuv422_output = true;
+		/* recursively entered and YUV422 failed, try YUV420 */
+		} else if (!aconnector->force_yuv420_output) {
+			drm_dbg_kms(connector->dev, "%s:%d Validation failed with %d, retrying w/ YUV420\n",
+				    __func__, __LINE__, dc_result);
+			aconnector->force_yuv420_output = true;
+		}
 		stream = create_validate_stream_for_sink(connector, drm_mode,
-						dm_state, old_stream);
+							 dm_state, old_stream);
+		aconnector->force_yuv422_output = false;
 		aconnector->force_yuv420_output = false;
+		break;
+	case DC_OK:
+		break;
+	default:
+		drm_dbg_kms(connector->dev, "%s:%d Unhandled validation failure %d\n",
+			    __func__, __LINE__, dc_result);
+		break;
 	}
 
 	return stream;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index ce74125c713e..2967cd28efe4 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -794,6 +794,7 @@ struct amdgpu_dm_connector {
 
 	bool fake_enable;
 	bool force_yuv420_output;
+	bool force_yuv422_output;
 	struct dsc_preferred_settings dsc_settings;
 	union dp_downstream_port_present mst_downstream_port_present;
 	/* Cached display modes */

From edae98a2bdf25d719297f5aa5dfbfc1b4d86bde5 Mon Sep 17 00:00:00 2001
From: Relja Vojvodic <rvojvodi@amd.com>
Date: Thu, 4 Sep 2025 15:38:24 -0400
Subject: [PATCH 1303/1455] drm/amd/display: Add DSC padding for OVT Support

[Why]
-Certain OVT timings require DSC configurations which divide the
horizontal active unevenly across DSC slices
-DSC slices must be even, so padding needs to be added to the active
to make this possible
-The pixel clock of the HW now needs to be increased to accommodate
the extra padded pixels
-To keep the line time the same, the blank of the HW timing needs to
be increased as well

[How]
-Calculate h_active padding, h_total padding, and pixel clock based
off of the original OVT timing and DSC calculations
-Store these values in the pipe and program HW with these modifications
-Added general support for cases where DSC slice config does not evenly
split the horizontal active by fixing some slice width calculations
-Updated PPS calculations for these cases

Reviewed-by: Chris Park <chris.park@amd.com>
Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Relja Vojvodic <rvojvodi@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/dc/core/dc_resource.c | 35 +++++++---------
 .../gpu/drm/amd/display/dc/dc_spl_translate.c |  2 +-
 .../dc/dml2/dml21/dml21_translation_helper.c  | 41 ++++++++-----------
 .../drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c  |  3 +-
 .../amd/display/dc/hwss/dcn32/dcn32_hwseq.c   |  2 +-
 .../amd/display/dc/hwss/dcn401/dcn401_hwseq.c |  9 ++--
 .../gpu/drm/amd/display/dc/inc/core_types.h   | 10 ++++-
 .../gpu/drm/amd/display/dc/link/link_dpms.c   |  2 +-
 .../dc/resource/dcn32/dcn32_resource.c        |  2 +-
 .../dc/resource/dcn401/dcn401_resource.c      |  3 ++
 10 files changed, 55 insertions(+), 54 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index 89c805457ed8..cbca3c67f439 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -95,7 +95,6 @@
 #define DC_LOGGER \
 	dc->ctx->logger
 #define DC_LOGGER_INIT(logger)
-
 #include "dml2/dml2_wrapper.h"
 
 #define UNABLE_TO_SPLIT -1
@@ -2149,7 +2148,7 @@ int resource_get_odm_slice_dst_width(struct pipe_ctx *otg_master,
 	h_active = timing->h_addressable +
 			timing->h_border_left +
 			timing->h_border_right +
-			otg_master->hblank_borrow;
+			otg_master->dsc_padding_params.dsc_hactive_padding;
 	width = h_active / count;
 
 	if (otg_master->stream_res.tg)
@@ -4267,39 +4266,33 @@ fail:
 	return res;
 }
 
+#if defined(CONFIG_DRM_AMD_DC_FP)
+#endif /* CONFIG_DRM_AMD_DC_FP */
+
 /**
- * decide_hblank_borrow - Decides the horizontal blanking borrow value for a given pipe context.
+ * calculate_timing_params_for_dsc_with_padding - Calculates timing parameters for DSC with padding.
  * @pipe_ctx: Pointer to the pipe context structure.
  *
- * This function calculates the horizontal blanking borrow value for a given pipe context based on the
+ * This function calculates the timing parameters for a given pipe context based on the
  * display stream compression (DSC) configuration. If the horizontal active pixels (hactive) are less
- * than the total width of the DSC slices, it sets the hblank_borrow value to the difference. If the
- * total horizontal timing minus the hblank_borrow value is less than 32, it resets the hblank_borrow
+ * than the total width of the DSC slices, it sets the dsc_hactive_padding value to the difference. If the
+ * total horizontal timing minus the dsc_hactive_padding value is less than 32, it resets the dsc_hactive_padding
  * value to 0.
  */
-static void decide_hblank_borrow(struct pipe_ctx *pipe_ctx)
+static void calculate_timing_params_for_dsc_with_padding(struct pipe_ctx *pipe_ctx)
 {
-	uint32_t hactive;
-	uint32_t ceil_slice_width;
 	struct dc_stream_state *stream = NULL;
 
 	if (!pipe_ctx)
 		return;
 
 	stream = pipe_ctx->stream;
+	pipe_ctx->dsc_padding_params.dsc_hactive_padding = 0;
+	pipe_ctx->dsc_padding_params.dsc_htotal_padding = 0;
 
-	if (stream->timing.flags.DSC) {
-		hactive = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right;
+	if (stream)
+		pipe_ctx->dsc_padding_params.dsc_pix_clk_100hz = stream->timing.pix_clk_100hz;
 
-		/* Assume if determined slices does not divide Hactive evenly, Hborrow is needed for padding*/
-		if (hactive % stream->timing.dsc_cfg.num_slices_h != 0) {
-			ceil_slice_width = (hactive / stream->timing.dsc_cfg.num_slices_h) + 1;
-			pipe_ctx->hblank_borrow = ceil_slice_width * stream->timing.dsc_cfg.num_slices_h - hactive;
-
-			if (stream->timing.h_total - hactive - pipe_ctx->hblank_borrow < 32)
-				pipe_ctx->hblank_borrow = 0;
-		}
-	}
 }
 
 /**
@@ -4342,7 +4335,7 @@ enum dc_status dc_validate_global_state(
 
 			/* Decide whether hblank borrow is needed and save it in pipe_ctx */
 			if (dc->debug.enable_hblank_borrow)
-				decide_hblank_borrow(pipe_ctx);
+				calculate_timing_params_for_dsc_with_padding(pipe_ctx);
 
 			if (dc->res_pool->funcs->patch_unknown_plane_state &&
 					pipe_ctx->plane_state &&
diff --git a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c
index 7f57661433eb..55704d4457ef 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c
@@ -128,7 +128,7 @@ void translate_SPL_in_params_from_pipe_ctx(struct pipe_ctx *pipe_ctx, struct spl
 	spl_in->odm_slice_index = resource_get_odm_slice_index(pipe_ctx);
 	// Make spl input basic out info output_size width point to stream h active
 	spl_in->basic_out.output_size.width =
-		stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right + pipe_ctx->hblank_borrow;
+		stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right + pipe_ctx->dsc_padding_params.dsc_hactive_padding;
 	// Make spl input basic out info output_size height point to v active
 	spl_in->basic_out.output_size.height =
 		stream->timing.v_addressable + stream->timing.v_border_bottom + stream->timing.v_border_top;
diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
index f6879e622271..bf5e7f4e0416 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c
@@ -84,25 +84,29 @@ static unsigned int calc_max_hardware_v_total(const struct dc_stream_state *stre
 
 static void populate_dml21_timing_config_from_stream_state(struct dml2_timing_cfg *timing,
 		struct dc_stream_state *stream,
+		struct pipe_ctx *pipe_ctx,
 		struct dml2_context *dml_ctx)
 {
 	unsigned int hblank_start, vblank_start, min_hardware_refresh_in_uhz;
+	uint32_t pix_clk_100hz;
 
-	timing->h_active = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right;
+	timing->h_active = stream->timing.h_addressable + stream->timing.h_border_left + stream->timing.h_border_right + pipe_ctx->dsc_padding_params.dsc_hactive_padding;
 	timing->v_active = stream->timing.v_addressable + stream->timing.v_border_bottom + stream->timing.v_border_top;
 	timing->h_front_porch = stream->timing.h_front_porch;
 	timing->v_front_porch = stream->timing.v_front_porch;
 	timing->pixel_clock_khz = stream->timing.pix_clk_100hz / 10;
+	if (pipe_ctx->dsc_padding_params.dsc_hactive_padding != 0)
+		timing->pixel_clock_khz = pipe_ctx->dsc_padding_params.dsc_pix_clk_100hz / 10;
 	if (stream->timing.timing_3d_format == TIMING_3D_FORMAT_HW_FRAME_PACKING)
 		timing->pixel_clock_khz *= 2;
-	timing->h_total = stream->timing.h_total;
+	timing->h_total = stream->timing.h_total + pipe_ctx->dsc_padding_params.dsc_htotal_padding;
 	timing->v_total = stream->timing.v_total;
 	timing->h_sync_width = stream->timing.h_sync_width;
 	timing->interlaced = stream->timing.flags.INTERLACE;
 
 	hblank_start = stream->timing.h_total - stream->timing.h_front_porch;
 
-	timing->h_blank_end = hblank_start - stream->timing.h_addressable
+	timing->h_blank_end = hblank_start - stream->timing.h_addressable - pipe_ctx->dsc_padding_params.dsc_hactive_padding
 		- stream->timing.h_border_left - stream->timing.h_border_right;
 
 	if (hblank_start < stream->timing.h_addressable)
@@ -121,8 +125,13 @@ static void populate_dml21_timing_config_from_stream_state(struct dml2_timing_cf
 	/* limit min refresh rate to DC cap */
 	min_hardware_refresh_in_uhz = stream->timing.min_refresh_in_uhz;
 	if (stream->ctx->dc->caps.max_v_total != 0) {
-		min_hardware_refresh_in_uhz = div64_u64((stream->timing.pix_clk_100hz * 100000000ULL),
-				(stream->timing.h_total * (long long)calc_max_hardware_v_total(stream)));
+		if (pipe_ctx->dsc_padding_params.dsc_hactive_padding != 0) {
+			pix_clk_100hz = pipe_ctx->dsc_padding_params.dsc_pix_clk_100hz;
+		} else {
+			pix_clk_100hz = stream->timing.pix_clk_100hz;
+		}
+		min_hardware_refresh_in_uhz = div64_u64((pix_clk_100hz * 100000000ULL),
+				(timing->h_total * (long long)calc_max_hardware_v_total(stream)));
 	}
 
 	timing->drr_config.min_refresh_uhz = max(stream->timing.min_refresh_in_uhz, min_hardware_refresh_in_uhz);
@@ -173,21 +182,6 @@ static void populate_dml21_timing_config_from_stream_state(struct dml2_timing_cf
 	timing->vblank_nom = timing->v_total - timing->v_active;
 }
 
-/**
- * adjust_dml21_hblank_timing_config_from_pipe_ctx - Adjusts the horizontal blanking timing configuration
- *                                                   based on the pipe context.
- * @timing: Pointer to the dml2_timing_cfg structure to be adjusted.
- * @pipe: Pointer to the pipe_ctx structure containing the horizontal blanking borrow value.
- *
- * This function modifies the horizontal active and blank end timings by adding and subtracting
- * the horizontal blanking borrow value from the pipe context, respectively.
- */
-static void adjust_dml21_hblank_timing_config_from_pipe_ctx(struct dml2_timing_cfg *timing, struct pipe_ctx *pipe)
-{
-	timing->h_active += pipe->hblank_borrow;
-	timing->h_blank_end -= pipe->hblank_borrow;
-}
-
 static void populate_dml21_output_config_from_stream_state(struct dml2_link_output_cfg *output,
 		struct dc_stream_state *stream, const struct pipe_ctx *pipe)
 {
@@ -487,7 +481,9 @@ static const struct scaler_data *get_scaler_data_for_plane(
 			temp_pipe->plane_state = pipe->plane_state;
 			temp_pipe->plane_res.scl_data.taps = pipe->plane_res.scl_data.taps;
 			temp_pipe->stream_res = pipe->stream_res;
-			temp_pipe->hblank_borrow = pipe->hblank_borrow;
+			temp_pipe->dsc_padding_params.dsc_hactive_padding = pipe->dsc_padding_params.dsc_hactive_padding;
+			temp_pipe->dsc_padding_params.dsc_htotal_padding = pipe->dsc_padding_params.dsc_htotal_padding;
+			temp_pipe->dsc_padding_params.dsc_pix_clk_100hz = pipe->dsc_padding_params.dsc_pix_clk_100hz;
 			dml_ctx->config.callbacks.build_scaling_params(temp_pipe);
 			break;
 		}
@@ -755,8 +751,7 @@ bool dml21_map_dc_state_into_dml_display_cfg(const struct dc *in_dc, struct dc_s
 			disp_cfg_stream_location = dml_dispcfg->num_streams++;
 
 		ASSERT(disp_cfg_stream_location >= 0 && disp_cfg_stream_location < __DML2_WRAPPER_MAX_STREAMS_PLANES__);
-		populate_dml21_timing_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, context->streams[stream_index], dml_ctx);
-		adjust_dml21_hblank_timing_config_from_pipe_ctx(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, &context->res_ctx.pipe_ctx[stream_index]);
+		populate_dml21_timing_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].timing, context->streams[stream_index], &context->res_ctx.pipe_ctx[stream_index], dml_ctx);
 		populate_dml21_output_config_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location].output, context->streams[stream_index], &context->res_ctx.pipe_ctx[stream_index]);
 		populate_dml21_stream_overrides_from_stream_state(&dml_dispcfg->stream_descriptors[disp_cfg_stream_location], context->streams[stream_index], &context->stream_status[stream_index]);
 
diff --git a/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c b/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c
index bd1b9aef6d5c..89f0d999bf35 100644
--- a/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c
+++ b/drivers/gpu/drm/amd/display/dc/dsc/dcn20/dcn20_dsc.c
@@ -406,9 +406,10 @@ bool dsc_prepare_config(const struct dsc_config *dsc_cfg, struct dsc_reg_values
 	dsc_reg_vals->alternate_ich_encoding_en = dsc_reg_vals->pps.dsc_version_minor == 1 ? 0 : 1;
 	dsc_reg_vals->ich_reset_at_eol = (dsc_cfg->is_odm || dsc_reg_vals->num_slices_h > 1) ? 0xF : 0;
 
+	// Need to find the ceiling value for the slice width
+	dsc_reg_vals->pps.slice_width = (dsc_cfg->pic_width + dsc_cfg->dc_dsc_cfg.num_slices_h - 1) / dsc_cfg->dc_dsc_cfg.num_slices_h;
 	// TODO: in addition to validating slice height (pic height must be divisible by slice height),
 	// see what happens when the same condition doesn't apply for slice_width/pic_width.
-	dsc_reg_vals->pps.slice_width = dsc_cfg->pic_width / dsc_cfg->dc_dsc_cfg.num_slices_h;
 	dsc_reg_vals->pps.slice_height = dsc_cfg->pic_height / dsc_cfg->dc_dsc_cfg.num_slices_v;
 
 	ASSERT(dsc_reg_vals->pps.slice_height * dsc_cfg->dc_dsc_cfg.num_slices_v == dsc_cfg->pic_height);
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
index 3a6f946643ef..f39292952702 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c
@@ -1052,7 +1052,7 @@ void dcn32_update_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 		}
 
 		/* Enable DSC hw block */
-		dsc_cfg.pic_width = (stream->timing.h_addressable + pipe_ctx->hblank_borrow +
+		dsc_cfg.pic_width = (stream->timing.h_addressable + pipe_ctx->dsc_padding_params.dsc_hactive_padding +
 				stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt;
 		dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom;
 		dsc_cfg.pixel_encoding = stream->timing.pixel_encoding;
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index 9956145e3d18..1b0b772fc5dd 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -810,9 +810,12 @@ enum dc_status dcn401_enable_stream_timing(
 	if (dc->hwseq->funcs.PLAT_58856_wa && (!dc_is_dp_signal(stream->signal)))
 		dc->hwseq->funcs.PLAT_58856_wa(context, pipe_ctx);
 
-	/* if we are borrowing from hblank, h_addressable needs to be adjusted */
-	if (dc->debug.enable_hblank_borrow)
-		patched_crtc_timing.h_addressable = patched_crtc_timing.h_addressable + pipe_ctx->hblank_borrow;
+	/* if we are padding, h_addressable needs to be adjusted */
+	if (dc->debug.enable_hblank_borrow) {
+		patched_crtc_timing.h_addressable = patched_crtc_timing.h_addressable + pipe_ctx->dsc_padding_params.dsc_hactive_padding;
+		patched_crtc_timing.h_total = patched_crtc_timing.h_total + pipe_ctx->dsc_padding_params.dsc_htotal_padding;
+		patched_crtc_timing.pix_clk_100hz = pipe_ctx->dsc_padding_params.dsc_pix_clk_100hz;
+	}
 
 	pipe_ctx->stream_res.tg->funcs->program_timing(
 		pipe_ctx->stream_res.tg,
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index d30f94c35f11..f49549147d99 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -437,6 +437,13 @@ enum p_state_switch_method {
 	P_STATE_V_BLANK_SUB_VP,
 };
 
+struct dsc_padding_params {
+	/* pixels borrowed from hblank to hactive */
+	uint8_t dsc_hactive_padding;
+	uint32_t dsc_htotal_padding;
+	uint32_t dsc_pix_clk_100hz;
+};
+
 struct pipe_ctx {
 	struct dc_plane_state *plane_state;
 	struct dc_stream_state *stream;
@@ -494,8 +501,7 @@ struct pipe_ctx {
 	/* subvp_index: only valid if the pipe is a SUBVP_MAIN*/
 	uint8_t subvp_index;
 	struct pixel_rate_divider pixel_rate_divider;
-	/* pixels borrowed from hblank to hactive */
-	uint8_t hblank_borrow;
+	struct dsc_padding_params dsc_padding_params;
 	/* next vupdate */
 	uint32_t next_vupdate;
 	uint32_t wait_frame_count;
diff --git a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
index 08ee8d2f777b..83419e1a9036 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_dpms.c
@@ -832,7 +832,7 @@ void link_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable)
 		enum optc_dsc_mode optc_dsc_mode;
 
 		/* Enable DSC hw block */
-		dsc_cfg.pic_width = (stream->timing.h_addressable + pipe_ctx->hblank_borrow +
+		dsc_cfg.pic_width = (stream->timing.h_addressable + pipe_ctx->dsc_padding_params.dsc_hactive_padding +
 				stream->timing.h_border_left + stream->timing.h_border_right) / opp_cnt;
 		dsc_cfg.pic_height = stream->timing.v_addressable + stream->timing.v_border_top + stream->timing.v_border_bottom;
 		dsc_cfg.pixel_encoding = stream->timing.pixel_encoding;
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
index 4acd97b3e4c9..8f80ccb846d7 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
@@ -2852,7 +2852,7 @@ struct pipe_ctx *dcn32_acquire_free_pipe_as_secondary_opp_head(
 		free_pipe->plane_res.xfm = pool->transforms[free_pipe_idx];
 		free_pipe->plane_res.dpp = pool->dpps[free_pipe_idx];
 		free_pipe->plane_res.mpcc_inst = pool->dpps[free_pipe_idx]->inst;
-		free_pipe->hblank_borrow = otg_master->hblank_borrow;
+		free_pipe->dsc_padding_params = otg_master->dsc_padding_params;
 		if (free_pipe->stream->timing.flags.DSC == 1) {
 			dcn20_acquire_dsc(free_pipe->stream->ctx->dc,
 					&new_ctx->res_ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
index d498c0983d74..1d18807e4749 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c
@@ -1699,6 +1699,9 @@ static void dcn401_build_pipe_pix_clk_params(struct pipe_ctx *pipe_ctx)
 
 	pixel_clk_params->requested_pix_clk_100hz = stream->timing.pix_clk_100hz;
 
+	if (pipe_ctx->dsc_padding_params.dsc_hactive_padding != 0)
+		pixel_clk_params->requested_pix_clk_100hz = pipe_ctx->dsc_padding_params.dsc_pix_clk_100hz;
+
 	if (!pipe_ctx->stream->ctx->dc->config.unify_link_enc_assignment)
 		link_enc = link_enc_cfg_get_link_enc(link);
 	if (link_enc)

From c062decc5e9b62afffe5ee9e44ce1f35962a750e Mon Sep 17 00:00:00 2001
From: Martin Leung <Martin.Leung@amd.com>
Date: Wed, 6 Aug 2025 17:40:55 -0400
Subject: [PATCH 1304/1455] Revert "drm/amd/display: Reduce Stack Usage by
 moving 'audio_output' into 'stream_res' v4"

This reverts commit 1cf1205ef268 ("drm/amd/display: Reduce Stack Usage by moving 'audio_output' into 'stream_res' v4")

Reason for revert: Causes DP compliance errors

Acked-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Martin Leung <Martin.Leung@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/hwss/dce110/dce110_hwseq.c | 32 +++++++++++--------
 .../gpu/drm/amd/display/dc/inc/core_types.h   |  5 ++-
 .../display/dc/link/accessories/link_dp_cts.c | 13 ++++----
 .../dc/resource/dcn31/dcn31_resource.c        |  5 +--
 .../dc/resource/dcn31/dcn31_resource.h        |  3 +-
 5 files changed, 31 insertions(+), 27 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index ea904d52af20..7b4a04704312 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -1601,17 +1601,19 @@ enum dc_status dce110_apply_single_controller_ctx_to_hw(
 	}
 
 	if (pipe_ctx->stream_res.audio != NULL) {
-		build_audio_output(context, pipe_ctx, &pipe_ctx->stream_res.audio_output);
+		struct audio_output audio_output = {0};
 
-		link_hwss->setup_audio_output(pipe_ctx, &pipe_ctx->stream_res.audio_output,
+		build_audio_output(context, pipe_ctx, &audio_output);
+
+		link_hwss->setup_audio_output(pipe_ctx, &audio_output,
 				pipe_ctx->stream_res.audio->inst);
 
 		pipe_ctx->stream_res.audio->funcs->az_configure(
 				pipe_ctx->stream_res.audio,
 				pipe_ctx->stream->signal,
-				&pipe_ctx->stream_res.audio_output.crtc_info,
+				&audio_output.crtc_info,
 				&pipe_ctx->stream->audio_info,
-				&pipe_ctx->stream_res.audio_output.dp_link_info);
+				&audio_output.dp_link_info);
 
 		if (dc->config.disable_hbr_audio_dp2)
 			if (pipe_ctx->stream_res.audio->funcs->az_disable_hbr_audio &&
@@ -2385,7 +2387,9 @@ static void dce110_setup_audio_dto(
 		if (pipe_ctx->stream->signal != SIGNAL_TYPE_HDMI_TYPE_A)
 			continue;
 		if (pipe_ctx->stream_res.audio != NULL) {
-			build_audio_output(context, pipe_ctx, &pipe_ctx->stream_res.audio_output);
+			struct audio_output audio_output;
+
+			build_audio_output(context, pipe_ctx, &audio_output);
 
 			if (dc->res_pool->dccg && dc->res_pool->dccg->funcs->set_audio_dtbclk_dto) {
 				struct dtbclk_dto_params dto_params = {0};
@@ -2396,14 +2400,14 @@ static void dce110_setup_audio_dto(
 				pipe_ctx->stream_res.audio->funcs->wall_dto_setup(
 						pipe_ctx->stream_res.audio,
 						pipe_ctx->stream->signal,
-						&pipe_ctx->stream_res.audio_output.crtc_info,
-						&pipe_ctx->stream_res.audio_output.pll_info);
+						&audio_output.crtc_info,
+						&audio_output.pll_info);
 			} else
 				pipe_ctx->stream_res.audio->funcs->wall_dto_setup(
 					pipe_ctx->stream_res.audio,
 					pipe_ctx->stream->signal,
-					&pipe_ctx->stream_res.audio_output.crtc_info,
-					&pipe_ctx->stream_res.audio_output.pll_info);
+					&audio_output.crtc_info,
+					&audio_output.pll_info);
 			break;
 		}
 	}
@@ -2423,15 +2427,15 @@ static void dce110_setup_audio_dto(
 				continue;
 
 			if (pipe_ctx->stream_res.audio != NULL) {
-				build_audio_output(context,
-						   pipe_ctx,
-						   &pipe_ctx->stream_res.audio_output);
+				struct audio_output audio_output = {0};
+
+				build_audio_output(context, pipe_ctx, &audio_output);
 
 				pipe_ctx->stream_res.audio->funcs->wall_dto_setup(
 					pipe_ctx->stream_res.audio,
 					pipe_ctx->stream->signal,
-					&pipe_ctx->stream_res.audio_output.crtc_info,
-					&pipe_ctx->stream_res.audio_output.pll_info);
+					&audio_output.crtc_info,
+					&audio_output.pll_info);
 				break;
 			}
 		}
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index f49549147d99..d11893f8c916 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -228,7 +228,8 @@ struct resource_funcs {
 	enum dc_status (*update_dc_state_for_encoder_switch)(struct dc_link *link,
 		struct dc_link_settings *link_setting,
 		uint8_t pipe_count,
-		struct pipe_ctx *pipes);
+		struct pipe_ctx *pipes,
+		struct audio_output *audio_output);
 };
 
 struct audio_support{
@@ -360,8 +361,6 @@ struct stream_resource {
 	uint8_t gsl_group;
 
 	struct test_pattern_params test_pattern_params;
-
-	struct audio_output audio_output;
 };
 
 struct plane_resource {
diff --git a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
index 23f41c99fa38..9e33bf937a69 100644
--- a/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
+++ b/drivers/gpu/drm/amd/display/dc/link/accessories/link_dp_cts.c
@@ -75,7 +75,7 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 	bool is_hpo_acquired;
 	uint8_t count;
 	int i;
-
+	struct audio_output audio_output[MAX_PIPES];
 	struct dc_stream_state *streams_on_link[MAX_PIPES];
 	int num_streams_on_link = 0;
 
@@ -101,7 +101,7 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 	if (needs_divider_update && link->dc->res_pool->funcs->update_dc_state_for_encoder_switch) {
 		link->dc->res_pool->funcs->update_dc_state_for_encoder_switch(link,
 				link_setting, count,
-				*pipes);
+				*pipes, &audio_output[0]);
 		for (i = 0; i < count; i++) {
 			pipes[i]->clock_source->funcs->program_pix_clk(
 					pipes[i]->clock_source,
@@ -113,16 +113,15 @@ static void dp_retrain_link_dp_test(struct dc_link *link,
 				const struct link_hwss *link_hwss = get_link_hwss(
 					link, &pipes[i]->link_res);
 
-				link_hwss->setup_audio_output(pipes[i],
-							      &pipes[i]->stream_res.audio_output,
-							      pipes[i]->stream_res.audio->inst);
+				link_hwss->setup_audio_output(pipes[i], &audio_output[i],
+						pipes[i]->stream_res.audio->inst);
 
 				pipes[i]->stream_res.audio->funcs->az_configure(
 						pipes[i]->stream_res.audio,
 						pipes[i]->stream->signal,
-						&pipes[i]->stream_res.audio_output.crtc_info,
+						&audio_output[i].crtc_info,
 						&pipes[i]->stream->audio_info,
-						&pipes[i]->stream_res.audio_output.dp_link_info);
+						&audio_output[i].dp_link_info);
 
 				if (link->dc->config.disable_hbr_audio_dp2 &&
 						pipes[i]->stream_res.audio->funcs->az_disable_hbr_audio &&
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
index ca17e5d8fdc2..3ed7f50554e2 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c
@@ -2239,7 +2239,8 @@ struct resource_pool *dcn31_create_resource_pool(
 enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link,
 	struct dc_link_settings *link_setting,
 	uint8_t pipe_count,
-	struct pipe_ctx *pipes)
+	struct pipe_ctx *pipes,
+	struct audio_output *audio_output)
 {
 	struct dc_state *state = link->dc->current_state;
 	int i;
@@ -2254,7 +2255,7 @@ enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link,
 
 		// Setup audio
 		if (pipes[i].stream_res.audio != NULL)
-			build_audio_output(state, &pipes[i], &pipes[i].stream_res.audio_output);
+			build_audio_output(state, &pipes[i], &audio_output[i]);
 	}
 #else
 	/* This DCN requires rate divider updates and audio reprogramming to allow DP1<-->DP2 link rate switching,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
index 7e8fde65528f..c32c85ef0ba4 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.h
@@ -69,7 +69,8 @@ unsigned int dcn31_get_det_buffer_size(
 enum dc_status dcn31_update_dc_state_for_encoder_switch(struct dc_link *link,
 	struct dc_link_settings *link_setting,
 	uint8_t pipe_count,
-	struct pipe_ctx *pipes);
+	struct pipe_ctx *pipes,
+	struct audio_output *audio_output);
 
 /*temp: B0 specific before switch to dcn313 headers*/
 #ifndef regPHYPLLF_PIXCLK_RESYNC_CNTL

From d7d0008ece3785ab19412be07d12f798c0feb5b1 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Sun, 7 Sep 2025 18:55:39 -0400
Subject: [PATCH 1305/1455] drm/amd/display: Promote DC to 3.2.350

This version brings along following updates:
- Add DSC padding for OVT support
- Setup pixel encoding for YCBCR422
- Fix dml ms order
- Rename header file link.h to link_service.h
- Fix DMUB loading sequence
- Modify link training policy

Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Acked-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index aa808675fead..8bee44cd4609 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.349"
+#define DC_VER "3.2.350"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 7a1e3cfe8c8aed4cad7097c13d68ea2dbe4b001d Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Thu, 4 Sep 2025 09:48:00 +0800
Subject: [PATCH 1306/1455] drm/amd/pm: unified smu feature cap for link reset

unified link reset smu feature cap

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c            |  7 +------
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h        |  6 +-----
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 10 +++++++++-
 3 files changed, 11 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index c9ba9625e85e..3a9b51014c7b 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -3536,15 +3536,10 @@ bool smu_mode1_reset_is_support(struct smu_context *smu)
 
 bool smu_link_reset_is_support(struct smu_context *smu)
 {
-	bool ret = false;
-
 	if (!smu->pm_enabled)
 		return false;
 
-	if (smu->ppt_funcs && smu->ppt_funcs->link_reset_is_support)
-		ret = smu->ppt_funcs->link_reset_is_support(smu);
-
-	return ret;
+	return smu_feature_cap_test(smu, SMU_FEATURE_CAP_ID__LINK_RESET);
 }
 
 int smu_mode1_reset(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index ba1bab4b7036..3fc6daf6662c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -529,6 +529,7 @@ enum smu_fw_status {
 #define SMU_WBRF_EVENT_HANDLING_PACE	10
 
 enum smu_feature_cap_id {
+	SMU_FEATURE_CAP_ID__LINK_RESET = 0,
 	SMU_FEATURE_CAP_ID__COUNT,
 };
 
@@ -1281,11 +1282,6 @@ struct pptable_funcs {
 	 */
 	bool (*mode1_reset_is_support)(struct smu_context *smu);
 
-	/**
-	 * @link_reset_is_support: Check if GPU supports link reset.
-	 */
-	bool (*link_reset_is_support)(struct smu_context *smu);
-
 	/**
 	 * @mode1_reset: Perform mode1 reset.
 	 *
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index cbe1f314917a..be372f5e76fd 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3227,6 +3227,14 @@ static int smu_v13_0_6_reset_vcn(struct smu_context *smu, uint32_t inst_mask)
 }
 
 
+static int smu_v13_0_6_post_init(struct smu_context *smu)
+{
+	if (smu_v13_0_6_is_link_reset_supported(smu))
+		smu_feature_cap_set(smu, SMU_FEATURE_CAP_ID__LINK_RESET);
+
+	return 0;
+}
+
 static int mca_smu_set_debug_mode(struct amdgpu_device *adev, bool enable)
 {
 	struct smu_context *smu = adev->powerplay.pp_handle;
@@ -3896,7 +3904,6 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.get_xcp_metrics = smu_v13_0_6_get_xcp_metrics,
 	.get_thermal_temperature_range = smu_v13_0_6_get_thermal_temperature_range,
 	.mode1_reset_is_support = smu_v13_0_6_is_mode1_reset_supported,
-	.link_reset_is_support = smu_v13_0_6_is_link_reset_supported,
 	.mode1_reset = smu_v13_0_6_mode1_reset,
 	.mode2_reset = smu_v13_0_6_mode2_reset,
 	.link_reset = smu_v13_0_6_link_reset,
@@ -3909,6 +3916,7 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.reset_sdma_is_supported = smu_v13_0_6_reset_sdma_is_supported,
 	.dpm_reset_vcn = smu_v13_0_6_reset_vcn,
 	.reset_vcn_is_supported = smu_v13_0_6_reset_vcn_is_supported,
+	.post_init = smu_v13_0_6_post_init,
 };
 
 void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)

From 8dbd641f8e0be8a34c755da518ab9aabea285c5f Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Thu, 4 Sep 2025 09:53:43 +0800
Subject: [PATCH 1307/1455] drm/amd/pm: unified smu feature cap for sdma reset

unified sdma reset smu feature cap

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c            | 7 +------
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h        | 5 +----
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 4 +++-
 3 files changed, 5 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 3a9b51014c7b..c4ce056a1613 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -4129,12 +4129,7 @@ int smu_send_rma_reason(struct smu_context *smu)
  */
 bool smu_reset_sdma_is_supported(struct smu_context *smu)
 {
-	bool ret = false;
-
-	if (smu->ppt_funcs && smu->ppt_funcs->reset_sdma_is_supported)
-		ret = smu->ppt_funcs->reset_sdma_is_supported(smu);
-
-	return ret;
+	return smu_feature_cap_test(smu, SMU_FEATURE_CAP_ID__SDMA_RESET);
 }
 
 int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 3fc6daf6662c..5f07a04fdebc 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -530,6 +530,7 @@ enum smu_fw_status {
 
 enum smu_feature_cap_id {
 	SMU_FEATURE_CAP_ID__LINK_RESET = 0,
+	SMU_FEATURE_CAP_ID__SDMA_RESET,
 	SMU_FEATURE_CAP_ID__COUNT,
 };
 
@@ -1432,10 +1433,6 @@ struct pptable_funcs {
 	 * @reset_sdma: message SMU to soft reset sdma instance.
 	 */
 	int (*reset_sdma)(struct smu_context *smu, uint32_t inst_mask);
-	/**
-	 * @reset_sdma_is_supported: Check if support resets the SDMA engine.
-	 */
-	bool (*reset_sdma_is_supported)(struct smu_context *smu);
 
 	/**
 	 * @reset_vcn: message SMU to soft reset vcn instance.
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index be372f5e76fd..73defde12e8a 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3232,6 +3232,9 @@ static int smu_v13_0_6_post_init(struct smu_context *smu)
 	if (smu_v13_0_6_is_link_reset_supported(smu))
 		smu_feature_cap_set(smu, SMU_FEATURE_CAP_ID__LINK_RESET);
 
+	if (smu_v13_0_6_reset_sdma_is_supported(smu))
+		smu_feature_cap_set(smu, SMU_FEATURE_CAP_ID__SDMA_RESET);
+
 	return 0;
 }
 
@@ -3913,7 +3916,6 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.send_hbm_bad_pages_num = smu_v13_0_6_smu_send_hbm_bad_page_num,
 	.send_rma_reason = smu_v13_0_6_send_rma_reason,
 	.reset_sdma = smu_v13_0_6_reset_sdma,
-	.reset_sdma_is_supported = smu_v13_0_6_reset_sdma_is_supported,
 	.dpm_reset_vcn = smu_v13_0_6_reset_vcn,
 	.reset_vcn_is_supported = smu_v13_0_6_reset_vcn_is_supported,
 	.post_init = smu_v13_0_6_post_init,

From 28af436a8fa69d4408eadc1beb84528cc216b30b Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Thu, 4 Sep 2025 09:57:25 +0800
Subject: [PATCH 1308/1455] drm/amd/pm: unified smu feature cap for vcn reset

unified vcn reset smu feature cap

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c            | 7 +------
 drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h        | 5 +----
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 4 +++-
 3 files changed, 5 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index c4ce056a1613..715579950dd8 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -4144,12 +4144,7 @@ int smu_reset_sdma(struct smu_context *smu, uint32_t inst_mask)
 
 bool smu_reset_vcn_is_supported(struct smu_context *smu)
 {
-	bool ret = false;
-
-	if (smu->ppt_funcs && smu->ppt_funcs->reset_vcn_is_supported)
-		ret = smu->ppt_funcs->reset_vcn_is_supported(smu);
-
-	return ret;
+	return smu_feature_cap_test(smu, SMU_FEATURE_CAP_ID__VCN_RESET);
 }
 
 int smu_reset_vcn(struct smu_context *smu, uint32_t inst_mask)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
index 5f07a04fdebc..582c186d8b62 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h
@@ -531,6 +531,7 @@ enum smu_fw_status {
 enum smu_feature_cap_id {
 	SMU_FEATURE_CAP_ID__LINK_RESET = 0,
 	SMU_FEATURE_CAP_ID__SDMA_RESET,
+	SMU_FEATURE_CAP_ID__VCN_RESET,
 	SMU_FEATURE_CAP_ID__COUNT,
 };
 
@@ -1438,10 +1439,6 @@ struct pptable_funcs {
 	 * @reset_vcn: message SMU to soft reset vcn instance.
 	 */
 	int (*dpm_reset_vcn)(struct smu_context *smu, uint32_t inst_mask);
-	/**
-	 * @reset_vcn_is_supported: Check if support resets vcn.
-	 */
-	bool (*reset_vcn_is_supported)(struct smu_context *smu);
 
 	/**
 	 * @get_ecc_table:  message SMU to get ECC INFO table.
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 73defde12e8a..bcd6525ca5e9 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3235,6 +3235,9 @@ static int smu_v13_0_6_post_init(struct smu_context *smu)
 	if (smu_v13_0_6_reset_sdma_is_supported(smu))
 		smu_feature_cap_set(smu, SMU_FEATURE_CAP_ID__SDMA_RESET);
 
+	if (smu_v13_0_6_reset_vcn_is_supported(smu))
+		smu_feature_cap_set(smu, SMU_FEATURE_CAP_ID__VCN_RESET);
+
 	return 0;
 }
 
@@ -3917,7 +3920,6 @@ static const struct pptable_funcs smu_v13_0_6_ppt_funcs = {
 	.send_rma_reason = smu_v13_0_6_send_rma_reason,
 	.reset_sdma = smu_v13_0_6_reset_sdma,
 	.dpm_reset_vcn = smu_v13_0_6_reset_vcn,
-	.reset_vcn_is_supported = smu_v13_0_6_reset_vcn_is_supported,
 	.post_init = smu_v13_0_6_post_init,
 };
 

From 5cefcbb306d6b4fa3ac5fba86db94cac81f0b457 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Thu, 4 Sep 2025 16:30:08 +0800
Subject: [PATCH 1309/1455] drm/amdgpu: adjust MES API used for suspend and
 resume

Use the suspend and resume API rather than remove queue
and add queue API.  The former just preempts the queue
while the latter remove it from the scheduler completely.
There is no need to do that, we only need preemption
in this case.

V2: replace queue_active with queue state
v3: set the suspend_fence_addr
v4: allocate another per queue buffer for the suspend fence, and  set the sequence number.
    also wait for the suspend fence. (Alex)
v5: use a wb slot (Alex)
v6: Change the timeout period. For MES, the default timeout  is  2100000; /* 2100 ms */ (Alex)

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 72 ++++++++++++++++++++++
 1 file changed, 72 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
index 3a4fd6de08ce..2db9b2c63693 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c
@@ -404,10 +404,82 @@ mes_userq_mqd_destroy(struct amdgpu_userq_mgr *uq_mgr,
 	amdgpu_userq_destroy_object(uq_mgr, &queue->mqd);
 }
 
+static int mes_userq_preempt(struct amdgpu_userq_mgr *uq_mgr,
+				struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct mes_suspend_gang_input queue_input;
+	struct amdgpu_userq_obj *ctx = &queue->fw_obj;
+	signed long timeout = 2100000; /* 2100 ms */
+	u64 fence_gpu_addr;
+	u32 fence_offset;
+	u64 *fence_ptr;
+	int i, r;
+
+	if (queue->state != AMDGPU_USERQ_STATE_MAPPED)
+		return 0;
+	r = amdgpu_device_wb_get(adev, &fence_offset);
+	if (r)
+		return r;
+
+	fence_gpu_addr = adev->wb.gpu_addr + (fence_offset * 4);
+	fence_ptr = (u64 *)&adev->wb.wb[fence_offset];
+	*fence_ptr = 0;
+
+	memset(&queue_input, 0x0, sizeof(struct mes_suspend_gang_input));
+	queue_input.gang_context_addr = ctx->gpu_addr + AMDGPU_USERQ_PROC_CTX_SZ;
+	queue_input.suspend_fence_addr = fence_gpu_addr;
+	queue_input.suspend_fence_value = 1;
+	amdgpu_mes_lock(&adev->mes);
+	r = adev->mes.funcs->suspend_gang(&adev->mes, &queue_input);
+	amdgpu_mes_unlock(&adev->mes);
+	if (r) {
+		DRM_ERROR("Failed to suspend gang: %d\n", r);
+		goto out;
+	}
+
+	for (i = 0; i < timeout; i++) {
+		if (*fence_ptr == 1)
+			goto out;
+		udelay(1);
+	}
+	r = -ETIMEDOUT;
+
+out:
+	amdgpu_device_wb_free(adev, fence_offset);
+	return r;
+}
+
+static int mes_userq_restore(struct amdgpu_userq_mgr *uq_mgr,
+				struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct mes_resume_gang_input queue_input;
+	struct amdgpu_userq_obj *ctx = &queue->fw_obj;
+	int r;
+
+	if (queue->state == AMDGPU_USERQ_STATE_HUNG)
+		return -EINVAL;
+	if (queue->state != AMDGPU_USERQ_STATE_PREEMPTED)
+		return 0;
+
+	memset(&queue_input, 0x0, sizeof(struct mes_resume_gang_input));
+	queue_input.gang_context_addr = ctx->gpu_addr + AMDGPU_USERQ_PROC_CTX_SZ;
+
+	amdgpu_mes_lock(&adev->mes);
+	r = adev->mes.funcs->resume_gang(&adev->mes, &queue_input);
+	amdgpu_mes_unlock(&adev->mes);
+	if (r)
+		dev_err(adev->dev, "Failed to resume queue, err (%d)\n", r);
+	return r;
+}
+
 const struct amdgpu_userq_funcs userq_mes_funcs = {
 	.mqd_create = mes_userq_mqd_create,
 	.mqd_destroy = mes_userq_mqd_destroy,
 	.unmap = mes_userq_unmap,
 	.map = mes_userq_map,
 	.detect_and_reset = mes_userq_detect_and_reset,
+	.preempt = mes_userq_preempt,
+	.restore = mes_userq_restore,
 };

From bb1d7f157e3d2357daf3146f7bcd03bcdbcefd25 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Tue, 5 Aug 2025 12:23:18 +0800
Subject: [PATCH 1310/1455] drm/amdgpu: Switch user queues to use
 preempt/restore for eviction

This patch modifies the user queue management to use preempt/restore
operations instead of full map/unmap for queue eviction scenarios where
applicable. The changes include:

1. Introduces new helper functions:
   - amdgpu_userqueue_preempt_helper()
   - amdgpu_userqueue_restore_helper()

2. Updates queue state management to track PREEMPTED state

3. Modifies eviction handling to use preempt instead of unmap:
   - amdgpu_userq_evict_all() now uses preempt_helper
   - amdgpu_userq_restore_all() now uses restore_helper

The preempt/restore approach provides better performance during queue
eviction by avoiding the overhead of full queue teardown and setup.
Full map/unmap operations are still used for initial setup/teardown
and system suspend scenarios.

v2: rename amdgpu_userqueue_restore_helper/amdgpu_userqueue_preempt_helper to
amdgpu_userq_restore_helper/amdgpu_userq_preempt_helper for consistency. (Alex)

v3: amdgpu_userq_stop_sched_for_enforce_isolation() and
amdgpu_userq_start_sched_for_enforce_isolation() should use preempt and restore (Alex)

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 53 ++++++++++++++++++++---
 1 file changed, 48 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index c3c1a714b06a..d2e59aecb3eb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -77,7 +77,7 @@ out_err:
 }
 
 static int
-amdgpu_userq_unmap_helper(struct amdgpu_userq_mgr *uq_mgr,
+amdgpu_userq_preempt_helper(struct amdgpu_userq_mgr *uq_mgr,
 			  struct amdgpu_usermode_queue *queue)
 {
 	struct amdgpu_device *adev = uq_mgr->adev;
@@ -86,6 +86,49 @@ amdgpu_userq_unmap_helper(struct amdgpu_userq_mgr *uq_mgr,
 	int r = 0;
 
 	if (queue->state == AMDGPU_USERQ_STATE_MAPPED) {
+		r = userq_funcs->preempt(uq_mgr, queue);
+		if (r) {
+			queue->state = AMDGPU_USERQ_STATE_HUNG;
+		} else {
+			queue->state = AMDGPU_USERQ_STATE_PREEMPTED;
+		}
+	}
+
+	return r;
+}
+
+static int
+amdgpu_userq_restore_helper(struct amdgpu_userq_mgr *uq_mgr,
+			struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	const struct amdgpu_userq_funcs *userq_funcs =
+		adev->userq_funcs[queue->queue_type];
+	int r = 0;
+
+	if (queue->state == AMDGPU_USERQ_STATE_PREEMPTED) {
+		r = userq_funcs->restore(uq_mgr, queue);
+		if (r) {
+			queue->state = AMDGPU_USERQ_STATE_HUNG;
+		} else {
+			queue->state = AMDGPU_USERQ_STATE_MAPPED;
+		}
+	}
+
+	return r;
+}
+
+static int
+amdgpu_userq_unmap_helper(struct amdgpu_userq_mgr *uq_mgr,
+			  struct amdgpu_usermode_queue *queue)
+{
+	struct amdgpu_device *adev = uq_mgr->adev;
+	const struct amdgpu_userq_funcs *userq_funcs =
+		adev->userq_funcs[queue->queue_type];
+	int r = 0;
+
+	if ((queue->state == AMDGPU_USERQ_STATE_MAPPED) ||
+		(queue->state == AMDGPU_USERQ_STATE_PREEMPTED)) {
 		r = userq_funcs->unmap(uq_mgr, queue);
 		if (r)
 			queue->state = AMDGPU_USERQ_STATE_HUNG;
@@ -653,7 +696,7 @@ amdgpu_userq_restore_all(struct amdgpu_userq_mgr *uq_mgr)
 
 	/* Resume all the queues for this process */
 	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id) {
-		r = amdgpu_userq_map_helper(uq_mgr, queue);
+		r = amdgpu_userq_restore_helper(uq_mgr, queue);
 		if (r)
 			ret = r;
 	}
@@ -810,7 +853,7 @@ amdgpu_userq_evict_all(struct amdgpu_userq_mgr *uq_mgr)
 
 	/* Try to unmap all the queues in this process ctx */
 	idr_for_each_entry(&uq_mgr->userq_idr, queue, queue_id) {
-		r = amdgpu_userq_unmap_helper(uq_mgr, queue);
+		r = amdgpu_userq_preempt_helper(uq_mgr, queue);
 		if (r)
 			ret = r;
 	}
@@ -995,7 +1038,7 @@ int amdgpu_userq_stop_sched_for_enforce_isolation(struct amdgpu_device *adev,
 			if (((queue->queue_type == AMDGPU_HW_IP_GFX) ||
 			     (queue->queue_type == AMDGPU_HW_IP_COMPUTE)) &&
 			    (queue->xcp_id == idx)) {
-				r = amdgpu_userq_unmap_helper(uqm, queue);
+				r = amdgpu_userq_preempt_helper(uqm, queue);
 				if (r)
 					ret = r;
 			}
@@ -1029,7 +1072,7 @@ int amdgpu_userq_start_sched_for_enforce_isolation(struct amdgpu_device *adev,
 			if (((queue->queue_type == AMDGPU_HW_IP_GFX) ||
 			     (queue->queue_type == AMDGPU_HW_IP_COMPUTE)) &&
 			    (queue->xcp_id == idx)) {
-				r = amdgpu_userq_map_helper(uqm, queue);
+				r = amdgpu_userq_restore_helper(uqm, queue);
 				if (r)
 					ret = r;
 			}

From 531df041f2a5296174abd8292d298eb62fe1ea97 Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <superm1@kernel.org>
Date: Mon, 11 Aug 2025 12:00:06 -0500
Subject: [PATCH 1311/1455] drm/amd: Avoid evicting resources at S5
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Normally resources are evicted on dGPUs at suspend or hibernate and
on APUs at hibernate.  These steps are unnecessary when using the S4
callbacks to put the system into S5.

Cc: AceLan Kao <acelan.kao@canonical.com>
Cc: Kai-Heng Feng <kaihengf@nvidia.com>
Cc: Mark Pearson <mpearson-lenovo@squebb.ca>
Cc: Denis Benato <benato.denis96@gmail.com>
Cc: Merthan Karakaş <m3rthn.k@gmail.com>
Tested-by: Eric Naim <dnaim@cachyos.org>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 81a48b3a1931..0fdfde3dcb9f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5072,6 +5072,10 @@ static int amdgpu_device_evict_resources(struct amdgpu_device *adev)
 	if (!adev->in_s4 && (adev->flags & AMD_IS_APU))
 		return 0;
 
+	/* No need to evict when going to S5 through S4 callbacks */
+	if (system_state == SYSTEM_POWER_OFF)
+		return 0;
+
 	ret = amdgpu_ttm_evict_resources(adev, TTM_PL_VRAM);
 	if (ret) {
 		dev_warn(adev->dev, "evicting device resources failed\n");

From 85442bac84666640354c6728fd5d5d241190d960 Mon Sep 17 00:00:00 2001
From: Shaoyun Liu <shaoyun.liu@amd.com>
Date: Thu, 11 Sep 2025 12:59:00 -0400
Subject: [PATCH 1312/1455] drm/amd/amdgpu: Fix the mes version that support
 inv_tlbs

MES pipe0 will do VM invalidation with engine set 5 when assign VMID to a process,
driver will submit inv_tlb package to mes pipe1. It might run into race condition
if both pipes use the same invalidate engine set. From MES version 0x83 it will use
invalidate engine set 6 for pipe1 to fix the issue

Signed-off-by: Shaoyun Liu <shaoyun.liu@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
index 76d3c40735b0..7cc16af1868b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
@@ -337,7 +337,7 @@ static void gmc_v12_0_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
 	int vmid, i;
 
 	if (adev->enable_uni_mes && adev->mes.ring[AMDGPU_MES_SCHED_PIPE].sched.ready &&
-	    (adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x81) {
+	    (adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x83) {
 		struct mes_inv_tlbs_pasid_input input = {0};
 		input.pasid = pasid;
 		input.flush_type = flush_type;

From 7ccaaf13197fddb525571d6646bc174405efb37c Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Fri, 22 Aug 2025 15:38:00 -0400
Subject: [PATCH 1313/1455] Revert "drm/amdkfd: return migration pages from
 copy function"

This reverts commit bd6093e2f1601c0c83906f5115a2efb6b93050b1.

migrate_vma_pages can fail if a CPU thread faults on the same page.
However, the page table is locked and only one of the new pages will
be inserted. The device driver will see that the MIGRATE_PFN_MIGRATE
bit is cleared if it loses the race.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 130 +++++++++++------------
 1 file changed, 65 insertions(+), 65 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 5d7eb0234002..f0eb4547318a 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -39,22 +39,22 @@
 #endif
 #define dev_fmt(fmt) "kfd_migrate: " fmt
 
-static uint64_t
-svm_migrate_direct_mapping_addr(struct amdgpu_device *adev, uint64_t addr)
+static u64
+svm_migrate_direct_mapping_addr(struct amdgpu_device *adev, u64 addr)
 {
 	return addr + amdgpu_ttm_domain_start(adev, TTM_PL_VRAM);
 }
 
 static int
-svm_migrate_gart_map(struct amdgpu_ring *ring, uint64_t npages,
-		     dma_addr_t *addr, uint64_t *gart_addr, uint64_t flags)
+svm_migrate_gart_map(struct amdgpu_ring *ring, u64 npages,
+		     dma_addr_t *addr, u64 *gart_addr, u64 flags)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_job *job;
 	unsigned int num_dw, num_bytes;
 	struct dma_fence *fence;
-	uint64_t src_addr, dst_addr;
-	uint64_t pte_flags;
+	u64 src_addr, dst_addr;
+	u64 pte_flags;
 	void *cpu_addr;
 	int r;
 
@@ -122,15 +122,15 @@ svm_migrate_gart_map(struct amdgpu_ring *ring, uint64_t npages,
 
 static int
 svm_migrate_copy_memory_gart(struct amdgpu_device *adev, dma_addr_t *sys,
-			     uint64_t *vram, uint64_t npages,
+			     u64 *vram, u64 npages,
 			     enum MIGRATION_COPY_DIR direction,
 			     struct dma_fence **mfence)
 {
-	const uint64_t GTT_MAX_PAGES = AMDGPU_GTT_MAX_TRANSFER_SIZE;
+	const u64 GTT_MAX_PAGES = AMDGPU_GTT_MAX_TRANSFER_SIZE;
 	struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
-	uint64_t gart_s, gart_d;
+	u64 gart_s, gart_d;
 	struct dma_fence *next;
-	uint64_t size;
+	u64 size;
 	int r;
 
 	mutex_lock(&adev->mman.gtt_window_lock);
@@ -260,30 +260,42 @@ static void svm_migrate_put_sys_page(unsigned long addr)
 	put_page(page);
 }
 
-static long
+static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate)
+{
+	unsigned long upages = 0;
+	unsigned long i;
+
+	for (i = 0; i < migrate->npages; i++) {
+		if (migrate->src[i] & MIGRATE_PFN_VALID &&
+		    !(migrate->src[i] & MIGRATE_PFN_MIGRATE))
+			upages++;
+	}
+	return upages;
+}
+
+static int
 svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 			 struct migrate_vma *migrate, struct dma_fence **mfence,
-			 dma_addr_t *scratch, uint64_t ttm_res_offset)
+			 dma_addr_t *scratch, u64 ttm_res_offset)
 {
-	uint64_t npages = migrate->npages;
+	u64 npages = migrate->npages;
 	struct amdgpu_device *adev = node->adev;
 	struct device *dev = adev->dev;
 	struct amdgpu_res_cursor cursor;
-	long mpages;
+	u64 mpages = 0;
 	dma_addr_t *src;
-	uint64_t *dst;
-	uint64_t i, j;
+	u64 *dst;
+	u64 i, j;
 	int r;
 
 	pr_debug("svms 0x%p [0x%lx 0x%lx 0x%llx]\n", prange->svms, prange->start,
 		 prange->last, ttm_res_offset);
 
 	src = scratch;
-	dst = (uint64_t *)(scratch + npages);
+	dst = (u64 *)(scratch + npages);
 
 	amdgpu_res_first(prange->ttm_res, ttm_res_offset,
 			 npages << PAGE_SHIFT, &cursor);
-	mpages = 0;
 	for (i = j = 0; (i < npages) && (mpages < migrate->cpages); i++) {
 		struct page *spage;
 
@@ -344,14 +356,13 @@ svm_migrate_copy_to_vram(struct kfd_node *node, struct svm_range *prange,
 out_free_vram_pages:
 	if (r) {
 		pr_debug("failed %d to copy memory to vram\n", r);
-		while (i-- && mpages) {
+		for (i = 0; i < npages && mpages; i++) {
 			if (!dst[i])
 				continue;
 			svm_migrate_put_vram_page(adev, dst[i]);
 			migrate->dst[i] = 0;
 			mpages--;
 		}
-		mpages = r;
 	}
 
 #ifdef DEBUG_FORCE_MIXED_DOMAINS
@@ -369,22 +380,22 @@ out_free_vram_pages:
 	}
 #endif
 
-	return mpages;
+	return r;
 }
 
 static long
 svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
-			struct vm_area_struct *vma, uint64_t start,
-			uint64_t end, uint32_t trigger, uint64_t ttm_res_offset)
+			struct vm_area_struct *vma, u64 start,
+			u64 end, uint32_t trigger, u64 ttm_res_offset)
 {
 	struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms);
-	uint64_t npages = (end - start) >> PAGE_SHIFT;
+	u64 npages = (end - start) >> PAGE_SHIFT;
 	struct amdgpu_device *adev = node->adev;
 	struct kfd_process_device *pdd;
 	struct dma_fence *mfence = NULL;
 	struct migrate_vma migrate = { 0 };
 	unsigned long cpages = 0;
-	long mpages = 0;
+	unsigned long mpages = 0;
 	dma_addr_t *scratch;
 	void *buf;
 	int r = -ENOMEM;
@@ -397,7 +408,7 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	migrate.pgmap_owner = SVM_ADEV_PGMAP_OWNER(adev);
 
 	buf = kvcalloc(npages,
-		       2 * sizeof(*migrate.src) + sizeof(uint64_t) + sizeof(dma_addr_t),
+		       2 * sizeof(*migrate.src) + sizeof(u64) + sizeof(dma_addr_t),
 		       GFP_KERNEL);
 	if (!buf)
 		goto out;
@@ -430,17 +441,15 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	else
 		pr_debug("0x%lx pages collected\n", cpages);
 
-	mpages = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset);
+	r = svm_migrate_copy_to_vram(node, prange, &migrate, &mfence, scratch, ttm_res_offset);
 	migrate_vma_pages(&migrate);
 
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
 
-	if (mpages >= 0)
-		pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n",
-			 mpages, cpages, migrate.npages);
-	else
-		r = mpages;
+	mpages = cpages - svm_migrate_unsuccessful_pages(&migrate);
+	pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
+		 mpages, cpages, migrate.npages);
 
 	svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages);
 
@@ -450,13 +459,14 @@ out_free:
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
 				    0, node->id, trigger, r);
 out:
-	if (!r && mpages > 0) {
+	if (!r && mpages) {
 		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_in, pdd->page_in + mpages);
-	}
 
-	return r ? r : mpages;
+		return mpages;
+	}
+	return r;
 }
 
 /**
@@ -480,7 +490,7 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 {
 	unsigned long addr, start, end;
 	struct vm_area_struct *vma;
-	uint64_t ttm_res_offset;
+	u64 ttm_res_offset;
 	struct kfd_node *node;
 	unsigned long mpages = 0;
 	long r = 0;
@@ -567,18 +577,17 @@ static void svm_migrate_page_free(struct page *page)
 	}
 }
 
-static long
+static int
 svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 			struct migrate_vma *migrate, struct dma_fence **mfence,
-			dma_addr_t *scratch, uint64_t npages)
+			dma_addr_t *scratch, u64 npages)
 {
 	struct device *dev = adev->dev;
-	uint64_t *src;
+	u64 *src;
 	dma_addr_t *dst;
 	struct page *dpage;
-	long mpages;
-	uint64_t i = 0, j;
-	uint64_t addr;
+	u64 i = 0, j;
+	u64 addr;
 	int r = 0;
 
 	pr_debug("svms 0x%p [0x%lx 0x%lx]\n", prange->svms, prange->start,
@@ -586,10 +595,9 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 
 	addr = migrate->start;
 
-	src = (uint64_t *)(scratch + npages);
+	src = (u64 *)(scratch + npages);
 	dst = scratch;
 
-	mpages = 0;
 	for (i = 0, j = 0; i < npages; i++, addr += PAGE_SIZE) {
 		struct page *spage;
 
@@ -638,7 +646,6 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 				     dst[i] >> PAGE_SHIFT, page_to_pfn(dpage));
 
 		migrate->dst[i] = migrate_pfn(page_to_pfn(dpage));
-		mpages++;
 		j++;
 	}
 
@@ -648,17 +655,13 @@ svm_migrate_copy_to_ram(struct amdgpu_device *adev, struct svm_range *prange,
 out_oom:
 	if (r) {
 		pr_debug("failed %d copy to ram\n", r);
-		while (i-- && mpages) {
-			if (!migrate->dst[i])
-				continue;
+		while (i--) {
 			svm_migrate_put_sys_page(dst[i]);
 			migrate->dst[i] = 0;
-			mpages--;
 		}
-		mpages = r;
 	}
 
-	return mpages;
+	return r;
 }
 
 /**
@@ -680,13 +683,14 @@ out_oom:
  */
 static long
 svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
-		       struct vm_area_struct *vma, uint64_t start, uint64_t end,
+		       struct vm_area_struct *vma, u64 start, u64 end,
 		       uint32_t trigger, struct page *fault_page)
 {
 	struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms);
-	uint64_t npages = (end - start) >> PAGE_SHIFT;
+	u64 npages = (end - start) >> PAGE_SHIFT;
+	unsigned long upages = npages;
 	unsigned long cpages = 0;
-	long mpages = 0;
+	unsigned long mpages = 0;
 	struct amdgpu_device *adev = node->adev;
 	struct kfd_process_device *pdd;
 	struct dma_fence *mfence = NULL;
@@ -706,7 +710,7 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 		migrate.flags = MIGRATE_VMA_SELECT_DEVICE_PRIVATE;
 
 	buf = kvcalloc(npages,
-		       2 * sizeof(*migrate.src) + sizeof(uint64_t) + sizeof(dma_addr_t),
+		       2 * sizeof(*migrate.src) + sizeof(u64) + sizeof(dma_addr_t),
 		       GFP_KERNEL);
 	if (!buf)
 		goto out;
@@ -740,15 +744,13 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 	else
 		pr_debug("0x%lx pages collected\n", cpages);
 
-	mpages = svm_migrate_copy_to_ram(adev, prange, &migrate, &mfence,
+	r = svm_migrate_copy_to_ram(adev, prange, &migrate, &mfence,
 				    scratch, npages);
 	migrate_vma_pages(&migrate);
 
-	if (mpages >= 0)
-		pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n",
-		 mpages, cpages, migrate.npages);
-	else
-		r = mpages;
+	upages = svm_migrate_unsuccessful_pages(&migrate);
+	pr_debug("unsuccessful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
+		 upages, cpages, migrate.npages);
 
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
@@ -761,7 +763,8 @@ out_free:
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
 				    node->id, 0, trigger, r);
 out:
-	if (!r && mpages > 0) {
+	if (!r && cpages) {
+		mpages = cpages - upages;
 		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_out, pdd->page_out + mpages);
@@ -844,9 +847,6 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 	}
 
 	if (r >= 0) {
-		WARN_ONCE(prange->vram_pages < mpages,
-			"Recorded vram pages(0x%llx) should not be less than migration pages(0x%lx).",
-			prange->vram_pages, mpages);
 		prange->vram_pages -= mpages;
 
 		/* prange does not have vram page set its actual_loc to system

From a450d225327845c9e780d81ec881c8813bbf1f7b Mon Sep 17 00:00:00 2001
From: James Zhu <James.Zhu@amd.com>
Date: Fri, 22 Aug 2025 15:38:01 -0400
Subject: [PATCH 1314/1455] drm/amdkfd: add function
 svm_migrate_successful_pages

to get migration pages. dst MIGRATE_PFN_VALID bit and src
MIGRATE_PFN_MIGRATE bit should always be set when migration success.

cpage includes src MIGRATE_PFN_MIGRATE bit set and MIGRATE_PFN_VALID
bit unset pages for both RAM and VRAM when memory is only allocated
without being populated before migration, those ram pages should be
counted as migrated pages and those vram pages should not be counted
as migrated pages. Here migration pages refer to how many vram pages
invloved. Current svm_migrate_unsuccessful_pages only covers the
unsuccessful case that source is on RAM.

So far, we only see two unsuccessful migration cases. Since we
can clearly identify successful migration cases through dst
MIGRATE_PFN_VALID bit and src MIGRATE_PFN_MIGRATE bit within this
prange, also eventually successful migration pages will be used,
so we can use function svm_migrate_successful_pages to replace
function svm_migrate_unsuccessful_pages.

Signed-off-by: James Zhu <James.Zhu@amd.com>
Reviewed-by: Philip Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 29 ++++++++++++------------
 1 file changed, 15 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index f0eb4547318a..d10c6673f4de 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -260,17 +260,17 @@ static void svm_migrate_put_sys_page(unsigned long addr)
 	put_page(page);
 }
 
-static unsigned long svm_migrate_unsuccessful_pages(struct migrate_vma *migrate)
+static unsigned long svm_migrate_successful_pages(struct migrate_vma *migrate)
 {
-	unsigned long upages = 0;
+	unsigned long mpages = 0;
 	unsigned long i;
 
 	for (i = 0; i < migrate->npages; i++) {
-		if (migrate->src[i] & MIGRATE_PFN_VALID &&
-		    !(migrate->src[i] & MIGRATE_PFN_MIGRATE))
-			upages++;
+		if (migrate->dst[i] & MIGRATE_PFN_VALID &&
+		    migrate->src[i] & MIGRATE_PFN_MIGRATE)
+			mpages++;
 	}
-	return upages;
+	return mpages;
 }
 
 static int
@@ -447,8 +447,8 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange,
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
 
-	mpages = cpages - svm_migrate_unsuccessful_pages(&migrate);
-	pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
+	mpages = svm_migrate_successful_pages(&migrate);
+	pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n",
 		 mpages, cpages, migrate.npages);
 
 	svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages);
@@ -688,7 +688,6 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 {
 	struct kfd_process *p = container_of(prange->svms, struct kfd_process, svms);
 	u64 npages = (end - start) >> PAGE_SHIFT;
-	unsigned long upages = npages;
 	unsigned long cpages = 0;
 	unsigned long mpages = 0;
 	struct amdgpu_device *adev = node->adev;
@@ -748,9 +747,9 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange,
 				    scratch, npages);
 	migrate_vma_pages(&migrate);
 
-	upages = svm_migrate_unsuccessful_pages(&migrate);
-	pr_debug("unsuccessful/cpages/npages 0x%lx/0x%lx/0x%lx\n",
-		 upages, cpages, migrate.npages);
+	mpages = svm_migrate_successful_pages(&migrate);
+	pr_debug("migrated/collected/requested 0x%lx/0x%lx/0x%lx\n",
+		 mpages, cpages, migrate.npages);
 
 	svm_migrate_copy_done(adev, mfence);
 	migrate_vma_finalize(&migrate);
@@ -763,8 +762,7 @@ out_free:
 				    start >> PAGE_SHIFT, end >> PAGE_SHIFT,
 				    node->id, 0, trigger, r);
 out:
-	if (!r && cpages) {
-		mpages = cpages - upages;
+	if (!r && mpages) {
 		pdd = svm_range_get_pdd_by_node(prange, node);
 		if (pdd)
 			WRITE_ONCE(pdd->page_out, pdd->page_out + mpages);
@@ -847,6 +845,9 @@ int svm_migrate_vram_to_ram(struct svm_range *prange, struct mm_struct *mm,
 	}
 
 	if (r >= 0) {
+		WARN_ONCE(prange->vram_pages < mpages,
+			  "Recorded vram pages(0x%llx) should not be less than migration pages(0x%lx).",
+			  prange->vram_pages, mpages);
 		prange->vram_pages -= mpages;
 
 		/* prange does not have vram page set its actual_loc to system

From 489f0f600ce2c0dae640df9035e1d82677d2580f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Mon, 25 Aug 2025 23:33:33 +0200
Subject: [PATCH 1315/1455] drm/amd/display: Fix DVI-D/HDMI adapters
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When the EDID has the HDMI bit, we should simply select
the HDMI signal type even on DVI ports.

For reference see, the legacy amdgpu display code:
amdgpu_atombios_encoder_get_encoder_mode
which selects ATOM_ENCODER_MODE_HDMI for the same case.

This commit fixes DVI connectors to work with DVI-D/HDMI
adapters so that they can now produce output over these
connectors for HDMI monitors with higher bandwidth modes.
With this change, even HDMI audio works through DVI.

For testing, I used a CAA-DMDHFD3 DVI-D/HDMI adapter
with the following GPUs:

Tahiti (DCE 6) - DC can now output 4K 30 Hz over DVI
Polaris 10 (DCE 11.2) - DC can now output 4K 60 Hz over DVI

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/link/link_detection.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
index b717e430051a..85303167a553 100644
--- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c
+++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c
@@ -1140,6 +1140,10 @@ static bool detect_link_and_local_sink(struct dc_link *link,
 		if (sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A &&
 		    !sink->edid_caps.edid_hdmi)
 			sink->sink_signal = SIGNAL_TYPE_DVI_SINGLE_LINK;
+		else if (dc_is_dvi_signal(sink->sink_signal) &&
+			 aud_support->hdmi_audio_native &&
+			 sink->edid_caps.edid_hdmi)
+			sink->sink_signal = SIGNAL_TYPE_HDMI_TYPE_A;
 
 		if (link->local_sink && dc_is_dp_signal(sink_caps.signal))
 			dp_trace_init(link);

From 0449726b58ea64ec96b95f95944f0a3650204059 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Mon, 25 Aug 2025 23:56:28 +0200
Subject: [PATCH 1316/1455] drm/amd/display: Keep PLL0 running on DCE 6.0 and
 6.4
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

DC can turn off the display clock when no displays are connected
or when all displays are off, for reference see:
- dce*_validate_bandwidth

DC also assumes that the DP clock is always on and never powers
it down, for reference see:
- dce110_clock_source_power_down

In case of DCE 6.0 and 6.4, PLL0 is the clock source for both
the engine clock and DP clock, for reference see:
- radeon_atom_pick_pll
- atombios_crtc_set_disp_eng_pll

Therefore, PLL0 should be always kept running on DCE 6.0 and 6.4.
This commit achieves that by ensuring that by setting the display
clock to the corresponding value in low power state instead of
zero.

This fixes a page flip timeout on SI with DC which happens when
all connected displays are blanked.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/resource/dce60/dce60_resource.c    | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
index 53b60044653f..c164d2500c2a 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
@@ -881,7 +881,16 @@ static enum dc_status dce60_validate_bandwidth(
 		context->bw_ctx.bw.dce.dispclk_khz = 681000;
 		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
 	} else {
-		context->bw_ctx.bw.dce.dispclk_khz = 0;
+		/* On DCE 6.0 and 6.4 the PLL0 is both the display engine clock and
+		 * the DP clock, and shouldn't be turned off. Just select the display
+		 * clock value from its low power mode.
+		 */
+		if (dc->ctx->dce_version == DCE_VERSION_6_0 ||
+			dc->ctx->dce_version == DCE_VERSION_6_4)
+			context->bw_ctx.bw.dce.dispclk_khz = 352000;
+		else
+			context->bw_ctx.bw.dce.dispclk_khz = 0;
+
 		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 

From c97a7dccb3ed680031011cfc1457506e6de49c9a Mon Sep 17 00:00:00 2001
From: Xi Ruoyao <xry111@xry111.site>
Date: Mon, 25 Aug 2025 16:52:11 +0800
Subject: [PATCH 1317/1455] drm/amd/display/dml2: Guard
 dml21_map_dc_state_into_dml_display_cfg with DC_FP_START

dml21_map_dc_state_into_dml_display_cfg calls (the call is usually
inlined by the compiler) populate_dml21_surface_config_from_plane_state
and populate_dml21_plane_config_from_plane_state which may use FPU.  In
a x86-64 build:

    $ objdump --disassemble=dml21_map_dc_state_into_dml_display_cfg \
    > drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.o |
    > grep %xmm -c
    63

Thus it needs to be guarded with DC_FP_START.  But we must note that the
current code quality of the in-kernel FPU use in AMD dml2 is very much
problematic: we are actually calling DC_FP_START in dml21_wrapper.c
here, and this translation unit is built with CC_FLAGS_FPU.  Strictly
speaking this does not make any sense: with CC_FLAGS_FPU the compiler is
allowed to generate FPU uses anywhere in the translated code, perhaps
out of the DC_FP_START guard.  This problematic pattern also occurs in
at least dml2_wrapper.c, dcn35_fpu.c, and dcn351_fpu.c.  Thus we really
need a careful audit and refactor for the in-kernel FPU uses, and this
patch is simply whacking a mole.  However per the reporter, whacking
this mole is enough to make a 9060XT "just work."

Reported-by: Asiacn <710187964@qq.com>
Closes: https://github.com/loongson-community/discussions/issues/102
Tested-by: Asiacn <710187964@qq.com>
Signed-off-by: Xi Ruoyao <xry111@xry111.site>
Reviewed-by: Huacai Chen <chenhuacai@loongson.cn>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
index 798abb2b2e67..08f7f03b1023 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c
@@ -224,7 +224,9 @@ static bool dml21_mode_check_and_programming(const struct dc *in_dc, struct dc_s
 	dml_ctx->config.svp_pstate.callbacks.release_phantom_streams_and_planes(in_dc, context);
 
 	/* Populate stream, plane mappings and other fields in display config. */
+	DC_FP_START();
 	result = dml21_map_dc_state_into_dml_display_cfg(in_dc, context, dml_ctx);
+	DC_FP_END();
 	if (!result)
 		return false;
 
@@ -279,7 +281,9 @@ static bool dml21_check_mode_support(const struct dc *in_dc, struct dc_state *co
 	dml_ctx->config.svp_pstate.callbacks.release_phantom_streams_and_planes(in_dc, context);
 
 	mode_support->dml2_instance = dml_init->dml2_instance;
+	DC_FP_START();
 	dml21_map_dc_state_into_dml_display_cfg(in_dc, context, dml_ctx);
+	DC_FP_END();
 	dml_ctx->v21.mode_programming.dml2_instance->scratch.build_mode_programming_locals.mode_programming_params.programming = dml_ctx->v21.mode_programming.programming;
 	DC_FP_START();
 	is_supported = dml2_check_mode_supported(mode_support);

From 7495962cbceb967e095233a5673ea71f3bcdee7e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Mon, 25 Aug 2025 23:56:29 +0200
Subject: [PATCH 1318/1455] drm/amd/display: Disable fastboot on DCE 6 too
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It already didn't work on DCE 8,
so there is no reason to assume it would on DCE 6.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
index 7b4a04704312..24184b4eb352 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c
@@ -1925,10 +1925,8 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
 
 	get_edp_streams(context, edp_streams, &edp_stream_num);
 
-	// Check fastboot support, disable on DCE8 because of blank screens
-	if (edp_num && edp_stream_num && dc->ctx->dce_version != DCE_VERSION_8_0 &&
-		    dc->ctx->dce_version != DCE_VERSION_8_1 &&
-		    dc->ctx->dce_version != DCE_VERSION_8_3) {
+	/* Check fastboot support, disable on DCE 6-8 because of blank screens */
+	if (edp_num && edp_stream_num && dc->ctx->dce_version < DCE_VERSION_10_0) {
 		for (i = 0; i < edp_num; i++) {
 			edp_link = edp_links[i];
 			if (edp_link != edp_streams[0]->link)

From 043c87d7d56e135393f8aab927148096e2d17589 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Mon, 25 Aug 2025 23:56:30 +0200
Subject: [PATCH 1319/1455] drm/amd/display: Disable VRR on DCE 6
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

DCE 6 was not advertised as being able to support VRR,
so let's mark it as unsupported for now.

The VRR implementation in amdgpu_dm depends on the VUPDATE
interrupt which is not registered for DCE 6.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 +++-
 drivers/gpu/drm/amd/display/dc/dc_helper.c        | 5 +++++
 drivers/gpu/drm/amd/display/dc/dm_services.h      | 2 ++
 3 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 7495122358bd..38fa6319265e 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -10834,6 +10834,8 @@ static void get_freesync_config_for_crtc(
 		} else {
 			config.state = VRR_STATE_INACTIVE;
 		}
+	} else {
+		config.state = VRR_STATE_UNSUPPORTED;
 	}
 out:
 	new_crtc_state->freesync_config = config;
@@ -12735,7 +12737,7 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector,
 
 	dm_con_state = to_dm_connector_state(connector->state);
 
-	if (!adev->dm.freesync_module)
+	if (!adev->dm.freesync_module || !dc_supports_vrr(sink->ctx->dce_version))
 		goto update;
 
 	edid = drm_edid_raw(drm_edid); // FIXME: Get rid of drm_edid_raw()
diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c
index 51e41aed7316..5a365bd19933 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c
@@ -755,3 +755,8 @@ char *dce_version_to_string(const int version)
 		return "Unknown";
 	}
 }
+
+bool dc_supports_vrr(const enum dce_version v)
+{
+	return v >= DCE_VERSION_8_0;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h
index 7b9c22c45453..7b398d4f4439 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services.h
@@ -311,4 +311,6 @@ void dm_dtn_log_end(struct dc_context *ctx,
 
 char *dce_version_to_string(const int version);
 
+bool dc_supports_vrr(const enum dce_version v);
+
 #endif /* __DM_SERVICES_H__ */

From 6cbe6e072c5d088101cd542ad8ef8541edeea5c3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Mon, 25 Aug 2025 23:56:31 +0200
Subject: [PATCH 1320/1455] drm/amd/display: Don't use non-registered VUPDATE
 on DCE 6
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The VUPDATE interrupt isn't registered on DCE 6, so don't try
to use that.

This fixes a page flip timeout after sleep/resume on DCE 6.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 20 ++++++++++++-------
 .../amd/display/amdgpu_dm/amdgpu_dm_crtc.c    | 16 +++++++++------
 2 files changed, 23 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 38fa6319265e..9dd2d1ed674d 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -3047,14 +3047,20 @@ static void dm_gpureset_toggle_interrupts(struct amdgpu_device *adev,
 				drm_warn(adev_to_drm(adev), "Failed to %s pflip interrupts\n",
 					 enable ? "enable" : "disable");
 
-			if (enable) {
-				if (amdgpu_dm_crtc_vrr_active(to_dm_crtc_state(acrtc->base.state)))
-					rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, true);
-			} else
-				rc = amdgpu_dm_crtc_set_vupdate_irq(&acrtc->base, false);
+			if (dc_supports_vrr(adev->dm.dc->ctx->dce_version)) {
+				if (enable) {
+					if (amdgpu_dm_crtc_vrr_active(
+							to_dm_crtc_state(acrtc->base.state)))
+						rc = amdgpu_dm_crtc_set_vupdate_irq(
+							&acrtc->base, true);
+				} else
+					rc = amdgpu_dm_crtc_set_vupdate_irq(
+							&acrtc->base, false);
 
-			if (rc)
-				drm_warn(adev_to_drm(adev), "Failed to %sable vupdate interrupt\n", enable ? "en" : "dis");
+				if (rc)
+					drm_warn(adev_to_drm(adev), "Failed to %sable vupdate interrupt\n",
+						enable ? "en" : "dis");
+			}
 
 			irq_source = IRQ_TYPE_VBLANK + acrtc->otg_inst;
 			/* During gpu-reset we disable and then enable vblank irq, so
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index 45feb404b097..466dccb355d7 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -317,13 +317,17 @@ static inline int amdgpu_dm_crtc_set_vblank(struct drm_crtc *crtc, bool enable)
 			dc->config.disable_ips != DMUB_IPS_DISABLE_ALL &&
 			sr_supported && vblank->config.disable_immediate)
 			drm_crtc_vblank_restore(crtc);
+	}
 
-		/* vblank irq on -> Only need vupdate irq in vrr mode */
-		if (amdgpu_dm_crtc_vrr_active(acrtc_state))
-			rc = amdgpu_dm_crtc_set_vupdate_irq(crtc, true);
-	} else {
-		/* vblank irq off -> vupdate irq off */
-		rc = amdgpu_dm_crtc_set_vupdate_irq(crtc, false);
+	if (dc_supports_vrr(dm->dc->ctx->dce_version)) {
+		if (enable) {
+			/* vblank irq on -> Only need vupdate irq in vrr mode */
+			if (amdgpu_dm_crtc_vrr_active(acrtc_state))
+				rc = amdgpu_dm_crtc_set_vupdate_irq(crtc, true);
+		} else {
+			/* vblank irq off -> vupdate irq off */
+			rc = amdgpu_dm_crtc_set_vupdate_irq(crtc, false);
+		}
 	}
 
 	if (rc)

From a9273da04fa033667a4d0ccfe46c4ba55721d7d1 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Fri, 5 Sep 2025 14:45:39 +0200
Subject: [PATCH 1321/1455] drm/amdgpu: add AMDGPU_IDS_FLAGS_GANG_SUBMIT
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add a UAPI flag indicating if gang submit is supported or not.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 4 ++++
 include/uapi/drm/amdgpu_drm.h           | 9 +++++----
 2 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 8a76960803c6..8676400834fc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -939,6 +939,10 @@ out:
 		if (adev->gfx.config.ta_cntl2_truncate_coord_mode)
 			dev_info->ids_flags |= AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD;
 
+		/* Gang submit is not supported under SRIOV currently */
+		if (!amdgpu_sriov_vf(adev))
+			dev_info->ids_flags |= AMDGPU_IDS_FLAGS_GANG_SUBMIT;
+
 		if (amdgpu_passthrough(adev))
 			dev_info->ids_flags |= (AMDGPU_IDS_FLAGS_MODE_PT <<
 						AMDGPU_IDS_FLAGS_MODE_SHIFT) &
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 85b3ca14f81e..cd7402e36b6d 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -1088,10 +1088,11 @@ struct drm_amdgpu_cs_chunk_cp_gfx_shadow {
  *  Query h/w info: Flag that this is integrated (a.h.a. fusion) GPU
  *
  */
-#define AMDGPU_IDS_FLAGS_FUSION         0x1
-#define AMDGPU_IDS_FLAGS_PREEMPTION     0x2
-#define AMDGPU_IDS_FLAGS_TMZ            0x4
-#define AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD 0x8
+#define AMDGPU_IDS_FLAGS_FUSION			0x01
+#define AMDGPU_IDS_FLAGS_PREEMPTION		0x02
+#define AMDGPU_IDS_FLAGS_TMZ			0x04
+#define AMDGPU_IDS_FLAGS_CONFORMANT_TRUNC_COORD	0x08
+#define AMDGPU_IDS_FLAGS_GANG_SUBMIT		0x10
 
 /*
  *  Query h/w info: Flag identifying VF/PF/PT mode

From 27405096232bc8e5042c6836920f793d31845694 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Fri, 6 Jun 2025 14:53:54 +0200
Subject: [PATCH 1322/1455] drm/amdgpu: revert "Implement new dummy vram
 manager"
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is should be unnecessary since a VRAM manager isn't mandatory in
the first place.

It could be that we have some missing checks inside AMDGPU or TTM but
those should then be fixed instead of worked around like that.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c      | 15 +++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c | 61 ++------------------
 2 files changed, 13 insertions(+), 63 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index f38bc9542cd6..4e2486dbc0a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -1943,11 +1943,13 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
 	}
 	adev->mman.initialized = true;
 
-	/* Initialize VRAM pool with all of VRAM divided into pages */
-	r = amdgpu_vram_mgr_init(adev);
-	if (r) {
-		dev_err(adev->dev, "Failed initializing VRAM heap.\n");
-		return r;
+	if (!adev->gmc.is_app_apu) {
+		/* Initialize VRAM pool with all of VRAM divided into pages */
+		r = amdgpu_vram_mgr_init(adev);
+		if (r) {
+			dev_err(adev->dev, "Failed initializing VRAM heap.\n");
+			return r;
+		}
 	}
 
 	/* Change the size here instead of the init above so only lpfn is affected */
@@ -2162,7 +2164,8 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 		drm_dev_exit(idx);
 	}
 
-	amdgpu_vram_mgr_fini(adev);
+	if (!adev->gmc.is_app_apu)
+		amdgpu_vram_mgr_fini(adev);
 	amdgpu_gtt_mgr_fini(adev);
 	amdgpu_preempt_mgr_fini(adev);
 	amdgpu_doorbell_fini(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
index e69db0a93378..a5adb2ed9b3c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c
@@ -425,45 +425,6 @@ out:
 	return ret;
 }
 
-static void amdgpu_dummy_vram_mgr_debug(struct ttm_resource_manager *man,
-				  struct drm_printer *printer)
-{
-	DRM_DEBUG_DRIVER("Dummy vram mgr debug\n");
-}
-
-static bool amdgpu_dummy_vram_mgr_compatible(struct ttm_resource_manager *man,
-				       struct ttm_resource *res,
-				       const struct ttm_place *place,
-				       size_t size)
-{
-	DRM_DEBUG_DRIVER("Dummy vram mgr compatible\n");
-	return false;
-}
-
-static bool amdgpu_dummy_vram_mgr_intersects(struct ttm_resource_manager *man,
-				       struct ttm_resource *res,
-				       const struct ttm_place *place,
-				       size_t size)
-{
-	DRM_DEBUG_DRIVER("Dummy vram mgr intersects\n");
-	return true;
-}
-
-static void amdgpu_dummy_vram_mgr_del(struct ttm_resource_manager *man,
-				struct ttm_resource *res)
-{
-	DRM_DEBUG_DRIVER("Dummy vram mgr deleted\n");
-}
-
-static int amdgpu_dummy_vram_mgr_new(struct ttm_resource_manager *man,
-			       struct ttm_buffer_object *tbo,
-			       const struct ttm_place *place,
-			       struct ttm_resource **res)
-{
-	DRM_DEBUG_DRIVER("Dummy vram mgr new\n");
-	return -ENOSPC;
-}
-
 /**
  * amdgpu_vram_mgr_new - allocate new ranges
  *
@@ -932,14 +893,6 @@ static void amdgpu_vram_mgr_debug(struct ttm_resource_manager *man,
 	mutex_unlock(&mgr->lock);
 }
 
-static const struct ttm_resource_manager_func amdgpu_dummy_vram_mgr_func = {
-	.alloc	= amdgpu_dummy_vram_mgr_new,
-	.free	= amdgpu_dummy_vram_mgr_del,
-	.intersects = amdgpu_dummy_vram_mgr_intersects,
-	.compatible = amdgpu_dummy_vram_mgr_compatible,
-	.debug	= amdgpu_dummy_vram_mgr_debug
-};
-
 static const struct ttm_resource_manager_func amdgpu_vram_mgr_func = {
 	.alloc	= amdgpu_vram_mgr_new,
 	.free	= amdgpu_vram_mgr_del,
@@ -973,16 +926,10 @@ int amdgpu_vram_mgr_init(struct amdgpu_device *adev)
 	INIT_LIST_HEAD(&mgr->allocated_vres_list);
 	mgr->default_page_size = PAGE_SIZE;
 
-	if (!adev->gmc.is_app_apu) {
-		man->func = &amdgpu_vram_mgr_func;
-
-		err = drm_buddy_init(&mgr->mm, man->size, PAGE_SIZE);
-		if (err)
-			return err;
-	} else {
-		man->func = &amdgpu_dummy_vram_mgr_func;
-		DRM_INFO("Setup dummy vram mgr\n");
-	}
+	man->func = &amdgpu_vram_mgr_func;
+	err = drm_buddy_init(&mgr->mm, man->size, PAGE_SIZE);
+	if (err)
+		return err;
 
 	ttm_set_driver_manager(&adev->mman.bdev, TTM_PL_VRAM, &mgr->manager);
 	ttm_resource_manager_set_used(man, true);

From 7840a699c513dfd20661b5d1d055011b3ab17e30 Mon Sep 17 00:00:00 2001
From: "Mario Limonciello (AMD)" <superm1@kernel.org>
Date: Fri, 12 Sep 2025 12:59:33 -0500
Subject: [PATCH 1323/1455] drm/amd: Drop unnecessary calls to
 smu_dpm_set_vpe_enable()

smu_hw_init() and smu_hw_fini() call smu_dpm_set_vpe_enable() for
APUs as part of startup and teardown.  These calls however are
not necessary because vpe_hw_init()/vpe_hw_fini() will call at
init/fini:

```
vpe_hw_init() / vpe_hw_fini()
  amdgpu_device_ip_set_powergating_state()
    vpe_set_powergating_state()
      amdgpu_dpm_enable_vpe()
        amdgpu_dpm_set_powergating_by_smu()
          smu_dpm_set_power_gate()
            smu_dpm_set_vpe_enable()
```

Drop the extra calls.

Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index 715579950dd8..bf2b38dd7e25 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -1925,7 +1925,6 @@ static int smu_hw_init(struct amdgpu_ip_block *ip_block)
 		for (i = 0; i < adev->vcn.num_vcn_inst; i++)
 			smu_dpm_set_vcn_enable(smu, true, i);
 		smu_dpm_set_jpeg_enable(smu, true);
-		smu_dpm_set_vpe_enable(smu, true);
 		smu_dpm_set_umsch_mm_enable(smu, true);
 		smu_set_mall_enable(smu);
 		smu_set_gfx_cgpg(smu, true);
@@ -2133,7 +2132,6 @@ static int smu_hw_fini(struct amdgpu_ip_block *ip_block)
 	}
 	smu_dpm_set_jpeg_enable(smu, false);
 	adev->jpeg.cur_state = AMD_PG_STATE_GATE;
-	smu_dpm_set_vpe_enable(smu, false);
 	smu_dpm_set_umsch_mm_enable(smu, false);
 
 	if (!smu->pm_enabled)

From 1a869168d91f1a1a2b0db22cea0295c67908e5d8 Mon Sep 17 00:00:00 2001
From: Zongyao Bai <zongyao.bai@intel.com>
Date: Tue, 16 Sep 2025 05:47:15 +0800
Subject: [PATCH 1324/1455] drm/xe/sysfs: Add cleanup action in
 xe_device_sysfs_init

On partial failure, some sysfs files created before the failure might
not be removed. Add common cleanup step to remove them all immediately,
as is should be harmless to attempt to remove non-existing files.

Fixes: 0e414bf7ad01 ("drm/xe: Expose PCIe link downgrade attributes")
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Stuart Summers <stuart.summers@intel.com>
Cc: Shuicheng Lin <shuicheng.lin@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Signed-off-by: Zongyao Bai <zongyao.bai@intel.com>
Reviewed-by: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250915214716.1327379-2-zongyao.bai@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_device_sysfs.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index 6ee422594b56..b7f8fcfed8d8 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -311,12 +311,16 @@ int xe_device_sysfs_init(struct xe_device *xe)
 	if (xe->info.platform == XE_BATTLEMAGE) {
 		ret = sysfs_create_files(&dev->kobj, auto_link_downgrade_attrs);
 		if (ret)
-			return ret;
+			goto cleanup;
 
 		ret = late_bind_create_files(dev);
 		if (ret)
-			return ret;
+			goto cleanup;
 	}
 
 	return devm_add_action_or_reset(dev, xe_device_sysfs_fini, xe);
+
+cleanup:
+	xe_device_sysfs_fini(xe);
+	return ret;
 }

From 3b09b11805bfee32d5a0000f5ede42c07237a6c4 Mon Sep 17 00:00:00 2001
From: John Harrison <John.C.Harrison@Intel.com>
Date: Tue, 9 Sep 2025 15:41:31 -0700
Subject: [PATCH 1325/1455] drm/xe/guc: Return an error code if the GuC load
 fails

Due to multiple explosion issues in the early days of the Xe driver,
the GuC load was hacked to never return a failure. That prevented
kernel panics and such initially, but now all it achieves is creating
more confusing errors when the driver tries to submit commands to a
GuC it already knows is not there. So fix that up.

As a stop-gap and to help with debug of load failures due to invalid
GuC init params, a wedge call had been added to the inner GuC load
function. The reason being that it leaves the GuC log accessible via
debugfs. However, for an end user, simply aborting the module load is
much cleaner than wedging and trying to continue. The wedge blocks
user submissions but it seems that various bits of the driver itself
still try to submit to a dead GuC and lots of subsequent errors occur.
And with regards to developers debugging why their particular code
change is being rejected by the GuC, it is trivial to either add the
wedge back in and hack the return code to zero again or to just do a
GuC log dump to dmesg.

v2: Add support for error injection testing and drop the now redundant
wedge call.

CC: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: Matt Atwood <matthew.s.atwood@intel.com>
Link: https://lore.kernel.org/r/20250909224132.536320-1-John.C.Harrison@Intel.com
---
 drivers/gpu/drm/xe/xe_guc.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
index 171b49817adf..00789844ea4d 100644
--- a/drivers/gpu/drm/xe/xe_guc.c
+++ b/drivers/gpu/drm/xe/xe_guc.c
@@ -1055,7 +1055,7 @@ static s32 guc_pc_get_cur_freq(struct xe_guc_pc *guc_pc)
 #endif
 #define GUC_LOAD_TIME_WARN_MS      200
 
-static void guc_wait_ucode(struct xe_guc *guc)
+static int guc_wait_ucode(struct xe_guc *guc)
 {
 	struct xe_gt *gt = guc_to_gt(guc);
 	struct xe_mmio *mmio = &gt->mmio;
@@ -1162,7 +1162,7 @@ static void guc_wait_ucode(struct xe_guc *guc)
 			break;
 		}
 
-		xe_device_declare_wedged(gt_to_xe(gt));
+		return -EPROTO;
 	} else if (delta_ms > GUC_LOAD_TIME_WARN_MS) {
 		xe_gt_warn(gt, "excessive init time: %lldms! [status = 0x%08X, timeouts = %d]\n",
 			   delta_ms, status, count);
@@ -1174,7 +1174,10 @@ static void guc_wait_ucode(struct xe_guc *guc)
 			  delta_ms, xe_guc_pc_get_act_freq(guc_pc), guc_pc_get_cur_freq(guc_pc),
 			  before_freq, status, count);
 	}
+
+	return 0;
 }
+ALLOW_ERROR_INJECTION(guc_wait_ucode, ERRNO);
 
 static int __xe_guc_upload(struct xe_guc *guc)
 {
@@ -1206,14 +1209,16 @@ static int __xe_guc_upload(struct xe_guc *guc)
 		goto out;
 
 	/* Wait for authentication */
-	guc_wait_ucode(guc);
+	ret = guc_wait_ucode(guc);
+	if (ret)
+		goto out;
 
 	xe_uc_fw_change_status(&guc->fw, XE_UC_FIRMWARE_RUNNING);
 	return 0;
 
 out:
 	xe_uc_fw_change_status(&guc->fw, XE_UC_FIRMWARE_LOAD_FAIL);
-	return 0	/* FIXME: ret, don't want to stop load currently */;
+	return ret;
 }
 
 static int vf_guc_min_load_for_hwconfig(struct xe_guc *guc)

From 9e6eb49ec13936461c697348c74e5450ac82707d Mon Sep 17 00:00:00 2001
From: Yang Li <yang.lee@linux.alibaba.com>
Date: Tue, 16 Sep 2025 10:10:39 +0800
Subject: [PATCH 1326/1455] drm/xe: Remove duplicate header files

Fix some duplicate includes in xe:
./drivers/gpu/drm/xe/xe_tlb_inval.c: xe_tlb_inval.h is included more than once.
./drivers/gpu/drm/xe/xe_pt.c: xe_tlb_inval_job.h is included more than once.

While at it, also sort the include lines alphabetically.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=24705
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=24706
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
[Reword commit message]
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250916021039.1632766-1-yang.lee@linux.alibaba.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_pt.c        | 3 +--
 drivers/gpu/drm/xe/xe_tlb_inval.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 01eea8eb1779..a1c88f9a6c76 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -13,14 +13,13 @@
 #include "xe_drm_client.h"
 #include "xe_exec_queue.h"
 #include "xe_gt.h"
-#include "xe_tlb_inval_job.h"
 #include "xe_migrate.h"
 #include "xe_pt_types.h"
 #include "xe_pt_walk.h"
 #include "xe_res_cursor.h"
 #include "xe_sched_job.h"
-#include "xe_sync.h"
 #include "xe_svm.h"
+#include "xe_sync.h"
 #include "xe_tlb_inval_job.h"
 #include "xe_trace.h"
 #include "xe_ttm_stolen_mgr.h"
diff --git a/drivers/gpu/drm/xe/xe_tlb_inval.c b/drivers/gpu/drm/xe/xe_tlb_inval.c
index e6e97b5a7b5c..918a59e686ea 100644
--- a/drivers/gpu/drm/xe/xe_tlb_inval.c
+++ b/drivers/gpu/drm/xe/xe_tlb_inval.c
@@ -10,11 +10,10 @@
 #include "xe_force_wake.h"
 #include "xe_gt.h"
 #include "xe_gt_printk.h"
+#include "xe_gt_stats.h"
 #include "xe_guc.h"
 #include "xe_guc_ct.h"
 #include "xe_guc_tlb_inval.h"
-#include "xe_gt_stats.h"
-#include "xe_tlb_inval.h"
 #include "xe_mmio.h"
 #include "xe_pm.h"
 #include "xe_tlb_inval.h"

From d7ddcf921e7d0d8ebe82e89635bc9dc26ba9540d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 27 Aug 2025 13:14:43 +0200
Subject: [PATCH 1327/1455] drm/amdgpu: reject gang submissions under SRIOV
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Gang submission means that the kernel driver guarantees that multiple
submissions are executed on the HW at the same time on different engines.

Background is that those submissions then depend on each other and each
can't finish stand alone.

SRIOV now uses world switch to preempt submissions on the engines to allow
sharing the HW resources between multiple VFs.

The problem is now that the SRIOV world switch can't know about such inter
dependencies and will cause a timeout if it waits for a partially running
gang submission.

To conclude SRIOV and gang submissions are fundamentally incompatible at
the moment. For now just disable them.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 2e93d570153c..defb511acc5a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -273,7 +273,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p,
 		}
 	}
 
-	if (!p->gang_size) {
+	if (!p->gang_size || (amdgpu_sriov_vf(p->adev) && p->gang_size > 1)) {
 		ret = -EINVAL;
 		goto free_all_kdata;
 	}

From 39203f5e6dcf2b0529ae526004e7a9c8ef453ae2 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 27 Aug 2025 09:28:40 +0200
Subject: [PATCH 1328/1455] drm/amdgpu: fix userq VM validation v4
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

That was actually complete nonsense and not validating the BOs
at all. The code just cleared all VM areas were it couldn't grab the
lock for a BO.

Try to fix this. Only compile tested at the moment.

v2: fix fence slot reservation as well as pointed out by Sunil.
    also validate PDs, PTs, per VM BOs and update PDEs
v3: grab the status_lock while working with the done list.
v4: rename functions, add some comments, fix waiting for updates to
    complete.
v4: rename amdgpu_vm_lock_done_list(), add some more comments

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 158 +++++++++++-----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c    |  35 +++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h    |   2 +
 3 files changed, 115 insertions(+), 80 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index d2e59aecb3eb..3dfe7f5f154b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -706,109 +706,107 @@ amdgpu_userq_restore_all(struct amdgpu_userq_mgr *uq_mgr)
 	return ret;
 }
 
-static int
-amdgpu_userq_validate_vm_bo(void *_unused, struct amdgpu_bo *bo)
+static int amdgpu_userq_validate_vm(void *param, struct amdgpu_bo *bo)
 {
 	struct ttm_operation_ctx ctx = { false, false };
-	int ret;
 
 	amdgpu_bo_placement_from_domain(bo, bo->allowed_domains);
-
-	ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
-	if (ret)
-		DRM_ERROR("Fail to validate\n");
-
-	return ret;
+	return ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 }
 
+/* Handle all BOs on the invalidated list, validate them and update the PTs */
 static int
-amdgpu_userq_validate_bos(struct amdgpu_userq_mgr *uq_mgr)
+amdgpu_userq_bo_validate(struct amdgpu_device *adev, struct drm_exec *exec,
+			 struct amdgpu_vm *vm)
 {
-	struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr);
-	struct amdgpu_vm *vm = &fpriv->vm;
-	struct amdgpu_device *adev = uq_mgr->adev;
+	struct ttm_operation_ctx ctx = { false, false };
 	struct amdgpu_bo_va *bo_va;
-	struct ww_acquire_ctx *ticket;
-	struct drm_exec exec;
 	struct amdgpu_bo *bo;
-	struct dma_resv *resv;
-	bool clear, unlock;
-	int ret = 0;
-
-	drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES, 0);
-	drm_exec_until_all_locked(&exec) {
-		ret = amdgpu_vm_lock_pd(vm, &exec, 2);
-		drm_exec_retry_on_contention(&exec);
-		if (unlikely(ret)) {
-			drm_file_err(uq_mgr->file, "Failed to lock PD\n");
-			goto unlock_all;
-		}
-
-		/* Lock the done list */
-		list_for_each_entry(bo_va, &vm->done, base.vm_status) {
-			bo = bo_va->base.bo;
-			if (!bo)
-				continue;
-
-			ret = drm_exec_lock_obj(&exec, &bo->tbo.base);
-			drm_exec_retry_on_contention(&exec);
-			if (unlikely(ret))
-				goto unlock_all;
-		}
-	}
+	int ret;
 
 	spin_lock(&vm->status_lock);
-	while (!list_empty(&vm->moved)) {
-		bo_va = list_first_entry(&vm->moved, struct amdgpu_bo_va,
-					 base.vm_status);
-		spin_unlock(&vm->status_lock);
-
-		/* Per VM BOs never need to bo cleared in the page tables */
-		ret = amdgpu_vm_bo_update(adev, bo_va, false);
-		if (ret)
-			goto unlock_all;
-		spin_lock(&vm->status_lock);
-	}
-
-	ticket = &exec.ticket;
 	while (!list_empty(&vm->invalidated)) {
-		bo_va = list_first_entry(&vm->invalidated, struct amdgpu_bo_va,
+		bo_va = list_first_entry(&vm->invalidated,
+					 struct amdgpu_bo_va,
 					 base.vm_status);
-		resv = bo_va->base.bo->tbo.base.resv;
 		spin_unlock(&vm->status_lock);
 
 		bo = bo_va->base.bo;
-		ret = amdgpu_userq_validate_vm_bo(NULL, bo);
-		if (ret) {
-			drm_file_err(uq_mgr->file, "Failed to validate BO\n");
-			goto unlock_all;
-		}
+		ret = drm_exec_prepare_obj(exec, &bo->tbo.base, 2);
+		if (unlikely(ret))
+			return ret;
 
-		/* Try to reserve the BO to avoid clearing its ptes */
-		if (!adev->debug_vm && dma_resv_trylock(resv)) {
-			clear = false;
-			unlock = true;
-		/* The caller is already holding the reservation lock */
-		} else if (dma_resv_locking_ctx(resv) == ticket) {
-			clear = false;
-			unlock = false;
-		/* Somebody else is using the BO right now */
-		} else {
-			clear = true;
-			unlock = false;
-		}
-
-		ret = amdgpu_vm_bo_update(adev, bo_va, clear);
-
-		if (unlock)
-			dma_resv_unlock(resv);
+		amdgpu_bo_placement_from_domain(bo, bo->allowed_domains);
+		ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 		if (ret)
-			goto unlock_all;
+			return ret;
+
+		/* This moves the bo_va to the done list */
+		ret = amdgpu_vm_bo_update(adev, bo_va, false);
+		if (ret)
+			return ret;
 
 		spin_lock(&vm->status_lock);
 	}
 	spin_unlock(&vm->status_lock);
 
+	return 0;
+}
+
+/* Make sure the whole VM is ready to be used */
+static int
+amdgpu_userq_vm_validate(struct amdgpu_userq_mgr *uq_mgr)
+{
+	struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr);
+	struct amdgpu_device *adev = uq_mgr->adev;
+	struct amdgpu_vm *vm = &fpriv->vm;
+	struct amdgpu_bo_va *bo_va;
+	struct drm_exec exec;
+	int ret;
+
+	drm_exec_init(&exec, DRM_EXEC_IGNORE_DUPLICATES, 0);
+	drm_exec_until_all_locked(&exec) {
+		ret = amdgpu_vm_lock_pd(vm, &exec, 1);
+		drm_exec_retry_on_contention(&exec);
+		if (unlikely(ret))
+			goto unlock_all;
+
+		ret = amdgpu_vm_lock_done_list(vm, &exec, 1);
+		drm_exec_retry_on_contention(&exec);
+		if (unlikely(ret))
+			goto unlock_all;
+
+		/* This validates PDs, PTs and per VM BOs */
+		ret = amdgpu_vm_validate(adev, vm, NULL,
+					 amdgpu_userq_validate_vm,
+					 NULL);
+		if (unlikely(ret))
+			goto unlock_all;
+
+		/* This locks and validates the remaining evicted BOs */
+		ret = amdgpu_userq_bo_validate(adev, &exec, vm);
+		drm_exec_retry_on_contention(&exec);
+		if (unlikely(ret))
+			goto unlock_all;
+	}
+
+	ret = amdgpu_vm_handle_moved(adev, vm, NULL);
+	if (ret)
+		goto unlock_all;
+
+	ret = amdgpu_vm_update_pdes(adev, vm, false);
+	if (ret)
+		goto unlock_all;
+
+	/*
+	 * We need to wait for all VM updates to finish before restarting the
+	 * queues. Using the done list like that is now ok since everything is
+	 * locked in place.
+	 */
+	list_for_each_entry(bo_va, &vm->done, base.vm_status)
+		dma_fence_wait(bo_va->last_pt_update, false);
+	dma_fence_wait(vm->last_update, false);
+
 	ret = amdgpu_eviction_fence_replace_fence(&fpriv->evf_mgr, &exec);
 	if (ret)
 		drm_file_err(uq_mgr->file, "Failed to replace eviction fence\n");
@@ -828,7 +826,7 @@ static void amdgpu_userq_restore_worker(struct work_struct *work)
 
 	mutex_lock(&uq_mgr->userq_mutex);
 
-	ret = amdgpu_userq_validate_bos(uq_mgr);
+	ret = amdgpu_userq_vm_validate(uq_mgr);
 	if (ret) {
 		drm_file_err(uq_mgr->file, "Failed to validate BOs to restore\n");
 		goto unlock;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index bd12d8ff15a4..9980c0cded94 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -484,6 +484,41 @@ int amdgpu_vm_lock_pd(struct amdgpu_vm *vm, struct drm_exec *exec,
 				    2 + num_fences);
 }
 
+/**
+ * amdgpu_vm_lock_done_list - lock all BOs on the done list
+ * @exec: drm execution context
+ * @num_fences: number of extra fences to reserve
+ *
+ * Lock the BOs on the done list in the DRM execution context.
+ */
+int amdgpu_vm_lock_done_list(struct amdgpu_vm *vm, struct drm_exec *exec,
+			     unsigned int num_fences)
+{
+	struct list_head *prev = &vm->done;
+	struct amdgpu_bo_va *bo_va;
+	struct amdgpu_bo *bo;
+	int ret;
+
+	/* We can only trust prev->next while holding the lock */
+	spin_lock(&vm->status_lock);
+	while (!list_is_head(prev->next, &vm->done)) {
+		bo_va = list_entry(prev->next, typeof(*bo_va), base.vm_status);
+		spin_unlock(&vm->status_lock);
+
+		bo = bo_va->base.bo;
+		if (bo) {
+			ret = drm_exec_prepare_obj(exec, &bo->tbo.base, 1);
+			if (unlikely(ret))
+				return ret;
+		}
+		spin_lock(&vm->status_lock);
+		prev = prev->next;
+	}
+	spin_unlock(&vm->status_lock);
+
+	return 0;
+}
+
 /**
  * amdgpu_vm_move_to_lru_tail - move all BOs to the end of LRU
  *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index e045c1590d78..3409904b5c63 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -491,6 +491,8 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm);
 void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm);
 int amdgpu_vm_lock_pd(struct amdgpu_vm *vm, struct drm_exec *exec,
 		      unsigned int num_fences);
+int amdgpu_vm_lock_done_list(struct amdgpu_vm *vm, struct drm_exec *exec,
+			     unsigned int num_fences);
 bool amdgpu_vm_ready(struct amdgpu_vm *vm);
 uint64_t amdgpu_vm_generation(struct amdgpu_device *adev, struct amdgpu_vm *vm);
 int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm,

From 70db83e2b9064bc427d1a7003e516b3400a8ceda Mon Sep 17 00:00:00 2001
From: James Flowers <bold.zone2373@fastmail.com>
Date: Fri, 12 Sep 2025 22:19:52 -0700
Subject: [PATCH 1329/1455] drm/amd/display: Use kmalloc_array() instead of
 kmalloc()

Documentation/process/deprecated.rst recommends against the use of kmalloc
with dynamic size calculations due to the risk of overflow and smaller
allocation being made than the caller was expecting. This could lead to
buffer overflow in code similar to the memcpy in
amdgpu_dm_plane_add_modifier().

Signed-off-by: James Flowers <bold.zone2373@fastmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
index eef51652ca35..08f629c64df3 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
@@ -146,7 +146,7 @@ static void amdgpu_dm_plane_add_modifier(uint64_t **mods, uint64_t *size, uint64
 
 	if (*cap - *size < 1) {
 		uint64_t new_cap = *cap * 2;
-		uint64_t *new_mods = kmalloc(new_cap * sizeof(uint64_t), GFP_KERNEL);
+		uint64_t *new_mods = kmalloc_array(new_cap, sizeof(uint64_t), GFP_KERNEL);
 
 		if (!new_mods) {
 			kfree(*mods);
@@ -732,7 +732,7 @@ static int amdgpu_dm_plane_get_plane_modifiers(struct amdgpu_device *adev, unsig
 	if (adev->family < AMDGPU_FAMILY_AI)
 		return 0;
 
-	*mods = kmalloc(capacity * sizeof(uint64_t), GFP_KERNEL);
+	*mods = kmalloc_array(capacity, sizeof(uint64_t), GFP_KERNEL);
 
 	if (plane_type == DRM_PLANE_TYPE_CURSOR) {
 		amdgpu_dm_plane_add_modifier(mods, &size, &capacity, DRM_FORMAT_MOD_LINEAR);

From 5b3eca05cfb02ed0dbd7097c056fb08ce5f76fe3 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:34 -0600
Subject: [PATCH 1330/1455] drm/amd/display: Use devm_i2c_add_adapter to
 simplify i2c cleanup logic

This commit replaces the utilization of i2c_add/del_adapter() with
devm_i2c_add_adapter() to reduce the amount of boilerplate. Using
devm_i2c_add_adapter() has the advantage of removing the manual
manipulation of the I2C adapter.

Suggested-by: Robert Beckett <bob.beckett@collabora.com>
Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 21 ++-----------------
 1 file changed, 2 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 9dd2d1ed674d..f76d2ef9ca5c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -2945,7 +2945,7 @@ static int dm_oem_i2c_hw_init(struct amdgpu_device *adev)
 			return -ENOMEM;
 		}
 
-		r = i2c_add_adapter(&oem_i2c->base);
+		r = devm_i2c_add_adapter(adev->dev, &oem_i2c->base);
 		if (r) {
 			drm_info(adev_to_drm(adev), "Failed to register oem i2c\n");
 			kfree(oem_i2c);
@@ -2957,17 +2957,6 @@ static int dm_oem_i2c_hw_init(struct amdgpu_device *adev)
 	return 0;
 }
 
-static void dm_oem_i2c_hw_fini(struct amdgpu_device *adev)
-{
-	struct amdgpu_display_manager *dm = &adev->dm;
-
-	if (dm->oem_i2c) {
-		i2c_del_adapter(&dm->oem_i2c->base);
-		kfree(dm->oem_i2c);
-		dm->oem_i2c = NULL;
-	}
-}
-
 /**
  * dm_hw_init() - Initialize DC device
  * @ip_block: Pointer to the amdgpu_ip_block for this hw instance.
@@ -3018,8 +3007,6 @@ static int dm_hw_fini(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 
-	dm_oem_i2c_hw_fini(adev);
-
 	amdgpu_dm_hpd_fini(adev);
 
 	amdgpu_dm_irq_fini(adev);
@@ -7373,10 +7360,6 @@ static void amdgpu_dm_connector_destroy(struct drm_connector *connector)
 	drm_dp_cec_unregister_connector(&aconnector->dm_dp_aux.aux);
 	drm_connector_unregister(connector);
 	drm_connector_cleanup(connector);
-	if (aconnector->i2c) {
-		i2c_del_adapter(&aconnector->i2c->base);
-		kfree(aconnector->i2c);
-	}
 	kfree(aconnector->dm_dp_aux.aux.name);
 
 	kfree(connector);
@@ -8734,7 +8717,7 @@ static int amdgpu_dm_connector_init(struct amdgpu_display_manager *dm,
 	}
 
 	aconnector->i2c = i2c;
-	res = i2c_add_adapter(&i2c->base);
+	res = devm_i2c_add_adapter(dm->adev->dev, &i2c->base);
 
 	if (res) {
 		drm_err(adev_to_drm(dm->adev), "Failed to register hw i2c %d\n", link->link_index);

From 0f36a3c6af9e137344a2060cf2510db7ff9920c0 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:35 -0600
Subject: [PATCH 1331/1455] drm/amdgpu/amdgpu_i2c: Use devm_i2c_add_adapter
 instead of i2c_add_adapter

This commit replaces i2c_add_adapter() with devm_i2c_add_adapter() and
removes part of the cleanup logic since the new function handles the i2c
removal.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c | 18 +++---------------
 1 file changed, 3 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c
index 57101d24422f..9cb72f0c5277 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c
@@ -184,7 +184,7 @@ struct amdgpu_i2c_chan *amdgpu_i2c_create(struct drm_device *dev,
 		snprintf(i2c->adapter.name, sizeof(i2c->adapter.name),
 			 "AMDGPU i2c hw bus %s", name);
 		i2c->adapter.algo = &amdgpu_atombios_i2c_algo;
-		ret = i2c_add_adapter(&i2c->adapter);
+		ret = devm_i2c_add_adapter(dev->dev, &i2c->adapter);
 		if (ret)
 			goto out_free;
 	} else {
@@ -215,15 +215,6 @@ out_free:
 
 }
 
-void amdgpu_i2c_destroy(struct amdgpu_i2c_chan *i2c)
-{
-	if (!i2c)
-		return;
-	WARN_ON(i2c->has_aux);
-	i2c_del_adapter(&i2c->adapter);
-	kfree(i2c);
-}
-
 void amdgpu_i2c_init(struct amdgpu_device *adev)
 {
 	if (!adev->is_atom_fw) {
@@ -248,12 +239,9 @@ void amdgpu_i2c_fini(struct amdgpu_device *adev)
 {
 	int i;
 
-	for (i = 0; i < AMDGPU_MAX_I2C_BUS; i++) {
-		if (adev->i2c_bus[i]) {
-			amdgpu_i2c_destroy(adev->i2c_bus[i]);
+	for (i = 0; i < AMDGPU_MAX_I2C_BUS; i++)
+		if (adev->i2c_bus[i])
 			adev->i2c_bus[i] = NULL;
-		}
-	}
 }
 
 /* looks up bus based on id */

From 63137c7c8cd1aee788067b9d3782dce6ea21c0df Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:36 -0600
Subject: [PATCH 1332/1455] drm/amdgpu: Use devm_i2c_add_adapter() in SMU V11

Instead of using i2c_add_adapter() and i2c_del_adapter() in the SMU V11,
use devm_i2c_add_adapter() to simplify the code path.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c
index dd2d66090d23..68aef47254a9 100644
--- a/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c
+++ b/drivers/gpu/drm/amd/amdgpu/smu_v11_0_i2c.c
@@ -743,7 +743,7 @@ int smu_v11_0_i2c_control_init(struct amdgpu_device *adev)
 	adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 
-	res = i2c_add_adapter(control);
+	res = devm_i2c_add_adapter(adev->dev, control);
 	if (res)
 		DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
 
@@ -752,9 +752,6 @@ int smu_v11_0_i2c_control_init(struct amdgpu_device *adev)
 
 void smu_v11_0_i2c_control_fini(struct amdgpu_device *adev)
 {
-	struct i2c_adapter *control = adev->pm.ras_eeprom_i2c_bus;
-
-	i2c_del_adapter(control);
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From f4dfc4447d57c1d03ded394797018ac463449d2c Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:37 -0600
Subject: [PATCH 1333/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the i2c
 init

Instead of using i2c_add_adapter() and i2c_del_adapter(), replace them
with devm_i2c_add_adapter() to simplify the i2c logic.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c    | 15 ++-------------
 1 file changed, 2 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
index b067147b7c41..18d5d0704509 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
@@ -1641,33 +1641,22 @@ static int aldebaran_i2c_control_init(struct smu_context *smu)
 	control->quirks = &aldebaran_i2c_control_quirks;
 	i2c_set_adapdata(control, smu_i2c);
 
-	res = i2c_add_adapter(control);
+	res = devm_i2c_add_adapter(adev->dev, control);
 	if (res) {
 		DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-		goto Out_err;
+		return res;
 	}
 
 	adev->pm.ras_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 
 	return 0;
-Out_err:
-	i2c_del_adapter(control);
-
-	return res;
 }
 
 static void aldebaran_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From 439158c47512910054d28a5b6d6b01280c089ab8 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:38 -0600
Subject: [PATCH 1334/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the
 Arcturus smu

The I2C init for Arcturus uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that Arcturus init initializes multiple
I2C buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load is interrupted. Since I2C init is required
for the correct load of amdgpu, it is safe to rely on
devm_i2c_add_adapter() to handle any previously initialized I2C adapter.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
index 599eddb5a67d..4fff78da81ff 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c
@@ -1745,10 +1745,10 @@ static int arcturus_i2c_control_init(struct smu_context *smu)
 		snprintf(control->name, sizeof(control->name), "AMDGPU SMU %d", i);
 		i2c_set_adapdata(control, smu_i2c);
 
-		res = i2c_add_adapter(control);
+		res = devm_i2c_add_adapter(adev->dev, control);
 		if (res) {
 			DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-			goto Out_err;
+			return res;
 		}
 	}
 
@@ -1756,27 +1756,12 @@ static int arcturus_i2c_control_init(struct smu_context *smu)
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[1].adapter;
 
 	return 0;
-Out_err:
-	for ( ; i >= 0; i--) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
-	return res;
 }
 
 static void arcturus_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From 9058cb7775083bbbc2795439be8cf18f98098c49 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:39 -0600
Subject: [PATCH 1335/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the
 Navi10 smu

The I2C init for Navi10 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that Navi10 init initializes multiple I2C
buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load is interrupted. Since I2C init is required
for the correct load of amdgpu, it is safe to rely on
devm_i2c_add_adapter() to handle any previously initialized I2C adapter.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c   | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
index aac202d0c30e..0028f10ead42 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c
@@ -3145,10 +3145,10 @@ static int navi10_i2c_control_init(struct smu_context *smu)
 		control->quirks = &navi10_i2c_control_quirks;
 		i2c_set_adapdata(control, smu_i2c);
 
-		res = i2c_add_adapter(control);
+		res = devm_i2c_add_adapter(adev->dev, control);
 		if (res) {
 			DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-			goto Out_err;
+			return res;
 		}
 	}
 
@@ -3156,27 +3156,12 @@ static int navi10_i2c_control_init(struct smu_context *smu)
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[1].adapter;
 
 	return 0;
-Out_err:
-	for ( ; i >= 0; i--) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
-	return res;
 }
 
 static void navi10_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From 13f785d37aea745de0e722c81ec67995456577fd Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:40 -0600
Subject: [PATCH 1336/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the
 Sienna smu

The I2C init for Sienna Cichlid uses i2c_add_adapter() and
i2c_del_adapter(), this commit replaces the use of these two functions
with devm_i2c_add_adapter(). Notice that Sienna Cichlid init initializes
multiple I2C buses in a loop; if something goes wrong, the previous
adapters are removed, and the amdgpu load is interrupted. Since I2C init
is required for the correct load of amdgpu, it is safe to rely on
devm_i2c_add_adapter() to handle any previously initialized I2C adapter.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/pm/swsmu/smu11/sienna_cichlid_ppt.c   | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
index d57591509aed..31c2c0386b1f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c
@@ -2648,10 +2648,10 @@ static int sienna_cichlid_i2c_control_init(struct smu_context *smu)
 		control->quirks = &sienna_cichlid_i2c_control_quirks;
 		i2c_set_adapdata(control, smu_i2c);
 
-		res = i2c_add_adapter(control);
+		res = devm_i2c_add_adapter(adev->dev, control);
 		if (res) {
 			DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-			goto Out_err;
+			return res;
 		}
 	}
 	/* assign the buses used for the FRU EEPROM and RAS EEPROM */
@@ -2660,27 +2660,12 @@ static int sienna_cichlid_i2c_control_init(struct smu_context *smu)
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 
 	return 0;
-Out_err:
-	for ( ; i >= 0; i--) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
-	return res;
 }
 
 static void sienna_cichlid_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From 4970883abd31e4da100a3bf73a74b41517fc552e Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:41 -0600
Subject: [PATCH 1337/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the V13
 smu

The I2C init for SMU_V13 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that SMU_V13 init initializes multiple
I2C buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load is interrupted. Since I2C init is required
for the correct load of amdgpu, it is safe to rely on
devm_i2c_add_adapter() to handle any previously initialized I2C adapter.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c  | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
index e084ed99ec0e..c1062e5f0393 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c
@@ -2825,10 +2825,10 @@ static int smu_v13_0_0_i2c_control_init(struct smu_context *smu)
 		control->quirks = &smu_v13_0_0_i2c_control_quirks;
 		i2c_set_adapdata(control, smu_i2c);
 
-		res = i2c_add_adapter(control);
+		res = devm_i2c_add_adapter(adev->dev, control);
 		if (res) {
 			DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-			goto Out_err;
+			return res;
 		}
 	}
 
@@ -2838,27 +2838,12 @@ static int smu_v13_0_0_i2c_control_init(struct smu_context *smu)
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 
 	return 0;
-Out_err:
-	for ( ; i >= 0; i--) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
-	return res;
 }
 
 static void smu_v13_0_0_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From c32da00612baa106d7cfea707411b1510239e66a Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:42 -0600
Subject: [PATCH 1338/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the
 V13_0_6 smu

The I2C init for V13_0_6 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that V13_0_6 init initializes multiple
I2C buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load is interrupted. Since I2C init is required
for the correct load of amdgpu, it is safe to rely on
devm_i2c_add_adapter() to handle any previously initialized I2C adapter.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index bcd6525ca5e9..fd1d8e48273f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -2494,10 +2494,10 @@ static int smu_v13_0_6_i2c_control_init(struct smu_context *smu)
 		control->quirks = &smu_v13_0_6_i2c_control_quirks;
 		i2c_set_adapdata(control, smu_i2c);
 
-		res = i2c_add_adapter(control);
+		res = devm_i2c_add_adapter(adev->dev, control);
 		if (res) {
 			DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-			goto Out_err;
+			return res;
 		}
 	}
 
@@ -2505,27 +2505,12 @@ static int smu_v13_0_6_i2c_control_init(struct smu_context *smu)
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 
 	return 0;
-Out_err:
-	for ( ; i >= 0; i--) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
-	return res;
 }
 
 static void smu_v13_0_6_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From 49e957b2899c2f1c4ea44527727e46923d057a86 Mon Sep 17 00:00:00 2001
From: Rodrigo Siqueira <siqueira@igalia.com>
Date: Wed, 10 Sep 2025 11:39:43 -0600
Subject: [PATCH 1339/1455] drm/amd/pm: Use devm_i2c_add_adapter() in the
 V14_0_2 smu

The I2C init for V14_0_2 uses i2c_add_adapter() and i2c_del_adapter(),
this commit replaces the use of these two functions with
devm_i2c_add_adapter(). Notice that V14_0_2 init initializes multiple
I2C buses in a loop; if something goes wrong, the previous adapters are
removed, and the amdgpu load is interrupted. Since I2C init is required
for the correct load of amdgpu, it is safe to rely on
devm_i2c_add_adapter() to handle any previously initialized I2C adapter.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c  | 19 ++-----------------
 1 file changed, 2 insertions(+), 17 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
index f32474af90b3..086501cc5213 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c
@@ -2087,10 +2087,10 @@ static int smu_v14_0_2_i2c_control_init(struct smu_context *smu)
 		control->quirks = &smu_v14_0_2_i2c_control_quirks;
 		i2c_set_adapdata(control, smu_i2c);
 
-		res = i2c_add_adapter(control);
+		res = devm_i2c_add_adapter(adev->dev, control);
 		if (res) {
 			DRM_ERROR("Failed to register hw i2c, err: %d\n", res);
-			goto Out_err;
+			return res;
 		}
 	}
 
@@ -2100,27 +2100,12 @@ static int smu_v14_0_2_i2c_control_init(struct smu_context *smu)
 	adev->pm.fru_eeprom_i2c_bus = &adev->pm.smu_i2c[0].adapter;
 
 	return 0;
-Out_err:
-	for ( ; i >= 0; i--) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
-	return res;
 }
 
 static void smu_v14_0_2_i2c_control_fini(struct smu_context *smu)
 {
 	struct amdgpu_device *adev = smu->adev;
-	int i;
 
-	for (i = 0; i < MAX_SMU_I2C_BUSES; i++) {
-		struct amdgpu_smu_i2c_bus *smu_i2c = &adev->pm.smu_i2c[i];
-		struct i2c_adapter *control = &smu_i2c->adapter;
-
-		i2c_del_adapter(control);
-	}
 	adev->pm.ras_eeprom_i2c_bus = NULL;
 	adev->pm.fru_eeprom_i2c_bus = NULL;
 }

From 1a4dd33cc6e1baaa81efdbe68227a19f51c50f20 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Mon, 15 Sep 2025 20:59:02 -0500
Subject: [PATCH 1340/1455] drm/amd: Only restore cached manual clock settings
 in restore if OD enabled
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

If OD is not enabled then restoring cached clock settings doesn't make
sense and actually leads to errors in resume.

Check if enabled before restoring settings.

Fixes: 4e9526924d09 ("drm/amd: Restore cached manual clock settings during resume")
Reported-by: Jérôme Lécuyer <jerome.4a4c@gmail.com>
Closes: https://lore.kernel.org/amd-gfx/0ffe2692-7bfa-4821-856e-dd0f18e2c32b@amd.com/T/#me6db8ddb192626360c462b7570ed7eba0c6c9733
Suggested-by: Jérôme Lécuyer <jerome.4a4c@gmail.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
index bf2b38dd7e25..fb8086859857 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c
@@ -2263,7 +2263,7 @@ static int smu_resume(struct amdgpu_ip_block *ip_block)
 			return ret;
 	}
 
-	if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL) {
+	if (smu_dpm_ctx->dpm_level == AMD_DPM_FORCED_LEVEL_MANUAL && smu->od_enabled) {
 		ret = smu_od_edit_dpm_table(smu, PP_OD_COMMIT_DPM_TABLE, NULL, 0);
 		if (ret)
 			return ret;

From e09a4fdfbb1cf21a6f32f03497f3307b03653b36 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Mon, 15 Sep 2025 17:42:44 +0800
Subject: [PATCH 1341/1455] drm/amd/pm: Rename amdgpu_hwmon_get_sensor_generic

Rename amdgpu_hwmon_get_sensor_generic to use for generic pm
interfaces

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/amdgpu_pm.c | 76 +++++++++++++++---------------
 1 file changed, 39 insertions(+), 37 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 3e935a89d279..57821e59d3cc 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -1421,9 +1421,9 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev,
 	return -EINVAL;
 }
 
-static int amdgpu_hwmon_get_sensor_generic(struct amdgpu_device *adev,
-					   enum amd_pp_sensors sensor,
-					   void *query)
+static int amdgpu_pm_get_sensor_generic(struct amdgpu_device *adev,
+					enum amd_pp_sensors sensor,
+					void *query)
 {
 	int r, size = sizeof(uint32_t);
 
@@ -1456,7 +1456,7 @@ static ssize_t amdgpu_get_gpu_busy_percent(struct device *dev,
 	unsigned int value;
 	int r;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_LOAD, &value);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_LOAD, &value);
 	if (r)
 		return r;
 
@@ -1480,7 +1480,7 @@ static ssize_t amdgpu_get_mem_busy_percent(struct device *dev,
 	unsigned int value;
 	int r;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MEM_LOAD, &value);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MEM_LOAD, &value);
 	if (r)
 		return r;
 
@@ -1504,7 +1504,7 @@ static ssize_t amdgpu_get_vcn_busy_percent(struct device *dev,
 	unsigned int value;
 	int r;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VCN_LOAD, &value);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VCN_LOAD, &value);
 	if (r)
 		return r;
 
@@ -1783,7 +1783,7 @@ static int amdgpu_show_powershift_percent(struct device *dev,
 	uint32_t ss_power;
 	int r = 0, i;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, sensor, (void *)&ss_power);
+	r = amdgpu_pm_get_sensor_generic(adev, sensor, (void *)&ss_power);
 	if (r == -EOPNOTSUPP) {
 		/* sensor not available on dGPU, try to read from APU */
 		adev = NULL;
@@ -1796,7 +1796,7 @@ static int amdgpu_show_powershift_percent(struct device *dev,
 		}
 		mutex_unlock(&mgpu_info.mutex);
 		if (adev)
-			r = amdgpu_hwmon_get_sensor_generic(adev, sensor, (void *)&ss_power);
+			r = amdgpu_pm_get_sensor_generic(adev, sensor, (void *)&ss_power);
 	}
 
 	if (r)
@@ -1906,11 +1906,11 @@ static int ss_bias_attr_update(struct amdgpu_device *adev, struct amdgpu_device_
 
 	if (!amdgpu_device_supports_smart_shift(adev))
 		*states = ATTR_STATE_UNSUPPORTED;
-	else if (amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_APU_SHARE,
-		 (void *)&ss_power))
+	else if (amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_APU_SHARE,
+					      (void *)&ss_power))
 		*states = ATTR_STATE_UNSUPPORTED;
-	else if (amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_DGPU_SHARE,
-		 (void *)&ss_power))
+	else if (amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_SS_DGPU_SHARE,
+					      (void *)&ss_power))
 		*states = ATTR_STATE_UNSUPPORTED;
 
 	return 0;
@@ -2636,18 +2636,18 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev,
 	switch (channel) {
 	case PP_TEMP_JUNCTION:
 		/* get current junction temperature */
-		r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_HOTSPOT_TEMP,
-					   (void *)&temp);
+		r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_HOTSPOT_TEMP,
+						 (void *)&temp);
 		break;
 	case PP_TEMP_EDGE:
 		/* get current edge temperature */
-		r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_EDGE_TEMP,
-					   (void *)&temp);
+		r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_EDGE_TEMP,
+						 (void *)&temp);
 		break;
 	case PP_TEMP_MEM:
 		/* get current memory temperature */
-		r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MEM_TEMP,
-					   (void *)&temp);
+		r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MEM_TEMP,
+						 (void *)&temp);
 		break;
 	default:
 		r = -EINVAL;
@@ -2909,8 +2909,8 @@ static ssize_t amdgpu_hwmon_get_fan1_min(struct device *dev,
 	u32 min_rpm = 0;
 	int r;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MIN_FAN_RPM,
-				   (void *)&min_rpm);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MIN_FAN_RPM,
+					 (void *)&min_rpm);
 
 	if (r)
 		return r;
@@ -2926,8 +2926,8 @@ static ssize_t amdgpu_hwmon_get_fan1_max(struct device *dev,
 	u32 max_rpm = 0;
 	int r;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MAX_FAN_RPM,
-				   (void *)&max_rpm);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MAX_FAN_RPM,
+					 (void *)&max_rpm);
 
 	if (r)
 		return r;
@@ -3060,8 +3060,8 @@ static ssize_t amdgpu_hwmon_show_vddgfx(struct device *dev,
 	int r;
 
 	/* get the voltage */
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDGFX,
-				   (void *)&vddgfx);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDGFX,
+					 (void *)&vddgfx);
 	if (r)
 		return r;
 
@@ -3077,8 +3077,8 @@ static ssize_t amdgpu_hwmon_show_vddboard(struct device *dev,
 	int r;
 
 	/* get the voltage */
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDBOARD,
-					    (void *)&vddboard);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDBOARD,
+					 (void *)&vddboard);
 	if (r)
 		return r;
 
@@ -3111,8 +3111,8 @@ static ssize_t amdgpu_hwmon_show_vddnb(struct device *dev,
 		return -EINVAL;
 
 	/* get the voltage */
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDNB,
-				   (void *)&vddnb);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDNB,
+					 (void *)&vddnb);
 	if (r)
 		return r;
 
@@ -3134,7 +3134,7 @@ static int amdgpu_hwmon_get_power(struct device *dev,
 	u32 query = 0;
 	int r;
 
-	r = amdgpu_hwmon_get_sensor_generic(adev, sensor, (void *)&query);
+	r = amdgpu_pm_get_sensor_generic(adev, sensor, (void *)&query);
 	if (r)
 		return r;
 
@@ -3284,8 +3284,8 @@ static ssize_t amdgpu_hwmon_show_sclk(struct device *dev,
 	int r;
 
 	/* get the sclk */
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GFX_SCLK,
-				   (void *)&sclk);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GFX_SCLK,
+					 (void *)&sclk);
 	if (r)
 		return r;
 
@@ -3308,8 +3308,8 @@ static ssize_t amdgpu_hwmon_show_mclk(struct device *dev,
 	int r;
 
 	/* get the sclk */
-	r = amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GFX_MCLK,
-				   (void *)&mclk);
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GFX_MCLK,
+					 (void *)&mclk);
 	if (r)
 		return r;
 
@@ -3607,10 +3607,12 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 
 	/* not all products support both average and instantaneous */
 	if (attr == &sensor_dev_attr_power1_average.dev_attr.attr &&
-	    amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER, (void *)&tmp) == -EOPNOTSUPP)
+	    amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_AVG_POWER,
+					 (void *)&tmp) == -EOPNOTSUPP)
 		return 0;
 	if (attr == &sensor_dev_attr_power1_input.dev_attr.attr &&
-	    amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER, (void *)&tmp) == -EOPNOTSUPP)
+	    amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPU_INPUT_POWER,
+					 (void *)&tmp) == -EOPNOTSUPP)
 		return 0;
 
 	/* hide max/min values if we can't both query and manage the fan */
@@ -3649,8 +3651,8 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj,
 	/* only few boards support vddboard */
 	if ((attr == &sensor_dev_attr_in2_input.dev_attr.attr ||
 	     attr == &sensor_dev_attr_in2_label.dev_attr.attr) &&
-	     amdgpu_hwmon_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDBOARD,
-					     (void *)&tmp) == -EOPNOTSUPP)
+	     amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_VDDBOARD,
+					  (void *)&tmp) == -EOPNOTSUPP)
 		return 0;
 
 	/* no mclk on APUs other than gc 9,4,3*/

From beae798b6829dbecda7fdcdb3e466b1059976960 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Wed, 27 Aug 2025 18:19:13 +0800
Subject: [PATCH 1342/1455] drm/amd/pm: Update pmfw headers for smu_v13_0_12

Update pmfw headers for smu_v13_0_12 to include node power limit

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h
index 1c407a8e96ee..bf6aa9620911 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_pmfw.h
@@ -191,7 +191,7 @@ typedef enum {
 
 #define SMU_METRICS_TABLE_VERSION 0x14
 
-#define SMU_SYSTEM_METRICS_TABLE_VERSION 0x0
+#define SMU_SYSTEM_METRICS_TABLE_VERSION 0x1
 
 typedef struct __attribute__((packed, aligned(4))) {
   uint64_t AccumulationCounter;
@@ -304,7 +304,12 @@ typedef struct {
   int16_t  SystemTemperatures[SYSTEM_TEMP_MAX_ENTRIES];     // Signed integer temperature value in Celsius, unused fields are set to 0xFFFF
   int16_t  NodeTemperatures[NODE_TEMP_MAX_TEMP_ENTRIES];    // Signed integer temperature value in Celsius, unused fields are set to 0xFFFF
   int16_t  VrTemperatures[SVI_MAX_TEMP_ENTRIES];            // Signed integer temperature value in Celsius
-  int16_t  spare[3];
+  int16_t  spare[7];
+
+  //NPM: NODE POWER MANAGEMENT
+  uint32_t NodePowerLimit;
+  uint32_t NodePower;
+  uint32_t GlobalPPTResidencyAcc;
 } SystemMetricsTable_t;
 #pragma pack(pop)
 
@@ -359,6 +364,9 @@ typedef struct {
 
   // General info
   uint32_t pldmVersion[2];
+
+  //Node Power Limit
+  uint32_t MaxNodePowerLimit;
 } StaticMetricsTable_t;
 #pragma pack(pop)
 

From 930595df251c24e44642651a1386f7ed53cffd10 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 27 Aug 2025 10:17:48 +0200
Subject: [PATCH 1343/1455] drm/amdgpu: remove check for BO reservation add
 assert instead
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

We should leave such checks to lockdep and not implement something
manually.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 9980c0cded94..d0c95fb0ef81 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -651,18 +651,7 @@ int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		spin_unlock(&vm->status_lock);
 
 		bo = bo_base->bo;
-
-		if (dma_resv_locking_ctx(bo->tbo.base.resv) != ticket) {
-			struct amdgpu_task_info *ti = amdgpu_vm_get_task_info_vm(vm);
-
-			pr_warn_ratelimited("Evicted user BO is not reserved\n");
-			if (ti) {
-				pr_warn_ratelimited("pid %d\n", ti->task.pid);
-				amdgpu_vm_put_task_info(ti);
-			}
-
-			return -EINVAL;
-		}
+		dma_resv_assert_held(bo->tbo.base.resv);
 
 		r = validate(param, bo);
 		if (r)

From df99f6d112493808c5e5d948ad482d267aad5b89 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 27 Aug 2025 11:45:45 +0200
Subject: [PATCH 1344/1455] drm/amdgpu: re-order and document VM code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Re-order fields in the VM structure and try to improve the
documentation a bit.

Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 30 ++++++++++++++++++++------
 1 file changed, 24 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 3409904b5c63..74e61e45778e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -349,12 +349,16 @@ struct amdgpu_vm {
 	/* Memory statistics for this vm, protected by status_lock */
 	struct amdgpu_mem_stats stats[__AMDGPU_PL_NUM];
 
+	/*
+	 * The following lists contain amdgpu_vm_bo_base objects for either
+	 * PDs, PTs or per VM BOs. The state transits are:
+	 *
+	 * evicted -> relocated (PDs, PTs) or moved (per VM BOs) -> idle
+	 */
+
 	/* Per-VM and PT BOs who needs a validation */
 	struct list_head	evicted;
 
-	/* BOs for user mode queues that need a validation */
-	struct list_head	evicted_user;
-
 	/* PT BOs which relocated and their parent need an update */
 	struct list_head	relocated;
 
@@ -364,15 +368,29 @@ struct amdgpu_vm {
 	/* All BOs of this VM not currently in the state machine */
 	struct list_head	idle;
 
+	/*
+	 * The following lists contain amdgpu_vm_bo_base objects for BOs which
+	 * have their own dma_resv object and not depend on the root PD. Their
+	 * state transits are:
+	 *
+	 * evicted_user or invalidated -> done
+	 */
+
+	/* BOs for user mode queues that need a validation */
+	struct list_head	evicted_user;
+
 	/* regular invalidated BOs, but not yet updated in the PT */
 	struct list_head	invalidated;
 
-	/* BO mappings freed, but not yet updated in the PT */
-	struct list_head	freed;
-
 	/* BOs which are invalidated, has been updated in the PTs */
 	struct list_head        done;
 
+	/*
+	 * This list contains amdgpu_bo_va_mapping objects which have been freed
+	 * but not updated in the PTs
+	 */
+	struct list_head	freed;
+
 	/* contains the page directory */
 	struct amdgpu_vm_bo_base     root;
 	struct dma_fence	*last_update;

From 626667321deb4c7a294725406faa3dd71c3d445d Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Tue, 9 Sep 2025 15:12:40 -0700
Subject: [PATCH 1345/1455] drm/xe: Fix error handling if PXP fails to start

Since the PXP start comes after __xe_exec_queue_init() has completed,
we need to cleanup what was done in that function in case of a PXP
start error.
__xe_exec_queue_init calls the submission backend init() function,
so we need to introduce an opposite for that. Unfortunately, while
we already have a fini() function pointer, it performs other
operations in addition to cleaning up what was done by the init().
Therefore, for clarity, the existing fini() has been renamed to
destroy(), while a new fini() has been added to only clean up what was
done by the init(), with the latter being called by the former (via
xe_exec_queue_fini).

Fixes: 72d479601d67 ("drm/xe/pxp/uapi: Add userspace and LRC support for PXP-using queues")
Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Cc: Matthew Brost <matthew.brost@intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250909221240.3711023-3-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/xe/xe_exec_queue.c           | 22 +++++---
 drivers/gpu/drm/xe/xe_exec_queue_types.h     |  8 ++-
 drivers/gpu/drm/xe/xe_execlist.c             | 25 +++++----
 drivers/gpu/drm/xe/xe_execlist_types.h       |  2 +-
 drivers/gpu/drm/xe/xe_guc_exec_queue_types.h |  4 +-
 drivers/gpu/drm/xe/xe_guc_submit.c           | 54 ++++++++++++--------
 6 files changed, 73 insertions(+), 42 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_exec_queue.c b/drivers/gpu/drm/xe/xe_exec_queue.c
index 063c89d981e5..37b2b93b73d6 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue.c
+++ b/drivers/gpu/drm/xe/xe_exec_queue.c
@@ -199,6 +199,16 @@ err_lrc:
 	return err;
 }
 
+static void __xe_exec_queue_fini(struct xe_exec_queue *q)
+{
+	int i;
+
+	q->ops->fini(q);
+
+	for (i = 0; i < q->width; ++i)
+		xe_lrc_put(q->lrc[i]);
+}
+
 struct xe_exec_queue *xe_exec_queue_create(struct xe_device *xe, struct xe_vm *vm,
 					   u32 logical_mask, u16 width,
 					   struct xe_hw_engine *hwe, u32 flags,
@@ -229,11 +239,13 @@ struct xe_exec_queue *xe_exec_queue_create(struct xe_device *xe, struct xe_vm *v
 	if (xe_exec_queue_uses_pxp(q)) {
 		err = xe_pxp_exec_queue_add(xe->pxp, q);
 		if (err)
-			goto err_post_alloc;
+			goto err_post_init;
 	}
 
 	return q;
 
+err_post_init:
+	__xe_exec_queue_fini(q);
 err_post_alloc:
 	__xe_exec_queue_free(q);
 	return ERR_PTR(err);
@@ -331,13 +343,11 @@ void xe_exec_queue_destroy(struct kref *ref)
 			xe_exec_queue_put(eq);
 	}
 
-	q->ops->fini(q);
+	q->ops->destroy(q);
 }
 
 void xe_exec_queue_fini(struct xe_exec_queue *q)
 {
-	int i;
-
 	/*
 	 * Before releasing our ref to lrc and xef, accumulate our run ticks
 	 * and wakeup any waiters.
@@ -346,9 +356,7 @@ void xe_exec_queue_fini(struct xe_exec_queue *q)
 	if (q->xef && atomic_dec_and_test(&q->xef->exec_queue.pending_removal))
 		wake_up_var(&q->xef->exec_queue.pending_removal);
 
-	for (i = 0; i < q->width; ++i)
-		xe_lrc_put(q->lrc[i]);
-
+	__xe_exec_queue_fini(q);
 	__xe_exec_queue_free(q);
 }
 
diff --git a/drivers/gpu/drm/xe/xe_exec_queue_types.h b/drivers/gpu/drm/xe/xe_exec_queue_types.h
index ba443a497b38..27b76cf9da89 100644
--- a/drivers/gpu/drm/xe/xe_exec_queue_types.h
+++ b/drivers/gpu/drm/xe/xe_exec_queue_types.h
@@ -181,8 +181,14 @@ struct xe_exec_queue_ops {
 	int (*init)(struct xe_exec_queue *q);
 	/** @kill: Kill inflight submissions for backend */
 	void (*kill)(struct xe_exec_queue *q);
-	/** @fini: Fini exec queue for submission backend */
+	/** @fini: Undoes the init() for submission backend */
 	void (*fini)(struct xe_exec_queue *q);
+	/**
+	 * @destroy: Destroy exec queue for submission backend. The backend
+	 * function must call xe_exec_queue_fini() (which will in turn call the
+	 * fini() backend function) to ensure the queue is properly cleaned up.
+	 */
+	void (*destroy)(struct xe_exec_queue *q);
 	/** @set_priority: Set priority for exec queue */
 	int (*set_priority)(struct xe_exec_queue *q,
 			    enum xe_exec_queue_priority priority);
diff --git a/drivers/gpu/drm/xe/xe_execlist.c b/drivers/gpu/drm/xe/xe_execlist.c
index 788f56b066b6..f83d421ac9d3 100644
--- a/drivers/gpu/drm/xe/xe_execlist.c
+++ b/drivers/gpu/drm/xe/xe_execlist.c
@@ -385,10 +385,20 @@ err_free:
 	return err;
 }
 
-static void execlist_exec_queue_fini_async(struct work_struct *w)
+static void execlist_exec_queue_fini(struct xe_exec_queue *q)
+{
+	struct xe_execlist_exec_queue *exl = q->execlist;
+
+	drm_sched_entity_fini(&exl->entity);
+	drm_sched_fini(&exl->sched);
+
+	kfree(exl);
+}
+
+static void execlist_exec_queue_destroy_async(struct work_struct *w)
 {
 	struct xe_execlist_exec_queue *ee =
-		container_of(w, struct xe_execlist_exec_queue, fini_async);
+		container_of(w, struct xe_execlist_exec_queue, destroy_async);
 	struct xe_exec_queue *q = ee->q;
 	struct xe_execlist_exec_queue *exl = q->execlist;
 	struct xe_device *xe = gt_to_xe(q->gt);
@@ -401,10 +411,6 @@ static void execlist_exec_queue_fini_async(struct work_struct *w)
 		list_del(&exl->active_link);
 	spin_unlock_irqrestore(&exl->port->lock, flags);
 
-	drm_sched_entity_fini(&exl->entity);
-	drm_sched_fini(&exl->sched);
-	kfree(exl);
-
 	xe_exec_queue_fini(q);
 }
 
@@ -413,10 +419,10 @@ static void execlist_exec_queue_kill(struct xe_exec_queue *q)
 	/* NIY */
 }
 
-static void execlist_exec_queue_fini(struct xe_exec_queue *q)
+static void execlist_exec_queue_destroy(struct xe_exec_queue *q)
 {
-	INIT_WORK(&q->execlist->fini_async, execlist_exec_queue_fini_async);
-	queue_work(system_unbound_wq, &q->execlist->fini_async);
+	INIT_WORK(&q->execlist->destroy_async, execlist_exec_queue_destroy_async);
+	queue_work(system_unbound_wq, &q->execlist->destroy_async);
 }
 
 static int execlist_exec_queue_set_priority(struct xe_exec_queue *q,
@@ -467,6 +473,7 @@ static const struct xe_exec_queue_ops execlist_exec_queue_ops = {
 	.init = execlist_exec_queue_init,
 	.kill = execlist_exec_queue_kill,
 	.fini = execlist_exec_queue_fini,
+	.destroy = execlist_exec_queue_destroy,
 	.set_priority = execlist_exec_queue_set_priority,
 	.set_timeslice = execlist_exec_queue_set_timeslice,
 	.set_preempt_timeout = execlist_exec_queue_set_preempt_timeout,
diff --git a/drivers/gpu/drm/xe/xe_execlist_types.h b/drivers/gpu/drm/xe/xe_execlist_types.h
index 415140936f11..92c4ba52db0c 100644
--- a/drivers/gpu/drm/xe/xe_execlist_types.h
+++ b/drivers/gpu/drm/xe/xe_execlist_types.h
@@ -42,7 +42,7 @@ struct xe_execlist_exec_queue {
 
 	bool has_run;
 
-	struct work_struct fini_async;
+	struct work_struct destroy_async;
 
 	enum xe_exec_queue_priority active_priority;
 	struct list_head active_link;
diff --git a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
index a3f421e2adc0..c30c0e3ccbbb 100644
--- a/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
+++ b/drivers/gpu/drm/xe/xe_guc_exec_queue_types.h
@@ -35,8 +35,8 @@ struct xe_guc_exec_queue {
 	struct xe_sched_msg static_msgs[MAX_STATIC_MSG_TYPE];
 	/** @lr_tdr: long running TDR worker */
 	struct work_struct lr_tdr;
-	/** @fini_async: do final fini async from this worker */
-	struct work_struct fini_async;
+	/** @destroy_async: do final destroy async from this worker */
+	struct work_struct destroy_async;
 	/** @resume_time: time of last resume */
 	u64 resume_time;
 	/** @state: GuC specific state for this xe_exec_queue */
diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
index 8cb8b93a37a5..53024eb5670b 100644
--- a/drivers/gpu/drm/xe/xe_guc_submit.c
+++ b/drivers/gpu/drm/xe/xe_guc_submit.c
@@ -1419,21 +1419,12 @@ rearm:
 	return DRM_GPU_SCHED_STAT_NO_HANG;
 }
 
-static void __guc_exec_queue_fini_async(struct work_struct *w)
+static void guc_exec_queue_fini(struct xe_exec_queue *q)
 {
-	struct xe_guc_exec_queue *ge =
-		container_of(w, struct xe_guc_exec_queue, fini_async);
-	struct xe_exec_queue *q = ge->q;
+	struct xe_guc_exec_queue *ge = q->guc;
 	struct xe_guc *guc = exec_queue_to_guc(q);
 
-	xe_pm_runtime_get(guc_to_xe(guc));
-	trace_xe_exec_queue_destroy(q);
-
 	release_guc_id(guc, q);
-	if (xe_exec_queue_is_lr(q))
-		cancel_work_sync(&ge->lr_tdr);
-	/* Confirm no work left behind accessing device structures */
-	cancel_delayed_work_sync(&ge->sched.base.work_tdr);
 	xe_sched_entity_fini(&ge->entity);
 	xe_sched_fini(&ge->sched);
 
@@ -1442,25 +1433,43 @@ static void __guc_exec_queue_fini_async(struct work_struct *w)
 	 * (timeline name).
 	 */
 	kfree_rcu(ge, rcu);
+}
+
+static void __guc_exec_queue_destroy_async(struct work_struct *w)
+{
+	struct xe_guc_exec_queue *ge =
+		container_of(w, struct xe_guc_exec_queue, destroy_async);
+	struct xe_exec_queue *q = ge->q;
+	struct xe_guc *guc = exec_queue_to_guc(q);
+
+	xe_pm_runtime_get(guc_to_xe(guc));
+	trace_xe_exec_queue_destroy(q);
+
+	if (xe_exec_queue_is_lr(q))
+		cancel_work_sync(&ge->lr_tdr);
+	/* Confirm no work left behind accessing device structures */
+	cancel_delayed_work_sync(&ge->sched.base.work_tdr);
+
 	xe_exec_queue_fini(q);
+
 	xe_pm_runtime_put(guc_to_xe(guc));
 }
 
-static void guc_exec_queue_fini_async(struct xe_exec_queue *q)
+static void guc_exec_queue_destroy_async(struct xe_exec_queue *q)
 {
 	struct xe_guc *guc = exec_queue_to_guc(q);
 	struct xe_device *xe = guc_to_xe(guc);
 
-	INIT_WORK(&q->guc->fini_async, __guc_exec_queue_fini_async);
+	INIT_WORK(&q->guc->destroy_async, __guc_exec_queue_destroy_async);
 
 	/* We must block on kernel engines so slabs are empty on driver unload */
 	if (q->flags & EXEC_QUEUE_FLAG_PERMANENT || exec_queue_wedged(q))
-		__guc_exec_queue_fini_async(&q->guc->fini_async);
+		__guc_exec_queue_destroy_async(&q->guc->destroy_async);
 	else
-		queue_work(xe->destroy_wq, &q->guc->fini_async);
+		queue_work(xe->destroy_wq, &q->guc->destroy_async);
 }
 
-static void __guc_exec_queue_fini(struct xe_guc *guc, struct xe_exec_queue *q)
+static void __guc_exec_queue_destroy(struct xe_guc *guc, struct xe_exec_queue *q)
 {
 	/*
 	 * Might be done from within the GPU scheduler, need to do async as we
@@ -1469,7 +1478,7 @@ static void __guc_exec_queue_fini(struct xe_guc *guc, struct xe_exec_queue *q)
 	 * this we and don't really care when everything is fini'd, just that it
 	 * is.
 	 */
-	guc_exec_queue_fini_async(q);
+	guc_exec_queue_destroy_async(q);
 }
 
 static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg)
@@ -1483,7 +1492,7 @@ static void __guc_exec_queue_process_msg_cleanup(struct xe_sched_msg *msg)
 	if (exec_queue_registered(q))
 		disable_scheduling_deregister(guc, q);
 	else
-		__guc_exec_queue_fini(guc, q);
+		__guc_exec_queue_destroy(guc, q);
 }
 
 static bool guc_exec_queue_allowed_to_change_state(struct xe_exec_queue *q)
@@ -1716,14 +1725,14 @@ static bool guc_exec_queue_try_add_msg(struct xe_exec_queue *q,
 #define STATIC_MSG_CLEANUP	0
 #define STATIC_MSG_SUSPEND	1
 #define STATIC_MSG_RESUME	2
-static void guc_exec_queue_fini(struct xe_exec_queue *q)
+static void guc_exec_queue_destroy(struct xe_exec_queue *q)
 {
 	struct xe_sched_msg *msg = q->guc->static_msgs + STATIC_MSG_CLEANUP;
 
 	if (!(q->flags & EXEC_QUEUE_FLAG_PERMANENT) && !exec_queue_wedged(q))
 		guc_exec_queue_add_msg(q, msg, CLEANUP);
 	else
-		__guc_exec_queue_fini(exec_queue_to_guc(q), q);
+		__guc_exec_queue_destroy(exec_queue_to_guc(q), q);
 }
 
 static int guc_exec_queue_set_priority(struct xe_exec_queue *q,
@@ -1853,6 +1862,7 @@ static const struct xe_exec_queue_ops guc_exec_queue_ops = {
 	.init = guc_exec_queue_init,
 	.kill = guc_exec_queue_kill,
 	.fini = guc_exec_queue_fini,
+	.destroy = guc_exec_queue_destroy,
 	.set_priority = guc_exec_queue_set_priority,
 	.set_timeslice = guc_exec_queue_set_timeslice,
 	.set_preempt_timeout = guc_exec_queue_set_preempt_timeout,
@@ -1874,7 +1884,7 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q)
 		if (exec_queue_extra_ref(q) || xe_exec_queue_is_lr(q))
 			xe_exec_queue_put(q);
 		else if (exec_queue_destroyed(q))
-			__guc_exec_queue_fini(guc, q);
+			__guc_exec_queue_destroy(guc, q);
 	}
 	if (q->guc->suspend_pending) {
 		set_exec_queue_suspended(q);
@@ -2203,7 +2213,7 @@ static void handle_deregister_done(struct xe_guc *guc, struct xe_exec_queue *q)
 	if (exec_queue_extra_ref(q) || xe_exec_queue_is_lr(q))
 		xe_exec_queue_put(q);
 	else
-		__guc_exec_queue_fini(guc, q);
+		__guc_exec_queue_destroy(guc, q);
 }
 
 int xe_guc_deregister_done_handler(struct xe_guc *guc, u32 *msg, u32 len)

From aaae483657ec9de5782169bc052dd6e8ee7f806c Mon Sep 17 00:00:00 2001
From: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Date: Tue, 9 Sep 2025 15:12:41 -0700
Subject: [PATCH 1346/1455] drm/xe: Allow error injection for
 xe_pxp_exec_queue_add

This will allow us to simulate this function returning an error like
we do for other functions called in the exec_queue_create path.

Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: John Harrison <John.C.Harrison@Intel.com>
Reviewed-by: John Harrison <John.C.Harrison@Intel.com>
Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
Link: https://lore.kernel.org/r/20250909221240.3711023-4-daniele.ceraolospurio@intel.com
---
 drivers/gpu/drm/xe/xe_pxp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/xe/xe_pxp.c b/drivers/gpu/drm/xe/xe_pxp.c
index 3d62008c99f1..bdbdbbf6a678 100644
--- a/drivers/gpu/drm/xe/xe_pxp.c
+++ b/drivers/gpu/drm/xe/xe_pxp.c
@@ -688,6 +688,7 @@ start:
 
 	return ret;
 }
+ALLOW_ERROR_INJECTION(xe_pxp_exec_queue_add, ERRNO);
 
 static void __pxp_exec_queue_remove(struct xe_pxp *pxp, struct xe_exec_queue *q, bool lock)
 {

From 5bb5258e357eb79719d1c998731b932c945cb52c Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 16 Sep 2025 19:16:45 +0200
Subject: [PATCH 1347/1455] drm/xe/tests: Add pre-GMDID IP descriptors to param
 generators

Recently introduced kunit parameter generators were based on
the existing arrays which have only GDMID-based IPs and didn't
take into account IP definitions from pre-GMDID era.

Add test only arrays with pre-GMDID IPs (as those will not change)
and extend param generators to start iterating over them.

 [ ] =================== xe_pci (2 subtests) ====================
 [ ] ==================== check_graphics_ip  ====================
 [ ] [PASSED] 12.00 Xe_LP
 [ ] [PASSED] 12.10 Xe_LP+
 [ ] [PASSED] 12.55 Xe_HPG
 [ ] [PASSED] 12.60 Xe_HPC
 [ ] [PASSED] 12.70 Xe_LPG
 [ ] [PASSED] 12.71 Xe_LPG
 [ ] [PASSED] 12.74 Xe_LPG+
 [ ] [PASSED] 20.01 Xe2_HPG
 [ ] [PASSED] 20.02 Xe2_HPG
 [ ] [PASSED] 20.04 Xe2_LPG
 [ ] [PASSED] 30.00 Xe3_LPG
 [ ] [PASSED] 30.01 Xe3_LPG
 [ ] [PASSED] 30.03 Xe3_LPG
 [ ] ================ [PASSED] check_graphics_ip ================
 [ ] ===================== check_media_ip  ======================
 [ ] [PASSED] 12.00 Xe_M
 [ ] [PASSED] 12.55 Xe_HPM
 [ ] [PASSED] 13.00 Xe_LPM+
 [ ] [PASSED] 13.01 Xe2_HPM
 [ ] [PASSED] 20.00 Xe2_LPM
 [ ] [PASSED] 30.00 Xe3_LPM
 [ ] [PASSED] 30.02 Xe3_LPM
 [ ] ================= [PASSED] check_media_ip ==================
 [ ] ===================== [PASSED] xe_pci ======================

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Matt Roper <matthew.d.roper@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250916171645.3335-1-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/tests/xe_pci.c | 35 +++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/gpu/drm/xe/tests/xe_pci.c b/drivers/gpu/drm/xe/tests/xe_pci.c
index f19b1a64128b..aa29ac759d5d 100644
--- a/drivers/gpu/drm/xe/tests/xe_pci.c
+++ b/drivers/gpu/drm/xe/tests/xe_pci.c
@@ -204,6 +204,27 @@ static void xe_ip_kunit_desc(const struct xe_ip *param, char *desc)
 		 param->verx100 / 100, param->verx100 % 100, param->name);
 }
 
+/*
+ * Pre-GMDID Graphics and Media IPs definitions.
+ *
+ * Mimic the way GMDID IPs are declared so the same
+ * param generator can be used for both
+ */
+static const struct xe_ip pre_gmdid_graphics_ips[] = {
+	graphics_ip_xelp,
+	graphics_ip_xelpp,
+	graphics_ip_xehpg,
+	graphics_ip_xehpc,
+};
+
+static const struct xe_ip pre_gmdid_media_ips[] = {
+	media_ip_xem,
+	media_ip_xehpm,
+};
+
+KUNIT_ARRAY_PARAM(pre_gmdid_graphics_ip, pre_gmdid_graphics_ips, xe_ip_kunit_desc);
+KUNIT_ARRAY_PARAM(pre_gmdid_media_ip, pre_gmdid_media_ips, xe_ip_kunit_desc);
+
 KUNIT_ARRAY_PARAM(graphics_ip, graphics_ips, xe_ip_kunit_desc);
 KUNIT_ARRAY_PARAM(media_ip, media_ips, xe_ip_kunit_desc);
 
@@ -232,6 +253,13 @@ KUNIT_ARRAY_PARAM(pci_id, pciidlist, xe_pci_id_kunit_desc);
  */
 const void *xe_pci_graphics_ip_gen_param(const void *prev, char *desc)
 {
+	const void *next = pre_gmdid_graphics_ip_gen_params(prev, desc);
+
+	if (next)
+		return next;
+	if (is_insidevar(prev, pre_gmdid_graphics_ips))
+		prev = NULL;
+
 	return graphics_ip_gen_params(prev, desc);
 }
 EXPORT_SYMBOL_IF_KUNIT(xe_pci_graphics_ip_gen_param);
@@ -249,6 +277,13 @@ EXPORT_SYMBOL_IF_KUNIT(xe_pci_graphics_ip_gen_param);
  */
 const void *xe_pci_media_ip_gen_param(const void *prev, char *desc)
 {
+	const void *next = pre_gmdid_media_ip_gen_params(prev, desc);
+
+	if (next)
+		return next;
+	if (is_insidevar(prev, pre_gmdid_media_ips))
+		prev = NULL;
+
 	return media_ip_gen_params(prev, desc);
 }
 EXPORT_SYMBOL_IF_KUNIT(xe_pci_media_ip_gen_param);

From 5959c4da17e01782ad09b226802ac1f1ef157a70 Mon Sep 17 00:00:00 2001
From: Shuicheng Lin <shuicheng.lin@intel.com>
Date: Thu, 11 Sep 2025 03:14:06 +0000
Subject: [PATCH 1348/1455] drm/xe: Misc refine for svm

These changes should have no functional impact.
1. Correct typo of "operation"in macro range_debug().
2. Combine 2 spin_lock() call in xe_svm_garbage_collector() into 1.
3. Drop redundant preferred_region_is_vram check in
   xe_svm_range_needs_migrate_to_vram().
4. Combine the devmem_possible check in xe_svm_handle_pagefault().
   need_vram includes the IS_DGFX() check, so there is no change for
   .devmem_only.

v2: revert !ctx.devmem_only change (Matt)
v3: rebase code and refine commit message.
v4: rebase code and refine commit message.

Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://lore.kernel.org/r/20250911031405.1371812-2-shuicheng.lin@intel.com
---
 drivers/gpu/drm/xe/xe_svm.c | 25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index 86e7c72c89f1..7f2f1f041f1d 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -50,11 +50,11 @@ static struct xe_vm *range_to_vm(struct drm_gpusvm_range *r)
 	return gpusvm_to_vm(r->gpusvm);
 }
 
-#define range_debug(r__, operaton__)					\
+#define range_debug(r__, operation__)					\
 	vm_dbg(&range_to_vm(&(r__)->base)->xe->drm,			\
 	       "%s: asid=%u, gpusvm=%p, vram=%d,%d, seqno=%lu, " \
 	       "start=0x%014lx, end=0x%014lx, size=%lu",		\
-	       (operaton__), range_to_vm(&(r__)->base)->usm.asid,	\
+	       (operation__), range_to_vm(&(r__)->base)->usm.asid,	\
 	       (r__)->base.gpusvm,					\
 	       xe_svm_range_in_vram((r__)) ? 1 : 0,			\
 	       xe_svm_range_has_vram_binding((r__)) ? 1 : 0,		\
@@ -347,8 +347,8 @@ static int xe_svm_garbage_collector(struct xe_vm *vm)
 	if (xe_vm_is_closed_or_banned(vm))
 		return -ENOENT;
 
-	spin_lock(&vm->svm.garbage_collector.lock);
 	for (;;) {
+		spin_lock(&vm->svm.garbage_collector.lock);
 		range = list_first_entry_or_null(&vm->svm.garbage_collector.range_list,
 						 typeof(*range),
 						 garbage_collector_link);
@@ -377,8 +377,6 @@ static int xe_svm_garbage_collector(struct xe_vm *vm)
 			else
 				return err;
 		}
-
-		spin_lock(&vm->svm.garbage_collector.lock);
 	}
 	spin_unlock(&vm->svm.garbage_collector.lock);
 
@@ -944,12 +942,12 @@ bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range, struct xe_vm
 
 	xe_assert(vm->xe, IS_DGFX(vm->xe));
 
-	if (preferred_region_is_vram && xe_svm_range_in_vram(range)) {
+	if (xe_svm_range_in_vram(range)) {
 		drm_info(&vm->xe->drm, "Range is already in VRAM\n");
 		return false;
 	}
 
-	if (preferred_region_is_vram && range_size < SZ_64K && !supports_4K_migration(vm->xe)) {
+	if (range_size < SZ_64K && !supports_4K_migration(vm->xe)) {
 		drm_dbg(&vm->xe->drm, "Platform doesn't support SZ_4K range migration\n");
 		return false;
 	}
@@ -1010,15 +1008,14 @@ static int __xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
 				     struct xe_gt *gt, u64 fault_addr,
 				     bool need_vram)
 {
+	int devmem_possible = IS_DGFX(vm->xe) &&
+		IS_ENABLED(CONFIG_DRM_XE_PAGEMAP);
 	struct drm_gpusvm_ctx ctx = {
 		.read_only = xe_vma_read_only(vma),
-		.devmem_possible = IS_DGFX(vm->xe) &&
-			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP),
-		.check_pages_threshold = IS_DGFX(vm->xe) &&
-			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ? SZ_64K : 0,
-		.devmem_only = need_vram && IS_ENABLED(CONFIG_DRM_XE_PAGEMAP),
-		.timeslice_ms = need_vram && IS_DGFX(vm->xe) &&
-			IS_ENABLED(CONFIG_DRM_XE_PAGEMAP) ?
+		.devmem_possible = devmem_possible,
+		.check_pages_threshold = devmem_possible ? SZ_64K : 0,
+		.devmem_only = need_vram && devmem_possible,
+		.timeslice_ms = need_vram && devmem_possible ?
 			vm->xe->atomic_svm_timeslice_ms : 0,
 	};
 	struct xe_validation_ctx vctx;

From 33fe111a35a40c62761cc4ee75509b50c598fa4f Mon Sep 17 00:00:00 2001
From: Shuicheng Lin <shuicheng.lin@intel.com>
Date: Thu, 11 Sep 2025 17:31:40 +0000
Subject: [PATCH 1349/1455] drm/xe/madvise: Fix ioctl argument check

It is "preferred_mem_loc" instead of "atomic" for the ATTR_PREFERRED_LOC
path.

Also include 2 minor changes with no functional impact.
1. Remove the redundant "attr.atomic_access" assignment.
2. Replace down_read_interruptible() with
   xe_svm_notifier_lock_interruptible() to pair with
   xe_svm_notifier_unlock().

Fixes: ada7486c5668 ("drm/xe: Implement madvise ioctl for xe")
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Link: https://lore.kernel.org/r/20250911173139.1405878-2-shuicheng.lin@intel.com
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
---
 drivers/gpu/drm/xe/xe_vm_madvise.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_vm_madvise.c b/drivers/gpu/drm/xe/xe_vm_madvise.c
index 153700743545..cad3cf627c3f 100644
--- a/drivers/gpu/drm/xe/xe_vm_madvise.c
+++ b/drivers/gpu/drm/xe/xe_vm_madvise.c
@@ -124,8 +124,6 @@ static void madvise_atomic(struct xe_device *xe, struct xe_vm *vm,
 			vmas[i]->attr.atomic_access = op->atomic.val;
 		}
 
-		vmas[i]->attr.atomic_access = op->atomic.val;
-
 		bo = xe_vma_bo(vmas[i]);
 		if (!bo || bo->attr.atomic_access == op->atomic.val)
 			continue;
@@ -253,7 +251,7 @@ static bool madvise_args_are_sane(struct xe_device *xe, const struct drm_xe_madv
 		if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.pad))
 			return false;
 
-		if (XE_IOCTL_DBG(xe, args->atomic.reserved))
+		if (XE_IOCTL_DBG(xe, args->preferred_mem_loc.reserved))
 			return false;
 		break;
 	}
@@ -407,7 +405,7 @@ int xe_vm_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *fil
 	}
 
 	if (madvise_range.has_svm_userptr_vmas) {
-		err = down_read_interruptible(&vm->svm.gpusvm.notifier_lock);
+		err = xe_svm_notifier_lock_interruptible(vm);
 		if (err)
 			goto err_fini;
 	}

From a2d6223d224f333f705ed8495bf8bebfbc585c35 Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 16 Sep 2025 19:00:28 +0200
Subject: [PATCH 1350/1455] drm/xe/vf: Don't expose sysfs attributes not
 applicable for VFs

VFs can't read BMG_PCIE_CAP(0x138340) register nor access PCODE
(already guarded by the info.skip_pcode flag) so we shouldn't
expose attributes that require any of them to avoid errors like:

 [] xe 0000:03:00.1: [drm] Tile0: GT0: VF is trying to read an \
                     inaccessible register 0x138340+0x0
 [] RIP: 0010:xe_gt_sriov_vf_read32+0x6c2/0x9a0 [xe]
 [] Call Trace:
 []  xe_mmio_read32+0x110/0x280 [xe]
 []  auto_link_downgrade_capable_show+0x2e/0x70 [xe]
 []  dev_attr_show+0x1a/0x70
 []  sysfs_kf_seq_show+0xaa/0x120
 []  kernfs_seq_show+0x41/0x60

Fixes: 0e414bf7ad01 ("drm/xe: Expose PCIe link downgrade attributes")
Fixes: cdc36b66cd41 ("drm/xe: Expose fan control and voltage regulator version")
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Lukasz Laguna <lukasz.laguna@intel.com>
Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250916170029.3313-2-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_device_sysfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index b7f8fcfed8d8..896484c8fbcc 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -308,7 +308,7 @@ int xe_device_sysfs_init(struct xe_device *xe)
 			return ret;
 	}
 
-	if (xe->info.platform == XE_BATTLEMAGE) {
+	if (xe->info.platform == XE_BATTLEMAGE && !IS_SRIOV_VF(xe)) {
 		ret = sysfs_create_files(&dev->kobj, auto_link_downgrade_attrs);
 		if (ret)
 			goto cleanup;

From fb3c27a69c4736a9a78bb344b56907f8fb172edc Mon Sep 17 00:00:00 2001
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
Date: Tue, 16 Sep 2025 19:00:29 +0200
Subject: [PATCH 1351/1455] drm/xe/sysfs: Simplify sysfs registration

Instead of manually maintaining each sysfs file define and use
attribute groups and register them using device managed function.
Then use is_visible() to filter-out unsupported attributes.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Raag Jadav <raag.jadav@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250916170029.3313-3-michal.wajdeczko@intel.com
---
 drivers/gpu/drm/xe/xe_device_sysfs.c | 104 +++++++++++----------------
 1 file changed, 41 insertions(+), 63 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_device_sysfs.c b/drivers/gpu/drm/xe/xe_device_sysfs.c
index 896484c8fbcc..c5151c86a98a 100644
--- a/drivers/gpu/drm/xe/xe_device_sysfs.c
+++ b/drivers/gpu/drm/xe/xe_device_sysfs.c
@@ -71,6 +71,15 @@ vram_d3cold_threshold_store(struct device *dev, struct device_attribute *attr,
 
 static DEVICE_ATTR_RW(vram_d3cold_threshold);
 
+static struct attribute *vram_attrs[] = {
+	&dev_attr_vram_d3cold_threshold.attr,
+	NULL
+};
+
+static const struct attribute_group vram_attr_group = {
+	.attrs = vram_attrs,
+};
+
 static ssize_t
 lb_fan_control_version_show(struct device *dev, struct device_attribute *attr, char *buf)
 {
@@ -149,8 +158,16 @@ out:
 }
 static DEVICE_ATTR_ADMIN_RO(lb_voltage_regulator_version);
 
-static int late_bind_create_files(struct device *dev)
+static struct attribute *late_bind_attrs[] = {
+	&dev_attr_lb_fan_control_version.attr,
+	&dev_attr_lb_voltage_regulator_version.attr,
+	NULL
+};
+
+static umode_t late_bind_attr_is_visible(struct kobject *kobj,
+					 struct attribute *attr, int n)
 {
+	struct device *dev = kobj_to_dev(kobj);
 	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
 	struct xe_tile *root = xe_device_get_root_tile(xe);
 	u32 cap = 0;
@@ -160,51 +177,25 @@ static int late_bind_create_files(struct device *dev)
 
 	ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0),
 			    &cap, NULL);
-	if (ret) {
-		if (ret == -ENXIO) {
-			drm_dbg(&xe->drm, "Late binding not supported by firmware\n");
-			ret = 0;
-		}
-		goto out;
-	}
-
-	if (REG_FIELD_GET(V1_FAN_SUPPORTED, cap)) {
-		ret = sysfs_create_file(&dev->kobj, &dev_attr_lb_fan_control_version.attr);
-		if (ret)
-			goto out;
-	}
-
-	if (REG_FIELD_GET(VR_PARAMS_SUPPORTED, cap))
-		ret = sysfs_create_file(&dev->kobj, &dev_attr_lb_voltage_regulator_version.attr);
-out:
 	xe_pm_runtime_put(xe);
-
-	return ret;
-}
-
-static void late_bind_remove_files(struct device *dev)
-{
-	struct xe_device *xe = pdev_to_xe_device(to_pci_dev(dev));
-	struct xe_tile *root = xe_device_get_root_tile(xe);
-	u32 cap = 0;
-	int ret;
-
-	xe_pm_runtime_get(xe);
-
-	ret = xe_pcode_read(root, PCODE_MBOX(PCODE_LATE_BINDING, GET_CAPABILITY_STATUS, 0),
-			    &cap, NULL);
 	if (ret)
-		goto out;
+		return 0;
 
-	if (REG_FIELD_GET(V1_FAN_SUPPORTED, cap))
-		sysfs_remove_file(&dev->kobj, &dev_attr_lb_fan_control_version.attr);
+	if (attr == &dev_attr_lb_fan_control_version.attr &&
+	    REG_FIELD_GET(V1_FAN_SUPPORTED, cap))
+		return attr->mode;
+	if (attr == &dev_attr_lb_voltage_regulator_version.attr &&
+	    REG_FIELD_GET(VR_PARAMS_SUPPORTED, cap))
+		return attr->mode;
 
-	if (REG_FIELD_GET(VR_PARAMS_SUPPORTED, cap))
-		sysfs_remove_file(&dev->kobj, &dev_attr_lb_voltage_regulator_version.attr);
-out:
-	xe_pm_runtime_put(xe);
+	return 0;
 }
 
+static const struct attribute_group late_bind_attr_group = {
+	.attrs = late_bind_attrs,
+	.is_visible = late_bind_attr_is_visible,
+};
+
 /**
  * DOC: PCIe Gen5 Limitations
  *
@@ -278,24 +269,15 @@ auto_link_downgrade_status_show(struct device *dev, struct device_attribute *att
 }
 static DEVICE_ATTR_ADMIN_RO(auto_link_downgrade_status);
 
-static const struct attribute *auto_link_downgrade_attrs[] = {
+static struct attribute *auto_link_downgrade_attrs[] = {
 	&dev_attr_auto_link_downgrade_capable.attr,
 	&dev_attr_auto_link_downgrade_status.attr,
 	NULL
 };
 
-static void xe_device_sysfs_fini(void *arg)
-{
-	struct xe_device *xe = arg;
-
-	if (xe->d3cold.capable)
-		sysfs_remove_file(&xe->drm.dev->kobj, &dev_attr_vram_d3cold_threshold.attr);
-
-	if (xe->info.platform == XE_BATTLEMAGE) {
-		sysfs_remove_files(&xe->drm.dev->kobj, auto_link_downgrade_attrs);
-		late_bind_remove_files(xe->drm.dev);
-	}
-}
+static const struct attribute_group auto_link_downgrade_attr_group = {
+	.attrs = auto_link_downgrade_attrs,
+};
 
 int xe_device_sysfs_init(struct xe_device *xe)
 {
@@ -303,24 +285,20 @@ int xe_device_sysfs_init(struct xe_device *xe)
 	int ret;
 
 	if (xe->d3cold.capable) {
-		ret = sysfs_create_file(&dev->kobj, &dev_attr_vram_d3cold_threshold.attr);
+		ret = devm_device_add_group(dev, &vram_attr_group);
 		if (ret)
 			return ret;
 	}
 
 	if (xe->info.platform == XE_BATTLEMAGE && !IS_SRIOV_VF(xe)) {
-		ret = sysfs_create_files(&dev->kobj, auto_link_downgrade_attrs);
+		ret = devm_device_add_group(dev, &auto_link_downgrade_attr_group);
 		if (ret)
-			goto cleanup;
+			return ret;
 
-		ret = late_bind_create_files(dev);
+		ret = devm_device_add_group(dev, &late_bind_attr_group);
 		if (ret)
-			goto cleanup;
+			return ret;
 	}
 
-	return devm_add_action_or_reset(dev, xe_device_sysfs_fini, xe);
-
-cleanup:
-	xe_device_sysfs_fini(xe);
-	return ret;
+	return 0;
 }

From 187e16f69de2ba30779b97ba8852b583fd7f5fad Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Thomas=20Hellstr=C3=B6m?= <thomas.hellstrom@linux.intel.com>
Date: Thu, 11 Sep 2025 10:03:24 +0200
Subject: [PATCH 1352/1455] drm/xe: Work around clang multiple goto-label error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When using drm_exec_retry_on_contention(), clang may consider
all labels for which we take addresses in a function as
potential retry goto targets, although strictly only one
is possible. It will then in some situations generate false
positive errors.

In this case, the compiler, for some architectures, consider the

might_lock(&m->job_mutex);

as a potential goto target from drm_exec_retry_on_contention(),
and errors.

Work around that by moving the xe_validate / drm_exec
transaction to a separate function.

v2:
- New commit message based on analysis of Nathan Chancellor

Fixes: 59eabff2a352 ("drm/xe: Convert xe_bo_create_pin_map() for exhaustive eviction")
Cc: Matthew Brost <matthew.brost@intel.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202509101853.nDmyxTEM-lkp@intel.com/
Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org> # build
Link: https://lore.kernel.org/r/20250911080324.180307-1-thomas.hellstrom@linux.intel.com
---
 drivers/gpu/drm/xe/xe_migrate.c | 29 +++++++++++++++++++----------
 1 file changed, 19 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 6fad5d469629..1d667fa36cf3 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -394,6 +394,24 @@ struct xe_migrate *xe_migrate_alloc(struct xe_tile *tile)
 	return m;
 }
 
+static int xe_migrate_lock_prepare_vm(struct xe_tile *tile, struct xe_migrate *m, struct xe_vm *vm)
+{
+	struct xe_device *xe = tile_to_xe(tile);
+	struct xe_validation_ctx ctx;
+	struct drm_exec exec;
+	int err = 0;
+
+	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {}, err) {
+		err = xe_vm_drm_exec_lock(vm, &exec);
+		drm_exec_retry_on_contention(&exec);
+		err = xe_migrate_prepare_vm(tile, m, vm, &exec);
+		drm_exec_retry_on_contention(&exec);
+		xe_validation_retry_on_oom(&ctx, &err);
+	}
+
+	return err;
+}
+
 /**
  * xe_migrate_init() - Initialize a migrate context
  * @m: The migration context
@@ -405,8 +423,6 @@ int xe_migrate_init(struct xe_migrate *m)
 	struct xe_tile *tile = m->tile;
 	struct xe_gt *primary_gt = tile->primary_gt;
 	struct xe_device *xe = tile_to_xe(tile);
-	struct xe_validation_ctx ctx;
-	struct drm_exec exec;
 	struct xe_vm *vm;
 	int err;
 
@@ -416,14 +432,7 @@ int xe_migrate_init(struct xe_migrate *m)
 	if (IS_ERR(vm))
 		return PTR_ERR(vm);
 
-	err = 0;
-	xe_validation_guard(&ctx, &xe->val, &exec, (struct xe_val_flags) {}, err) {
-		err = xe_vm_drm_exec_lock(vm, &exec);
-		drm_exec_retry_on_contention(&exec);
-		err = xe_migrate_prepare_vm(tile, m, vm, &exec);
-		drm_exec_retry_on_contention(&exec);
-		xe_validation_retry_on_oom(&ctx, &err);
-	}
+	err = xe_migrate_lock_prepare_vm(tile, m, vm);
 	if (err)
 		return err;
 

From 4072b16dd841fad544b3efac1a5d2f99682a1ee2 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Mon, 15 Sep 2025 17:53:19 +0800
Subject: [PATCH 1353/1455] drm/amd/pm: Allow system metrics table in 1vf mode

Allow fetching system metrics table in 1VF mode

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 2 +-
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 0bec12b348ce..f10228de416c 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -137,7 +137,7 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(SetThrottlingPolicy,                 PPSMC_MSG_SetThrottlingPolicy,             0),
 	MSG_MAP(ResetSDMA,                           PPSMC_MSG_ResetSDMA,                       0),
 	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           1),
-	MSG_MAP(GetSystemMetricsTable,               PPSMC_MSG_GetSystemMetricsTable,           0),
+	MSG_MAP(GetSystemMetricsTable,               PPSMC_MSG_GetSystemMetricsTable,           1),
 };
 
 int smu_v13_0_12_tables_init(struct smu_context *smu)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index fd1d8e48273f..4e2a2393fff5 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -356,6 +356,8 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu)
 	if (fw_ver >= 0x04560700) {
 		if (!amdgpu_sriov_vf(smu->adev))
 			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));
+		else if (fw_ver >= 0x04560900)
+			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));
 	} else {
 		smu_v13_0_12_tables_fini(smu);
 	}

From ef612f58d944ce4a468fa533335035a6e5163eda Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Wed, 27 Aug 2025 21:22:13 +0800
Subject: [PATCH 1354/1455] drm/amd/pm: Add sysfs node for node power

Add sysfs node to expose node power limit for smu_v13_0_12

v2: Remove support check from visible function (Kevin)

v3: Update comments (Kevin)
    Remove sysfs remove file, change format specifier
    for sysfs_emit, use attribute_group.name (Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/include/kgd_pp_interface.h    |   4 +
 drivers/gpu/drm/amd/pm/amdgpu_pm.c            | 151 +++++++++++++++++-
 2 files changed, 153 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
index 2f7e4b5bebf3..2b0cdb2a2775 100644
--- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h
@@ -162,6 +162,10 @@ enum amd_pp_sensors {
 	AMDGPU_PP_SENSOR_PEAK_PSTATE_SCLK,
 	AMDGPU_PP_SENSOR_PEAK_PSTATE_MCLK,
 	AMDGPU_PP_SENSOR_VCN_LOAD,
+	AMDGPU_PP_SENSOR_NODEPOWERLIMIT,
+	AMDGPU_PP_SENSOR_NODEPOWER,
+	AMDGPU_PP_SENSOR_GPPTRESIDENCY,
+	AMDGPU_PP_SENSOR_MAXNODEPOWERLIMIT,
 };
 
 enum amd_pp_task {
diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
index 57821e59d3cc..b5fbb0fd1dc0 100644
--- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c
@@ -2081,8 +2081,9 @@ static int pp_dpm_clk_default_attr_update(struct amdgpu_device *adev, struct amd
  * for user application to monitor various board reated attributes.
  *
  * The amdgpu driver provides a sysfs API for reporting board attributes. Presently,
- * only two types of attributes are reported, baseboard temperature and
- * gpu board temperature. Both of them are reported as binary files.
+ * seven types of attributes are reported. Baseboard temperature and
+ * gpu board temperature are reported as binary files. Npm status, current node power limit,
+ * max node power limit, node power and global ppt residency is reported as ASCII text file.
  *
  * * .. code-block:: console
  *
@@ -2090,6 +2091,15 @@ static int pp_dpm_clk_default_attr_update(struct amdgpu_device *adev, struct amd
  *
  *      hexdump /sys/bus/pci/devices/.../board/gpuboard_temp
  *
+ *      hexdump /sys/bus/pci/devices/.../board/npm_status
+ *
+ *      hexdump /sys/bus/pci/devices/.../board/cur_node_power_limit
+ *
+ *      hexdump /sys/bus/pci/devices/.../board/max_node_power_limit
+ *
+ *      hexdump /sys/bus/pci/devices/.../board/node_power
+ *
+ *      hexdump /sys/bus/pci/devices/.../board/global_ppt_resid
  */
 
 /**
@@ -2168,8 +2178,129 @@ out:
 	return size;
 }
 
+/**
+ * DOC: cur_node_power_limit
+ *
+ * The amdgpu driver provides a sysfs API for retrieving current node power limit.
+ * The file cur_node_power_limit is used for this.
+ */
+static ssize_t amdgpu_show_cur_node_power_limit(struct device *dev,
+						struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	u32 nplimit;
+	int r;
+
+	/* get the current node power limit */
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_NODEPOWERLIMIT,
+					 (void *)&nplimit);
+	if (r)
+		return r;
+
+	return sysfs_emit(buf, "%u\n", nplimit);
+}
+
+/**
+ * DOC: node_power
+ *
+ * The amdgpu driver provides a sysfs API for retrieving current node power.
+ * The file node_power is used for this.
+ */
+static ssize_t amdgpu_show_node_power(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	u32 npower;
+	int r;
+
+	/* get the node power */
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_NODEPOWER,
+					 (void *)&npower);
+	if (r)
+		return r;
+
+	return sysfs_emit(buf, "%u\n", npower);
+}
+
+/**
+ * DOC: npm_status
+ *
+ * The amdgpu driver provides a sysfs API for retrieving current node power management status.
+ * The file npm_status is used for this. It shows the status as enabled or disabled based on
+ * current node power value. If node power is zero, status is disabled else enabled.
+ */
+static ssize_t amdgpu_show_npm_status(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	u32 npower;
+	int r;
+
+	/* get the node power */
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_NODEPOWER,
+					 (void *)&npower);
+	if (r)
+		return r;
+
+	return sysfs_emit(buf, "%s\n", npower ? "enabled" : "disabled");
+}
+
+/**
+ * DOC: global_ppt_resid
+ *
+ * The amdgpu driver provides a sysfs API for retrieving global ppt residency.
+ * The file global_ppt_resid is used for this.
+ */
+static ssize_t amdgpu_show_global_ppt_resid(struct device *dev,
+					    struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	u32 gpptresid;
+	int r;
+
+	/* get the global ppt residency */
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_GPPTRESIDENCY,
+					 (void *)&gpptresid);
+	if (r)
+		return r;
+
+	return sysfs_emit(buf, "%u\n", gpptresid);
+}
+
+/**
+ * DOC: max_node_power_limit
+ *
+ * The amdgpu driver provides a sysfs API for retrieving maximum node power limit.
+ * The file max_node_power_limit is used for this.
+ */
+static ssize_t amdgpu_show_max_node_power_limit(struct device *dev,
+						struct device_attribute *attr, char *buf)
+{
+	struct drm_device *ddev = dev_get_drvdata(dev);
+	struct amdgpu_device *adev = drm_to_adev(ddev);
+	u32 max_nplimit;
+	int r;
+
+	/* get the max node power limit */
+	r = amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MAXNODEPOWERLIMIT,
+					 (void *)&max_nplimit);
+	if (r)
+		return r;
+
+	return sysfs_emit(buf, "%u\n", max_nplimit);
+}
+
 static DEVICE_ATTR(baseboard_temp, 0444, amdgpu_get_baseboard_temp_metrics, NULL);
 static DEVICE_ATTR(gpuboard_temp, 0444, amdgpu_get_gpuboard_temp_metrics, NULL);
+static DEVICE_ATTR(cur_node_power_limit, 0444, amdgpu_show_cur_node_power_limit, NULL);
+static DEVICE_ATTR(node_power, 0444, amdgpu_show_node_power, NULL);
+static DEVICE_ATTR(global_ppt_resid, 0444, amdgpu_show_global_ppt_resid, NULL);
+static DEVICE_ATTR(max_node_power_limit, 0444, amdgpu_show_max_node_power_limit, NULL);
+static DEVICE_ATTR(npm_status, 0444, amdgpu_show_npm_status, NULL);
 
 static struct attribute *board_attrs[] = {
 	&dev_attr_baseboard_temp.attr,
@@ -4534,6 +4665,7 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 {
 	enum amdgpu_sriov_vf_mode mode;
 	uint32_t mask = 0;
+	uint32_t tmp;
 	int ret;
 
 	if (adev->pm.sysfs_initialized)
@@ -4600,6 +4732,21 @@ int amdgpu_pm_sysfs_init(struct amdgpu_device *adev)
 					    &amdgpu_board_attr_group);
 		if (ret)
 			goto err_out0;
+		if (amdgpu_pm_get_sensor_generic(adev, AMDGPU_PP_SENSOR_MAXNODEPOWERLIMIT,
+						 (void *)&tmp) != -EOPNOTSUPP) {
+			sysfs_add_file_to_group(&adev->dev->kobj,
+						&dev_attr_cur_node_power_limit.attr,
+						amdgpu_board_attr_group.name);
+			sysfs_add_file_to_group(&adev->dev->kobj, &dev_attr_node_power.attr,
+						amdgpu_board_attr_group.name);
+			sysfs_add_file_to_group(&adev->dev->kobj, &dev_attr_global_ppt_resid.attr,
+						amdgpu_board_attr_group.name);
+			sysfs_add_file_to_group(&adev->dev->kobj,
+						&dev_attr_max_node_power_limit.attr,
+						amdgpu_board_attr_group.name);
+			sysfs_add_file_to_group(&adev->dev->kobj, &dev_attr_npm_status.attr,
+						amdgpu_board_attr_group.name);
+		}
 	}
 
 	adev->pm.sysfs_initialized = true;

From acae8ad69bb513fdc3c111b8794fd6a36bf147c0 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Fri, 29 Aug 2025 12:25:54 +0800
Subject: [PATCH 1355/1455] drm/amd/pm: Fetch npm data from system metrics
 table

Fetch npm data from system metrics table for smu_v13_0_12

v3: Remove intermittent type for npm data, remove node id check,
move npm caps check to npm_get_data function (Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c | 47 +++++++++++++++++++
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c  |  9 ++++
 .../drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h  |  5 ++
 3 files changed, 61 insertions(+)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index f10228de416c..1842a33b2bce 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -341,6 +341,9 @@ int smu_v13_0_12_setup_driver_pptable(struct smu_context *smu)
 			static_metrics->pldmVersion[0] != 0xFFFFFFFF)
 			smu->adev->firmware.pldm_version =
 				static_metrics->pldmVersion[0];
+		if (smu_v13_0_6_cap_supported(smu, SMU_CAP(NPM_METRICS)))
+			pptable->MaxNodePowerLimit =
+				SMUQ10_ROUND(static_metrics->MaxNodePowerLimit);
 		smu_v13_0_12_init_xgmi_data(smu, static_metrics);
 		pptable->Init = true;
 	}
@@ -580,6 +583,50 @@ static bool smu_v13_0_12_is_temp_metrics_supported(struct smu_context *smu,
 	return false;
 }
 
+int smu_v13_0_12_get_npm_data(struct smu_context *smu,
+			      enum amd_pp_sensors sensor,
+			      uint32_t *value)
+{
+	struct smu_table_context *smu_table = &smu->smu_table;
+	struct PPTable_t *pptable =
+		(struct PPTable_t *)smu_table->driver_pptable;
+	struct smu_table *tables = smu_table->tables;
+	SystemMetricsTable_t *metrics;
+	struct smu_table *sys_table;
+	int ret;
+
+	if (!smu_v13_0_6_cap_supported(smu, SMU_CAP(NPM_METRICS)))
+		return -EOPNOTSUPP;
+
+	if (sensor == AMDGPU_PP_SENSOR_MAXNODEPOWERLIMIT) {
+		*value = pptable->MaxNodePowerLimit;
+		return 0;
+	}
+
+	ret = smu_v13_0_12_get_system_metrics_table(smu);
+	if (ret)
+		return ret;
+
+	sys_table = &tables[SMU_TABLE_PMFW_SYSTEM_METRICS];
+	metrics = (SystemMetricsTable_t *)sys_table->cache.buffer;
+
+	switch (sensor) {
+	case AMDGPU_PP_SENSOR_NODEPOWERLIMIT:
+		*value = SMUQ10_ROUND(metrics->NodePowerLimit);
+		break;
+	case AMDGPU_PP_SENSOR_NODEPOWER:
+		*value = SMUQ10_ROUND(metrics->NodePower);
+		break;
+	case AMDGPU_PP_SENSOR_GPPTRESIDENCY:
+		*value = SMUQ10_ROUND(metrics->GlobalPPTResidencyAcc);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return ret;
+}
+
 static ssize_t smu_v13_0_12_get_temp_metrics(struct smu_context *smu,
 					     enum smu_temp_metric_type type, void *table)
 {
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 4e2a2393fff5..86c215894f49 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -1801,6 +1801,15 @@ static int smu_v13_0_6_read_sensor(struct smu_context *smu,
 			ret = -EOPNOTSUPP;
 			break;
 		}
+	case AMDGPU_PP_SENSOR_NODEPOWERLIMIT:
+	case AMDGPU_PP_SENSOR_NODEPOWER:
+	case AMDGPU_PP_SENSOR_GPPTRESIDENCY:
+	case AMDGPU_PP_SENSOR_MAXNODEPOWERLIMIT:
+		ret = smu_v13_0_12_get_npm_data(smu, sensor, (uint32_t *)data);
+		if (ret)
+			return ret;
+		*size = 4;
+		break;
 	case AMDGPU_PP_SENSOR_GPU_AVG_POWER:
 	default:
 		ret = -EOPNOTSUPP;
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
index 4652fcd5e068..7ef5f3e66c27 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.h
@@ -49,6 +49,7 @@ struct PPTable_t {
 	uint32_t MaxLclkDpmRange;
 	uint32_t MinLclkDpmRange;
 	uint64_t PublicSerialNumber_AID;
+	uint32_t MaxNodePowerLimit;
 	bool Init;
 };
 
@@ -70,6 +71,7 @@ enum smu_v13_0_6_caps {
 	SMU_CAP(BOARD_VOLTAGE),
 	SMU_CAP(PLDM_VERSION),
 	SMU_CAP(TEMP_METRICS),
+	SMU_CAP(NPM_METRICS),
 	SMU_CAP(ALL),
 };
 
@@ -91,6 +93,9 @@ ssize_t smu_v13_0_12_get_xcp_metrics(struct smu_context *smu,
 				     void *smu_metrics);
 int smu_v13_0_12_tables_init(struct smu_context *smu);
 void smu_v13_0_12_tables_fini(struct smu_context *smu);
+int smu_v13_0_12_get_npm_data(struct smu_context *smu,
+			      enum amd_pp_sensors sensor,
+			      uint32_t *value);
 extern const struct cmn2asic_mapping smu_v13_0_12_feature_mask_map[];
 extern const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[];
 extern const struct smu_temp_funcs smu_v13_0_12_temp_funcs;

From 34f10da667090c66a4efcda895dd7b1766c1e2a1 Mon Sep 17 00:00:00 2001
From: Asad Kamal <asad.kamal@amd.com>
Date: Mon, 15 Sep 2025 20:28:49 +0800
Subject: [PATCH 1356/1455] drm/amd/pm: Enable npm metrics data

Enable npm metrics data for smu_v13_0_12

v3: Add node id check for setting NPM_CAPS (Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 86c215894f49..349b6b8be010 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -354,9 +354,11 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu)
 	}
 
 	if (fw_ver >= 0x04560700) {
-		if (!amdgpu_sriov_vf(smu->adev))
+		if (fw_ver >= 0x04560900) {
 			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));
-		else if (fw_ver >= 0x04560900)
+			if (smu->adev->gmc.xgmi.physical_node_id == 0)
+				smu_v13_0_6_cap_set(smu, SMU_CAP(NPM_METRICS));
+		} else if (!amdgpu_sriov_vf(smu->adev))
 			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));
 	} else {
 		smu_v13_0_12_tables_fini(smu);

From 18f769ff368c6b133a41b44ae673c329eb5d9f46 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 20 Aug 2025 16:04:18 -0400
Subject: [PATCH 1357/1455] drm/amdgpu: remove non-DC DCE 11 code
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

DC has been the default for ~8 years now and supports
many things that the non-DC code does not (audio, DP MST, etc.).
No DCE 11.x IPs ever supported analog encoders so that is not
an issue.  Finally drop this code.

Acked-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/Makefile      |    1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c |    8 -
 drivers/gpu/drm/amd/amdgpu/dce_v11_0.c   | 3817 ----------------------
 drivers/gpu/drm/amd/amdgpu/dce_v11_0.h   |   32 -
 drivers/gpu/drm/amd/amdgpu/vi.c          |    7 -
 5 files changed, 3865 deletions(-)
 delete mode 100644 drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
 delete mode 100644 drivers/gpu/drm/amd/amdgpu/dce_v11_0.h

diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 2d0fea87af79..64e7acff8f18 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -138,7 +138,6 @@ amdgpu-y += \
 # add DCE block
 amdgpu-y += \
 	dce_v10_0.o \
-	dce_v11_0.o \
 	amdgpu_vkms.o
 
 # add GFX block
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
index 155bb9891a17..79bad9cbe2ab 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c
@@ -14,7 +14,6 @@
 #include "dce_v8_0.h"
 #endif
 #include "dce_v10_0.h"
-#include "dce_v11_0.h"
 #include "ivsrcid/ivsrcid_vislands30.h"
 #include "amdgpu_vkms.h"
 #include "amdgpu_display.h"
@@ -581,13 +580,6 @@ static int amdgpu_vkms_hw_init(struct amdgpu_ip_block *ip_block)
 	case CHIP_TONGA:
 		dce_v10_0_disable_dce(adev);
 		break;
-	case CHIP_CARRIZO:
-	case CHIP_STONEY:
-	case CHIP_POLARIS10:
-	case CHIP_POLARIS11:
-	case CHIP_VEGAM:
-		dce_v11_0_disable_dce(adev);
-		break;
 	case CHIP_TOPAZ:
 #ifdef CONFIG_DRM_AMDGPU_SI
 	case CHIP_HAINAN:
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
deleted file mode 100644
index e84608891300..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ /dev/null
@@ -1,3817 +0,0 @@
-/*
- * Copyright 2014 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- */
-
-#include <drm/drm_edid.h>
-#include <drm/drm_fourcc.h>
-#include <drm/drm_modeset_helper.h>
-#include <drm/drm_modeset_helper_vtables.h>
-#include <drm/drm_vblank.h>
-
-#include "amdgpu.h"
-#include "amdgpu_pm.h"
-#include "amdgpu_i2c.h"
-#include "vid.h"
-#include "atom.h"
-#include "amdgpu_atombios.h"
-#include "atombios_crtc.h"
-#include "atombios_encoders.h"
-#include "amdgpu_pll.h"
-#include "amdgpu_connectors.h"
-#include "amdgpu_display.h"
-#include "dce_v11_0.h"
-
-#include "dce/dce_11_0_d.h"
-#include "dce/dce_11_0_sh_mask.h"
-#include "dce/dce_11_0_enum.h"
-#include "oss/oss_3_0_d.h"
-#include "oss/oss_3_0_sh_mask.h"
-#include "gmc/gmc_8_1_d.h"
-#include "gmc/gmc_8_1_sh_mask.h"
-
-#include "ivsrcid/ivsrcid_vislands30.h"
-
-static void dce_v11_0_set_display_funcs(struct amdgpu_device *adev);
-static void dce_v11_0_set_irq_funcs(struct amdgpu_device *adev);
-static void dce_v11_0_hpd_int_ack(struct amdgpu_device *adev, int hpd);
-
-static const u32 crtc_offsets[] =
-{
-	CRTC0_REGISTER_OFFSET,
-	CRTC1_REGISTER_OFFSET,
-	CRTC2_REGISTER_OFFSET,
-	CRTC3_REGISTER_OFFSET,
-	CRTC4_REGISTER_OFFSET,
-	CRTC5_REGISTER_OFFSET,
-	CRTC6_REGISTER_OFFSET
-};
-
-static const u32 hpd_offsets[] =
-{
-	HPD0_REGISTER_OFFSET,
-	HPD1_REGISTER_OFFSET,
-	HPD2_REGISTER_OFFSET,
-	HPD3_REGISTER_OFFSET,
-	HPD4_REGISTER_OFFSET,
-	HPD5_REGISTER_OFFSET
-};
-
-static const uint32_t dig_offsets[] = {
-	DIG0_REGISTER_OFFSET,
-	DIG1_REGISTER_OFFSET,
-	DIG2_REGISTER_OFFSET,
-	DIG3_REGISTER_OFFSET,
-	DIG4_REGISTER_OFFSET,
-	DIG5_REGISTER_OFFSET,
-	DIG6_REGISTER_OFFSET,
-	DIG7_REGISTER_OFFSET,
-	DIG8_REGISTER_OFFSET
-};
-
-static const struct {
-	uint32_t        reg;
-	uint32_t        vblank;
-	uint32_t        vline;
-	uint32_t        hpd;
-
-} interrupt_status_offsets[] = { {
-	.reg = mmDISP_INTERRUPT_STATUS,
-	.vblank = DISP_INTERRUPT_STATUS__LB_D1_VBLANK_INTERRUPT_MASK,
-	.vline = DISP_INTERRUPT_STATUS__LB_D1_VLINE_INTERRUPT_MASK,
-	.hpd = DISP_INTERRUPT_STATUS__DC_HPD1_INTERRUPT_MASK
-}, {
-	.reg = mmDISP_INTERRUPT_STATUS_CONTINUE,
-	.vblank = DISP_INTERRUPT_STATUS_CONTINUE__LB_D2_VBLANK_INTERRUPT_MASK,
-	.vline = DISP_INTERRUPT_STATUS_CONTINUE__LB_D2_VLINE_INTERRUPT_MASK,
-	.hpd = DISP_INTERRUPT_STATUS_CONTINUE__DC_HPD2_INTERRUPT_MASK
-}, {
-	.reg = mmDISP_INTERRUPT_STATUS_CONTINUE2,
-	.vblank = DISP_INTERRUPT_STATUS_CONTINUE2__LB_D3_VBLANK_INTERRUPT_MASK,
-	.vline = DISP_INTERRUPT_STATUS_CONTINUE2__LB_D3_VLINE_INTERRUPT_MASK,
-	.hpd = DISP_INTERRUPT_STATUS_CONTINUE2__DC_HPD3_INTERRUPT_MASK
-}, {
-	.reg = mmDISP_INTERRUPT_STATUS_CONTINUE3,
-	.vblank = DISP_INTERRUPT_STATUS_CONTINUE3__LB_D4_VBLANK_INTERRUPT_MASK,
-	.vline = DISP_INTERRUPT_STATUS_CONTINUE3__LB_D4_VLINE_INTERRUPT_MASK,
-	.hpd = DISP_INTERRUPT_STATUS_CONTINUE3__DC_HPD4_INTERRUPT_MASK
-}, {
-	.reg = mmDISP_INTERRUPT_STATUS_CONTINUE4,
-	.vblank = DISP_INTERRUPT_STATUS_CONTINUE4__LB_D5_VBLANK_INTERRUPT_MASK,
-	.vline = DISP_INTERRUPT_STATUS_CONTINUE4__LB_D5_VLINE_INTERRUPT_MASK,
-	.hpd = DISP_INTERRUPT_STATUS_CONTINUE4__DC_HPD5_INTERRUPT_MASK
-}, {
-	.reg = mmDISP_INTERRUPT_STATUS_CONTINUE5,
-	.vblank = DISP_INTERRUPT_STATUS_CONTINUE5__LB_D6_VBLANK_INTERRUPT_MASK,
-	.vline = DISP_INTERRUPT_STATUS_CONTINUE5__LB_D6_VLINE_INTERRUPT_MASK,
-	.hpd = DISP_INTERRUPT_STATUS_CONTINUE5__DC_HPD6_INTERRUPT_MASK
-} };
-
-static const u32 cz_golden_settings_a11[] =
-{
-	mmCRTC_DOUBLE_BUFFER_CONTROL, 0x00010101, 0x00010000,
-	mmFBC_MISC, 0x1f311fff, 0x14300000,
-};
-
-static const u32 cz_mgcg_cgcg_init[] =
-{
-	mmXDMA_CLOCK_GATING_CNTL, 0xffffffff, 0x00000100,
-	mmXDMA_MEM_POWER_CNTL, 0x00000101, 0x00000000,
-};
-
-static const u32 stoney_golden_settings_a11[] =
-{
-	mmCRTC_DOUBLE_BUFFER_CONTROL, 0x00010101, 0x00010000,
-	mmFBC_MISC, 0x1f311fff, 0x14302000,
-};
-
-static const u32 polaris11_golden_settings_a11[] =
-{
-	mmDCI_CLK_CNTL, 0x00000080, 0x00000000,
-	mmFBC_DEBUG_COMP, 0x000000f0, 0x00000070,
-	mmFBC_DEBUG1, 0xffffffff, 0x00000008,
-	mmFBC_MISC, 0x9f313fff, 0x14302008,
-	mmHDMI_CONTROL, 0x313f031f, 0x00000011,
-};
-
-static const u32 polaris10_golden_settings_a11[] =
-{
-	mmDCI_CLK_CNTL, 0x00000080, 0x00000000,
-	mmFBC_DEBUG_COMP, 0x000000f0, 0x00000070,
-	mmFBC_MISC, 0x9f313fff, 0x14302008,
-	mmHDMI_CONTROL, 0x313f031f, 0x00000011,
-};
-
-static void dce_v11_0_init_golden_registers(struct amdgpu_device *adev)
-{
-	switch (adev->asic_type) {
-	case CHIP_CARRIZO:
-		amdgpu_device_program_register_sequence(adev,
-							cz_mgcg_cgcg_init,
-							ARRAY_SIZE(cz_mgcg_cgcg_init));
-		amdgpu_device_program_register_sequence(adev,
-							cz_golden_settings_a11,
-							ARRAY_SIZE(cz_golden_settings_a11));
-		break;
-	case CHIP_STONEY:
-		amdgpu_device_program_register_sequence(adev,
-							stoney_golden_settings_a11,
-							ARRAY_SIZE(stoney_golden_settings_a11));
-		break;
-	case CHIP_POLARIS11:
-	case CHIP_POLARIS12:
-		amdgpu_device_program_register_sequence(adev,
-							polaris11_golden_settings_a11,
-							ARRAY_SIZE(polaris11_golden_settings_a11));
-		break;
-	case CHIP_POLARIS10:
-	case CHIP_VEGAM:
-		amdgpu_device_program_register_sequence(adev,
-							polaris10_golden_settings_a11,
-							ARRAY_SIZE(polaris10_golden_settings_a11));
-		break;
-	default:
-		break;
-	}
-}
-
-static u32 dce_v11_0_audio_endpt_rreg(struct amdgpu_device *adev,
-				     u32 block_offset, u32 reg)
-{
-	unsigned long flags;
-	u32 r;
-
-	spin_lock_irqsave(&adev->audio_endpt_idx_lock, flags);
-	WREG32(mmAZALIA_F0_CODEC_ENDPOINT_INDEX + block_offset, reg);
-	r = RREG32(mmAZALIA_F0_CODEC_ENDPOINT_DATA + block_offset);
-	spin_unlock_irqrestore(&adev->audio_endpt_idx_lock, flags);
-
-	return r;
-}
-
-static void dce_v11_0_audio_endpt_wreg(struct amdgpu_device *adev,
-				      u32 block_offset, u32 reg, u32 v)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&adev->audio_endpt_idx_lock, flags);
-	WREG32(mmAZALIA_F0_CODEC_ENDPOINT_INDEX + block_offset, reg);
-	WREG32(mmAZALIA_F0_CODEC_ENDPOINT_DATA + block_offset, v);
-	spin_unlock_irqrestore(&adev->audio_endpt_idx_lock, flags);
-}
-
-static u32 dce_v11_0_vblank_get_counter(struct amdgpu_device *adev, int crtc)
-{
-	if (crtc < 0 || crtc >= adev->mode_info.num_crtc)
-		return 0;
-	else
-		return RREG32(mmCRTC_STATUS_FRAME_COUNT + crtc_offsets[crtc]);
-}
-
-static void dce_v11_0_pageflip_interrupt_init(struct amdgpu_device *adev)
-{
-	unsigned i;
-
-	/* Enable pflip interrupts */
-	for (i = 0; i < adev->mode_info.num_crtc; i++)
-		amdgpu_irq_get(adev, &adev->pageflip_irq, i);
-}
-
-static void dce_v11_0_pageflip_interrupt_fini(struct amdgpu_device *adev)
-{
-	unsigned i;
-
-	/* Disable pflip interrupts */
-	for (i = 0; i < adev->mode_info.num_crtc; i++)
-		amdgpu_irq_put(adev, &adev->pageflip_irq, i);
-}
-
-/**
- * dce_v11_0_page_flip - pageflip callback.
- *
- * @adev: amdgpu_device pointer
- * @crtc_id: crtc to cleanup pageflip on
- * @crtc_base: new address of the crtc (GPU MC address)
- * @async: asynchronous flip
- *
- * Triggers the actual pageflip by updating the primary
- * surface base address.
- */
-static void dce_v11_0_page_flip(struct amdgpu_device *adev,
-				int crtc_id, u64 crtc_base, bool async)
-{
-	struct amdgpu_crtc *amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
-	struct drm_framebuffer *fb = amdgpu_crtc->base.primary->fb;
-	u32 tmp;
-
-	/* flip immediate for async, default is vsync */
-	tmp = RREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, GRPH_FLIP_CONTROL,
-			    GRPH_SURFACE_UPDATE_IMMEDIATE_EN, async ? 1 : 0);
-	WREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-	/* update pitch */
-	WREG32(mmGRPH_PITCH + amdgpu_crtc->crtc_offset,
-	       fb->pitches[0] / fb->format->cpp[0]);
-	/* update the scanout addresses */
-	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(crtc_base));
-	/* writing to the low address triggers the update */
-	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(crtc_base));
-	/* post the write */
-	RREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset);
-}
-
-static int dce_v11_0_crtc_get_scanoutpos(struct amdgpu_device *adev, int crtc,
-					u32 *vbl, u32 *position)
-{
-	if ((crtc < 0) || (crtc >= adev->mode_info.num_crtc))
-		return -EINVAL;
-
-	*vbl = RREG32(mmCRTC_V_BLANK_START_END + crtc_offsets[crtc]);
-	*position = RREG32(mmCRTC_STATUS_POSITION + crtc_offsets[crtc]);
-
-	return 0;
-}
-
-/**
- * dce_v11_0_hpd_sense - hpd sense callback.
- *
- * @adev: amdgpu_device pointer
- * @hpd: hpd (hotplug detect) pin
- *
- * Checks if a digital monitor is connected (evergreen+).
- * Returns true if connected, false if not connected.
- */
-static bool dce_v11_0_hpd_sense(struct amdgpu_device *adev,
-			       enum amdgpu_hpd_id hpd)
-{
-	bool connected = false;
-
-	if (hpd >= adev->mode_info.num_hpd)
-		return connected;
-
-	if (RREG32(mmDC_HPD_INT_STATUS + hpd_offsets[hpd]) &
-	    DC_HPD_INT_STATUS__DC_HPD_SENSE_MASK)
-		connected = true;
-
-	return connected;
-}
-
-/**
- * dce_v11_0_hpd_set_polarity - hpd set polarity callback.
- *
- * @adev: amdgpu_device pointer
- * @hpd: hpd (hotplug detect) pin
- *
- * Set the polarity of the hpd pin (evergreen+).
- */
-static void dce_v11_0_hpd_set_polarity(struct amdgpu_device *adev,
-				      enum amdgpu_hpd_id hpd)
-{
-	u32 tmp;
-	bool connected = dce_v11_0_hpd_sense(adev, hpd);
-
-	if (hpd >= adev->mode_info.num_hpd)
-		return;
-
-	tmp = RREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd]);
-	if (connected)
-		tmp = REG_SET_FIELD(tmp, DC_HPD_INT_CONTROL, DC_HPD_INT_POLARITY, 0);
-	else
-		tmp = REG_SET_FIELD(tmp, DC_HPD_INT_CONTROL, DC_HPD_INT_POLARITY, 1);
-	WREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd], tmp);
-}
-
-/**
- * dce_v11_0_hpd_init - hpd setup callback.
- *
- * @adev: amdgpu_device pointer
- *
- * Setup the hpd pins used by the card (evergreen+).
- * Enable the pin, set the polarity, and enable the hpd interrupts.
- */
-static void dce_v11_0_hpd_init(struct amdgpu_device *adev)
-{
-	struct drm_device *dev = adev_to_drm(adev);
-	struct drm_connector *connector;
-	struct drm_connector_list_iter iter;
-	u32 tmp;
-
-	drm_connector_list_iter_begin(dev, &iter);
-	drm_for_each_connector_iter(connector, &iter) {
-		struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
-
-		if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
-			continue;
-
-		if (connector->connector_type == DRM_MODE_CONNECTOR_eDP ||
-		    connector->connector_type == DRM_MODE_CONNECTOR_LVDS) {
-			/* don't try to enable hpd on eDP or LVDS avoid breaking the
-			 * aux dp channel on imac and help (but not completely fix)
-			 * https://bugzilla.redhat.com/show_bug.cgi?id=726143
-			 * also avoid interrupt storms during dpms.
-			 */
-			tmp = RREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[amdgpu_connector->hpd.hpd]);
-			tmp = REG_SET_FIELD(tmp, DC_HPD_INT_CONTROL, DC_HPD_INT_EN, 0);
-			WREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[amdgpu_connector->hpd.hpd], tmp);
-			continue;
-		}
-
-		tmp = RREG32(mmDC_HPD_CONTROL + hpd_offsets[amdgpu_connector->hpd.hpd]);
-		tmp = REG_SET_FIELD(tmp, DC_HPD_CONTROL, DC_HPD_EN, 1);
-		WREG32(mmDC_HPD_CONTROL + hpd_offsets[amdgpu_connector->hpd.hpd], tmp);
-
-		tmp = RREG32(mmDC_HPD_TOGGLE_FILT_CNTL + hpd_offsets[amdgpu_connector->hpd.hpd]);
-		tmp = REG_SET_FIELD(tmp, DC_HPD_TOGGLE_FILT_CNTL,
-				    DC_HPD_CONNECT_INT_DELAY,
-				    AMDGPU_HPD_CONNECT_INT_DELAY_IN_MS);
-		tmp = REG_SET_FIELD(tmp, DC_HPD_TOGGLE_FILT_CNTL,
-				    DC_HPD_DISCONNECT_INT_DELAY,
-				    AMDGPU_HPD_DISCONNECT_INT_DELAY_IN_MS);
-		WREG32(mmDC_HPD_TOGGLE_FILT_CNTL + hpd_offsets[amdgpu_connector->hpd.hpd], tmp);
-
-		dce_v11_0_hpd_int_ack(adev, amdgpu_connector->hpd.hpd);
-		dce_v11_0_hpd_set_polarity(adev, amdgpu_connector->hpd.hpd);
-		amdgpu_irq_get(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
-	}
-	drm_connector_list_iter_end(&iter);
-}
-
-/**
- * dce_v11_0_hpd_fini - hpd tear down callback.
- *
- * @adev: amdgpu_device pointer
- *
- * Tear down the hpd pins used by the card (evergreen+).
- * Disable the hpd interrupts.
- */
-static void dce_v11_0_hpd_fini(struct amdgpu_device *adev)
-{
-	struct drm_device *dev = adev_to_drm(adev);
-	struct drm_connector *connector;
-	struct drm_connector_list_iter iter;
-	u32 tmp;
-
-	drm_connector_list_iter_begin(dev, &iter);
-	drm_for_each_connector_iter(connector, &iter) {
-		struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
-
-		if (amdgpu_connector->hpd.hpd >= adev->mode_info.num_hpd)
-			continue;
-
-		tmp = RREG32(mmDC_HPD_CONTROL + hpd_offsets[amdgpu_connector->hpd.hpd]);
-		tmp = REG_SET_FIELD(tmp, DC_HPD_CONTROL, DC_HPD_EN, 0);
-		WREG32(mmDC_HPD_CONTROL + hpd_offsets[amdgpu_connector->hpd.hpd], tmp);
-
-		amdgpu_irq_put(adev, &adev->hpd_irq, amdgpu_connector->hpd.hpd);
-	}
-	drm_connector_list_iter_end(&iter);
-}
-
-static u32 dce_v11_0_hpd_get_gpio_reg(struct amdgpu_device *adev)
-{
-	return mmDC_GPIO_HPD_A;
-}
-
-static bool dce_v11_0_is_display_hung(struct amdgpu_device *adev)
-{
-	u32 crtc_hung = 0;
-	u32 crtc_status[6];
-	u32 i, j, tmp;
-
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		tmp = RREG32(mmCRTC_CONTROL + crtc_offsets[i]);
-		if (REG_GET_FIELD(tmp, CRTC_CONTROL, CRTC_MASTER_EN)) {
-			crtc_status[i] = RREG32(mmCRTC_STATUS_HV_COUNT + crtc_offsets[i]);
-			crtc_hung |= (1 << i);
-		}
-	}
-
-	for (j = 0; j < 10; j++) {
-		for (i = 0; i < adev->mode_info.num_crtc; i++) {
-			if (crtc_hung & (1 << i)) {
-				tmp = RREG32(mmCRTC_STATUS_HV_COUNT + crtc_offsets[i]);
-				if (tmp != crtc_status[i])
-					crtc_hung &= ~(1 << i);
-			}
-		}
-		if (crtc_hung == 0)
-			return false;
-		udelay(100);
-	}
-
-	return true;
-}
-
-static void dce_v11_0_set_vga_render_state(struct amdgpu_device *adev,
-					   bool render)
-{
-	u32 tmp;
-
-	/* Lockout access through VGA aperture*/
-	tmp = RREG32(mmVGA_HDP_CONTROL);
-	if (render)
-		tmp = REG_SET_FIELD(tmp, VGA_HDP_CONTROL, VGA_MEMORY_DISABLE, 0);
-	else
-		tmp = REG_SET_FIELD(tmp, VGA_HDP_CONTROL, VGA_MEMORY_DISABLE, 1);
-	WREG32(mmVGA_HDP_CONTROL, tmp);
-
-	/* disable VGA render */
-	tmp = RREG32(mmVGA_RENDER_CONTROL);
-	if (render)
-		tmp = REG_SET_FIELD(tmp, VGA_RENDER_CONTROL, VGA_VSTATUS_CNTL, 1);
-	else
-		tmp = REG_SET_FIELD(tmp, VGA_RENDER_CONTROL, VGA_VSTATUS_CNTL, 0);
-	WREG32(mmVGA_RENDER_CONTROL, tmp);
-}
-
-static int dce_v11_0_get_num_crtc (struct amdgpu_device *adev)
-{
-	int num_crtc = 0;
-
-	switch (adev->asic_type) {
-	case CHIP_CARRIZO:
-		num_crtc = 3;
-		break;
-	case CHIP_STONEY:
-		num_crtc = 2;
-		break;
-	case CHIP_POLARIS10:
-	case CHIP_VEGAM:
-		num_crtc = 6;
-		break;
-	case CHIP_POLARIS11:
-	case CHIP_POLARIS12:
-		num_crtc = 5;
-		break;
-	default:
-		num_crtc = 0;
-	}
-	return num_crtc;
-}
-
-void dce_v11_0_disable_dce(struct amdgpu_device *adev)
-{
-	/*Disable VGA render and enabled crtc, if has DCE engine*/
-	if (amdgpu_atombios_has_dce_engine_info(adev)) {
-		u32 tmp;
-		int crtc_enabled, i;
-
-		dce_v11_0_set_vga_render_state(adev, false);
-
-		/*Disable crtc*/
-		for (i = 0; i < dce_v11_0_get_num_crtc(adev); i++) {
-			crtc_enabled = REG_GET_FIELD(RREG32(mmCRTC_CONTROL + crtc_offsets[i]),
-									 CRTC_CONTROL, CRTC_MASTER_EN);
-			if (crtc_enabled) {
-				WREG32(mmCRTC_UPDATE_LOCK + crtc_offsets[i], 1);
-				tmp = RREG32(mmCRTC_CONTROL + crtc_offsets[i]);
-				tmp = REG_SET_FIELD(tmp, CRTC_CONTROL, CRTC_MASTER_EN, 0);
-				WREG32(mmCRTC_CONTROL + crtc_offsets[i], tmp);
-				WREG32(mmCRTC_UPDATE_LOCK + crtc_offsets[i], 0);
-			}
-		}
-	}
-}
-
-static void dce_v11_0_program_fmt(struct drm_encoder *encoder)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
-	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
-	int bpc = 0;
-	u32 tmp = 0;
-	enum amdgpu_connector_dither dither = AMDGPU_FMT_DITHER_DISABLE;
-
-	if (connector) {
-		struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
-		bpc = amdgpu_connector_get_monitor_bpc(connector);
-		dither = amdgpu_connector->dither;
-	}
-
-	/* LVDS/eDP FMT is set up by atom */
-	if (amdgpu_encoder->devices & ATOM_DEVICE_LCD_SUPPORT)
-		return;
-
-	/* not needed for analog */
-	if ((amdgpu_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1) ||
-	    (amdgpu_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2))
-		return;
-
-	if (bpc == 0)
-		return;
-
-	switch (bpc) {
-	case 6:
-		if (dither == AMDGPU_FMT_DITHER_ENABLE) {
-			/* XXX sort out optimal dither settings */
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_FRAME_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_HIGHPASS_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_EN, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_DEPTH, 0);
-		} else {
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_EN, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_DEPTH, 0);
-		}
-		break;
-	case 8:
-		if (dither == AMDGPU_FMT_DITHER_ENABLE) {
-			/* XXX sort out optimal dither settings */
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_FRAME_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_HIGHPASS_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_RGB_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_EN, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_DEPTH, 1);
-		} else {
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_EN, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_DEPTH, 1);
-		}
-		break;
-	case 10:
-		if (dither == AMDGPU_FMT_DITHER_ENABLE) {
-			/* XXX sort out optimal dither settings */
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_FRAME_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_HIGHPASS_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_RGB_RANDOM_ENABLE, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_EN, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_SPATIAL_DITHER_DEPTH, 2);
-		} else {
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_EN, 1);
-			tmp = REG_SET_FIELD(tmp, FMT_BIT_DEPTH_CONTROL, FMT_TRUNCATE_DEPTH, 2);
-		}
-		break;
-	default:
-		/* not needed */
-		break;
-	}
-
-	WREG32(mmFMT_BIT_DEPTH_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-}
-
-
-/* display watermark setup */
-/**
- * dce_v11_0_line_buffer_adjust - Set up the line buffer
- *
- * @adev: amdgpu_device pointer
- * @amdgpu_crtc: the selected display controller
- * @mode: the current display mode on the selected display
- * controller
- *
- * Setup up the line buffer allocation for
- * the selected display controller (CIK).
- * Returns the line buffer size in pixels.
- */
-static u32 dce_v11_0_line_buffer_adjust(struct amdgpu_device *adev,
-				       struct amdgpu_crtc *amdgpu_crtc,
-				       struct drm_display_mode *mode)
-{
-	u32 tmp, buffer_alloc, i, mem_cfg;
-	u32 pipe_offset = amdgpu_crtc->crtc_id;
-	/*
-	 * Line Buffer Setup
-	 * There are 6 line buffers, one for each display controllers.
-	 * There are 3 partitions per LB. Select the number of partitions
-	 * to enable based on the display width.  For display widths larger
-	 * than 4096, you need use to use 2 display controllers and combine
-	 * them using the stereo blender.
-	 */
-	if (amdgpu_crtc->base.enabled && mode) {
-		if (mode->crtc_hdisplay < 1920) {
-			mem_cfg = 1;
-			buffer_alloc = 2;
-		} else if (mode->crtc_hdisplay < 2560) {
-			mem_cfg = 2;
-			buffer_alloc = 2;
-		} else if (mode->crtc_hdisplay < 4096) {
-			mem_cfg = 0;
-			buffer_alloc = (adev->flags & AMD_IS_APU) ? 2 : 4;
-		} else {
-			DRM_DEBUG_KMS("Mode too big for LB!\n");
-			mem_cfg = 0;
-			buffer_alloc = (adev->flags & AMD_IS_APU) ? 2 : 4;
-		}
-	} else {
-		mem_cfg = 1;
-		buffer_alloc = 0;
-	}
-
-	tmp = RREG32(mmLB_MEMORY_CTRL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, LB_MEMORY_CTRL, LB_MEMORY_CONFIG, mem_cfg);
-	WREG32(mmLB_MEMORY_CTRL + amdgpu_crtc->crtc_offset, tmp);
-
-	tmp = RREG32(mmPIPE0_DMIF_BUFFER_CONTROL + pipe_offset);
-	tmp = REG_SET_FIELD(tmp, PIPE0_DMIF_BUFFER_CONTROL, DMIF_BUFFERS_ALLOCATED, buffer_alloc);
-	WREG32(mmPIPE0_DMIF_BUFFER_CONTROL + pipe_offset, tmp);
-
-	for (i = 0; i < adev->usec_timeout; i++) {
-		tmp = RREG32(mmPIPE0_DMIF_BUFFER_CONTROL + pipe_offset);
-		if (REG_GET_FIELD(tmp, PIPE0_DMIF_BUFFER_CONTROL, DMIF_BUFFERS_ALLOCATION_COMPLETED))
-			break;
-		udelay(1);
-	}
-
-	if (amdgpu_crtc->base.enabled && mode) {
-		switch (mem_cfg) {
-		case 0:
-		default:
-			return 4096 * 2;
-		case 1:
-			return 1920 * 2;
-		case 2:
-			return 2560 * 2;
-		}
-	}
-
-	/* controller not enabled, so no lb used */
-	return 0;
-}
-
-/**
- * cik_get_number_of_dram_channels - get the number of dram channels
- *
- * @adev: amdgpu_device pointer
- *
- * Look up the number of video ram channels (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the number of dram channels
- */
-static u32 cik_get_number_of_dram_channels(struct amdgpu_device *adev)
-{
-	u32 tmp = RREG32(mmMC_SHARED_CHMAP);
-
-	switch (REG_GET_FIELD(tmp, MC_SHARED_CHMAP, NOOFCHAN)) {
-	case 0:
-	default:
-		return 1;
-	case 1:
-		return 2;
-	case 2:
-		return 4;
-	case 3:
-		return 8;
-	case 4:
-		return 3;
-	case 5:
-		return 6;
-	case 6:
-		return 10;
-	case 7:
-		return 12;
-	case 8:
-		return 16;
-	}
-}
-
-struct dce10_wm_params {
-	u32 dram_channels; /* number of dram channels */
-	u32 yclk;          /* bandwidth per dram data pin in kHz */
-	u32 sclk;          /* engine clock in kHz */
-	u32 disp_clk;      /* display clock in kHz */
-	u32 src_width;     /* viewport width */
-	u32 active_time;   /* active display time in ns */
-	u32 blank_time;    /* blank time in ns */
-	bool interlaced;    /* mode is interlaced */
-	fixed20_12 vsc;    /* vertical scale ratio */
-	u32 num_heads;     /* number of active crtcs */
-	u32 bytes_per_pixel; /* bytes per pixel display + overlay */
-	u32 lb_size;       /* line buffer allocated to pipe */
-	u32 vtaps;         /* vertical scaler taps */
-};
-
-/**
- * dce_v11_0_dram_bandwidth - get the dram bandwidth
- *
- * @wm: watermark calculation data
- *
- * Calculate the raw dram bandwidth (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the dram bandwidth in MBytes/s
- */
-static u32 dce_v11_0_dram_bandwidth(struct dce10_wm_params *wm)
-{
-	/* Calculate raw DRAM Bandwidth */
-	fixed20_12 dram_efficiency; /* 0.7 */
-	fixed20_12 yclk, dram_channels, bandwidth;
-	fixed20_12 a;
-
-	a.full = dfixed_const(1000);
-	yclk.full = dfixed_const(wm->yclk);
-	yclk.full = dfixed_div(yclk, a);
-	dram_channels.full = dfixed_const(wm->dram_channels * 4);
-	a.full = dfixed_const(10);
-	dram_efficiency.full = dfixed_const(7);
-	dram_efficiency.full = dfixed_div(dram_efficiency, a);
-	bandwidth.full = dfixed_mul(dram_channels, yclk);
-	bandwidth.full = dfixed_mul(bandwidth, dram_efficiency);
-
-	return dfixed_trunc(bandwidth);
-}
-
-/**
- * dce_v11_0_dram_bandwidth_for_display - get the dram bandwidth for display
- *
- * @wm: watermark calculation data
- *
- * Calculate the dram bandwidth used for display (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the dram bandwidth for display in MBytes/s
- */
-static u32 dce_v11_0_dram_bandwidth_for_display(struct dce10_wm_params *wm)
-{
-	/* Calculate DRAM Bandwidth and the part allocated to display. */
-	fixed20_12 disp_dram_allocation; /* 0.3 to 0.7 */
-	fixed20_12 yclk, dram_channels, bandwidth;
-	fixed20_12 a;
-
-	a.full = dfixed_const(1000);
-	yclk.full = dfixed_const(wm->yclk);
-	yclk.full = dfixed_div(yclk, a);
-	dram_channels.full = dfixed_const(wm->dram_channels * 4);
-	a.full = dfixed_const(10);
-	disp_dram_allocation.full = dfixed_const(3); /* XXX worse case value 0.3 */
-	disp_dram_allocation.full = dfixed_div(disp_dram_allocation, a);
-	bandwidth.full = dfixed_mul(dram_channels, yclk);
-	bandwidth.full = dfixed_mul(bandwidth, disp_dram_allocation);
-
-	return dfixed_trunc(bandwidth);
-}
-
-/**
- * dce_v11_0_data_return_bandwidth - get the data return bandwidth
- *
- * @wm: watermark calculation data
- *
- * Calculate the data return bandwidth used for display (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the data return bandwidth in MBytes/s
- */
-static u32 dce_v11_0_data_return_bandwidth(struct dce10_wm_params *wm)
-{
-	/* Calculate the display Data return Bandwidth */
-	fixed20_12 return_efficiency; /* 0.8 */
-	fixed20_12 sclk, bandwidth;
-	fixed20_12 a;
-
-	a.full = dfixed_const(1000);
-	sclk.full = dfixed_const(wm->sclk);
-	sclk.full = dfixed_div(sclk, a);
-	a.full = dfixed_const(10);
-	return_efficiency.full = dfixed_const(8);
-	return_efficiency.full = dfixed_div(return_efficiency, a);
-	a.full = dfixed_const(32);
-	bandwidth.full = dfixed_mul(a, sclk);
-	bandwidth.full = dfixed_mul(bandwidth, return_efficiency);
-
-	return dfixed_trunc(bandwidth);
-}
-
-/**
- * dce_v11_0_dmif_request_bandwidth - get the dmif bandwidth
- *
- * @wm: watermark calculation data
- *
- * Calculate the dmif bandwidth used for display (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the dmif bandwidth in MBytes/s
- */
-static u32 dce_v11_0_dmif_request_bandwidth(struct dce10_wm_params *wm)
-{
-	/* Calculate the DMIF Request Bandwidth */
-	fixed20_12 disp_clk_request_efficiency; /* 0.8 */
-	fixed20_12 disp_clk, bandwidth;
-	fixed20_12 a, b;
-
-	a.full = dfixed_const(1000);
-	disp_clk.full = dfixed_const(wm->disp_clk);
-	disp_clk.full = dfixed_div(disp_clk, a);
-	a.full = dfixed_const(32);
-	b.full = dfixed_mul(a, disp_clk);
-
-	a.full = dfixed_const(10);
-	disp_clk_request_efficiency.full = dfixed_const(8);
-	disp_clk_request_efficiency.full = dfixed_div(disp_clk_request_efficiency, a);
-
-	bandwidth.full = dfixed_mul(b, disp_clk_request_efficiency);
-
-	return dfixed_trunc(bandwidth);
-}
-
-/**
- * dce_v11_0_available_bandwidth - get the min available bandwidth
- *
- * @wm: watermark calculation data
- *
- * Calculate the min available bandwidth used for display (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the min available bandwidth in MBytes/s
- */
-static u32 dce_v11_0_available_bandwidth(struct dce10_wm_params *wm)
-{
-	/* Calculate the Available bandwidth. Display can use this temporarily but not in average. */
-	u32 dram_bandwidth = dce_v11_0_dram_bandwidth(wm);
-	u32 data_return_bandwidth = dce_v11_0_data_return_bandwidth(wm);
-	u32 dmif_req_bandwidth = dce_v11_0_dmif_request_bandwidth(wm);
-
-	return min(dram_bandwidth, min(data_return_bandwidth, dmif_req_bandwidth));
-}
-
-/**
- * dce_v11_0_average_bandwidth - get the average available bandwidth
- *
- * @wm: watermark calculation data
- *
- * Calculate the average available bandwidth used for display (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the average available bandwidth in MBytes/s
- */
-static u32 dce_v11_0_average_bandwidth(struct dce10_wm_params *wm)
-{
-	/* Calculate the display mode Average Bandwidth
-	 * DisplayMode should contain the source and destination dimensions,
-	 * timing, etc.
-	 */
-	fixed20_12 bpp;
-	fixed20_12 line_time;
-	fixed20_12 src_width;
-	fixed20_12 bandwidth;
-	fixed20_12 a;
-
-	a.full = dfixed_const(1000);
-	line_time.full = dfixed_const(wm->active_time + wm->blank_time);
-	line_time.full = dfixed_div(line_time, a);
-	bpp.full = dfixed_const(wm->bytes_per_pixel);
-	src_width.full = dfixed_const(wm->src_width);
-	bandwidth.full = dfixed_mul(src_width, bpp);
-	bandwidth.full = dfixed_mul(bandwidth, wm->vsc);
-	bandwidth.full = dfixed_div(bandwidth, line_time);
-
-	return dfixed_trunc(bandwidth);
-}
-
-/**
- * dce_v11_0_latency_watermark - get the latency watermark
- *
- * @wm: watermark calculation data
- *
- * Calculate the latency watermark (CIK).
- * Used for display watermark bandwidth calculations
- * Returns the latency watermark in ns
- */
-static u32 dce_v11_0_latency_watermark(struct dce10_wm_params *wm)
-{
-	/* First calculate the latency in ns */
-	u32 mc_latency = 2000; /* 2000 ns. */
-	u32 available_bandwidth = dce_v11_0_available_bandwidth(wm);
-	u32 worst_chunk_return_time = (512 * 8 * 1000) / available_bandwidth;
-	u32 cursor_line_pair_return_time = (128 * 4 * 1000) / available_bandwidth;
-	u32 dc_latency = 40000000 / wm->disp_clk; /* dc pipe latency */
-	u32 other_heads_data_return_time = ((wm->num_heads + 1) * worst_chunk_return_time) +
-		(wm->num_heads * cursor_line_pair_return_time);
-	u32 latency = mc_latency + other_heads_data_return_time + dc_latency;
-	u32 max_src_lines_per_dst_line, lb_fill_bw, line_fill_time;
-	u32 tmp, dmif_size = 12288;
-	fixed20_12 a, b, c;
-
-	if (wm->num_heads == 0)
-		return 0;
-
-	a.full = dfixed_const(2);
-	b.full = dfixed_const(1);
-	if ((wm->vsc.full > a.full) ||
-	    ((wm->vsc.full > b.full) && (wm->vtaps >= 3)) ||
-	    (wm->vtaps >= 5) ||
-	    ((wm->vsc.full >= a.full) && wm->interlaced))
-		max_src_lines_per_dst_line = 4;
-	else
-		max_src_lines_per_dst_line = 2;
-
-	a.full = dfixed_const(available_bandwidth);
-	b.full = dfixed_const(wm->num_heads);
-	a.full = dfixed_div(a, b);
-	tmp = div_u64((u64) dmif_size * (u64) wm->disp_clk, mc_latency + 512);
-	tmp = min(dfixed_trunc(a), tmp);
-
-	lb_fill_bw = min(tmp, wm->disp_clk * wm->bytes_per_pixel / 1000);
-
-	a.full = dfixed_const(max_src_lines_per_dst_line * wm->src_width * wm->bytes_per_pixel);
-	b.full = dfixed_const(1000);
-	c.full = dfixed_const(lb_fill_bw);
-	b.full = dfixed_div(c, b);
-	a.full = dfixed_div(a, b);
-	line_fill_time = dfixed_trunc(a);
-
-	if (line_fill_time < wm->active_time)
-		return latency;
-	else
-		return latency + (line_fill_time - wm->active_time);
-
-}
-
-/**
- * dce_v11_0_average_bandwidth_vs_dram_bandwidth_for_display - check
- * average and available dram bandwidth
- *
- * @wm: watermark calculation data
- *
- * Check if the display average bandwidth fits in the display
- * dram bandwidth (CIK).
- * Used for display watermark bandwidth calculations
- * Returns true if the display fits, false if not.
- */
-static bool dce_v11_0_average_bandwidth_vs_dram_bandwidth_for_display(struct dce10_wm_params *wm)
-{
-	if (dce_v11_0_average_bandwidth(wm) <=
-	    (dce_v11_0_dram_bandwidth_for_display(wm) / wm->num_heads))
-		return true;
-	else
-		return false;
-}
-
-/**
- * dce_v11_0_average_bandwidth_vs_available_bandwidth - check
- * average and available bandwidth
- *
- * @wm: watermark calculation data
- *
- * Check if the display average bandwidth fits in the display
- * available bandwidth (CIK).
- * Used for display watermark bandwidth calculations
- * Returns true if the display fits, false if not.
- */
-static bool dce_v11_0_average_bandwidth_vs_available_bandwidth(struct dce10_wm_params *wm)
-{
-	if (dce_v11_0_average_bandwidth(wm) <=
-	    (dce_v11_0_available_bandwidth(wm) / wm->num_heads))
-		return true;
-	else
-		return false;
-}
-
-/**
- * dce_v11_0_check_latency_hiding - check latency hiding
- *
- * @wm: watermark calculation data
- *
- * Check latency hiding (CIK).
- * Used for display watermark bandwidth calculations
- * Returns true if the display fits, false if not.
- */
-static bool dce_v11_0_check_latency_hiding(struct dce10_wm_params *wm)
-{
-	u32 lb_partitions = wm->lb_size / wm->src_width;
-	u32 line_time = wm->active_time + wm->blank_time;
-	u32 latency_tolerant_lines;
-	u32 latency_hiding;
-	fixed20_12 a;
-
-	a.full = dfixed_const(1);
-	if (wm->vsc.full > a.full)
-		latency_tolerant_lines = 1;
-	else {
-		if (lb_partitions <= (wm->vtaps + 1))
-			latency_tolerant_lines = 1;
-		else
-			latency_tolerant_lines = 2;
-	}
-
-	latency_hiding = (latency_tolerant_lines * line_time + wm->blank_time);
-
-	if (dce_v11_0_latency_watermark(wm) <= latency_hiding)
-		return true;
-	else
-		return false;
-}
-
-/**
- * dce_v11_0_program_watermarks - program display watermarks
- *
- * @adev: amdgpu_device pointer
- * @amdgpu_crtc: the selected display controller
- * @lb_size: line buffer size
- * @num_heads: number of display controllers in use
- *
- * Calculate and program the display watermarks for the
- * selected display controller (CIK).
- */
-static void dce_v11_0_program_watermarks(struct amdgpu_device *adev,
-					struct amdgpu_crtc *amdgpu_crtc,
-					u32 lb_size, u32 num_heads)
-{
-	struct drm_display_mode *mode = &amdgpu_crtc->base.mode;
-	struct dce10_wm_params wm_low, wm_high;
-	u32 active_time;
-	u32 line_time = 0;
-	u32 latency_watermark_a = 0, latency_watermark_b = 0;
-	u32 tmp, wm_mask, lb_vblank_lead_lines = 0;
-
-	if (amdgpu_crtc->base.enabled && num_heads && mode) {
-		active_time = (u32) div_u64((u64)mode->crtc_hdisplay * 1000000,
-					    (u32)mode->clock);
-		line_time = (u32) div_u64((u64)mode->crtc_htotal * 1000000,
-					  (u32)mode->clock);
-		line_time = min_t(u32, line_time, 65535);
-
-		/* watermark for high clocks */
-		if (adev->pm.dpm_enabled) {
-			wm_high.yclk =
-				amdgpu_dpm_get_mclk(adev, false) * 10;
-			wm_high.sclk =
-				amdgpu_dpm_get_sclk(adev, false) * 10;
-		} else {
-			wm_high.yclk = adev->pm.current_mclk * 10;
-			wm_high.sclk = adev->pm.current_sclk * 10;
-		}
-
-		wm_high.disp_clk = mode->clock;
-		wm_high.src_width = mode->crtc_hdisplay;
-		wm_high.active_time = active_time;
-		wm_high.blank_time = line_time - wm_high.active_time;
-		wm_high.interlaced = false;
-		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
-			wm_high.interlaced = true;
-		wm_high.vsc = amdgpu_crtc->vsc;
-		wm_high.vtaps = 1;
-		if (amdgpu_crtc->rmx_type != RMX_OFF)
-			wm_high.vtaps = 2;
-		wm_high.bytes_per_pixel = 4; /* XXX: get this from fb config */
-		wm_high.lb_size = lb_size;
-		wm_high.dram_channels = cik_get_number_of_dram_channels(adev);
-		wm_high.num_heads = num_heads;
-
-		/* set for high clocks */
-		latency_watermark_a = min_t(u32, dce_v11_0_latency_watermark(&wm_high), 65535);
-
-		/* possibly force display priority to high */
-		/* should really do this at mode validation time... */
-		if (!dce_v11_0_average_bandwidth_vs_dram_bandwidth_for_display(&wm_high) ||
-		    !dce_v11_0_average_bandwidth_vs_available_bandwidth(&wm_high) ||
-		    !dce_v11_0_check_latency_hiding(&wm_high) ||
-		    (adev->mode_info.disp_priority == 2)) {
-			DRM_DEBUG_KMS("force priority to high\n");
-		}
-
-		/* watermark for low clocks */
-		if (adev->pm.dpm_enabled) {
-			wm_low.yclk =
-				amdgpu_dpm_get_mclk(adev, true) * 10;
-			wm_low.sclk =
-				amdgpu_dpm_get_sclk(adev, true) * 10;
-		} else {
-			wm_low.yclk = adev->pm.current_mclk * 10;
-			wm_low.sclk = adev->pm.current_sclk * 10;
-		}
-
-		wm_low.disp_clk = mode->clock;
-		wm_low.src_width = mode->crtc_hdisplay;
-		wm_low.active_time = active_time;
-		wm_low.blank_time = line_time - wm_low.active_time;
-		wm_low.interlaced = false;
-		if (mode->flags & DRM_MODE_FLAG_INTERLACE)
-			wm_low.interlaced = true;
-		wm_low.vsc = amdgpu_crtc->vsc;
-		wm_low.vtaps = 1;
-		if (amdgpu_crtc->rmx_type != RMX_OFF)
-			wm_low.vtaps = 2;
-		wm_low.bytes_per_pixel = 4; /* XXX: get this from fb config */
-		wm_low.lb_size = lb_size;
-		wm_low.dram_channels = cik_get_number_of_dram_channels(adev);
-		wm_low.num_heads = num_heads;
-
-		/* set for low clocks */
-		latency_watermark_b = min_t(u32, dce_v11_0_latency_watermark(&wm_low), 65535);
-
-		/* possibly force display priority to high */
-		/* should really do this at mode validation time... */
-		if (!dce_v11_0_average_bandwidth_vs_dram_bandwidth_for_display(&wm_low) ||
-		    !dce_v11_0_average_bandwidth_vs_available_bandwidth(&wm_low) ||
-		    !dce_v11_0_check_latency_hiding(&wm_low) ||
-		    (adev->mode_info.disp_priority == 2)) {
-			DRM_DEBUG_KMS("force priority to high\n");
-		}
-		lb_vblank_lead_lines = DIV_ROUND_UP(lb_size, mode->crtc_hdisplay);
-	}
-
-	/* select wm A */
-	wm_mask = RREG32(mmDPG_WATERMARK_MASK_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(wm_mask, DPG_WATERMARK_MASK_CONTROL, URGENCY_WATERMARK_MASK, 1);
-	WREG32(mmDPG_WATERMARK_MASK_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-	tmp = RREG32(mmDPG_PIPE_URGENCY_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, DPG_PIPE_URGENCY_CONTROL, URGENCY_LOW_WATERMARK, latency_watermark_a);
-	tmp = REG_SET_FIELD(tmp, DPG_PIPE_URGENCY_CONTROL, URGENCY_HIGH_WATERMARK, line_time);
-	WREG32(mmDPG_PIPE_URGENCY_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-	/* select wm B */
-	tmp = REG_SET_FIELD(wm_mask, DPG_WATERMARK_MASK_CONTROL, URGENCY_WATERMARK_MASK, 2);
-	WREG32(mmDPG_WATERMARK_MASK_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-	tmp = RREG32(mmDPG_PIPE_URGENCY_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, DPG_PIPE_URGENCY_CONTROL, URGENCY_LOW_WATERMARK, latency_watermark_b);
-	tmp = REG_SET_FIELD(tmp, DPG_PIPE_URGENCY_CONTROL, URGENCY_HIGH_WATERMARK, line_time);
-	WREG32(mmDPG_PIPE_URGENCY_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-	/* restore original selection */
-	WREG32(mmDPG_WATERMARK_MASK_CONTROL + amdgpu_crtc->crtc_offset, wm_mask);
-
-	/* save values for DPM */
-	amdgpu_crtc->line_time = line_time;
-
-	/* Save number of lines the linebuffer leads before the scanout */
-	amdgpu_crtc->lb_vblank_lead_lines = lb_vblank_lead_lines;
-}
-
-/**
- * dce_v11_0_bandwidth_update - program display watermarks
- *
- * @adev: amdgpu_device pointer
- *
- * Calculate and program the display watermarks and line
- * buffer allocation (CIK).
- */
-static void dce_v11_0_bandwidth_update(struct amdgpu_device *adev)
-{
-	struct drm_display_mode *mode = NULL;
-	u32 num_heads = 0, lb_size;
-	int i;
-
-	amdgpu_display_update_priority(adev);
-
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		if (adev->mode_info.crtcs[i]->base.enabled)
-			num_heads++;
-	}
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		mode = &adev->mode_info.crtcs[i]->base.mode;
-		lb_size = dce_v11_0_line_buffer_adjust(adev, adev->mode_info.crtcs[i], mode);
-		dce_v11_0_program_watermarks(adev, adev->mode_info.crtcs[i],
-					    lb_size, num_heads);
-	}
-}
-
-static void dce_v11_0_audio_get_connected_pins(struct amdgpu_device *adev)
-{
-	int i;
-	u32 offset, tmp;
-
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++) {
-		offset = adev->mode_info.audio.pin[i].offset;
-		tmp = RREG32_AUDIO_ENDPT(offset,
-					 ixAZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_CONFIGURATION_DEFAULT);
-		if (((tmp &
-		AZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_CONFIGURATION_DEFAULT__PORT_CONNECTIVITY_MASK) >>
-		AZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_CONFIGURATION_DEFAULT__PORT_CONNECTIVITY__SHIFT) == 1)
-			adev->mode_info.audio.pin[i].connected = false;
-		else
-			adev->mode_info.audio.pin[i].connected = true;
-	}
-}
-
-static struct amdgpu_audio_pin *dce_v11_0_audio_get_pin(struct amdgpu_device *adev)
-{
-	int i;
-
-	dce_v11_0_audio_get_connected_pins(adev);
-
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++) {
-		if (adev->mode_info.audio.pin[i].connected)
-			return &adev->mode_info.audio.pin[i];
-	}
-	DRM_ERROR("No connected audio pins found!\n");
-	return NULL;
-}
-
-static void dce_v11_0_afmt_audio_select_pin(struct drm_encoder *encoder)
-{
-	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	u32 tmp;
-
-	if (!dig || !dig->afmt || !dig->afmt->pin)
-		return;
-
-	tmp = RREG32(mmAFMT_AUDIO_SRC_CONTROL + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, AFMT_AUDIO_SRC_CONTROL, AFMT_AUDIO_SRC_SELECT, dig->afmt->pin->id);
-	WREG32(mmAFMT_AUDIO_SRC_CONTROL + dig->afmt->offset, tmp);
-}
-
-static void dce_v11_0_audio_write_latency_fields(struct drm_encoder *encoder,
-						struct drm_display_mode *mode)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	struct drm_connector *connector;
-	struct drm_connector_list_iter iter;
-	struct amdgpu_connector *amdgpu_connector = NULL;
-	u32 tmp;
-	int interlace = 0;
-
-	if (!dig || !dig->afmt || !dig->afmt->pin)
-		return;
-
-	drm_connector_list_iter_begin(dev, &iter);
-	drm_for_each_connector_iter(connector, &iter) {
-		if (connector->encoder == encoder) {
-			amdgpu_connector = to_amdgpu_connector(connector);
-			break;
-		}
-	}
-	drm_connector_list_iter_end(&iter);
-
-	if (!amdgpu_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
-		interlace = 1;
-	if (connector->latency_present[interlace]) {
-		tmp = REG_SET_FIELD(0, AZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC,
-				    VIDEO_LIPSYNC, connector->video_latency[interlace]);
-		tmp = REG_SET_FIELD(0, AZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC,
-				    AUDIO_LIPSYNC, connector->audio_latency[interlace]);
-	} else {
-		tmp = REG_SET_FIELD(0, AZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC,
-				    VIDEO_LIPSYNC, 0);
-		tmp = REG_SET_FIELD(0, AZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC,
-				    AUDIO_LIPSYNC, 0);
-	}
-	WREG32_AUDIO_ENDPT(dig->afmt->pin->offset,
-			   ixAZALIA_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp);
-}
-
-static void dce_v11_0_audio_write_speaker_allocation(struct drm_encoder *encoder)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	struct drm_connector *connector;
-	struct drm_connector_list_iter iter;
-	struct amdgpu_connector *amdgpu_connector = NULL;
-	u32 tmp;
-	u8 *sadb = NULL;
-	int sad_count;
-
-	if (!dig || !dig->afmt || !dig->afmt->pin)
-		return;
-
-	drm_connector_list_iter_begin(dev, &iter);
-	drm_for_each_connector_iter(connector, &iter) {
-		if (connector->encoder == encoder) {
-			amdgpu_connector = to_amdgpu_connector(connector);
-			break;
-		}
-	}
-	drm_connector_list_iter_end(&iter);
-
-	if (!amdgpu_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	sad_count = drm_edid_to_speaker_allocation(amdgpu_connector->edid, &sadb);
-	if (sad_count < 0) {
-		DRM_ERROR("Couldn't read Speaker Allocation Data Block: %d\n", sad_count);
-		sad_count = 0;
-	}
-
-	/* program the speaker allocation */
-	tmp = RREG32_AUDIO_ENDPT(dig->afmt->pin->offset,
-				 ixAZALIA_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
-	tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER,
-			    DP_CONNECTION, 0);
-	/* set HDMI mode */
-	tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER,
-			    HDMI_CONNECTION, 1);
-	if (sad_count)
-		tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER,
-				    SPEAKER_ALLOCATION, sadb[0]);
-	else
-		tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER,
-				    SPEAKER_ALLOCATION, 5); /* stereo */
-	WREG32_AUDIO_ENDPT(dig->afmt->pin->offset,
-			   ixAZALIA_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
-
-	kfree(sadb);
-}
-
-static void dce_v11_0_audio_write_sad_regs(struct drm_encoder *encoder)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	struct drm_connector *connector;
-	struct drm_connector_list_iter iter;
-	struct amdgpu_connector *amdgpu_connector = NULL;
-	struct cea_sad *sads;
-	int i, sad_count;
-
-	static const u16 eld_reg_to_type[][2] = {
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0, HDMI_AUDIO_CODING_TYPE_PCM },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR1, HDMI_AUDIO_CODING_TYPE_AC3 },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR2, HDMI_AUDIO_CODING_TYPE_MPEG1 },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR3, HDMI_AUDIO_CODING_TYPE_MP3 },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR4, HDMI_AUDIO_CODING_TYPE_MPEG2 },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR5, HDMI_AUDIO_CODING_TYPE_AAC_LC },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR6, HDMI_AUDIO_CODING_TYPE_DTS },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR7, HDMI_AUDIO_CODING_TYPE_ATRAC },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR9, HDMI_AUDIO_CODING_TYPE_EAC3 },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR10, HDMI_AUDIO_CODING_TYPE_DTS_HD },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR11, HDMI_AUDIO_CODING_TYPE_MLP },
-		{ ixAZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
-	};
-
-	if (!dig || !dig->afmt || !dig->afmt->pin)
-		return;
-
-	drm_connector_list_iter_begin(dev, &iter);
-	drm_for_each_connector_iter(connector, &iter) {
-		if (connector->encoder == encoder) {
-			amdgpu_connector = to_amdgpu_connector(connector);
-			break;
-		}
-	}
-	drm_connector_list_iter_end(&iter);
-
-	if (!amdgpu_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return;
-	}
-
-	sad_count = drm_edid_to_sad(amdgpu_connector->edid, &sads);
-	if (sad_count < 0)
-		DRM_ERROR("Couldn't read SADs: %d\n", sad_count);
-	if (sad_count <= 0)
-		return;
-	BUG_ON(!sads);
-
-	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
-		u32 tmp = 0;
-		u8 stereo_freqs = 0;
-		int max_channels = -1;
-		int j;
-
-		for (j = 0; j < sad_count; j++) {
-			struct cea_sad *sad = &sads[j];
-
-			if (sad->format == eld_reg_to_type[i][1]) {
-				if (sad->channels > max_channels) {
-					tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-							    MAX_CHANNELS, sad->channels);
-					tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-							    DESCRIPTOR_BYTE_2, sad->byte2);
-					tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-							    SUPPORTED_FREQUENCIES, sad->freq);
-					max_channels = sad->channels;
-				}
-
-				if (sad->format == HDMI_AUDIO_CODING_TYPE_PCM)
-					stereo_freqs |= sad->freq;
-				else
-					break;
-			}
-		}
-
-		tmp = REG_SET_FIELD(tmp, AZALIA_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR0,
-				    SUPPORTED_FREQUENCIES_STEREO, stereo_freqs);
-		WREG32_AUDIO_ENDPT(dig->afmt->pin->offset, eld_reg_to_type[i][0], tmp);
-	}
-
-	kfree(sads);
-}
-
-static void dce_v11_0_audio_enable(struct amdgpu_device *adev,
-				  struct amdgpu_audio_pin *pin,
-				  bool enable)
-{
-	if (!pin)
-		return;
-
-	WREG32_AUDIO_ENDPT(pin->offset, ixAZALIA_F0_CODEC_PIN_CONTROL_HOT_PLUG_CONTROL,
-			   enable ? AZALIA_F0_CODEC_PIN_CONTROL_HOT_PLUG_CONTROL__AUDIO_ENABLED_MASK : 0);
-}
-
-static const u32 pin_offsets[] =
-{
-	AUD0_REGISTER_OFFSET,
-	AUD1_REGISTER_OFFSET,
-	AUD2_REGISTER_OFFSET,
-	AUD3_REGISTER_OFFSET,
-	AUD4_REGISTER_OFFSET,
-	AUD5_REGISTER_OFFSET,
-	AUD6_REGISTER_OFFSET,
-	AUD7_REGISTER_OFFSET,
-};
-
-static int dce_v11_0_audio_init(struct amdgpu_device *adev)
-{
-	int i;
-
-	if (!amdgpu_audio)
-		return 0;
-
-	adev->mode_info.audio.enabled = true;
-
-	switch (adev->asic_type) {
-	case CHIP_CARRIZO:
-	case CHIP_STONEY:
-		adev->mode_info.audio.num_pins = 7;
-		break;
-	case CHIP_POLARIS10:
-	case CHIP_VEGAM:
-		adev->mode_info.audio.num_pins = 8;
-		break;
-	case CHIP_POLARIS11:
-	case CHIP_POLARIS12:
-		adev->mode_info.audio.num_pins = 6;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++) {
-		adev->mode_info.audio.pin[i].channels = -1;
-		adev->mode_info.audio.pin[i].rate = -1;
-		adev->mode_info.audio.pin[i].bits_per_sample = -1;
-		adev->mode_info.audio.pin[i].status_bits = 0;
-		adev->mode_info.audio.pin[i].category_code = 0;
-		adev->mode_info.audio.pin[i].connected = false;
-		adev->mode_info.audio.pin[i].offset = pin_offsets[i];
-		adev->mode_info.audio.pin[i].id = i;
-		/* disable audio.  it will be set up later */
-		/* XXX remove once we switch to ip funcs */
-		dce_v11_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-	}
-
-	return 0;
-}
-
-static void dce_v11_0_audio_fini(struct amdgpu_device *adev)
-{
-	if (!amdgpu_audio)
-		return;
-
-	if (!adev->mode_info.audio.enabled)
-		return;
-
-	adev->mode_info.audio.enabled = false;
-}
-
-/*
- * update the N and CTS parameters for a given pixel clock rate
- */
-static void dce_v11_0_afmt_update_ACR(struct drm_encoder *encoder, uint32_t clock)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_afmt_acr acr = amdgpu_afmt_acr(clock);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	u32 tmp;
-
-	tmp = RREG32(mmHDMI_ACR_32_0 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_32_0, HDMI_ACR_CTS_32, acr.cts_32khz);
-	WREG32(mmHDMI_ACR_32_0 + dig->afmt->offset, tmp);
-	tmp = RREG32(mmHDMI_ACR_32_1 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_32_1, HDMI_ACR_N_32, acr.n_32khz);
-	WREG32(mmHDMI_ACR_32_1 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_ACR_44_0 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_44_0, HDMI_ACR_CTS_44, acr.cts_44_1khz);
-	WREG32(mmHDMI_ACR_44_0 + dig->afmt->offset, tmp);
-	tmp = RREG32(mmHDMI_ACR_44_1 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_44_1, HDMI_ACR_N_44, acr.n_44_1khz);
-	WREG32(mmHDMI_ACR_44_1 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_ACR_48_0 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_48_0, HDMI_ACR_CTS_48, acr.cts_48khz);
-	WREG32(mmHDMI_ACR_48_0 + dig->afmt->offset, tmp);
-	tmp = RREG32(mmHDMI_ACR_48_1 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_48_1, HDMI_ACR_N_48, acr.n_48khz);
-	WREG32(mmHDMI_ACR_48_1 + dig->afmt->offset, tmp);
-
-}
-
-/*
- * build a HDMI Video Info Frame
- */
-static void dce_v11_0_afmt_update_avi_infoframe(struct drm_encoder *encoder,
-					       void *buffer, size_t size)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	uint8_t *frame = buffer + 3;
-	uint8_t *header = buffer;
-
-	WREG32(mmAFMT_AVI_INFO0 + dig->afmt->offset,
-		frame[0x0] | (frame[0x1] << 8) | (frame[0x2] << 16) | (frame[0x3] << 24));
-	WREG32(mmAFMT_AVI_INFO1 + dig->afmt->offset,
-		frame[0x4] | (frame[0x5] << 8) | (frame[0x6] << 16) | (frame[0x7] << 24));
-	WREG32(mmAFMT_AVI_INFO2 + dig->afmt->offset,
-		frame[0x8] | (frame[0x9] << 8) | (frame[0xA] << 16) | (frame[0xB] << 24));
-	WREG32(mmAFMT_AVI_INFO3 + dig->afmt->offset,
-		frame[0xC] | (frame[0xD] << 8) | (header[1] << 24));
-}
-
-static void dce_v11_0_audio_set_dto(struct drm_encoder *encoder, u32 clock)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
-	u32 dto_phase = 24 * 1000;
-	u32 dto_modulo = clock;
-	u32 tmp;
-
-	if (!dig || !dig->afmt)
-		return;
-
-	/* XXX two dtos; generally use dto0 for hdmi */
-	/* Express [24MHz / target pixel clock] as an exact rational
-	 * number (coefficient of two integer numbers.  DCCG_AUDIO_DTOx_PHASE
-	 * is the numerator, DCCG_AUDIO_DTOx_MODULE is the denominator
-	 */
-	tmp = RREG32(mmDCCG_AUDIO_DTO_SOURCE);
-	tmp = REG_SET_FIELD(tmp, DCCG_AUDIO_DTO_SOURCE, DCCG_AUDIO_DTO0_SOURCE_SEL,
-			    amdgpu_crtc->crtc_id);
-	WREG32(mmDCCG_AUDIO_DTO_SOURCE, tmp);
-	WREG32(mmDCCG_AUDIO_DTO0_PHASE, dto_phase);
-	WREG32(mmDCCG_AUDIO_DTO0_MODULE, dto_modulo);
-}
-
-/*
- * update the info frames with the data from the current display mode
- */
-static void dce_v11_0_afmt_setmode(struct drm_encoder *encoder,
-				  struct drm_display_mode *mode)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
-	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
-	struct hdmi_avi_infoframe frame;
-	ssize_t err;
-	u32 tmp;
-	int bpc = 8;
-
-	if (!dig || !dig->afmt)
-		return;
-
-	/* Silent, r600_hdmi_enable will raise WARN for us */
-	if (!dig->afmt->enabled)
-		return;
-
-	/* hdmi deep color mode general control packets setup, if bpc > 8 */
-	if (encoder->crtc) {
-		struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(encoder->crtc);
-		bpc = amdgpu_crtc->bpc;
-	}
-
-	/* disable audio prior to setting up hw */
-	dig->afmt->pin = dce_v11_0_audio_get_pin(adev);
-	dce_v11_0_audio_enable(adev, dig->afmt->pin, false);
-
-	dce_v11_0_audio_set_dto(encoder, mode->clock);
-
-	tmp = RREG32(mmHDMI_VBI_PACKET_CONTROL + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_VBI_PACKET_CONTROL, HDMI_NULL_SEND, 1);
-	WREG32(mmHDMI_VBI_PACKET_CONTROL + dig->afmt->offset, tmp); /* send null packets when required */
-
-	WREG32(mmAFMT_AUDIO_CRC_CONTROL + dig->afmt->offset, 0x1000);
-
-	tmp = RREG32(mmHDMI_CONTROL + dig->afmt->offset);
-	switch (bpc) {
-	case 0:
-	case 6:
-	case 8:
-	case 16:
-	default:
-		tmp = REG_SET_FIELD(tmp, HDMI_CONTROL, HDMI_DEEP_COLOR_ENABLE, 0);
-		tmp = REG_SET_FIELD(tmp, HDMI_CONTROL, HDMI_DEEP_COLOR_DEPTH, 0);
-		DRM_DEBUG("%s: Disabling hdmi deep color for %d bpc.\n",
-			  connector->name, bpc);
-		break;
-	case 10:
-		tmp = REG_SET_FIELD(tmp, HDMI_CONTROL, HDMI_DEEP_COLOR_ENABLE, 1);
-		tmp = REG_SET_FIELD(tmp, HDMI_CONTROL, HDMI_DEEP_COLOR_DEPTH, 1);
-		DRM_DEBUG("%s: Enabling hdmi deep color 30 for 10 bpc.\n",
-			  connector->name);
-		break;
-	case 12:
-		tmp = REG_SET_FIELD(tmp, HDMI_CONTROL, HDMI_DEEP_COLOR_ENABLE, 1);
-		tmp = REG_SET_FIELD(tmp, HDMI_CONTROL, HDMI_DEEP_COLOR_DEPTH, 2);
-		DRM_DEBUG("%s: Enabling hdmi deep color 36 for 12 bpc.\n",
-			  connector->name);
-		break;
-	}
-	WREG32(mmHDMI_CONTROL + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_VBI_PACKET_CONTROL + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_VBI_PACKET_CONTROL, HDMI_NULL_SEND, 1); /* send null packets when required */
-	tmp = REG_SET_FIELD(tmp, HDMI_VBI_PACKET_CONTROL, HDMI_GC_SEND, 1); /* send general control packets */
-	tmp = REG_SET_FIELD(tmp, HDMI_VBI_PACKET_CONTROL, HDMI_GC_CONT, 1); /* send general control packets every frame */
-	WREG32(mmHDMI_VBI_PACKET_CONTROL + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_INFOFRAME_CONTROL0 + dig->afmt->offset);
-	/* enable audio info frames (frames won't be set until audio is enabled) */
-	tmp = REG_SET_FIELD(tmp, HDMI_INFOFRAME_CONTROL0, HDMI_AUDIO_INFO_SEND, 1);
-	/* required for audio info values to be updated */
-	tmp = REG_SET_FIELD(tmp, HDMI_INFOFRAME_CONTROL0, HDMI_AUDIO_INFO_CONT, 1);
-	WREG32(mmHDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmAFMT_INFOFRAME_CONTROL0 + dig->afmt->offset);
-	/* required for audio info values to be updated */
-	tmp = REG_SET_FIELD(tmp, AFMT_INFOFRAME_CONTROL0, AFMT_AUDIO_INFO_UPDATE, 1);
-	WREG32(mmAFMT_INFOFRAME_CONTROL0 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_INFOFRAME_CONTROL1 + dig->afmt->offset);
-	/* anything other than 0 */
-	tmp = REG_SET_FIELD(tmp, HDMI_INFOFRAME_CONTROL1, HDMI_AUDIO_INFO_LINE, 2);
-	WREG32(mmHDMI_INFOFRAME_CONTROL1 + dig->afmt->offset, tmp);
-
-	WREG32(mmHDMI_GC + dig->afmt->offset, 0); /* unset HDMI_GC_AVMUTE */
-
-	tmp = RREG32(mmHDMI_AUDIO_PACKET_CONTROL + dig->afmt->offset);
-	/* set the default audio delay */
-	tmp = REG_SET_FIELD(tmp, HDMI_AUDIO_PACKET_CONTROL, HDMI_AUDIO_DELAY_EN, 1);
-	/* should be suffient for all audio modes and small enough for all hblanks */
-	tmp = REG_SET_FIELD(tmp, HDMI_AUDIO_PACKET_CONTROL, HDMI_AUDIO_PACKETS_PER_LINE, 3);
-	WREG32(mmHDMI_AUDIO_PACKET_CONTROL + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmAFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset);
-	/* allow 60958 channel status fields to be updated */
-	tmp = REG_SET_FIELD(tmp, AFMT_AUDIO_PACKET_CONTROL, AFMT_60958_CS_UPDATE, 1);
-	WREG32(mmAFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_ACR_PACKET_CONTROL + dig->afmt->offset);
-	if (bpc > 8)
-		/* clear SW CTS value */
-		tmp = REG_SET_FIELD(tmp, HDMI_ACR_PACKET_CONTROL, HDMI_ACR_SOURCE, 0);
-	else
-		/* select SW CTS value */
-		tmp = REG_SET_FIELD(tmp, HDMI_ACR_PACKET_CONTROL, HDMI_ACR_SOURCE, 1);
-	/* allow hw to sent ACR packets when required */
-	tmp = REG_SET_FIELD(tmp, HDMI_ACR_PACKET_CONTROL, HDMI_ACR_AUTO_SEND, 1);
-	WREG32(mmHDMI_ACR_PACKET_CONTROL + dig->afmt->offset, tmp);
-
-	dce_v11_0_afmt_update_ACR(encoder, mode->clock);
-
-	tmp = RREG32(mmAFMT_60958_0 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_0, AFMT_60958_CS_CHANNEL_NUMBER_L, 1);
-	WREG32(mmAFMT_60958_0 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmAFMT_60958_1 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_1, AFMT_60958_CS_CHANNEL_NUMBER_R, 2);
-	WREG32(mmAFMT_60958_1 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmAFMT_60958_2 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_2, AFMT_60958_CS_CHANNEL_NUMBER_2, 3);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_2, AFMT_60958_CS_CHANNEL_NUMBER_3, 4);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_2, AFMT_60958_CS_CHANNEL_NUMBER_4, 5);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_2, AFMT_60958_CS_CHANNEL_NUMBER_5, 6);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_2, AFMT_60958_CS_CHANNEL_NUMBER_6, 7);
-	tmp = REG_SET_FIELD(tmp, AFMT_60958_2, AFMT_60958_CS_CHANNEL_NUMBER_7, 8);
-	WREG32(mmAFMT_60958_2 + dig->afmt->offset, tmp);
-
-	dce_v11_0_audio_write_speaker_allocation(encoder);
-
-	WREG32(mmAFMT_AUDIO_PACKET_CONTROL2 + dig->afmt->offset,
-	       (0xff << AFMT_AUDIO_PACKET_CONTROL2__AFMT_AUDIO_CHANNEL_ENABLE__SHIFT));
-
-	dce_v11_0_afmt_audio_select_pin(encoder);
-	dce_v11_0_audio_write_sad_regs(encoder);
-	dce_v11_0_audio_write_latency_fields(encoder, mode);
-
-	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, connector, mode);
-	if (err < 0) {
-		DRM_ERROR("failed to setup AVI infoframe: %zd\n", err);
-		return;
-	}
-
-	err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer));
-	if (err < 0) {
-		DRM_ERROR("failed to pack AVI infoframe: %zd\n", err);
-		return;
-	}
-
-	dce_v11_0_afmt_update_avi_infoframe(encoder, buffer, sizeof(buffer));
-
-	tmp = RREG32(mmHDMI_INFOFRAME_CONTROL0 + dig->afmt->offset);
-	/* enable AVI info frames */
-	tmp = REG_SET_FIELD(tmp, HDMI_INFOFRAME_CONTROL0, HDMI_AVI_INFO_SEND, 1);
-	/* required for audio info values to be updated */
-	tmp = REG_SET_FIELD(tmp, HDMI_INFOFRAME_CONTROL0, HDMI_AVI_INFO_CONT, 1);
-	WREG32(mmHDMI_INFOFRAME_CONTROL0 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmHDMI_INFOFRAME_CONTROL1 + dig->afmt->offset);
-	tmp = REG_SET_FIELD(tmp, HDMI_INFOFRAME_CONTROL1, HDMI_AVI_INFO_LINE, 2);
-	WREG32(mmHDMI_INFOFRAME_CONTROL1 + dig->afmt->offset, tmp);
-
-	tmp = RREG32(mmAFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset);
-	/* send audio packets */
-	tmp = REG_SET_FIELD(tmp, AFMT_AUDIO_PACKET_CONTROL, AFMT_AUDIO_SAMPLE_SEND, 1);
-	WREG32(mmAFMT_AUDIO_PACKET_CONTROL + dig->afmt->offset, tmp);
-
-	WREG32(mmAFMT_RAMP_CONTROL0 + dig->afmt->offset, 0x00FFFFFF);
-	WREG32(mmAFMT_RAMP_CONTROL1 + dig->afmt->offset, 0x007FFFFF);
-	WREG32(mmAFMT_RAMP_CONTROL2 + dig->afmt->offset, 0x00000001);
-	WREG32(mmAFMT_RAMP_CONTROL3 + dig->afmt->offset, 0x00000001);
-
-	/* enable audio after to setting up hw */
-	dce_v11_0_audio_enable(adev, dig->afmt->pin, true);
-}
-
-static void dce_v11_0_afmt_enable(struct drm_encoder *encoder, bool enable)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-
-	if (!dig || !dig->afmt)
-		return;
-
-	/* Silent, r600_hdmi_enable will raise WARN for us */
-	if (enable && dig->afmt->enabled)
-		return;
-	if (!enable && !dig->afmt->enabled)
-		return;
-
-	if (!enable && dig->afmt->pin) {
-		dce_v11_0_audio_enable(adev, dig->afmt->pin, false);
-		dig->afmt->pin = NULL;
-	}
-
-	dig->afmt->enabled = enable;
-
-	DRM_DEBUG("%sabling AFMT interface @ 0x%04X for encoder 0x%x\n",
-		  enable ? "En" : "Dis", dig->afmt->offset, amdgpu_encoder->encoder_id);
-}
-
-static int dce_v11_0_afmt_init(struct amdgpu_device *adev)
-{
-	int i;
-
-	for (i = 0; i < adev->mode_info.num_dig; i++)
-		adev->mode_info.afmt[i] = NULL;
-
-	/* DCE11 has audio blocks tied to DIG encoders */
-	for (i = 0; i < adev->mode_info.num_dig; i++) {
-		adev->mode_info.afmt[i] = kzalloc(sizeof(struct amdgpu_afmt), GFP_KERNEL);
-		if (adev->mode_info.afmt[i]) {
-			adev->mode_info.afmt[i]->offset = dig_offsets[i];
-			adev->mode_info.afmt[i]->id = i;
-		} else {
-			int j;
-			for (j = 0; j < i; j++) {
-				kfree(adev->mode_info.afmt[j]);
-				adev->mode_info.afmt[j] = NULL;
-			}
-			return -ENOMEM;
-		}
-	}
-	return 0;
-}
-
-static void dce_v11_0_afmt_fini(struct amdgpu_device *adev)
-{
-	int i;
-
-	for (i = 0; i < adev->mode_info.num_dig; i++) {
-		kfree(adev->mode_info.afmt[i]);
-		adev->mode_info.afmt[i] = NULL;
-	}
-}
-
-static const u32 vga_control_regs[6] =
-{
-	mmD1VGA_CONTROL,
-	mmD2VGA_CONTROL,
-	mmD3VGA_CONTROL,
-	mmD4VGA_CONTROL,
-	mmD5VGA_CONTROL,
-	mmD6VGA_CONTROL,
-};
-
-static void dce_v11_0_vga_enable(struct drm_crtc *crtc, bool enable)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	u32 vga_control;
-
-	vga_control = RREG32(vga_control_regs[amdgpu_crtc->crtc_id]) & ~1;
-	if (enable)
-		WREG32(vga_control_regs[amdgpu_crtc->crtc_id], vga_control | 1);
-	else
-		WREG32(vga_control_regs[amdgpu_crtc->crtc_id], vga_control);
-}
-
-static void dce_v11_0_grph_enable(struct drm_crtc *crtc, bool enable)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-
-	if (enable)
-		WREG32(mmGRPH_ENABLE + amdgpu_crtc->crtc_offset, 1);
-	else
-		WREG32(mmGRPH_ENABLE + amdgpu_crtc->crtc_offset, 0);
-}
-
-static int dce_v11_0_crtc_do_set_base(struct drm_crtc *crtc,
-				     struct drm_framebuffer *fb,
-				     int x, int y, int atomic)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct drm_framebuffer *target_fb;
-	struct drm_gem_object *obj;
-	struct amdgpu_bo *abo;
-	uint64_t fb_location, tiling_flags;
-	uint32_t fb_format, fb_pitch_pixels;
-	u32 fb_swap = REG_SET_FIELD(0, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP, ENDIAN_NONE);
-	u32 pipe_config;
-	u32 tmp, viewport_w, viewport_h;
-	int r;
-	bool bypass_lut = false;
-
-	/* no fb bound */
-	if (!atomic && !crtc->primary->fb) {
-		DRM_DEBUG_KMS("No FB bound\n");
-		return 0;
-	}
-
-	if (atomic)
-		target_fb = fb;
-	else
-		target_fb = crtc->primary->fb;
-
-	/* If atomic, assume fb object is pinned & idle & fenced and
-	 * just update base pointers
-	 */
-	obj = target_fb->obj[0];
-	abo = gem_to_amdgpu_bo(obj);
-	r = amdgpu_bo_reserve(abo, false);
-	if (unlikely(r != 0))
-		return r;
-
-	if (!atomic) {
-		abo->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
-		r = amdgpu_bo_pin(abo, AMDGPU_GEM_DOMAIN_VRAM);
-		if (unlikely(r != 0)) {
-			amdgpu_bo_unreserve(abo);
-			return -EINVAL;
-		}
-	}
-	fb_location = amdgpu_bo_gpu_offset(abo);
-
-	amdgpu_bo_get_tiling_flags(abo, &tiling_flags);
-	amdgpu_bo_unreserve(abo);
-
-	pipe_config = AMDGPU_TILING_GET(tiling_flags, PIPE_CONFIG);
-
-	switch (target_fb->format->format) {
-	case DRM_FORMAT_C8:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 0);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 0);
-		break;
-	case DRM_FORMAT_XRGB4444:
-	case DRM_FORMAT_ARGB4444:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 1);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 2);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN16);
-#endif
-		break;
-	case DRM_FORMAT_XRGB1555:
-	case DRM_FORMAT_ARGB1555:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 1);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 0);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN16);
-#endif
-		break;
-	case DRM_FORMAT_BGRX5551:
-	case DRM_FORMAT_BGRA5551:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 1);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 5);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN16);
-#endif
-		break;
-	case DRM_FORMAT_RGB565:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 1);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 1);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN16);
-#endif
-		break;
-	case DRM_FORMAT_XRGB8888:
-	case DRM_FORMAT_ARGB8888:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 2);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 0);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN32);
-#endif
-		break;
-	case DRM_FORMAT_XRGB2101010:
-	case DRM_FORMAT_ARGB2101010:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 2);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 1);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN32);
-#endif
-		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
-		bypass_lut = true;
-		break;
-	case DRM_FORMAT_BGRX1010102:
-	case DRM_FORMAT_BGRA1010102:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 2);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 4);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN32);
-#endif
-		/* Greater 8 bpc fb needs to bypass hw-lut to retain precision */
-		bypass_lut = true;
-		break;
-	case DRM_FORMAT_XBGR8888:
-	case DRM_FORMAT_ABGR8888:
-		fb_format = REG_SET_FIELD(0, GRPH_CONTROL, GRPH_DEPTH, 2);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_FORMAT, 0);
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_RED_CROSSBAR, 2);
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_BLUE_CROSSBAR, 2);
-#ifdef __BIG_ENDIAN
-		fb_swap = REG_SET_FIELD(fb_swap, GRPH_SWAP_CNTL, GRPH_ENDIAN_SWAP,
-					ENDIAN_8IN32);
-#endif
-		break;
-	default:
-		DRM_ERROR("Unsupported screen format %p4cc\n",
-			  &target_fb->format->format);
-		return -EINVAL;
-	}
-
-	if (AMDGPU_TILING_GET(tiling_flags, ARRAY_MODE) == ARRAY_2D_TILED_THIN1) {
-		unsigned bankw, bankh, mtaspect, tile_split, num_banks;
-
-		bankw = AMDGPU_TILING_GET(tiling_flags, BANK_WIDTH);
-		bankh = AMDGPU_TILING_GET(tiling_flags, BANK_HEIGHT);
-		mtaspect = AMDGPU_TILING_GET(tiling_flags, MACRO_TILE_ASPECT);
-		tile_split = AMDGPU_TILING_GET(tiling_flags, TILE_SPLIT);
-		num_banks = AMDGPU_TILING_GET(tiling_flags, NUM_BANKS);
-
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_NUM_BANKS, num_banks);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_ARRAY_MODE,
-					  ARRAY_2D_TILED_THIN1);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_TILE_SPLIT,
-					  tile_split);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_BANK_WIDTH, bankw);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_BANK_HEIGHT, bankh);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_MACRO_TILE_ASPECT,
-					  mtaspect);
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_MICRO_TILE_MODE,
-					  ADDR_SURF_MICRO_TILING_DISPLAY);
-	} else if (AMDGPU_TILING_GET(tiling_flags, ARRAY_MODE) == ARRAY_1D_TILED_THIN1) {
-		fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_ARRAY_MODE,
-					  ARRAY_1D_TILED_THIN1);
-	}
-
-	fb_format = REG_SET_FIELD(fb_format, GRPH_CONTROL, GRPH_PIPE_CONFIG,
-				  pipe_config);
-
-	dce_v11_0_vga_enable(crtc, false);
-
-	/* Make sure surface address is updated at vertical blank rather than
-	 * horizontal blank
-	 */
-	tmp = RREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, GRPH_FLIP_CONTROL,
-			    GRPH_SURFACE_UPDATE_H_RETRACE_EN, 0);
-	WREG32(mmGRPH_FLIP_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(fb_location));
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(fb_location));
-	WREG32(mmGRPH_PRIMARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       (u32)fb_location & GRPH_PRIMARY_SURFACE_ADDRESS__GRPH_PRIMARY_SURFACE_ADDRESS_MASK);
-	WREG32(mmGRPH_SECONDARY_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       (u32) fb_location & GRPH_SECONDARY_SURFACE_ADDRESS__GRPH_SECONDARY_SURFACE_ADDRESS_MASK);
-	WREG32(mmGRPH_CONTROL + amdgpu_crtc->crtc_offset, fb_format);
-	WREG32(mmGRPH_SWAP_CNTL + amdgpu_crtc->crtc_offset, fb_swap);
-
-	/*
-	 * The LUT only has 256 slots for indexing by a 8 bpc fb. Bypass the LUT
-	 * for > 8 bpc scanout to avoid truncation of fb indices to 8 msb's, to
-	 * retain the full precision throughout the pipeline.
-	 */
-	tmp = RREG32(mmGRPH_LUT_10BIT_BYPASS + amdgpu_crtc->crtc_offset);
-	if (bypass_lut)
-		tmp = REG_SET_FIELD(tmp, GRPH_LUT_10BIT_BYPASS, GRPH_LUT_10BIT_BYPASS_EN, 1);
-	else
-		tmp = REG_SET_FIELD(tmp, GRPH_LUT_10BIT_BYPASS, GRPH_LUT_10BIT_BYPASS_EN, 0);
-	WREG32(mmGRPH_LUT_10BIT_BYPASS + amdgpu_crtc->crtc_offset, tmp);
-
-	if (bypass_lut)
-		DRM_DEBUG_KMS("Bypassing hardware LUT due to 10 bit fb scanout.\n");
-
-	WREG32(mmGRPH_SURFACE_OFFSET_X + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmGRPH_SURFACE_OFFSET_Y + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmGRPH_X_START + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmGRPH_Y_START + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmGRPH_X_END + amdgpu_crtc->crtc_offset, target_fb->width);
-	WREG32(mmGRPH_Y_END + amdgpu_crtc->crtc_offset, target_fb->height);
-
-	fb_pitch_pixels = target_fb->pitches[0] / target_fb->format->cpp[0];
-	WREG32(mmGRPH_PITCH + amdgpu_crtc->crtc_offset, fb_pitch_pixels);
-
-	dce_v11_0_grph_enable(crtc, true);
-
-	WREG32(mmLB_DESKTOP_HEIGHT + amdgpu_crtc->crtc_offset,
-	       target_fb->height);
-
-	x &= ~3;
-	y &= ~1;
-	WREG32(mmVIEWPORT_START + amdgpu_crtc->crtc_offset,
-	       (x << 16) | y);
-	viewport_w = crtc->mode.hdisplay;
-	viewport_h = (crtc->mode.vdisplay + 1) & ~1;
-	WREG32(mmVIEWPORT_SIZE + amdgpu_crtc->crtc_offset,
-	       (viewport_w << 16) | viewport_h);
-
-	/* set pageflip to happen anywhere in vblank interval */
-	WREG32(mmCRTC_MASTER_UPDATE_MODE + amdgpu_crtc->crtc_offset, 0);
-
-	if (!atomic && fb && fb != crtc->primary->fb) {
-		abo = gem_to_amdgpu_bo(fb->obj[0]);
-		r = amdgpu_bo_reserve(abo, true);
-		if (unlikely(r != 0))
-			return r;
-		amdgpu_bo_unpin(abo);
-		amdgpu_bo_unreserve(abo);
-	}
-
-	/* Bytes per pixel may have changed */
-	dce_v11_0_bandwidth_update(adev);
-
-	return 0;
-}
-
-static void dce_v11_0_set_interleave(struct drm_crtc *crtc,
-				     struct drm_display_mode *mode)
-{
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	u32 tmp;
-
-	tmp = RREG32(mmLB_DATA_FORMAT + amdgpu_crtc->crtc_offset);
-	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
-		tmp = REG_SET_FIELD(tmp, LB_DATA_FORMAT, INTERLEAVE_EN, 1);
-	else
-		tmp = REG_SET_FIELD(tmp, LB_DATA_FORMAT, INTERLEAVE_EN, 0);
-	WREG32(mmLB_DATA_FORMAT + amdgpu_crtc->crtc_offset, tmp);
-}
-
-static void dce_v11_0_crtc_load_lut(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	u16 *r, *g, *b;
-	int i;
-	u32 tmp;
-
-	DRM_DEBUG_KMS("%d\n", amdgpu_crtc->crtc_id);
-
-	tmp = RREG32(mmINPUT_CSC_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, INPUT_CSC_CONTROL, INPUT_CSC_GRPH_MODE, 0);
-	WREG32(mmINPUT_CSC_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	tmp = RREG32(mmPRESCALE_GRPH_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, PRESCALE_GRPH_CONTROL, GRPH_PRESCALE_BYPASS, 1);
-	WREG32(mmPRESCALE_GRPH_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	tmp = RREG32(mmINPUT_GAMMA_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, INPUT_GAMMA_CONTROL, GRPH_INPUT_GAMMA_MODE, 0);
-	WREG32(mmINPUT_GAMMA_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	WREG32(mmDC_LUT_CONTROL + amdgpu_crtc->crtc_offset, 0);
-
-	WREG32(mmDC_LUT_BLACK_OFFSET_BLUE + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmDC_LUT_BLACK_OFFSET_GREEN + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmDC_LUT_BLACK_OFFSET_RED + amdgpu_crtc->crtc_offset, 0);
-
-	WREG32(mmDC_LUT_WHITE_OFFSET_BLUE + amdgpu_crtc->crtc_offset, 0xffff);
-	WREG32(mmDC_LUT_WHITE_OFFSET_GREEN + amdgpu_crtc->crtc_offset, 0xffff);
-	WREG32(mmDC_LUT_WHITE_OFFSET_RED + amdgpu_crtc->crtc_offset, 0xffff);
-
-	WREG32(mmDC_LUT_RW_MODE + amdgpu_crtc->crtc_offset, 0);
-	WREG32(mmDC_LUT_WRITE_EN_MASK + amdgpu_crtc->crtc_offset, 0x00000007);
-
-	WREG32(mmDC_LUT_RW_INDEX + amdgpu_crtc->crtc_offset, 0);
-	r = crtc->gamma_store;
-	g = r + crtc->gamma_size;
-	b = g + crtc->gamma_size;
-	for (i = 0; i < 256; i++) {
-		WREG32(mmDC_LUT_30_COLOR + amdgpu_crtc->crtc_offset,
-		       ((*r++ & 0xffc0) << 14) |
-		       ((*g++ & 0xffc0) << 4) |
-		       (*b++ >> 6));
-	}
-
-	tmp = RREG32(mmDEGAMMA_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, DEGAMMA_CONTROL, GRPH_DEGAMMA_MODE, 0);
-	tmp = REG_SET_FIELD(tmp, DEGAMMA_CONTROL, CURSOR_DEGAMMA_MODE, 0);
-	tmp = REG_SET_FIELD(tmp, DEGAMMA_CONTROL, CURSOR2_DEGAMMA_MODE, 0);
-	WREG32(mmDEGAMMA_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	tmp = RREG32(mmGAMUT_REMAP_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, GAMUT_REMAP_CONTROL, GRPH_GAMUT_REMAP_MODE, 0);
-	WREG32(mmGAMUT_REMAP_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	tmp = RREG32(mmREGAMMA_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, REGAMMA_CONTROL, GRPH_REGAMMA_MODE, 0);
-	WREG32(mmREGAMMA_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	tmp = RREG32(mmOUTPUT_CSC_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, OUTPUT_CSC_CONTROL, OUTPUT_CSC_GRPH_MODE, 0);
-	WREG32(mmOUTPUT_CSC_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-
-	/* XXX match this to the depth of the crtc fmt block, move to modeset? */
-	WREG32(mmDENORM_CONTROL + amdgpu_crtc->crtc_offset, 0);
-	/* XXX this only needs to be programmed once per crtc at startup,
-	 * not sure where the best place for it is
-	 */
-	tmp = RREG32(mmALPHA_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, ALPHA_CONTROL, CURSOR_ALPHA_BLND_ENA, 1);
-	WREG32(mmALPHA_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-}
-
-static int dce_v11_0_pick_dig_encoder(struct drm_encoder *encoder)
-{
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-
-	switch (amdgpu_encoder->encoder_id) {
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
-		if (dig->linkb)
-			return 1;
-		else
-			return 0;
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1:
-		if (dig->linkb)
-			return 3;
-		else
-			return 2;
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY2:
-		if (dig->linkb)
-			return 5;
-		else
-			return 4;
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY3:
-		return 6;
-	default:
-		DRM_ERROR("invalid encoder_id: 0x%x\n", amdgpu_encoder->encoder_id);
-		return 0;
-	}
-}
-
-/**
- * dce_v11_0_pick_pll - Allocate a PPLL for use by the crtc.
- *
- * @crtc: drm crtc
- *
- * Returns the PPLL (Pixel PLL) to be used by the crtc.  For DP monitors
- * a single PPLL can be used for all DP crtcs/encoders.  For non-DP
- * monitors a dedicated PPLL must be used.  If a particular board has
- * an external DP PLL, return ATOM_PPLL_INVALID to skip PLL programming
- * as there is no need to program the PLL itself.  If we are not able to
- * allocate a PLL, return ATOM_PPLL_INVALID to skip PLL programming to
- * avoid messing up an existing monitor.
- *
- * Asic specific PLL information
- *
- * DCE 10.x
- * Tonga
- * - PPLL1, PPLL2 are available for all UNIPHY (both DP and non-DP)
- * CI
- * - PPLL0, PPLL1, PPLL2 are available for all UNIPHY (both DP and non-DP) and DAC
- *
- */
-static u32 dce_v11_0_pick_pll(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	u32 pll_in_use;
-	int pll;
-
-	if ((adev->asic_type == CHIP_POLARIS10) ||
-	    (adev->asic_type == CHIP_POLARIS11) ||
-	    (adev->asic_type == CHIP_POLARIS12) ||
-	    (adev->asic_type == CHIP_VEGAM)) {
-		struct amdgpu_encoder *amdgpu_encoder =
-			to_amdgpu_encoder(amdgpu_crtc->encoder);
-		struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-
-		if (ENCODER_MODE_IS_DP(amdgpu_atombios_encoder_get_encoder_mode(amdgpu_crtc->encoder)))
-			return ATOM_DP_DTO;
-
-		switch (amdgpu_encoder->encoder_id) {
-		case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
-			if (dig->linkb)
-				return ATOM_COMBOPHY_PLL1;
-			else
-				return ATOM_COMBOPHY_PLL0;
-		case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1:
-			if (dig->linkb)
-				return ATOM_COMBOPHY_PLL3;
-			else
-				return ATOM_COMBOPHY_PLL2;
-		case ENCODER_OBJECT_ID_INTERNAL_UNIPHY2:
-			if (dig->linkb)
-				return ATOM_COMBOPHY_PLL5;
-			else
-				return ATOM_COMBOPHY_PLL4;
-		default:
-			DRM_ERROR("invalid encoder_id: 0x%x\n", amdgpu_encoder->encoder_id);
-			return ATOM_PPLL_INVALID;
-		}
-	}
-
-	if (ENCODER_MODE_IS_DP(amdgpu_atombios_encoder_get_encoder_mode(amdgpu_crtc->encoder))) {
-		if (adev->clock.dp_extclk)
-			/* skip PPLL programming if using ext clock */
-			return ATOM_PPLL_INVALID;
-		else {
-			/* use the same PPLL for all DP monitors */
-			pll = amdgpu_pll_get_shared_dp_ppll(crtc);
-			if (pll != ATOM_PPLL_INVALID)
-				return pll;
-		}
-	} else {
-		/* use the same PPLL for all monitors with the same clock */
-		pll = amdgpu_pll_get_shared_nondp_ppll(crtc);
-		if (pll != ATOM_PPLL_INVALID)
-			return pll;
-	}
-
-	/* XXX need to determine what plls are available on each DCE11 part */
-	pll_in_use = amdgpu_pll_get_use_mask(crtc);
-	if (adev->flags & AMD_IS_APU) {
-		if (!(pll_in_use & (1 << ATOM_PPLL1)))
-			return ATOM_PPLL1;
-		if (!(pll_in_use & (1 << ATOM_PPLL0)))
-			return ATOM_PPLL0;
-		DRM_ERROR("unable to allocate a PPLL\n");
-		return ATOM_PPLL_INVALID;
-	} else {
-		if (!(pll_in_use & (1 << ATOM_PPLL2)))
-			return ATOM_PPLL2;
-		if (!(pll_in_use & (1 << ATOM_PPLL1)))
-			return ATOM_PPLL1;
-		if (!(pll_in_use & (1 << ATOM_PPLL0)))
-			return ATOM_PPLL0;
-		DRM_ERROR("unable to allocate a PPLL\n");
-		return ATOM_PPLL_INVALID;
-	}
-	return ATOM_PPLL_INVALID;
-}
-
-static void dce_v11_0_lock_cursor(struct drm_crtc *crtc, bool lock)
-{
-	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	uint32_t cur_lock;
-
-	cur_lock = RREG32(mmCUR_UPDATE + amdgpu_crtc->crtc_offset);
-	if (lock)
-		cur_lock = REG_SET_FIELD(cur_lock, CUR_UPDATE, CURSOR_UPDATE_LOCK, 1);
-	else
-		cur_lock = REG_SET_FIELD(cur_lock, CUR_UPDATE, CURSOR_UPDATE_LOCK, 0);
-	WREG32(mmCUR_UPDATE + amdgpu_crtc->crtc_offset, cur_lock);
-}
-
-static void dce_v11_0_hide_cursor(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
-	u32 tmp;
-
-	tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 0);
-	WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-}
-
-static void dce_v11_0_show_cursor(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
-	u32 tmp;
-
-	WREG32(mmCUR_SURFACE_ADDRESS_HIGH + amdgpu_crtc->crtc_offset,
-	       upper_32_bits(amdgpu_crtc->cursor_addr));
-	WREG32(mmCUR_SURFACE_ADDRESS + amdgpu_crtc->crtc_offset,
-	       lower_32_bits(amdgpu_crtc->cursor_addr));
-
-	tmp = RREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset);
-	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_EN, 1);
-	tmp = REG_SET_FIELD(tmp, CUR_CONTROL, CURSOR_MODE, 2);
-	WREG32(mmCUR_CONTROL + amdgpu_crtc->crtc_offset, tmp);
-}
-
-static int dce_v11_0_cursor_move_locked(struct drm_crtc *crtc,
-					int x, int y)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct amdgpu_device *adev = drm_to_adev(crtc->dev);
-	int xorigin = 0, yorigin = 0;
-
-	amdgpu_crtc->cursor_x = x;
-	amdgpu_crtc->cursor_y = y;
-
-	/* avivo cursor are offset into the total surface */
-	x += crtc->x;
-	y += crtc->y;
-	DRM_DEBUG("x %d y %d c->x %d c->y %d\n", x, y, crtc->x, crtc->y);
-
-	if (x < 0) {
-		xorigin = min(-x, amdgpu_crtc->max_cursor_width - 1);
-		x = 0;
-	}
-	if (y < 0) {
-		yorigin = min(-y, amdgpu_crtc->max_cursor_height - 1);
-		y = 0;
-	}
-
-	WREG32(mmCUR_POSITION + amdgpu_crtc->crtc_offset, (x << 16) | y);
-	WREG32(mmCUR_HOT_SPOT + amdgpu_crtc->crtc_offset, (xorigin << 16) | yorigin);
-	WREG32(mmCUR_SIZE + amdgpu_crtc->crtc_offset,
-	       ((amdgpu_crtc->cursor_width - 1) << 16) | (amdgpu_crtc->cursor_height - 1));
-
-	return 0;
-}
-
-static int dce_v11_0_crtc_cursor_move(struct drm_crtc *crtc,
-				      int x, int y)
-{
-	int ret;
-
-	dce_v11_0_lock_cursor(crtc, true);
-	ret = dce_v11_0_cursor_move_locked(crtc, x, y);
-	dce_v11_0_lock_cursor(crtc, false);
-
-	return ret;
-}
-
-static int dce_v11_0_crtc_cursor_set2(struct drm_crtc *crtc,
-				      struct drm_file *file_priv,
-				      uint32_t handle,
-				      uint32_t width,
-				      uint32_t height,
-				      int32_t hot_x,
-				      int32_t hot_y)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_gem_object *obj;
-	struct amdgpu_bo *aobj;
-	int ret;
-
-	if (!handle) {
-		/* turn off cursor */
-		dce_v11_0_hide_cursor(crtc);
-		obj = NULL;
-		goto unpin;
-	}
-
-	if ((width > amdgpu_crtc->max_cursor_width) ||
-	    (height > amdgpu_crtc->max_cursor_height)) {
-		DRM_ERROR("bad cursor width or height %d x %d\n", width, height);
-		return -EINVAL;
-	}
-
-	obj = drm_gem_object_lookup(file_priv, handle);
-	if (!obj) {
-		DRM_ERROR("Cannot find cursor object %x for crtc %d\n", handle, amdgpu_crtc->crtc_id);
-		return -ENOENT;
-	}
-
-	aobj = gem_to_amdgpu_bo(obj);
-	ret = amdgpu_bo_reserve(aobj, false);
-	if (ret != 0) {
-		drm_gem_object_put(obj);
-		return ret;
-	}
-
-	aobj->flags |= AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS;
-	ret = amdgpu_bo_pin(aobj, AMDGPU_GEM_DOMAIN_VRAM);
-	amdgpu_bo_unreserve(aobj);
-	if (ret) {
-		DRM_ERROR("Failed to pin new cursor BO (%d)\n", ret);
-		drm_gem_object_put(obj);
-		return ret;
-	}
-	amdgpu_crtc->cursor_addr = amdgpu_bo_gpu_offset(aobj);
-
-	dce_v11_0_lock_cursor(crtc, true);
-
-	if (width != amdgpu_crtc->cursor_width ||
-	    height != amdgpu_crtc->cursor_height ||
-	    hot_x != amdgpu_crtc->cursor_hot_x ||
-	    hot_y != amdgpu_crtc->cursor_hot_y) {
-		int x, y;
-
-		x = amdgpu_crtc->cursor_x + amdgpu_crtc->cursor_hot_x - hot_x;
-		y = amdgpu_crtc->cursor_y + amdgpu_crtc->cursor_hot_y - hot_y;
-
-		dce_v11_0_cursor_move_locked(crtc, x, y);
-
-		amdgpu_crtc->cursor_width = width;
-		amdgpu_crtc->cursor_height = height;
-		amdgpu_crtc->cursor_hot_x = hot_x;
-		amdgpu_crtc->cursor_hot_y = hot_y;
-	}
-
-	dce_v11_0_show_cursor(crtc);
-	dce_v11_0_lock_cursor(crtc, false);
-
-unpin:
-	if (amdgpu_crtc->cursor_bo) {
-		struct amdgpu_bo *aobj = gem_to_amdgpu_bo(amdgpu_crtc->cursor_bo);
-		ret = amdgpu_bo_reserve(aobj, true);
-		if (likely(ret == 0)) {
-			amdgpu_bo_unpin(aobj);
-			amdgpu_bo_unreserve(aobj);
-		}
-		drm_gem_object_put(amdgpu_crtc->cursor_bo);
-	}
-
-	amdgpu_crtc->cursor_bo = obj;
-	return 0;
-}
-
-static void dce_v11_0_cursor_reset(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-
-	if (amdgpu_crtc->cursor_bo) {
-		dce_v11_0_lock_cursor(crtc, true);
-
-		dce_v11_0_cursor_move_locked(crtc, amdgpu_crtc->cursor_x,
-					     amdgpu_crtc->cursor_y);
-
-		dce_v11_0_show_cursor(crtc);
-
-		dce_v11_0_lock_cursor(crtc, false);
-	}
-}
-
-static int dce_v11_0_crtc_gamma_set(struct drm_crtc *crtc, u16 *red, u16 *green,
-				    u16 *blue, uint32_t size,
-				    struct drm_modeset_acquire_ctx *ctx)
-{
-	dce_v11_0_crtc_load_lut(crtc);
-
-	return 0;
-}
-
-static void dce_v11_0_crtc_destroy(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-
-	drm_crtc_cleanup(crtc);
-	kfree(amdgpu_crtc);
-}
-
-static const struct drm_crtc_funcs dce_v11_0_crtc_funcs = {
-	.cursor_set2 = dce_v11_0_crtc_cursor_set2,
-	.cursor_move = dce_v11_0_crtc_cursor_move,
-	.gamma_set = dce_v11_0_crtc_gamma_set,
-	.set_config = amdgpu_display_crtc_set_config,
-	.destroy = dce_v11_0_crtc_destroy,
-	.page_flip_target = amdgpu_display_crtc_page_flip_target,
-	.get_vblank_counter = amdgpu_get_vblank_counter_kms,
-	.enable_vblank = amdgpu_enable_vblank_kms,
-	.disable_vblank = amdgpu_disable_vblank_kms,
-	.get_vblank_timestamp = drm_crtc_vblank_helper_get_vblank_timestamp,
-};
-
-static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode)
-{
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	unsigned type;
-
-	switch (mode) {
-	case DRM_MODE_DPMS_ON:
-		amdgpu_crtc->enabled = true;
-		amdgpu_atombios_crtc_enable(crtc, ATOM_ENABLE);
-		dce_v11_0_vga_enable(crtc, true);
-		amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE);
-		dce_v11_0_vga_enable(crtc, false);
-		/* Make sure VBLANK and PFLIP interrupts are still enabled */
-		type = amdgpu_display_crtc_idx_to_irq_type(adev,
-						amdgpu_crtc->crtc_id);
-		amdgpu_irq_update(adev, &adev->crtc_irq, type);
-		amdgpu_irq_update(adev, &adev->pageflip_irq, type);
-		drm_crtc_vblank_on(crtc);
-		dce_v11_0_crtc_load_lut(crtc);
-		break;
-	case DRM_MODE_DPMS_STANDBY:
-	case DRM_MODE_DPMS_SUSPEND:
-	case DRM_MODE_DPMS_OFF:
-		drm_crtc_vblank_off(crtc);
-		if (amdgpu_crtc->enabled) {
-			dce_v11_0_vga_enable(crtc, true);
-			amdgpu_atombios_crtc_blank(crtc, ATOM_ENABLE);
-			dce_v11_0_vga_enable(crtc, false);
-		}
-		amdgpu_atombios_crtc_enable(crtc, ATOM_DISABLE);
-		amdgpu_crtc->enabled = false;
-		break;
-	}
-	/* adjust pm to dpms */
-	amdgpu_dpm_compute_clocks(adev);
-}
-
-static void dce_v11_0_crtc_prepare(struct drm_crtc *crtc)
-{
-	/* disable crtc pair power gating before programming */
-	amdgpu_atombios_crtc_powergate(crtc, ATOM_DISABLE);
-	amdgpu_atombios_crtc_lock(crtc, ATOM_ENABLE);
-	dce_v11_0_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
-}
-
-static void dce_v11_0_crtc_commit(struct drm_crtc *crtc)
-{
-	dce_v11_0_crtc_dpms(crtc, DRM_MODE_DPMS_ON);
-	amdgpu_atombios_crtc_lock(crtc, ATOM_DISABLE);
-}
-
-static void dce_v11_0_crtc_disable(struct drm_crtc *crtc)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-	struct amdgpu_atom_ss ss;
-	int i;
-
-	dce_v11_0_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
-	if (crtc->primary->fb) {
-		int r;
-		struct amdgpu_bo *abo;
-
-		abo = gem_to_amdgpu_bo(crtc->primary->fb->obj[0]);
-		r = amdgpu_bo_reserve(abo, true);
-		if (unlikely(r))
-			DRM_ERROR("failed to reserve abo before unpin\n");
-		else {
-			amdgpu_bo_unpin(abo);
-			amdgpu_bo_unreserve(abo);
-		}
-	}
-	/* disable the GRPH */
-	dce_v11_0_grph_enable(crtc, false);
-
-	amdgpu_atombios_crtc_powergate(crtc, ATOM_ENABLE);
-
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		if (adev->mode_info.crtcs[i] &&
-		    adev->mode_info.crtcs[i]->enabled &&
-		    i != amdgpu_crtc->crtc_id &&
-		    amdgpu_crtc->pll_id == adev->mode_info.crtcs[i]->pll_id) {
-			/* one other crtc is using this pll don't turn
-			 * off the pll
-			 */
-			goto done;
-		}
-	}
-
-	switch (amdgpu_crtc->pll_id) {
-	case ATOM_PPLL0:
-	case ATOM_PPLL1:
-	case ATOM_PPLL2:
-		/* disable the ppll */
-		amdgpu_atombios_crtc_program_pll(crtc, amdgpu_crtc->crtc_id, amdgpu_crtc->pll_id,
-						 0, 0, ATOM_DISABLE, 0, 0, 0, 0, 0, false, &ss);
-		break;
-	case ATOM_COMBOPHY_PLL0:
-	case ATOM_COMBOPHY_PLL1:
-	case ATOM_COMBOPHY_PLL2:
-	case ATOM_COMBOPHY_PLL3:
-	case ATOM_COMBOPHY_PLL4:
-	case ATOM_COMBOPHY_PLL5:
-		/* disable the ppll */
-		amdgpu_atombios_crtc_program_pll(crtc, ATOM_CRTC_INVALID, amdgpu_crtc->pll_id,
-						 0, 0, ATOM_DISABLE, 0, 0, 0, 0, 0, false, &ss);
-		break;
-	default:
-		break;
-	}
-done:
-	amdgpu_crtc->pll_id = ATOM_PPLL_INVALID;
-	amdgpu_crtc->adjusted_clock = 0;
-	amdgpu_crtc->encoder = NULL;
-	amdgpu_crtc->connector = NULL;
-}
-
-static int dce_v11_0_crtc_mode_set(struct drm_crtc *crtc,
-				  struct drm_display_mode *mode,
-				  struct drm_display_mode *adjusted_mode,
-				  int x, int y, struct drm_framebuffer *old_fb)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-
-	if (!amdgpu_crtc->adjusted_clock)
-		return -EINVAL;
-
-	if ((adev->asic_type == CHIP_POLARIS10) ||
-	    (adev->asic_type == CHIP_POLARIS11) ||
-	    (adev->asic_type == CHIP_POLARIS12) ||
-	    (adev->asic_type == CHIP_VEGAM)) {
-		struct amdgpu_encoder *amdgpu_encoder =
-			to_amdgpu_encoder(amdgpu_crtc->encoder);
-		int encoder_mode =
-			amdgpu_atombios_encoder_get_encoder_mode(amdgpu_crtc->encoder);
-
-		/* SetPixelClock calculates the plls and ss values now */
-		amdgpu_atombios_crtc_program_pll(crtc, amdgpu_crtc->crtc_id,
-						 amdgpu_crtc->pll_id,
-						 encoder_mode, amdgpu_encoder->encoder_id,
-						 adjusted_mode->clock, 0, 0, 0, 0,
-						 amdgpu_crtc->bpc, amdgpu_crtc->ss_enabled, &amdgpu_crtc->ss);
-	} else {
-		amdgpu_atombios_crtc_set_pll(crtc, adjusted_mode);
-	}
-	amdgpu_atombios_crtc_set_dtd_timing(crtc, adjusted_mode);
-	dce_v11_0_crtc_do_set_base(crtc, old_fb, x, y, 0);
-	amdgpu_atombios_crtc_overscan_setup(crtc, mode, adjusted_mode);
-	amdgpu_atombios_crtc_scaler_setup(crtc);
-	dce_v11_0_cursor_reset(crtc);
-	/* update the hw version fpr dpm */
-	amdgpu_crtc->hw_mode = *adjusted_mode;
-
-	return 0;
-}
-
-static bool dce_v11_0_crtc_mode_fixup(struct drm_crtc *crtc,
-				     const struct drm_display_mode *mode,
-				     struct drm_display_mode *adjusted_mode)
-{
-	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct drm_encoder *encoder;
-
-	/* assign the encoder to the amdgpu crtc to avoid repeated lookups later */
-	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-		if (encoder->crtc == crtc) {
-			amdgpu_crtc->encoder = encoder;
-			amdgpu_crtc->connector = amdgpu_get_connector_for_encoder(encoder);
-			break;
-		}
-	}
-	if ((amdgpu_crtc->encoder == NULL) || (amdgpu_crtc->connector == NULL)) {
-		amdgpu_crtc->encoder = NULL;
-		amdgpu_crtc->connector = NULL;
-		return false;
-	}
-	if (!amdgpu_display_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode))
-		return false;
-	if (amdgpu_atombios_crtc_prepare_pll(crtc, adjusted_mode))
-		return false;
-	/* pick pll */
-	amdgpu_crtc->pll_id = dce_v11_0_pick_pll(crtc);
-	/* if we can't get a PPLL for a non-DP encoder, fail */
-	if ((amdgpu_crtc->pll_id == ATOM_PPLL_INVALID) &&
-	    !ENCODER_MODE_IS_DP(amdgpu_atombios_encoder_get_encoder_mode(amdgpu_crtc->encoder)))
-		return false;
-
-	return true;
-}
-
-static int dce_v11_0_crtc_set_base(struct drm_crtc *crtc, int x, int y,
-				  struct drm_framebuffer *old_fb)
-{
-	return dce_v11_0_crtc_do_set_base(crtc, old_fb, x, y, 0);
-}
-
-static int dce_v11_0_crtc_set_base_atomic(struct drm_crtc *crtc,
-					 struct drm_framebuffer *fb,
-					 int x, int y, enum mode_set_atomic state)
-{
-	return dce_v11_0_crtc_do_set_base(crtc, fb, x, y, 1);
-}
-
-static const struct drm_crtc_helper_funcs dce_v11_0_crtc_helper_funcs = {
-	.dpms = dce_v11_0_crtc_dpms,
-	.mode_fixup = dce_v11_0_crtc_mode_fixup,
-	.mode_set = dce_v11_0_crtc_mode_set,
-	.mode_set_base = dce_v11_0_crtc_set_base,
-	.mode_set_base_atomic = dce_v11_0_crtc_set_base_atomic,
-	.prepare = dce_v11_0_crtc_prepare,
-	.commit = dce_v11_0_crtc_commit,
-	.disable = dce_v11_0_crtc_disable,
-	.get_scanout_position = amdgpu_crtc_get_scanout_position,
-};
-
-static void dce_v11_0_panic_flush(struct drm_plane *plane)
-{
-	struct drm_framebuffer *fb;
-	struct amdgpu_crtc *amdgpu_crtc;
-	struct amdgpu_device *adev;
-	uint32_t fb_format;
-
-	if (!plane->fb)
-		return;
-
-	fb = plane->fb;
-	amdgpu_crtc = to_amdgpu_crtc(plane->crtc);
-	adev = drm_to_adev(fb->dev);
-
-	/* Disable DC tiling */
-	fb_format = RREG32(mmGRPH_CONTROL + amdgpu_crtc->crtc_offset);
-	fb_format &= ~GRPH_CONTROL__GRPH_ARRAY_MODE_MASK;
-	WREG32(mmGRPH_CONTROL + amdgpu_crtc->crtc_offset, fb_format);
-
-}
-
-static const struct drm_plane_helper_funcs dce_v11_0_drm_primary_plane_helper_funcs = {
-	.get_scanout_buffer = amdgpu_display_get_scanout_buffer,
-	.panic_flush = dce_v11_0_panic_flush,
-};
-
-static int dce_v11_0_crtc_init(struct amdgpu_device *adev, int index)
-{
-	struct amdgpu_crtc *amdgpu_crtc;
-
-	amdgpu_crtc = kzalloc(sizeof(struct amdgpu_crtc) +
-			      (AMDGPUFB_CONN_LIMIT * sizeof(struct drm_connector *)), GFP_KERNEL);
-	if (amdgpu_crtc == NULL)
-		return -ENOMEM;
-
-	drm_crtc_init(adev_to_drm(adev), &amdgpu_crtc->base, &dce_v11_0_crtc_funcs);
-
-	drm_mode_crtc_set_gamma_size(&amdgpu_crtc->base, 256);
-	amdgpu_crtc->crtc_id = index;
-	adev->mode_info.crtcs[index] = amdgpu_crtc;
-
-	amdgpu_crtc->max_cursor_width = 128;
-	amdgpu_crtc->max_cursor_height = 128;
-	adev_to_drm(adev)->mode_config.cursor_width = amdgpu_crtc->max_cursor_width;
-	adev_to_drm(adev)->mode_config.cursor_height = amdgpu_crtc->max_cursor_height;
-
-	switch (amdgpu_crtc->crtc_id) {
-	case 0:
-	default:
-		amdgpu_crtc->crtc_offset = CRTC0_REGISTER_OFFSET;
-		break;
-	case 1:
-		amdgpu_crtc->crtc_offset = CRTC1_REGISTER_OFFSET;
-		break;
-	case 2:
-		amdgpu_crtc->crtc_offset = CRTC2_REGISTER_OFFSET;
-		break;
-	case 3:
-		amdgpu_crtc->crtc_offset = CRTC3_REGISTER_OFFSET;
-		break;
-	case 4:
-		amdgpu_crtc->crtc_offset = CRTC4_REGISTER_OFFSET;
-		break;
-	case 5:
-		amdgpu_crtc->crtc_offset = CRTC5_REGISTER_OFFSET;
-		break;
-	}
-
-	amdgpu_crtc->pll_id = ATOM_PPLL_INVALID;
-	amdgpu_crtc->adjusted_clock = 0;
-	amdgpu_crtc->encoder = NULL;
-	amdgpu_crtc->connector = NULL;
-	drm_crtc_helper_add(&amdgpu_crtc->base, &dce_v11_0_crtc_helper_funcs);
-	drm_plane_helper_add(amdgpu_crtc->base.primary, &dce_v11_0_drm_primary_plane_helper_funcs);
-
-	return 0;
-}
-
-static int dce_v11_0_early_init(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-
-	adev->audio_endpt_rreg = &dce_v11_0_audio_endpt_rreg;
-	adev->audio_endpt_wreg = &dce_v11_0_audio_endpt_wreg;
-
-	dce_v11_0_set_display_funcs(adev);
-
-	adev->mode_info.num_crtc = dce_v11_0_get_num_crtc(adev);
-
-	switch (adev->asic_type) {
-	case CHIP_CARRIZO:
-		adev->mode_info.num_hpd = 6;
-		adev->mode_info.num_dig = 9;
-		break;
-	case CHIP_STONEY:
-		adev->mode_info.num_hpd = 6;
-		adev->mode_info.num_dig = 9;
-		break;
-	case CHIP_POLARIS10:
-	case CHIP_VEGAM:
-		adev->mode_info.num_hpd = 6;
-		adev->mode_info.num_dig = 6;
-		break;
-	case CHIP_POLARIS11:
-	case CHIP_POLARIS12:
-		adev->mode_info.num_hpd = 5;
-		adev->mode_info.num_dig = 5;
-		break;
-	default:
-		/* FIXME: not supported yet */
-		return -EINVAL;
-	}
-
-	dce_v11_0_set_irq_funcs(adev);
-
-	return 0;
-}
-
-static int dce_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
-{
-	int r, i;
-	struct amdgpu_device *adev = ip_block->adev;
-
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, i + 1, &adev->crtc_irq);
-		if (r)
-			return r;
-	}
-
-	for (i = VISLANDS30_IV_SRCID_D1_GRPH_PFLIP; i < 20; i += 2) {
-		r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, i, &adev->pageflip_irq);
-		if (r)
-			return r;
-	}
-
-	/* HPD hotplug */
-	r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, VISLANDS30_IV_SRCID_HOTPLUG_DETECT_A, &adev->hpd_irq);
-	if (r)
-		return r;
-
-	adev_to_drm(adev)->mode_config.funcs = &amdgpu_mode_funcs;
-
-	adev_to_drm(adev)->mode_config.async_page_flip = true;
-
-	adev_to_drm(adev)->mode_config.max_width = 16384;
-	adev_to_drm(adev)->mode_config.max_height = 16384;
-
-	adev_to_drm(adev)->mode_config.preferred_depth = 24;
-	adev_to_drm(adev)->mode_config.prefer_shadow = 1;
-
-	adev_to_drm(adev)->mode_config.fb_modifiers_not_supported = true;
-
-	r = amdgpu_display_modeset_create_props(adev);
-	if (r)
-		return r;
-
-	adev_to_drm(adev)->mode_config.max_width = 16384;
-	adev_to_drm(adev)->mode_config.max_height = 16384;
-
-
-	/* allocate crtcs */
-	for (i = 0; i < adev->mode_info.num_crtc; i++) {
-		r = dce_v11_0_crtc_init(adev, i);
-		if (r)
-			return r;
-	}
-
-	if (amdgpu_atombios_get_connector_info_from_object_table(adev))
-		amdgpu_display_print_display_setup(adev_to_drm(adev));
-	else
-		return -EINVAL;
-
-	/* setup afmt */
-	r = dce_v11_0_afmt_init(adev);
-	if (r)
-		return r;
-
-	r = dce_v11_0_audio_init(adev);
-	if (r)
-		return r;
-
-	/* Disable vblank IRQs aggressively for power-saving */
-	/* XXX: can this be enabled for DC? */
-	adev_to_drm(adev)->vblank_disable_immediate = true;
-
-	r = drm_vblank_init(adev_to_drm(adev), adev->mode_info.num_crtc);
-	if (r)
-		return r;
-
-	INIT_DELAYED_WORK(&adev->hotplug_work,
-		  amdgpu_display_hotplug_work_func);
-
-	drm_kms_helper_poll_init(adev_to_drm(adev));
-
-	adev->mode_info.mode_config_initialized = true;
-	return 0;
-}
-
-static int dce_v11_0_sw_fini(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-
-	drm_edid_free(adev->mode_info.bios_hardcoded_edid);
-
-	drm_kms_helper_poll_fini(adev_to_drm(adev));
-
-	dce_v11_0_audio_fini(adev);
-
-	dce_v11_0_afmt_fini(adev);
-
-	drm_mode_config_cleanup(adev_to_drm(adev));
-	adev->mode_info.mode_config_initialized = false;
-
-	return 0;
-}
-
-static int dce_v11_0_hw_init(struct amdgpu_ip_block *ip_block)
-{
-	int i;
-	struct amdgpu_device *adev = ip_block->adev;
-
-	dce_v11_0_init_golden_registers(adev);
-
-	/* disable vga render */
-	dce_v11_0_set_vga_render_state(adev, false);
-	/* init dig PHYs, disp eng pll */
-	amdgpu_atombios_crtc_powergate_init(adev);
-	amdgpu_atombios_encoder_init_dig(adev);
-	if ((adev->asic_type == CHIP_POLARIS10) ||
-	    (adev->asic_type == CHIP_POLARIS11) ||
-	    (adev->asic_type == CHIP_POLARIS12) ||
-	    (adev->asic_type == CHIP_VEGAM)) {
-		amdgpu_atombios_crtc_set_dce_clock(adev, adev->clock.default_dispclk,
-						   DCE_CLOCK_TYPE_DISPCLK, ATOM_GCK_DFS);
-		amdgpu_atombios_crtc_set_dce_clock(adev, 0,
-						   DCE_CLOCK_TYPE_DPREFCLK, ATOM_GCK_DFS);
-	} else {
-		amdgpu_atombios_crtc_set_disp_eng_pll(adev, adev->clock.default_dispclk);
-	}
-
-	/* initialize hpd */
-	dce_v11_0_hpd_init(adev);
-
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++) {
-		dce_v11_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-	}
-
-	dce_v11_0_pageflip_interrupt_init(adev);
-
-	return 0;
-}
-
-static int dce_v11_0_hw_fini(struct amdgpu_ip_block *ip_block)
-{
-	int i;
-	struct amdgpu_device *adev = ip_block->adev;
-
-	dce_v11_0_hpd_fini(adev);
-
-	for (i = 0; i < adev->mode_info.audio.num_pins; i++) {
-		dce_v11_0_audio_enable(adev, &adev->mode_info.audio.pin[i], false);
-	}
-
-	dce_v11_0_pageflip_interrupt_fini(adev);
-
-	flush_delayed_work(&adev->hotplug_work);
-
-	return 0;
-}
-
-static int dce_v11_0_suspend(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int r;
-
-	r = amdgpu_display_suspend_helper(adev);
-	if (r)
-		return r;
-
-	adev->mode_info.bl_level =
-		amdgpu_atombios_encoder_get_backlight_level_from_reg(adev);
-
-	return dce_v11_0_hw_fini(ip_block);
-}
-
-static int dce_v11_0_resume(struct amdgpu_ip_block *ip_block)
-{
-	struct amdgpu_device *adev = ip_block->adev;
-	int ret;
-
-	amdgpu_atombios_encoder_set_backlight_level_to_reg(adev,
-							   adev->mode_info.bl_level);
-
-	ret = dce_v11_0_hw_init(ip_block);
-
-	/* turn on the BL */
-	if (adev->mode_info.bl_encoder) {
-		u8 bl_level = amdgpu_display_backlight_get_level(adev,
-								  adev->mode_info.bl_encoder);
-		amdgpu_display_backlight_set_level(adev, adev->mode_info.bl_encoder,
-						    bl_level);
-	}
-	if (ret)
-		return ret;
-
-	return amdgpu_display_resume_helper(adev);
-}
-
-static bool dce_v11_0_is_idle(struct amdgpu_ip_block *ip_block)
-{
-	return true;
-}
-
-static int dce_v11_0_soft_reset(struct amdgpu_ip_block *ip_block)
-{
-	u32 srbm_soft_reset = 0, tmp;
-	struct amdgpu_device *adev = ip_block->adev;
-
-	if (dce_v11_0_is_display_hung(adev))
-		srbm_soft_reset |= SRBM_SOFT_RESET__SOFT_RESET_DC_MASK;
-
-	if (srbm_soft_reset) {
-		tmp = RREG32(mmSRBM_SOFT_RESET);
-		tmp |= srbm_soft_reset;
-		dev_info(adev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp);
-		WREG32(mmSRBM_SOFT_RESET, tmp);
-		tmp = RREG32(mmSRBM_SOFT_RESET);
-
-		udelay(50);
-
-		tmp &= ~srbm_soft_reset;
-		WREG32(mmSRBM_SOFT_RESET, tmp);
-		tmp = RREG32(mmSRBM_SOFT_RESET);
-
-		/* Wait a little for things to settle down */
-		udelay(50);
-	}
-	return 0;
-}
-
-static void dce_v11_0_set_crtc_vblank_interrupt_state(struct amdgpu_device *adev,
-						     int crtc,
-						     enum amdgpu_interrupt_state state)
-{
-	u32 lb_interrupt_mask;
-
-	if (crtc >= adev->mode_info.num_crtc) {
-		DRM_DEBUG("invalid crtc %d\n", crtc);
-		return;
-	}
-
-	switch (state) {
-	case AMDGPU_IRQ_STATE_DISABLE:
-		lb_interrupt_mask = RREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc]);
-		lb_interrupt_mask = REG_SET_FIELD(lb_interrupt_mask, LB_INTERRUPT_MASK,
-						  VBLANK_INTERRUPT_MASK, 0);
-		WREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc], lb_interrupt_mask);
-		break;
-	case AMDGPU_IRQ_STATE_ENABLE:
-		lb_interrupt_mask = RREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc]);
-		lb_interrupt_mask = REG_SET_FIELD(lb_interrupt_mask, LB_INTERRUPT_MASK,
-						  VBLANK_INTERRUPT_MASK, 1);
-		WREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc], lb_interrupt_mask);
-		break;
-	default:
-		break;
-	}
-}
-
-static void dce_v11_0_set_crtc_vline_interrupt_state(struct amdgpu_device *adev,
-						    int crtc,
-						    enum amdgpu_interrupt_state state)
-{
-	u32 lb_interrupt_mask;
-
-	if (crtc >= adev->mode_info.num_crtc) {
-		DRM_DEBUG("invalid crtc %d\n", crtc);
-		return;
-	}
-
-	switch (state) {
-	case AMDGPU_IRQ_STATE_DISABLE:
-		lb_interrupt_mask = RREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc]);
-		lb_interrupt_mask = REG_SET_FIELD(lb_interrupt_mask, LB_INTERRUPT_MASK,
-						  VLINE_INTERRUPT_MASK, 0);
-		WREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc], lb_interrupt_mask);
-		break;
-	case AMDGPU_IRQ_STATE_ENABLE:
-		lb_interrupt_mask = RREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc]);
-		lb_interrupt_mask = REG_SET_FIELD(lb_interrupt_mask, LB_INTERRUPT_MASK,
-						  VLINE_INTERRUPT_MASK, 1);
-		WREG32(mmLB_INTERRUPT_MASK + crtc_offsets[crtc], lb_interrupt_mask);
-		break;
-	default:
-		break;
-	}
-}
-
-static int dce_v11_0_set_hpd_irq_state(struct amdgpu_device *adev,
-					struct amdgpu_irq_src *source,
-					unsigned hpd,
-					enum amdgpu_interrupt_state state)
-{
-	u32 tmp;
-
-	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hpd %d\n", hpd);
-		return 0;
-	}
-
-	switch (state) {
-	case AMDGPU_IRQ_STATE_DISABLE:
-		tmp = RREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd]);
-		tmp = REG_SET_FIELD(tmp, DC_HPD_INT_CONTROL, DC_HPD_INT_EN, 0);
-		WREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd], tmp);
-		break;
-	case AMDGPU_IRQ_STATE_ENABLE:
-		tmp = RREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd]);
-		tmp = REG_SET_FIELD(tmp, DC_HPD_INT_CONTROL, DC_HPD_INT_EN, 1);
-		WREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd], tmp);
-		break;
-	default:
-		break;
-	}
-
-	return 0;
-}
-
-static int dce_v11_0_set_crtc_irq_state(struct amdgpu_device *adev,
-					struct amdgpu_irq_src *source,
-					unsigned type,
-					enum amdgpu_interrupt_state state)
-{
-	switch (type) {
-	case AMDGPU_CRTC_IRQ_VBLANK1:
-		dce_v11_0_set_crtc_vblank_interrupt_state(adev, 0, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VBLANK2:
-		dce_v11_0_set_crtc_vblank_interrupt_state(adev, 1, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VBLANK3:
-		dce_v11_0_set_crtc_vblank_interrupt_state(adev, 2, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VBLANK4:
-		dce_v11_0_set_crtc_vblank_interrupt_state(adev, 3, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VBLANK5:
-		dce_v11_0_set_crtc_vblank_interrupt_state(adev, 4, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VBLANK6:
-		dce_v11_0_set_crtc_vblank_interrupt_state(adev, 5, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VLINE1:
-		dce_v11_0_set_crtc_vline_interrupt_state(adev, 0, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VLINE2:
-		dce_v11_0_set_crtc_vline_interrupt_state(adev, 1, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VLINE3:
-		dce_v11_0_set_crtc_vline_interrupt_state(adev, 2, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VLINE4:
-		dce_v11_0_set_crtc_vline_interrupt_state(adev, 3, state);
-		break;
-	case AMDGPU_CRTC_IRQ_VLINE5:
-		dce_v11_0_set_crtc_vline_interrupt_state(adev, 4, state);
-		break;
-	 case AMDGPU_CRTC_IRQ_VLINE6:
-		dce_v11_0_set_crtc_vline_interrupt_state(adev, 5, state);
-		break;
-	default:
-		break;
-	}
-	return 0;
-}
-
-static int dce_v11_0_set_pageflip_irq_state(struct amdgpu_device *adev,
-					    struct amdgpu_irq_src *src,
-					    unsigned type,
-					    enum amdgpu_interrupt_state state)
-{
-	u32 reg;
-
-	if (type >= adev->mode_info.num_crtc) {
-		DRM_ERROR("invalid pageflip crtc %d\n", type);
-		return -EINVAL;
-	}
-
-	reg = RREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type]);
-	if (state == AMDGPU_IRQ_STATE_DISABLE)
-		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
-		       reg & ~GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
-	else
-		WREG32(mmGRPH_INTERRUPT_CONTROL + crtc_offsets[type],
-		       reg | GRPH_INTERRUPT_CONTROL__GRPH_PFLIP_INT_MASK_MASK);
-
-	return 0;
-}
-
-static int dce_v11_0_pageflip_irq(struct amdgpu_device *adev,
-				  struct amdgpu_irq_src *source,
-				  struct amdgpu_iv_entry *entry)
-{
-	unsigned long flags;
-	unsigned crtc_id;
-	struct amdgpu_crtc *amdgpu_crtc;
-	struct amdgpu_flip_work *works;
-
-	crtc_id = (entry->src_id - 8) >> 1;
-	amdgpu_crtc = adev->mode_info.crtcs[crtc_id];
-
-	if (crtc_id >= adev->mode_info.num_crtc) {
-		DRM_ERROR("invalid pageflip crtc %d\n", crtc_id);
-		return -EINVAL;
-	}
-
-	if (RREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id]) &
-	    GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_OCCURRED_MASK)
-		WREG32(mmGRPH_INTERRUPT_STATUS + crtc_offsets[crtc_id],
-		       GRPH_INTERRUPT_STATUS__GRPH_PFLIP_INT_CLEAR_MASK);
-
-	/* IRQ could occur when in initial stage */
-	if(amdgpu_crtc == NULL)
-		return 0;
-
-	spin_lock_irqsave(&adev_to_drm(adev)->event_lock, flags);
-	works = amdgpu_crtc->pflip_works;
-	if (amdgpu_crtc->pflip_status != AMDGPU_FLIP_SUBMITTED){
-		DRM_DEBUG_DRIVER("amdgpu_crtc->pflip_status = %d != "
-						 "AMDGPU_FLIP_SUBMITTED(%d)\n",
-						 amdgpu_crtc->pflip_status,
-						 AMDGPU_FLIP_SUBMITTED);
-		spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
-		return 0;
-	}
-
-	/* page flip completed. clean up */
-	amdgpu_crtc->pflip_status = AMDGPU_FLIP_NONE;
-	amdgpu_crtc->pflip_works = NULL;
-
-	/* wakeup usersapce */
-	if(works->event)
-		drm_crtc_send_vblank_event(&amdgpu_crtc->base, works->event);
-
-	spin_unlock_irqrestore(&adev_to_drm(adev)->event_lock, flags);
-
-	drm_crtc_vblank_put(&amdgpu_crtc->base);
-	schedule_work(&works->unpin_work);
-
-	return 0;
-}
-
-static void dce_v11_0_hpd_int_ack(struct amdgpu_device *adev,
-				  int hpd)
-{
-	u32 tmp;
-
-	if (hpd >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("invalid hpd %d\n", hpd);
-		return;
-	}
-
-	tmp = RREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd]);
-	tmp = REG_SET_FIELD(tmp, DC_HPD_INT_CONTROL, DC_HPD_INT_ACK, 1);
-	WREG32(mmDC_HPD_INT_CONTROL + hpd_offsets[hpd], tmp);
-}
-
-static void dce_v11_0_crtc_vblank_int_ack(struct amdgpu_device *adev,
-					  int crtc)
-{
-	u32 tmp;
-
-	if (crtc < 0 || crtc >= adev->mode_info.num_crtc) {
-		DRM_DEBUG("invalid crtc %d\n", crtc);
-		return;
-	}
-
-	tmp = RREG32(mmLB_VBLANK_STATUS + crtc_offsets[crtc]);
-	tmp = REG_SET_FIELD(tmp, LB_VBLANK_STATUS, VBLANK_ACK, 1);
-	WREG32(mmLB_VBLANK_STATUS + crtc_offsets[crtc], tmp);
-}
-
-static void dce_v11_0_crtc_vline_int_ack(struct amdgpu_device *adev,
-					 int crtc)
-{
-	u32 tmp;
-
-	if (crtc < 0 || crtc >= adev->mode_info.num_crtc) {
-		DRM_DEBUG("invalid crtc %d\n", crtc);
-		return;
-	}
-
-	tmp = RREG32(mmLB_VLINE_STATUS + crtc_offsets[crtc]);
-	tmp = REG_SET_FIELD(tmp, LB_VLINE_STATUS, VLINE_ACK, 1);
-	WREG32(mmLB_VLINE_STATUS + crtc_offsets[crtc], tmp);
-}
-
-static int dce_v11_0_crtc_irq(struct amdgpu_device *adev,
-				struct amdgpu_irq_src *source,
-				struct amdgpu_iv_entry *entry)
-{
-	unsigned crtc = entry->src_id - 1;
-	uint32_t disp_int = RREG32(interrupt_status_offsets[crtc].reg);
-	unsigned int irq_type = amdgpu_display_crtc_idx_to_irq_type(adev,
-								    crtc);
-
-	switch (entry->src_data[0]) {
-	case 0: /* vblank */
-		if (disp_int & interrupt_status_offsets[crtc].vblank)
-			dce_v11_0_crtc_vblank_int_ack(adev, crtc);
-		else
-			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
-
-		if (amdgpu_irq_enabled(adev, source, irq_type)) {
-			drm_handle_vblank(adev_to_drm(adev), crtc);
-		}
-		DRM_DEBUG("IH: D%d vblank\n", crtc + 1);
-
-		break;
-	case 1: /* vline */
-		if (disp_int & interrupt_status_offsets[crtc].vline)
-			dce_v11_0_crtc_vline_int_ack(adev, crtc);
-		else
-			DRM_DEBUG("IH: IH event w/o asserted irq bit?\n");
-
-		DRM_DEBUG("IH: D%d vline\n", crtc + 1);
-
-		break;
-	default:
-		DRM_DEBUG("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data[0]);
-		break;
-	}
-
-	return 0;
-}
-
-static int dce_v11_0_hpd_irq(struct amdgpu_device *adev,
-			     struct amdgpu_irq_src *source,
-			     struct amdgpu_iv_entry *entry)
-{
-	uint32_t disp_int, mask;
-	unsigned hpd;
-
-	if (entry->src_data[0] >= adev->mode_info.num_hpd) {
-		DRM_DEBUG("Unhandled interrupt: %d %d\n", entry->src_id, entry->src_data[0]);
-		return 0;
-	}
-
-	hpd = entry->src_data[0];
-	disp_int = RREG32(interrupt_status_offsets[hpd].reg);
-	mask = interrupt_status_offsets[hpd].hpd;
-
-	if (disp_int & mask) {
-		dce_v11_0_hpd_int_ack(adev, hpd);
-		schedule_delayed_work(&adev->hotplug_work, 0);
-		DRM_DEBUG("IH: HPD%d\n", hpd + 1);
-	}
-
-	return 0;
-}
-
-static int dce_v11_0_set_clockgating_state(struct amdgpu_ip_block *ip_block,
-					  enum amd_clockgating_state state)
-{
-	return 0;
-}
-
-static int dce_v11_0_set_powergating_state(struct amdgpu_ip_block *ip_block,
-					  enum amd_powergating_state state)
-{
-	return 0;
-}
-
-static const struct amd_ip_funcs dce_v11_0_ip_funcs = {
-	.name = "dce_v11_0",
-	.early_init = dce_v11_0_early_init,
-	.sw_init = dce_v11_0_sw_init,
-	.sw_fini = dce_v11_0_sw_fini,
-	.hw_init = dce_v11_0_hw_init,
-	.hw_fini = dce_v11_0_hw_fini,
-	.suspend = dce_v11_0_suspend,
-	.resume = dce_v11_0_resume,
-	.is_idle = dce_v11_0_is_idle,
-	.soft_reset = dce_v11_0_soft_reset,
-	.set_clockgating_state = dce_v11_0_set_clockgating_state,
-	.set_powergating_state = dce_v11_0_set_powergating_state,
-};
-
-static void dce_v11_0_encoder_mode_set(struct drm_encoder *encoder,
-			  struct drm_display_mode *mode,
-			  struct drm_display_mode *adjusted_mode)
-{
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-
-	amdgpu_encoder->pixel_clock = adjusted_mode->clock;
-
-	/* need to call this here rather than in prepare() since we need some crtc info */
-	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
-
-	/* set scaler clears this on some chips */
-	dce_v11_0_set_interleave(encoder->crtc, mode);
-
-	if (amdgpu_atombios_encoder_get_encoder_mode(encoder) == ATOM_ENCODER_MODE_HDMI) {
-		dce_v11_0_afmt_enable(encoder, true);
-		dce_v11_0_afmt_setmode(encoder, adjusted_mode);
-	}
-}
-
-static void dce_v11_0_encoder_prepare(struct drm_encoder *encoder)
-{
-	struct amdgpu_device *adev = drm_to_adev(encoder->dev);
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct drm_connector *connector = amdgpu_get_connector_for_encoder(encoder);
-
-	if ((amdgpu_encoder->active_device &
-	     (ATOM_DEVICE_DFP_SUPPORT | ATOM_DEVICE_LCD_SUPPORT)) ||
-	    (amdgpu_encoder_get_dp_bridge_encoder_id(encoder) !=
-	     ENCODER_OBJECT_ID_NONE)) {
-		struct amdgpu_encoder_atom_dig *dig = amdgpu_encoder->enc_priv;
-		if (dig) {
-			dig->dig_encoder = dce_v11_0_pick_dig_encoder(encoder);
-			if (amdgpu_encoder->active_device & ATOM_DEVICE_DFP_SUPPORT)
-				dig->afmt = adev->mode_info.afmt[dig->dig_encoder];
-		}
-	}
-
-	amdgpu_atombios_scratch_regs_lock(adev, true);
-
-	if (connector) {
-		struct amdgpu_connector *amdgpu_connector = to_amdgpu_connector(connector);
-
-		/* select the clock/data port if it uses a router */
-		if (amdgpu_connector->router.cd_valid)
-			amdgpu_i2c_router_select_cd_port(amdgpu_connector);
-
-		/* turn eDP panel on for mode set */
-		if (connector->connector_type == DRM_MODE_CONNECTOR_eDP)
-			amdgpu_atombios_encoder_set_edp_panel_power(connector,
-							     ATOM_TRANSMITTER_ACTION_POWER_ON);
-	}
-
-	/* this is needed for the pll/ss setup to work correctly in some cases */
-	amdgpu_atombios_encoder_set_crtc_source(encoder);
-	/* set up the FMT blocks */
-	dce_v11_0_program_fmt(encoder);
-}
-
-static void dce_v11_0_encoder_commit(struct drm_encoder *encoder)
-{
-	struct drm_device *dev = encoder->dev;
-	struct amdgpu_device *adev = drm_to_adev(dev);
-
-	/* need to call this here as we need the crtc set up */
-	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
-	amdgpu_atombios_scratch_regs_lock(adev, false);
-}
-
-static void dce_v11_0_encoder_disable(struct drm_encoder *encoder)
-{
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	struct amdgpu_encoder_atom_dig *dig;
-
-	amdgpu_atombios_encoder_dpms(encoder, DRM_MODE_DPMS_OFF);
-
-	if (amdgpu_atombios_encoder_is_digital(encoder)) {
-		if (amdgpu_atombios_encoder_get_encoder_mode(encoder) == ATOM_ENCODER_MODE_HDMI)
-			dce_v11_0_afmt_enable(encoder, false);
-		dig = amdgpu_encoder->enc_priv;
-		dig->dig_encoder = -1;
-	}
-	amdgpu_encoder->active_device = 0;
-}
-
-/* these are handled by the primary encoders */
-static void dce_v11_0_ext_prepare(struct drm_encoder *encoder)
-{
-
-}
-
-static void dce_v11_0_ext_commit(struct drm_encoder *encoder)
-{
-
-}
-
-static void
-dce_v11_0_ext_mode_set(struct drm_encoder *encoder,
-		      struct drm_display_mode *mode,
-		      struct drm_display_mode *adjusted_mode)
-{
-
-}
-
-static void dce_v11_0_ext_disable(struct drm_encoder *encoder)
-{
-
-}
-
-static void
-dce_v11_0_ext_dpms(struct drm_encoder *encoder, int mode)
-{
-
-}
-
-static const struct drm_encoder_helper_funcs dce_v11_0_ext_helper_funcs = {
-	.dpms = dce_v11_0_ext_dpms,
-	.prepare = dce_v11_0_ext_prepare,
-	.mode_set = dce_v11_0_ext_mode_set,
-	.commit = dce_v11_0_ext_commit,
-	.disable = dce_v11_0_ext_disable,
-	/* no detect for TMDS/LVDS yet */
-};
-
-static const struct drm_encoder_helper_funcs dce_v11_0_dig_helper_funcs = {
-	.dpms = amdgpu_atombios_encoder_dpms,
-	.mode_fixup = amdgpu_atombios_encoder_mode_fixup,
-	.prepare = dce_v11_0_encoder_prepare,
-	.mode_set = dce_v11_0_encoder_mode_set,
-	.commit = dce_v11_0_encoder_commit,
-	.disable = dce_v11_0_encoder_disable,
-	.detect = amdgpu_atombios_encoder_dig_detect,
-};
-
-static const struct drm_encoder_helper_funcs dce_v11_0_dac_helper_funcs = {
-	.dpms = amdgpu_atombios_encoder_dpms,
-	.mode_fixup = amdgpu_atombios_encoder_mode_fixup,
-	.prepare = dce_v11_0_encoder_prepare,
-	.mode_set = dce_v11_0_encoder_mode_set,
-	.commit = dce_v11_0_encoder_commit,
-	.detect = amdgpu_atombios_encoder_dac_detect,
-};
-
-static void dce_v11_0_encoder_destroy(struct drm_encoder *encoder)
-{
-	struct amdgpu_encoder *amdgpu_encoder = to_amdgpu_encoder(encoder);
-	if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT))
-		amdgpu_atombios_encoder_fini_backlight(amdgpu_encoder);
-	kfree(amdgpu_encoder->enc_priv);
-	drm_encoder_cleanup(encoder);
-	kfree(amdgpu_encoder);
-}
-
-static const struct drm_encoder_funcs dce_v11_0_encoder_funcs = {
-	.destroy = dce_v11_0_encoder_destroy,
-};
-
-static void dce_v11_0_encoder_add(struct amdgpu_device *adev,
-				 uint32_t encoder_enum,
-				 uint32_t supported_device,
-				 u16 caps)
-{
-	struct drm_device *dev = adev_to_drm(adev);
-	struct drm_encoder *encoder;
-	struct amdgpu_encoder *amdgpu_encoder;
-
-	/* see if we already added it */
-	list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) {
-		amdgpu_encoder = to_amdgpu_encoder(encoder);
-		if (amdgpu_encoder->encoder_enum == encoder_enum) {
-			amdgpu_encoder->devices |= supported_device;
-			return;
-		}
-
-	}
-
-	/* add a new one */
-	amdgpu_encoder = kzalloc(sizeof(struct amdgpu_encoder), GFP_KERNEL);
-	if (!amdgpu_encoder)
-		return;
-
-	encoder = &amdgpu_encoder->base;
-	switch (adev->mode_info.num_crtc) {
-	case 1:
-		encoder->possible_crtcs = 0x1;
-		break;
-	case 2:
-	default:
-		encoder->possible_crtcs = 0x3;
-		break;
-	case 3:
-		encoder->possible_crtcs = 0x7;
-		break;
-	case 4:
-		encoder->possible_crtcs = 0xf;
-		break;
-	case 5:
-		encoder->possible_crtcs = 0x1f;
-		break;
-	case 6:
-		encoder->possible_crtcs = 0x3f;
-		break;
-	}
-
-	amdgpu_encoder->enc_priv = NULL;
-
-	amdgpu_encoder->encoder_enum = encoder_enum;
-	amdgpu_encoder->encoder_id = (encoder_enum & OBJECT_ID_MASK) >> OBJECT_ID_SHIFT;
-	amdgpu_encoder->devices = supported_device;
-	amdgpu_encoder->rmx_type = RMX_OFF;
-	amdgpu_encoder->underscan_type = UNDERSCAN_OFF;
-	amdgpu_encoder->is_ext_encoder = false;
-	amdgpu_encoder->caps = caps;
-
-	switch (amdgpu_encoder->encoder_id) {
-	case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC1:
-	case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DAC2:
-		drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-				 DRM_MODE_ENCODER_DAC, NULL);
-		drm_encoder_helper_add(encoder, &dce_v11_0_dac_helper_funcs);
-		break;
-	case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1:
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1:
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY2:
-	case ENCODER_OBJECT_ID_INTERNAL_UNIPHY3:
-		if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) {
-			amdgpu_encoder->rmx_type = RMX_FULL;
-			drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-					 DRM_MODE_ENCODER_LVDS, NULL);
-			amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_lcd_info(amdgpu_encoder);
-		} else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT)) {
-			drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-					 DRM_MODE_ENCODER_DAC, NULL);
-			amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder);
-		} else {
-			drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-					 DRM_MODE_ENCODER_TMDS, NULL);
-			amdgpu_encoder->enc_priv = amdgpu_atombios_encoder_get_dig_info(amdgpu_encoder);
-		}
-		drm_encoder_helper_add(encoder, &dce_v11_0_dig_helper_funcs);
-		break;
-	case ENCODER_OBJECT_ID_SI170B:
-	case ENCODER_OBJECT_ID_CH7303:
-	case ENCODER_OBJECT_ID_EXTERNAL_SDVOA:
-	case ENCODER_OBJECT_ID_EXTERNAL_SDVOB:
-	case ENCODER_OBJECT_ID_TITFP513:
-	case ENCODER_OBJECT_ID_VT1623:
-	case ENCODER_OBJECT_ID_HDMI_SI1930:
-	case ENCODER_OBJECT_ID_TRAVIS:
-	case ENCODER_OBJECT_ID_NUTMEG:
-		/* these are handled by the primary encoders */
-		amdgpu_encoder->is_ext_encoder = true;
-		if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT))
-			drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-					 DRM_MODE_ENCODER_LVDS, NULL);
-		else if (amdgpu_encoder->devices & (ATOM_DEVICE_CRT_SUPPORT))
-			drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-					 DRM_MODE_ENCODER_DAC, NULL);
-		else
-			drm_encoder_init(dev, encoder, &dce_v11_0_encoder_funcs,
-					 DRM_MODE_ENCODER_TMDS, NULL);
-		drm_encoder_helper_add(encoder, &dce_v11_0_ext_helper_funcs);
-		break;
-	}
-}
-
-static const struct amdgpu_display_funcs dce_v11_0_display_funcs = {
-	.bandwidth_update = &dce_v11_0_bandwidth_update,
-	.vblank_get_counter = &dce_v11_0_vblank_get_counter,
-	.backlight_set_level = &amdgpu_atombios_encoder_set_backlight_level,
-	.backlight_get_level = &amdgpu_atombios_encoder_get_backlight_level,
-	.hpd_sense = &dce_v11_0_hpd_sense,
-	.hpd_set_polarity = &dce_v11_0_hpd_set_polarity,
-	.hpd_get_gpio_reg = &dce_v11_0_hpd_get_gpio_reg,
-	.page_flip = &dce_v11_0_page_flip,
-	.page_flip_get_scanoutpos = &dce_v11_0_crtc_get_scanoutpos,
-	.add_encoder = &dce_v11_0_encoder_add,
-	.add_connector = &amdgpu_connector_add,
-};
-
-static void dce_v11_0_set_display_funcs(struct amdgpu_device *adev)
-{
-	adev->mode_info.funcs = &dce_v11_0_display_funcs;
-}
-
-static const struct amdgpu_irq_src_funcs dce_v11_0_crtc_irq_funcs = {
-	.set = dce_v11_0_set_crtc_irq_state,
-	.process = dce_v11_0_crtc_irq,
-};
-
-static const struct amdgpu_irq_src_funcs dce_v11_0_pageflip_irq_funcs = {
-	.set = dce_v11_0_set_pageflip_irq_state,
-	.process = dce_v11_0_pageflip_irq,
-};
-
-static const struct amdgpu_irq_src_funcs dce_v11_0_hpd_irq_funcs = {
-	.set = dce_v11_0_set_hpd_irq_state,
-	.process = dce_v11_0_hpd_irq,
-};
-
-static void dce_v11_0_set_irq_funcs(struct amdgpu_device *adev)
-{
-	if (adev->mode_info.num_crtc > 0)
-		adev->crtc_irq.num_types = AMDGPU_CRTC_IRQ_VLINE1 + adev->mode_info.num_crtc;
-	else
-		adev->crtc_irq.num_types = 0;
-	adev->crtc_irq.funcs = &dce_v11_0_crtc_irq_funcs;
-
-	adev->pageflip_irq.num_types = adev->mode_info.num_crtc;
-	adev->pageflip_irq.funcs = &dce_v11_0_pageflip_irq_funcs;
-
-	adev->hpd_irq.num_types = adev->mode_info.num_hpd;
-	adev->hpd_irq.funcs = &dce_v11_0_hpd_irq_funcs;
-}
-
-const struct amdgpu_ip_block_version dce_v11_0_ip_block =
-{
-	.type = AMD_IP_BLOCK_TYPE_DCE,
-	.major = 11,
-	.minor = 0,
-	.rev = 0,
-	.funcs = &dce_v11_0_ip_funcs,
-};
-
-const struct amdgpu_ip_block_version dce_v11_2_ip_block =
-{
-	.type = AMD_IP_BLOCK_TYPE_DCE,
-	.major = 11,
-	.minor = 2,
-	.rev = 0,
-	.funcs = &dce_v11_0_ip_funcs,
-};
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.h b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.h
deleted file mode 100644
index 0d878ca3acba..000000000000
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * Copyright 2014 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- */
-
-#ifndef __DCE_V11_0_H__
-#define __DCE_V11_0_H__
-
-extern const struct amdgpu_ip_block_version dce_v11_0_ip_block;
-extern const struct amdgpu_ip_block_version dce_v11_2_ip_block;
-
-void dce_v11_0_disable_dce(struct amdgpu_device *adev);
-
-#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 9b3510e53112..a611a7345125 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -67,7 +67,6 @@
 #include "sdma_v2_4.h"
 #include "sdma_v3_0.h"
 #include "dce_v10_0.h"
-#include "dce_v11_0.h"
 #include "iceland_ih.h"
 #include "tonga_ih.h"
 #include "cz_ih.h"
@@ -2124,8 +2123,6 @@ int vi_set_ip_blocks(struct amdgpu_device *adev)
 		else if (amdgpu_device_has_dc_support(adev))
 			amdgpu_device_ip_block_add(adev, &dm_ip_block);
 #endif
-		else
-			amdgpu_device_ip_block_add(adev, &dce_v11_2_ip_block);
 		amdgpu_device_ip_block_add(adev, &uvd_v6_3_ip_block);
 		amdgpu_device_ip_block_add(adev, &vce_v3_4_ip_block);
 		break;
@@ -2142,8 +2139,6 @@ int vi_set_ip_blocks(struct amdgpu_device *adev)
 		else if (amdgpu_device_has_dc_support(adev))
 			amdgpu_device_ip_block_add(adev, &dm_ip_block);
 #endif
-		else
-			amdgpu_device_ip_block_add(adev, &dce_v11_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &uvd_v6_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &vce_v3_1_ip_block);
 #if defined(CONFIG_DRM_AMD_ACP)
@@ -2163,8 +2158,6 @@ int vi_set_ip_blocks(struct amdgpu_device *adev)
 		else if (amdgpu_device_has_dc_support(adev))
 			amdgpu_device_ip_block_add(adev, &dm_ip_block);
 #endif
-		else
-			amdgpu_device_ip_block_add(adev, &dce_v11_0_ip_block);
 		amdgpu_device_ip_block_add(adev, &uvd_v6_2_ip_block);
 		amdgpu_device_ip_block_add(adev, &vce_v3_4_ip_block);
 #if defined(CONFIG_DRM_AMD_ACP)

From f1fdeb3d07a4c2dcfc8cf0e20c0bddb68e8bc607 Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Tue, 19 Aug 2025 12:51:28 +0800
Subject: [PATCH 1358/1455] drm/amdgpu: Introduce VF critical region check for
 RAS poison injection

The SRIOV guest send requet to host to check whether the poison
injection address is in VF critical region or not via mabox.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c    | 55 +++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h    |  3 ++
 drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h |  5 ++
 drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c       | 14 ++++++
 drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h       |  2 +
 5 files changed, 79 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 13f0cdeb59c4..3328ab63376b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -828,11 +828,14 @@ static void amdgpu_virt_init_ras(struct amdgpu_device *adev)
 {
 	ratelimit_state_init(&adev->virt.ras.ras_error_cnt_rs, 5 * HZ, 1);
 	ratelimit_state_init(&adev->virt.ras.ras_cper_dump_rs, 5 * HZ, 1);
+	ratelimit_state_init(&adev->virt.ras.ras_chk_criti_rs, 5 * HZ, 1);
 
 	ratelimit_set_flags(&adev->virt.ras.ras_error_cnt_rs,
 			    RATELIMIT_MSG_ON_RELEASE);
 	ratelimit_set_flags(&adev->virt.ras.ras_cper_dump_rs,
 			    RATELIMIT_MSG_ON_RELEASE);
+	ratelimit_set_flags(&adev->virt.ras.ras_chk_criti_rs,
+			    RATELIMIT_MSG_ON_RELEASE);
 
 	mutex_init(&adev->virt.ras.ras_telemetry_mutex);
 
@@ -1501,3 +1504,55 @@ void amdgpu_virt_request_bad_pages(struct amdgpu_device *adev)
 	if (virt->ops && virt->ops->req_bad_pages)
 		virt->ops->req_bad_pages(adev);
 }
+
+static int amdgpu_virt_cache_chk_criti_hit(struct amdgpu_device *adev,
+					   struct amdsriov_ras_telemetry *host_telemetry,
+					   bool *hit)
+{
+	struct amd_sriov_ras_chk_criti *tmp = NULL;
+	uint32_t checksum, used_size;
+
+	checksum = host_telemetry->header.checksum;
+	used_size = host_telemetry->header.used_size;
+
+	if (used_size > (AMD_SRIOV_RAS_TELEMETRY_SIZE_KB << 10))
+		return 0;
+
+	tmp = kmemdup(&host_telemetry->body.chk_criti, used_size, GFP_KERNEL);
+	if (!tmp)
+		return -ENOMEM;
+
+	if (checksum != amd_sriov_msg_checksum(tmp, used_size, 0, 0))
+		goto out;
+
+	if (hit)
+		*hit = tmp->hit ? true : false;
+
+out:
+	kfree(tmp);
+
+	return 0;
+}
+
+int amdgpu_virt_check_vf_critical_region(struct amdgpu_device *adev, u64 addr, bool *hit)
+{
+	struct amdgpu_virt *virt = &adev->virt;
+	int r = -EPERM;
+
+	if (!virt->ops || !virt->ops->req_ras_chk_criti)
+		return -EOPNOTSUPP;
+
+	/* Host allows 15 ras telemetry requests per 60 seconds. Afterwhich, the Host
+	 * will ignore incoming guest messages. Ratelimit the guest messages to
+	 * prevent guest self DOS.
+	 */
+	if (__ratelimit(&virt->ras.ras_chk_criti_rs)) {
+		mutex_lock(&virt->ras.ras_telemetry_mutex);
+		if (!virt->ops->req_ras_chk_criti(adev, addr))
+			r = amdgpu_virt_cache_chk_criti_hit(
+				adev, virt->fw_reserve.ras_telemetry, hit);
+		mutex_unlock(&virt->ras.ras_telemetry_mutex);
+	}
+
+	return r;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 3afbdf7b487a..d1172c8e58c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -98,6 +98,7 @@ struct amdgpu_virt_ops {
 	int (*req_ras_err_count)(struct amdgpu_device *adev);
 	int (*req_ras_cper_dump)(struct amdgpu_device *adev, u64 vf_rptr);
 	int (*req_bad_pages)(struct amdgpu_device *adev);
+	int (*req_ras_chk_criti)(struct amdgpu_device *adev, u64 addr);
 };
 
 /*
@@ -252,6 +253,7 @@ struct amdgpu_virt_ras_err_handler_data {
 struct amdgpu_virt_ras {
 	struct ratelimit_state ras_error_cnt_rs;
 	struct ratelimit_state ras_cper_dump_rs;
+	struct ratelimit_state ras_chk_criti_rs;
 	struct mutex ras_telemetry_mutex;
 	uint64_t cper_rptr;
 };
@@ -453,4 +455,5 @@ int amdgpu_virt_ras_telemetry_post_reset(struct amdgpu_device *adev);
 bool amdgpu_virt_ras_telemetry_block_en(struct amdgpu_device *adev,
 					enum amdgpu_ras_block block);
 void amdgpu_virt_request_bad_pages(struct amdgpu_device *adev);
+int amdgpu_virt_check_vf_critical_region(struct amdgpu_device *adev, u64 addr, bool *hit);
 #endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
index 33edad1f9dcd..3a79ed7d8031 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgv_sriovmsg.h
@@ -405,12 +405,17 @@ struct amd_sriov_ras_cper_dump {
 	uint32_t buf[];
 };
 
+struct amd_sriov_ras_chk_criti {
+	uint32_t hit;
+};
+
 struct amdsriov_ras_telemetry {
 	struct amd_sriov_ras_telemetry_header header;
 
 	union {
 		struct amd_sriov_ras_telemetry_error_count error_count;
 		struct amd_sriov_ras_cper_dump cper_dump;
+		struct amd_sriov_ras_chk_criti chk_criti;
 	} body;
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
index 457972aa5632..e5282a5d05d9 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
@@ -202,6 +202,9 @@ send_request:
 	case IDH_REQ_RAS_CPER_DUMP:
 		event = IDH_RAS_CPER_DUMP_READY;
 		break;
+	case IDH_REQ_RAS_CHK_CRITI:
+		event = IDH_REQ_RAS_CHK_CRITI_READY;
+		break;
 	default:
 		break;
 	}
@@ -556,6 +559,16 @@ static int xgpu_nv_req_ras_bad_pages(struct amdgpu_device *adev)
 	return xgpu_nv_send_access_requests(adev, IDH_REQ_RAS_BAD_PAGES);
 }
 
+static int xgpu_nv_check_vf_critical_region(struct amdgpu_device *adev, u64 addr)
+{
+	uint32_t addr_hi, addr_lo;
+
+	addr_hi = (uint32_t)(addr >> 32);
+	addr_lo = (uint32_t)(addr & 0xFFFFFFFF);
+	return xgpu_nv_send_access_requests_with_param(
+		adev, IDH_REQ_RAS_CHK_CRITI, addr_hi, addr_lo, 0);
+}
+
 const struct amdgpu_virt_ops xgpu_nv_virt_ops = {
 	.req_full_gpu	= xgpu_nv_request_full_gpu_access,
 	.rel_full_gpu	= xgpu_nv_release_full_gpu_access,
@@ -569,4 +582,5 @@ const struct amdgpu_virt_ops xgpu_nv_virt_ops = {
 	.req_ras_err_count = xgpu_nv_req_ras_err_count,
 	.req_ras_cper_dump = xgpu_nv_req_ras_cper_dump,
 	.req_bad_pages = xgpu_nv_req_ras_bad_pages,
+	.req_ras_chk_criti = xgpu_nv_check_vf_critical_region
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h
index 5808689562cc..c1083e5e41e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.h
@@ -43,6 +43,7 @@ enum idh_request {
 	IDH_REQ_RAS_ERROR_COUNT = 203,
 	IDH_REQ_RAS_CPER_DUMP = 204,
 	IDH_REQ_RAS_BAD_PAGES = 205,
+	IDH_REQ_RAS_CHK_CRITI = 206
 };
 
 enum idh_event {
@@ -62,6 +63,7 @@ enum idh_event {
 	IDH_RAS_BAD_PAGES_READY = 15,
 	IDH_RAS_BAD_PAGES_NOTIFICATION = 16,
 	IDH_UNRECOV_ERR_NOTIFICATION = 17,
+	IDH_REQ_RAS_CHK_CRITI_READY = 18,
 
 	IDH_TEXT_MESSAGE = 255,
 };

From 4bfa8609934dbf39bbe6e75b4f971469384b50b1 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 17 Sep 2025 12:42:09 -0400
Subject: [PATCH 1359/1455] drm/amdkfd: add proper handling for S0ix

When in S0i3, the GFX state is retained, so all we need to do
is stop the runlist so GFX can enter gfxoff.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Tested-by: David Perry <david.perry@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c | 16 +++++++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 12 ++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_device.c    | 36 ++++++++++++++++++++++
 3 files changed, 60 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index fbe7616555c8..a2879d2b7c8e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -250,16 +250,24 @@ void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
 
 void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc)
 {
-	if (adev->kfd.dev)
-		kgd2kfd_suspend(adev->kfd.dev, suspend_proc);
+	if (adev->kfd.dev) {
+		if (adev->in_s0ix)
+			kgd2kfd_stop_sched_all_nodes(adev->kfd.dev);
+		else
+			kgd2kfd_suspend(adev->kfd.dev, suspend_proc);
+	}
 }
 
 int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc)
 {
 	int r = 0;
 
-	if (adev->kfd.dev)
-		r = kgd2kfd_resume(adev->kfd.dev, resume_proc);
+	if (adev->kfd.dev) {
+		if (adev->in_s0ix)
+			r = kgd2kfd_start_sched_all_nodes(adev->kfd.dev);
+		else
+			r = kgd2kfd_resume(adev->kfd.dev, resume_proc);
+	}
 
 	return r;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 127927b16ee2..9e120c934cc1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -428,7 +428,9 @@ void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask);
 int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd);
 void kgd2kfd_unlock_kfd(struct kfd_dev *kfd);
 int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id);
+int kgd2kfd_start_sched_all_nodes(struct kfd_dev *kfd);
 int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id);
+int kgd2kfd_stop_sched_all_nodes(struct kfd_dev *kfd);
 bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id);
 bool kgd2kfd_vmfault_fast_path(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry,
 			       bool retry_fault);
@@ -518,11 +520,21 @@ static inline int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id)
 	return 0;
 }
 
+static inline int kgd2kfd_start_sched_all_nodes(struct kfd_dev *kfd)
+{
+	return 0;
+}
+
 static inline int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id)
 {
 	return 0;
 }
 
+static inline int kgd2kfd_stop_sched_all_nodes(struct kfd_dev *kfd)
+{
+	return 0;
+}
+
 static inline bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id)
 {
 	return false;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 7e749f9b6d69..349c351e242b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -1550,6 +1550,25 @@ int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id)
 	return ret;
 }
 
+int kgd2kfd_start_sched_all_nodes(struct kfd_dev *kfd)
+{
+	struct kfd_node *node;
+	int i, r;
+
+	if (!kfd->init_complete)
+		return 0;
+
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kfd->nodes[i];
+		r = node->dqm->ops.unhalt(node->dqm);
+		if (r) {
+			dev_err(kfd_device, "Error in starting scheduler\n");
+			return r;
+		}
+	}
+	return 0;
+}
+
 int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id)
 {
 	struct kfd_node *node;
@@ -1567,6 +1586,23 @@ int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id)
 	return node->dqm->ops.halt(node->dqm);
 }
 
+int kgd2kfd_stop_sched_all_nodes(struct kfd_dev *kfd)
+{
+	struct kfd_node *node;
+	int i, r;
+
+	if (!kfd->init_complete)
+		return 0;
+
+	for (i = 0; i < kfd->num_nodes; i++) {
+		node = kfd->nodes[i];
+		r = node->dqm->ops.halt(node->dqm);
+		if (r)
+			return r;
+	}
+	return 0;
+}
+
 bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id)
 {
 	struct kfd_node *node;

From 1ed511fb760848e3ccdcb936f4257c094e95a43c Mon Sep 17 00:00:00 2001
From: Xiang Liu <xiang.liu@amd.com>
Date: Tue, 19 Aug 2025 13:06:24 +0800
Subject: [PATCH 1360/1455] drm/amdgpu: Check VF critical region before RAS
 poison injection

Check VF critical region before RAS poison injection to ensure that the
poison injection will not hit the VF critical region.

Signed-off-by: Xiang Liu <xiang.liu@amd.com>
Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 663b96583bfd..e0ee21150860 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -219,10 +219,17 @@ static int amdgpu_check_address_validity(struct amdgpu_device *adev,
 	struct amdgpu_vram_block_info blk_info;
 	uint64_t page_pfns[32] = {0};
 	int i, ret, count;
+	bool hit = false;
 
 	if (amdgpu_ip_version(adev, UMC_HWIP, 0) < IP_VERSION(12, 0, 0))
 		return 0;
 
+	if (amdgpu_sriov_vf(adev)) {
+		if (amdgpu_virt_check_vf_critical_region(adev, address, &hit))
+			return -EPERM;
+		return hit ? -EACCES : 0;
+	}
+
 	if ((address >= adev->gmc.mc_vram_size) ||
 	    (address >= RAS_UMC_INJECT_ADDR_LIMIT))
 		return -EFAULT;

From f05c03ffc786968ec728422728f5226522886d9c Mon Sep 17 00:00:00 2001
From: "Joe.Wang" <joe.wang@amd.com>
Date: Wed, 17 Sep 2025 14:58:49 +0800
Subject: [PATCH 1361/1455] drm/amdgpu: Fix PRT flag for gfx12

AMDGPU_PTE_PRT_GFX12 flag is missed during pageTable rework, add it back.

Fixes: 6716a823d18d ("drm/amdgpu: rework how PTE flags are generated v3")
Signed-off-by: Joe Wang <joe.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
index 7cc16af1868b..404cc8c2ff2c 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c
@@ -521,6 +521,7 @@ static void gmc_v12_0_get_vm_pte(struct amdgpu_device *adev,
 		*flags &= ~AMDGPU_PTE_NOALLOC;
 
 	if (vm_flags & AMDGPU_VM_PAGE_PRT) {
+		*flags |= AMDGPU_PTE_PRT_GFX12;
 		*flags |= AMDGPU_PTE_SNOOPED;
 		*flags |= AMDGPU_PTE_SYSTEM;
 		*flags |= AMDGPU_PTE_IS_PTE;

From 846de1384a6ab5576601f206b9a94786276ec204 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 17 Sep 2025 12:42:10 -0400
Subject: [PATCH 1362/1455] drm/amdgpu/userq: Optimize S0ix handling

In S0i3, GFX state is retained, so it's preferrable to
preempt queues rather than unmapping them as the overhead
is lower.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Tested-by: David Perry <david.perry@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 3dfe7f5f154b..22a5827b0772 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -977,7 +977,10 @@ int amdgpu_userq_suspend(struct amdgpu_device *adev)
 		cancel_delayed_work_sync(&uqm->resume_work);
 		mutex_lock(&uqm->userq_mutex);
 		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
-			r = amdgpu_userq_unmap_helper(uqm, queue);
+			if (adev->in_s0ix)
+				r = amdgpu_userq_preempt_helper(uqm, queue);
+			else
+				r = amdgpu_userq_unmap_helper(uqm, queue);
 			if (r)
 				ret = r;
 		}
@@ -1002,7 +1005,10 @@ int amdgpu_userq_resume(struct amdgpu_device *adev)
 	list_for_each_entry_safe(uqm, tmp, &adev->userq_mgr_list, list) {
 		mutex_lock(&uqm->userq_mutex);
 		idr_for_each_entry(&uqm->userq_idr, queue, queue_id) {
-			r = amdgpu_userq_map_helper(uqm, queue);
+			if (adev->in_s0ix)
+				r = amdgpu_userq_restore_helper(uqm, queue);
+			else
+				r = amdgpu_userq_map_helper(uqm, queue);
 			if (r)
 				ret = r;
 		}

From f8b367e6fa1716cab7cc232b9e3dff29187fc99d Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Wed, 17 Sep 2025 12:42:11 -0400
Subject: [PATCH 1363/1455] drm/amdgpu: suspend KFD and KGD user queues for
 S0ix

We need to make sure the user queues are preempted so
GFX can enter gfxoff.

Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
Tested-by: David Perry <david.perry@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 24 +++++++++-------------
 1 file changed, 10 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 0fdfde3dcb9f..a77000c2e0bb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -5200,7 +5200,7 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 	adev->in_suspend = true;
 
 	if (amdgpu_sriov_vf(adev)) {
-		if (!adev->in_s0ix && !adev->in_runpm)
+		if (!adev->in_runpm)
 			amdgpu_amdkfd_suspend_process(adev);
 		amdgpu_virt_fini_data_exchange(adev);
 		r = amdgpu_virt_request_full_gpu(adev, false);
@@ -5220,10 +5220,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients)
 
 	amdgpu_device_ip_suspend_phase1(adev);
 
-	if (!adev->in_s0ix) {
-		amdgpu_amdkfd_suspend(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
-		amdgpu_userq_suspend(adev);
-	}
+	amdgpu_amdkfd_suspend(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
+	amdgpu_userq_suspend(adev);
 
 	r = amdgpu_device_evict_resources(adev);
 	if (r)
@@ -5318,15 +5316,13 @@ int amdgpu_device_resume(struct drm_device *dev, bool notify_clients)
 		goto exit;
 	}
 
-	if (!adev->in_s0ix) {
-		r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
-		if (r)
-			goto exit;
+	r = amdgpu_amdkfd_resume(adev, !amdgpu_sriov_vf(adev) && !adev->in_runpm);
+	if (r)
+		goto exit;
 
-		r = amdgpu_userq_resume(adev);
-		if (r)
-			goto exit;
-	}
+	r = amdgpu_userq_resume(adev);
+	if (r)
+		goto exit;
 
 	r = amdgpu_device_ip_late_init(adev);
 	if (r)
@@ -5339,7 +5335,7 @@ exit:
 		amdgpu_virt_init_data_exchange(adev);
 		amdgpu_virt_release_full_gpu(adev, true);
 
-		if (!adev->in_s0ix && !r && !adev->in_runpm)
+		if (!r && !adev->in_runpm)
 			r = amdgpu_amdkfd_resume_process(adev);
 	}
 

From 0aa09d8a6cab0f7d284e61eff28ac3be4d324c20 Mon Sep 17 00:00:00 2001
From: Sunil Khatri <sunil.khatri@amd.com>
Date: Thu, 18 Sep 2025 09:33:51 +0530
Subject: [PATCH 1364/1455] drm/amdgpu: add missing comment for the new
 argument
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In function 'amdgpu_vm_lock_done_list' update the comment
for the new argument 'vm'.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202509180211.UAqME0zj-lkp@intel.com/
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index d0c95fb0ef81..1ccd348bf48b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -486,6 +486,7 @@ int amdgpu_vm_lock_pd(struct amdgpu_vm *vm, struct drm_exec *exec,
 
 /**
  * amdgpu_vm_lock_done_list - lock all BOs on the done list
+ * @vm: vm providing the BOs
  * @exec: drm execution context
  * @num_fences: number of extra fences to reserve
  *

From 8d5b7009aabc27e626e4167fedf1e1c1c3d6b143 Mon Sep 17 00:00:00 2001
From: Alexander Usyskin <alexander.usyskin@intel.com>
Date: Fri, 5 Sep 2025 21:19:45 +0530
Subject: [PATCH 1365/1455] mei: bus: add mei_cldev_mtu interface

Add a new helper function that allows MEI client drivers
to query the maximum transmission unit (MTU) for a connected
MEI client.

This is useful for clients that need to transmit large payloads,
such as firmware blobs, allowing them to determine the maximum
message size that can be safely sent before starting transmission and
size of the buffer to allocate when receiving data.

Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20250905154953.3974335-2-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/misc/mei/bus.c     | 13 +++++++++++++
 include/linux/mei_cl_bus.h |  1 +
 2 files changed, 14 insertions(+)

diff --git a/drivers/misc/mei/bus.c b/drivers/misc/mei/bus.c
index 5cc3ad07d5be..09aae8f9d225 100644
--- a/drivers/misc/mei/bus.c
+++ b/drivers/misc/mei/bus.c
@@ -614,6 +614,19 @@ u8 mei_cldev_ver(const struct mei_cl_device *cldev)
 }
 EXPORT_SYMBOL_GPL(mei_cldev_ver);
 
+/**
+ * mei_cldev_mtu - max message that client can send and receive
+ *
+ * @cldev: mei client device
+ *
+ * Return: mtu or 0 if client is not connected
+ */
+size_t mei_cldev_mtu(const struct mei_cl_device *cldev)
+{
+	return mei_cl_mtu(cldev->cl);
+}
+EXPORT_SYMBOL_GPL(mei_cldev_mtu);
+
 /**
  * mei_cldev_enabled - check whether the device is enabled
  *
diff --git a/include/linux/mei_cl_bus.h b/include/linux/mei_cl_bus.h
index 725fd7727422..a82755e1fc40 100644
--- a/include/linux/mei_cl_bus.h
+++ b/include/linux/mei_cl_bus.h
@@ -113,6 +113,7 @@ int mei_cldev_register_notif_cb(struct mei_cl_device *cldev,
 				mei_cldev_cb_t notif_cb);
 
 u8 mei_cldev_ver(const struct mei_cl_device *cldev);
+size_t mei_cldev_mtu(const struct mei_cl_device *cldev);
 
 void *mei_cldev_get_drvdata(const struct mei_cl_device *cldev);
 void mei_cldev_set_drvdata(struct mei_cl_device *cldev, void *data);

From 741eeabb7c78c555c4c8e39df91b2b8e8d6f5ec6 Mon Sep 17 00:00:00 2001
From: Alexander Usyskin <alexander.usyskin@intel.com>
Date: Fri, 5 Sep 2025 21:19:46 +0530
Subject: [PATCH 1366/1455] mei: late_bind: add late binding component driver

Introduce a new MEI client driver to support Late Binding firmware
upload/update for Intel discrete graphics platforms.

Late Binding is a runtime firmware upload/update mechanism that allows
payloads, such as fan control and voltage regulator, to be securely
delivered and applied without requiring SPI flash updates or
system reboots. This driver enables the Xe graphics driver and other
user-space tools to push such firmware blobs to the authentication
firmware via the MEI interface.

The driver handles authentication, versioning, and communication
with the authentication firmware, which in turn coordinates with
the PUnit/PCODE to apply the payload.

This is a foundational component for enabling dynamic, secure,
and re-entrant configuration updates on platforms like Battlemage.

Cc: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Anshuman Gupta <anshuman.gupta@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20250905154953.3974335-3-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/misc/mei/Kconfig                   |  13 +
 drivers/misc/mei/Makefile                  |   1 +
 drivers/misc/mei/mei_lb.c                  | 312 +++++++++++++++++++++
 include/drm/intel/i915_component.h         |   1 +
 include/drm/intel/intel_lb_mei_interface.h |  70 +++++
 5 files changed, 397 insertions(+)
 create mode 100644 drivers/misc/mei/mei_lb.c
 create mode 100644 include/drm/intel/intel_lb_mei_interface.h

diff --git a/drivers/misc/mei/Kconfig b/drivers/misc/mei/Kconfig
index 7575fee96cc6..f8b04e49e4ba 100644
--- a/drivers/misc/mei/Kconfig
+++ b/drivers/misc/mei/Kconfig
@@ -81,6 +81,19 @@ config INTEL_MEI_VSC
 	  This driver can also be built as a module. If so, the module
 	  will be called mei-vsc.
 
+config INTEL_MEI_LB
+	tristate "Intel Late Binding (LB) support on ME Interface"
+	depends on INTEL_MEI_ME
+	depends on DRM_XE
+	help
+	  Enable support for Intel Late Binding (LB) via the MEI interface.
+
+	  Late Binding is a method for applying firmware updates at runtime,
+	  allowing the Intel Xe driver to load firmware payloads such as
+	  fan controller or voltage regulator. These firmware updates are
+	  authenticated and versioned, and do not require firmware flashing
+	  or system reboot.
+
 source "drivers/misc/mei/hdcp/Kconfig"
 source "drivers/misc/mei/pxp/Kconfig"
 source "drivers/misc/mei/gsc_proxy/Kconfig"
diff --git a/drivers/misc/mei/Makefile b/drivers/misc/mei/Makefile
index 6f9fdbf1a495..a203ed766b33 100644
--- a/drivers/misc/mei/Makefile
+++ b/drivers/misc/mei/Makefile
@@ -31,6 +31,7 @@ CFLAGS_mei-trace.o = -I$(src)
 obj-$(CONFIG_INTEL_MEI_HDCP) += hdcp/
 obj-$(CONFIG_INTEL_MEI_PXP) += pxp/
 obj-$(CONFIG_INTEL_MEI_GSC_PROXY) += gsc_proxy/
+obj-$(CONFIG_INTEL_MEI_LB) += mei_lb.o
 
 obj-$(CONFIG_INTEL_MEI_VSC_HW) += mei-vsc-hw.o
 mei-vsc-hw-y := vsc-tp.o
diff --git a/drivers/misc/mei/mei_lb.c b/drivers/misc/mei/mei_lb.c
new file mode 100644
index 000000000000..77686b108d3c
--- /dev/null
+++ b/drivers/misc/mei/mei_lb.c
@@ -0,0 +1,312 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2025 Intel Corporation
+ */
+
+#include <linux/component.h>
+#include <linux/mei_cl_bus.h>
+#include <linux/module.h>
+#include <linux/overflow.h>
+#include <linux/pci.h>
+#include <linux/slab.h>
+#include <linux/uuid.h>
+
+#include <drm/intel/i915_component.h>
+#include <drm/intel/intel_lb_mei_interface.h>
+
+#include "mkhi.h"
+
+/**
+ * DOC: Late Binding Firmware Update/Upload
+ *
+ * Late Binding is a firmware update/upload mechanism that allows configuration
+ * payloads to be securely delivered and applied at runtime, rather than
+ * being embedded in the system firmware image (e.g., IFWI or SPI flash).
+ *
+ * This mechanism is used to update device-level configuration such as:
+ * - Fan controller
+ * - Voltage regulator (VR)
+ *
+ * Key Characteristics:
+ * ---------------------
+ * - Runtime Delivery:
+ *   Firmware blobs are loaded by the host driver (e.g., Xe KMD)
+ *   after the GPU or SoC has booted.
+ *
+ * - Secure and Authenticated:
+ *   All payloads are signed and verified by the authentication firmware.
+ *
+ * - No Firmware Flashing Required:
+ *   Updates are applied in volatile memory and do not require SPI flash
+ *   modification or system reboot.
+ *
+ * - Re-entrant:
+ *   Multiple updates of the same or different types can be applied
+ *   sequentially within a single boot session.
+ *
+ * - Version Controlled:
+ *   Each payload includes version and security version number (SVN)
+ *   metadata to support anti-rollback enforcement.
+ *
+ * Upload Flow:
+ * ------------
+ * 1. Host driver (KMD or user-space tool) loads the late binding firmware.
+ * 2. Firmware is passed to the MEI interface and forwarded to
+ *    authentication firmware.
+ * 3. Authentication firmware authenticates the payload and extracts
+ *    command and data arrays.
+ * 4. Authentication firmware delivers the configuration to PUnit/PCODE.
+ * 5. Status is returned back to the host via MEI.
+ */
+
+#define INTEL_LB_CMD	0x12
+#define INTEL_LB_RSP	(INTEL_LB_CMD | 0x80)
+
+#define INTEL_LB_SEND_TIMEOUT_MSEC 3000
+#define INTEL_LB_RECV_TIMEOUT_MSEC 3000
+
+/**
+ * struct mei_lb_req - Late Binding request structure
+ * @header: MKHI message header (see struct mkhi_msg_hdr)
+ * @type: Type of the Late Binding payload
+ * @flags: Flags to be passed to the authentication firmware (e.g. %INTEL_LB_FLAGS_IS_PERSISTENT)
+ * @reserved: Reserved for future use by authentication firmware, must be set to 0
+ * @payload_size: Size of the payload data in bytes
+ * @payload: Payload data to be sent to the authentication firmware
+ */
+struct mei_lb_req {
+	struct mkhi_msg_hdr header;
+	__le32 type;
+	__le32 flags;
+	__le32 reserved[2];
+	__le32 payload_size;
+	u8 payload[] __counted_by(payload_size);
+} __packed;
+
+/**
+ * struct mei_lb_rsp - Late Binding response structure
+ * @header: MKHI message header (see struct mkhi_msg_hdr)
+ * @type: Type of the Late Binding payload
+ * @reserved: Reserved for future use by authentication firmware, must be set to 0
+ * @status: Status returned by authentication firmware (see &enum intel_lb_status)
+ */
+struct mei_lb_rsp {
+	struct mkhi_msg_hdr header;
+	__le32 type;
+	__le32 reserved[2];
+	__le32 status;
+} __packed;
+
+static bool mei_lb_check_response(const struct device *dev, ssize_t bytes,
+				  struct mei_lb_rsp *rsp)
+{
+	/*
+	 * Received message size may be smaller than the full message size when
+	 * reply contains only MKHI header with result field set to the error code.
+	 * Check the header size and content first to output exact error, if needed,
+	 * and then process to the whole message.
+	 */
+	if (bytes < sizeof(rsp->header)) {
+		dev_err(dev, "Received less than header size from the firmware: %zd < %zu\n",
+			bytes, sizeof(rsp->header));
+		return false;
+	}
+	if (rsp->header.group_id != MKHI_GROUP_ID_GFX) {
+		dev_err(dev, "Mismatch group id: 0x%x instead of 0x%x\n",
+			rsp->header.group_id, MKHI_GROUP_ID_GFX);
+		return false;
+	}
+	if (rsp->header.command != INTEL_LB_RSP) {
+		dev_err(dev, "Mismatch command: 0x%x instead of 0x%x\n",
+			rsp->header.command, INTEL_LB_RSP);
+		return false;
+	}
+	if (rsp->header.result) {
+		dev_err(dev, "Error in result: 0x%x\n", rsp->header.result);
+		return false;
+	}
+	if (bytes < sizeof(*rsp)) {
+		dev_err(dev, "Received less than message size from the firmware: %zd < %zu\n",
+			bytes, sizeof(*rsp));
+		return false;
+	}
+
+	return true;
+}
+
+static int mei_lb_push_payload(struct device *dev,
+			       enum intel_lb_type type, u32 flags,
+			       const void *payload, size_t payload_size)
+{
+	struct mei_cl_device *cldev;
+	struct mei_lb_req *req = NULL;
+	struct mei_lb_rsp rsp;
+	size_t req_size;
+	ssize_t bytes;
+	int ret;
+
+	cldev = to_mei_cl_device(dev);
+
+	ret = mei_cldev_enable(cldev);
+	if (ret) {
+		dev_dbg(dev, "Failed to enable firmware client. %d\n", ret);
+		return ret;
+	}
+
+	req_size = struct_size(req, payload, payload_size);
+	if (req_size > mei_cldev_mtu(cldev)) {
+		dev_err(dev, "Payload is too big: %zu\n", payload_size);
+		ret = -EMSGSIZE;
+		goto end;
+	}
+
+	req = kmalloc(req_size, GFP_KERNEL);
+	if (!req) {
+		ret = -ENOMEM;
+		goto end;
+	}
+
+	req->header.group_id = MKHI_GROUP_ID_GFX;
+	req->header.command = INTEL_LB_CMD;
+	req->type = cpu_to_le32(type);
+	req->flags = cpu_to_le32(flags);
+	req->reserved[0] = 0;
+	req->reserved[1] = 0;
+	req->payload_size = cpu_to_le32(payload_size);
+	memcpy(req->payload, payload, payload_size);
+
+	bytes = mei_cldev_send_timeout(cldev, (u8 *)req, req_size,
+				       INTEL_LB_SEND_TIMEOUT_MSEC);
+	if (bytes < 0) {
+		dev_err(dev, "Failed to send late binding request to firmware. %zd\n", bytes);
+		ret = bytes;
+		goto end;
+	}
+
+	bytes = mei_cldev_recv_timeout(cldev, (u8 *)&rsp, sizeof(rsp),
+				       INTEL_LB_RECV_TIMEOUT_MSEC);
+	if (bytes < 0) {
+		dev_err(dev, "Failed to receive late binding reply from MEI firmware. %zd\n",
+			bytes);
+		ret = bytes;
+		goto end;
+	}
+	if (!mei_lb_check_response(dev, bytes, &rsp)) {
+		dev_err(dev, "Bad response from the firmware. header: %02x %02x %02x %02x\n",
+			rsp.header.group_id, rsp.header.command,
+			rsp.header.reserved, rsp.header.result);
+		ret = -EPROTO;
+		goto end;
+	}
+
+	dev_dbg(dev, "status = %u\n", le32_to_cpu(rsp.status));
+	ret = (int)le32_to_cpu(rsp.status);
+end:
+	mei_cldev_disable(cldev);
+	kfree(req);
+	return ret;
+}
+
+static const struct intel_lb_component_ops mei_lb_ops = {
+	.push_payload = mei_lb_push_payload,
+};
+
+static int mei_lb_component_master_bind(struct device *dev)
+{
+	return component_bind_all(dev, (void *)&mei_lb_ops);
+}
+
+static void mei_lb_component_master_unbind(struct device *dev)
+{
+	component_unbind_all(dev, (void *)&mei_lb_ops);
+}
+
+static const struct component_master_ops mei_lb_component_master_ops = {
+	.bind = mei_lb_component_master_bind,
+	.unbind = mei_lb_component_master_unbind,
+};
+
+static int mei_lb_component_match(struct device *dev, int subcomponent,
+				  void *data)
+{
+	/*
+	 * This function checks if requester is Intel %PCI_CLASS_DISPLAY_VGA or
+	 * %PCI_CLASS_DISPLAY_OTHER device, and checks if the requester is the
+	 * grand parent of mei_if i.e. late bind MEI device
+	 */
+	struct device *base = data;
+	struct pci_dev *pdev;
+
+	if (!dev)
+		return 0;
+
+	if (!dev_is_pci(dev))
+		return 0;
+
+	pdev = to_pci_dev(dev);
+
+	if (pdev->vendor != PCI_VENDOR_ID_INTEL)
+		return 0;
+
+	if (pdev->class != (PCI_CLASS_DISPLAY_VGA << 8) &&
+	    pdev->class != (PCI_CLASS_DISPLAY_OTHER << 8))
+		return 0;
+
+	if (subcomponent != INTEL_COMPONENT_LB)
+		return 0;
+
+	base = base->parent;
+	if (!base) /* mei device */
+		return 0;
+
+	base = base->parent; /* pci device */
+
+	return !!base && dev == base;
+}
+
+static int mei_lb_probe(struct mei_cl_device *cldev,
+			const struct mei_cl_device_id *id)
+{
+	struct component_match *master_match = NULL;
+	int ret;
+
+	component_match_add_typed(&cldev->dev, &master_match,
+				  mei_lb_component_match, &cldev->dev);
+	if (IS_ERR_OR_NULL(master_match))
+		return -ENOMEM;
+
+	ret = component_master_add_with_match(&cldev->dev,
+					      &mei_lb_component_master_ops,
+					      master_match);
+	if (ret < 0)
+		dev_err(&cldev->dev, "Failed to add late binding master component. %d\n", ret);
+
+	return ret;
+}
+
+static void mei_lb_remove(struct mei_cl_device *cldev)
+{
+	component_master_del(&cldev->dev, &mei_lb_component_master_ops);
+}
+
+#define MEI_GUID_MKHI UUID_LE(0xe2c2afa2, 0x3817, 0x4d19, \
+			      0x9d, 0x95, 0x6, 0xb1, 0x6b, 0x58, 0x8a, 0x5d)
+
+static const struct mei_cl_device_id mei_lb_tbl[] = {
+	{ .uuid = MEI_GUID_MKHI, .version = MEI_CL_VERSION_ANY },
+	{ }
+};
+MODULE_DEVICE_TABLE(mei, mei_lb_tbl);
+
+static struct mei_cl_driver mei_lb_driver = {
+	.id_table = mei_lb_tbl,
+	.name = "mei_lb",
+	.probe = mei_lb_probe,
+	.remove	= mei_lb_remove,
+};
+
+module_mei_cl_driver(mei_lb_driver);
+
+MODULE_AUTHOR("Intel Corporation");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("MEI Late Binding Firmware Update/Upload");
diff --git a/include/drm/intel/i915_component.h b/include/drm/intel/i915_component.h
index 4ea3b17aa143..8082db222e00 100644
--- a/include/drm/intel/i915_component.h
+++ b/include/drm/intel/i915_component.h
@@ -31,6 +31,7 @@ enum i915_component_type {
 	I915_COMPONENT_HDCP,
 	I915_COMPONENT_PXP,
 	I915_COMPONENT_GSC_PROXY,
+	INTEL_COMPONENT_LB,
 };
 
 /* MAX_PORT is the number of port
diff --git a/include/drm/intel/intel_lb_mei_interface.h b/include/drm/intel/intel_lb_mei_interface.h
new file mode 100644
index 000000000000..d65be2cba2ab
--- /dev/null
+++ b/include/drm/intel/intel_lb_mei_interface.h
@@ -0,0 +1,70 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright (c) 2025 Intel Corporation
+ */
+
+#ifndef _INTEL_LB_MEI_INTERFACE_H_
+#define _INTEL_LB_MEI_INTERFACE_H_
+
+#include <linux/types.h>
+
+struct device;
+
+/**
+ * define INTEL_LB_FLAG_IS_PERSISTENT - Mark the payload as persistent
+ *
+ * This flag indicates that the late binding payload should be stored
+ * persistently in flash across warm resets.
+ */
+#define INTEL_LB_FLAG_IS_PERSISTENT	BIT(0)
+
+/**
+ * enum intel_lb_type - enum to determine late binding payload type
+ * @INTEL_LB_TYPE_FAN_CONTROL: Fan controller configuration
+ */
+enum intel_lb_type {
+	INTEL_LB_TYPE_FAN_CONTROL = 1,
+};
+
+/**
+ * enum intel_lb_status - Status codes returned on late binding transmissions
+ * @INTEL_LB_STATUS_SUCCESS: Operation completed successfully
+ * @INTEL_LB_STATUS_4ID_MISMATCH: Mismatch in the expected 4ID (firmware identity/token)
+ * @INTEL_LB_STATUS_ARB_FAILURE: Arbitration failure (e.g. conflicting access or state)
+ * @INTEL_LB_STATUS_GENERAL_ERROR: General firmware error not covered by other codes
+ * @INTEL_LB_STATUS_INVALID_PARAMS: One or more input parameters are invalid
+ * @INTEL_LB_STATUS_INVALID_SIGNATURE: Payload has an invalid or untrusted signature
+ * @INTEL_LB_STATUS_INVALID_PAYLOAD: Payload contents are not accepted by firmware
+ * @INTEL_LB_STATUS_TIMEOUT: Operation timed out before completion
+ */
+enum intel_lb_status {
+	INTEL_LB_STATUS_SUCCESS           = 0,
+	INTEL_LB_STATUS_4ID_MISMATCH      = 1,
+	INTEL_LB_STATUS_ARB_FAILURE       = 2,
+	INTEL_LB_STATUS_GENERAL_ERROR     = 3,
+	INTEL_LB_STATUS_INVALID_PARAMS    = 4,
+	INTEL_LB_STATUS_INVALID_SIGNATURE = 5,
+	INTEL_LB_STATUS_INVALID_PAYLOAD   = 6,
+	INTEL_LB_STATUS_TIMEOUT           = 7,
+};
+
+/**
+ * struct intel_lb_component_ops - Ops for late binding services
+ */
+struct intel_lb_component_ops {
+	/**
+	 * push_payload - Sends a payload to the authentication firmware
+	 * @dev: Device struct corresponding to the mei device
+	 * @type: Payload type (see &enum intel_lb_type)
+	 * @flags: Payload flags bitmap (e.g. %INTEL_LB_FLAGS_IS_PERSISTENT)
+	 * @payload: Pointer to payload buffer
+	 * @payload_size: Payload buffer size in bytes
+	 *
+	 * Return: 0 success, negative errno value on transport failure,
+	 *         positive status returned by firmware
+	 */
+	int (*push_payload)(struct device *dev, u32 type, u32 flags,
+			    const void *payload, size_t payload_size);
+};
+
+#endif /* _INTEL_LB_MEI_INTERFACE_H_ */

From 918bd789d62e6ecbcbc37b2c631ee9127f17bfa9 Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:47 +0530
Subject: [PATCH 1367/1455] drm/xe/xe_late_bind_fw: Introduce xe_late_bind_fw

Introduce xe_late_bind_fw to enable firmware loading for the devices,
such as the fan controller, during the driver probe. Typically,
firmware for such devices are part of IFWI flash image but can be
replaced at probe after OEM tuning.
This patch binds mei late binding component to enable firmware loading.

v2:
 - Add devm_add_action_or_reset to remove the component (Daniele)
 - Add INTEL_MEI_GSC check in xe_late_bind_init() (Daniele)
v3:
 - Fail driver probe if late bind initialization fails,
   add has_late_bind flag (Daniele)
v4:
 - %s/I915_COMPONENT_LATE_BIND/INTEL_COMPONENT_LATE_BIND/
v6:
 - rebased
v7:
 - rebased
 - In xe_late_bind_init, use drm_err when returning an error to
   stop the probe (Lucas)
 - Use imperative mode in commit message (Lucas)

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-4-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/Makefile                |  1 +
 drivers/gpu/drm/xe/xe_device.c             |  5 ++
 drivers/gpu/drm/xe/xe_device_types.h       |  6 ++
 drivers/gpu/drm/xe/xe_late_bind_fw.c       | 84 ++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_late_bind_fw.h       | 15 ++++
 drivers/gpu/drm/xe/xe_late_bind_fw_types.h | 33 +++++++++
 drivers/gpu/drm/xe/xe_pci.c                |  2 +
 drivers/gpu/drm/xe/xe_pci_types.h          |  1 +
 8 files changed, 147 insertions(+)
 create mode 100644 drivers/gpu/drm/xe/xe_late_bind_fw.c
 create mode 100644 drivers/gpu/drm/xe/xe_late_bind_fw.h
 create mode 100644 drivers/gpu/drm/xe/xe_late_bind_fw_types.h

diff --git a/drivers/gpu/drm/xe/Makefile b/drivers/gpu/drm/xe/Makefile
index 041195ca90e6..55b517154279 100644
--- a/drivers/gpu/drm/xe/Makefile
+++ b/drivers/gpu/drm/xe/Makefile
@@ -84,6 +84,7 @@ xe-y += xe_bb.o \
 	xe_hw_error.o \
 	xe_hw_fence.o \
 	xe_irq.o \
+	xe_late_bind_fw.o \
 	xe_lrc.o \
 	xe_migrate.o \
 	xe_mmio.o \
diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
index a4d12ee7d575..fdb7b7498920 100644
--- a/drivers/gpu/drm/xe/xe_device.c
+++ b/drivers/gpu/drm/xe/xe_device.c
@@ -45,6 +45,7 @@
 #include "xe_hwmon.h"
 #include "xe_i2c.h"
 #include "xe_irq.h"
+#include "xe_late_bind_fw.h"
 #include "xe_mmio.h"
 #include "xe_module.h"
 #include "xe_nvm.h"
@@ -903,6 +904,10 @@ int xe_device_probe(struct xe_device *xe)
 	if (err)
 		return err;
 
+	err = xe_late_bind_init(&xe->late_bind);
+	if (err)
+		return err;
+
 	err = xe_oa_init(xe);
 	if (err)
 		return err;
diff --git a/drivers/gpu/drm/xe/xe_device_types.h b/drivers/gpu/drm/xe/xe_device_types.h
index 02e313cf0ae6..772a2e50c919 100644
--- a/drivers/gpu/drm/xe/xe_device_types.h
+++ b/drivers/gpu/drm/xe/xe_device_types.h
@@ -14,6 +14,7 @@
 
 #include "xe_devcoredump_types.h"
 #include "xe_heci_gsc.h"
+#include "xe_late_bind_fw_types.h"
 #include "xe_lmtt_types.h"
 #include "xe_memirq_types.h"
 #include "xe_oa_types.h"
@@ -280,6 +281,8 @@ struct xe_device {
 		u8 has_heci_cscfi:1;
 		/** @info.has_heci_gscfi: device has heci gscfi */
 		u8 has_heci_gscfi:1;
+		/** @info.has_late_bind: Device has firmware late binding support */
+		u8 has_late_bind:1;
 		/** @info.has_llc: Device has a shared CPU+GPU last level cache */
 		u8 has_llc:1;
 		/** @info.has_mbx_power_limits: Device has support to manage power limits using
@@ -533,6 +536,9 @@ struct xe_device {
 	/** @nvm: discrete graphics non-volatile memory */
 	struct intel_dg_nvm_dev *nvm;
 
+	/** @late_bind: xe mei late bind interface */
+	struct xe_late_bind late_bind;
+
 	/** @oa: oa observation subsystem */
 	struct xe_oa oa;
 
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.c b/drivers/gpu/drm/xe/xe_late_bind_fw.c
new file mode 100644
index 000000000000..5f386f860728
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.c
@@ -0,0 +1,84 @@
+// SPDX-License-Identifier: MIT
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#include <linux/component.h>
+#include <linux/delay.h>
+
+#include <drm/drm_managed.h>
+#include <drm/intel/i915_component.h>
+#include <drm/intel/intel_lb_mei_interface.h>
+#include <drm/drm_print.h>
+
+#include "xe_device.h"
+#include "xe_late_bind_fw.h"
+
+static struct xe_device *
+late_bind_to_xe(struct xe_late_bind *late_bind)
+{
+	return container_of(late_bind, struct xe_device, late_bind);
+}
+
+static int xe_late_bind_component_bind(struct device *xe_kdev,
+				       struct device *mei_kdev, void *data)
+{
+	struct xe_device *xe = kdev_to_xe_device(xe_kdev);
+	struct xe_late_bind *late_bind = &xe->late_bind;
+
+	late_bind->component.ops = data;
+	late_bind->component.mei_dev = mei_kdev;
+
+	return 0;
+}
+
+static void xe_late_bind_component_unbind(struct device *xe_kdev,
+					  struct device *mei_kdev, void *data)
+{
+	struct xe_device *xe = kdev_to_xe_device(xe_kdev);
+	struct xe_late_bind *late_bind = &xe->late_bind;
+
+	late_bind->component.ops = NULL;
+}
+
+static const struct component_ops xe_late_bind_component_ops = {
+	.bind   = xe_late_bind_component_bind,
+	.unbind = xe_late_bind_component_unbind,
+};
+
+static void xe_late_bind_remove(void *arg)
+{
+	struct xe_late_bind *late_bind = arg;
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+
+	component_del(xe->drm.dev, &xe_late_bind_component_ops);
+}
+
+/**
+ * xe_late_bind_init() - add xe mei late binding component
+ * @late_bind: pointer to late bind structure.
+ *
+ * Return: 0 if the initialization was successful, a negative errno otherwise.
+ */
+int xe_late_bind_init(struct xe_late_bind *late_bind)
+{
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+	int err;
+
+	if (!xe->info.has_late_bind)
+		return 0;
+
+	if (!IS_ENABLED(CONFIG_INTEL_MEI_LB) || !IS_ENABLED(CONFIG_INTEL_MEI_GSC)) {
+		drm_info(&xe->drm, "Can't init xe mei late bind missing mei component\n");
+		return 0;
+	}
+
+	err = component_add_typed(xe->drm.dev, &xe_late_bind_component_ops,
+				  INTEL_COMPONENT_LB);
+	if (err < 0) {
+		drm_err(&xe->drm, "Failed to add mei late bind component (%pe)\n", ERR_PTR(err));
+		return err;
+	}
+
+	return devm_add_action_or_reset(xe->drm.dev, xe_late_bind_remove, late_bind);
+}
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.h b/drivers/gpu/drm/xe/xe_late_bind_fw.h
new file mode 100644
index 000000000000..4c73571c3e62
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_LATE_BIND_FW_H_
+#define _XE_LATE_BIND_FW_H_
+
+#include <linux/types.h>
+
+struct xe_late_bind;
+
+int xe_late_bind_init(struct xe_late_bind *late_bind);
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
new file mode 100644
index 000000000000..f79e5aefed94
--- /dev/null
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
@@ -0,0 +1,33 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright © 2025 Intel Corporation
+ */
+
+#ifndef _XE_LATE_BIND_TYPES_H_
+#define _XE_LATE_BIND_TYPES_H_
+
+#include <linux/iosys-map.h>
+#include <linux/mutex.h>
+#include <linux/types.h>
+
+/**
+ * struct xe_late_bind_component - Late Binding services component
+ * @mei_dev: device that provide Late Binding service.
+ * @ops: Ops implemented by Late Binding driver, used by Xe driver.
+ *
+ * Communication between Xe and MEI drivers for Late Binding services
+ */
+struct xe_late_bind_component {
+	struct device *mei_dev;
+	const struct late_bind_component_ops *ops;
+};
+
+/**
+ * struct xe_late_bind
+ */
+struct xe_late_bind {
+	/** @component: struct for communication with mei component */
+	struct xe_late_bind_component component;
+};
+
+#endif
diff --git a/drivers/gpu/drm/xe/xe_pci.c b/drivers/gpu/drm/xe/xe_pci.c
index dea027e175b6..be91343829dd 100644
--- a/drivers/gpu/drm/xe/xe_pci.c
+++ b/drivers/gpu/drm/xe/xe_pci.c
@@ -334,6 +334,7 @@ static const struct xe_device_desc bmg_desc = {
 	.has_mbx_power_limits = true,
 	.has_gsc_nvm = 1,
 	.has_heci_cscfi = 1,
+	.has_late_bind = true,
 	.has_sriov = true,
 	.max_gt_per_tile = 2,
 	.needs_scratch = true,
@@ -587,6 +588,7 @@ static int xe_info_init_early(struct xe_device *xe,
 	xe->info.has_gsc_nvm = desc->has_gsc_nvm;
 	xe->info.has_heci_gscfi = desc->has_heci_gscfi;
 	xe->info.has_heci_cscfi = desc->has_heci_cscfi;
+	xe->info.has_late_bind = desc->has_late_bind;
 	xe->info.has_llc = desc->has_llc;
 	xe->info.has_pxp = desc->has_pxp;
 	xe->info.has_sriov = desc->has_sriov;
diff --git a/drivers/gpu/drm/xe/xe_pci_types.h b/drivers/gpu/drm/xe/xe_pci_types.h
index b63002fc0f67..9b9766a3baa3 100644
--- a/drivers/gpu/drm/xe/xe_pci_types.h
+++ b/drivers/gpu/drm/xe/xe_pci_types.h
@@ -39,6 +39,7 @@ struct xe_device_desc {
 	u8 has_gsc_nvm:1;
 	u8 has_heci_gscfi:1;
 	u8 has_heci_cscfi:1;
+	u8 has_late_bind:1;
 	u8 has_llc:1;
 	u8 has_mbx_power_limits:1;
 	u8 has_pxp:1;

From 45832bf9c10f309b579f81b05bacc44fbb4d81b4 Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:48 +0530
Subject: [PATCH 1368/1455] drm/xe/xe_late_bind_fw: Initialize late binding
 firmware

Search for late binding firmware binaries and populate the meta data of
firmware structures.

v2 (Daniele):
 - drm_err if firmware size is more than max pay load size
 - s/request_firmware/firmware_request_nowarn/ as firmware will
   not be available for all possible cards
v3 (Daniele):
 - init firmware from within xe_late_bind_init, propagate error
 - switch late_bind_fw to array to handle multiple firmware types
v4 (Daniele):
 - Alloc payload dynamically, fix nits
v6 (Daniele)
 - %s/MAX_PAYLOAD_SIZE/XE_LB_MAX_PAYLOAD_SIZE/

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-5-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_late_bind_fw.c       | 100 ++++++++++++++++++++-
 drivers/gpu/drm/xe/xe_late_bind_fw_types.h |  30 +++++++
 2 files changed, 129 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.c b/drivers/gpu/drm/xe/xe_late_bind_fw.c
index 5f386f860728..2a9255e73747 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.c
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.c
@@ -5,6 +5,7 @@
 
 #include <linux/component.h>
 #include <linux/delay.h>
+#include <linux/firmware.h>
 
 #include <drm/drm_managed.h>
 #include <drm/intel/i915_component.h>
@@ -13,6 +14,16 @@
 
 #include "xe_device.h"
 #include "xe_late_bind_fw.h"
+#include "xe_pcode.h"
+#include "xe_pcode_api.h"
+
+static const u32 fw_id_to_type[] = {
+		[XE_LB_FW_FAN_CONTROL] = INTEL_LB_TYPE_FAN_CONTROL,
+	};
+
+static const char * const fw_id_to_name[] = {
+		[XE_LB_FW_FAN_CONTROL] = "fan_control",
+	};
 
 static struct xe_device *
 late_bind_to_xe(struct xe_late_bind *late_bind)
@@ -20,6 +31,89 @@ late_bind_to_xe(struct xe_late_bind *late_bind)
 	return container_of(late_bind, struct xe_device, late_bind);
 }
 
+static int xe_late_bind_fw_num_fans(struct xe_late_bind *late_bind)
+{
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+	struct xe_tile *root_tile = xe_device_get_root_tile(xe);
+	u32 uval;
+
+	if (!xe_pcode_read(root_tile,
+			   PCODE_MBOX(FAN_SPEED_CONTROL, FSC_READ_NUM_FANS, 0), &uval, NULL))
+		return uval;
+	else
+		return 0;
+}
+
+static int __xe_late_bind_fw_init(struct xe_late_bind *late_bind, u32 fw_id)
+{
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+	struct pci_dev *pdev = to_pci_dev(xe->drm.dev);
+	struct xe_late_bind_fw *lb_fw;
+	const struct firmware *fw;
+	u32 num_fans;
+	int ret;
+
+	if (fw_id >= XE_LB_FW_MAX_ID)
+		return -EINVAL;
+
+	lb_fw = &late_bind->late_bind_fw[fw_id];
+
+	lb_fw->id = fw_id;
+	lb_fw->type = fw_id_to_type[lb_fw->id];
+	lb_fw->flags &= ~INTEL_LB_FLAG_IS_PERSISTENT;
+
+	if (lb_fw->type == INTEL_LB_TYPE_FAN_CONTROL) {
+		num_fans = xe_late_bind_fw_num_fans(late_bind);
+		drm_dbg(&xe->drm, "Number of Fans: %d\n", num_fans);
+		if (!num_fans)
+			return 0;
+	}
+
+	snprintf(lb_fw->blob_path, sizeof(lb_fw->blob_path), "xe/%s_8086_%04x_%04x_%04x.bin",
+		 fw_id_to_name[lb_fw->id], pdev->device,
+		 pdev->subsystem_vendor, pdev->subsystem_device);
+
+	drm_dbg(&xe->drm, "Request late binding firmware %s\n", lb_fw->blob_path);
+	ret = firmware_request_nowarn(&fw, lb_fw->blob_path, xe->drm.dev);
+	if (ret) {
+		drm_dbg(&xe->drm, "%s late binding fw not available for current device",
+			fw_id_to_name[lb_fw->id]);
+		return 0;
+	}
+
+	if (fw->size > XE_LB_MAX_PAYLOAD_SIZE) {
+		drm_err(&xe->drm, "Firmware %s size %zu is larger than max pay load size %u\n",
+			lb_fw->blob_path, fw->size, XE_LB_MAX_PAYLOAD_SIZE);
+		release_firmware(fw);
+		return -ENODATA;
+	}
+
+	lb_fw->payload_size = fw->size;
+	lb_fw->payload = drmm_kzalloc(&xe->drm, lb_fw->payload_size, GFP_KERNEL);
+	if (!lb_fw->payload) {
+		release_firmware(fw);
+		return -ENOMEM;
+	}
+
+	memcpy((void *)lb_fw->payload, fw->data, lb_fw->payload_size);
+	release_firmware(fw);
+
+	return 0;
+}
+
+static int xe_late_bind_fw_init(struct xe_late_bind *late_bind)
+{
+	int ret;
+	int fw_id;
+
+	for (fw_id = 0; fw_id < XE_LB_FW_MAX_ID; fw_id++) {
+		ret = __xe_late_bind_fw_init(late_bind, fw_id);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
 static int xe_late_bind_component_bind(struct device *xe_kdev,
 				       struct device *mei_kdev, void *data)
 {
@@ -80,5 +174,9 @@ int xe_late_bind_init(struct xe_late_bind *late_bind)
 		return err;
 	}
 
-	return devm_add_action_or_reset(xe->drm.dev, xe_late_bind_remove, late_bind);
+	err = devm_add_action_or_reset(xe->drm.dev, xe_late_bind_remove, late_bind);
+	if (err)
+		return err;
+
+	return xe_late_bind_fw_init(late_bind);
 }
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
index f79e5aefed94..c4a8042f2600 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
@@ -10,6 +10,34 @@
 #include <linux/mutex.h>
 #include <linux/types.h>
 
+#define XE_LB_MAX_PAYLOAD_SIZE SZ_4K
+
+/**
+ * xe_late_bind_fw_id - enum to determine late binding fw index
+ */
+enum xe_late_bind_fw_id {
+	XE_LB_FW_FAN_CONTROL = 0,
+	XE_LB_FW_MAX_ID
+};
+
+/**
+ * struct xe_late_bind_fw
+ */
+struct xe_late_bind_fw {
+	/** @id: firmware index */
+	u32 id;
+	/** @blob_path: firmware binary path */
+	char blob_path[PATH_MAX];
+	/** @type: firmware type */
+	u32  type;
+	/** @flags: firmware flags */
+	u32  flags;
+	/** @payload: to store the late binding blob */
+	const u8  *payload;
+	/** @payload_size: late binding blob payload_size */
+	size_t payload_size;
+};
+
 /**
  * struct xe_late_bind_component - Late Binding services component
  * @mei_dev: device that provide Late Binding service.
@@ -28,6 +56,8 @@ struct xe_late_bind_component {
 struct xe_late_bind {
 	/** @component: struct for communication with mei component */
 	struct xe_late_bind_component component;
+	/** @late_bind_fw: late binding firmware array */
+	struct xe_late_bind_fw late_bind_fw[XE_LB_FW_MAX_ID];
 };
 
 #endif

From 691a54ad94792cd155620e6bac1b33d126efbf1a Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:49 +0530
Subject: [PATCH 1369/1455] drm/xe/xe_late_bind_fw: Load late binding firmware

Load late binding firmware

v2:
 - s/EAGAIN/EBUSY/
 - Flush worker in suspend and driver unload (Daniele)
v3:
 - Use retry interval of 6s, in steps of 200ms, to allow
   other OS components release MEI CL handle (Sasha)
v4:
 - return -ENODEV if component not added (Daniele)
 - parse and print status returned by csc
v5:
 - Use payload to check firmware valid (Daniele)
 - Obtain the RPM reference before scheduling the worker to
   ensure the device remains awake until the worker completes
   firmware loading (Rodrigo)
v6:
 - In case of error donot re-attempt fw download (Daniele)
v7 (Rodrigo):
 - Rename of mei structs and callback.

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-6-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_late_bind_fw.c       | 157 ++++++++++++++++++++-
 drivers/gpu/drm/xe/xe_late_bind_fw.h       |   1 +
 drivers/gpu/drm/xe/xe_late_bind_fw_types.h |   9 +-
 3 files changed, 165 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.c b/drivers/gpu/drm/xe/xe_late_bind_fw.c
index 2a9255e73747..bb161d99602e 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.c
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.c
@@ -16,6 +16,20 @@
 #include "xe_late_bind_fw.h"
 #include "xe_pcode.h"
 #include "xe_pcode_api.h"
+#include "xe_pm.h"
+
+/*
+ * The component should load quite quickly in most cases, but it could take
+ * a bit. Using a very big timeout just to cover the worst case scenario
+ */
+#define LB_INIT_TIMEOUT_MS 20000
+
+/*
+ * Retry interval set to 6 seconds, in steps of 200 ms, to allow time for
+ * other OS components to release the MEI CL handle
+ */
+#define LB_FW_LOAD_RETRY_MAXCOUNT 30
+#define LB_FW_LOAD_RETRY_PAUSE_MS 200
 
 static const u32 fw_id_to_type[] = {
 		[XE_LB_FW_FAN_CONTROL] = INTEL_LB_TYPE_FAN_CONTROL,
@@ -31,6 +45,30 @@ late_bind_to_xe(struct xe_late_bind *late_bind)
 	return container_of(late_bind, struct xe_device, late_bind);
 }
 
+static const char *xe_late_bind_parse_status(uint32_t status)
+{
+	switch (status) {
+	case INTEL_LB_STATUS_SUCCESS:
+		return "success";
+	case INTEL_LB_STATUS_4ID_MISMATCH:
+		return "4Id Mismatch";
+	case INTEL_LB_STATUS_ARB_FAILURE:
+		return "ARB Failure";
+	case INTEL_LB_STATUS_GENERAL_ERROR:
+		return "General Error";
+	case INTEL_LB_STATUS_INVALID_PARAMS:
+		return "Invalid Params";
+	case INTEL_LB_STATUS_INVALID_SIGNATURE:
+		return "Invalid Signature";
+	case INTEL_LB_STATUS_INVALID_PAYLOAD:
+		return "Invalid Payload";
+	case INTEL_LB_STATUS_TIMEOUT:
+		return "Timeout";
+	default:
+		return "Unknown error";
+	}
+}
+
 static int xe_late_bind_fw_num_fans(struct xe_late_bind *late_bind)
 {
 	struct xe_device *xe = late_bind_to_xe(late_bind);
@@ -44,6 +82,101 @@ static int xe_late_bind_fw_num_fans(struct xe_late_bind *late_bind)
 		return 0;
 }
 
+static void xe_late_bind_wait_for_worker_completion(struct xe_late_bind *late_bind)
+{
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+	struct xe_late_bind_fw *lbfw;
+	int fw_id;
+
+	for (fw_id = 0; fw_id < XE_LB_FW_MAX_ID; fw_id++) {
+		lbfw = &late_bind->late_bind_fw[fw_id];
+		if (lbfw->payload && late_bind->wq) {
+			drm_dbg(&xe->drm, "Flush work: load %s firmware\n",
+				fw_id_to_name[lbfw->id]);
+			flush_work(&lbfw->work);
+		}
+	}
+}
+
+static void xe_late_bind_work(struct work_struct *work)
+{
+	struct xe_late_bind_fw *lbfw = container_of(work, struct xe_late_bind_fw, work);
+	struct xe_late_bind *late_bind = container_of(lbfw, struct xe_late_bind,
+						      late_bind_fw[lbfw->id]);
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+	int retry = LB_FW_LOAD_RETRY_MAXCOUNT;
+	int ret;
+	int slept;
+
+	xe_device_assert_mem_access(xe);
+
+	/* we can queue this before the component is bound */
+	for (slept = 0; slept < LB_INIT_TIMEOUT_MS; slept += 100) {
+		if (late_bind->component.ops)
+			break;
+		msleep(100);
+	}
+
+	if (!late_bind->component.ops) {
+		drm_err(&xe->drm, "Late bind component not bound\n");
+		/* Do not re-attempt fw load */
+		drmm_kfree(&xe->drm, (void *)lbfw->payload);
+		lbfw->payload = NULL;
+		goto out;
+	}
+
+	drm_dbg(&xe->drm, "Load %s firmware\n", fw_id_to_name[lbfw->id]);
+
+	do {
+		ret = late_bind->component.ops->push_payload(late_bind->component.mei_dev,
+							     lbfw->type,
+							     lbfw->flags,
+							     lbfw->payload,
+							     lbfw->payload_size);
+		if (!ret)
+			break;
+		msleep(LB_FW_LOAD_RETRY_PAUSE_MS);
+	} while (--retry && ret == -EBUSY);
+
+	if (!ret) {
+		drm_dbg(&xe->drm, "Load %s firmware successful\n",
+			fw_id_to_name[lbfw->id]);
+		goto out;
+	}
+
+	if (ret > 0)
+		drm_err(&xe->drm, "Load %s firmware failed with err %d, %s\n",
+			fw_id_to_name[lbfw->id], ret, xe_late_bind_parse_status(ret));
+	else
+		drm_err(&xe->drm, "Load %s firmware failed with err %d",
+			fw_id_to_name[lbfw->id], ret);
+	/* Do not re-attempt fw load */
+	drmm_kfree(&xe->drm, (void *)lbfw->payload);
+	lbfw->payload = NULL;
+
+out:
+	xe_pm_runtime_put(xe);
+}
+
+int xe_late_bind_fw_load(struct xe_late_bind *late_bind)
+{
+	struct xe_device *xe = late_bind_to_xe(late_bind);
+	struct xe_late_bind_fw *lbfw;
+	int fw_id;
+
+	if (!late_bind->component_added)
+		return -ENODEV;
+
+	for (fw_id = 0; fw_id < XE_LB_FW_MAX_ID; fw_id++) {
+		lbfw = &late_bind->late_bind_fw[fw_id];
+		if (lbfw->payload) {
+			xe_pm_runtime_get_noresume(xe);
+			queue_work(late_bind->wq, &lbfw->work);
+		}
+	}
+	return 0;
+}
+
 static int __xe_late_bind_fw_init(struct xe_late_bind *late_bind, u32 fw_id)
 {
 	struct xe_device *xe = late_bind_to_xe(late_bind);
@@ -97,6 +230,7 @@ static int __xe_late_bind_fw_init(struct xe_late_bind *late_bind, u32 fw_id)
 
 	memcpy((void *)lb_fw->payload, fw->data, lb_fw->payload_size);
 	release_firmware(fw);
+	INIT_WORK(&lb_fw->work, xe_late_bind_work);
 
 	return 0;
 }
@@ -106,11 +240,16 @@ static int xe_late_bind_fw_init(struct xe_late_bind *late_bind)
 	int ret;
 	int fw_id;
 
+	late_bind->wq = alloc_ordered_workqueue("late-bind-ordered-wq", 0);
+	if (!late_bind->wq)
+		return -ENOMEM;
+
 	for (fw_id = 0; fw_id < XE_LB_FW_MAX_ID; fw_id++) {
 		ret = __xe_late_bind_fw_init(late_bind, fw_id);
 		if (ret)
 			return ret;
 	}
+
 	return 0;
 }
 
@@ -132,6 +271,8 @@ static void xe_late_bind_component_unbind(struct device *xe_kdev,
 	struct xe_device *xe = kdev_to_xe_device(xe_kdev);
 	struct xe_late_bind *late_bind = &xe->late_bind;
 
+	xe_late_bind_wait_for_worker_completion(late_bind);
+
 	late_bind->component.ops = NULL;
 }
 
@@ -145,7 +286,15 @@ static void xe_late_bind_remove(void *arg)
 	struct xe_late_bind *late_bind = arg;
 	struct xe_device *xe = late_bind_to_xe(late_bind);
 
+	xe_late_bind_wait_for_worker_completion(late_bind);
+
+	late_bind->component_added = false;
+
 	component_del(xe->drm.dev, &xe_late_bind_component_ops);
+	if (late_bind->wq) {
+		destroy_workqueue(late_bind->wq);
+		late_bind->wq = NULL;
+	}
 }
 
 /**
@@ -174,9 +323,15 @@ int xe_late_bind_init(struct xe_late_bind *late_bind)
 		return err;
 	}
 
+	late_bind->component_added = true;
+
 	err = devm_add_action_or_reset(xe->drm.dev, xe_late_bind_remove, late_bind);
 	if (err)
 		return err;
 
-	return xe_late_bind_fw_init(late_bind);
+	err = xe_late_bind_fw_init(late_bind);
+	if (err)
+		return err;
+
+	return xe_late_bind_fw_load(late_bind);
 }
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.h b/drivers/gpu/drm/xe/xe_late_bind_fw.h
index 4c73571c3e62..28d56ed2bfdc 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.h
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.h
@@ -11,5 +11,6 @@
 struct xe_late_bind;
 
 int xe_late_bind_init(struct xe_late_bind *late_bind);
+int xe_late_bind_fw_load(struct xe_late_bind *late_bind);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
index c4a8042f2600..5c0574aff7b9 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
@@ -9,6 +9,7 @@
 #include <linux/iosys-map.h>
 #include <linux/mutex.h>
 #include <linux/types.h>
+#include <linux/workqueue.h>
 
 #define XE_LB_MAX_PAYLOAD_SIZE SZ_4K
 
@@ -36,6 +37,8 @@ struct xe_late_bind_fw {
 	const u8  *payload;
 	/** @payload_size: late binding blob payload_size */
 	size_t payload_size;
+	/** @work: worker to upload latebind blob */
+	struct work_struct work;
 };
 
 /**
@@ -47,7 +50,7 @@ struct xe_late_bind_fw {
  */
 struct xe_late_bind_component {
 	struct device *mei_dev;
-	const struct late_bind_component_ops *ops;
+	const struct intel_lb_component_ops *ops;
 };
 
 /**
@@ -58,6 +61,10 @@ struct xe_late_bind {
 	struct xe_late_bind_component component;
 	/** @late_bind_fw: late binding firmware array */
 	struct xe_late_bind_fw late_bind_fw[XE_LB_FW_MAX_ID];
+	/** @wq: workqueue to submit request to download late bind blob */
+	struct workqueue_struct *wq;
+	/** @component_added: whether the component has been added */
+	bool component_added;
 };
 
 #endif

From 69ac1bb8fca5ea5fea40364ded86b4ceae2ffa21 Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:50 +0530
Subject: [PATCH 1370/1455] drm/xe/xe_late_bind_fw: Reload late binding fw in
 rpm resume

Reload late binding fw during runtime resume.

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-7-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_late_bind_fw.c | 2 +-
 drivers/gpu/drm/xe/xe_late_bind_fw.h | 1 +
 drivers/gpu/drm/xe/xe_pm.c           | 4 ++++
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.c b/drivers/gpu/drm/xe/xe_late_bind_fw.c
index bb161d99602e..d4d64677bf48 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.c
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.c
@@ -82,7 +82,7 @@ static int xe_late_bind_fw_num_fans(struct xe_late_bind *late_bind)
 		return 0;
 }
 
-static void xe_late_bind_wait_for_worker_completion(struct xe_late_bind *late_bind)
+void xe_late_bind_wait_for_worker_completion(struct xe_late_bind *late_bind)
 {
 	struct xe_device *xe = late_bind_to_xe(late_bind);
 	struct xe_late_bind_fw *lbfw;
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.h b/drivers/gpu/drm/xe/xe_late_bind_fw.h
index 28d56ed2bfdc..07e437390539 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.h
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.h
@@ -12,5 +12,6 @@ struct xe_late_bind;
 
 int xe_late_bind_init(struct xe_late_bind *late_bind);
 int xe_late_bind_fw_load(struct xe_late_bind *late_bind);
+void xe_late_bind_wait_for_worker_completion(struct xe_late_bind *late_bind);
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index a303ed5780b4..d34e1c1be35b 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -21,6 +21,7 @@
 #include "xe_gt_idle.h"
 #include "xe_i2c.h"
 #include "xe_irq.h"
+#include "xe_late_bind_fw.h"
 #include "xe_pcode.h"
 #include "xe_pxp.h"
 #include "xe_sriov_vf_ccs.h"
@@ -601,6 +602,9 @@ int xe_pm_runtime_resume(struct xe_device *xe)
 	if (IS_VF_CCS_READY(xe))
 		xe_sriov_vf_ccs_register_context(xe);
 
+	if (xe->d3cold.allowed)
+		xe_late_bind_fw_load(&xe->late_bind);
+
 out:
 	xe_rpm_lockmap_release(xe);
 	xe_pm_write_callback_task(xe, NULL);

From 02f52f6d924037a20e90a280363afba10e425367 Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:51 +0530
Subject: [PATCH 1371/1455] drm/xe/xe_late_bind_fw: Reload late binding fw
 during system resume

Reload late binding fw during resume from system suspend

v2:
  - Unconditionally reload late binding fw (Rodrigo)
  - Flush worker during system suspend

Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-8-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_pm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_pm.c b/drivers/gpu/drm/xe/xe_pm.c
index d34e1c1be35b..d6625c71115b 100644
--- a/drivers/gpu/drm/xe/xe_pm.c
+++ b/drivers/gpu/drm/xe/xe_pm.c
@@ -130,6 +130,8 @@ int xe_pm_suspend(struct xe_device *xe)
 	if (err)
 		goto err;
 
+	xe_late_bind_wait_for_worker_completion(&xe->late_bind);
+
 	for_each_gt(gt, xe, id)
 		xe_gt_suspend_prepare(gt);
 
@@ -217,6 +219,8 @@ int xe_pm_resume(struct xe_device *xe)
 	if (IS_VF_CCS_READY(xe))
 		xe_sriov_vf_ccs_register_context(xe);
 
+	xe_late_bind_fw_load(&xe->late_bind);
+
 	drm_dbg(&xe->drm, "Device resumed\n");
 	return 0;
 err:

From 67de7982d5053f5a277f86e4560fea98dd95dcdf Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:52 +0530
Subject: [PATCH 1372/1455] drm/xe/xe_late_bind_fw: Introduce debug fs node to
 disable late binding

Introduce a debug filesystem node to disable late binding fw reload
during the system or runtime resume. This is intended for situations
where the late binding fw needs to be loaded from user mode,
perticularly for validation purpose.
Note that xe kmd doesn't participate in late binding flow from user
space. Binary loaded from the userspace will be lost upon entering to
D3 cold hence user space app need to handle this situation.

v2:
  - s/(uval == 1) ? true : false/!!uval/ (Daniele)
v3:
  - Refine the commit message (Daniele)

Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-9-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_debugfs.c            | 41 ++++++++++++++++++++++
 drivers/gpu/drm/xe/xe_late_bind_fw.c       |  3 ++
 drivers/gpu/drm/xe/xe_late_bind_fw_types.h |  2 ++
 3 files changed, 46 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_debugfs.c b/drivers/gpu/drm/xe/xe_debugfs.c
index d67e32ebbe29..cd977dbd1ef6 100644
--- a/drivers/gpu/drm/xe/xe_debugfs.c
+++ b/drivers/gpu/drm/xe/xe_debugfs.c
@@ -331,6 +331,44 @@ static const struct file_operations atomic_svm_timeslice_ms_fops = {
 	.write = atomic_svm_timeslice_ms_set,
 };
 
+static ssize_t disable_late_binding_show(struct file *f, char __user *ubuf,
+					 size_t size, loff_t *pos)
+{
+	struct xe_device *xe = file_inode(f)->i_private;
+	struct xe_late_bind *late_bind = &xe->late_bind;
+	char buf[32];
+	int len;
+
+	len = scnprintf(buf, sizeof(buf), "%d\n", late_bind->disable);
+
+	return simple_read_from_buffer(ubuf, size, pos, buf, len);
+}
+
+static ssize_t disable_late_binding_set(struct file *f, const char __user *ubuf,
+					size_t size, loff_t *pos)
+{
+	struct xe_device *xe = file_inode(f)->i_private;
+	struct xe_late_bind *late_bind = &xe->late_bind;
+	u32 uval;
+	ssize_t ret;
+
+	ret = kstrtouint_from_user(ubuf, size, sizeof(uval), &uval);
+	if (ret)
+		return ret;
+
+	if (uval > 1)
+		return -EINVAL;
+
+	late_bind->disable = !!uval;
+	return size;
+}
+
+static const struct file_operations disable_late_binding_fops = {
+	.owner = THIS_MODULE,
+	.read = disable_late_binding_show,
+	.write = disable_late_binding_set,
+};
+
 void xe_debugfs_register(struct xe_device *xe)
 {
 	struct ttm_device *bdev = &xe->ttm;
@@ -364,6 +402,9 @@ void xe_debugfs_register(struct xe_device *xe)
 	debugfs_create_file("atomic_svm_timeslice_ms", 0600, root, xe,
 			    &atomic_svm_timeslice_ms_fops);
 
+	debugfs_create_file("disable_late_binding", 0600, root, xe,
+			    &disable_late_binding_fops);
+
 	for (mem_type = XE_PL_VRAM0; mem_type <= XE_PL_VRAM1; ++mem_type) {
 		man = ttm_manager_type(bdev, mem_type);
 
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.c b/drivers/gpu/drm/xe/xe_late_bind_fw.c
index d4d64677bf48..0f062008ca83 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.c
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.c
@@ -167,6 +167,9 @@ int xe_late_bind_fw_load(struct xe_late_bind *late_bind)
 	if (!late_bind->component_added)
 		return -ENODEV;
 
+	if (late_bind->disable)
+		return 0;
+
 	for (fw_id = 0; fw_id < XE_LB_FW_MAX_ID; fw_id++) {
 		lbfw = &late_bind->late_bind_fw[fw_id];
 		if (lbfw->payload) {
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
index 5c0574aff7b9..158dc1abe072 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
@@ -65,6 +65,8 @@ struct xe_late_bind {
 	struct workqueue_struct *wq;
 	/** @component_added: whether the component has been added */
 	bool component_added;
+	/** @disable: to block late binding reload during pm resume flow*/
+	bool disable;
 };
 
 #endif

From efa29317a55392a4a1e7934426fb602e50dc55fc Mon Sep 17 00:00:00 2001
From: Badal Nilawar <badal.nilawar@intel.com>
Date: Fri, 5 Sep 2025 21:19:53 +0530
Subject: [PATCH 1373/1455] drm/xe/xe_late_bind_fw: Extract and print version
 info

Extract and print version info of the late binding binary.

v2: Some refinements (Daniele)

Signed-off-by: Badal Nilawar <badal.nilawar@intel.com>
Reviewed-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250905154953.3974335-10-badal.nilawar@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_late_bind_fw.c       | 124 +++++++++++++++++++++
 drivers/gpu/drm/xe/xe_late_bind_fw_types.h |   3 +
 drivers/gpu/drm/xe/xe_uc_fw_abi.h          |  66 +++++++++++
 3 files changed, 193 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw.c b/drivers/gpu/drm/xe/xe_late_bind_fw.c
index 0f062008ca83..38f3feb2aecd 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw.c
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw.c
@@ -45,6 +45,121 @@ late_bind_to_xe(struct xe_late_bind *late_bind)
 	return container_of(late_bind, struct xe_device, late_bind);
 }
 
+static struct xe_device *
+late_bind_fw_to_xe(struct xe_late_bind_fw *lb_fw)
+{
+	return container_of(lb_fw, struct xe_device, late_bind.late_bind_fw[lb_fw->id]);
+}
+
+/* Refer to the "Late Bind based Firmware Layout" documentation entry for details */
+static int parse_cpd_header(struct xe_late_bind_fw *lb_fw,
+			    const void *data, size_t size, const char *manifest_entry)
+{
+	struct xe_device *xe = late_bind_fw_to_xe(lb_fw);
+	const struct gsc_cpd_header_v2 *header = data;
+	const struct gsc_manifest_header *manifest;
+	const struct gsc_cpd_entry *entry;
+	size_t min_size = sizeof(*header);
+	u32 offset;
+	int i;
+
+	/* manifest_entry is mandatory */
+	xe_assert(xe, manifest_entry);
+
+	if (size < min_size || header->header_marker != GSC_CPD_HEADER_MARKER)
+		return -ENOENT;
+
+	if (header->header_length < sizeof(struct gsc_cpd_header_v2)) {
+		drm_err(&xe->drm, "%s late binding fw: Invalid CPD header length %u!\n",
+			fw_id_to_name[lb_fw->id], header->header_length);
+		return -EINVAL;
+	}
+
+	min_size = header->header_length + sizeof(struct gsc_cpd_entry) * header->num_of_entries;
+	if (size < min_size) {
+		drm_err(&xe->drm, "%s late binding fw: too small! %zu < %zu\n",
+			fw_id_to_name[lb_fw->id], size, min_size);
+		return -ENODATA;
+	}
+
+	/* Look for the manifest first */
+	entry = (void *)header + header->header_length;
+	for (i = 0; i < header->num_of_entries; i++, entry++)
+		if (strcmp(entry->name, manifest_entry) == 0)
+			offset = entry->offset & GSC_CPD_ENTRY_OFFSET_MASK;
+
+	if (!offset) {
+		drm_err(&xe->drm, "%s late binding fw: Failed to find manifest_entry\n",
+			fw_id_to_name[lb_fw->id]);
+		return -ENODATA;
+	}
+
+	min_size = offset + sizeof(struct gsc_manifest_header);
+	if (size < min_size) {
+		drm_err(&xe->drm, "%s late binding fw: too small! %zu < %zu\n",
+			fw_id_to_name[lb_fw->id], size, min_size);
+		return -ENODATA;
+	}
+
+	manifest = data + offset;
+
+	lb_fw->version = manifest->fw_version;
+
+	return 0;
+}
+
+/* Refer to the "Late Bind based Firmware Layout" documentation entry for details */
+static int parse_lb_layout(struct xe_late_bind_fw *lb_fw,
+			   const void *data, size_t size, const char *fpt_entry)
+{
+	struct xe_device *xe = late_bind_fw_to_xe(lb_fw);
+	const struct csc_fpt_header *header = data;
+	const struct csc_fpt_entry *entry;
+	size_t min_size = sizeof(*header);
+	u32 offset;
+	int i;
+
+	/* fpt_entry is mandatory */
+	xe_assert(xe, fpt_entry);
+
+	if (size < min_size || header->header_marker != CSC_FPT_HEADER_MARKER)
+		return -ENOENT;
+
+	if (header->header_length < sizeof(struct csc_fpt_header)) {
+		drm_err(&xe->drm, "%s late binding fw: Invalid FPT header length %u!\n",
+			fw_id_to_name[lb_fw->id], header->header_length);
+		return -EINVAL;
+	}
+
+	min_size = header->header_length + sizeof(struct csc_fpt_entry) * header->num_of_entries;
+	if (size < min_size) {
+		drm_err(&xe->drm, "%s late binding fw: too small! %zu < %zu\n",
+			fw_id_to_name[lb_fw->id], size, min_size);
+		return -ENODATA;
+	}
+
+	/* Look for the cpd header first */
+	entry = (void *)header + header->header_length;
+	for (i = 0; i < header->num_of_entries; i++, entry++)
+		if (strcmp(entry->name, fpt_entry) == 0)
+			offset = entry->offset;
+
+	if (!offset) {
+		drm_err(&xe->drm, "%s late binding fw: Failed to find fpt_entry\n",
+			fw_id_to_name[lb_fw->id]);
+		return -ENODATA;
+	}
+
+	min_size = offset + sizeof(struct gsc_cpd_header_v2);
+	if (size < min_size) {
+		drm_err(&xe->drm, "%s late binding fw: too small! %zu < %zu\n",
+			fw_id_to_name[lb_fw->id], size, min_size);
+		return -ENODATA;
+	}
+
+	return parse_cpd_header(lb_fw, data + offset, size - offset, "LTES.man");
+}
+
 static const char *xe_late_bind_parse_status(uint32_t status)
 {
 	switch (status) {
@@ -224,6 +339,10 @@ static int __xe_late_bind_fw_init(struct xe_late_bind *late_bind, u32 fw_id)
 		return -ENODATA;
 	}
 
+	ret = parse_lb_layout(lb_fw, fw->data, fw->size, "LTES");
+	if (ret)
+		return ret;
+
 	lb_fw->payload_size = fw->size;
 	lb_fw->payload = drmm_kzalloc(&xe->drm, lb_fw->payload_size, GFP_KERNEL);
 	if (!lb_fw->payload) {
@@ -231,6 +350,11 @@ static int __xe_late_bind_fw_init(struct xe_late_bind *late_bind, u32 fw_id)
 		return -ENOMEM;
 	}
 
+	drm_info(&xe->drm, "Using %s firmware from %s version %u.%u.%u.%u\n",
+		 fw_id_to_name[lb_fw->id], lb_fw->blob_path,
+		 lb_fw->version.major, lb_fw->version.minor,
+		 lb_fw->version.hotfix, lb_fw->version.build);
+
 	memcpy((void *)lb_fw->payload, fw->data, lb_fw->payload_size);
 	release_firmware(fw);
 	INIT_WORK(&lb_fw->work, xe_late_bind_work);
diff --git a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
index 158dc1abe072..0f5da89ce98b 100644
--- a/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
+++ b/drivers/gpu/drm/xe/xe_late_bind_fw_types.h
@@ -10,6 +10,7 @@
 #include <linux/mutex.h>
 #include <linux/types.h>
 #include <linux/workqueue.h>
+#include "xe_uc_fw_abi.h"
 
 #define XE_LB_MAX_PAYLOAD_SIZE SZ_4K
 
@@ -39,6 +40,8 @@ struct xe_late_bind_fw {
 	size_t payload_size;
 	/** @work: worker to upload latebind blob */
 	struct work_struct work;
+	/** @version: late binding blob manifest version */
+	struct gsc_version version;
 };
 
 /**
diff --git a/drivers/gpu/drm/xe/xe_uc_fw_abi.h b/drivers/gpu/drm/xe/xe_uc_fw_abi.h
index faceb437fd2f..3c9a63d13032 100644
--- a/drivers/gpu/drm/xe/xe_uc_fw_abi.h
+++ b/drivers/gpu/drm/xe/xe_uc_fw_abi.h
@@ -336,4 +336,70 @@ struct gsc_manifest_header {
 	u32 exponent_size; /* in dwords */
 } __packed;
 
+/**
+ * DOC: Late binding Firmware Layout
+ *
+ * The Late binding binary starts with FPT header, which contains locations
+ * of various partitions of the binary. Here we're interested in finding out
+ * manifest version. To the manifest version, we need to locate CPD header
+ * one of the entry in CPD header points to manifest header. Manifest header
+ * contains the version.
+ *
+ *      +================================================+
+ *      |  FPT Header                                    |
+ *      +================================================+
+ *      |  FPT entries[]                                 |
+ *      |      entry1                                    |
+ *      |      ...                                       |
+ *      |      entryX                                    |
+ *      |          "LTES"                                |
+ *      |          ...                                   |
+ *      |          offset  >-----------------------------|------o
+ *      +================================================+      |
+ *                                                              |
+ *      +================================================+      |
+ *      |  CPD Header                                    |<-----o
+ *      +================================================+
+ *      |  CPD entries[]                                 |
+ *      |      entry1                                    |
+ *      |      ...                                       |
+ *      |      entryX                                    |
+ *      |          "LTES.man"                            |
+ *      |           ...                                  |
+ *      |           offset  >----------------------------|------o
+ *      +================================================+      |
+ *                                                              |
+ *      +================================================+      |
+ *      |  Manifest Header                               |<-----o
+ *      |      ...                                       |
+ *      |      FW version                                |
+ *      |      ...                                       |
+ *      +================================================+
+ */
+
+/* FPT Headers */
+struct csc_fpt_header {
+	u32 header_marker;
+#define CSC_FPT_HEADER_MARKER 0x54504624
+	u32 num_of_entries;
+	u8 header_version;
+	u8 entry_version;
+	u8 header_length; /* in bytes */
+	u8 flags;
+	u16 ticks_to_add;
+	u16 tokens_to_add;
+	u32 uma_size;
+	u32 crc32;
+	struct gsc_version fitc_version;
+} __packed;
+
+struct csc_fpt_entry {
+	u8 name[4]; /* partition name */
+	u32 reserved1;
+	u32 offset; /* offset from beginning of CSE region */
+	u32 length; /* partition length in bytes */
+	u32 reserved2[3];
+	u32 partition_flags;
+} __packed;
+
 #endif

From 59e4405e9ee2b318342d252422a82dd863b89ef4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Wed, 27 Aug 2025 11:45:45 +0200
Subject: [PATCH 1374/1455] drm/amdgpu: revert to old status lock handling v3
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

It turned out that protecting the status of each bo_va with a
spinlock was just hiding problems instead of solving them.

Revert the whole approach, add a separate stats_lock and lockdep
assertions that the correct reservation lock is held all over the place.

This not only allows for better checks if a state transition is properly
protected by a lock, but also switching back to using list macros to
iterate over the state of lists protected by the dma_resv lock of the
root PD.

v2: re-add missing check
v3: split into two patches

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c |   8 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c    | 168 +++++++++++-----------
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h    |  15 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c |   4 -
 4 files changed, 93 insertions(+), 102 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index 22a5827b0772..a22e6025de61 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -724,12 +724,12 @@ amdgpu_userq_bo_validate(struct amdgpu_device *adev, struct drm_exec *exec,
 	struct amdgpu_bo *bo;
 	int ret;
 
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->invalidated_lock);
 	while (!list_empty(&vm->invalidated)) {
 		bo_va = list_first_entry(&vm->invalidated,
 					 struct amdgpu_bo_va,
 					 base.vm_status);
-		spin_unlock(&vm->status_lock);
+		spin_unlock(&vm->invalidated_lock);
 
 		bo = bo_va->base.bo;
 		ret = drm_exec_prepare_obj(exec, &bo->tbo.base, 2);
@@ -746,9 +746,9 @@ amdgpu_userq_bo_validate(struct amdgpu_device *adev, struct drm_exec *exec,
 		if (ret)
 			return ret;
 
-		spin_lock(&vm->status_lock);
+		spin_lock(&vm->invalidated_lock);
 	}
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->invalidated_lock);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 1ccd348bf48b..1f8b43253eea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -127,6 +127,17 @@ struct amdgpu_vm_tlb_seq_struct {
 	struct dma_fence_cb cb;
 };
 
+/**
+ * amdgpu_vm_assert_locked - check if VM is correctly locked
+ * @vm: the VM which schould be tested
+ *
+ * Asserts that the VM root PD is locked.
+ */
+static void amdgpu_vm_assert_locked(struct amdgpu_vm *vm)
+{
+	dma_resv_assert_held(vm->root.bo->tbo.base.resv);
+}
+
 /**
  * amdgpu_vm_set_pasid - manage pasid and vm ptr mapping
  *
@@ -143,6 +154,8 @@ int amdgpu_vm_set_pasid(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 {
 	int r;
 
+	amdgpu_vm_assert_locked(vm);
+
 	if (vm->pasid == pasid)
 		return 0;
 
@@ -181,12 +194,11 @@ static void amdgpu_vm_bo_evicted(struct amdgpu_vm_bo_base *vm_bo)
 	struct amdgpu_bo *bo = vm_bo->bo;
 
 	vm_bo->moved = true;
-	spin_lock(&vm_bo->vm->status_lock);
+	amdgpu_vm_assert_locked(vm);
 	if (bo->tbo.type == ttm_bo_type_kernel)
 		list_move(&vm_bo->vm_status, &vm->evicted);
 	else
 		list_move_tail(&vm_bo->vm_status, &vm->evicted);
-	spin_unlock(&vm_bo->vm->status_lock);
 }
 /**
  * amdgpu_vm_bo_moved - vm_bo is moved
@@ -198,9 +210,8 @@ static void amdgpu_vm_bo_evicted(struct amdgpu_vm_bo_base *vm_bo)
  */
 static void amdgpu_vm_bo_moved(struct amdgpu_vm_bo_base *vm_bo)
 {
-	spin_lock(&vm_bo->vm->status_lock);
+	amdgpu_vm_assert_locked(vm_bo->vm);
 	list_move(&vm_bo->vm_status, &vm_bo->vm->moved);
-	spin_unlock(&vm_bo->vm->status_lock);
 }
 
 /**
@@ -213,9 +224,8 @@ static void amdgpu_vm_bo_moved(struct amdgpu_vm_bo_base *vm_bo)
  */
 static void amdgpu_vm_bo_idle(struct amdgpu_vm_bo_base *vm_bo)
 {
-	spin_lock(&vm_bo->vm->status_lock);
+	amdgpu_vm_assert_locked(vm_bo->vm);
 	list_move(&vm_bo->vm_status, &vm_bo->vm->idle);
-	spin_unlock(&vm_bo->vm->status_lock);
 	vm_bo->moved = false;
 }
 
@@ -229,9 +239,9 @@ static void amdgpu_vm_bo_idle(struct amdgpu_vm_bo_base *vm_bo)
  */
 static void amdgpu_vm_bo_invalidated(struct amdgpu_vm_bo_base *vm_bo)
 {
-	spin_lock(&vm_bo->vm->status_lock);
+	spin_lock(&vm_bo->vm->invalidated_lock);
 	list_move(&vm_bo->vm_status, &vm_bo->vm->invalidated);
-	spin_unlock(&vm_bo->vm->status_lock);
+	spin_unlock(&vm_bo->vm->invalidated_lock);
 }
 
 /**
@@ -244,10 +254,9 @@ static void amdgpu_vm_bo_invalidated(struct amdgpu_vm_bo_base *vm_bo)
  */
 static void amdgpu_vm_bo_evicted_user(struct amdgpu_vm_bo_base *vm_bo)
 {
+	amdgpu_vm_assert_locked(vm_bo->vm);
 	vm_bo->moved = true;
-	spin_lock(&vm_bo->vm->status_lock);
 	list_move(&vm_bo->vm_status, &vm_bo->vm->evicted_user);
-	spin_unlock(&vm_bo->vm->status_lock);
 }
 
 /**
@@ -260,13 +269,11 @@ static void amdgpu_vm_bo_evicted_user(struct amdgpu_vm_bo_base *vm_bo)
  */
 static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo)
 {
-	if (vm_bo->bo->parent) {
-		spin_lock(&vm_bo->vm->status_lock);
+	amdgpu_vm_assert_locked(vm_bo->vm);
+	if (vm_bo->bo->parent)
 		list_move(&vm_bo->vm_status, &vm_bo->vm->relocated);
-		spin_unlock(&vm_bo->vm->status_lock);
-	} else {
+	else
 		amdgpu_vm_bo_idle(vm_bo);
-	}
 }
 
 /**
@@ -279,9 +286,8 @@ static void amdgpu_vm_bo_relocated(struct amdgpu_vm_bo_base *vm_bo)
  */
 static void amdgpu_vm_bo_done(struct amdgpu_vm_bo_base *vm_bo)
 {
-	spin_lock(&vm_bo->vm->status_lock);
+	amdgpu_vm_assert_locked(vm_bo->vm);
 	list_move(&vm_bo->vm_status, &vm_bo->vm->done);
-	spin_unlock(&vm_bo->vm->status_lock);
 }
 
 /**
@@ -295,10 +301,13 @@ static void amdgpu_vm_bo_reset_state_machine(struct amdgpu_vm *vm)
 {
 	struct amdgpu_vm_bo_base *vm_bo, *tmp;
 
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->invalidated_lock);
 	list_splice_init(&vm->done, &vm->invalidated);
 	list_for_each_entry(vm_bo, &vm->invalidated, vm_status)
 		vm_bo->moved = true;
+	spin_unlock(&vm->invalidated_lock);
+
+	amdgpu_vm_assert_locked(vm_bo->vm);
 	list_for_each_entry_safe(vm_bo, tmp, &vm->idle, vm_status) {
 		struct amdgpu_bo *bo = vm_bo->bo;
 
@@ -308,14 +317,13 @@ static void amdgpu_vm_bo_reset_state_machine(struct amdgpu_vm *vm)
 		else if (bo->parent)
 			list_move(&vm_bo->vm_status, &vm_bo->vm->relocated);
 	}
-	spin_unlock(&vm->status_lock);
 }
 
 /**
  * amdgpu_vm_update_shared - helper to update shared memory stat
  * @base: base structure for tracking BO usage in a VM
  *
- * Takes the vm status_lock and updates the shared memory stat. If the basic
+ * Takes the vm stats_lock and updates the shared memory stat. If the basic
  * stat changed (e.g. buffer was moved) amdgpu_vm_update_stats need to be called
  * as well.
  */
@@ -327,7 +335,8 @@ static void amdgpu_vm_update_shared(struct amdgpu_vm_bo_base *base)
 	uint32_t bo_memtype = amdgpu_bo_mem_stats_placement(bo);
 	bool shared;
 
-	spin_lock(&vm->status_lock);
+	dma_resv_assert_held(bo->tbo.base.resv);
+	spin_lock(&vm->stats_lock);
 	shared = drm_gem_object_is_shared_for_memory_stats(&bo->tbo.base);
 	if (base->shared != shared) {
 		base->shared = shared;
@@ -339,7 +348,7 @@ static void amdgpu_vm_update_shared(struct amdgpu_vm_bo_base *base)
 			vm->stats[bo_memtype].drm.private += size;
 		}
 	}
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->stats_lock);
 }
 
 /**
@@ -364,11 +373,11 @@ void amdgpu_vm_bo_update_shared(struct amdgpu_bo *bo)
  *        be bo->tbo.resource
  * @sign: if we should add (+1) or subtract (-1) from the stat
  *
- * Caller need to have the vm status_lock held. Useful for when multiple update
+ * Caller need to have the vm stats_lock held. Useful for when multiple update
  * need to happen at the same time.
  */
 static void amdgpu_vm_update_stats_locked(struct amdgpu_vm_bo_base *base,
-			    struct ttm_resource *res, int sign)
+					  struct ttm_resource *res, int sign)
 {
 	struct amdgpu_vm *vm = base->vm;
 	struct amdgpu_bo *bo = base->bo;
@@ -392,7 +401,8 @@ static void amdgpu_vm_update_stats_locked(struct amdgpu_vm_bo_base *base,
 		 */
 		if (bo->flags & AMDGPU_GEM_CREATE_DISCARDABLE)
 			vm->stats[res_memtype].drm.purgeable += size;
-		if (!(bo->preferred_domains & amdgpu_mem_type_to_domain(res_memtype)))
+		if (!(bo->preferred_domains &
+		      amdgpu_mem_type_to_domain(res_memtype)))
 			vm->stats[bo_memtype].evicted += size;
 	}
 }
@@ -411,9 +421,9 @@ void amdgpu_vm_update_stats(struct amdgpu_vm_bo_base *base,
 {
 	struct amdgpu_vm *vm = base->vm;
 
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->stats_lock);
 	amdgpu_vm_update_stats_locked(base, res, sign);
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->stats_lock);
 }
 
 /**
@@ -439,10 +449,10 @@ void amdgpu_vm_bo_base_init(struct amdgpu_vm_bo_base *base,
 	base->next = bo->vm_bo;
 	bo->vm_bo = base;
 
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->stats_lock);
 	base->shared = drm_gem_object_is_shared_for_memory_stats(&bo->tbo.base);
 	amdgpu_vm_update_stats_locked(base, bo->tbo.resource, +1);
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->stats_lock);
 
 	if (!amdgpu_vm_is_bo_always_valid(vm, bo))
 		return;
@@ -501,10 +511,10 @@ int amdgpu_vm_lock_done_list(struct amdgpu_vm *vm, struct drm_exec *exec,
 	int ret;
 
 	/* We can only trust prev->next while holding the lock */
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->invalidated_lock);
 	while (!list_is_head(prev->next, &vm->done)) {
 		bo_va = list_entry(prev->next, typeof(*bo_va), base.vm_status);
-		spin_unlock(&vm->status_lock);
+		spin_unlock(&vm->invalidated_lock);
 
 		bo = bo_va->base.bo;
 		if (bo) {
@@ -512,10 +522,10 @@ int amdgpu_vm_lock_done_list(struct amdgpu_vm *vm, struct drm_exec *exec,
 			if (unlikely(ret))
 				return ret;
 		}
-		spin_lock(&vm->status_lock);
+		spin_lock(&vm->invalidated_lock);
 		prev = prev->next;
 	}
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->invalidated_lock);
 
 	return 0;
 }
@@ -611,7 +621,7 @@ int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		       void *param)
 {
 	uint64_t new_vm_generation = amdgpu_vm_generation(adev, vm);
-	struct amdgpu_vm_bo_base *bo_base;
+	struct amdgpu_vm_bo_base *bo_base, *tmp;
 	struct amdgpu_bo *bo;
 	int r;
 
@@ -624,13 +634,7 @@ int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			return r;
 	}
 
-	spin_lock(&vm->status_lock);
-	while (!list_empty(&vm->evicted)) {
-		bo_base = list_first_entry(&vm->evicted,
-					   struct amdgpu_vm_bo_base,
-					   vm_status);
-		spin_unlock(&vm->status_lock);
-
+	list_for_each_entry_safe(bo_base, tmp, &vm->evicted, vm_status) {
 		bo = bo_base->bo;
 
 		r = validate(param, bo);
@@ -643,26 +647,21 @@ int amdgpu_vm_validate(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			vm->update_funcs->map_table(to_amdgpu_bo_vm(bo));
 			amdgpu_vm_bo_relocated(bo_base);
 		}
-		spin_lock(&vm->status_lock);
 	}
-	while (ticket && !list_empty(&vm->evicted_user)) {
-		bo_base = list_first_entry(&vm->evicted_user,
-					   struct amdgpu_vm_bo_base,
-					   vm_status);
-		spin_unlock(&vm->status_lock);
 
-		bo = bo_base->bo;
-		dma_resv_assert_held(bo->tbo.base.resv);
+	if (ticket) {
+		list_for_each_entry_safe(bo_base, tmp, &vm->evicted_user,
+					 vm_status) {
+			bo = bo_base->bo;
+			dma_resv_assert_held(bo->tbo.base.resv);
 
-		r = validate(param, bo);
-		if (r)
-			return r;
+			r = validate(param, bo);
+			if (r)
+				return r;
 
-		amdgpu_vm_bo_invalidated(bo_base);
-
-		spin_lock(&vm->status_lock);
+			amdgpu_vm_bo_invalidated(bo_base);
+		}
 	}
-	spin_unlock(&vm->status_lock);
 
 	amdgpu_vm_eviction_lock(vm);
 	vm->evicting = false;
@@ -685,13 +684,13 @@ bool amdgpu_vm_ready(struct amdgpu_vm *vm)
 {
 	bool ret;
 
+	amdgpu_vm_assert_locked(vm);
+
 	amdgpu_vm_eviction_lock(vm);
 	ret = !vm->evicting;
 	amdgpu_vm_eviction_unlock(vm);
 
-	spin_lock(&vm->status_lock);
 	ret &= list_empty(&vm->evicted);
-	spin_unlock(&vm->status_lock);
 
 	spin_lock(&vm->immediate.lock);
 	ret &= !vm->immediate.stopped;
@@ -982,16 +981,13 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 			  struct amdgpu_vm *vm, bool immediate)
 {
 	struct amdgpu_vm_update_params params;
-	struct amdgpu_vm_bo_base *entry;
+	struct amdgpu_vm_bo_base *entry, *tmp;
 	bool flush_tlb_needed = false;
-	LIST_HEAD(relocated);
 	int r, idx;
 
-	spin_lock(&vm->status_lock);
-	list_splice_init(&vm->relocated, &relocated);
-	spin_unlock(&vm->status_lock);
+	amdgpu_vm_assert_locked(vm);
 
-	if (list_empty(&relocated))
+	if (list_empty(&vm->relocated))
 		return 0;
 
 	if (!drm_dev_enter(adev_to_drm(adev), &idx))
@@ -1006,7 +1002,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 	if (r)
 		goto error;
 
-	list_for_each_entry(entry, &relocated, vm_status) {
+	list_for_each_entry(entry, &vm->relocated, vm_status) {
 		/* vm_flush_needed after updating moved PDEs */
 		flush_tlb_needed |= entry->moved;
 
@@ -1022,9 +1018,7 @@ int amdgpu_vm_update_pdes(struct amdgpu_device *adev,
 	if (flush_tlb_needed)
 		atomic64_inc(&vm->tlb_seq);
 
-	while (!list_empty(&relocated)) {
-		entry = list_first_entry(&relocated, struct amdgpu_vm_bo_base,
-					 vm_status);
+	list_for_each_entry_safe(entry, tmp, &vm->relocated, vm_status) {
 		amdgpu_vm_bo_idle(entry);
 	}
 
@@ -1250,9 +1244,9 @@ error_free:
 void amdgpu_vm_get_memory(struct amdgpu_vm *vm,
 			  struct amdgpu_mem_stats stats[__AMDGPU_PL_NUM])
 {
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->stats_lock);
 	memcpy(stats, vm->stats, sizeof(*stats) * __AMDGPU_PL_NUM);
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->stats_lock);
 }
 
 /**
@@ -1619,29 +1613,24 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev,
 			   struct amdgpu_vm *vm,
 			   struct ww_acquire_ctx *ticket)
 {
-	struct amdgpu_bo_va *bo_va;
+	struct amdgpu_bo_va *bo_va, *tmp;
 	struct dma_resv *resv;
 	bool clear, unlock;
 	int r;
 
-	spin_lock(&vm->status_lock);
-	while (!list_empty(&vm->moved)) {
-		bo_va = list_first_entry(&vm->moved, struct amdgpu_bo_va,
-					 base.vm_status);
-		spin_unlock(&vm->status_lock);
-
+	list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) {
 		/* Per VM BOs never need to bo cleared in the page tables */
 		r = amdgpu_vm_bo_update(adev, bo_va, false);
 		if (r)
 			return r;
-		spin_lock(&vm->status_lock);
 	}
 
+	spin_lock(&vm->invalidated_lock);
 	while (!list_empty(&vm->invalidated)) {
 		bo_va = list_first_entry(&vm->invalidated, struct amdgpu_bo_va,
 					 base.vm_status);
 		resv = bo_va->base.bo->tbo.base.resv;
-		spin_unlock(&vm->status_lock);
+		spin_unlock(&vm->invalidated_lock);
 
 		/* Try to reserve the BO to avoid clearing its ptes */
 		if (!adev->debug_vm && dma_resv_trylock(resv)) {
@@ -1673,9 +1662,9 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev,
 		     bo_va->base.bo->tbo.resource->mem_type == TTM_PL_SYSTEM))
 			amdgpu_vm_bo_evicted_user(&bo_va->base);
 
-		spin_lock(&vm->status_lock);
+		spin_lock(&vm->invalidated_lock);
 	}
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->invalidated_lock);
 
 	return 0;
 }
@@ -2204,9 +2193,9 @@ void amdgpu_vm_bo_del(struct amdgpu_device *adev,
 		}
 	}
 
-	spin_lock(&vm->status_lock);
+	spin_lock(&vm->invalidated_lock);
 	list_del(&bo_va->base.vm_status);
-	spin_unlock(&vm->status_lock);
+	spin_unlock(&vm->invalidated_lock);
 
 	list_for_each_entry_safe(mapping, next, &bo_va->valids, list) {
 		list_del(&mapping->list);
@@ -2314,10 +2303,10 @@ void amdgpu_vm_bo_move(struct amdgpu_bo *bo, struct ttm_resource *new_mem,
 	for (bo_base = bo->vm_bo; bo_base; bo_base = bo_base->next) {
 		struct amdgpu_vm *vm = bo_base->vm;
 
-		spin_lock(&vm->status_lock);
+		spin_lock(&vm->stats_lock);
 		amdgpu_vm_update_stats_locked(bo_base, bo->tbo.resource, -1);
 		amdgpu_vm_update_stats_locked(bo_base, new_mem, +1);
-		spin_unlock(&vm->status_lock);
+		spin_unlock(&vm->stats_lock);
 	}
 
 	amdgpu_vm_bo_invalidate(bo, evicted);
@@ -2584,11 +2573,12 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	INIT_LIST_HEAD(&vm->relocated);
 	INIT_LIST_HEAD(&vm->moved);
 	INIT_LIST_HEAD(&vm->idle);
+	spin_lock_init(&vm->invalidated_lock);
 	INIT_LIST_HEAD(&vm->invalidated);
-	spin_lock_init(&vm->status_lock);
 	INIT_LIST_HEAD(&vm->freed);
 	INIT_LIST_HEAD(&vm->done);
 	INIT_KFIFO(vm->faults);
+	spin_lock_init(&vm->stats_lock);
 
 	r = amdgpu_vm_init_entities(adev, vm);
 	if (r)
@@ -3053,7 +3043,8 @@ void amdgpu_debugfs_vm_bo_info(struct amdgpu_vm *vm, struct seq_file *m)
 	unsigned int total_done_objs = 0;
 	unsigned int id = 0;
 
-	spin_lock(&vm->status_lock);
+	amdgpu_vm_assert_locked(vm);
+
 	seq_puts(m, "\tIdle BOs:\n");
 	list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) {
 		if (!bo_va->base.bo)
@@ -3091,11 +3082,13 @@ void amdgpu_debugfs_vm_bo_info(struct amdgpu_vm *vm, struct seq_file *m)
 	id = 0;
 
 	seq_puts(m, "\tInvalidated BOs:\n");
+	spin_lock(&vm->invalidated_lock);
 	list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) {
 		if (!bo_va->base.bo)
 			continue;
 		total_invalidated += amdgpu_bo_print_info(id++,	bo_va->base.bo, m);
 	}
+	spin_unlock(&vm->invalidated_lock);
 	total_invalidated_objs = id;
 	id = 0;
 
@@ -3105,7 +3098,6 @@ void amdgpu_debugfs_vm_bo_info(struct amdgpu_vm *vm, struct seq_file *m)
 			continue;
 		total_done += amdgpu_bo_print_info(id++, bo_va->base.bo, m);
 	}
-	spin_unlock(&vm->status_lock);
 	total_done_objs = id;
 
 	seq_printf(m, "\tTotal idle size:        %12lld\tobjs:\t%d\n", total_idle,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 74e61e45778e..829b400cb8c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -203,11 +203,11 @@ struct amdgpu_vm_bo_base {
 	/* protected by bo being reserved */
 	struct amdgpu_vm_bo_base	*next;
 
-	/* protected by vm status_lock */
+	/* protected by vm reservation and invalidated_lock */
 	struct list_head		vm_status;
 
 	/* if the bo is counted as shared in mem stats
-	 * protected by vm status_lock */
+	 * protected by vm BO being reserved */
 	bool				shared;
 
 	/* protected by the BO being reserved */
@@ -343,10 +343,8 @@ struct amdgpu_vm {
 	bool			evicting;
 	unsigned int		saved_flags;
 
-	/* Lock to protect vm_bo add/del/move on all lists of vm */
-	spinlock_t		status_lock;
-
-	/* Memory statistics for this vm, protected by status_lock */
+	/* Memory statistics for this vm, protected by stats_lock */
+	spinlock_t		stats_lock;
 	struct amdgpu_mem_stats stats[__AMDGPU_PL_NUM];
 
 	/*
@@ -354,6 +352,8 @@ struct amdgpu_vm {
 	 * PDs, PTs or per VM BOs. The state transits are:
 	 *
 	 * evicted -> relocated (PDs, PTs) or moved (per VM BOs) -> idle
+	 *
+	 * Lists are protected by the root PD dma_resv lock.
 	 */
 
 	/* Per-VM and PT BOs who needs a validation */
@@ -374,7 +374,10 @@ struct amdgpu_vm {
 	 * state transits are:
 	 *
 	 * evicted_user or invalidated -> done
+	 *
+	 * Lists are protected by the invalidated_lock.
 	 */
+	spinlock_t		invalidated_lock;
 
 	/* BOs for user mode queues that need a validation */
 	struct list_head	evicted_user;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
index 30022123b0bf..f57c48b74274 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_pt.c
@@ -541,9 +541,7 @@ static void amdgpu_vm_pt_free(struct amdgpu_vm_bo_base *entry)
 	entry->bo->vm_bo = NULL;
 	ttm_bo_set_bulk_move(&entry->bo->tbo, NULL);
 
-	spin_lock(&entry->vm->status_lock);
 	list_del(&entry->vm_status);
-	spin_unlock(&entry->vm->status_lock);
 	amdgpu_bo_unref(&entry->bo);
 }
 
@@ -587,7 +585,6 @@ static void amdgpu_vm_pt_add_list(struct amdgpu_vm_update_params *params,
 	struct amdgpu_vm_pt_cursor seek;
 	struct amdgpu_vm_bo_base *entry;
 
-	spin_lock(&params->vm->status_lock);
 	for_each_amdgpu_vm_pt_dfs_safe(params->adev, params->vm, cursor, seek, entry) {
 		if (entry && entry->bo)
 			list_move(&entry->vm_status, &params->tlb_flush_waitlist);
@@ -595,7 +592,6 @@ static void amdgpu_vm_pt_add_list(struct amdgpu_vm_update_params *params,
 
 	/* enter start node now */
 	list_move(&cursor->entry->vm_status, &params->tlb_flush_waitlist);
-	spin_unlock(&params->vm->status_lock);
 }
 
 /**

From cc9a8e238e42c1f43b98c097995137d644b69245 Mon Sep 17 00:00:00 2001
From: Guangshuo Li <lgs201920130244@gmail.com>
Date: Thu, 18 Sep 2025 18:57:05 +0800
Subject: [PATCH 1375/1455] drm/amdgpu/atom: Check kcalloc() for WS buffer in
 amdgpu_atom_execute_table_locked()

kcalloc() may fail. When WS is non-zero and allocation fails, ectx.ws
remains NULL while ectx.ws_size is set, leading to a potential NULL
pointer dereference in atom_get_src_int() when accessing WS entries.

Return -ENOMEM on allocation failure to avoid the NULL dereference.

Signed-off-by: Guangshuo Li <lgs201920130244@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/atom.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index 1c994d0cc50b..be5d67c2c7a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -1246,6 +1246,10 @@ static int amdgpu_atom_execute_table_locked(struct atom_context *ctx, int index,
 	ectx.last_jump_jiffies = 0;
 	if (ws) {
 		ectx.ws = kcalloc(4, ws, GFP_KERNEL);
+		if (!ectx.ws) {
+			ret = -ENOMEM;
+			goto free;
+		}
 		ectx.ws_size = ws;
 	} else {
 		ectx.ws = NULL;

From a490c8d77d500b5981e739be3d59c60cfe382536 Mon Sep 17 00:00:00 2001
From: Matthew Schwartz <matthew.schwartz@linux.dev>
Date: Thu, 11 Sep 2025 10:48:51 -0700
Subject: [PATCH 1376/1455] drm/amd/display: Only restore backlight after
 amdgpu_dm_init or dm_resume

On clients that utilize AMD_PRIVATE_COLOR properties for HDR support,
brightness sliders can include a hardware controlled portion and a
gamma-based portion. This is the case on the Steam Deck OLED when using
gamescope with Steam as a client.

When a user sets a brightness level while HDR is active, the gamma-based
portion and/or hardware portion are adjusted to achieve the desired
brightness. However, when a modeset takes place while the gamma-based
portion is in-use, restoring the hardware brightness level overrides the
user's overall brightness level and results in a mismatch between what
the slider reports and the display's current brightness.

To avoid overriding gamma-based brightness, only restore HW backlight
level after boot or resume. This ensures that the backlight level is
set correctly after the DC layer resets it while avoiding interference
with subsequent modesets.

Fixes: 7875afafba84 ("drm/amd/display: Fix brightness level not retained over reboot")
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4551
Signed-off-by: Matthew Schwartz <matthew.schwartz@linux.dev>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 12 ++++++++----
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  7 +++++++
 2 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index f76d2ef9ca5c..8170923b3c76 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -2081,6 +2081,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
 
 	dc_hardware_init(adev->dm.dc);
 
+	adev->dm.restore_backlight = true;
+
 	adev->dm.hpd_rx_offload_wq = hpd_rx_irq_create_workqueue(adev);
 	if (!adev->dm.hpd_rx_offload_wq) {
 		drm_err(adev_to_drm(adev), "failed to create hpd rx offload workqueue.\n");
@@ -3436,6 +3438,7 @@ static int dm_resume(struct amdgpu_ip_block *ip_block)
 		dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D0);
 
 		dc_resume(dm->dc);
+		adev->dm.restore_backlight = true;
 
 		amdgpu_dm_irq_resume_early(adev);
 
@@ -9927,7 +9930,6 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state,
 	bool mode_set_reset_required = false;
 	u32 i;
 	struct dc_commit_streams_params params = {dc_state->streams, dc_state->stream_count};
-	bool set_backlight_level = false;
 
 	/* Disable writeback */
 	for_each_old_connector_in_state(state, connector, old_con_state, i) {
@@ -10047,7 +10049,6 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state,
 			acrtc->hw_mode = new_crtc_state->mode;
 			crtc->hwmode = new_crtc_state->mode;
 			mode_set_reset_required = true;
-			set_backlight_level = true;
 		} else if (modereset_required(new_crtc_state)) {
 			drm_dbg_atomic(dev,
 				       "Atomic commit: RESET. crtc id %d:[%p]\n",
@@ -10104,13 +10105,16 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state,
 	 * to fix a flicker issue.
 	 * It will cause the dm->actual_brightness is not the current panel brightness
 	 * level. (the dm->brightness is the correct panel level)
-	 * So we set the backlight level with dm->brightness value after set mode
+	 * So we set the backlight level with dm->brightness value after initial
+	 * set mode. Use restore_backlight flag to avoid setting backlight level
+	 * for every subsequent mode set.
 	 */
-	if (set_backlight_level) {
+	if (dm->restore_backlight) {
 		for (i = 0; i < dm->num_of_edps; i++) {
 			if (dm->backlight_dev[i])
 				amdgpu_dm_backlight_set_level(dm, i, dm->brightness[i]);
 		}
+		dm->restore_backlight = false;
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 2967cd28efe4..69441980c3fa 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -625,6 +625,13 @@ struct amdgpu_display_manager {
 	 */
 	u32 actual_brightness[AMDGPU_DM_MAX_NUM_EDP];
 
+	/**
+	 * @restore_backlight:
+	 *
+	 * Flag to indicate whether to restore backlight after modeset.
+	 */
+	bool restore_backlight;
+
 	/**
 	 * @aux_hpd_discon_quirk:
 	 *

From 7166cc3a6aae9aaa529f52e1468006e67e3e6846 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:38 -0700
Subject: [PATCH 1377/1455] drm/xe/configfs: Extract function to parse engine

Move the part that copies the engine to a local buffer so it can be
shared in future for other configfs attributes parsing an engine.

Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-1-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 32 +++++++++++++++++++++-----------
 1 file changed, 21 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index e52808e3199f..8db7a2af1101 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -283,24 +283,34 @@ static bool lookup_engine_mask(const char *pattern, u64 *mask)
 	return false;
 }
 
+static int parse_engine(const char *s, const char *end_chars, u64 *mask)
+{
+	char buf[MAX_ENGINE_CLASS_CHARS + MAX_ENGINE_INSTANCE_CHARS + 1];
+	size_t len;
+
+	len = strcspn(s, end_chars);
+	if (len >= sizeof(buf))
+		return -EINVAL;
+
+	memcpy(buf, s, len);
+	buf[len] = '\0';
+
+	if (!lookup_engine_mask(buf, mask))
+		return -ENOENT;
+
+	return len;
+}
+
 static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 				     size_t len)
 {
 	struct xe_config_group_device *dev = to_xe_config_group_device(item);
-	size_t patternlen, p;
+	ssize_t patternlen, p;
 	u64 mask, val = 0;
 
 	for (p = 0; p < len; p += patternlen + 1) {
-		char buf[MAX_ENGINE_CLASS_CHARS + MAX_ENGINE_INSTANCE_CHARS + 1];
-
-		patternlen = strcspn(page + p, ",\n");
-		if (patternlen >= sizeof(buf))
-			return -EINVAL;
-
-		memcpy(buf, page + p, patternlen);
-		buf[patternlen] = '\0';
-
-		if (!lookup_engine_mask(buf, &mask))
+		patternlen = parse_engine(page + p, ",\n", &mask);
+		if (patternlen < 0)
 			return -EINVAL;
 
 		val |= mask;

From e2a9854d806e1cb95861a0f3c7125caaefd7ef03 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:39 -0700
Subject: [PATCH 1378/1455] drm/xe/configfs: Allow to select by class only

For a future configfs attribute, it's desirable to select by engine mask
only as the instance doesn't make sense.

Rename the function lookup_engine_mask() to lookup_engine_info() and
make it return the entry. This allows parse_engine() to still return an
item if the caller wants to allow parsing a class-only string like
"rcs", "bcs", "ccs", etc.

Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-2-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 50 ++++++++++++++++++++++----------
 1 file changed, 35 insertions(+), 15 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 8db7a2af1101..45fdb6cf26b5 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -152,6 +152,7 @@ static void set_device_defaults(struct xe_config_device *config)
 struct engine_info {
 	const char *cls;
 	u64 mask;
+	enum xe_engine_class engine_class;
 };
 
 /* Some helpful macros to aid on the sizing of buffer allocation when parsing */
@@ -159,12 +160,12 @@ struct engine_info {
 #define MAX_ENGINE_INSTANCE_CHARS 2
 
 static const struct engine_info engine_info[] = {
-	{ .cls = "rcs", .mask = XE_HW_ENGINE_RCS_MASK },
-	{ .cls = "bcs", .mask = XE_HW_ENGINE_BCS_MASK },
-	{ .cls = "vcs", .mask = XE_HW_ENGINE_VCS_MASK },
-	{ .cls = "vecs", .mask = XE_HW_ENGINE_VECS_MASK },
-	{ .cls = "ccs", .mask = XE_HW_ENGINE_CCS_MASK },
-	{ .cls = "gsccs", .mask = XE_HW_ENGINE_GSCCS_MASK },
+	{ .cls = "rcs", .mask = XE_HW_ENGINE_RCS_MASK, .engine_class = XE_ENGINE_CLASS_RENDER },
+	{ .cls = "bcs", .mask = XE_HW_ENGINE_BCS_MASK, .engine_class = XE_ENGINE_CLASS_COPY },
+	{ .cls = "vcs", .mask = XE_HW_ENGINE_VCS_MASK, .engine_class = XE_ENGINE_CLASS_VIDEO_DECODE },
+	{ .cls = "vecs", .mask = XE_HW_ENGINE_VECS_MASK, .engine_class = XE_ENGINE_CLASS_VIDEO_ENHANCE },
+	{ .cls = "ccs", .mask = XE_HW_ENGINE_CCS_MASK, .engine_class = XE_ENGINE_CLASS_COMPUTE },
+	{ .cls = "gsccs", .mask = XE_HW_ENGINE_GSCCS_MASK, .engine_class = XE_ENGINE_CLASS_OTHER },
 };
 
 static struct xe_config_group_device *to_xe_config_group_device(struct config_item *item)
@@ -253,7 +254,18 @@ static ssize_t engines_allowed_show(struct config_item *item, char *page)
 	return p - page;
 }
 
-static bool lookup_engine_mask(const char *pattern, u64 *mask)
+/*
+ * Lookup engine_info. If @mask is not NULL, reduce the mask according to the
+ * instance in @pattern.
+ *
+ * Examples of inputs:
+ * - lookup_engine_info("rcs0", &mask): return "rcs" entry from @engine_info and
+ *   mask == BIT_ULL(XE_HW_ENGINE_RCS0)
+ * - lookup_engine_info("rcs*", &mask): return "rcs" entry from @engine_info and
+ *   mask == XE_HW_ENGINE_RCS_MASK
+ * - lookup_engine_info("rcs", NULL): return "rcs" entry from @engine_info
+ */
+static const struct engine_info *lookup_engine_info(const char *pattern, u64 *mask)
 {
 	for (size_t i = 0; i < ARRAY_SIZE(engine_info); i++) {
 		u8 instance;
@@ -263,29 +275,33 @@ static bool lookup_engine_mask(const char *pattern, u64 *mask)
 			continue;
 
 		pattern += strlen(engine_info[i].cls);
+		if (!mask && !*pattern)
+			return &engine_info[i];
 
 		if (!strcmp(pattern, "*")) {
 			*mask = engine_info[i].mask;
-			return true;
+			return &engine_info[i];
 		}
 
 		if (kstrtou8(pattern, 10, &instance))
-			return false;
+			return NULL;
 
 		bit = __ffs64(engine_info[i].mask) + instance;
 		if (bit >= fls64(engine_info[i].mask))
-			return false;
+			return NULL;
 
 		*mask = BIT_ULL(bit);
-		return true;
+		return &engine_info[i];
 	}
 
-	return false;
+	return NULL;
 }
 
-static int parse_engine(const char *s, const char *end_chars, u64 *mask)
+static int parse_engine(const char *s, const char *end_chars, u64 *mask,
+			const struct engine_info **pinfo)
 {
 	char buf[MAX_ENGINE_CLASS_CHARS + MAX_ENGINE_INSTANCE_CHARS + 1];
+	const struct engine_info *info;
 	size_t len;
 
 	len = strcspn(s, end_chars);
@@ -295,9 +311,13 @@ static int parse_engine(const char *s, const char *end_chars, u64 *mask)
 	memcpy(buf, s, len);
 	buf[len] = '\0';
 
-	if (!lookup_engine_mask(buf, mask))
+	info = lookup_engine_info(buf, mask);
+	if (!info)
 		return -ENOENT;
 
+	if (pinfo)
+		*pinfo = info;
+
 	return len;
 }
 
@@ -309,7 +329,7 @@ static ssize_t engines_allowed_store(struct config_item *item, const char *page,
 	u64 mask, val = 0;
 
 	for (p = 0; p < len; p += patternlen + 1) {
-		patternlen = parse_engine(page + p, ",\n", &mask);
+		patternlen = parse_engine(page + p, ",\n", &mask, NULL);
 		if (patternlen < 0)
 			return -EINVAL;
 

From 6c6988c5e03df145e6c5cabc50d67b11f5aa2e56 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:40 -0700
Subject: [PATCH 1379/1455] drm/xe/lrc: Allow to add user commands on context
 switch

During validation it's useful to allows additional commands to be
executed on context switch. Fetch the commands from configfs (to be
added) and add them to the WA BB.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-3-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 13 +++++++++++++
 drivers/gpu/drm/xe/xe_configfs.h |  6 ++++++
 drivers/gpu/drm/xe/xe_lrc.c      | 32 ++++++++++++++++++++++++++++++++
 3 files changed, 51 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 45fdb6cf26b5..d86c75af0327 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -633,6 +633,19 @@ bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev)
 	return ret;
 }
 
+/**
+ * xe_configfs_get_ctx_restore_post_bb - get configfs ctx_restore_post_bb setting
+ * @pdev: pci device
+ *
+ * Return: post_ctx_restore setting in configfs
+ */
+u32 xe_configfs_get_ctx_restore_post_bb(struct pci_dev *pdev,
+					enum xe_engine_class class,
+					const u32 **cs)
+{
+	return 0;
+}
+
 int __init xe_configfs_init(void)
 {
 	int ret;
diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
index 1402e863b71c..eff2645b5f59 100644
--- a/drivers/gpu/drm/xe/xe_configfs.h
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -8,6 +8,8 @@
 #include <linux/limits.h>
 #include <linux/types.h>
 
+#include <xe_hw_engine_types.h>
+
 struct pci_dev;
 
 #if IS_ENABLED(CONFIG_CONFIGFS_FS)
@@ -17,6 +19,8 @@ void xe_configfs_check_device(struct pci_dev *pdev);
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
 bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev);
+u32 xe_configfs_get_ctx_restore_post_bb(struct pci_dev *pdev, enum xe_engine_class,
+					const u32 **cs);
 #else
 static inline int xe_configfs_init(void) { return 0; }
 static inline void xe_configfs_exit(void) { }
@@ -24,6 +28,8 @@ static inline void xe_configfs_check_device(struct pci_dev *pdev) { }
 static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; }
 static inline u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { return U64_MAX; }
 static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
+static inline u32 xe_configfs_get_ctx_restore_post_bb(struct pci_dev *pdev, enum xe_engine_class,
+						      const u32 **cs) { return 0; }
 #endif
 
 #endif
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 6d52e0eb97f5..c706585611d5 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -8,6 +8,7 @@
 #include <generated/xe_wa_oob.h>
 
 #include <linux/ascii85.h>
+#include <linux/panic.h>
 
 #include "instructions/xe_mi_commands.h"
 #include "instructions/xe_gfxpipe_commands.h"
@@ -16,6 +17,7 @@
 #include "regs/xe_lrc_layout.h"
 #include "xe_bb.h"
 #include "xe_bo.h"
+#include "xe_configfs.h"
 #include "xe_device.h"
 #include "xe_drm_client.h"
 #include "xe_exec_queue_types.h"
@@ -1102,6 +1104,35 @@ static ssize_t setup_timestamp_wa(struct xe_lrc *lrc, struct xe_hw_engine *hwe,
 	return cmd - batch;
 }
 
+static ssize_t setup_configfs_post_ctx_restore_bb(struct xe_lrc *lrc,
+						  struct xe_hw_engine *hwe,
+						  u32 *batch, size_t max_len)
+{
+	struct xe_device *xe = gt_to_xe(lrc->gt);
+	const u32 *user_batch;
+	u32 *cmd = batch;
+	u32 count;
+
+	count = xe_configfs_get_ctx_restore_post_bb(to_pci_dev(xe->drm.dev),
+						    hwe->class, &user_batch);
+	if (!count)
+		return 0;
+
+	if (count > max_len)
+		return -ENOSPC;
+
+	/*
+	 * This should be used only for tests and validation. Taint the kernel
+	 * as anything could be submitted directly in context switches
+	 */
+	add_taint(TAINT_TEST, LOCKDEP_STILL_OK);
+
+	memcpy(cmd, user_batch, count * sizeof(u32));
+	cmd += count;
+
+	return cmd - batch;
+}
+
 static ssize_t setup_invalidate_state_cache_wa(struct xe_lrc *lrc,
 					       struct xe_hw_engine *hwe,
 					       u32 *batch, size_t max_len)
@@ -1203,6 +1234,7 @@ int xe_lrc_setup_wa_bb_with_scratch(struct xe_lrc *lrc, struct xe_hw_engine *hwe
 		{ .setup = setup_timestamp_wa },
 		{ .setup = setup_invalidate_state_cache_wa },
 		{ .setup = setup_utilization_wa },
+		{ .setup = setup_configfs_post_ctx_restore_bb },
 	};
 	struct bo_setup_state state = {
 		.lrc = lrc,

From 39ac06f70062b6867836ea631e196cf2f60c1513 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:41 -0700
Subject: [PATCH 1380/1455] drm/xe/configfs: Add post context restore bb

Allow the user to specify commands to execute during a context restore.
Currently it's possible to parse 2 types of actions:

	- cmd: the instructions are added as is to the bb
	- reg: just use the address and value, without worrying about
	  encoding the right LRI instruction. This is possibly the most
	  useful use case, so added a dedicated action for that.

This also prepares for future BBs: mid context restore and rc6 context
restore that can re-use the same parsing functions.

Reviewed-by: Raag Jadav <raag.jadav@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-4-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 280 ++++++++++++++++++++++++++++++-
 1 file changed, 278 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index d86c75af0327..9a30dc958c35 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -4,6 +4,7 @@
  */
 
 #include <linux/bitops.h>
+#include <linux/ctype.h>
 #include <linux/configfs.h>
 #include <linux/cleanup.h>
 #include <linux/find.h>
@@ -12,6 +13,7 @@
 #include <linux/pci.h>
 #include <linux/string.h>
 
+#include "instructions/xe_mi_commands.h"
 #include "xe_configfs.h"
 #include "xe_hw_engine_types.h"
 #include "xe_module.h"
@@ -115,6 +117,37 @@
  *
  * This attribute can only be set before binding to the device.
  *
+ * Context restore BB
+ * ------------------
+ *
+ * Allow to execute a batch buffer during any context switches. When the
+ * GPU is restoring the context, it executes additional commands. It's useful
+ * for testing additional workarounds and validating certain HW behaviors: it's
+ * not intended for normal execution and will taint the kernel with TAINT_TEST
+ * when used.
+ *
+ * Currently this is implemented only for post context restore. Examples:
+ *
+ * #. Execute a LRI command to write 0xDEADBEEF to register 0x4f10::
+ *
+ *	# echo 'rcs cmd 11000001 4F100 DEADBEEF' \
+ *		> /sys/kernel/config/xe/0000:03:00.0/ctx_restore_post_bb
+ *
+ * #. Load certain values in a couple of registers (it can be used as a simpler
+ *    alternative to the `cmd`) action::
+ *
+ *	# cat > /sys/kernel/config/xe/0000:03:00.0/ctx_restore_post_bb <<EOF
+ *	rcs reg 4F100 DEADBEEF
+ *	rcs reg 4F104 FFFFFFFF
+ *	EOF
+ *
+ *    .. note::
+ *
+ *       When using multiple lines, make sure to use a command that is
+ *       implemented with a single write syscall, like HEREDOC.
+ *
+ * This attribute can only be set before binding to the device.
+ *
  * Remove devices
  * ==============
  *
@@ -123,11 +156,18 @@
  *	# rmdir /sys/kernel/config/xe/0000:03:00.0/
  */
 
+/* Similar to struct xe_bb, but not tied to HW (yet) */
+struct wa_bb {
+	u32 *cs;
+	u32 len; /* in dwords */
+};
+
 struct xe_config_group_device {
 	struct config_group group;
 
 	struct xe_config_device {
 		u64 engines_allowed;
+		struct wa_bb ctx_restore_post_bb[XE_ENGINE_CLASS_MAX];
 		bool survivability_mode;
 		bool enable_psmi;
 	} config;
@@ -371,11 +411,233 @@ static ssize_t enable_psmi_store(struct config_item *item, const char *page, siz
 	return len;
 }
 
+static bool wa_bb_read_advance(bool dereference, char **p,
+			       const char *append, size_t len,
+			       size_t *max_size)
+{
+	if (dereference) {
+		if (len >= *max_size)
+			return false;
+		*max_size -= len;
+		if (append)
+			memcpy(*p, append, len);
+	}
+
+	*p += len;
+
+	return true;
+}
+
+static ssize_t wa_bb_show(struct xe_config_group_device *dev,
+			  struct wa_bb wa_bb[static XE_ENGINE_CLASS_MAX],
+			  char *data, size_t sz)
+{
+	char *p = data;
+
+	guard(mutex)(&dev->lock);
+
+	for (size_t i = 0; i < ARRAY_SIZE(engine_info); i++) {
+		enum xe_engine_class ec = engine_info[i].engine_class;
+		size_t len;
+
+		if (!wa_bb[ec].len)
+			continue;
+
+		len = snprintf(p, sz, "%s:", engine_info[i].cls);
+		if (!wa_bb_read_advance(data, &p, NULL, len, &sz))
+			return -ENOBUFS;
+
+		for (size_t j = 0; j < wa_bb[ec].len; j++) {
+			len = snprintf(p, sz, " %08x", wa_bb[ec].cs[j]);
+			if (!wa_bb_read_advance(data, &p, NULL, len, &sz))
+				return -ENOBUFS;
+		}
+
+		if (!wa_bb_read_advance(data, &p, "\n", 1, &sz))
+			return -ENOBUFS;
+	}
+
+	if (!wa_bb_read_advance(data, &p, "", 1, &sz))
+		return -ENOBUFS;
+
+	/* Reserve one more to match check for '\0' */
+	if (!data)
+		p++;
+
+	return p - data;
+}
+
+static ssize_t ctx_restore_post_bb_show(struct config_item *item, char *page)
+{
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+
+	return wa_bb_show(dev, dev->config.ctx_restore_post_bb, page, SZ_4K);
+}
+
+static void wa_bb_append(struct wa_bb *wa_bb, u32 val)
+{
+	if (wa_bb->cs)
+		wa_bb->cs[wa_bb->len] = val;
+
+	wa_bb->len++;
+}
+
+static ssize_t parse_hex(const char *line, u32 *pval)
+{
+	char numstr[12];
+	const char *p;
+	ssize_t numlen;
+
+	p = line + strspn(line, " \t");
+	if (!*p || *p == '\n')
+		return 0;
+
+	numlen = strcspn(p, " \t\n");
+	if (!numlen || numlen >= sizeof(numstr) - 1)
+		return -EINVAL;
+
+	memcpy(numstr, p, numlen);
+	numstr[numlen] = '\0';
+	p += numlen;
+
+	if (kstrtou32(numstr, 16, pval))
+		return -EINVAL;
+
+	return p - line;
+}
+
+/*
+ * Parse lines with the format
+ *
+ *	<engine-class> cmd <u32> <u32...>
+ *	<engine-class> reg <u32_addr> <u32_val>
+ *
+ * and optionally save them in @wa_bb[i].cs is non-NULL.
+ *
+ * Return the number of dwords parsed.
+ */
+static ssize_t parse_wa_bb_lines(const char *lines,
+				 struct wa_bb wa_bb[static XE_ENGINE_CLASS_MAX])
+{
+	ssize_t dwords = 0, ret;
+	const char *p;
+
+	for (p = lines; *p; p++) {
+		const struct engine_info *info = NULL;
+		u32 val, val2;
+
+		/* Also allow empty lines */
+		p += strspn(p, " \t\n");
+		if (!*p)
+			break;
+
+		ret = parse_engine(p, " \t\n", NULL, &info);
+		if (ret < 0)
+			return ret;
+
+		p += ret;
+		p += strspn(p, " \t");
+
+		if (str_has_prefix(p, "cmd")) {
+			for (p += strlen("cmd"); *p;) {
+				ret = parse_hex(p, &val);
+				if (ret < 0)
+					return -EINVAL;
+				if (!ret)
+					break;
+
+				p += ret;
+				dwords++;
+				wa_bb_append(&wa_bb[info->engine_class], val);
+			}
+		} else if (str_has_prefix(p, "reg")) {
+			p += strlen("reg");
+			ret = parse_hex(p, &val);
+			if (ret <= 0)
+				return -EINVAL;
+
+			p += ret;
+			ret = parse_hex(p, &val2);
+			if (ret <= 0)
+				return -EINVAL;
+
+			p += ret;
+			dwords += 3;
+			wa_bb_append(&wa_bb[info->engine_class],
+				     MI_LOAD_REGISTER_IMM | MI_LRI_NUM_REGS(1));
+			wa_bb_append(&wa_bb[info->engine_class], val);
+			wa_bb_append(&wa_bb[info->engine_class], val2);
+		} else {
+			return -EINVAL;
+		}
+	}
+
+	return dwords;
+}
+
+static ssize_t wa_bb_store(struct wa_bb wa_bb[static XE_ENGINE_CLASS_MAX],
+			   struct xe_config_group_device *dev,
+			   const char *page, size_t len)
+{
+	/* tmp_wa_bb must match wa_bb's size */
+	struct wa_bb tmp_wa_bb[XE_ENGINE_CLASS_MAX] = { };
+	ssize_t count, class;
+	u32 *tmp;
+
+	/* 1. Count dwords - wa_bb[i].cs is NULL for all classes */
+	count = parse_wa_bb_lines(page, tmp_wa_bb);
+	if (count < 0)
+		return count;
+
+	guard(mutex)(&dev->lock);
+
+	if (is_bound(dev))
+		return -EBUSY;
+
+	/*
+	 * 2. Allocate a u32 array and set the pointers to the right positions
+	 * according to the length of each class' wa_bb
+	 */
+	tmp = krealloc(wa_bb[0].cs, count * sizeof(u32), GFP_KERNEL);
+	if (!tmp)
+		return -ENOMEM;
+
+	if (!count) {
+		memset(wa_bb, 0, sizeof(tmp_wa_bb));
+		return len;
+	}
+
+	for (class = 0, count = 0; class < XE_ENGINE_CLASS_MAX; ++class) {
+		tmp_wa_bb[class].cs = tmp + count;
+		count += tmp_wa_bb[class].len;
+		tmp_wa_bb[class].len = 0;
+	}
+
+	/* 3. Parse wa_bb lines again, this time saving the values */
+	count = parse_wa_bb_lines(page, tmp_wa_bb);
+	if (count < 0)
+		return count;
+
+	memcpy(wa_bb, tmp_wa_bb, sizeof(tmp_wa_bb));
+
+	return len;
+}
+
+static ssize_t ctx_restore_post_bb_store(struct config_item *item,
+					 const char *data, size_t sz)
+{
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+
+	return wa_bb_store(dev->config.ctx_restore_post_bb, dev, data, sz);
+}
+
+CONFIGFS_ATTR(, ctx_restore_post_bb);
 CONFIGFS_ATTR(, enable_psmi);
 CONFIGFS_ATTR(, engines_allowed);
 CONFIGFS_ATTR(, survivability_mode);
 
 static struct configfs_attribute *xe_config_device_attrs[] = {
+	&attr_ctx_restore_post_bb,
 	&attr_enable_psmi,
 	&attr_engines_allowed,
 	&attr_survivability_mode,
@@ -387,6 +649,8 @@ static void xe_config_device_release(struct config_item *item)
 	struct xe_config_group_device *dev = to_xe_config_group_device(item);
 
 	mutex_destroy(&dev->lock);
+
+	kfree(dev->config.ctx_restore_post_bb[0].cs);
 	kfree(dev);
 }
 
@@ -636,14 +900,26 @@ bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev)
 /**
  * xe_configfs_get_ctx_restore_post_bb - get configfs ctx_restore_post_bb setting
  * @pdev: pci device
+ * @class: hw engine class
+ * @cs: pointer to the bb to use - only valid during probe
  *
- * Return: post_ctx_restore setting in configfs
+ * Return: Number of dwords used in the post_ctx_restore setting in configfs
  */
 u32 xe_configfs_get_ctx_restore_post_bb(struct pci_dev *pdev,
 					enum xe_engine_class class,
 					const u32 **cs)
 {
-	return 0;
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
+	u32 len;
+
+	if (!dev)
+		return 0;
+
+	*cs = dev->config.ctx_restore_post_bb[class].cs;
+	len = dev->config.ctx_restore_post_bb[class].len;
+	config_group_put(&dev->group);
+
+	return len;
 }
 
 int __init xe_configfs_init(void)

From c9dfd66cb91ef32f76e51f75e315c07907df2b85 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:42 -0700
Subject: [PATCH 1381/1455] drm/xe/lrc: Allow INDIRECT_CTX for more engine
 classes

Currently it's only allowed for render and compute. Going forward we
want to enable it for more engine classes. Let the XE_LRC_FLAG_INDIRECT_CTX
flag (and thus gt_engine_needs_indirect_ctx()) be the deciding factor
for its availability.

While at it, add the missing const to rcs_funcs array. Since
CTX_INDIRECT_CTX_OFFSET_DEFAULT already matches the HW default and
gt_engine_needs_indirect_ctx() only ever enables it for rcs/ccs, there
is no change in behavior, it's only preparation for future use case.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-5-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/regs/xe_lrc_layout.h |  3 ---
 drivers/gpu/drm/xe/xe_lrc.c             | 14 ++++++++++----
 2 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/xe/regs/xe_lrc_layout.h b/drivers/gpu/drm/xe/regs/xe_lrc_layout.h
index 1b101edb838b..b5eff383902c 100644
--- a/drivers/gpu/drm/xe/regs/xe_lrc_layout.h
+++ b/drivers/gpu/drm/xe/regs/xe_lrc_layout.h
@@ -40,7 +40,4 @@
 #define INDIRECT_CTX_RING_START_UDW	(0x08 + 1)
 #define INDIRECT_CTX_RING_CTL		(0x0a + 1)
 
-#define CTX_INDIRECT_CTX_OFFSET_MASK	REG_GENMASK(15, 6)
-#define CTX_INDIRECT_CTX_OFFSET_DEFAULT	REG_FIELD_PREP(CTX_INDIRECT_CTX_OFFSET_MASK, 0xd)
-
 #endif
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index c706585611d5..0ab99c210d88 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -1281,9 +1281,11 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 static int
 setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 {
-	static struct bo_setup rcs_funcs[] = {
+	static const struct bo_setup rcs_funcs[] = {
 		{ .setup = setup_timestamp_wa },
 	};
+	static const struct bo_setup xcs_funcs[] = {
+	};
 	struct bo_setup_state state = {
 		.lrc = lrc,
 		.hwe = hwe,
@@ -1300,6 +1302,9 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 	    hwe->class == XE_ENGINE_CLASS_COMPUTE) {
 		state.funcs = rcs_funcs;
 		state.num_funcs = ARRAY_SIZE(rcs_funcs);
+	} else {
+		state.funcs = xcs_funcs;
+		state.num_funcs = ARRAY_SIZE(xcs_funcs);
 	}
 
 	if (xe_gt_WARN_ON(lrc->gt, !state.funcs))
@@ -1326,14 +1331,15 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 	finish_bo(&state);
 	kfree(state.buffer);
 
+	/*
+	 * Enable INDIRECT_CTX leaving INDIRECT_CTX_OFFSET at its default: it
+	 * varies per engine class, but the default is good enough
+	 */
 	xe_lrc_write_ctx_reg(lrc,
 			     CTX_CS_INDIRECT_CTX,
 			     (xe_bo_ggtt_addr(lrc->bo) + state.offset) |
 			     /* Size in CLs. */
 			     (state.written * sizeof(u32) / 64));
-	xe_lrc_write_ctx_reg(lrc,
-			     CTX_CS_INDIRECT_CTX_OFFSET,
-			     CTX_INDIRECT_CTX_OFFSET_DEFAULT);
 
 	return 0;
 }

From 7a4756b2fd0431f109b07b904a7442f00838abf9 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:43 -0700
Subject: [PATCH 1382/1455] drm/xe/lrc: Allow to add user commands mid context
 switch

Like done for post-context-restore commands, allow to add commands from
configfs in the middle of context restore. Since currently the indirect
ctx hardcodes the offset to CTX_INDIRECT_CTX_OFFSET_DEFAULT, this is
executed in the very beginning of engine context restore.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-6-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 15 +++++++++++++
 drivers/gpu/drm/xe/xe_configfs.h |  4 ++++
 drivers/gpu/drm/xe/xe_lrc.c      | 37 ++++++++++++++++++++++++++++++++
 3 files changed, 56 insertions(+)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 9a30dc958c35..90bc5a4b5da7 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -897,6 +897,21 @@ bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev)
 	return ret;
 }
 
+/**
+ * xe_configfs_get_ctx_restore_mid_bb - get configfs ctx_restore_mid_bb setting
+ * @pdev: pci device
+ * @class: hw engine class
+ * @cs: pointer to the bb to use - only valid during probe
+ *
+ * Return: Number of dwords used in the mid_ctx_restore setting in configfs
+ */
+u32 xe_configfs_get_ctx_restore_mid_bb(struct pci_dev *pdev,
+				       enum xe_engine_class class,
+				       const u32 **cs)
+{
+	return 0;
+}
+
 /**
  * xe_configfs_get_ctx_restore_post_bb - get configfs ctx_restore_post_bb setting
  * @pdev: pci device
diff --git a/drivers/gpu/drm/xe/xe_configfs.h b/drivers/gpu/drm/xe/xe_configfs.h
index eff2645b5f59..c61e0e47ed94 100644
--- a/drivers/gpu/drm/xe/xe_configfs.h
+++ b/drivers/gpu/drm/xe/xe_configfs.h
@@ -19,6 +19,8 @@ void xe_configfs_check_device(struct pci_dev *pdev);
 bool xe_configfs_get_survivability_mode(struct pci_dev *pdev);
 u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev);
 bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev);
+u32 xe_configfs_get_ctx_restore_mid_bb(struct pci_dev *pdev, enum xe_engine_class,
+				       const u32 **cs);
 u32 xe_configfs_get_ctx_restore_post_bb(struct pci_dev *pdev, enum xe_engine_class,
 					const u32 **cs);
 #else
@@ -28,6 +30,8 @@ static inline void xe_configfs_check_device(struct pci_dev *pdev) { }
 static inline bool xe_configfs_get_survivability_mode(struct pci_dev *pdev) { return false; }
 static inline u64 xe_configfs_get_engines_allowed(struct pci_dev *pdev) { return U64_MAX; }
 static inline bool xe_configfs_get_psmi_enabled(struct pci_dev *pdev) { return false; }
+static inline u32 xe_configfs_get_ctx_restore_mid_bb(struct pci_dev *pdev, enum xe_engine_class,
+						     const u32 **cs) { return 0; }
 static inline u32 xe_configfs_get_ctx_restore_post_bb(struct pci_dev *pdev, enum xe_engine_class,
 						      const u32 **cs) { return 0; }
 #endif
diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
index 0ab99c210d88..47e9df775072 100644
--- a/drivers/gpu/drm/xe/xe_lrc.c
+++ b/drivers/gpu/drm/xe/xe_lrc.c
@@ -77,11 +77,17 @@ lrc_to_xe(struct xe_lrc *lrc)
 static bool
 gt_engine_needs_indirect_ctx(struct xe_gt *gt, enum xe_engine_class class)
 {
+	struct xe_device *xe = gt_to_xe(gt);
+
 	if (XE_GT_WA(gt, 16010904313) &&
 	    (class == XE_ENGINE_CLASS_RENDER ||
 	     class == XE_ENGINE_CLASS_COMPUTE))
 		return true;
 
+	if (xe_configfs_get_ctx_restore_mid_bb(to_pci_dev(xe->drm.dev),
+					       class, NULL))
+		return true;
+
 	return false;
 }
 
@@ -1133,6 +1139,35 @@ static ssize_t setup_configfs_post_ctx_restore_bb(struct xe_lrc *lrc,
 	return cmd - batch;
 }
 
+static ssize_t setup_configfs_mid_ctx_restore_bb(struct xe_lrc *lrc,
+						 struct xe_hw_engine *hwe,
+						 u32 *batch, size_t max_len)
+{
+	struct xe_device *xe = gt_to_xe(lrc->gt);
+	const u32 *user_batch;
+	u32 *cmd = batch;
+	u32 count;
+
+	count = xe_configfs_get_ctx_restore_mid_bb(to_pci_dev(xe->drm.dev),
+						   hwe->class, &user_batch);
+	if (!count)
+		return 0;
+
+	if (count > max_len)
+		return -ENOSPC;
+
+	/*
+	 * This should be used only for tests and validation. Taint the kernel
+	 * as anything could be submitted directly in context switches
+	 */
+	add_taint(TAINT_TEST, LOCKDEP_STILL_OK);
+
+	memcpy(cmd, user_batch, count * sizeof(u32));
+	cmd += count;
+
+	return cmd - batch;
+}
+
 static ssize_t setup_invalidate_state_cache_wa(struct xe_lrc *lrc,
 					       struct xe_hw_engine *hwe,
 					       u32 *batch, size_t max_len)
@@ -1283,8 +1318,10 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
 {
 	static const struct bo_setup rcs_funcs[] = {
 		{ .setup = setup_timestamp_wa },
+		{ .setup = setup_configfs_mid_ctx_restore_bb },
 	};
 	static const struct bo_setup xcs_funcs[] = {
+		{ .setup = setup_configfs_mid_ctx_restore_bb },
 	};
 	struct bo_setup_state state = {
 		.lrc = lrc,

From b30d5de3d40c3fa642079bac0d91f17091c5f877 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Tue, 16 Sep 2025 14:15:44 -0700
Subject: [PATCH 1383/1455] drm/xe/configfs: Add mid context restore bb

Like done for post context restore, allow the user to add commands to
the middle of context restore, at the beginning of engine restore
commands.

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/20250916-wa-bb-cmds-v5-7-306bddbc15da@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 46 +++++++++++++++++++++++++++++---
 1 file changed, 42 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 90bc5a4b5da7..640df5bee457 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -126,13 +126,21 @@
  * not intended for normal execution and will taint the kernel with TAINT_TEST
  * when used.
  *
- * Currently this is implemented only for post context restore. Examples:
+ * Currently this is implemented only for post and mid context restore.
+ * Examples:
  *
- * #. Execute a LRI command to write 0xDEADBEEF to register 0x4f10::
+ * #. Execute a LRI command to write 0xDEADBEEF to register 0x4f10 after the
+ *    normal context restore::
  *
  *	# echo 'rcs cmd 11000001 4F100 DEADBEEF' \
  *		> /sys/kernel/config/xe/0000:03:00.0/ctx_restore_post_bb
  *
+ * #. Execute a LRI command to write 0xDEADBEEF to register 0x4f10 at the
+ *    beginning of the context restore::
+ *
+ *	# echo 'rcs cmd 11000001 4F100 DEADBEEF' \
+ *		> /sys/kernel/config/xe/0000:03:00.0/ctx_restore_mid_bb
+
  * #. Load certain values in a couple of registers (it can be used as a simpler
  *    alternative to the `cmd`) action::
  *
@@ -146,7 +154,7 @@
  *       When using multiple lines, make sure to use a command that is
  *       implemented with a single write syscall, like HEREDOC.
  *
- * This attribute can only be set before binding to the device.
+ * These attributes can only be set before binding to the device.
  *
  * Remove devices
  * ==============
@@ -168,6 +176,7 @@ struct xe_config_group_device {
 	struct xe_config_device {
 		u64 engines_allowed;
 		struct wa_bb ctx_restore_post_bb[XE_ENGINE_CLASS_MAX];
+		struct wa_bb ctx_restore_mid_bb[XE_ENGINE_CLASS_MAX];
 		bool survivability_mode;
 		bool enable_psmi;
 	} config;
@@ -467,6 +476,13 @@ static ssize_t wa_bb_show(struct xe_config_group_device *dev,
 	return p - data;
 }
 
+static ssize_t ctx_restore_mid_bb_show(struct config_item *item, char *page)
+{
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+
+	return wa_bb_show(dev, dev->config.ctx_restore_mid_bb, page, SZ_4K);
+}
+
 static ssize_t ctx_restore_post_bb_show(struct config_item *item, char *page)
 {
 	struct xe_config_group_device *dev = to_xe_config_group_device(item);
@@ -623,6 +639,14 @@ static ssize_t wa_bb_store(struct wa_bb wa_bb[static XE_ENGINE_CLASS_MAX],
 	return len;
 }
 
+static ssize_t ctx_restore_mid_bb_store(struct config_item *item,
+					const char *data, size_t sz)
+{
+	struct xe_config_group_device *dev = to_xe_config_group_device(item);
+
+	return wa_bb_store(dev->config.ctx_restore_mid_bb, dev, data, sz);
+}
+
 static ssize_t ctx_restore_post_bb_store(struct config_item *item,
 					 const char *data, size_t sz)
 {
@@ -631,12 +655,14 @@ static ssize_t ctx_restore_post_bb_store(struct config_item *item,
 	return wa_bb_store(dev->config.ctx_restore_post_bb, dev, data, sz);
 }
 
+CONFIGFS_ATTR(, ctx_restore_mid_bb);
 CONFIGFS_ATTR(, ctx_restore_post_bb);
 CONFIGFS_ATTR(, enable_psmi);
 CONFIGFS_ATTR(, engines_allowed);
 CONFIGFS_ATTR(, survivability_mode);
 
 static struct configfs_attribute *xe_config_device_attrs[] = {
+	&attr_ctx_restore_mid_bb,
 	&attr_ctx_restore_post_bb,
 	&attr_enable_psmi,
 	&attr_engines_allowed,
@@ -909,7 +935,19 @@ u32 xe_configfs_get_ctx_restore_mid_bb(struct pci_dev *pdev,
 				       enum xe_engine_class class,
 				       const u32 **cs)
 {
-	return 0;
+	struct xe_config_group_device *dev = find_xe_config_group_device(pdev);
+	u32 len;
+
+	if (!dev)
+		return 0;
+
+	if (cs)
+		*cs = dev->config.ctx_restore_mid_bb[class].cs;
+
+	len = dev->config.ctx_restore_mid_bb[class].len;
+	config_group_put(&dev->group);
+
+	return len;
 }
 
 /**

From 3e6339a19cfc93bd3fec2179a2e766ab1cf39e16 Mon Sep 17 00:00:00 2001
From: Liu Ying <victor.liu@nxp.com>
Date: Wed, 6 Aug 2025 16:41:21 +0800
Subject: [PATCH 1384/1455] drm/bridge: waveshare-dsi: Fix bailout for
 devm_drm_bridge_alloc()

devm_drm_bridge_alloc() returns ERR_PTR on failure instead of a
NULL pointer, so use IS_ERR() to check the returned pointer and
turn proper error code on failure by using PTR_ERR().

Fixes: dbdea37add13 ("drm: bridge: Add waveshare DSI2DPI unit driver")
Signed-off-by: Liu Ying <victor.liu@nxp.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250806084121.510207-1-victor.liu@nxp.com
---
 drivers/gpu/drm/bridge/waveshare-dsi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/bridge/waveshare-dsi.c b/drivers/gpu/drm/bridge/waveshare-dsi.c
index 01c70e7d3d3b..43f4e7412d72 100644
--- a/drivers/gpu/drm/bridge/waveshare-dsi.c
+++ b/drivers/gpu/drm/bridge/waveshare-dsi.c
@@ -147,8 +147,8 @@ static int ws_bridge_probe(struct i2c_client *i2c)
 	int ret;
 
 	ws = devm_drm_bridge_alloc(dev, struct ws_bridge, bridge, &ws_bridge_bridge_funcs);
-	if (!ws)
-		return -ENOMEM;
+	if (IS_ERR(ws))
+		return PTR_ERR(ws);
 
 	ws->dev = dev;
 

From d9b2623319fa20c2206754284291817488329648 Mon Sep 17 00:00:00 2001
From: Lucas De Marchi <lucas.demarchi@intel.com>
Date: Fri, 12 Sep 2025 14:54:51 -0700
Subject: [PATCH 1385/1455] drm/xe: Fix build with CONFIG_MODULES=n

When building with CONFIG_MODULES=n, the __exit functions are dropped.
However our init functions may call them for error handling, so they are
not good candidates for the exit sections.

Fix this error reported by 0day:

	ld.lld: error: relocation refers to a symbol in a discarded section: xe_configfs_exit
	>>> defined in vmlinux.a(drivers/gpu/drm/xe/xe_configfs.o)
	>>> referenced by xe_module.c
	>>>               drivers/gpu/drm/xe/xe_module.o:(init_funcs) in archive vmlinux.a

This is the only exit function using __exit. Drop it to fix the build.

Cc: Riana Tauro <riana.tauro@intel.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202506092221.1FmUQmI8-lkp@intel.com/
Fixes: 16280ded45fb ("drm/xe: Add configfs to enable survivability mode")
Reviewed-by: Balasubramani Vivekanandan <balasubramani.vivekanandan@intel.com>
Link: https://lore.kernel.org/r/20250912-fix-nomodule-build-v1-1-d11b70a92516@intel.com
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
---
 drivers/gpu/drm/xe/xe_configfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/xe/xe_configfs.c b/drivers/gpu/drm/xe/xe_configfs.c
index 640df5bee457..8a9b950e7a6d 100644
--- a/drivers/gpu/drm/xe/xe_configfs.c
+++ b/drivers/gpu/drm/xe/xe_configfs.c
@@ -990,7 +990,7 @@ int __init xe_configfs_init(void)
 	return 0;
 }
 
-void __exit xe_configfs_exit(void)
+void xe_configfs_exit(void)
 {
 	configfs_unregister_subsystem(&xe_configfs);
 	mutex_destroy(&xe_configfs.su_mutex);

From 0c4932f6ddf815618fa34f7403df682aed7862b5 Mon Sep 17 00:00:00 2001
From: LiangCheng Wang <zaq14760@gmail.com>
Date: Mon, 22 Sep 2025 10:57:34 +0800
Subject: [PATCH 1386/1455] drm/tiny: pixpaper: Fix missing dependency on
 DRM_GEM_SHMEM_HELPER

The driver uses drm_gem_shmem_prime_import_no_map() and
drm_gem_shmem_dumb_create(), but the Kconfig currently selects
DRM_GEM_DMA_HELPER instead of DRM_GEM_SHMEM_HELPER. This causes
link failures when DRM_GEM_SHMEM_HELPER is not enabled.

Select DRM_GEM_SHMEM_HELPER to fix the build.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202509220320.gfFZjmyg-lkp@intel.com/
Fixes: c9e70639f591 ("drm: tiny: Add support for Mayqueen Pixpaper e-ink panel")
Reviewed-by: Thomas Zimmermann <tzimmermann@suse.de>
Signed-off-by: LiangCheng Wang <zaq14760@gmail.com>
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://lore.kernel.org/r/20250922-bar-v1-1-b2a1f54ace82@gmail.com
---
 drivers/gpu/drm/tiny/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/tiny/Kconfig b/drivers/gpu/drm/tiny/Kconfig
index 94a5bf61a115..7d9e85e932d7 100644
--- a/drivers/gpu/drm/tiny/Kconfig
+++ b/drivers/gpu/drm/tiny/Kconfig
@@ -86,7 +86,7 @@ config DRM_PIXPAPER
         tristate "DRM support for PIXPAPER display panels"
         depends on DRM && SPI
         select DRM_CLIENT_SELECTION
-        select DRM_GEM_DMA_HELPER
+        select DRM_GEM_SHMEM_HELPER
         select DRM_KMS_HELPER
         help
 	  DRM driver for the Mayqueen Pixpaper e-ink display panel.

From b29c22b8dafd951664a491bf629e615d81513197 Mon Sep 17 00:00:00 2001
From: Lijo Lazar <lijo.lazar@amd.com>
Date: Thu, 18 Sep 2025 17:52:04 +0530
Subject: [PATCH 1387/1455] drm/amdgpu: Fix vbios build number parsing logic

It's not necessary that the build string and atom header section has a
difference of 32 bytes. Use the remaining bytes in the section as copy
limit.

Fixes: d6fa80266178 ("drm/amdgpu: Add vbios build number interface")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/atom.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c
index be5d67c2c7a1..7a063e44d429 100644
--- a/drivers/gpu/drm/amd/amdgpu/atom.c
+++ b/drivers/gpu/drm/amd/amdgpu/atom.c
@@ -1502,7 +1502,7 @@ static void atom_get_vbios_build(struct atom_context *ctx)
 {
 	unsigned char *atom_rom_hdr;
 	unsigned char *str;
-	uint16_t base;
+	uint16_t base, len;
 
 	base = CU16(ATOM_ROM_TABLE_PTR);
 	atom_rom_hdr = CSTR(base);
@@ -1515,8 +1515,9 @@ static void atom_get_vbios_build(struct atom_context *ctx)
 	while (str < atom_rom_hdr && *str++)
 		;
 
-	if ((str + STRLEN_NORMAL) < atom_rom_hdr)
-		strscpy(ctx->build_num, str, STRLEN_NORMAL);
+	len = min(atom_rom_hdr - str, STRLEN_NORMAL);
+	if (len)
+		strscpy(ctx->build_num, str, len);
 }
 
 struct atom_context *amdgpu_atom_parse(struct card_info *card, void *bios)

From ae4d627e43ccecc403e7378811289b33de38e67d Mon Sep 17 00:00:00 2001
From: Yang Wang <kevinyang.wang@amd.com>
Date: Fri, 19 Sep 2025 09:44:25 +0800
Subject: [PATCH 1388/1455] drm/amd/pm: place the smu 13.0.0 pptable header
 into the correct folder

Place the smu 13.0.0 pptable header in the correct folder

Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Mangesh Gadre <mangesh.gadre@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/{ => swsmu}/inc/smu_v13_0_0_pptable.h | 0
 1 file changed, 0 insertions(+), 0 deletions(-)
 rename drivers/gpu/drm/amd/pm/{ => swsmu}/inc/smu_v13_0_0_pptable.h (100%)

diff --git a/drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_0_pptable.h
similarity index 100%
rename from drivers/gpu/drm/amd/pm/inc/smu_v13_0_0_pptable.h
rename to drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0_0_pptable.h

From c5b3cc417b0260abc74ed32f6baa626c9de917c0 Mon Sep 17 00:00:00 2001
From: Sunil Khatri <sunil.khatri@amd.com>
Date: Wed, 17 Sep 2025 20:12:43 +0530
Subject: [PATCH 1389/1455] drm/amdgpu: use hmm_pfns instead of array of pages
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

we dont need to allocate local array of pages to hold
the pages returned by the hmm, instead we could use
the hmm_range structure itself to get to hmm_pfn
and get the required pages directly.

This avoids call to alloc/free quite a lot.

Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Suggested-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Acked-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c  | 10 +++++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h   |  1 -
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c        | 30 ++++---------------
 drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c       |  5 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c       | 11 +------
 drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.h       |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c       |  8 ++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h       |  5 ++--
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c          |  2 +-
 9 files changed, 25 insertions(+), 49 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
index c3b34a410375..7c54fe6b0f5d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c
@@ -1089,7 +1089,7 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr,
 		return 0;
 	}
 
-	ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages, &range);
+	ret = amdgpu_ttm_tt_get_user_pages(bo, &range);
 	if (ret) {
 		if (ret == -EAGAIN)
 			pr_debug("Failed to get user pages, try again\n");
@@ -1103,6 +1103,9 @@ static int init_user_pages(struct kgd_mem *mem, uint64_t user_addr,
 		pr_err("%s: Failed to reserve BO\n", __func__);
 		goto release_out;
 	}
+
+	amdgpu_ttm_tt_set_user_pages(bo->tbo.ttm, range);
+
 	amdgpu_bo_placement_from_domain(bo, mem->domain);
 	ret = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 	if (ret)
@@ -2565,8 +2568,7 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info,
 		}
 
 		/* Get updated user pages */
-		ret = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages,
-						   &mem->range);
+		ret = amdgpu_ttm_tt_get_user_pages(bo, &mem->range);
 		if (ret) {
 			pr_debug("Failed %d to get user pages\n", ret);
 
@@ -2595,6 +2597,8 @@ static int update_invalid_user_pages(struct amdkfd_process_info *process_info,
 			ret = 0;
 		}
 
+		amdgpu_ttm_tt_set_user_pages(bo->tbo.ttm, mem->range);
+
 		mutex_lock(&process_info->notifier_lock);
 
 		/* Mark the BO as valid unless it was invalidated
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h
index 555cd6d877c3..a716c9886c74 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.h
@@ -38,7 +38,6 @@ struct amdgpu_bo_list_entry {
 	struct amdgpu_bo		*bo;
 	struct amdgpu_bo_va		*bo_va;
 	uint32_t			priority;
-	struct page			**user_pages;
 	struct hmm_range		*range;
 	bool				user_invalidated;
 };
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index defb511acc5a..744e6ff69814 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -29,6 +29,7 @@
 #include <linux/pagemap.h>
 #include <linux/sync_file.h>
 #include <linux/dma-buf.h>
+#include <linux/hmm.h>
 
 #include <drm/amdgpu_drm.h>
 #include <drm/drm_syncobj.h>
@@ -885,24 +886,12 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 		struct amdgpu_bo *bo = e->bo;
 		int i;
 
-		e->user_pages = kvcalloc(bo->tbo.ttm->num_pages,
-					 sizeof(struct page *),
-					 GFP_KERNEL);
-		if (!e->user_pages) {
-			drm_err(adev_to_drm(p->adev), "kvmalloc_array failure\n");
-			r = -ENOMEM;
+		r = amdgpu_ttm_tt_get_user_pages(bo, &e->range);
+		if (r)
 			goto out_free_user_pages;
-		}
-
-		r = amdgpu_ttm_tt_get_user_pages(bo, e->user_pages, &e->range);
-		if (r) {
-			kvfree(e->user_pages);
-			e->user_pages = NULL;
-			goto out_free_user_pages;
-		}
 
 		for (i = 0; i < bo->tbo.ttm->num_pages; i++) {
-			if (bo->tbo.ttm->pages[i] != e->user_pages[i]) {
+			if (bo->tbo.ttm->pages[i] != hmm_pfn_to_page(e->range->hmm_pfns[i])) {
 				userpage_invalidated = true;
 				break;
 			}
@@ -946,7 +935,7 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 		}
 
 		if (amdgpu_ttm_tt_is_userptr(e->bo->tbo.ttm) &&
-		    e->user_invalidated && e->user_pages) {
+		    e->user_invalidated) {
 			amdgpu_bo_placement_from_domain(e->bo,
 							AMDGPU_GEM_DOMAIN_CPU);
 			r = ttm_bo_validate(&e->bo->tbo, &e->bo->placement,
@@ -955,11 +944,8 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 				goto out_free_user_pages;
 
 			amdgpu_ttm_tt_set_user_pages(e->bo->tbo.ttm,
-						     e->user_pages);
+						     e->range);
 		}
-
-		kvfree(e->user_pages);
-		e->user_pages = NULL;
 	}
 
 	amdgpu_cs_get_threshold_for_moves(p->adev, &p->bytes_moved_threshold,
@@ -1001,11 +987,7 @@ out_free_user_pages:
 	amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) {
 		struct amdgpu_bo *bo = e->bo;
 
-		if (!e->user_pages)
-			continue;
 		amdgpu_ttm_tt_get_user_pages_done(bo->tbo.ttm, e->range);
-		kvfree(e->user_pages);
-		e->user_pages = NULL;
 		e->range = NULL;
 	}
 	mutex_unlock(&p->bo_list->bo_list_mutex);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 1679c2c3d505..b7ebae289bea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -572,8 +572,7 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 		goto release_object;
 
 	if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) {
-		r = amdgpu_ttm_tt_get_user_pages(bo, bo->tbo.ttm->pages,
-						 &range);
+		r = amdgpu_ttm_tt_get_user_pages(bo, &range);
 		if (r)
 			goto release_object;
 
@@ -581,6 +580,8 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 		if (r)
 			goto user_pages_done;
 
+		amdgpu_ttm_tt_set_user_pages(bo->tbo.ttm, range);
+
 		amdgpu_bo_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT);
 		r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx);
 		amdgpu_bo_unreserve(bo);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c
index e36fede7f74c..2c6a6b858112 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c
@@ -167,13 +167,12 @@ void amdgpu_hmm_unregister(struct amdgpu_bo *bo)
 
 int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
 			       uint64_t start, uint64_t npages, bool readonly,
-			       void *owner, struct page **pages,
+			       void *owner,
 			       struct hmm_range **phmm_range)
 {
 	struct hmm_range *hmm_range;
 	unsigned long end;
 	unsigned long timeout;
-	unsigned long i;
 	unsigned long *pfns;
 	int r = 0;
 
@@ -222,14 +221,6 @@ retry:
 	hmm_range->start = start;
 	hmm_range->hmm_pfns = pfns;
 
-	/*
-	 * Due to default_flags, all pages are HMM_PFN_VALID or
-	 * hmm_range_fault() fails. FIXME: The pages cannot be touched outside
-	 * the notifier_lock, and mmu_interval_read_retry() must be done first.
-	 */
-	for (i = 0; pages && i < npages; i++)
-		pages[i] = hmm_pfn_to_page(pfns[i]);
-
 	*phmm_range = hmm_range;
 
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.h
index e2edcd010ccc..953e1d06de20 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.h
@@ -33,7 +33,7 @@
 
 int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier,
 			       uint64_t start, uint64_t npages, bool readonly,
-			       void *owner, struct page **pages,
+			       void *owner,
 			       struct hmm_range **phmm_range);
 bool amdgpu_hmm_range_get_pages_done(struct hmm_range *hmm_range);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index 71b6691edab4..aa9ee5dffa45 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -708,7 +708,7 @@ struct amdgpu_ttm_tt {
  * Calling function must call amdgpu_ttm_tt_userptr_range_done() once and only
  * once afterwards to stop HMM tracking
  */
-int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages,
+int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo,
 				 struct hmm_range **range)
 {
 	struct ttm_tt *ttm = bo->tbo.ttm;
@@ -745,7 +745,7 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages,
 
 	readonly = amdgpu_ttm_tt_is_readonly(ttm);
 	r = amdgpu_hmm_range_get_pages(&bo->notifier, start, ttm->num_pages,
-				       readonly, NULL, pages, range);
+				       readonly, NULL, range);
 out_unlock:
 	mmap_read_unlock(mm);
 	if (r)
@@ -797,12 +797,12 @@ bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm,
  * that backs user memory and will ultimately be mapped into the device
  * address space.
  */
-void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages)
+void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct hmm_range *range)
 {
 	unsigned long i;
 
 	for (i = 0; i < ttm->num_pages; ++i)
-		ttm->pages[i] = pages ? pages[i] : NULL;
+		ttm->pages[i] = range ? hmm_pfn_to_page(range->hmm_pfns[i]) : NULL;
 }
 
 /*
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index 1c5c7836ce4f..0be2728aa872 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -191,7 +191,7 @@ void amdgpu_ttm_recover_gart(struct ttm_buffer_object *tbo);
 uint64_t amdgpu_ttm_domain_start(struct amdgpu_device *adev, uint32_t type);
 
 #if IS_ENABLED(CONFIG_DRM_AMDGPU_USERPTR)
-int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages,
+int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo,
 				 struct hmm_range **range);
 void amdgpu_ttm_tt_discard_user_pages(struct ttm_tt *ttm,
 				      struct hmm_range *range);
@@ -199,7 +199,6 @@ bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm,
 				       struct hmm_range *range);
 #else
 static inline int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo,
-					       struct page **pages,
 					       struct hmm_range **range)
 {
 	return -EPERM;
@@ -215,7 +214,7 @@ static inline bool amdgpu_ttm_tt_get_user_pages_done(struct ttm_tt *ttm,
 }
 #endif
 
-void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct page **pages);
+void amdgpu_ttm_tt_set_user_pages(struct ttm_tt *ttm, struct hmm_range *range);
 int amdgpu_ttm_tt_get_userptr(const struct ttm_buffer_object *tbo,
 			      uint64_t *user_addr);
 int amdgpu_ttm_tt_set_userptr(struct ttm_buffer_object *bo,
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 68ba239b2e5d..273f42e3afdd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -1738,7 +1738,7 @@ static int svm_range_validate_and_map(struct mm_struct *mm,
 
 			WRITE_ONCE(p->svms.faulting_task, current);
 			r = amdgpu_hmm_range_get_pages(&prange->notifier, addr, npages,
-						       readonly, owner, NULL,
+						       readonly, owner,
 						       &hmm_range);
 			WRITE_ONCE(p->svms.faulting_task, NULL);
 			if (r)

From 1fb710793ce2619223adffaf981b1ff13cd48f17 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 18 Sep 2025 19:48:00 -0500
Subject: [PATCH 1390/1455] drm/amdgpu: Enable MES lr_compute_wa by default

The MES set resources packet has an optional bit 'lr_compute_wa'
which can be used for preventing MES hangs on long compute jobs.

Set this bit by default.

Co-developed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/mes_v11_0.c        | 6 ++++++
 drivers/gpu/drm/amd/amdgpu/mes_v12_0.c        | 5 +++++
 drivers/gpu/drm/amd/include/mes_v11_api_def.h | 3 ++-
 drivers/gpu/drm/amd/include/mes_v12_api_def.h | 3 ++-
 4 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
index 3b91ea601add..e82188431f79 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
@@ -713,6 +713,12 @@ static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes)
 	mes_set_hw_res_pkt.enable_reg_active_poll = 1;
 	mes_set_hw_res_pkt.enable_level_process_quantum_check = 1;
 	mes_set_hw_res_pkt.oversubscription_timer = 50;
+	if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x7f)
+		mes_set_hw_res_pkt.enable_lr_compute_wa = 1;
+	else
+		dev_info_once(mes->adev->dev,
+			      "MES FW version must be >= 0x7f to enable LR compute workaround.\n");
+
 	if (amdgpu_mes_log_enable) {
 		mes_set_hw_res_pkt.enable_mes_event_int_logging = 1;
 		mes_set_hw_res_pkt.event_intr_history_gpu_mc_ptr =
diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
index 998893dff08e..aff06f06aeee 100644
--- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
@@ -769,6 +769,11 @@ static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe)
 	mes_set_hw_res_pkt.use_different_vmid_compute = 1;
 	mes_set_hw_res_pkt.enable_reg_active_poll = 1;
 	mes_set_hw_res_pkt.enable_level_process_quantum_check = 1;
+	if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x82)
+		mes_set_hw_res_pkt.enable_lr_compute_wa = 1;
+	else
+		dev_info_once(adev->dev,
+			      "MES FW version must be >= 0x82 to enable LR compute workaround.\n");
 
 	/*
 	 * Keep oversubscribe timer for sdma . When we have unmapped doorbell
diff --git a/drivers/gpu/drm/amd/include/mes_v11_api_def.h b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
index 15680c3f4970..ab1cfc92dbeb 100644
--- a/drivers/gpu/drm/amd/include/mes_v11_api_def.h
+++ b/drivers/gpu/drm/amd/include/mes_v11_api_def.h
@@ -238,7 +238,8 @@ union MESAPI_SET_HW_RESOURCES {
 				uint32_t enable_mes_sch_stb_log : 1;
 				uint32_t limit_single_process : 1;
 				uint32_t is_strix_tmz_wa_enabled  :1;
-				uint32_t reserved : 13;
+				uint32_t enable_lr_compute_wa : 1;
+				uint32_t reserved : 12;
 			};
 			uint32_t	uint32_t_all;
 		};
diff --git a/drivers/gpu/drm/amd/include/mes_v12_api_def.h b/drivers/gpu/drm/amd/include/mes_v12_api_def.h
index c04bd351b250..69611c7e30e3 100644
--- a/drivers/gpu/drm/amd/include/mes_v12_api_def.h
+++ b/drivers/gpu/drm/amd/include/mes_v12_api_def.h
@@ -287,7 +287,8 @@ union MESAPI_SET_HW_RESOURCES {
 				uint32_t limit_single_process : 1;
 				uint32_t unmapped_doorbell_handling: 2;
 				uint32_t enable_mes_fence_int: 1;
-				uint32_t reserved : 10;
+				uint32_t enable_lr_compute_wa : 1;
+				uint32_t reserved : 9;
 			};
 			uint32_t uint32_all;
 		};

From 854b9ab637d72bad8afe8f11cf6edd9aab0d1cf9 Mon Sep 17 00:00:00 2001
From: Sonny Jiang <sonny.jiang@amd.com>
Date: Thu, 18 Sep 2025 12:34:48 -0400
Subject: [PATCH 1391/1455] drm/amdgpu: Update amdgpu_vcn5_fw_shared for
 vcn_5_0_1

Align vcn5_fw_shared structure with FW

Signed-off-by: Sonny Jiang <sonny.jiang@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index bebfc2b34afe..dc8a17bcc3c8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -501,7 +501,7 @@ struct amdgpu_vcn5_fw_shared {
 	struct amdgpu_fw_shared_rb_setup rb_setup;
 	struct amdgpu_fw_shared_smu_interface_info smu_dpm_interface;
 	struct amdgpu_fw_shared_drm_key_wa drm_key_wa;
-	uint8_t pad3[9];
+	uint8_t pad3[404];
 };
 
 #define VCN_BLOCK_ENCODE_DISABLE_MASK 0x80

From c82b8f96eb972eb313e20ede5eb6617bc393c490 Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Mon, 15 Sep 2025 11:22:52 -0400
Subject: [PATCH 1392/1455] drm/amd/display: Disable stutter when programming
 watermarks on dcn32

[WHY&HOW]
Reprogramming watermarks with stutter allowed can cause instability on
some ASICs. Disable it prior to raising watermarks (prepare bandwidth),
then re-enable after lowering (optimize bandwidth).

Reviewed-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h           |  1 +
 .../display/dc/hubbub/dcn32/dcn32_hubbub.c    | 37 ++++++++++++++++---
 .../dc/resource/dcn32/dcn32_resource.c        |  1 +
 .../dc/resource/dcn32/dcn32_resource.h        |  3 +-
 4 files changed, 36 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 9d15ba591772..62d132bf3028 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1163,6 +1163,7 @@ struct dc_debug_options {
 	unsigned int auxless_alpm_lfps_silence_ns;
 	unsigned int auxless_alpm_lfps_t1t2_us;
 	short auxless_alpm_lfps_t1t2_offset_us;
+	bool disable_stutter_for_wm_program;
 };
 
 
diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
index 92957398ac0a..3b71bfaca291 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
@@ -28,6 +28,7 @@
 #include "dcn32_hubbub.h"
 #include "dm_services.h"
 #include "reg_helper.h"
+#include "dal_asic_id.h"
 
 
 #define CTX \
@@ -72,6 +73,14 @@ static void dcn32_init_crb(struct hubbub *hubbub)
 	REG_UPDATE(DCHUBBUB_DEBUG_CTRL_0, DET_DEPTH, 0x47F);
 }
 
+static void hubbub32_set_sdp_control(struct hubbub *hubbub, bool dc_control)
+{
+	struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub);
+
+	REG_UPDATE(DCHUBBUB_SDPIF_CFG0,
+			SDPIF_PORT_CONTROL, dc_control);
+}
+
 void hubbub32_set_request_limit(struct hubbub *hubbub, int memory_channel_count, int words_per_channel)
 {
 	struct dcn20_hubbub *hubbub2 = TO_DCN20_HUBBUB(hubbub);
@@ -754,8 +763,17 @@ static bool hubbub32_program_watermarks(
 		unsigned int refclk_mhz,
 		bool safe_to_lower)
 {
+	struct dc *dc = hubbub->ctx->dc;
 	bool wm_pending = false;
 
+	if (!safe_to_lower && dc->debug.disable_stutter_for_wm_program &&
+			(ASICREV_IS_GC_11_0_0(dc->ctx->asic_id.hw_internal_rev) ||
+			ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev))) {
+		/* before raising watermarks, SDP control give to DF, stutter must be disabled */
+		hubbub32_set_sdp_control(hubbub, false);
+		hubbub1_allow_self_refresh_control(hubbub, false);
+	}
+
 	if (hubbub32_program_urgent_watermarks(hubbub, watermarks, refclk_mhz, safe_to_lower))
 		wm_pending = true;
 
@@ -786,10 +804,20 @@ static bool hubbub32_program_watermarks(
 	REG_UPDATE(DCHUBBUB_ARB_DF_REQ_OUTSTAND,
 			DCHUBBUB_ARB_MIN_REQ_OUTSTAND, 0x1FF);*/
 
-	if (safe_to_lower || hubbub->ctx->dc->debug.disable_stutter)
-		hubbub1_allow_self_refresh_control(hubbub, !hubbub->ctx->dc->debug.disable_stutter);
+	if (safe_to_lower) {
+		/* after lowering watermarks, stutter setting is restored, SDP control given to DC */
+		hubbub1_allow_self_refresh_control(hubbub, !dc->debug.disable_stutter);
 
-	hubbub32_force_usr_retraining_allow(hubbub, hubbub->ctx->dc->debug.force_usr_allow);
+		if (dc->debug.disable_stutter_for_wm_program &&
+				(ASICREV_IS_GC_11_0_0(dc->ctx->asic_id.hw_internal_rev) ||
+				ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev))) {
+			hubbub32_set_sdp_control(hubbub, true);
+		}
+	} else if (dc->debug.disable_stutter) {
+		hubbub1_allow_self_refresh_control(hubbub, !dc->debug.disable_stutter);
+	}
+
+	hubbub32_force_usr_retraining_allow(hubbub, dc->debug.force_usr_allow);
 
 	return wm_pending;
 }
@@ -974,8 +1002,7 @@ void hubbub32_init(struct hubbub *hubbub)
 	ignore the "df_pre_cstate_req" from the SDP port control.
 	only the DCN will determine when to connect the SDP port
 	*/
-	REG_UPDATE(DCHUBBUB_SDPIF_CFG0,
-			SDPIF_PORT_CONTROL, 1);
+	hubbub32_set_sdp_control(hubbub, true);
 	/*Set SDP's max outstanding request to 512
 	must set the register back to 0 (max outstanding = 256) in zero frame buffer mode*/
 	REG_UPDATE(DCHUBBUB_SDPIF_CFG1,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
index 8f80ccb846d7..3965a7f1b64b 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c
@@ -739,6 +739,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.fpo_vactive_min_active_margin_us = 200,
 	.fpo_vactive_max_blank_us = 1000,
 	.enable_legacy_fast_update = false,
+	.disable_stutter_for_wm_program = true
 };
 
 static struct dce_aux *dcn32_aux_engine_create(
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
index 20d714596021..99f0432288b4 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h
@@ -1230,7 +1230,8 @@ unsigned int dcn32_get_max_hw_cursor_size(const struct dc *dc,
       SR(DCHUBBUB_ARB_MALL_CNTL),                                              \
       SR(DCN_VM_FAULT_ADDR_MSB), SR(DCN_VM_FAULT_ADDR_LSB),                    \
       SR(DCN_VM_FAULT_CNTL), SR(DCN_VM_FAULT_STATUS),                          \
-      SR(SDPIF_REQUEST_RATE_LIMIT)
+      SR(SDPIF_REQUEST_RATE_LIMIT),                                            \
+      SR(DCHUBBUB_SDPIF_CFG0)
 
 /* DCCG */
 

From 54980f3c63ed3e5cca3d251416581193c90eae76 Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Mon, 15 Sep 2025 11:35:37 -0400
Subject: [PATCH 1393/1455] drm/amd/display: Add missing post flip calls

[WHY&HOW]
dc_post_update_surfaces_to_stream needs to be called after a full update
completes in order to optimize clocks and watermarks for power. Add
missing calls before idle entry is requested to ensure optimal power.

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c      | 3 +--
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c | 8 ++++++--
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 7012b2c692b3..09427e056055 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -417,8 +417,7 @@ static inline bool update_planes_and_stream_adapter(struct dc *dc,
 	/*
 	 * Previous frame finished and HW is ready for optimization.
 	 */
-	if (update_type == UPDATE_TYPE_FAST)
-		dc_post_update_surfaces_to_stream(dc);
+	dc_post_update_surfaces_to_stream(dc);
 
 	return dc_update_planes_and_stream(dc,
 					   array_of_surface_update,
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
index 466dccb355d7..1ec9d03ad747 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c
@@ -218,8 +218,10 @@ static void amdgpu_dm_idle_worker(struct work_struct *work)
 			break;
 		}
 
-		if (idle_work->enable)
+		if (idle_work->enable) {
+			dc_post_update_surfaces_to_stream(idle_work->dm->dc);
 			dc_allow_idle_optimizations(idle_work->dm->dc, true);
+		}
 		mutex_unlock(&idle_work->dm->dc_lock);
 	}
 	idle_work->dm->idle_workqueue->running = false;
@@ -273,8 +275,10 @@ static void amdgpu_dm_crtc_vblank_control_worker(struct work_struct *work)
 			vblank_work->acrtc->dm_irq_params.allow_sr_entry);
 	}
 
-	if (dm->active_vblank_irq_count == 0)
+	if (dm->active_vblank_irq_count == 0) {
+		dc_post_update_surfaces_to_stream(dm->dc);
 		dc_allow_idle_optimizations(dm->dc, true);
+	}
 
 	mutex_unlock(&dm->dc_lock);
 

From c8bedab2d9a1a0daa49ac20f9928a943f7205582 Mon Sep 17 00:00:00 2001
From: Karthi Kandasamy <karthi.kandasamy@amd.com>
Date: Wed, 3 Sep 2025 14:16:27 +0200
Subject: [PATCH 1394/1455] drm/amd/display: Add AVI infoframe copy in
 copy_stream_update_to_stream

[WHY]
Ensure AVI infoframe updates from stream updates are applied to the active
stream so OS overrides are not lost.

[HOW]
Copy avi_infopacket to stream when valid flag is set.
Follow existing infopacket copy pattern and perform a basic validity check before assignment.

Reviewed-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Karthi Kandasamy <karthi.kandasamy@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c          | 7 ++++++-
 drivers/gpu/drm/amd/display/dc/core/dc_resource.c | 6 ++++++
 drivers/gpu/drm/amd/display/dc/dc_stream.h        | 3 +++
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index c4dd52ed377d..974d63af1c70 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -3313,6 +3313,9 @@ static void copy_stream_update_to_stream(struct dc *dc,
 	if (update->adaptive_sync_infopacket)
 		stream->adaptive_sync_infopacket = *update->adaptive_sync_infopacket;
 
+	if (update->avi_infopacket)
+		stream->avi_infopacket = *update->avi_infopacket;
+
 	if (update->dither_option)
 		stream->dither_option = *update->dither_option;
 
@@ -3607,7 +3610,8 @@ static void commit_planes_do_stream_update(struct dc *dc,
 					stream_update->vsp_infopacket ||
 					stream_update->hfvsif_infopacket ||
 					stream_update->adaptive_sync_infopacket ||
-					stream_update->vtem_infopacket) {
+					stream_update->vtem_infopacket ||
+					stream_update->avi_infopacket) {
 				resource_build_info_frame(pipe_ctx);
 				dc->hwss.update_info_frame(pipe_ctx);
 
@@ -5079,6 +5083,7 @@ static bool full_update_required(struct dc *dc,
 			stream_update->hfvsif_infopacket ||
 			stream_update->vtem_infopacket ||
 			stream_update->adaptive_sync_infopacket ||
+			stream_update->avi_infopacket ||
 			stream_update->dpms_off ||
 			stream_update->allow_freesync ||
 			stream_update->vrr_active_variable ||
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index cbca3c67f439..bc5dedf5f60c 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -4410,8 +4410,14 @@ static void set_avi_info_frame(
 	unsigned int fr_ind = pipe_ctx->stream->timing.fr_index;
 	enum dc_timing_3d_format format;
 
+	if (stream->avi_infopacket.valid) {
+		*info_packet = stream->avi_infopacket;
+		return;
+	}
+
 	memset(&hdmi_info, 0, sizeof(union hdmi_info_packet));
 
+
 	color_space = pipe_ctx->stream->output_color_space;
 	if (color_space == COLOR_SPACE_UNKNOWN)
 		color_space = (stream->timing.pixel_encoding == PIXEL_ENCODING_RGB) ?
diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h
index 5fc6fea211de..76cf9fdedab0 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_stream.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h
@@ -203,6 +203,7 @@ struct dc_stream_state {
 	struct dc_info_packet hfvsif_infopacket;
 	struct dc_info_packet vtem_infopacket;
 	struct dc_info_packet adaptive_sync_infopacket;
+	struct dc_info_packet avi_infopacket;
 	uint8_t dsc_packed_pps[128];
 	struct rect src; /* composition area */
 	struct rect dst; /* stream addressable area */
@@ -335,6 +336,8 @@ struct dc_stream_update {
 	struct dc_info_packet *hfvsif_infopacket;
 	struct dc_info_packet *vtem_infopacket;
 	struct dc_info_packet *adaptive_sync_infopacket;
+	struct dc_info_packet *avi_infopacket;
+
 	bool *dpms_off;
 	bool integer_scaling_update;
 	bool *allow_freesync;

From 74d70e309d155550c9bd8bce74812875df47fb7b Mon Sep 17 00:00:00 2001
From: Paul Hsieh <Paul.Hsieh@amd.com>
Date: Fri, 5 Sep 2025 11:38:21 +0800
Subject: [PATCH 1395/1455] drm/amd/display: Add monitor patch to read psr cap
 again
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why & How]
According to the vendor’s requirement, after each OUI write,
the PSR cap must be read; otherwise, the vendor will default
to using PSRSU. But its PSR cap indicates that it only supports
PSR1.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Paul Hsieh <Paul.Hsieh@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc_types.h          |  1 +
 .../dc/link/protocols/link_edp_panel_control.c     | 14 ++++++++++++++
 2 files changed, 15 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h
index 619834a328a3..b5aa03a3e39c 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_types.h
@@ -1217,6 +1217,7 @@ struct dc_panel_config {
 		bool rc_disable;
 		bool rc_allow_static_screen;
 		bool rc_allow_fullscreen_VPB;
+		bool read_psrcap_again;
 		unsigned int replay_enable_option;
 	} psr;
 	/* ABM */
diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
index 8b7b87b21c2e..5e806edbb9f6 100644
--- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
+++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c
@@ -703,6 +703,20 @@ bool edp_setup_psr(struct dc_link *link,
 	if (!link)
 		return false;
 
+	/* This is a workaround: some vendors require the source to
+	 * read the PSR cap; otherwise, the vendor's PSR feature will
+	 * fall back to its default behavior, causing a misconfiguration
+	 * of this feature.
+	 */
+	if (link->panel_config.psr.read_psrcap_again) {
+		dm_helpers_dp_read_dpcd(
+			link->ctx,
+			link,
+			DP_PSR_SUPPORT,
+			&link->dpcd_caps.psr_info.psr_version,
+			sizeof(link->dpcd_caps.psr_info.psr_version));
+	}
+
 	//Clear PSR cfg
 	memset(&psr_configuration, 0, sizeof(psr_configuration));
 	dm_helpers_dp_write_dpcd(

From 6cec25f5b5660602b1953038cf40968b2d71c403 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Thu, 4 Sep 2025 13:49:35 -0500
Subject: [PATCH 1396/1455] drm/amd/display: Handle interpolation for first
 data point

[Why]
If the first data point for a custom brightness curve is not 0% luminance
then the first few luminance values will be ignored.

[How]
Check signal is below first data point and if so do linear interpolation to
0 instead.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 09427e056055..271ea1615178 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -4828,6 +4828,16 @@ static void convert_custom_brightness(const struct amdgpu_dm_backlight_caps *cap
 	if (!caps->data_points)
 		return;
 
+	/*
+	 * Handle the case where brightness is below the first data point
+	 * Interpolate between (0,0) and (first_signal, first_lum)
+	 */
+	if (brightness < caps->luminance_data[0].input_signal) {
+		lum = DIV_ROUND_CLOSEST(caps->luminance_data[0].luminance * brightness,
+					caps->luminance_data[0].input_signal);
+		goto scale;
+	}
+
 	left = 0;
 	right = caps->data_points - 1;
 	while (left <= right) {

From f082daf08f2ff313bdf9cf929a28f6d888117986 Mon Sep 17 00:00:00 2001
From: Lo-an Chen <lo-an.chen@amd.com>
Date: Mon, 25 Aug 2025 18:16:24 +0800
Subject: [PATCH 1397/1455] drm/amd/display: Init dispclk from bootup clock for
 DCN314

[Why]
Driver does not pick up and save vbios's clocks during init clocks,
the dispclk in clk_mgr will keep 0 until the first update clocks.
In some cases, OS changes the timing in the second set mode
(lower the pixel clock), causing the driver to lower the dispclk
in prepare bandwidth, which is illegal and causes grey screen.

[How]
1. Dump and save the vbios's clocks, and init the dispclk in
dcn314_init_clocks.
2. Fix the condition in dcn314_update_clocks, regarding a 0kHz value.

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Signed-off-by: Lo-an Chen <lo-an.chen@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.c        | 142 +++++++++++++++++-
 .../dc/clk_mgr/dcn314/dcn314_clk_mgr.h        |   5 +
 .../dc/resource/dcn314/dcn314_resource.c      |   1 +
 3 files changed, 143 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
index 790bbd8235b1..9e63fa72101c 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.c
@@ -77,6 +77,7 @@ static const struct IP_BASE CLK_BASE = { { { { 0x00016C00, 0x02401800, 0, 0, 0,
 #undef DC_LOGGER
 #define DC_LOGGER \
 	clk_mgr->base.base.ctx->logger
+
 #define regCLK1_CLK_PLL_REQ			0x0237
 #define regCLK1_CLK_PLL_REQ_BASE_IDX		0
 
@@ -87,8 +88,70 @@ static const struct IP_BASE CLK_BASE = { { { { 0x00016C00, 0x02401800, 0, 0, 0,
 #define CLK1_CLK_PLL_REQ__PllSpineDiv_MASK	0x0000F000L
 #define CLK1_CLK_PLL_REQ__FbMult_frac_MASK	0xFFFF0000L
 
+#define regCLK1_CLK0_DFS_CNTL				0x0269
+#define regCLK1_CLK0_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK1_DFS_CNTL				0x026c
+#define regCLK1_CLK1_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK2_DFS_CNTL				0x026f
+#define regCLK1_CLK2_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK3_DFS_CNTL				0x0272
+#define regCLK1_CLK3_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK4_DFS_CNTL				0x0275
+#define regCLK1_CLK4_DFS_CNTL_BASE_IDX		0
+#define regCLK1_CLK5_DFS_CNTL				0x0278
+#define regCLK1_CLK5_DFS_CNTL_BASE_IDX		0
+
+#define regCLK1_CLK0_CURRENT_CNT			0x02fb
+#define regCLK1_CLK0_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK1_CURRENT_CNT			0x02fc
+#define regCLK1_CLK1_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK2_CURRENT_CNT			0x02fd
+#define regCLK1_CLK2_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK3_CURRENT_CNT			0x02fe
+#define regCLK1_CLK3_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK4_CURRENT_CNT			0x02ff
+#define regCLK1_CLK4_CURRENT_CNT_BASE_IDX	0
+#define regCLK1_CLK5_CURRENT_CNT			0x0300
+#define regCLK1_CLK5_CURRENT_CNT_BASE_IDX	0
+
+#define regCLK1_CLK0_BYPASS_CNTL			0x028a
+#define regCLK1_CLK0_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK1_BYPASS_CNTL			0x0293
+#define regCLK1_CLK1_BYPASS_CNTL_BASE_IDX	0
 #define regCLK1_CLK2_BYPASS_CNTL			0x029c
 #define regCLK1_CLK2_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK3_BYPASS_CNTL			0x02a5
+#define regCLK1_CLK3_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK4_BYPASS_CNTL			0x02ae
+#define regCLK1_CLK4_BYPASS_CNTL_BASE_IDX	0
+#define regCLK1_CLK5_BYPASS_CNTL			0x02b7
+#define regCLK1_CLK5_BYPASS_CNTL_BASE_IDX	0
+
+#define regCLK1_CLK0_DS_CNTL				0x0283
+#define regCLK1_CLK0_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK1_DS_CNTL				0x028c
+#define regCLK1_CLK1_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK2_DS_CNTL				0x0295
+#define regCLK1_CLK2_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK3_DS_CNTL				0x029e
+#define regCLK1_CLK3_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK4_DS_CNTL				0x02a7
+#define regCLK1_CLK4_DS_CNTL_BASE_IDX		0
+#define regCLK1_CLK5_DS_CNTL				0x02b0
+#define regCLK1_CLK5_DS_CNTL_BASE_IDX		0
+
+#define regCLK1_CLK0_ALLOW_DS				0x0284
+#define regCLK1_CLK0_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK1_ALLOW_DS				0x028d
+#define regCLK1_CLK1_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK2_ALLOW_DS				0x0296
+#define regCLK1_CLK2_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK3_ALLOW_DS				0x029f
+#define regCLK1_CLK3_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK4_ALLOW_DS				0x02a8
+#define regCLK1_CLK4_ALLOW_DS_BASE_IDX		0
+#define regCLK1_CLK5_ALLOW_DS				0x02b1
+#define regCLK1_CLK5_ALLOW_DS_BASE_IDX		0
 
 #define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_SEL__SHIFT	0x0
 #define CLK1_CLK2_BYPASS_CNTL__CLK2_BYPASS_DIV__SHIFT	0x10
@@ -185,6 +248,8 @@ void dcn314_init_clocks(struct clk_mgr *clk_mgr)
 {
 	struct clk_mgr_internal *clk_mgr_int = TO_CLK_MGR_INTERNAL(clk_mgr);
 	uint32_t ref_dtbclk = clk_mgr->clks.ref_dtbclk_khz;
+	struct clk_mgr_dcn314 *clk_mgr_dcn314 = TO_CLK_MGR_DCN314(clk_mgr_int);
+	struct clk_log_info log_info = {0};
 
 	memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks));
 	// Assumption is that boot state always supports pstate
@@ -200,6 +265,9 @@ void dcn314_init_clocks(struct clk_mgr *clk_mgr)
 			dce_adjust_dp_ref_freq_for_ss(clk_mgr_int, clk_mgr->dprefclk_khz);
 	else
 		clk_mgr->dp_dto_source_clock_in_khz = clk_mgr->dprefclk_khz;
+
+	dcn314_dump_clk_registers(&clk_mgr->boot_snapshot, &clk_mgr_dcn314->base.base, &log_info);
+	clk_mgr->clks.dispclk_khz =  clk_mgr->boot_snapshot.dispclk * 1000;
 }
 
 void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
@@ -218,6 +286,8 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
 	if (dc->work_arounds.skip_clock_update)
 		return;
 
+	display_count = dcn314_get_active_display_cnt_wa(dc, context);
+
 	/*
 	 * if it is safe to lower, but we are already in the lower state, we don't have to do anything
 	 * also if safe to lower is false, we just go in the higher state
@@ -236,7 +306,6 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
 		}
 		/* check that we're not already in lower */
 		if (clk_mgr_base->clks.pwr_state != DCN_PWR_STATE_LOW_POWER) {
-			display_count = dcn314_get_active_display_cnt_wa(dc, context);
 			/* if we can go lower, go lower */
 			if (display_count == 0) {
 				union display_idle_optimization_u idle_info = { 0 };
@@ -293,11 +362,19 @@ void dcn314_update_clocks(struct clk_mgr *clk_mgr_base,
 		update_dppclk = true;
 	}
 
-	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz)) {
+	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz) &&
+	    (new_clocks->dispclk_khz > 0 || (safe_to_lower && display_count == 0))) {
+		int requested_dispclk_khz = new_clocks->dispclk_khz;
+
 		dcn314_disable_otg_wa(clk_mgr_base, context, safe_to_lower, true);
 
+		/* Clamp the requested clock to PMFW based on their limit. */
+		if (dc->debug.min_disp_clk_khz > 0 && requested_dispclk_khz < dc->debug.min_disp_clk_khz)
+			requested_dispclk_khz = dc->debug.min_disp_clk_khz;
+
+		dcn314_smu_set_dispclk(clk_mgr, requested_dispclk_khz);
 		clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
-		dcn314_smu_set_dispclk(clk_mgr, clk_mgr_base->clks.dispclk_khz);
+
 		dcn314_disable_otg_wa(clk_mgr_base, context, safe_to_lower, false);
 
 		update_dispclk = true;
@@ -385,10 +462,65 @@ bool dcn314_are_clock_states_equal(struct dc_clocks *a,
 	return true;
 }
 
-static void dcn314_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
+
+static void dcn314_dump_clk_registers_internal(struct dcn35_clk_internal *internal, struct clk_mgr *clk_mgr_base)
+{
+	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+
+	// read dtbclk
+	internal->CLK1_CLK4_CURRENT_CNT = REG_READ(CLK1_CLK4_CURRENT_CNT);
+	internal->CLK1_CLK4_BYPASS_CNTL = REG_READ(CLK1_CLK4_BYPASS_CNTL);
+
+	// read dcfclk
+	internal->CLK1_CLK3_CURRENT_CNT = REG_READ(CLK1_CLK3_CURRENT_CNT);
+	internal->CLK1_CLK3_BYPASS_CNTL = REG_READ(CLK1_CLK3_BYPASS_CNTL);
+
+	// read dcf deep sleep divider
+	internal->CLK1_CLK3_DS_CNTL = REG_READ(CLK1_CLK3_DS_CNTL);
+	internal->CLK1_CLK3_ALLOW_DS = REG_READ(CLK1_CLK3_ALLOW_DS);
+
+	// read dppclk
+	internal->CLK1_CLK1_CURRENT_CNT = REG_READ(CLK1_CLK1_CURRENT_CNT);
+	internal->CLK1_CLK1_BYPASS_CNTL = REG_READ(CLK1_CLK1_BYPASS_CNTL);
+
+	// read dprefclk
+	internal->CLK1_CLK2_CURRENT_CNT = REG_READ(CLK1_CLK2_CURRENT_CNT);
+	internal->CLK1_CLK2_BYPASS_CNTL = REG_READ(CLK1_CLK2_BYPASS_CNTL);
+
+	// read dispclk
+	internal->CLK1_CLK0_CURRENT_CNT = REG_READ(CLK1_CLK0_CURRENT_CNT);
+	internal->CLK1_CLK0_BYPASS_CNTL = REG_READ(CLK1_CLK0_BYPASS_CNTL);
+}
+
+void dcn314_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
 		struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info)
 {
-	return;
+
+	struct dcn35_clk_internal internal = {0};
+
+	dcn314_dump_clk_registers_internal(&internal, clk_mgr_base);
+
+	regs_and_bypass->dcfclk = internal.CLK1_CLK3_CURRENT_CNT / 10;
+	regs_and_bypass->dcf_deep_sleep_divider = internal.CLK1_CLK3_DS_CNTL / 10;
+	regs_and_bypass->dcf_deep_sleep_allow = internal.CLK1_CLK3_ALLOW_DS;
+	regs_and_bypass->dprefclk = internal.CLK1_CLK2_CURRENT_CNT / 10;
+	regs_and_bypass->dispclk = internal.CLK1_CLK0_CURRENT_CNT / 10;
+	regs_and_bypass->dppclk = internal.CLK1_CLK1_CURRENT_CNT / 10;
+	regs_and_bypass->dtbclk = internal.CLK1_CLK4_CURRENT_CNT / 10;
+
+	regs_and_bypass->dppclk_bypass = internal.CLK1_CLK1_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dppclk_bypass < 0 || regs_and_bypass->dppclk_bypass > 4)
+		regs_and_bypass->dppclk_bypass = 0;
+	regs_and_bypass->dcfclk_bypass = internal.CLK1_CLK3_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dcfclk_bypass < 0 || regs_and_bypass->dcfclk_bypass > 4)
+		regs_and_bypass->dcfclk_bypass = 0;
+	regs_and_bypass->dispclk_bypass = internal.CLK1_CLK0_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dispclk_bypass < 0 || regs_and_bypass->dispclk_bypass > 4)
+		regs_and_bypass->dispclk_bypass = 0;
+	regs_and_bypass->dprefclk_bypass = internal.CLK1_CLK2_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dprefclk_bypass < 0 || regs_and_bypass->dprefclk_bypass > 4)
+		regs_and_bypass->dprefclk_bypass = 0;
+
 }
 
 static struct clk_bw_params dcn314_bw_params = {
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
index 002c28e80720..0577eb527bc3 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn314/dcn314_clk_mgr.h
@@ -65,4 +65,9 @@ void dcn314_clk_mgr_construct(struct dc_context *ctx,
 
 void dcn314_clk_mgr_destroy(struct clk_mgr_internal *clk_mgr_int);
 
+
+void dcn314_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
+		struct clk_mgr *clk_mgr_base, struct clk_log_info *log_info);
+
+
 #endif //__DCN314_CLK_MGR_H__
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
index 663c49cce4aa..d4917a35b991 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c
@@ -927,6 +927,7 @@ static const struct dc_debug_options debug_defaults_drv = {
 	.enable_legacy_fast_update = true,
 	.using_dml2 = false,
 	.disable_dsc_power_gate = true,
+	.min_disp_clk_khz = 100000,
 };
 
 static const struct dc_panel_config panel_config_defaults = {

From 3451021a9e153e3f3067c90429c8317c126e159e Mon Sep 17 00:00:00 2001
From: Ivan Lipski <ivan.lipski@amd.com>
Date: Tue, 9 Sep 2025 14:15:05 -0400
Subject: [PATCH 1398/1455] drm/amd/display: Enable DTM v3 on dGPUs with DCN
 3.1+

[Why&How]
Right now, only selected APUs have enabled DTM v3, which allows to use
newer firmware for content protection.

We want to enable it on the dGPUs starting with DCN 3.2

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c   | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
index 58e084f52526..19038f336155 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_hdcp.c
@@ -768,14 +768,18 @@ struct hdcp_workqueue *hdcp_create_workqueue(struct amdgpu_device *adev,
 		struct mod_hdcp_ddc_funcs *ddc_funcs = &config->ddc.funcs;
 
 		config->psp.handle = &adev->psp;
-		if (dc->ctx->dce_version == DCN_VERSION_3_1 ||
+		if (dc->ctx->dce_version == DCN_VERSION_3_1  ||
 		    dc->ctx->dce_version == DCN_VERSION_3_14 ||
 		    dc->ctx->dce_version == DCN_VERSION_3_15 ||
-		    dc->ctx->dce_version == DCN_VERSION_3_5 ||
+		    dc->ctx->dce_version == DCN_VERSION_3_16 ||
+		    dc->ctx->dce_version == DCN_VERSION_3_2  ||
+		    dc->ctx->dce_version == DCN_VERSION_3_21 ||
+		    dc->ctx->dce_version == DCN_VERSION_3_5  ||
 		    dc->ctx->dce_version == DCN_VERSION_3_51 ||
-		    dc->ctx->dce_version == DCN_VERSION_3_6 ||
-		    dc->ctx->dce_version == DCN_VERSION_3_16)
+		    dc->ctx->dce_version == DCN_VERSION_3_6  ||
+		    dc->ctx->dce_version == DCN_VERSION_4_01)
 			config->psp.caps.dtm_v3_supported = 1;
+
 		config->ddc.handle = dc_get_link_at_index(dc, i);
 
 		ddc_funcs->write_i2c = lp_write_i2c;

From 0bf6b216d4783cb51f9af05a49d3cce4fc22dc24 Mon Sep 17 00:00:00 2001
From: Sridevi Arvindekar <sarvinde@amd.com>
Date: Wed, 10 Sep 2025 11:04:07 -0400
Subject: [PATCH 1399/1455] drm/amd/display: Fix for test crash due to power
 gating

[Why/How]
Call power gating routine only if it is defined.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Sridevi Arvindekar <sarvinde@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index 417f2679723e..95af58cbb92a 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -3131,7 +3131,8 @@ void dcn20_fpga_init_hw(struct dc *dc)
 		res_pool->dccg->funcs->dccg_init(res_pool->dccg);
 
 	//Enable ability to power gate / don't force power on permanently
-	hws->funcs.enable_power_gating_plane(hws, true);
+	if (hws->funcs.enable_power_gating_plane)
+		hws->funcs.enable_power_gating_plane(hws, true);
 
 	// Specific to FPGA dccg and registers
 	REG_WRITE(RBBMIF_TIMEOUT_DIS, 0xFFFFFFFF);

From 9e5d4a5e27c6dc4e1b4fc9d654d13de12b8ce156 Mon Sep 17 00:00:00 2001
From: Alvin Lee <Alvin.Lee2@amd.com>
Date: Tue, 9 Sep 2025 16:03:08 -0400
Subject: [PATCH 1400/1455] drm/amd/display: Use mpc.preblend flag to indicate
 preblend

[Description]
Modifications in per asic capability means mpc.preblend flag should be used
to indicate preblend. Update relevant paths to use this flag.

Fixes: 39923050615c ("drm/amd/display: Clear DPP 3DLUT Cap")
Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c | 2 +-
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index c7387af725d6..b0ef157f1333 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -821,7 +821,7 @@ int amdgpu_dm_verify_lut3d_size(struct amdgpu_device *adev,
 	struct dm_plane_state *dm_plane_state = to_dm_plane_state(plane_state);
 	const struct drm_color_lut *shaper = NULL, *lut3d = NULL;
 	uint32_t exp_size, size, dim_size = MAX_COLOR_3DLUT_SIZE;
-	bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut;
+	bool has_3dlut = adev->dm.dc->caps.color.dpp.hw_3d_lut || adev->dm.dc->caps.color.mpc.preblend;
 
 	/* shaper LUT is only available if 3D LUT color caps */
 	exp_size = has_3dlut ? MAX_COLOR_LUT_ENTRIES : 0;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
index 08f629c64df3..e027798ece03 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_plane.c
@@ -1633,7 +1633,7 @@ dm_atomic_plane_attach_color_mgmt_properties(struct amdgpu_display_manager *dm,
 		drm_object_attach_property(&plane->base,
 					   dm->adev->mode_info.plane_ctm_property, 0);
 
-	if (dpp_color_caps.hw_3d_lut) {
+	if (dpp_color_caps.hw_3d_lut || dm->dc->caps.color.mpc.preblend) {
 		drm_object_attach_property(&plane->base,
 					   mode_info.plane_shaper_lut_property, 0);
 		drm_object_attach_property(&plane->base,

From b65cf4baeb24bdb5fee747679ee88f1ade5c1d6c Mon Sep 17 00:00:00 2001
From: Allen Li <wei-guang.li@amd.com>
Date: Fri, 5 Sep 2025 16:58:38 +0800
Subject: [PATCH 1401/1455] drm/amd/display: Add fast sync field in ultra sleep
 more for DMUB

[Why&How]
We need to inform DMUB whether fast sync in ultra sleep mode is supported,
so that it can disable desync error detection when the it is not enabled.
This helps prevent unexpected desync errors when transitioning out of
ultra sleep mode.

Add fast sync in ultra sleep mode field in replay copy setting command.

Reviewed-by: Robin Chen <robin.chen@amd.com>
Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Allen Li <wei-guang.li@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c | 1 +
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h  | 6 +++++-
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
index 9e2a473a8852..f9542edff14b 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dmub_replay.c
@@ -169,6 +169,7 @@ static bool dmub_replay_copy_settings(struct dmub_replay *dmub,
 	copy_settings_data->max_deviation_line			= link->dpcd_caps.pr_info.max_deviation_line;
 	copy_settings_data->smu_optimizations_en		= link->replay_settings.replay_smu_opt_enable;
 	copy_settings_data->replay_timing_sync_supported = link->replay_settings.config.replay_timing_sync_supported;
+	copy_settings_data->replay_support_fast_resync_in_ultra_sleep_mode = link->replay_settings.config.replay_support_fast_resync_in_ultra_sleep_mode;
 
 	copy_settings_data->debug.bitfields.enable_ips_visual_confirm = dc->dc->debug.enable_ips_visual_confirm;
 
diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index e65747f7f12f..6db19e49a05c 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -4142,10 +4142,14 @@ struct dmub_cmd_replay_copy_settings_data {
 	 * @hpo_link_enc_inst: HPO link encoder instance
 	 */
 	uint8_t hpo_link_enc_inst;
+	/**
+	 * Determines if fast sync in ultra sleep mode is enabled/disabled.
+	 */
+	uint8_t replay_support_fast_resync_in_ultra_sleep_mode;
 	/**
 	 * @pad: Align structure to 4 byte boundary.
 	 */
-	uint8_t pad[2];
+	uint8_t pad[1];
 };
 
 

From d19f570cdf94caa5fad0f1403466e01a7c430b8f Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Fri, 22 Aug 2025 13:23:18 -0400
Subject: [PATCH 1402/1455] drm/amd/display: Isolate dcn401 SMU functions

[WHY&HOW]
SMU interfaces are not backwards and forwards compatible, so they should
be isolated per version.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/clk_mgr/dcn401/dcn401_clk_mgr.c        |  26 ++--
 .../clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c   | 118 ++++++++++++++++++
 .../clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.h   |  10 +-
 3 files changed, 140 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
index 47461f249e83..306016c1f109 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr.c
@@ -162,7 +162,7 @@ static void dcn401_init_single_clock(struct clk_mgr_internal *clk_mgr, PPCLK_e c
 	unsigned int i;
 	char *entry_i = (char *)entry_0;
 
-	uint32_t ret = dcn30_smu_get_dpm_freq_by_index(clk_mgr, clk, 0xFF);
+	uint32_t ret = dcn401_smu_get_dpm_freq_by_index(clk_mgr, clk, 0xFF);
 
 	if (ret & (1 << 31))
 		/* fine-grained, only min and max */
@@ -174,7 +174,7 @@ static void dcn401_init_single_clock(struct clk_mgr_internal *clk_mgr, PPCLK_e c
 
 	/* if the initial message failed, num_levels will be 0 */
 	for (i = 0; i < *num_levels && i < ARRAY_SIZE(clk_mgr->base.bw_params->clk_table.entries); i++) {
-		*((unsigned int *)entry_i) = (dcn30_smu_get_dpm_freq_by_index(clk_mgr, clk, i) & 0xFFFF);
+		*((unsigned int *)entry_i) = (dcn401_smu_get_dpm_freq_by_index(clk_mgr, clk, i) & 0xFFFF);
 		entry_i += sizeof(clk_mgr->base.bw_params->clk_table.entries[0]);
 	}
 }
@@ -231,20 +231,20 @@ void dcn401_init_clocks(struct clk_mgr *clk_mgr_base)
 	clk_mgr->smu_present = false;
 	clk_mgr->dpm_present = false;
 
-	if (!clk_mgr_base->force_smu_not_present && dcn30_smu_get_smu_version(clk_mgr, &clk_mgr->smu_ver))
+	if (!clk_mgr_base->force_smu_not_present && dcn401_smu_get_smu_version(clk_mgr, &clk_mgr->smu_ver))
 		clk_mgr->smu_present = true;
 
 	if (!clk_mgr->smu_present)
 		return;
 
-	dcn30_smu_check_driver_if_version(clk_mgr);
-	dcn30_smu_check_msg_header_version(clk_mgr);
+	dcn401_smu_check_driver_if_version(clk_mgr);
+	dcn401_smu_check_msg_header_version(clk_mgr);
 
 	/* DCFCLK */
 	dcn401_init_single_clock(clk_mgr, PPCLK_DCFCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dcfclk_mhz,
 			&num_entries_per_clk->num_dcfclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DCFCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DCFCLK);
 	if (num_entries_per_clk->num_dcfclk_levels && clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_dcfclk_levels - 1].dcfclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.dcfclk_mhz = 0;
@@ -253,7 +253,7 @@ void dcn401_init_clocks(struct clk_mgr *clk_mgr_base)
 	dcn401_init_single_clock(clk_mgr, PPCLK_SOCCLK,
 					&clk_mgr_base->bw_params->clk_table.entries[0].socclk_mhz,
 					&num_entries_per_clk->num_socclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_SOCCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_SOCCLK);
 	if (num_entries_per_clk->num_socclk_levels && clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_socclk_levels - 1].socclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.socclk_mhz = 0;
@@ -263,7 +263,7 @@ void dcn401_init_clocks(struct clk_mgr *clk_mgr_base)
 		dcn401_init_single_clock(clk_mgr, PPCLK_DTBCLK,
 				&clk_mgr_base->bw_params->clk_table.entries[0].dtbclk_mhz,
 				&num_entries_per_clk->num_dtbclk_levels);
-		clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DTBCLK);
+		clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DTBCLK);
 		if (num_entries_per_clk->num_dtbclk_levels && clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz ==
 				clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_dtbclk_levels - 1].dtbclk_mhz)
 			clk_mgr_base->bw_params->dc_mode_limit.dtbclk_mhz = 0;
@@ -273,7 +273,7 @@ void dcn401_init_clocks(struct clk_mgr *clk_mgr_base)
 	dcn401_init_single_clock(clk_mgr, PPCLK_DISPCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].dispclk_mhz,
 			&num_entries_per_clk->num_dispclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DISPCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_DISPCLK);
 	if (num_entries_per_clk->num_dispclk_levels && clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_dispclk_levels - 1].dispclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.dispclk_mhz = 0;
@@ -1318,8 +1318,8 @@ static void dcn401_notify_wm_ranges(struct clk_mgr *clk_mgr_base)
 			table->Watermarks.WatermarkRow[i].WmSetting = i;
 			table->Watermarks.WatermarkRow[i].Flags = clk_mgr->base.bw_params->wm_table.nv_entries[i].pmfw_breakdown.wm_type;
 		}
-	dcn30_smu_set_dram_addr_high(clk_mgr, clk_mgr->wm_range_table_addr >> 32);
-	dcn30_smu_set_dram_addr_low(clk_mgr, clk_mgr->wm_range_table_addr & 0xFFFFFFFF);
+	dcn401_smu_set_dram_addr_high(clk_mgr, clk_mgr->wm_range_table_addr >> 32);
+	dcn401_smu_set_dram_addr_low(clk_mgr, clk_mgr->wm_range_table_addr & 0xFFFFFFFF);
 	dcn401_smu_transfer_wm_table_dram_2_smu(clk_mgr);
 }
 
@@ -1390,7 +1390,7 @@ static void dcn401_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 				clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_memclk_levels - 1].memclk_mhz;
 	}
 
-	clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK);
 	if (num_entries_per_clk->num_memclk_levels && clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_memclk_levels - 1].memclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.memclk_mhz = 0;
@@ -1399,7 +1399,7 @@ static void dcn401_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base)
 	dcn401_init_single_clock(clk_mgr, PPCLK_FCLK,
 			&clk_mgr_base->bw_params->clk_table.entries[0].fclk_mhz,
 			&num_entries_per_clk->num_fclk_levels);
-	clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK);
+	clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = dcn401_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_FCLK);
 	if (num_entries_per_clk->num_fclk_levels && clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz ==
 			clk_mgr_base->bw_params->clk_table.entries[num_entries_per_clk->num_fclk_levels - 1].fclk_mhz)
 		clk_mgr_base->bw_params->dc_mode_limit.fclk_mhz = 0;
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c
index 21c35528f61f..347a0d66d653 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c
@@ -139,6 +139,59 @@ static bool dcn401_smu_send_msg_with_param_delay(struct clk_mgr_internal *clk_mg
 	return false;
 }
 
+bool dcn401_smu_get_smu_version(struct clk_mgr_internal *clk_mgr, unsigned int *version)
+{
+	smu_print("SMU Get SMU version\n");
+
+	if (dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetSmuVersion, 0, version)) {
+
+		smu_print("SMU version: %d\n", *version);
+
+		return true;
+	}
+
+	return false;
+}
+
+/* Message output should match SMU11_DRIVER_IF_VERSION in smu11_driver_if.h */
+bool dcn401_smu_check_driver_if_version(struct clk_mgr_internal *clk_mgr)
+{
+	uint32_t response = 0;
+
+	smu_print("SMU Check driver if version\n");
+
+	if (dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetDriverIfVersion, 0, &response)) {
+
+		smu_print("SMU driver if version: %d\n", response);
+
+		if (response == SMU14_DRIVER_IF_VERSION)
+			return true;
+	}
+
+	return false;
+}
+
+/* Message output should match DALSMC_VERSION in dalsmc.h */
+bool dcn401_smu_check_msg_header_version(struct clk_mgr_internal *clk_mgr)
+{
+	uint32_t response = 0;
+
+	smu_print("SMU Check msg header version\n");
+
+	if (dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetMsgHeaderVersion, 0, &response)) {
+
+		smu_print("SMU msg header version: %d\n", response);
+
+		if (response == DALSMC_VERSION)
+			return true;
+	}
+
+	return false;
+}
+
 void dcn401_smu_send_fclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool support)
 {
 	smu_print("FCLK P-state support value is : %d\n", support);
@@ -163,6 +216,22 @@ void dcn401_smu_send_cab_for_uclk_message(struct clk_mgr_internal *clk_mgr, unsi
 	smu_print("Numways for SubVP : %d\n", num_ways);
 }
 
+void dcn401_smu_set_dram_addr_high(struct clk_mgr_internal *clk_mgr, uint32_t addr_high)
+{
+	smu_print("SMU Set DRAM addr high: %d\n", addr_high);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_SetDalDramAddrHigh, addr_high, NULL);
+}
+
+void dcn401_smu_set_dram_addr_low(struct clk_mgr_internal *clk_mgr, uint32_t addr_low)
+{
+	smu_print("SMU Set DRAM addr low: %d\n", addr_low);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_SetDalDramAddrLow, addr_low, NULL);
+}
+
 void dcn401_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr)
 {
 	smu_print("SMU Transfer WM table DRAM 2 SMU\n");
@@ -348,3 +417,52 @@ unsigned int dcn401_smu_get_num_of_umc_channels(struct clk_mgr_internal *clk_mgr
 
 	return response;
 }
+
+/*
+ * Frequency in MHz returned in lower 16 bits for valid DPM level
+ *
+ * Call with dpm_level = 0xFF to query features, return value will be:
+ *     Bits 7:0 - number of DPM levels
+ *     Bit   28 - 1 = auto DPM on
+ *     Bit   29 - 1 = sweep DPM on
+ *     Bit   30 - 1 = forced DPM on
+ *     Bit   31 - 0 = discrete, 1 = fine-grained
+ *
+ * With fine-grained DPM, only min and max frequencies will be reported
+ *
+ * Returns 0 on failure
+ */
+unsigned int dcn401_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint8_t dpm_level)
+{
+	uint32_t response = 0;
+
+	/* bits 23:16 for clock type, lower 8 bits for DPM level */
+	uint32_t param = (clk << 16) | dpm_level;
+
+	smu_print("SMU Get dpm freq by index: clk = %d, dpm_level = %d\n", clk, dpm_level);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetDpmFreqByIndex, param, &response);
+
+	smu_print("SMU dpm freq: %d MHz\n", response);
+
+	return response;
+}
+
+/* Returns the max DPM frequency in DC mode in MHz, 0 on failure */
+unsigned int dcn401_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk)
+{
+	uint32_t response = 0;
+
+	/* bits 23:16 for clock type */
+	uint32_t param = clk << 16;
+
+	smu_print("SMU Get DC mode max DPM freq: clk = %d\n", clk);
+
+	dcn401_smu_send_msg_with_param(clk_mgr,
+			DALSMC_MSG_GetDcModeMaxDpmFreq, param, &response);
+
+	smu_print("SMU DC mode max DMP freq: %d MHz\n", response);
+
+	return response;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.h b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.h
index e02eb1294b37..4f5ac603e822 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.h
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.h
@@ -7,11 +7,17 @@
 
 #include "os_types.h"
 #include "core_types.h"
-#include "dcn32/dcn32_clk_mgr_smu_msg.h"
 
+struct clk_mgr_internal;
+
+bool dcn401_smu_get_smu_version(struct clk_mgr_internal *clk_mgr, unsigned int *version);
+bool dcn401_smu_check_driver_if_version(struct clk_mgr_internal *clk_mgr);
+bool dcn401_smu_check_msg_header_version(struct clk_mgr_internal *clk_mgr);
 void dcn401_smu_send_fclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool support);
 void dcn401_smu_send_uclk_pstate_message(struct clk_mgr_internal *clk_mgr, bool support);
 void dcn401_smu_send_cab_for_uclk_message(struct clk_mgr_internal *clk_mgr, unsigned int num_ways);
+void dcn401_smu_set_dram_addr_high(struct clk_mgr_internal *clk_mgr, uint32_t addr_high);
+void dcn401_smu_set_dram_addr_low(struct clk_mgr_internal *clk_mgr, uint32_t addr_low);
 void dcn401_smu_transfer_wm_table_dram_2_smu(struct clk_mgr_internal *clk_mgr);
 void dcn401_smu_set_pme_workaround(struct clk_mgr_internal *clk_mgr);
 unsigned int dcn401_smu_set_hard_min_by_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint16_t freq_mhz);
@@ -29,5 +35,7 @@ bool dcn401_smu_set_subvp_uclk_fclk_hardmin(struct clk_mgr_internal *clk_mgr,
 void dcn401_smu_set_min_deep_sleep_dcef_clk(struct clk_mgr_internal *clk_mgr, uint32_t freq_mhz);
 void dcn401_smu_set_num_of_displays(struct clk_mgr_internal *clk_mgr, uint32_t num_displays);
 unsigned int dcn401_smu_get_num_of_umc_channels(struct clk_mgr_internal *clk_mgr);
+unsigned int dcn401_smu_get_dc_mode_max_dpm_freq(struct clk_mgr_internal *clk_mgr, uint32_t clk);
+unsigned int dcn401_smu_get_dpm_freq_by_index(struct clk_mgr_internal *clk_mgr, uint32_t clk, uint8_t dpm_level);
 
 #endif /* __DCN401_CLK_MGR_SMU_MSG_H_ */

From adb441ca64a2f3972eb2919e2e340c83a3212479 Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Wed, 10 Sep 2025 10:55:48 -0400
Subject: [PATCH 1403/1455] drm/amd/display: Refactor SMU tracing

[WHY&HOW]
Add new tracing and performance measurements for SMU messaging.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/amdgpu_dm/amdgpu_dm_services.c   |  4 ++--
 .../display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c |  2 +-
 .../display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c |  5 +++--
 .../dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c       | 12 ++++++++----
 drivers/gpu/drm/amd/display/dc/dm_services.h         | 11 ++++++-----
 5 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
index 132de4071efd..8550d5e8b753 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_services.c
@@ -53,11 +53,11 @@ void dm_perf_trace_timestamp(const char *func_name, unsigned int line, struct dc
 				    func_name, line);
 }
 
-void dm_trace_smu_msg(uint32_t msg_id, uint32_t param_in, struct dc_context *ctx)
+void dm_trace_smu_enter(uint32_t msg_id, uint32_t param_in, unsigned int delay, struct dc_context *ctx)
 {
 }
 
-void dm_trace_smu_delay(uint32_t delay, struct dc_context *ctx)
+void dm_trace_smu_exit(bool success, uint32_t response, struct dc_context *ctx)
 {
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c
index 3253115a153d..827bc2431d5d 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr_smu_msg.c
@@ -69,7 +69,7 @@ static uint32_t dcn30_smu_wait_for_response(struct clk_mgr_internal *clk_mgr, un
 
 	/* handle DALSMC_Result_CmdRejectedBusy? */
 
-	TRACE_SMU_DELAY(delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
+	TRACE_SMU_MSG_DELAY(0, 0, delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
 
 	return reg;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c
index cf2d35363e8b..5d80fdf63ffc 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn32/dcn32_clk_mgr_smu_msg.c
@@ -63,7 +63,8 @@ static uint32_t dcn32_smu_wait_for_response(struct clk_mgr_internal *clk_mgr, un
 			udelay(delay_us);
 	} while (max_retries--);
 
-	TRACE_SMU_DELAY(delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
+	TRACE_SMU_MSG_DELAY(0, 0, delay_us * (initial_max_retries - max_retries), clk_mgr->base.ctx);
+
 
 	return reg;
 }
@@ -120,7 +121,7 @@ static uint32_t dcn32_smu_wait_for_response_delay(struct clk_mgr_internal *clk_m
 		*total_delay_us += delay_us;
 	} while (max_retries--);
 
-	TRACE_SMU_DELAY(*total_delay_us, clk_mgr->base.ctx);
+	TRACE_SMU_MSG_DELAY(0, 0, *total_delay_us, clk_mgr->base.ctx);
 
 	return reg;
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c
index 347a0d66d653..3a263840893e 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn401/dcn401_clk_mgr_smu_msg.c
@@ -57,6 +57,8 @@ static bool dcn401_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, uin
 	/* Wait for response register to be ready */
 	dcn401_smu_wait_for_response(clk_mgr, 10, 200000);
 
+	TRACE_SMU_MSG_ENTER(msg_id, param_in, clk_mgr->base.ctx);
+
 	/* Clear response register */
 	REG_WRITE(DAL_RESP_REG, 0);
 
@@ -71,9 +73,11 @@ static bool dcn401_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, uin
 		if (param_out)
 			*param_out = REG_READ(DAL_ARG_REG);
 
+		TRACE_SMU_MSG_EXIT(true, param_out ? *param_out : 0, clk_mgr->base.ctx);
 		return true;
 	}
 
+	TRACE_SMU_MSG_EXIT(false, 0, clk_mgr->base.ctx);
 	return false;
 }
 
@@ -102,8 +106,6 @@ static uint32_t dcn401_smu_wait_for_response_delay(struct clk_mgr_internal *clk_
 		*total_delay_us += delay_us;
 	} while (max_retries--);
 
-	TRACE_SMU_DELAY(*total_delay_us, clk_mgr->base.ctx);
-
 	return reg;
 }
 
@@ -115,6 +117,8 @@ static bool dcn401_smu_send_msg_with_param_delay(struct clk_mgr_internal *clk_mg
 	/* Wait for response register to be ready */
 	dcn401_smu_wait_for_response_delay(clk_mgr, 10, 200000, &delay1_us);
 
+	TRACE_SMU_MSG_ENTER(msg_id, param_in, clk_mgr->base.ctx);
+
 	/* Clear response register */
 	REG_WRITE(DAL_RESP_REG, 0);
 
@@ -124,18 +128,18 @@ static bool dcn401_smu_send_msg_with_param_delay(struct clk_mgr_internal *clk_mg
 	/* Trigger the message transaction by writing the message ID */
 	REG_WRITE(DAL_MSG_REG, msg_id);
 
-	TRACE_SMU_MSG(msg_id, param_in, clk_mgr->base.ctx);
-
 	/* Wait for response */
 	if (dcn401_smu_wait_for_response_delay(clk_mgr, 10, 200000, &delay2_us) == DALSMC_Result_OK) {
 		if (param_out)
 			*param_out = REG_READ(DAL_ARG_REG);
 
 		*total_delay_us = delay1_us + delay2_us;
+		TRACE_SMU_MSG_EXIT(true, param_out ? *param_out : 0, clk_mgr->base.ctx);
 		return true;
 	}
 
 	*total_delay_us = delay1_us + 2000000;
+	TRACE_SMU_MSG_EXIT(false, 0, clk_mgr->base.ctx);
 	return false;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h
index 7b398d4f4439..fbbf9c757b3c 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services.h
@@ -277,12 +277,13 @@ void dm_perf_trace_timestamp(const char *func_name, unsigned int line, struct dc
 /*
  * SMU message tracing
  */
-void dm_trace_smu_msg(uint32_t msg_id, uint32_t param_in, struct dc_context *ctx);
-void dm_trace_smu_delay(uint32_t delay, struct dc_context *ctx);
-
-#define TRACE_SMU_MSG(msg_id, param_in, ctx)	dm_trace_smu_msg(msg_id, param_in, ctx)
-#define TRACE_SMU_DELAY(response_delay, ctx)	dm_trace_smu_delay(response_delay, ctx)
+void dm_trace_smu_enter(uint32_t msg_id, uint32_t param_in, unsigned int delay, struct dc_context *ctx);
+void dm_trace_smu_exit(bool success, uint32_t response, struct dc_context *ctx);
 
+#define TRACE_SMU_MSG_DELAY(msg_id, param_in, delay, ctx)	dm_trace_smu_enter(msg_id, param_in, delay, ctx)
+#define TRACE_SMU_MSG(msg_id, param_in, ctx)	dm_trace_smu_enter(msg_id, param_in, 0, ctx)
+#define TRACE_SMU_MSG_ENTER(msg_id, param_in, ctx)	dm_trace_smu_enter(msg_id, param_in, 0, ctx)
+#define TRACE_SMU_MSG_EXIT(success, response, ctx)	dm_trace_smu_exit(success, response, ctx)
 
 /*
  * DMUB Interfaces

From c480d074b3ff27a8d582505dfd53c9b091b63d57 Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Wed, 10 Sep 2025 16:55:23 -0400
Subject: [PATCH 1404/1455] drm/amd/display: Set wm_pending when disable
 stutter w/a used

[WHY&HOW]
When stutter is disabled prior to watermark programming due to a w/a,
wm_pending should be returned as true.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
index 3b71bfaca291..4d4ca6d77bbd 100644
--- a/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
+++ b/drivers/gpu/drm/amd/display/dc/hubbub/dcn32/dcn32_hubbub.c
@@ -770,6 +770,7 @@ static bool hubbub32_program_watermarks(
 			(ASICREV_IS_GC_11_0_0(dc->ctx->asic_id.hw_internal_rev) ||
 			ASICREV_IS_GC_11_0_3(dc->ctx->asic_id.hw_internal_rev))) {
 		/* before raising watermarks, SDP control give to DF, stutter must be disabled */
+		wm_pending = true;
 		hubbub32_set_sdp_control(hubbub, false);
 		hubbub1_allow_self_refresh_control(hubbub, false);
 	}

From 19d8f3192f78ce38e710da2df8bad2bc972a7a5a Mon Sep 17 00:00:00 2001
From: Dillon Varone <Dillon.Varone@amd.com>
Date: Thu, 11 Sep 2025 13:52:52 -0400
Subject: [PATCH 1405/1455] drm/amd/display: Remove wm_optimized_required

[WHY&HOW]
This is a redundant field that is identically to optimized_required, so
just replace it.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/core/dc.c                  | 5 +----
 drivers/gpu/drm/amd/display/dc/dc.h                       | 1 -
 drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c   | 2 +-
 drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c   | 8 ++++----
 drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c | 8 ++++----
 5 files changed, 10 insertions(+), 14 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 974d63af1c70..5f2d5638c819 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -460,7 +460,7 @@ bool dc_stream_adjust_vmin_vmax(struct dc *dc,
 	 * avoid conflicting with firmware updates.
 	 */
 	if (dc->ctx->dce_version > DCE_VERSION_MAX) {
-		if ((dc->optimized_required || dc->wm_optimized_required) &&
+		if (dc->optimized_required &&
 			(stream->adjust.v_total_max != adjust->v_total_max ||
 			stream->adjust.v_total_min != adjust->v_total_min)) {
 			stream->adjust.timing_adjust_pending = true;
@@ -2577,7 +2577,6 @@ void dc_post_update_surfaces_to_stream(struct dc *dc)
 	}
 
 	dc->optimized_required = false;
-	dc->wm_optimized_required = false;
 }
 
 bool dc_set_generic_gpio_for_stereo(bool enable,
@@ -3056,8 +3055,6 @@ enum surface_update_type dc_check_update_surfaces_for_stream(
 		} else if (memcmp(&dc->current_state->bw_ctx.bw.dcn.clk, &dc->clk_mgr->clks, offsetof(struct dc_clocks, prev_p_state_change_support)) != 0) {
 			dc->optimized_required = true;
 		}
-
-		dc->optimized_required |= dc->wm_optimized_required;
 	}
 
 	return type;
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 62d132bf3028..bd910b0a523a 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1736,7 +1736,6 @@ struct dc {
 
 	/* Require to optimize clocks and bandwidth for added/removed planes */
 	bool optimized_required;
-	bool wm_optimized_required;
 	bool idle_optimizations_allowed;
 	bool enable_c20_dtm_b0;
 
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
index 74f5e05f9cb4..e9fe97f0c4ea 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c
@@ -3347,7 +3347,7 @@ void dcn10_prepare_bandwidth(
 			context,
 			false);
 
-	dc->wm_optimized_required = hubbub->funcs->program_watermarks(hubbub,
+	dc->optimized_required = hubbub->funcs->program_watermarks(hubbub,
 			&context->bw_ctx.bw.dcn.watermarks,
 			dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 			true);
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index 95af58cbb92a..717d2c4daa88 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -2390,10 +2390,10 @@ void dcn20_prepare_bandwidth(
 	}
 
 	/* program dchubbub watermarks:
-	 * For assigning wm_optimized_required, use |= operator since we don't want
+	 * For assigning optimized_required, use |= operator since we don't want
 	 * to clear the value if the optimize has not happened yet
 	 */
-	dc->wm_optimized_required |= hubbub->funcs->program_watermarks(hubbub,
+	dc->optimized_required |= hubbub->funcs->program_watermarks(hubbub,
 					&context->bw_ctx.bw.dcn.watermarks,
 					dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 					false);
@@ -2406,10 +2406,10 @@ void dcn20_prepare_bandwidth(
 	if (hubbub->funcs->program_compbuf_size) {
 		if (context->bw_ctx.dml.ip.min_comp_buffer_size_kbytes) {
 			compbuf_size_kb = context->bw_ctx.dml.ip.min_comp_buffer_size_kbytes;
-			dc->wm_optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.dml.ip.min_comp_buffer_size_kbytes);
+			dc->optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.dml.ip.min_comp_buffer_size_kbytes);
 		} else {
 			compbuf_size_kb = context->bw_ctx.bw.dcn.compbuf_size_kb;
-			dc->wm_optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.bw.dcn.compbuf_size_kb);
+			dc->optimized_required |= (compbuf_size_kb != dc->current_state->bw_ctx.bw.dcn.compbuf_size_kb);
 		}
 
 		hubbub->funcs->program_compbuf_size(hubbub, compbuf_size_kb, false);
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index 1b0b772fc5dd..625653ce5556 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -1383,22 +1383,22 @@ void dcn401_prepare_bandwidth(struct dc *dc,
 			false);
 
 	/* program dchubbub watermarks:
-	 * For assigning wm_optimized_required, use |= operator since we don't want
+	 * For assigning optimized_required, use |= operator since we don't want
 	 * to clear the value if the optimize has not happened yet
 	 */
-	dc->wm_optimized_required |= hubbub->funcs->program_watermarks(hubbub,
+	dc->optimized_required |= hubbub->funcs->program_watermarks(hubbub,
 					&context->bw_ctx.bw.dcn.watermarks,
 					dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000,
 					false);
 	/* update timeout thresholds */
 	if (hubbub->funcs->program_arbiter) {
-		dc->wm_optimized_required |= hubbub->funcs->program_arbiter(hubbub, &context->bw_ctx.bw.dcn.arb_regs, false);
+		dc->optimized_required |= hubbub->funcs->program_arbiter(hubbub, &context->bw_ctx.bw.dcn.arb_regs, false);
 	}
 
 	/* decrease compbuf size */
 	if (hubbub->funcs->program_compbuf_segments) {
 		compbuf_size = context->bw_ctx.bw.dcn.arb_regs.compbuf_size;
-		dc->wm_optimized_required |= (compbuf_size != dc->current_state->bw_ctx.bw.dcn.arb_regs.compbuf_size);
+		dc->optimized_required |= (compbuf_size != dc->current_state->bw_ctx.bw.dcn.arb_regs.compbuf_size);
 
 		hubbub->funcs->program_compbuf_segments(hubbub, compbuf_size, false);
 	}

From 35bcc9168f3ce6416cbf3f776758be0937f84cb3 Mon Sep 17 00:00:00 2001
From: Charlene Liu <Charlene.Liu@amd.com>
Date: Thu, 11 Sep 2025 17:52:11 -0400
Subject: [PATCH 1406/1455] drm/amd/display: Insert dccg log for easy debug

[why]
Log for sequence tracking

Reviewed-by: Ovidiu (Ovi) Bunea <ovidiu.bunea@amd.com>
Reviewed-by: Yihan Zhu <yihan.zhu@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../amd/display/dc/dccg/dcn35/dcn35_dccg.c    | 24 ++++++++++++++++---
 1 file changed, 21 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
index 0ce9489ac6b7..de6d62401362 100644
--- a/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
+++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn35/dcn35_dccg.c
@@ -39,6 +39,7 @@
 
 #define CTX \
 	dccg_dcn->base.ctx
+#include "logger_types.h"
 #define DC_LOGGER \
 	dccg->ctx->logger
 
@@ -1136,7 +1137,7 @@ static void dcn35_set_dppclk_enable(struct dccg *dccg,
 	default:
 		break;
 	}
-	//DC_LOG_DEBUG("%s: dpp_inst(%d) DPPCLK_EN = %d\n", __func__, dpp_inst, enable);
+	DC_LOG_DEBUG("%s: dpp_inst(%d) DPPCLK_EN = %d\n", __func__, dpp_inst, enable);
 
 }
 
@@ -1406,6 +1407,10 @@ static void dccg35_set_dtbclk_dto(
 		 * PIPEx_DTO_SRC_SEL should not be programmed during DTBCLK update since OTG may still be on, and the
 		 * programming is handled in program_pix_clk() regardless, so it can be removed from here.
 		 */
+		DC_LOG_DEBUG("%s: OTG%d DTBCLK DTO enabled: pixclk_khz=%d, ref_dtbclk_khz=%d, req_dtbclk_khz=%d, phase=%d, modulo=%d\n",
+				__func__, params->otg_inst, params->pixclk_khz,
+				params->ref_dtbclk_khz, req_dtbclk_khz, phase, modulo);
+
 	} else {
 		switch (params->otg_inst) {
 		case 0:
@@ -1431,6 +1436,8 @@ static void dccg35_set_dtbclk_dto(
 
 		REG_WRITE(DTBCLK_DTO_MODULO[params->otg_inst], 0);
 		REG_WRITE(DTBCLK_DTO_PHASE[params->otg_inst], 0);
+
+		DC_LOG_DEBUG("%s: OTG%d DTBCLK DTO disabled\n", __func__, params->otg_inst);
 	}
 }
 
@@ -1475,6 +1482,8 @@ static void dccg35_set_dpstreamclk(
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	DC_LOG_DEBUG("%s: dp_hpo_inst(%d) DPSTREAMCLK_EN = %d, DPSTREAMCLK_SRC_SEL = %d\n",
+			__func__, dp_hpo_inst, (src == REFCLK) ? 0 : 1, otg_inst);
 }
 
 
@@ -1514,6 +1523,8 @@ static void dccg35_set_dpstreamclk_root_clock_gating(
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	DC_LOG_DEBUG("%s: dp_hpo_inst(%d) DPSTREAMCLK_ROOT_GATE_DISABLE = %d\n",
+			__func__, dp_hpo_inst, enable ? 1 : 0);
 }
 
 
@@ -1553,7 +1564,7 @@ static void dccg35_set_physymclk_root_clock_gating(
 		BREAK_TO_DEBUGGER();
 		return;
 	}
-	//DC_LOG_DEBUG("%s: dpp_inst(%d) PHYESYMCLK_ROOT_GATE_DISABLE:\n", __func__, phy_inst, enable ? 0 : 1);
+	DC_LOG_DEBUG("%s: dpp_inst(%d) PHYESYMCLK_ROOT_GATE_DISABLE: %d\n", __func__, phy_inst, enable ? 0 : 1);
 
 }
 
@@ -1626,6 +1637,8 @@ static void dccg35_set_physymclk(
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+	DC_LOG_DEBUG("%s: phy_inst(%d) PHYxSYMCLK_EN = %d, PHYxSYMCLK_SRC_SEL = %d\n",
+			__func__, phy_inst, force_enable ? 1 : 0, clk_src);
 }
 
 static void dccg35_set_valid_pixel_rate(
@@ -1673,6 +1686,7 @@ static void dccg35_dpp_root_clock_control(
 	}
 
 	dccg->dpp_clock_gated[dpp_inst] = !clock_on;
+	DC_LOG_DEBUG("%s: dpp_inst(%d) clock_on = %d\n", __func__, dpp_inst, clock_on);
 }
 
 static void dccg35_disable_symclk32_se(
@@ -1731,6 +1745,7 @@ static void dccg35_disable_symclk32_se(
 		BREAK_TO_DEBUGGER();
 		return;
 	}
+
 }
 
 static void dccg35_init_cb(struct dccg *dccg)
@@ -1738,7 +1753,6 @@ static void dccg35_init_cb(struct dccg *dccg)
 	(void)dccg;
 	/* Any RCG should be done when driver enter low power mode*/
 }
-
 void dccg35_init(struct dccg *dccg)
 {
 	int otg_inst;
@@ -1753,6 +1767,8 @@ void dccg35_init(struct dccg *dccg)
 		for (otg_inst = 0; otg_inst < 2; otg_inst++) {
 			dccg31_disable_symclk32_le(dccg, otg_inst);
 			dccg31_set_symclk32_le_root_clock_gating(dccg, otg_inst, false);
+			DC_LOG_DEBUG("%s: OTG%d SYMCLK32_LE disabled and root clock gating disabled\n",
+					__func__, otg_inst);
 		}
 
 //	if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se)
@@ -1765,6 +1781,8 @@ void dccg35_init(struct dccg *dccg)
 			dccg35_set_dpstreamclk(dccg, REFCLK, otg_inst,
 						otg_inst);
 			dccg35_set_dpstreamclk_root_clock_gating(dccg, otg_inst, false);
+			DC_LOG_DEBUG("%s: OTG%d DPSTREAMCLK disabled and root clock gating disabled\n",
+					__func__, otg_inst);
 		}
 
 /*

From f1fd8a9ac2aa5118f76baf28e6ca4d6962a485be Mon Sep 17 00:00:00 2001
From: Charlene Liu <Charlene.Liu@amd.com>
Date: Thu, 11 Sep 2025 19:20:45 -0400
Subject: [PATCH 1407/1455] drm/amd/display: Correct sw cache timing to ensure
 dispclk ramping

[why]
Current driver will cache the dispclk right after send cmd to pmfw,
but actual clock not reached yet.

Change to only cache the dispclk setting after HW reached to the real clock.
Also give some range as it might be in bypass clock setting.

Reviewed-by: Yihan Zhu <yihan.zhu@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c  | 30 +++++++++++++------
 1 file changed, 21 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
index 86edf11b8c5a..6fc8a7491684 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
@@ -387,6 +387,7 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	bool update_dispclk = false;
 	bool dpp_clock_lowered = false;
 	int all_active_disps = 0;
+	int actual_dppclk = 0;
 
 	if (dc->work_arounds.skip_clock_update)
 		return;
@@ -472,14 +473,13 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
 			dpp_clock_lowered = true;
-		clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz;
 		update_dppclk = true;
 	}
 
 	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz) &&
 	    (new_clocks->dispclk_khz > 0 || (safe_to_lower && display_count == 0))) {
 		int requested_dispclk_khz = new_clocks->dispclk_khz;
-
+		int actual_dispclk;
 		dcn35_disable_otg_wa(clk_mgr_base, context, safe_to_lower, true);
 
 		/* Clamp the requested clock to PMFW based on their limit. */
@@ -487,7 +487,11 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 			requested_dispclk_khz = dc->debug.min_disp_clk_khz;
 
 		dcn35_smu_set_dispclk(clk_mgr, requested_dispclk_khz);
-		clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
+		actual_dispclk = REG_READ(CLK1_CLK0_CURRENT_CNT);
+
+		/*pmfw might set bypass clock which is higher than hardmin*/
+		if (actual_dispclk >= new_clocks->dispclk_khz)
+			clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
 
 		dcn35_disable_otg_wa(clk_mgr_base, context, safe_to_lower, false);
 
@@ -505,13 +509,20 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	if (dpp_clock_lowered) {
 		// increase per DPP DTO before lowering global dppclk
 		dcn35_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower);
-		dcn35_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz);
+		dcn35_smu_set_dppclk(clk_mgr, new_clocks->dppclk_khz);
 	} else {
 		// increase global DPPCLK before lowering per DPP DTO
 		if (update_dppclk || update_dispclk)
-			dcn35_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz);
+			dcn35_smu_set_dppclk(clk_mgr, new_clocks->dppclk_khz);
 		dcn35_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower);
 	}
+	if (update_dppclk) {
+		actual_dppclk = REG_READ(CLK1_CLK1_CURRENT_CNT);
+
+		/*pmfw might set bypass clock which is higher than hardmin*/
+		if (actual_dppclk >= new_clocks->dppclk_khz)
+			clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz;
+	}
 
 	// notify PMFW of bandwidth per DPIA tunnel
 	if (dc->debug.notify_dpia_hr_bw)
@@ -551,7 +562,7 @@ static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr)
 	 * since fractional part is only 16 bit in register definition but is 32 bit
 	 * in our fix point definiton, need to shift left by 16 to obtain correct value
 	 */
-	pll_req.value |= fbmult_frac_val << 16;
+	pll_req.value |= (long long) fbmult_frac_val << 16;
 
 	/* multiply by REFCLK period */
 	pll_req = dc_fixpt_mul_int(pll_req, clk_mgr->dfs_ref_freq_khz);
@@ -778,7 +789,8 @@ static void dcn35_build_watermark_ranges(struct clk_bw_params *bw_params, struct
 			table->WatermarkRow[WM_DCFCLK][num_valid_sets].MaxClock = 0xFFFF;
 
 			/* Modify previous watermark range to cover up to max */
-			table->WatermarkRow[WM_DCFCLK][num_valid_sets - 1].MaxClock = 0xFFFF;
+			if (num_valid_sets > 0)
+				table->WatermarkRow[WM_DCFCLK][num_valid_sets - 1].MaxClock = 0xFFFF;
 		}
 		num_valid_sets++;
 	}
@@ -939,8 +951,8 @@ static void dcn35_clk_mgr_helper_populate_bw_params(struct clk_mgr_internal *clk
 	       is_valid_clock_value(min_dram_speed_mts));
 
 	/* dispclk and dppclk can be max at any voltage, same number of levels for both */
-	if (clock_table->NumDispClkLevelsEnabled <= NUM_DISPCLK_DPM_LEVELS &&
-	    clock_table->NumDispClkLevelsEnabled <= NUM_DPPCLK_DPM_LEVELS) {
+	if (clock_table->NumDispClkLevelsEnabled <= NUM_DISPCLK_DPM_LEVELS) {
+		/*numDispclk is the same as numDPPclk*/
 		max_dispclk = find_max_clk_value(clock_table->DispClocks,
 			clock_table->NumDispClkLevelsEnabled);
 		max_dppclk = find_max_clk_value(clock_table->DppClocks,

From 550038edef7f933591e10e32f94a5e9557b023b8 Mon Sep 17 00:00:00 2001
From: Charlene Liu <Charlene.Liu@amd.com>
Date: Fri, 12 Sep 2025 12:37:30 -0400
Subject: [PATCH 1408/1455] drm/amd/display: Revert "correct sw cache timing to
 ensure dispclk ramping"

[why]
Need consider SSC enabled case

This reverts commit f1fd8a9ac2aa5118f76baf28e6ca4d6962a485be.

Reviewed-by: Ovidiu (Ovi) Bunea <ovidiu.bunea@amd.com>
Reviewed-by: Chris Park <chris.park@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c  | 30 ++++++-------------
 1 file changed, 9 insertions(+), 21 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
index 6fc8a7491684..86edf11b8c5a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
@@ -387,7 +387,6 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	bool update_dispclk = false;
 	bool dpp_clock_lowered = false;
 	int all_active_disps = 0;
-	int actual_dppclk = 0;
 
 	if (dc->work_arounds.skip_clock_update)
 		return;
@@ -473,13 +472,14 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->base.clks.dppclk_khz)) {
 		if (clk_mgr->base.clks.dppclk_khz > new_clocks->dppclk_khz)
 			dpp_clock_lowered = true;
+		clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz;
 		update_dppclk = true;
 	}
 
 	if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr_base->clks.dispclk_khz) &&
 	    (new_clocks->dispclk_khz > 0 || (safe_to_lower && display_count == 0))) {
 		int requested_dispclk_khz = new_clocks->dispclk_khz;
-		int actual_dispclk;
+
 		dcn35_disable_otg_wa(clk_mgr_base, context, safe_to_lower, true);
 
 		/* Clamp the requested clock to PMFW based on their limit. */
@@ -487,11 +487,7 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 			requested_dispclk_khz = dc->debug.min_disp_clk_khz;
 
 		dcn35_smu_set_dispclk(clk_mgr, requested_dispclk_khz);
-		actual_dispclk = REG_READ(CLK1_CLK0_CURRENT_CNT);
-
-		/*pmfw might set bypass clock which is higher than hardmin*/
-		if (actual_dispclk >= new_clocks->dispclk_khz)
-			clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
+		clk_mgr_base->clks.dispclk_khz = new_clocks->dispclk_khz;
 
 		dcn35_disable_otg_wa(clk_mgr_base, context, safe_to_lower, false);
 
@@ -509,20 +505,13 @@ void dcn35_update_clocks(struct clk_mgr *clk_mgr_base,
 	if (dpp_clock_lowered) {
 		// increase per DPP DTO before lowering global dppclk
 		dcn35_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower);
-		dcn35_smu_set_dppclk(clk_mgr, new_clocks->dppclk_khz);
+		dcn35_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz);
 	} else {
 		// increase global DPPCLK before lowering per DPP DTO
 		if (update_dppclk || update_dispclk)
-			dcn35_smu_set_dppclk(clk_mgr, new_clocks->dppclk_khz);
+			dcn35_smu_set_dppclk(clk_mgr, clk_mgr_base->clks.dppclk_khz);
 		dcn35_update_clocks_update_dpp_dto(clk_mgr, context, safe_to_lower);
 	}
-	if (update_dppclk) {
-		actual_dppclk = REG_READ(CLK1_CLK1_CURRENT_CNT);
-
-		/*pmfw might set bypass clock which is higher than hardmin*/
-		if (actual_dppclk >= new_clocks->dppclk_khz)
-			clk_mgr_base->clks.dppclk_khz = new_clocks->dppclk_khz;
-	}
 
 	// notify PMFW of bandwidth per DPIA tunnel
 	if (dc->debug.notify_dpia_hr_bw)
@@ -562,7 +551,7 @@ static int get_vco_frequency_from_reg(struct clk_mgr_internal *clk_mgr)
 	 * since fractional part is only 16 bit in register definition but is 32 bit
 	 * in our fix point definiton, need to shift left by 16 to obtain correct value
 	 */
-	pll_req.value |= (long long) fbmult_frac_val << 16;
+	pll_req.value |= fbmult_frac_val << 16;
 
 	/* multiply by REFCLK period */
 	pll_req = dc_fixpt_mul_int(pll_req, clk_mgr->dfs_ref_freq_khz);
@@ -789,8 +778,7 @@ static void dcn35_build_watermark_ranges(struct clk_bw_params *bw_params, struct
 			table->WatermarkRow[WM_DCFCLK][num_valid_sets].MaxClock = 0xFFFF;
 
 			/* Modify previous watermark range to cover up to max */
-			if (num_valid_sets > 0)
-				table->WatermarkRow[WM_DCFCLK][num_valid_sets - 1].MaxClock = 0xFFFF;
+			table->WatermarkRow[WM_DCFCLK][num_valid_sets - 1].MaxClock = 0xFFFF;
 		}
 		num_valid_sets++;
 	}
@@ -951,8 +939,8 @@ static void dcn35_clk_mgr_helper_populate_bw_params(struct clk_mgr_internal *clk
 	       is_valid_clock_value(min_dram_speed_mts));
 
 	/* dispclk and dppclk can be max at any voltage, same number of levels for both */
-	if (clock_table->NumDispClkLevelsEnabled <= NUM_DISPCLK_DPM_LEVELS) {
-		/*numDispclk is the same as numDPPclk*/
+	if (clock_table->NumDispClkLevelsEnabled <= NUM_DISPCLK_DPM_LEVELS &&
+	    clock_table->NumDispClkLevelsEnabled <= NUM_DPPCLK_DPM_LEVELS) {
 		max_dispclk = find_max_clk_value(clock_table->DispClocks,
 			clock_table->NumDispClkLevelsEnabled);
 		max_dppclk = find_max_clk_value(clock_table->DppClocks,

From d43cc4ea1f9d720ab4bf06806f79260bfe981508 Mon Sep 17 00:00:00 2001
From: Leo Li <sunpeng.li@amd.com>
Date: Fri, 12 Sep 2025 11:01:50 -0400
Subject: [PATCH 1409/1455] drm/amd/display: Init DCN35 clocks from pre-os HW
 values

[Why]
We did not initialize dc clocks with boot-time hw values during init.
This lead to incorrect clock values in dc, causing `dcn35_update_clocks`
to make incorrect updates.

[How]
Correctly initialize DC with pre-os clk values from HW.
s/dump/save/ as that accurately reflects the purpose of the functions.

Fixes: 8774029f76b9 ("drm/amd/display: Add DCN35 CLK_MGR")
Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Leo Li <sunpeng.li@amd.com>
Signed-off-by: Fangzhi Zuo <Jerry.Zuo@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c  | 121 +++++++++++++++++-
 1 file changed, 119 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
index 86edf11b8c5a..b11383fba35f 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c
@@ -587,9 +587,118 @@ bool dcn35_are_clock_states_equal(struct dc_clocks *a,
 	return true;
 }
 
-static void dcn35_dump_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
+static void dcn35_save_clk_registers_internal(struct dcn35_clk_internal *internal, struct clk_mgr *clk_mgr_base)
+{
+	struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base);
+
+	// read dtbclk
+	internal->CLK1_CLK4_CURRENT_CNT = REG_READ(CLK1_CLK4_CURRENT_CNT);
+	internal->CLK1_CLK4_BYPASS_CNTL = REG_READ(CLK1_CLK4_BYPASS_CNTL);
+
+	// read dcfclk
+	internal->CLK1_CLK3_CURRENT_CNT = REG_READ(CLK1_CLK3_CURRENT_CNT);
+	internal->CLK1_CLK3_BYPASS_CNTL = REG_READ(CLK1_CLK3_BYPASS_CNTL);
+
+	// read dcf deep sleep divider
+	internal->CLK1_CLK3_DS_CNTL = REG_READ(CLK1_CLK3_DS_CNTL);
+	internal->CLK1_CLK3_ALLOW_DS = REG_READ(CLK1_CLK3_ALLOW_DS);
+
+	// read dppclk
+	internal->CLK1_CLK1_CURRENT_CNT = REG_READ(CLK1_CLK1_CURRENT_CNT);
+	internal->CLK1_CLK1_BYPASS_CNTL = REG_READ(CLK1_CLK1_BYPASS_CNTL);
+
+	// read dprefclk
+	internal->CLK1_CLK2_CURRENT_CNT = REG_READ(CLK1_CLK2_CURRENT_CNT);
+	internal->CLK1_CLK2_BYPASS_CNTL = REG_READ(CLK1_CLK2_BYPASS_CNTL);
+
+	// read dispclk
+	internal->CLK1_CLK0_CURRENT_CNT = REG_READ(CLK1_CLK0_CURRENT_CNT);
+	internal->CLK1_CLK0_BYPASS_CNTL = REG_READ(CLK1_CLK0_BYPASS_CNTL);
+}
+
+static void dcn35_save_clk_registers(struct clk_state_registers_and_bypass *regs_and_bypass,
 		struct clk_mgr_dcn35 *clk_mgr)
 {
+	struct dcn35_clk_internal internal = {0};
+	char *bypass_clks[5] = {"0x0 DFS", "0x1 REFCLK", "0x2 ERROR", "0x3 400 FCH", "0x4 600 FCH"};
+
+	dcn35_save_clk_registers_internal(&internal, &clk_mgr->base.base);
+
+	regs_and_bypass->dcfclk = internal.CLK1_CLK3_CURRENT_CNT / 10;
+	regs_and_bypass->dcf_deep_sleep_divider = internal.CLK1_CLK3_DS_CNTL / 10;
+	regs_and_bypass->dcf_deep_sleep_allow = internal.CLK1_CLK3_ALLOW_DS;
+	regs_and_bypass->dprefclk = internal.CLK1_CLK2_CURRENT_CNT / 10;
+	regs_and_bypass->dispclk = internal.CLK1_CLK0_CURRENT_CNT / 10;
+	regs_and_bypass->dppclk = internal.CLK1_CLK1_CURRENT_CNT / 10;
+	regs_and_bypass->dtbclk = internal.CLK1_CLK4_CURRENT_CNT / 10;
+
+	regs_and_bypass->dppclk_bypass = internal.CLK1_CLK1_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dppclk_bypass < 0 || regs_and_bypass->dppclk_bypass > 4)
+		regs_and_bypass->dppclk_bypass = 0;
+	regs_and_bypass->dcfclk_bypass = internal.CLK1_CLK3_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dcfclk_bypass < 0 || regs_and_bypass->dcfclk_bypass > 4)
+		regs_and_bypass->dcfclk_bypass = 0;
+	regs_and_bypass->dispclk_bypass = internal.CLK1_CLK0_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dispclk_bypass < 0 || regs_and_bypass->dispclk_bypass > 4)
+		regs_and_bypass->dispclk_bypass = 0;
+	regs_and_bypass->dprefclk_bypass = internal.CLK1_CLK2_BYPASS_CNTL & 0x0007;
+	if (regs_and_bypass->dprefclk_bypass < 0 || regs_and_bypass->dprefclk_bypass > 4)
+		regs_and_bypass->dprefclk_bypass = 0;
+
+	if (clk_mgr->base.base.ctx->dc->debug.pstate_enabled) {
+		DC_LOG_SMU("clk_type,clk_value,deepsleep_cntl,deepsleep_allow,bypass\n");
+
+		DC_LOG_SMU("dcfclk,%d,%d,%d,%s\n",
+				   regs_and_bypass->dcfclk,
+				   regs_and_bypass->dcf_deep_sleep_divider,
+				   regs_and_bypass->dcf_deep_sleep_allow,
+				   bypass_clks[(int) regs_and_bypass->dcfclk_bypass]);
+
+		DC_LOG_SMU("dprefclk,%d,N/A,N/A,%s\n",
+			regs_and_bypass->dprefclk,
+			bypass_clks[(int) regs_and_bypass->dprefclk_bypass]);
+
+		DC_LOG_SMU("dispclk,%d,N/A,N/A,%s\n",
+			regs_and_bypass->dispclk,
+			bypass_clks[(int) regs_and_bypass->dispclk_bypass]);
+
+		// REGISTER VALUES
+		DC_LOG_SMU("reg_name,value,clk_type");
+
+		DC_LOG_SMU("CLK1_CLK3_CURRENT_CNT,%d,dcfclk",
+				internal.CLK1_CLK3_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK4_CURRENT_CNT,%d,dtbclk",
+					internal.CLK1_CLK4_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK3_DS_CNTL,%d,dcf_deep_sleep_divider",
+					internal.CLK1_CLK3_DS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK3_ALLOW_DS,%d,dcf_deep_sleep_allow",
+					internal.CLK1_CLK3_ALLOW_DS);
+
+		DC_LOG_SMU("CLK1_CLK2_CURRENT_CNT,%d,dprefclk",
+					internal.CLK1_CLK2_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK0_CURRENT_CNT,%d,dispclk",
+					internal.CLK1_CLK0_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK1_CURRENT_CNT,%d,dppclk",
+					internal.CLK1_CLK1_CURRENT_CNT);
+
+		DC_LOG_SMU("CLK1_CLK3_BYPASS_CNTL,%d,dcfclk_bypass",
+					internal.CLK1_CLK3_BYPASS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK2_BYPASS_CNTL,%d,dprefclk_bypass",
+					internal.CLK1_CLK2_BYPASS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK0_BYPASS_CNTL,%d,dispclk_bypass",
+					internal.CLK1_CLK0_BYPASS_CNTL);
+
+		DC_LOG_SMU("CLK1_CLK1_BYPASS_CNTL,%d,dppclk_bypass",
+					internal.CLK1_CLK1_BYPASS_CNTL);
+
+	}
 }
 
 static bool dcn35_is_spll_ssc_enabled(struct clk_mgr *clk_mgr_base)
@@ -623,6 +732,7 @@ static void init_clk_states(struct clk_mgr *clk_mgr)
 void dcn35_init_clocks(struct clk_mgr *clk_mgr)
 {
 	struct clk_mgr_internal *clk_mgr_int = TO_CLK_MGR_INTERNAL(clk_mgr);
+	struct clk_mgr_dcn35 *clk_mgr_dcn35 = TO_CLK_MGR_DCN35(clk_mgr_int);
 
 	init_clk_states(clk_mgr);
 
@@ -633,6 +743,13 @@ void dcn35_init_clocks(struct clk_mgr *clk_mgr)
 	else
 		clk_mgr->dp_dto_source_clock_in_khz = clk_mgr->dprefclk_khz;
 
+	dcn35_save_clk_registers(&clk_mgr->boot_snapshot, clk_mgr_dcn35);
+
+	clk_mgr->clks.ref_dtbclk_khz =  clk_mgr->boot_snapshot.dtbclk * 10;
+	if (clk_mgr->boot_snapshot.dtbclk > 59000) {
+		/*dtbclk enabled based on */
+		clk_mgr->clks.dtbclk_en = true;
+	}
 }
 static struct clk_bw_params dcn35_bw_params = {
 	.vram_type = Ddr4MemType,
@@ -1323,7 +1440,7 @@ void dcn35_clk_mgr_construct(
 		dcn35_bw_params.wm_table = ddr5_wm_table;
 	}
 	/* Saved clocks configured at boot for debug purposes */
-	dcn35_dump_clk_registers(&clk_mgr->base.base.boot_snapshot, clk_mgr);
+	dcn35_save_clk_registers(&clk_mgr->base.base.boot_snapshot, clk_mgr);
 
 	clk_mgr->base.base.dprefclk_khz = dcn35_smu_get_dprefclk(&clk_mgr->base);
 	clk_mgr->base.base.clks.ref_dtbclk_khz = 600000;

From dff184cdba518cef0c0eb6ac8ed115cb83cd26a8 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 12 Sep 2025 16:15:44 -0400
Subject: [PATCH 1410/1455] drm/amd/display: [FW Promotion] Release 0.1.28.0

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
index 6db19e49a05c..92248224b713 100644
--- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
+++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h
@@ -4143,7 +4143,7 @@ struct dmub_cmd_replay_copy_settings_data {
 	 */
 	uint8_t hpo_link_enc_inst;
 	/**
-	 * Determines if fast sync in ultra sleep mode is enabled/disabled.
+	 * Determines if fast resync in ultra sleep mode is enabled/disabled.
 	 */
 	uint8_t replay_support_fast_resync_in_ultra_sleep_mode;
 	/**

From cceb54222ea00cec4960ad346a918920c24e75c5 Mon Sep 17 00:00:00 2001
From: Taimur Hassan <Syed.Hassan@amd.com>
Date: Fri, 12 Sep 2025 18:23:48 -0500
Subject: [PATCH 1411/1455] drm/amd/display: Promote DC to 3.2.351

- Disable stutter when programming watermarks on dcn32
- Improve brightness calculations
- Fix saving vbios clocks during init for DCN314
- Enable DTM 3 on DCN3.1+ dGPUs
- Add new ultra sleep field in DMUB
- Isolate DCN401 SMU functions
- Refactor and add logging of SMU functions.
- Add dccg logging
- Fix DCN35 clocks initialization

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Ivan Lipski <ivan.lipski@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index bd910b0a523a..79de318c85e5 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -55,7 +55,7 @@ struct aux_payload;
 struct set_config_cmd_payload;
 struct dmub_notification;
 
-#define DC_VER "3.2.350"
+#define DC_VER "3.2.351"
 
 /**
  * MAX_SURFACES - representative of the upper bound of surfaces that can be piped to a single CRTC

From 2f9c63883730a0bfecb086e6e59246933f936ca1 Mon Sep 17 00:00:00 2001
From: Melissa Wen <mwen@igalia.com>
Date: Thu, 11 Sep 2025 14:21:19 -0300
Subject: [PATCH 1412/1455] drm/amd/display: update color on atomic commit time

Use `atomic_commit_setup` to change the DC stream state. It's a
preparation to remove from `atomic_check` changes in CRTC color
components of DC stream state and prevent DC to commit TEST_ONLY
changes.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/4444
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Melissa Wen <mwen@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 36 ++++++++++++++++++-
 1 file changed, 35 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 271ea1615178..c0a0efdf44de 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -233,6 +233,7 @@ static int amdgpu_dm_encoder_init(struct drm_device *dev,
 
 static int amdgpu_dm_connector_get_modes(struct drm_connector *connector);
 
+static int amdgpu_dm_atomic_setup_commit(struct drm_atomic_state *state);
 static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state);
 
 static int amdgpu_dm_atomic_check(struct drm_device *dev,
@@ -3636,7 +3637,7 @@ static const struct drm_mode_config_funcs amdgpu_dm_mode_funcs = {
 
 static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = {
 	.atomic_commit_tail = amdgpu_dm_atomic_commit_tail,
-	.atomic_commit_setup = drm_dp_mst_atomic_setup_commit,
+	.atomic_commit_setup = amdgpu_dm_atomic_setup_commit,
 };
 
 static void update_connector_ext_caps(struct amdgpu_dm_connector *aconnector)
@@ -10362,6 +10363,39 @@ static void amdgpu_dm_update_hdcp(struct drm_atomic_state *state)
 	}
 }
 
+static int amdgpu_dm_atomic_setup_commit(struct drm_atomic_state *state)
+{
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *old_crtc_state, *new_crtc_state;
+	struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state;
+	int i, ret;
+
+	ret = drm_dp_mst_atomic_setup_commit(state);
+	if (ret)
+		return ret;
+
+	for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
+		dm_old_crtc_state = to_dm_crtc_state(old_crtc_state);
+		dm_new_crtc_state = to_dm_crtc_state(new_crtc_state);
+		/*
+		 * Color management settings. We also update color properties
+		 * when a modeset is needed, to ensure it gets reprogrammed.
+		 */
+		if (dm_new_crtc_state->base.active && dm_new_crtc_state->stream &&
+		    (dm_new_crtc_state->base.color_mgmt_changed ||
+		     dm_old_crtc_state->regamma_tf != dm_new_crtc_state->regamma_tf ||
+		     drm_atomic_crtc_needs_modeset(new_crtc_state))) {
+			ret = amdgpu_dm_update_crtc_color_mgmt(dm_new_crtc_state);
+			if (ret) {
+				drm_dbg_atomic(state->dev, "Failed to update color state\n");
+				return ret;
+			}
+		}
+	}
+
+	return 0;
+}
+
 /**
  * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation.
  * @state: The atomic state to commit

From 86a54e45fd91953e2b0e1aa15932112f135e4821 Mon Sep 17 00:00:00 2001
From: Rahul Kumar <rk0006818@gmail.com>
Date: Thu, 18 Sep 2025 12:42:00 +0530
Subject: [PATCH 1413/1455] drm/amdgpu: Use kmalloc_array() instead of
 kmalloc()

Documentation/process/deprecated.rst recommends against the use of
kmalloc with dynamic size calculations due to the risk of overflow and
smaller allocation being made than the caller was expecting.

Replace kmalloc() with kmalloc_array() in amdgpu_amdkfd_gfx_v10.c,
amdgpu_amdkfd_gfx_v10_3.c, amdgpu_amdkfd_gfx_v11.c and
amdgpu_amdkfd_gfx_v12.c to make the intended allocation size clearer
and avoid potential overflow issues.

Suggested-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Rahul Kumar <rk0006818@gmail.com>
Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c   | 4 ++--
 drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v12.c   | 4 ++--
 4 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
index 04ef0ca10541..0239114fb6c4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10.c
@@ -352,7 +352,7 @@ static int kgd_hqd_dump(struct amdgpu_device *adev,
 		(*dump)[i++][1] = RREG32_SOC15_IP(GC, addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
@@ -449,7 +449,7 @@ static int kgd_hqd_sdma_dump(struct amdgpu_device *adev,
 #undef HQD_N_REGS
 #define HQD_N_REGS (19+6+7+10)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
index 6d08bc2781a3..f2278a0937ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v10_3.c
@@ -338,7 +338,7 @@ static int hqd_dump_v10_3(struct amdgpu_device *adev,
 		(*dump)[i++][1] = RREG32_SOC15_IP(GC, addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
@@ -435,7 +435,7 @@ static int hqd_sdma_dump_v10_3(struct amdgpu_device *adev,
 #undef HQD_N_REGS
 #define HQD_N_REGS (19+6+7+12)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
index e0e6a6a49d90..aaccf0b9947d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v11.c
@@ -323,7 +323,7 @@ static int hqd_dump_v11(struct amdgpu_device *adev,
 		(*dump)[i++][1] = RREG32(addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
@@ -420,7 +420,7 @@ static int hqd_sdma_dump_v11(struct amdgpu_device *adev,
 #undef HQD_N_REGS
 #define HQD_N_REGS (7+11+1+12+12)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v12.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v12.c
index 6f0dc23c901b..e0ceab400b2d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v12.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v12.c
@@ -115,7 +115,7 @@ static int hqd_dump_v12(struct amdgpu_device *adev,
 		(*dump)[i++][1] = RREG32(addr);		\
 	} while (0)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 
@@ -146,7 +146,7 @@ static int hqd_sdma_dump_v12(struct amdgpu_device *adev,
 #undef HQD_N_REGS
 #define HQD_N_REGS (last_reg - first_reg + 1)
 
-	*dump = kmalloc(HQD_N_REGS*2*sizeof(uint32_t), GFP_KERNEL);
+	*dump = kmalloc_array(HQD_N_REGS, sizeof(**dump), GFP_KERNEL);
 	if (*dump == NULL)
 		return -ENOMEM;
 

From 2330437da0994321020777c605a2a8cb0ecb7001 Mon Sep 17 00:00:00 2001
From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Mon, 24 Mar 2025 15:10:44 +0800
Subject: [PATCH 1414/1455] drm/amd/ras: Add rascore status definition

Add rascore status definition.

V5:
  Merge the previous empty files.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/ras/rascore/Makefile      |  0
 .../gpu/drm/amd/ras/rascore/ras_core_status.h | 37 +++++++++++++++++++
 2 files changed, 37 insertions(+)
 create mode 100644 drivers/gpu/drm/amd/ras/rascore/Makefile
 create mode 100644 drivers/gpu/drm/amd/ras/rascore/ras_core_status.h

diff --git a/drivers/gpu/drm/amd/ras/rascore/Makefile b/drivers/gpu/drm/amd/ras/rascore/Makefile
new file mode 100644
index 000000000000..e69de29bb2d1
diff --git a/drivers/gpu/drm/amd/ras/rascore/ras_core_status.h b/drivers/gpu/drm/amd/ras/rascore/ras_core_status.h
new file mode 100644
index 000000000000..144fbe4ceb9a
--- /dev/null
+++ b/drivers/gpu/drm/amd/ras/rascore/ras_core_status.h
@@ -0,0 +1,37 @@
+/* SPDX-License-Identifier: MIT */
+/*
+ * Copyright 2025 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __RAS_CORE_STATUS_H__
+#define __RAS_CORE_STATUS_H__
+
+#define RAS_CORE_OK                       0
+#define RAS_CORE_NOT_SUPPORTED            248
+#define RAS_CORE_FAIL_ERROR_QUERY         249
+#define RAS_CORE_FAIL_ERROR_INJECTION     250
+#define RAS_CORE_FAIL_FATAL_RECOVERY      251
+#define RAS_CORE_FAIL_POISON_CONSUMPTION  252
+#define RAS_CORE_FAIL_POISON_CREATION     253
+#define RAS_CORE_FAIL_NO_VALID_BANKS      254
+#define RAS_CORE_GPU_IN_MODE1_RESET       255
+#endif

From 51cb93aa0c4a9bb126b76f6e9fd640d88de25cee Mon Sep 17 00:00:00 2001
From: Melissa Wen <mwen@igalia.com>
Date: Thu, 11 Sep 2025 14:21:20 -0300
Subject: [PATCH 1415/1455] drm/amd/display: change dc stream color settings
 only in atomic commit

Don't update DC stream color components during atomic check. The driver
will continue validating the new CRTC color state but will not change DC
stream color components. The DC stream color state will only be
programmed at commit time in the `atomic_setup_commit` stage.

It fixes gamma LUT loss reported by KDE users when changing brightness
quickly or changing Display settings (such as overscan) with nightlight
on and HDR. As KWin can do a test commit with color settings different
from those that should be applied in a non-test-only commit, if the
driver changes DC stream color state in atomic check, this state can be
eventually HW programmed in commit tail, instead of the respective state
set by the non-blocking commit.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4444
Reported-by: Xaver Hugl <xaver.hugl@gmail.com>
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c |  2 +-
 .../gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h |  2 +
 .../amd/display/amdgpu_dm/amdgpu_dm_color.c   | 86 ++++++++++++++-----
 3 files changed, 66 insertions(+), 24 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c0a0efdf44de..951eaeb7f793 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -11210,7 +11210,7 @@ skip_modeset:
 	if (dm_new_crtc_state->base.color_mgmt_changed ||
 	    dm_old_crtc_state->regamma_tf != dm_new_crtc_state->regamma_tf ||
 	    drm_atomic_crtc_needs_modeset(new_crtc_state)) {
-		ret = amdgpu_dm_update_crtc_color_mgmt(dm_new_crtc_state);
+		ret = amdgpu_dm_check_crtc_color_mgmt(dm_new_crtc_state, true);
 		if (ret)
 			goto fail;
 	}
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index c41c0ee4687f..009f206226f0 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -1054,6 +1054,8 @@ void amdgpu_dm_init_color_mod(void);
 int amdgpu_dm_create_color_properties(struct amdgpu_device *adev);
 int amdgpu_dm_verify_lut_sizes(const struct drm_crtc_state *crtc_state);
 int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc);
+int amdgpu_dm_check_crtc_color_mgmt(struct dm_crtc_state *crtc,
+				    bool check_only);
 int amdgpu_dm_update_plane_color_mgmt(struct dm_crtc_state *crtc,
 				      struct drm_plane_state *plane_state,
 				      struct dc_plane_state *dc_plane_state);
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index b0ef157f1333..a4ac6d442278 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -566,12 +566,11 @@ static int __set_output_tf(struct dc_transfer_func *func,
 	return res ? 0 : -ENOMEM;
 }
 
-static int amdgpu_dm_set_atomic_regamma(struct dc_stream_state *stream,
+static int amdgpu_dm_set_atomic_regamma(struct dc_transfer_func *out_tf,
 					const struct drm_color_lut *regamma_lut,
 					uint32_t regamma_size, bool has_rom,
 					enum dc_transfer_func_predefined tf)
 {
-	struct dc_transfer_func *out_tf = &stream->out_transfer_func;
 	int ret = 0;
 
 	if (regamma_size || tf != TRANSFER_FUNCTION_LINEAR) {
@@ -885,33 +884,33 @@ int amdgpu_dm_verify_lut_sizes(const struct drm_crtc_state *crtc_state)
 }
 
 /**
- * amdgpu_dm_update_crtc_color_mgmt: Maps DRM color management to DC stream.
+ * amdgpu_dm_check_crtc_color_mgmt: Check if DRM color props are programmable by DC.
  * @crtc: amdgpu_dm crtc state
+ * @check_only: only check color state without update dc stream
  *
- * With no plane level color management properties we're free to use any
- * of the HW blocks as long as the CRTC CTM always comes before the
- * CRTC RGM and after the CRTC DGM.
- *
- * - The CRTC RGM block will be placed in the RGM LUT block if it is non-linear.
- * - The CRTC DGM block will be placed in the DGM LUT block if it is non-linear.
- * - The CRTC CTM will be placed in the gamut remap block if it is non-linear.
+ * This function just verifies CRTC LUT sizes, if there is enough space for
+ * output transfer function and if its parameters can be calculated by AMD
+ * color module. It also adjusts some settings for programming CRTC degamma at
+ * plane stage, using plane DGM block.
  *
  * The RGM block is typically more fully featured and accurate across
  * all ASICs - DCE can't support a custom non-linear CRTC DGM.
  *
  * For supporting both plane level color management and CRTC level color
- * management at once we have to either restrict the usage of CRTC properties
- * or blend adjustments together.
+ * management at once we have to either restrict the usage of some CRTC
+ * properties or blend adjustments together.
  *
  * Returns:
- * 0 on success. Error code if setup fails.
+ * 0 on success. Error code if validation fails.
  */
-int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
+
+int amdgpu_dm_check_crtc_color_mgmt(struct dm_crtc_state *crtc,
+				    bool check_only)
 {
 	struct dc_stream_state *stream = crtc->stream;
 	struct amdgpu_device *adev = drm_to_adev(crtc->base.state->dev);
 	bool has_rom = adev->asic_type <= CHIP_RAVEN;
-	struct drm_color_ctm *ctm = NULL;
+	struct dc_transfer_func *out_tf;
 	const struct drm_color_lut *degamma_lut, *regamma_lut;
 	uint32_t degamma_size, regamma_size;
 	bool has_regamma, has_degamma;
@@ -940,6 +939,14 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 	crtc->cm_has_degamma = false;
 	crtc->cm_is_degamma_srgb = false;
 
+	if (check_only) {
+		out_tf = kvzalloc(sizeof(*out_tf), GFP_KERNEL);
+		if (!out_tf)
+			return -ENOMEM;
+	} else {
+		out_tf = &stream->out_transfer_func;
+	}
+
 	/* Setup regamma and degamma. */
 	if (is_legacy) {
 		/*
@@ -954,8 +961,8 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 		 * inverse color ramp in legacy userspace.
 		 */
 		crtc->cm_is_degamma_srgb = true;
-		stream->out_transfer_func.type = TF_TYPE_DISTRIBUTED_POINTS;
-		stream->out_transfer_func.tf = TRANSFER_FUNCTION_SRGB;
+		out_tf->type = TF_TYPE_DISTRIBUTED_POINTS;
+		out_tf->tf = TRANSFER_FUNCTION_SRGB;
 		/*
 		 * Note: although we pass has_rom as parameter here, we never
 		 * actually use ROM because the color module only takes the ROM
@@ -963,16 +970,12 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 		 *
 		 * See more in mod_color_calculate_regamma_params()
 		 */
-		r = __set_legacy_tf(&stream->out_transfer_func, regamma_lut,
+		r = __set_legacy_tf(out_tf, regamma_lut,
 				    regamma_size, has_rom);
-		if (r)
-			return r;
 	} else {
 		regamma_size = has_regamma ? regamma_size : 0;
-		r = amdgpu_dm_set_atomic_regamma(stream, regamma_lut,
+		r = amdgpu_dm_set_atomic_regamma(out_tf, regamma_lut,
 						 regamma_size, has_rom, tf);
-		if (r)
-			return r;
 	}
 
 	/*
@@ -981,6 +984,43 @@ int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
 	 * have to place the CTM in the OCSC in that case.
 	 */
 	crtc->cm_has_degamma = has_degamma;
+	if (check_only)
+		kvfree(out_tf);
+
+	return r;
+}
+
+/**
+ * amdgpu_dm_update_crtc_color_mgmt: Maps DRM color management to DC stream.
+ * @crtc: amdgpu_dm crtc state
+ *
+ * With no plane level color management properties we're free to use any
+ * of the HW blocks as long as the CRTC CTM always comes before the
+ * CRTC RGM and after the CRTC DGM.
+ *
+ * - The CRTC RGM block will be placed in the RGM LUT block if it is non-linear.
+ * - The CRTC DGM block will be placed in the DGM LUT block if it is non-linear.
+ * - The CRTC CTM will be placed in the gamut remap block if it is non-linear.
+ *
+ * The RGM block is typically more fully featured and accurate across
+ * all ASICs - DCE can't support a custom non-linear CRTC DGM.
+ *
+ * For supporting both plane level color management and CRTC level color
+ * management at once we have to either restrict the usage of CRTC properties
+ * or blend adjustments together.
+ *
+ * Returns:
+ * 0 on success. Error code if setup fails.
+ */
+int amdgpu_dm_update_crtc_color_mgmt(struct dm_crtc_state *crtc)
+{
+	struct dc_stream_state *stream = crtc->stream;
+	struct drm_color_ctm *ctm = NULL;
+	int ret;
+
+	ret = amdgpu_dm_check_crtc_color_mgmt(crtc, false);
+	if (ret)
+		return ret;
 
 	/* Setup CRTC CTM. */
 	if (crtc->base.ctm) {

From 43f06e8165c4f6e16ab32ede845171ac66d4eaaa Mon Sep 17 00:00:00 2001
From: Kuan-Wei Chiu <visitorckw@gmail.com>
Date: Tue, 9 Sep 2025 17:20:57 +0800
Subject: [PATCH 1416/1455] drm/amd/display: Optimize remove_duplicates() from
 O(N^2) to O(N)

Replace the previous O(N^2) implementation of remove_duplicates() with
a O(N) version using a fast/slow pointer approach. The new version
keeps only the first occurrence of each element and compacts the array
in place, improving efficiency without changing functionality.

Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.c   | 21 ++++++++++---------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.c
index e763c8e45da8..1b9579a32ff2 100644
--- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.c
+++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn3.c
@@ -48,18 +48,19 @@ static void set_reserved_time_on_all_planes_with_stream_index(struct display_con
 
 static void remove_duplicates(double *list_a, int *list_a_size)
 {
-	int cur_element = 0;
-	// For all elements b[i] in list_b[]
-	while (cur_element < *list_a_size - 1) {
-		if (list_a[cur_element] == list_a[cur_element + 1]) {
-			for (int j = cur_element + 1; j < *list_a_size - 1; j++) {
-				list_a[j] = list_a[j + 1];
-			}
-			*list_a_size = *list_a_size - 1;
-		} else {
-			cur_element++;
+	int j = 0;
+
+	if (*list_a_size == 0)
+		return;
+
+	for (int i = 1; i < *list_a_size; i++) {
+		if (list_a[j] != list_a[i]) {
+			j++;
+			list_a[j] = list_a[i];
 		}
 	}
+
+	*list_a_size = j + 1;
 }
 
 static bool increase_mpc_combine_factor(unsigned int *mpc_combine_factor, unsigned int limit)

From 752e6f283ec59ae007aa15a93d5a4b2eefa8cec9 Mon Sep 17 00:00:00 2001
From: Melissa Wen <mwen@igalia.com>
Date: Mon, 1 Sep 2025 18:51:05 -0300
Subject: [PATCH 1417/1455] drm/amd/display: remove output_tf_change flag

Remove this flag as the driver stopped managing it individually since
commit a4056c2a6344 ("drm/amd/display: use HW hdr mult for brightness
boost"). After some back and forth it was reintroduced as a condition to
`set_output_transfer_func()` in [1]. Without direct management, this
flag only changes value when all surface update flags are set true on
UPDATE_TYPE_FULL with no output TF status meaning.

Fixes: bb622e0c0044 ("drm/amd/display: program output tf when required") [1]
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/dc/dc.h                       | 1 -
 drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c   | 6 ++----
 drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c | 6 ++----
 3 files changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 79de318c85e5..98f0b6b3c213 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -1392,7 +1392,6 @@ union surface_update_flags {
 		uint32_t in_transfer_func_change:1;
 		uint32_t input_csc_change:1;
 		uint32_t coeff_reduction_change:1;
-		uint32_t output_tf_change:1;
 		uint32_t pixel_format_change:1;
 		uint32_t plane_size_change:1;
 		uint32_t gamut_remap_change:1;
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
index 717d2c4daa88..9477c9f9e196 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c
@@ -1982,10 +1982,8 @@ static void dcn20_program_pipe(
 	 * updating on slave planes
 	 */
 	if (pipe_ctx->update_flags.bits.enable ||
-		pipe_ctx->update_flags.bits.plane_changed ||
-		pipe_ctx->stream->update_flags.bits.out_tf ||
-		(pipe_ctx->plane_state &&
-			pipe_ctx->plane_state->update_flags.bits.output_tf_change))
+	    pipe_ctx->update_flags.bits.plane_changed ||
+	    pipe_ctx->stream->update_flags.bits.out_tf)
 		hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream);
 
 	/* If the pipe has been enabled or has a different opp, we
diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
index 625653ce5556..7c276c319086 100644
--- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
+++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c
@@ -2032,10 +2032,8 @@ void dcn401_program_pipe(
 	 * updating on slave planes
 	 */
 	if (pipe_ctx->update_flags.bits.enable ||
-		pipe_ctx->update_flags.bits.plane_changed ||
-		pipe_ctx->stream->update_flags.bits.out_tf ||
-		(pipe_ctx->plane_state &&
-			pipe_ctx->plane_state->update_flags.bits.output_tf_change))
+	    pipe_ctx->update_flags.bits.plane_changed ||
+	    pipe_ctx->stream->update_flags.bits.out_tf)
 		hws->funcs.set_output_transfer_func(dc, pipe_ctx, pipe_ctx->stream);
 
 	/* If the pipe has been enabled or has a different opp, we

From 0c1f3fe9a5f899ac95114e68959a35454af1523d Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 2 May 2025 16:22:45 -0400
Subject: [PATCH 1418/1455] Documentation: add initial documenation for user
 queues

Add an initial documentation page for user mode queues.

Reviewed-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 Documentation/gpu/amdgpu/index.rst |   1 +
 Documentation/gpu/amdgpu/userq.rst | 203 +++++++++++++++++++++++++++++
 2 files changed, 204 insertions(+)
 create mode 100644 Documentation/gpu/amdgpu/userq.rst

diff --git a/Documentation/gpu/amdgpu/index.rst b/Documentation/gpu/amdgpu/index.rst
index bb2894b5edaf..45523e9860fc 100644
--- a/Documentation/gpu/amdgpu/index.rst
+++ b/Documentation/gpu/amdgpu/index.rst
@@ -12,6 +12,7 @@ Next (GCN), Radeon DNA (RDNA), and Compute DNA (CDNA) architectures.
    module-parameters
    gc/index
    display/index
+   userq
    flashing
    xgmi
    ras
diff --git a/Documentation/gpu/amdgpu/userq.rst b/Documentation/gpu/amdgpu/userq.rst
new file mode 100644
index 000000000000..ca3ea71f7888
--- /dev/null
+++ b/Documentation/gpu/amdgpu/userq.rst
@@ -0,0 +1,203 @@
+==================
+ User Mode Queues
+==================
+
+Introduction
+============
+
+Similar to the KFD, GPU engine queues move into userspace.  The idea is to let
+user processes manage their submissions to the GPU engines directly, bypassing
+IOCTL calls to the driver to submit work.  This reduces overhead and also allows
+the GPU to submit work to itself.  Applications can set up work graphs of jobs
+across multiple GPU engines without needing trips through the CPU.
+
+UMDs directly interface with firmware via per application shared memory areas.
+The main vehicle for this is queue.  A queue is a ring buffer with a read
+pointer (rptr) and a write pointer (wptr).  The UMD writes IP specific packets
+into the queue and the firmware processes those packets, kicking off work on the
+GPU engines.  The CPU in the application (or another queue or device) updates
+the wptr to tell the firmware how far into the ring buffer to process packets
+and the rtpr provides feedback to the UMD on how far the firmware has progressed
+in executing those packets.  When the wptr and the rptr are equal, the queue is
+idle.
+
+Theory of Operation
+===================
+
+The various engines on modern AMD GPUs support multiple queues per engine with a
+scheduling firmware which handles dynamically scheduling user queues on the
+available hardware queue slots.  When the number of user queues outnumbers the
+available hardware queue slots, the scheduling firmware dynamically maps and
+unmaps queues based on priority and time quanta.  The state of each user queue
+is managed in the kernel driver in an MQD (Memory Queue Descriptor).  This is a
+buffer in GPU accessible memory that stores the state of a user queue.  The
+scheduling firmware uses the MQD to load the queue state into an HQD (Hardware
+Queue Descriptor) when a user queue is mapped.  Each user queue requires a
+number of additional buffers which represent the ring buffer and any metadata
+needed by the engine for runtime operation.  On most engines this consists of
+the ring buffer itself, a rptr buffer (where the firmware will shadow the rptr
+to userspace), a wptr buffer (where the application will write the wptr for the
+firmware to fetch it), and a doorbell.  A doorbell is a piece of one of the
+device's MMIO BARs which can be mapped to specific user queues.  When the
+application writes to the doorbell, it will signal the firmware to take some
+action. Writing to the doorbell wakes the firmware and causes it to fetch the
+wptr and start processing the packets in the queue. Each 4K page of the doorbell
+BAR supports specific offset ranges for specific engines.  The doorbell of a
+queue must be mapped into the aperture aligned to the IP used by the queue
+(e.g., GFX, VCN, SDMA, etc.).  These doorbell apertures are set up via NBIO
+registers.  Doorbells are 32 bit or 64 bit (depending on the engine) chunks of
+the doorbell BAR.  A 4K doorbell page provides 512 64-bit doorbells for up to
+512 user queues.  A subset of each page is reserved for each IP type supported
+on the device.  The user can query the doorbell ranges for each IP via the INFO
+IOCTL.  See the IOCTL Interfaces section for more information.
+
+When an application wants to create a user queue, it allocates the necessary
+buffers for the queue (ring buffer, wptr and rptr, context save areas, etc.).
+These can be separate buffers or all part of one larger buffer.  The application
+would map the buffer(s) into its GPUVM and use the GPU virtual addresses of for
+the areas of memory they want to use for the user queue.  They would also
+allocate a doorbell page for the doorbells used by the user queues.  The
+application would then populate the MQD in the USERQ IOCTL structure with the
+GPU virtual addresses and doorbell index they want to use.  The user can also
+specify the attributes for the user queue (priority, whether the queue is secure
+for protected content, etc.).  The application would then call the USERQ
+CREATE IOCTL to create the queue using the specified MQD details in the IOCTL.
+The kernel driver then validates the MQD provided by the application and
+translates the MQD into the engine specific MQD format for the IP.  The IP
+specific MQD would be allocated and the queue would be added to the run list
+maintained by the scheduling firmware.  Once the queue has been created, the
+application can write packets directly into the queue, update the wptr, and
+write to the doorbell offset to kick off work in the user queue.
+
+When the application is done with the user queue, it would call the USERQ
+FREE IOCTL to destroy it.  The kernel driver would preempt the queue and
+remove it from the scheduling firmware's run list.  Then the IP specific MQD
+would be freed and the user queue state would be cleaned up.
+
+Some engines may require the aggregated doorbell too if the engine does not
+support doorbells from unmapped queues.  The aggregated doorbell is a special
+page of doorbell space which wakes the scheduler.  In cases where the engine may
+be oversubscribed, some queues may not be mapped.  If the doorbell is rung when
+the queue is not mapped, the engine firmware may miss the request.  Some
+scheduling firmware may work around this by polling wptr shadows when the
+hardware is oversubscribed, other engines may support doorbell updates from
+unmapped queues.  In the event that one of these options is not available, the
+kernel driver will map a page of aggregated doorbell space into each GPUVM
+space.  The UMD will then update the doorbell and wptr as normal and then write
+to the aggregated doorbell as well.
+
+Special Packets
+---------------
+
+In order to support legacy implicit synchronization, as well as mixed user and
+kernel queues, we need a synchronization mechanism that is secure.  Because
+kernel queues or memory management tasks depend on kernel fences, we need a way
+for user queues to update memory that the kernel can use for a fence, that can't
+be messed with by a bad actor.  To support this, we've added a protected fence
+packet.  This packet works by writing a monotonically increasing value to
+a memory location that only privileged clients have write access to. User
+queues only have read access.  When this packet is executed, the memory location
+is updated and other queues (kernel or user) can see the results.  The
+user application would submit this packet in their command stream.  The actual
+packet format varies from IP to IP (GFX/Compute, SDMA, VCN, etc.), but the
+behavior is the same.  The packet submission is handled in userspace.  The
+kernel driver sets up the privileged memory used for each user queue when it
+sets the queues up when the application creates them.
+
+
+Memory Management
+=================
+
+It is assumed that all buffers mapped into the GPUVM space for the process are
+valid when engines on the GPU are running.  The kernel driver will only allow
+user queues to run when all buffers are mapped.  If there is a memory event that
+requires buffer migration, the kernel driver will preempt the user queues,
+migrate buffers to where they need to be, update the GPUVM page tables and
+invaldidate the TLB, and then resume the user queues.
+
+Interaction with Kernel Queues
+==============================
+
+Depending on the IP and the scheduling firmware, you can enable kernel queues
+and user queues at the same time, however, you are limited by the HQD slots.
+Kernel queues are always mapped so any work that goes into kernel queues will
+take priority.  This limits the available HQD slots for user queues.
+
+Not all IPs will support user queues on all GPUs.  As such, UMDs will need to
+support both user queues and kernel queues depending on the IP.  For example, a
+GPU may support user queues for GFX, compute, and SDMA, but not for VCN, JPEG,
+and VPE.  UMDs need to support both.  The kernel driver provides a way to
+determine if user queues and kernel queues are supported on a per IP basis.
+UMDs can query this information via the INFO IOCTL and determine whether to use
+kernel queues or user queues for each IP.
+
+Queue Resets
+============
+
+For most engines, queues can be reset individually.  GFX, compute, and SDMA
+queues can be reset individually.  When a hung queue is detected, it can be
+reset either via the scheduling firmware or MMIO.  Since there are no kernel
+fences for most user queues, they will usually only be detected when some other
+event happens; e.g., a memory event which requires migration of buffers.  When
+the queues are preempted, if the queue is hung, the preemption will fail.
+Driver will then look up the queues that failed to preempt and reset them and
+record which queues are hung.
+
+On the UMD side, we will add a USERQ QUERY_STATUS IOCTL to query the queue
+status.  UMD will provide the queue id in the IOCTL and the kernel driver
+will check if it has already recorded the queue as hung (e.g., due to failed
+peemption) and report back the status.
+
+IOCTL Interfaces
+================
+
+GPU virtual addresses used for queues and related data (rptrs, wptrs, context
+save areas, etc.) should be validated by the kernel mode driver to prevent the
+user from specifying invalid GPU virtual addresses.  If the user provides
+invalid GPU virtual addresses or doorbell indicies, the IOCTL should return an
+error message.  These buffers should also be tracked in the kernel driver so
+that if the user attempts to unmap the buffer(s) from the GPUVM, the umap call
+would return an error.
+
+INFO
+----
+There are several new INFO queries related to user queues in order to query the
+size of user queue meta data needed for a user queue (e.g., context save areas
+or shadow buffers), whether kernel or user queues or both are supported
+for each IP type, and the offsets for each IP type in each doorbell page.
+
+USERQ
+-----
+The USERQ IOCTL is used for creating, freeing, and querying the status of user
+queues.  It supports 3 opcodes:
+
+1. CREATE - Create a user queue.  The application provides an MQD-like structure
+   that defines the type of queue and associated metadata and flags for that
+   queue type.  Returns the queue id.
+2. FREE - Free a user queue.
+3. QUERY_STATUS - Query that status of a queue.  Used to check if the queue is
+   healthy or not.  E.g., if the queue has been reset. (WIP)
+
+USERQ_SIGNAL
+------------
+The USERQ_SIGNAL IOCTL is used to provide a list of sync objects to be signaled.
+
+USERQ_WAIT
+----------
+The USERQ_WAIT IOCTL is used to provide a list of sync object to be waited on.
+
+Kernel and User Queues
+======================
+
+In order to properly validate and test performance, we have a driver option to
+select what type of queues are enabled (kernel queues, user queues or both).
+The user_queue driver parameter allows you to enable kernel queues only (0),
+user queues and kernel queues (1), and user queues only (2).  Enabling user
+queues only will free up static queue assignments that would otherwise be used
+by kernel queues for use by the scheduling firmware.  Some kernel queues are
+required for kernel driver operation and they will always be created.  When the
+kernel queues are not enabled, they are not registered with the drm scheduler
+and the CS IOCTL will reject any incoming command submissions which target those
+queue types.  Kernel queues only mirrors the behavior on all existing GPUs.
+Enabling both queues allows for backwards compatibility with old userspace while
+still supporting user queues.

From eb6910cdaa8a1b5af11df6fd2e9a0b2b9f72a028 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 6 Aug 2025 16:03:13 +0800
Subject: [PATCH 1419/1455] drm/amdgpu: Refactor VCN v5.0.1 HW init into
 separate instance function

Split the per-instance initialization code from vcn_v5_0_1_hw_init()
into a new vcn_v5_0_1_hw_init_inst() function. This improves code
organization by:

1. Separating the instance-specific initialization logic
2. Making the main init function more readable
3. Following the pattern used in queue reset

The SR-IOV specific initialization remains in the main function since
it has different requirements.

Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 27 +++++++++++++++++--------
 1 file changed, 19 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 9c281ba6bced..3677ea9ffa43 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -253,6 +253,23 @@ static int vcn_v5_0_1_sw_fini(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
+static int vcn_v5_0_1_hw_init_inst(struct amdgpu_device *adev, int i)
+{
+	struct amdgpu_ring *ring;
+	int vcn_inst;
+
+	vcn_inst = GET_INST(VCN, i);
+	ring = &adev->vcn.inst[i].ring_enc[0];
+
+	if (ring->use_doorbell)
+		adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
+			((adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
+			11 * vcn_inst),
+			adev->vcn.inst[i].aid_id);
+
+	return 0;
+}
+
 /**
  * vcn_v5_0_1_hw_init - start and test VCN block
  *
@@ -264,7 +281,7 @@ static int vcn_v5_0_1_hw_init(struct amdgpu_ip_block *ip_block)
 {
 	struct amdgpu_device *adev = ip_block->adev;
 	struct amdgpu_ring *ring;
-	int i, r, vcn_inst;
+	int i, r;
 
 	if (amdgpu_sriov_vf(adev)) {
 		r = vcn_v5_0_1_start_sriov(adev);
@@ -282,14 +299,8 @@ static int vcn_v5_0_1_hw_init(struct amdgpu_ip_block *ip_block)
 		if (RREG32_SOC15(VCN, GET_INST(VCN, 0), regVCN_RRMT_CNTL) & 0x100)
 			adev->vcn.caps |= AMDGPU_VCN_CAPS(RRMT_ENABLED);
 		for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
-			vcn_inst = GET_INST(VCN, i);
 			ring = &adev->vcn.inst[i].ring_enc[0];
-
-			if (ring->use_doorbell)
-				adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell,
-					((adev->doorbell_index.vcn.vcn_ring0_1 << 1) +
-					11 * vcn_inst),
-					adev->vcn.inst[i].aid_id);
+			vcn_v5_0_1_hw_init_inst(adev, i);
 
 			/* Re-init fw_shared, if required */
 			vcn_v5_0_1_fw_shared_init(adev, i);

From dc704458dd31ba45bf9cac3fe3e3ab546dabed20 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 6 Aug 2025 16:20:28 +0800
Subject: [PATCH 1420/1455] drm/amdgpu: Add ring reset support for VCN v5.0.1

Implement the ring reset callback for VCN v5.0.1 to properly handle
hardware recovery when encountering GPU hangs. The new functionality:

1. Adds vcn_v5_0_1_ring_reset() function that:
   - Prepares for reset using amdgpu_ring_reset_helper_begin()
   - Performs VCN instance reset via amdgpu_dpm_reset_vcn()
   - Re-initializes hardware through vcn_v5_0_1_hw_init_inst()
   - Restarts DPG mode with vcn_v5_0_1_start_dpg_mode()
   - Completes reset with amdgpu_ring_reset_helper_end()

2. Hooks the reset function into the unified ring functions via:
   - Adding .reset = vcn_v5_0_1_ring_reset to vcn_v5_0_1_unified_ring_vm_funcs

3. Maintains existing behavior for SR-IOV VF cases by checking RRMT status

This provides proper hardware recovery capabilities for VCN 5.0.1 IP block
during fault conditions, matching functionality available in other VCN versions.

v2: Remove the RRMT_ENABLED cap setting in the reset function
    and replace adev->vcn.inst[ring->me].indirect_sram with vinst->indirect_sram (Lijo)

Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 26 +++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 3677ea9ffa43..11b931153ea1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -1284,6 +1284,31 @@ static void vcn_v5_0_1_unified_ring_set_wptr(struct amdgpu_ring *ring)
 	}
 }
 
+static int vcn_v5_0_1_ring_reset(struct amdgpu_ring *ring,
+				 unsigned int vmid,
+				 struct amdgpu_fence *timedout_fence)
+{
+	int r = 0;
+	int vcn_inst;
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vcn_inst *vinst = &adev->vcn.inst[ring->me];
+
+	amdgpu_ring_reset_helper_begin(ring, timedout_fence);
+
+	vcn_inst = GET_INST(VCN, ring->me);
+	r = amdgpu_dpm_reset_vcn(adev, 1 << vcn_inst);
+
+	if (r) {
+		DRM_DEV_ERROR(adev->dev, "VCN reset fail : %d\n", r);
+		return r;
+	}
+
+	vcn_v5_0_1_hw_init_inst(adev, ring->me);
+	vcn_v5_0_1_start_dpg_mode(vinst, vinst->indirect_sram);
+
+	return amdgpu_ring_reset_helper_end(ring, timedout_fence);
+}
+
 static const struct amdgpu_ring_funcs vcn_v5_0_1_unified_ring_vm_funcs = {
 	.type = AMDGPU_RING_TYPE_VCN_ENC,
 	.align_mask = 0x3f,
@@ -1312,6 +1337,7 @@ static const struct amdgpu_ring_funcs vcn_v5_0_1_unified_ring_vm_funcs = {
 	.emit_wreg = vcn_v4_0_3_enc_ring_emit_wreg,
 	.emit_reg_wait = vcn_v4_0_3_enc_ring_emit_reg_wait,
 	.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+	.reset = vcn_v5_0_1_ring_reset,
 };
 
 /**

From 5886090032ec891ccd6bf3c7e9d1ae3fe43ecc34 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Fri, 15 Aug 2025 23:44:11 +0800
Subject: [PATCH 1421/1455] drm/amdgpu: Move VCN reset mask setup to late_init
 for VCN 5.0.1

This patch moves the initialization of the VCN supported_reset mask from
sw_init to a new late_init function for VCN 5.0.1. The change ensures
that all necessary hardware and firmware initialization is complete
before determining the supported reset types.

Key changes:
- Added vcn_v5_0_1_late_init() function to handle late initialization
- Moved supported_reset mask setup from sw_init to late_init
- Added check for per-queue reset support via amdgpu_dpm_reset_vcn_is_supported()
- Updated ip_funcs to use the new late_init function

This change helps ensure proper reset behavior by waiting until all
dependencies are initialized before determining available reset types.

Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Ruili Ji <ruiliji2@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c | 25 ++++++++++++++++++++-----
 1 file changed, 20 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
index 11b931153ea1..714350cabf2f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_1.c
@@ -113,6 +113,25 @@ static int vcn_v5_0_1_early_init(struct amdgpu_ip_block *ip_block)
 	return 0;
 }
 
+static int vcn_v5_0_1_late_init(struct amdgpu_ip_block *ip_block)
+{
+	struct amdgpu_device *adev = ip_block->adev;
+
+	adev->vcn.supported_reset =
+		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
+
+	switch (amdgpu_ip_version(adev, MP0_HWIP, 0)) {
+	case IP_VERSION(13, 0, 12):
+		if ((adev->psp.sos.fw_version >= 0x00450025) && amdgpu_dpm_reset_vcn_is_supported(adev))
+			adev->vcn.supported_reset |= AMDGPU_RESET_TYPE_PER_QUEUE;
+		break;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
 static void vcn_v5_0_1_fw_shared_init(struct amdgpu_device *adev, int inst_idx)
 {
 	struct amdgpu_vcn5_fw_shared *fw_shared;
@@ -187,10 +206,6 @@ static int vcn_v5_0_1_sw_init(struct amdgpu_ip_block *ip_block)
 		vcn_v5_0_1_fw_shared_init(adev, i);
 	}
 
-	/* TODO: Add queue reset mask when FW fully supports it */
-	adev->vcn.supported_reset =
-		amdgpu_get_soft_full_reset_mask(&adev->vcn.inst[0].ring_enc[0]);
-
 	if (amdgpu_sriov_vf(adev)) {
 		r = amdgpu_virt_alloc_mm_table(adev);
 		if (r)
@@ -1541,7 +1556,7 @@ static void vcn_v5_0_1_set_irq_funcs(struct amdgpu_device *adev)
 static const struct amd_ip_funcs vcn_v5_0_1_ip_funcs = {
 	.name = "vcn_v5_0_1",
 	.early_init = vcn_v5_0_1_early_init,
-	.late_init = NULL,
+	.late_init = vcn_v5_0_1_late_init,
 	.sw_init = vcn_v5_0_1_sw_init,
 	.sw_fini = vcn_v5_0_1_sw_fini,
 	.hw_init = vcn_v5_0_1_hw_init,

From 4c709ccc478b4b711ee071146aae98eb7fa9f625 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Mon, 22 Sep 2025 09:40:51 +0800
Subject: [PATCH 1422/1455] drm/amd/pm: Add VCN reset message support for SMU
 v13.0.12

This commit adds support for VCN reset functionality in SMU v13.0.12 by:

1. Adding two new PPSMC messages in smu_v13_0_12_ppsmc.h:
   - PPSMC_MSG_ResetVCN (0x5E)
   - Updates PPSMC_Message_Count to 0x5F to account for new messages

2. Adding message mapping for ResetVCN in smu_v13_0_12_ppt.c:
   - Maps SMU_MSG_ResetVCN to PPSMC_MSG_ResetVCN

These changes enable proper VCN reset handling through the SMU firmware
interface for compatible AMD GPUs.

v2: Added fw version check to support vcn queue reset.

Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Sonny Jiang <sonny.jiang@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h | 3 ++-
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c         | 1 +
 drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c          | 3 +++
 3 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h
index aff2776a8b6f..4b066c42e0ec 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_12_ppsmc.h
@@ -120,7 +120,8 @@
 #define PPSMC_MSG_GetBadPageSeverity                0x5B
 #define PPSMC_MSG_GetSystemMetricsTable             0x5C
 #define PPSMC_MSG_GetSystemMetricsVersion           0x5D
-#define PPSMC_Message_Count                         0x5E
+#define PPSMC_MSG_ResetVCN                          0x5E
+#define PPSMC_Message_Count                         0x5F
 
 //PPSMC Reset Types for driver msg argument
 #define PPSMC_RESET_TYPE_DRIVER_MODE_1_RESET        0x1
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
index 1842a33b2bce..cb3fea9e8cf3 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_12_ppt.c
@@ -136,6 +136,7 @@ const struct cmn2asic_msg_mapping smu_v13_0_12_message_map[SMU_MSG_MAX_COUNT] =
 	MSG_MAP(RmaDueToBadPageThreshold,            PPSMC_MSG_RmaDueToBadPageThreshold,        0),
 	MSG_MAP(SetThrottlingPolicy,                 PPSMC_MSG_SetThrottlingPolicy,             0),
 	MSG_MAP(ResetSDMA,                           PPSMC_MSG_ResetSDMA,                       0),
+	MSG_MAP(ResetVCN,                            PPSMC_MSG_ResetVCN,                        0),
 	MSG_MAP(GetStaticMetricsTable,               PPSMC_MSG_GetStaticMetricsTable,           1),
 	MSG_MAP(GetSystemMetricsTable,               PPSMC_MSG_GetSystemMetricsTable,           1),
 };
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 349b6b8be010..cbe5b06438c1 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -353,6 +353,9 @@ static void smu_v13_0_12_init_caps(struct smu_context *smu)
 		smu_v13_0_6_cap_set(smu, SMU_CAP(PLDM_VERSION));
 	}
 
+	if (fw_ver > 0x04560900)
+		smu_v13_0_6_cap_set(smu, SMU_CAP(VCN_RESET));
+
 	if (fw_ver >= 0x04560700) {
 		if (fw_ver >= 0x04560900) {
 			smu_v13_0_6_cap_set(smu, SMU_CAP(TEMP_METRICS));

From 9f5067531c9b79318c4e48a933cb2694f53f3de2 Mon Sep 17 00:00:00 2001
From: Tomer Tayar <tomer.tayar@intel.com>
Date: Sun, 26 May 2024 16:32:32 +0300
Subject: [PATCH 1423/1455] accel/habanalabs: return ENOMEM if less than
 requested pages were pinned

EFAULT is currently returned if less than requested user pages are
pinned. This value means a "bad address" which might be confusing to
the user, as the address of the given user memory is not necessarily
"bad".

Modify the return value to ENOMEM, as "out of memory" is more suitable
in this case.

Signed-off-by: Tomer Tayar <tomer.tayar@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/memory.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/common/memory.c b/drivers/accel/habanalabs/common/memory.c
index 61472a381904..48d2d598a387 100644
--- a/drivers/accel/habanalabs/common/memory.c
+++ b/drivers/accel/habanalabs/common/memory.c
@@ -2332,7 +2332,7 @@ static int get_user_memory(struct hl_device *hdev, u64 addr, u64 size,
 		if (rc < 0)
 			goto destroy_pages;
 		npages = rc;
-		rc = -EFAULT;
+		rc = -ENOMEM;
 		goto put_pages;
 	}
 	userptr->npages = npages;

From b4fd8e56c9a3b614370fde2d45aec1032eb67ddd Mon Sep 17 00:00:00 2001
From: Vered Yavniely <vered.yavniely@intel.com>
Date: Tue, 18 Jun 2024 19:58:30 +0300
Subject: [PATCH 1424/1455] accel/habanalabs/gaudi2: fix BMON disable
 configuration

Change the BMON_CR register value back to its original state before
enabling, so that BMON does not continue to collect information
after being disabled.

Signed-off-by: Vered Yavniely <vered.yavniely@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/gaudi2/gaudi2_coresight.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2_coresight.c b/drivers/accel/habanalabs/gaudi2/gaudi2_coresight.c
index 2423620ff358..bc3c57bda5cd 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2_coresight.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2_coresight.c
@@ -2426,7 +2426,7 @@ static int gaudi2_config_bmon(struct hl_device *hdev, struct hl_debug_params *pa
 		WREG32(base_reg + mmBMON_ADDRH_E3_OFFSET, 0);
 		WREG32(base_reg + mmBMON_REDUCTION_OFFSET, 0);
 		WREG32(base_reg + mmBMON_STM_TRC_OFFSET, 0x7 | (0xA << 8));
-		WREG32(base_reg + mmBMON_CR_OFFSET, 0x77 | 0xf << 24);
+		WREG32(base_reg + mmBMON_CR_OFFSET, 0x41);
 	}
 
 	return 0;

From 5295be6c4ea4e1ffd38ab0ab131a65afc6b78e9f Mon Sep 17 00:00:00 2001
From: Vitaly Margolin <vitaly.margolin@intel.com>
Date: Sun, 23 Jun 2024 09:19:15 +0300
Subject: [PATCH 1425/1455] accel/habanalabs: add generic message type to get
 error counters

Add a new CPUCP generic message type to retrieve HBM, SRAM and critical
error counters from the device.

Signed-off-by: Vitaly Margolin <vitaly.margolin@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/habanalabs_ioctl.c | 3 +++
 include/linux/habanalabs/cpucp_if.h                | 2 ++
 2 files changed, 5 insertions(+)

diff --git a/drivers/accel/habanalabs/common/habanalabs_ioctl.c b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
index dc80ca921d90..ed0dee5aa9be 100644
--- a/drivers/accel/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
@@ -961,6 +961,9 @@ static int send_fw_generic_request(struct hl_device *hdev, struct hl_info_args *
 	case HL_PASSTHROUGH_VERSIONS:
 		need_input_buff = false;
 		break;
+	case  HL_GET_ERR_COUNTERS_CMD:
+		need_input_buff = true;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/include/linux/habanalabs/cpucp_if.h b/include/linux/habanalabs/cpucp_if.h
index 7ed3fdd55dda..29c50e7427d1 100644
--- a/include/linux/habanalabs/cpucp_if.h
+++ b/include/linux/habanalabs/cpucp_if.h
@@ -1425,9 +1425,11 @@ struct cpucp_monitor_dump {
  * from "pkt_subidx" field in struct cpucp_packet.
  *
  * HL_PASSTHROUGHT_VERSIONS	- Fetch all firmware versions.
+ * HL_GET_ERR_COUNTERS_CMD	- Command to get error counters
  */
 enum hl_passthrough_type {
 	HL_PASSTHROUGH_VERSIONS,
+	HL_GET_ERR_COUNTERS_CMD,
 };
 
 #endif /* CPUCP_IF_H */

From 214e26a43f237358eec59c7c4e5ed1eaf2bfe472 Mon Sep 17 00:00:00 2001
From: Ariel Suller <ariel.suller@intel.com>
Date: Tue, 25 Jun 2024 16:26:59 +0300
Subject: [PATCH 1426/1455] accel/habanalabs/gaudi2: stringify engine/queue ids

Print engine/queue names instead of numerical engine/queue IDs to make
logs and debug output more readable.

Signed-off-by: Ariel Suller <ariel.suller@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c  | 366 +++++++++++++++++++++-
 drivers/accel/habanalabs/gaudi2/gaudi2P.h |   9 +
 2 files changed, 367 insertions(+), 8 deletions(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 5722e4128d3c..39c4cc12ffdf 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -728,6 +728,354 @@ static const int gaudi2_dma_core_async_event_id[] = {
 	[DMA_CORE_ID_KDMA] = GAUDI2_EVENT_KDMA0_CORE,
 };
 
+const char *gaudi2_engine_id_str[] = {
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_EDMA_0),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_EDMA_1),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_MME),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_0),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_1),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_2),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_3),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_4),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_5),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_DEC_0),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_DEC_1),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_EDMA_0),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_EDMA_1),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_MME),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_TPC_0),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_TPC_1),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_TPC_2),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_TPC_3),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_TPC_4),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_TPC_5),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_DEC_0),
+	__stringify(GAUDI2_DCORE1_ENGINE_ID_DEC_1),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_EDMA_0),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_EDMA_1),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_MME),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_TPC_0),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_TPC_1),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_TPC_2),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_TPC_3),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_TPC_4),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_TPC_5),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_DEC_0),
+	__stringify(GAUDI2_DCORE2_ENGINE_ID_DEC_1),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_EDMA_0),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_EDMA_1),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_MME),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_TPC_0),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_TPC_1),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_TPC_2),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_TPC_3),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_TPC_4),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_TPC_5),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_DEC_0),
+	__stringify(GAUDI2_DCORE3_ENGINE_ID_DEC_1),
+	__stringify(GAUDI2_DCORE0_ENGINE_ID_TPC_6),
+	__stringify(GAUDI2_ENGINE_ID_PDMA_0),
+	__stringify(GAUDI2_ENGINE_ID_PDMA_1),
+	__stringify(GAUDI2_ENGINE_ID_ROT_0),
+	__stringify(GAUDI2_ENGINE_ID_ROT_1),
+	__stringify(GAUDI2_PCIE_ENGINE_ID_DEC_0),
+	__stringify(GAUDI2_PCIE_ENGINE_ID_DEC_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC0_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC0_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC1_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC1_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC2_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC2_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC3_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC3_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC4_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC4_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC5_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC5_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC6_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC6_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC7_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC7_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC8_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC8_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC9_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC9_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC10_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC10_1),
+	__stringify(GAUDI2_ENGINE_ID_NIC11_0),
+	__stringify(GAUDI2_ENGINE_ID_NIC11_1),
+	__stringify(GAUDI2_ENGINE_ID_PCIE),
+	__stringify(GAUDI2_ENGINE_ID_PSOC),
+	__stringify(GAUDI2_ENGINE_ID_ARC_FARM),
+	__stringify(GAUDI2_ENGINE_ID_KDMA),
+	__stringify(GAUDI2_ENGINE_ID_SIZE),
+};
+
+const char *gaudi2_queue_id_str[] = {
+	__stringify(GAUDI2_QUEUE_ID_PDMA_0_0),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_0_1),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_0_2),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_0_3),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_1_0),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_1_1),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_1_2),
+	__stringify(GAUDI2_QUEUE_ID_PDMA_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_EDMA_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_MME_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_MME_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_MME_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_MME_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_2_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_2_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_2_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_2_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_3_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_3_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_3_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_3_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_4_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_4_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_4_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_4_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_5_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_5_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_5_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_5_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_6_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_6_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_6_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE0_TPC_6_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_EDMA_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_MME_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_MME_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_MME_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_MME_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_2_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_2_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_2_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_2_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_3_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_3_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_3_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_3_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_4_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_4_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_4_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_4_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_5_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_5_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_5_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE1_TPC_5_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_EDMA_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_MME_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_MME_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_MME_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_MME_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_2_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_2_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_2_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_2_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_3_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_3_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_3_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_3_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_4_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_4_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_4_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_4_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_5_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_5_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_5_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE2_TPC_5_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_EDMA_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_MME_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_MME_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_MME_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_MME_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_0_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_0_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_0_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_0_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_1_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_1_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_1_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_1_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_2_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_2_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_2_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_2_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_3_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_3_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_3_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_3_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_4_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_4_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_4_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_4_3),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_5_0),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_5_1),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_5_2),
+	__stringify(GAUDI2_QUEUE_ID_DCORE3_TPC_5_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_0_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_0_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_0_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_0_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_1_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_1_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_1_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_1_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_2_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_2_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_2_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_2_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_3_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_3_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_3_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_3_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_4_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_4_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_4_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_4_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_5_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_5_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_5_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_5_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_6_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_6_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_6_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_6_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_7_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_7_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_7_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_7_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_8_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_8_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_8_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_8_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_9_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_9_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_9_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_9_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_10_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_10_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_10_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_10_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_11_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_11_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_11_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_11_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_12_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_12_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_12_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_12_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_13_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_13_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_13_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_13_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_14_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_14_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_14_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_14_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_15_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_15_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_15_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_15_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_16_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_16_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_16_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_16_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_17_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_17_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_17_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_17_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_18_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_18_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_18_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_18_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_19_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_19_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_19_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_19_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_20_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_20_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_20_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_20_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_21_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_21_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_21_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_21_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_22_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_22_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_22_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_22_3),
+	__stringify(GAUDI2_QUEUE_ID_NIC_23_0),
+	__stringify(GAUDI2_QUEUE_ID_NIC_23_1),
+	__stringify(GAUDI2_QUEUE_ID_NIC_23_2),
+	__stringify(GAUDI2_QUEUE_ID_NIC_23_3),
+	__stringify(GAUDI2_QUEUE_ID_ROT_0_0),
+	__stringify(GAUDI2_QUEUE_ID_ROT_0_1),
+	__stringify(GAUDI2_QUEUE_ID_ROT_0_2),
+	__stringify(GAUDI2_QUEUE_ID_ROT_0_3),
+	__stringify(GAUDI2_QUEUE_ID_ROT_1_0),
+	__stringify(GAUDI2_QUEUE_ID_ROT_1_1),
+	__stringify(GAUDI2_QUEUE_ID_ROT_1_2),
+	__stringify(GAUDI2_QUEUE_ID_ROT_1_3),
+	__stringify(GAUDI2_QUEUE_ID_CPU_PQ),
+	__stringify(GAUDI2_QUEUE_ID_SIZE),
+};
+
 static const char * const gaudi2_qm_sei_error_cause[GAUDI2_NUM_OF_QM_SEI_ERR_CAUSE] = {
 	"qman sei intr",
 	"arc sei intr"
@@ -6774,7 +7122,8 @@ static int gaudi2_validate_cb_address(struct hl_device *hdev, struct hl_cs_parse
 	struct gaudi2_device *gaudi2 = hdev->asic_specific;
 
 	if (!gaudi2_is_queue_enabled(hdev, parser->hw_queue_id)) {
-		dev_err(hdev->dev, "h/w queue %d is disabled\n", parser->hw_queue_id);
+		dev_err(hdev->dev, "h/w queue %s is disabled\n",
+		GAUDI2_QUEUE_ID_TO_STR(parser->hw_queue_id));
 		return -EINVAL;
 	}
 
@@ -7026,7 +7375,8 @@ static int gaudi2_test_queue_send_msg_short(struct hl_device *hdev, u32 hw_queue
 	rc = hl_hw_queue_send_cb_no_cmpl(hdev, hw_queue_id, pkt_size, msg_info->dma_addr);
 	if (rc)
 		dev_err(hdev->dev,
-			"Failed to send msg_short packet to H/W queue %d\n", hw_queue_id);
+			"Failed to send msg_short packet to H/W queue %s\n",
+			GAUDI2_QUEUE_ID_TO_STR(hw_queue_id));
 
 	return rc;
 }
@@ -7052,8 +7402,8 @@ static int gaudi2_test_queue_wait_completion(struct hl_device *hdev, u32 hw_queu
 			timeout_usec);
 
 	if (rc == -ETIMEDOUT) {
-		dev_err(hdev->dev, "H/W queue %d test failed (SOB_OBJ_0 == 0x%x)\n",
-			hw_queue_id, tmp);
+		dev_err(hdev->dev, "H/W queue %s test failed (SOB_OBJ_0 == 0x%x)\n",
+			GAUDI2_QUEUE_ID_TO_STR(hw_queue_id), tmp);
 		rc = -EIO;
 	}
 
@@ -9603,8 +9953,8 @@ static int hl_arc_event_handle(struct hl_device *hdev, u16 event_type,
 		q = (struct hl_engine_arc_dccm_queue_full_irq *) &payload;
 
 		gaudi2_print_event(hdev, event_type, true,
-				"ARC DCCM Full event: EngId: %u, Intr_type: %u, Qidx: %u",
-				engine_id, intr_type, q->queue_index);
+				"ARC DCCM Full event: Eng: %s, Intr_type: %u, Qidx: %u",
+				GAUDI2_ENG_ID_TO_STR(engine_id), intr_type, q->queue_index);
 		return 1;
 	default:
 		gaudi2_print_event(hdev, event_type, true, "Unknown ARC event type");
@@ -10296,8 +10646,8 @@ static int gaudi2_memset_memory_chunk_using_edma_qm(struct hl_device *hdev,
 
 	rc = hl_hw_queue_send_cb_no_cmpl(hdev, hw_queue_id, pkt_size, phys_addr);
 	if (rc)
-		dev_err(hdev->dev, "Failed to send lin_dma packet to H/W queue %d\n",
-				hw_queue_id);
+		dev_err(hdev->dev, "Failed to send lin_dma packet to H/W queue %s\n",
+				GAUDI2_QUEUE_ID_TO_STR(hw_queue_id));
 
 	return rc;
 }
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2P.h b/drivers/accel/habanalabs/gaudi2/gaudi2P.h
index 05117272cac7..bdf5c1bd2d63 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2P.h
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2P.h
@@ -240,6 +240,15 @@
 #define GAUDI2_NUM_TESTED_QS		(GAUDI2_QUEUE_ID_CPU_PQ - GAUDI2_QUEUE_ID_PDMA_0_0)
 
 
+extern const char *gaudi2_engine_id_str[];
+extern const char *gaudi2_queue_id_str[];
+
+#define GAUDI2_ENG_ID_TO_STR(initiator) ((initiator) >= GAUDI2_ENGINE_ID_SIZE ? "not found" :	\
+						gaudi2_engine_id_str[initiator])
+
+#define GAUDI2_QUEUE_ID_TO_STR(initiator) ((initiator) >= GAUDI2_QUEUE_ID_SIZE ? "not found" :	\
+						gaudi2_queue_id_str[initiator])
+
 enum gaudi2_reserved_sob_id {
 	GAUDI2_RESERVED_SOB_CS_COMPLETION_FIRST,
 	GAUDI2_RESERVED_SOB_CS_COMPLETION_LAST =

From b5cddeb0dc6521dfee669e74967c6d28b1fd0d9e Mon Sep 17 00:00:00 2001
From: Sharley Calzolari <sharley.calzolari@intel.com>
Date: Thu, 18 Jul 2024 18:01:52 +0300
Subject: [PATCH 1427/1455] accel/habanalabs/gaudi2: add support for logging
 register accesses from debugfs

Add infrastructure for logging the last configuration register accesses
that occur via debugfs read/write operations. At interrupt time, these
log entries can be dumped to dmesg, which helps in diagnosing the cause
of RAZWI and ADDR_DEC interrupts.

The logging is implemented as a ring buffer of access entries, with each
entry recording timestamp and access details. To ensure correctness
under concurrent access, operations are now protected using spinlocks.
Entries are copied under lock and then printed after releasing it, which
minimizes time spent in the critical section.

Signed-off-by: Sharley Calzolari <sharley.calzolari@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/debugfs.c    | 111 +++++++++++++++++++
 drivers/accel/habanalabs/common/habanalabs.h |  37 ++++++-
 drivers/accel/habanalabs/gaudi2/gaudi2.c     |   1 +
 3 files changed, 148 insertions(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/common/debugfs.c b/drivers/accel/habanalabs/common/debugfs.c
index 4b391807e5f2..4d975e8059ca 100644
--- a/drivers/accel/habanalabs/common/debugfs.c
+++ b/drivers/accel/habanalabs/common/debugfs.c
@@ -788,6 +788,113 @@ static void hl_access_host_mem(struct hl_device *hdev, u64 addr, u64 *val,
 	}
 }
 
+static void dump_cfg_access_entry(struct hl_device *hdev,
+				  struct hl_debugfs_cfg_access_entry *entry)
+{
+	char *access_type = "";
+	struct tm tm;
+
+	switch (entry->debugfs_type) {
+	case DEBUGFS_READ32:
+		access_type = "READ32 from";
+		break;
+	case DEBUGFS_WRITE32:
+		access_type = "WRITE32 to";
+		break;
+	case DEBUGFS_READ64:
+		access_type = "READ64 from";
+		break;
+	case DEBUGFS_WRITE64:
+		access_type = "WRITE64 to";
+		break;
+	default:
+		dev_err(hdev->dev, "Invalid DEBUGFS access type (%u)\n", entry->debugfs_type);
+		return;
+	}
+
+	time64_to_tm(entry->seconds_since_epoch, 0, &tm);
+	dev_info(hdev->dev,
+		"%ld-%02d-%02d %02d:%02d:%02d (UTC): %s %#llx\n", tm.tm_year + 1900, tm.tm_mon + 1,
+		tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec, access_type, entry->addr);
+}
+
+void hl_debugfs_cfg_access_history_dump(struct hl_device *hdev)
+{
+	struct hl_debugfs_cfg_access *dbgfs = &hdev->debugfs_cfg_accesses;
+	u32 i, head, count = 0;
+	time64_t entry_time, now;
+	unsigned long flags;
+
+	now = ktime_get_real_seconds();
+
+	spin_lock_irqsave(&dbgfs->lock, flags);
+	head = dbgfs->head;
+	if (head == 0)
+		i = HL_DBGFS_CFG_ACCESS_HIST_LEN - 1;
+	else
+		i = head - 1;
+
+	/* Walk back until timeout or invalid entry */
+	while (dbgfs->cfg_access_list[i].valid) {
+		entry_time = dbgfs->cfg_access_list[i].seconds_since_epoch;
+		/* Stop when entry is older than timeout */
+		if (now - entry_time > HL_DBGFS_CFG_ACCESS_HIST_TIMEOUT_SEC)
+			break;
+
+		/* print single entry under lock */
+		{
+			struct hl_debugfs_cfg_access_entry entry = dbgfs->cfg_access_list[i];
+			/*
+			 * We copy the entry out under lock and then print after
+			 * releasing the lock to minimize time under lock.
+			 */
+			spin_unlock_irqrestore(&dbgfs->lock, flags);
+			dump_cfg_access_entry(hdev, &entry);
+			spin_lock_irqsave(&dbgfs->lock, flags);
+		}
+
+		/* mark consumed */
+		dbgfs->cfg_access_list[i].valid = false;
+
+		if (i == 0)
+			i = HL_DBGFS_CFG_ACCESS_HIST_LEN - 1;
+		else
+			i--;
+		count++;
+		if (count >= HL_DBGFS_CFG_ACCESS_HIST_LEN)
+			break;
+	}
+	spin_unlock_irqrestore(&dbgfs->lock, flags);
+}
+
+static void check_if_cfg_access_and_log(struct hl_device *hdev, u64 addr, size_t access_size,
+					enum debugfs_access_type access_type)
+{
+	struct hl_debugfs_cfg_access *dbgfs_cfg_accesses = &hdev->debugfs_cfg_accesses;
+	struct pci_mem_region *mem_reg = &hdev->pci_mem_region[PCI_REGION_CFG];
+	struct hl_debugfs_cfg_access_entry *new_entry;
+	unsigned long flags;
+
+	/* Check if address is in config memory */
+	if (addr >= mem_reg->region_base &&
+		mem_reg->region_size >= access_size &&
+		addr <= mem_reg->region_base + mem_reg->region_size - access_size) {
+
+		spin_lock_irqsave(&dbgfs_cfg_accesses->lock, flags);
+
+		new_entry = &dbgfs_cfg_accesses->cfg_access_list[dbgfs_cfg_accesses->head];
+		new_entry->seconds_since_epoch = ktime_get_real_seconds();
+		new_entry->addr = addr;
+		new_entry->debugfs_type = access_type;
+		new_entry->valid = true;
+		dbgfs_cfg_accesses->head = (dbgfs_cfg_accesses->head + 1)
+						% HL_DBGFS_CFG_ACCESS_HIST_LEN;
+
+		spin_unlock_irqrestore(&dbgfs_cfg_accesses->lock, flags);
+
+	}
+}
+
 static int hl_access_mem(struct hl_device *hdev, u64 addr, u64 *val,
 				enum debugfs_access_type acc_type)
 {
@@ -805,6 +912,7 @@ static int hl_access_mem(struct hl_device *hdev, u64 addr, u64 *val,
 			return rc;
 	}
 
+	check_if_cfg_access_and_log(hdev, addr, acc_size, acc_type);
 	rc = hl_access_dev_mem_by_region(hdev, addr, val, acc_type, &found);
 	if (rc) {
 		dev_err(hdev->dev,
@@ -1762,6 +1870,9 @@ int hl_debugfs_device_init(struct hl_device *hdev)
 	spin_lock_init(&dev_entry->userptr_spinlock);
 	mutex_init(&dev_entry->ctx_mem_hash_mutex);
 
+	spin_lock_init(&hdev->debugfs_cfg_accesses.lock);
+	hdev->debugfs_cfg_accesses.head = 0; /* already zero by alloc but explicit init is fine */
+
 	return 0;
 }
 
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 6f27ce4fa01b..122ade172bb4 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -90,7 +90,9 @@ struct hl_fpriv;
 #define HL_COMMON_USER_CQ_INTERRUPT_ID	0xFFF
 #define HL_COMMON_DEC_INTERRUPT_ID	0xFFE
 
-#define HL_STATE_DUMP_HIST_LEN		5
+#define HL_STATE_DUMP_HIST_LEN			5
+#define HL_DBGFS_CFG_ACCESS_HIST_LEN		20
+#define HL_DBGFS_CFG_ACCESS_HIST_TIMEOUT_SEC	2 /* 2s */
 
 /* Default value for device reset trigger , an invalid value */
 #define HL_RESET_TRIGGER_DEFAULT	0xFF
@@ -2436,6 +2438,32 @@ struct hl_dbg_device_entry {
 	u8				i2c_len;
 };
 
+/**
+ * struct hl_debugfs_cfg_access_entry - single debugfs config access object, member of
+ * hl_debugfs_cfg_access.
+ * @seconds_since_epoch: seconds since January 1, 1970, used for time comparisons.
+ * @debugfs_type: the debugfs operation requested, can be READ32, WRITE32, READ64 or WRITE64.
+ * @addr: the requested address to access.
+ * @valid: if set, this entry has valid data for dumping at interrupt time.
+ */
+struct hl_debugfs_cfg_access_entry {
+	ktime_t				seconds_since_epoch;
+	enum debugfs_access_type	debugfs_type;
+	u64				addr;
+	bool				valid;
+};
+
+/**
+ * struct hl_debugfs_cfg_access - saves debugfs config region access requests history.
+ * @cfg_access_list: list of objects describing config region access requests.
+ * @head: next valid index to add new entry to in cfg_access_list.
+ */
+struct hl_debugfs_cfg_access {
+	struct hl_debugfs_cfg_access_entry	cfg_access_list[HL_DBGFS_CFG_ACCESS_HIST_LEN];
+	u32					head;
+	spinlock_t			lock; /* protects head and entries */
+};
+
 /**
  * struct hl_hw_obj_name_entry - single hw object name, member of
  * hl_state_dump_specs
@@ -3281,6 +3309,7 @@ struct eq_heartbeat_debug_info {
  * @hl_chip_info: ASIC's sensors information.
  * @device_status_description: device status description.
  * @hl_debugfs: device's debugfs manager.
+ * @debugfs_cfg_accesses: list of last debugfs config region accesses.
  * @cb_pool: list of pre allocated CBs.
  * @cb_pool_lock: protects the CB pool.
  * @internal_cb_pool_virt_addr: internal command buffer pool virtual address.
@@ -3461,6 +3490,7 @@ struct hl_device {
 	struct hwmon_chip_info		*hl_chip_info;
 
 	struct hl_dbg_device_entry	hl_debugfs;
+	struct hl_debugfs_cfg_access	debugfs_cfg_accesses;
 
 	struct list_head		cb_pool;
 	spinlock_t			cb_pool_lock;
@@ -4110,6 +4140,7 @@ void hl_debugfs_add_ctx_mem_hash(struct hl_device *hdev, struct hl_ctx *ctx);
 void hl_debugfs_remove_ctx_mem_hash(struct hl_device *hdev, struct hl_ctx *ctx);
 void hl_debugfs_set_state_dump(struct hl_device *hdev, char *data,
 					unsigned long length);
+void hl_debugfs_cfg_access_history_dump(struct hl_device *hdev);
 
 #else
 
@@ -4185,6 +4216,10 @@ static inline void hl_debugfs_set_state_dump(struct hl_device *hdev,
 {
 }
 
+static inline void hl_debugfs_cfg_access_history_dump(struct hl_device *hdev)
+{
+}
+
 #endif
 
 /* Security */
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 39c4cc12ffdf..d8f40f2c967c 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -10610,6 +10610,7 @@ reset_device:
 	if (event_mask & HL_NOTIFIER_EVENT_GENERAL_HW_ERR)
 		hl_handle_critical_hw_err(hdev, event_type, &event_mask);
 
+	hl_debugfs_cfg_access_history_dump(hdev);
 	event_mask |= HL_NOTIFIER_EVENT_DEVICE_RESET;
 	hl_device_cond_reset(hdev, reset_flags, event_mask);
 }

From d0dd796becb08b164fb2d791dca4a76127b3e89a Mon Sep 17 00:00:00 2001
From: Tomer Tayar <tomer.tayar@intel.com>
Date: Thu, 25 Jul 2024 11:39:58 +0300
Subject: [PATCH 1428/1455] accel/habanalabs: clarify ctx use after
 hl_ctx_put() in dmabuf release

In hl_release_dmabuf(), ctx is dereferenced after calling hl_ctx_put()
to obtain the compute device file.

This is safe because the dma-buf object holds a file reference taken in
export_dmabuf(), and the file release (which drops another ctx reference)
can only happen after we drop that file reference via fput(). Thus, this
hl_ctx_put() call cannot be the last one at this point.

Add a comment explaining this to avoid confusion.

Signed-off-by: Tomer Tayar <tomer.tayar@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/memory.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/common/memory.c b/drivers/accel/habanalabs/common/memory.c
index 48d2d598a387..633db4bff46f 100644
--- a/drivers/accel/habanalabs/common/memory.c
+++ b/drivers/accel/habanalabs/common/memory.c
@@ -1837,7 +1837,12 @@ static void hl_release_dmabuf(struct dma_buf *dmabuf)
 	atomic_dec(&ctx->hdev->dmabuf_export_cnt);
 	hl_ctx_put(ctx);
 
-	/* Paired with get_file() in export_dmabuf() */
+	/*
+	 * Paired with get_file() in export_dmabuf().
+	 * 'ctx' can be still used here to get the file pointer, even after hl_ctx_put() was called,
+	 * because releasing the compute device file involves another reference decrement, and it
+	 * would be possible only after calling fput().
+	 */
 	fput(ctx->hpriv->file_priv->filp);
 
 	kfree(hl_dmabuf);

From cade027efa9b358719dcb9305b845d905c3eafce Mon Sep 17 00:00:00 2001
From: Tomer Tayar <tomer.tayar@intel.com>
Date: Wed, 31 Jul 2024 13:56:34 +0300
Subject: [PATCH 1429/1455] accel/habanalabs: fix typo in trace output (cms ->
 cmd)

Fix a typo in TP_printk format string of habanalabs tracepoint:
replace "cms" with "cmd".

Signed-off-by: Tomer Tayar <tomer.tayar@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 include/trace/events/habanalabs.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/trace/events/habanalabs.h b/include/trace/events/habanalabs.h
index 4a2bb2c896d1..fa0d2c6bace4 100644
--- a/include/trace/events/habanalabs.h
+++ b/include/trace/events/habanalabs.h
@@ -145,7 +145,7 @@ DECLARE_EVENT_CLASS(habanalabs_comms_template,
 		__entry->op_str = op_str;
 	),
 
-	TP_printk("%s: cms: %s",
+	TP_printk("%s: cmd: %s",
 		__get_str(dname),
 		__entry->op_str)
 );

From 083c53a85490b4cf5ed63876b09c067358085e20 Mon Sep 17 00:00:00 2001
From: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Date: Thu, 15 Aug 2024 11:24:24 +0300
Subject: [PATCH 1430/1455] accel/habanalabs: disable device access after
 CPLD_SHUTDOWN

After a CPLD shutdown event the device becomes unusable. Prevent further
device access once this event is received.

Signed-off-by: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/device.c     | 25 ++++++++++++++++++++
 drivers/accel/habanalabs/common/habanalabs.h |  3 +++
 2 files changed, 28 insertions(+)

diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index 80fa08bf57bd..c6830c8026ac 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -1630,6 +1630,11 @@ int hl_device_reset(struct hl_device *hdev, u32 flags)
 	from_watchdog_thread = !!(flags & HL_DRV_RESET_FROM_WD_THR);
 	reset_upon_device_release = hdev->reset_upon_device_release && from_dev_release;
 
+	if (hdev->cpld_shutdown) {
+		dev_err(hdev->dev, "Cannot reset device, cpld is shutdown! Device is NOT usable\n");
+		return -EIO;
+	}
+
 	if (!hard_reset && (hl_device_status(hdev) == HL_DEVICE_STATUS_MALFUNCTION)) {
 		dev_dbg(hdev->dev, "soft-reset isn't supported on a malfunctioning device\n");
 		return 0;
@@ -2576,6 +2581,14 @@ void hl_device_fini(struct hl_device *hdev)
 	if (rc)
 		dev_err(hdev->dev, "hw_fini failed in device fini while removing device %d\n", rc);
 
+	/* Reset the H/W (if it accessible). It will be in idle state after this returns */
+	if (!hdev->cpld_shutdown) {
+		rc = hdev->asic_funcs->hw_fini(hdev, true, false);
+		if (rc)
+			dev_err(hdev->dev,
+				"hw_fini failed in device fini while removing device %d\n", rc);
+	}
+
 	hdev->fw_loader.fw_comp_loaded = FW_TYPE_NONE;
 
 	/* Release kernel context */
@@ -2943,3 +2956,15 @@ void hl_handle_clk_change_event(struct hl_device *hdev, u16 event_type, u64 *eve
 
 	mutex_unlock(&clk_throttle->lock);
 }
+
+void hl_eq_cpld_shutdown_event_handle(struct hl_device *hdev, u16 event_id, u64 *event_mask)
+{
+	hl_handle_critical_hw_err(hdev, event_id, event_mask);
+	*event_mask |= HL_NOTIFIER_EVENT_DEVICE_UNAVAILABLE;
+
+	/* Avoid any new accesses to the H/W */
+	hdev->disabled = true;
+	hdev->cpld_shutdown = true;
+	hl_cn_hard_reset_prepare(hdev);
+	hl_cn_stop(hdev);
+}
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 122ade172bb4..33aa385c2693 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -3386,6 +3386,7 @@ struct eq_heartbeat_debug_info {
  *                    addresses.
  * @is_in_dram_scrub: true if dram scrub operation is on going.
  * @disabled: is device disabled.
+ * @cpld_shutdown: is cpld shutdown.
  * @late_init_done: is late init stage was done during initialization.
  * @hwmon_initialized: is H/W monitor sensors was initialized.
  * @reset_on_lockup: true if a reset should be done in case of stuck CS, false
@@ -3562,6 +3563,7 @@ struct hl_device {
 	u16				cpu_pci_msb_addr;
 	u8				is_in_dram_scrub;
 	u8				disabled;
+	u8				cpld_shutdown;
 	u8				late_init_done;
 	u8				hwmon_initialized;
 	u8				reset_on_lockup;
@@ -4119,6 +4121,7 @@ void hl_init_cpu_for_irq(struct hl_device *hdev);
 void hl_set_irq_affinity(struct hl_device *hdev, int irq);
 void hl_eq_heartbeat_event_handle(struct hl_device *hdev);
 void hl_handle_clk_change_event(struct hl_device *hdev, u16 event_type, u64 *event_mask);
+void hl_eq_cpld_shutdown_event_handle(struct hl_device *hdev, u16 event_id, u64 *event_mask);
 
 #ifdef CONFIG_DEBUG_FS
 

From 0529b191ac9c907225593e12a68f1718a60b2fba Mon Sep 17 00:00:00 2001
From: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Date: Thu, 15 Aug 2024 15:24:24 +0300
Subject: [PATCH 1431/1455] accel/habanalabs/gaudi2: use the CPLD_SHUTDOWN
 event handler

After CPLD shutdown event the device is not usable anymore. The common
CPLD_SHUTDOWN event handler disables any subsequent device access.

Signed-off-by: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/device.c | 2 --
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 4 ++--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/accel/habanalabs/common/device.c b/drivers/accel/habanalabs/common/device.c
index c6830c8026ac..999c92d7036e 100644
--- a/drivers/accel/habanalabs/common/device.c
+++ b/drivers/accel/habanalabs/common/device.c
@@ -2965,6 +2965,4 @@ void hl_eq_cpld_shutdown_event_handle(struct hl_device *hdev, u16 event_id, u64
 	/* Avoid any new accesses to the H/W */
 	hdev->disabled = true;
 	hdev->cpld_shutdown = true;
-	hl_cn_hard_reset_prepare(hdev);
-	hl_cn_stop(hdev);
 }
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index d8f40f2c967c..72bfc5e639e4 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -5184,7 +5184,7 @@ static void gaudi2_halt_engines(struct hl_device *hdev, bool hard_reset, bool fw
 	else
 		wait_timeout_ms = GAUDI2_RESET_WAIT_MSEC;
 
-	if (fw_reset)
+	if (fw_reset || hdev->cpld_shutdown)
 		goto skip_engines;
 
 	gaudi2_stop_dma_qmans(hdev);
@@ -10522,7 +10522,7 @@ static void gaudi2_handle_eqe(struct hl_device *hdev, struct hl_eq_entry *eq_ent
 		dev_err(hdev->dev, "CPLD shutdown event, reset reason: 0x%llx\n",
 						le64_to_cpu(eq_entry->data[0]));
 		error_count = GAUDI2_NA_EVENT_CAUSE;
-		event_mask |= HL_NOTIFIER_EVENT_GENERAL_HW_ERR;
+		hl_eq_cpld_shutdown_event_handle(hdev, event_type, &event_mask);
 		break;
 
 	case GAUDI2_EVENT_CPU_PKT_SANITY_FAILED:

From 0668db41b5d8a834a04be3b281fa9452dcf3dc18 Mon Sep 17 00:00:00 2001
From: Ilia Levi <ilia.levi@intel.com>
Date: Mon, 19 Aug 2024 12:13:07 +0300
Subject: [PATCH 1432/1455] accel/habanalabs: remove old interface variation of
 'access_ok()'

The access_ok() API no longer requires the VERIFY_WRITE argument,
and the use of the old interface with VERIFY_WRITE is deprecated.

Clean up the habanalabs memory manager to use the modern access_ok()
interface consistently. This removes old #ifdef guards and aligns the
driver with current upstream kernel APIs.

Signed-off-by: Ilia Levi <ilia.levi@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/memory_mgr.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/accel/habanalabs/common/memory_mgr.c b/drivers/accel/habanalabs/common/memory_mgr.c
index 99cd83139d46..4401beb99e42 100644
--- a/drivers/accel/habanalabs/common/memory_mgr.c
+++ b/drivers/accel/habanalabs/common/memory_mgr.c
@@ -259,13 +259,8 @@ int hl_mem_mgr_mmap(struct hl_mem_mgr *mmg, struct vm_area_struct *vma,
 		goto put_mem;
 	}
 
-#ifdef _HAS_TYPE_ARG_IN_ACCESS_OK
-	if (!access_ok(VERIFY_WRITE, (void __user *)(uintptr_t)vma->vm_start,
-		       user_mem_size)) {
-#else
 	if (!access_ok((void __user *)(uintptr_t)vma->vm_start,
 		       user_mem_size)) {
-#endif
 		dev_err(mmg->dev, "%s: User pointer is invalid - 0x%lx\n",
 			buf->behavior->topic, vma->vm_start);
 

From 513024d5a0e34fd34247043f1876b6138ca52847 Mon Sep 17 00:00:00 2001
From: Moti Haimovski <moti.haimovski@intel.com>
Date: Sun, 8 Sep 2024 15:01:26 +0300
Subject: [PATCH 1433/1455] accel/habanalabs: support mapping cb with
 vmalloc-backed coherent memory

When IOMMU is enabled, dma_alloc_coherent() with GFP_USER may return
addresses from the vmalloc range. If such an address is mapped without
VM_MIXEDMAP, vm_insert_page() will trigger a BUG_ON due to the
VM_PFNMAP restriction.

Fix this by checking for vmalloc addresses and setting VM_MIXEDMAP
in the VMA before mapping. This ensures safe mapping and avoids kernel
crashes. The memory is still driver-allocated and cannot be accessed
directly by userspace.

Signed-off-by: Moti Haimovski  <moti.haimovski@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/gaudi/gaudi.c   | 19 +++++++++++++++++++
 drivers/accel/habanalabs/gaudi2/gaudi2.c |  7 +++++++
 2 files changed, 26 insertions(+)

diff --git a/drivers/accel/habanalabs/gaudi/gaudi.c b/drivers/accel/habanalabs/gaudi/gaudi.c
index fa893a9b826e..34771d75da9d 100644
--- a/drivers/accel/habanalabs/gaudi/gaudi.c
+++ b/drivers/accel/habanalabs/gaudi/gaudi.c
@@ -4168,10 +4168,29 @@ static int gaudi_mmap(struct hl_device *hdev, struct vm_area_struct *vma,
 	vm_flags_set(vma, VM_IO | VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP |
 			VM_DONTCOPY | VM_NORESERVE);
 
+#ifdef _HAS_DMA_MMAP_COHERENT
+	/*
+	 * If dma_alloc_coherent() returns a vmalloc address, set VM_MIXEDMAP
+	 * so vm_insert_page() can handle it safely. Without this, the kernel
+	 * may BUG_ON due to VM_PFNMAP.
+	 */
+	if (is_vmalloc_addr(cpu_addr))
+		vm_flags_set(vma, VM_MIXEDMAP);
+
 	rc = dma_mmap_coherent(hdev->dev, vma, cpu_addr,
 				(dma_addr - HOST_PHYS_BASE), size);
 	if (rc)
 		dev_err(hdev->dev, "dma_mmap_coherent error %d", rc);
+#else
+
+	rc = remap_pfn_range(vma, vma->vm_start,
+				virt_to_phys(cpu_addr) >> PAGE_SHIFT,
+				size, vma->vm_page_prot);
+	if (rc)
+		dev_err(hdev->dev, "remap_pfn_range error %d", rc);
+
+ #endif
+
 
 	return rc;
 }
diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 72bfc5e639e4..03dde63ed875 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -6832,6 +6832,13 @@ static int gaudi2_mmap(struct hl_device *hdev, struct vm_area_struct *vma,
 			VM_DONTCOPY | VM_NORESERVE);
 
 #ifdef _HAS_DMA_MMAP_COHERENT
+	/*
+	 * If dma_alloc_coherent() returns a vmalloc address, set VM_MIXEDMAP
+	 * so vm_insert_page() can handle it safely. Without this, the kernel
+	 * may BUG_ON due to VM_PFNMAP.
+	 */
+	if (is_vmalloc_addr(cpu_addr))
+		vm_flags_set(vma, VM_MIXEDMAP);
 
 	rc = dma_mmap_coherent(hdev->dev, vma, cpu_addr, dma_addr, size);
 	if (rc)

From 8cbacc9a2703c661c5e15bfbaeb1d4dcb58eb621 Mon Sep 17 00:00:00 2001
From: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Date: Mon, 9 Sep 2024 18:21:22 +0300
Subject: [PATCH 1434/1455] accel/habanalabs: add NVMe Direct I/O (HLDIO)
 infrastructure
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Introduce NVMe Direct I/O (HLDIO) infrastructure to support
peer‑to‑peer DMA in the habanalabs driver. This adds internal helpers
and data structures to enable direct transfers between NVMe storage
and device memory.

The feature is built only when CONFIG_HL_HLDIO is enabled. A debugfs
interface is also provided for functional validation.

Signed-off-by: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Reviewed-by: Farah Kassabri <farah.kassabri@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/Kconfig             |  23 +
 drivers/accel/habanalabs/common/Makefile     |   5 +
 drivers/accel/habanalabs/common/debugfs.c    |   3 +
 drivers/accel/habanalabs/common/habanalabs.h |  12 +-
 drivers/accel/habanalabs/common/hldio.c      | 437 +++++++++++++++++++
 drivers/accel/habanalabs/common/hldio.h      | 146 +++++++
 6 files changed, 624 insertions(+), 2 deletions(-)
 create mode 100644 drivers/accel/habanalabs/common/hldio.c
 create mode 100644 drivers/accel/habanalabs/common/hldio.h

diff --git a/drivers/accel/habanalabs/Kconfig b/drivers/accel/habanalabs/Kconfig
index 1919fbb169c7..6d1506acbd72 100644
--- a/drivers/accel/habanalabs/Kconfig
+++ b/drivers/accel/habanalabs/Kconfig
@@ -27,3 +27,26 @@ config DRM_ACCEL_HABANALABS
 
 	  To compile this driver as a module, choose M here: the
 	  module will be called habanalabs.
+
+if DRM_ACCEL_HABANALABS
+
+config HL_HLDIO
+	bool "Habanalabs NVMe Direct I/O (HLDIO)"
+	depends on PCI_P2PDMA
+	depends on BLOCK
+	help
+	  Enable NVMe peer-to-peer direct I/O support for Habanalabs AI
+	  accelerators.
+
+	  This allows direct data transfers between NVMe storage devices
+	  and Habanalabs accelerators without involving system memory,
+	  using PCI peer-to-peer DMA capabilities.
+
+	  Requirements:
+	  - CONFIG_PCI_P2PDMA=y
+	  - NVMe device and Habanalabs accelerator under same PCI root complex
+	  - IOMMU disabled or in passthrough mode
+	  - Hardware supporting PCI P2P DMA
+
+	  If unsure, say N
+endif # DRM_ACCEL_HABANALABS
diff --git a/drivers/accel/habanalabs/common/Makefile b/drivers/accel/habanalabs/common/Makefile
index e6abffea9f87..b6d00de09db5 100644
--- a/drivers/accel/habanalabs/common/Makefile
+++ b/drivers/accel/habanalabs/common/Makefile
@@ -13,3 +13,8 @@ HL_COMMON_FILES := common/habanalabs_drv.o common/device.o common/context.o \
 		common/command_submission.o common/firmware_if.o \
 		common/security.o common/state_dump.o \
 		common/memory_mgr.o common/decoder.o
+
+# Conditionally add HLDIO support
+ifdef CONFIG_HL_HLDIO
+HL_COMMON_FILES += common/hldio.o
+endif
\ No newline at end of file
diff --git a/drivers/accel/habanalabs/common/debugfs.c b/drivers/accel/habanalabs/common/debugfs.c
index 4d975e8059ca..f3600df064d1 100644
--- a/drivers/accel/habanalabs/common/debugfs.c
+++ b/drivers/accel/habanalabs/common/debugfs.c
@@ -1891,6 +1891,7 @@ void hl_debugfs_device_fini(struct hl_device *hdev)
 		vfree(entry->state_dump[i]);
 
 	kfree(entry->entry_arr);
+
 }
 
 void hl_debugfs_add_device(struct hl_device *hdev)
@@ -1903,6 +1904,7 @@ void hl_debugfs_add_device(struct hl_device *hdev)
 
 	if (!hdev->asic_prop.fw_security_enabled)
 		add_secured_nodes(dev_entry, dev_entry->root);
+
 }
 
 void hl_debugfs_add_file(struct hl_fpriv *hpriv)
@@ -2035,3 +2037,4 @@ void hl_debugfs_set_state_dump(struct hl_device *hdev, char *data,
 
 	up_write(&dev_entry->state_dump_sem);
 }
+
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 33aa385c2693..6be185d7be58 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -704,6 +704,7 @@ struct hl_hints_range {
  * @supports_advanced_cpucp_rc: true if new cpucp opcodes are supported.
  * @supports_engine_modes: true if changing engines/engine_cores modes is supported.
  * @support_dynamic_resereved_fw_size: true if we support dynamic reserved size for fw.
+ * @supports_nvme: indicates whether the asic supports NVMe P2P DMA.
  */
 struct asic_fixed_properties {
 	struct hw_queue_properties	*hw_queues_props;
@@ -824,6 +825,7 @@ struct asic_fixed_properties {
 	u8				supports_advanced_cpucp_rc;
 	u8				supports_engine_modes;
 	u8				support_dynamic_resereved_fw_size;
+	u8				supports_nvme;
 };
 
 /**
@@ -2276,6 +2278,9 @@ struct hl_vm {
 	u8			init_done;
 };
 
+#ifdef CONFIG_HL_HLDIO
+#include "hldio.h"
+#endif
 
 /*
  * DEBUG, PROFILING STRUCTURE
@@ -2346,7 +2351,6 @@ struct hl_fpriv {
 	struct mutex			ctx_lock;
 };
 
-
 /*
  * DebugFS
  */
@@ -2374,6 +2378,7 @@ struct hl_debugfs_entry {
 	struct hl_dbg_device_entry	*dev_entry;
 };
 
+
 /**
  * struct hl_dbg_device_entry - ASIC specific debugfs manager.
  * @root: root dentry.
@@ -3334,6 +3339,7 @@ struct eq_heartbeat_debug_info {
  * @captured_err_info: holds information about errors.
  * @reset_info: holds current device reset information.
  * @heartbeat_debug_info: counters used to debug heartbeat failures.
+ * @hldio: describes habanalabs direct storage interaction interface.
  * @irq_affinity_mask: mask of available CPU cores for user and decoder interrupt handling.
  * @stream_master_qid_arr: pointer to array with QIDs of master streams.
  * @fw_inner_major_ver: the major of current loaded preboot inner version.
@@ -3527,7 +3533,9 @@ struct hl_device {
 	struct hl_reset_info		reset_info;
 
 	struct eq_heartbeat_debug_info	heartbeat_debug_info;
-
+#ifdef CONFIG_HL_HLDIO
+	struct hl_dio			hldio;
+#endif
 	cpumask_t			irq_affinity_mask;
 
 	u32				*stream_master_qid_arr;
diff --git a/drivers/accel/habanalabs/common/hldio.c b/drivers/accel/habanalabs/common/hldio.c
new file mode 100644
index 000000000000..083ae5610875
--- /dev/null
+++ b/drivers/accel/habanalabs/common/hldio.c
@@ -0,0 +1,437 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Copyright 2024 HabanaLabs, Ltd.
+ * All Rights Reserved.
+ */
+
+#include "habanalabs.h"
+#include "hldio.h"
+#include <generated/uapi/linux/version.h>
+#include <linux/pci-p2pdma.h>
+#include <linux/blkdev.h>
+#include <linux/vmalloc.h>
+
+/*
+ * NVMe Direct I/O implementation for habanalabs driver
+ *
+ * ASSUMPTIONS
+ * ===========
+ * 1. No IOMMU (well, technically it can work with IOMMU, but it is *almost useless).
+ * 2. Only READ operations (can extend in the future).
+ * 3. No sparse files (can overcome this in the future).
+ * 4. Kernel version >= 6.9
+ * 5. Requiring page alignment is OK (I don't see a solution to this one right,
+ *    now, how do we read partial pages?)
+ * 6. Kernel compiled with CONFIG_PCI_P2PDMA. This requires a CUSTOM kernel.
+ *    Theoretically I have a slight idea on how this could be solvable, but it
+ *    is probably inacceptable for the upstream. Also may not work in the end.
+ * 7. Either make sure our cards and disks are under the same PCI bridge, or
+ *    compile a custom kernel to hack around this.
+ */
+
+#define IO_STABILIZE_TIMEOUT 10000000 /* 10 seconds in microseconds */
+
+/*
+ * This struct contains all the useful data I could milk out of the file handle
+ * provided by the user.
+ * @TODO: right now it is retrieved on each IO, but can be done once with some
+ * dedicated IOCTL, call it for example HL_REGISTER_HANDLE.
+ */
+struct hl_dio_fd {
+	/* Back pointer in case we need it in async completion */
+	struct hl_ctx *ctx;
+	/* Associated fd struct */
+	struct file *filp;
+};
+
+/*
+ * This is a single IO descriptor
+ */
+struct hl_direct_io {
+	struct hl_dio_fd f;
+	struct kiocb kio;
+	struct bio_vec *bv;
+	struct iov_iter iter;
+	u64 device_va;
+	u64 off_bytes;
+	u64 len_bytes;
+	u32 type;
+};
+
+bool hl_device_supports_nvme(struct hl_device *hdev)
+{
+	return hdev->asic_prop.supports_nvme;
+}
+
+static int hl_dio_fd_register(struct hl_ctx *ctx, int fd, struct hl_dio_fd *f)
+{
+	struct hl_device *hdev = ctx->hdev;
+	struct block_device *bd;
+	struct super_block *sb;
+	struct inode *inode;
+	struct gendisk *gd;
+	struct device *disk_dev;
+	int rc;
+
+	f->filp = fget(fd);
+	if (!f->filp) {
+		rc = -ENOENT;
+		goto out;
+	}
+
+	if (!(f->filp->f_flags & O_DIRECT)) {
+		dev_err(hdev->dev, "file is not in the direct mode\n");
+		rc = -EINVAL;
+		goto fput;
+	}
+
+	if (!f->filp->f_op->read_iter) {
+		dev_err(hdev->dev, "read iter is not supported, need to fall back to legacy\n");
+		rc = -EINVAL;
+		goto fput;
+	}
+
+	inode = file_inode(f->filp);
+	sb = inode->i_sb;
+	bd = sb->s_bdev;
+	gd = bd->bd_disk;
+
+	if (inode->i_blocks << sb->s_blocksize_bits < i_size_read(inode)) {
+		dev_err(hdev->dev, "sparse files are not currently supported\n");
+		rc = -EINVAL;
+		goto fput;
+	}
+
+	if (!bd || !gd) {
+		dev_err(hdev->dev, "invalid block device\n");
+		rc = -ENODEV;
+		goto fput;
+	}
+	/* Get the underlying device from the block device */
+	disk_dev = disk_to_dev(gd);
+	if (!dma_pci_p2pdma_supported(disk_dev)) {
+		dev_err(hdev->dev, "device does not support PCI P2P DMA\n");
+		rc = -EOPNOTSUPP;
+		goto fput;
+	}
+
+	/*
+	 * @TODO: Maybe we need additional checks here
+	 */
+
+	f->ctx = ctx;
+	rc = 0;
+
+	goto out;
+fput:
+	fput(f->filp);
+out:
+	return rc;
+}
+
+static void hl_dio_fd_unregister(struct hl_dio_fd *f)
+{
+	fput(f->filp);
+}
+
+static long hl_dio_count_io(struct hl_device *hdev)
+{
+	s64 sum = 0;
+	int i;
+
+	for_each_possible_cpu(i)
+		sum += per_cpu(*hdev->hldio.inflight_ios, i);
+
+	return sum;
+}
+
+static bool hl_dio_get_iopath(struct hl_ctx *ctx)
+{
+	struct hl_device *hdev = ctx->hdev;
+
+	if (hdev->hldio.io_enabled) {
+		this_cpu_inc(*hdev->hldio.inflight_ios);
+
+		/* Avoid race conditions */
+		if (!hdev->hldio.io_enabled) {
+			this_cpu_dec(*hdev->hldio.inflight_ios);
+			return false;
+		}
+
+		hl_ctx_get(ctx);
+
+		return true;
+	}
+
+	return false;
+}
+
+static void hl_dio_put_iopath(struct hl_ctx *ctx)
+{
+	struct hl_device *hdev = ctx->hdev;
+
+	hl_ctx_put(ctx);
+	this_cpu_dec(*hdev->hldio.inflight_ios);
+}
+
+static void hl_dio_set_io_enabled(struct hl_device *hdev, bool enabled)
+{
+	hdev->hldio.io_enabled = enabled;
+}
+
+static bool hl_dio_validate_io(struct hl_device *hdev, struct hl_direct_io *io)
+{
+	if ((u64)io->device_va & ~PAGE_MASK) {
+		dev_dbg(hdev->dev, "device address must be 4K aligned\n");
+		return false;
+	}
+
+	if (io->len_bytes & ~PAGE_MASK) {
+		dev_dbg(hdev->dev, "IO length must be 4K aligned\n");
+		return false;
+	}
+
+	if (io->off_bytes & ~PAGE_MASK) {
+		dev_dbg(hdev->dev, "IO offset must be 4K aligned\n");
+		return false;
+	}
+
+	return true;
+}
+
+static struct page *hl_dio_va2page(struct hl_device *hdev, struct hl_ctx *ctx, u64 device_va)
+{
+	struct hl_dio *hldio = &hdev->hldio;
+	u64 device_pa;
+	int rc, i;
+
+	rc = hl_mmu_va_to_pa(ctx, device_va, &device_pa);
+	if (rc) {
+		dev_err(hdev->dev, "device virtual address translation error: %#llx (%d)",
+				device_va, rc);
+		return NULL;
+	}
+
+	for (i = 0 ; i < hldio->np2prs ; ++i) {
+		if (device_pa >= hldio->p2prs[i].device_pa &&
+		    device_pa < hldio->p2prs[i].device_pa + hldio->p2prs[i].size)
+			return hldio->p2prs[i].p2ppages[(device_pa - hldio->p2prs[i].device_pa) >>
+				PAGE_SHIFT];
+	}
+
+	return NULL;
+}
+
+static ssize_t hl_direct_io(struct hl_device *hdev, struct hl_direct_io *io)
+{
+	u64 npages, device_va;
+	ssize_t rc;
+	int i;
+
+	if (!hl_dio_validate_io(hdev, io))
+		return -EINVAL;
+
+	if (!hl_dio_get_iopath(io->f.ctx)) {
+		dev_info(hdev->dev, "can't schedule a new IO, IO is disabled\n");
+		return -ESHUTDOWN;
+	}
+
+	init_sync_kiocb(&io->kio, io->f.filp);
+	io->kio.ki_pos = io->off_bytes;
+
+	npages = (io->len_bytes >> PAGE_SHIFT);
+
+	/* @TODO: this can be implemented smarter, vmalloc in iopath is not
+	 * ideal. Maybe some variation of genpool. Number of pages may differ
+	 * greatly, so maybe even use pools of different sizes and chose the
+	 * closest one.
+	 */
+	io->bv = vzalloc(npages * sizeof(struct bio_vec));
+	if (!io->bv)
+		return -ENOMEM;
+
+	for (i = 0, device_va = io->device_va; i < npages ; ++i, device_va += PAGE_SIZE) {
+		io->bv[i].bv_page = hl_dio_va2page(hdev, io->f.ctx, device_va);
+		if (!io->bv[i].bv_page) {
+			dev_err(hdev->dev, "error getting page struct for device va %#llx",
+					device_va);
+			rc = -EFAULT;
+			goto cleanup;
+		}
+		io->bv[i].bv_offset = 0;
+		io->bv[i].bv_len = PAGE_SIZE;
+	}
+
+	iov_iter_bvec(&io->iter, io->type, io->bv, 1, io->len_bytes);
+	if (io->f.filp->f_op && io->f.filp->f_op->read_iter)
+		rc = io->f.filp->f_op->read_iter(&io->kio, &io->iter);
+	else
+		rc = -EINVAL;
+
+cleanup:
+	vfree(io->bv);
+	hl_dio_put_iopath(io->f.ctx);
+
+	dev_dbg(hdev->dev, "IO ended with %ld\n", rc);
+
+	return rc;
+}
+
+/*
+ * @TODO: This function can be used as a callback for io completion under
+ * kio->ki_complete in order to implement async IO.
+ * Note that on more recent kernels there is no ret2.
+ */
+__maybe_unused static void hl_direct_io_complete(struct kiocb *kio, long ret, long ret2)
+{
+	struct hl_direct_io *io = container_of(kio, struct hl_direct_io, kio);
+
+	dev_dbg(io->f.ctx->hdev->dev, "IO completed with %ld\n", ret);
+
+	/* Do something to copy result to user / notify completion */
+
+	hl_dio_put_iopath(io->f.ctx);
+
+	hl_dio_fd_unregister(&io->f);
+}
+
+/*
+ * DMA disk to ASIC, wait for results. Must be invoked from the user context
+ */
+int hl_dio_ssd2hl(struct hl_device *hdev, struct hl_ctx *ctx, int fd,
+		  u64 device_va, off_t off_bytes, size_t len_bytes,
+		  size_t *len_read)
+{
+	struct hl_direct_io *io;
+	ssize_t rc;
+
+	dev_dbg(hdev->dev, "SSD2HL fd=%d va=%#llx len=%#lx\n", fd, device_va, len_bytes);
+
+	io = kzalloc(sizeof(*io), GFP_KERNEL);
+	if (!io) {
+		rc = -ENOMEM;
+		goto out;
+	}
+
+	*io = (struct hl_direct_io){
+		.device_va = device_va,
+		.len_bytes = len_bytes,
+		.off_bytes = off_bytes,
+		.type = READ,
+	};
+
+	rc = hl_dio_fd_register(ctx, fd, &io->f);
+	if (rc)
+		goto kfree_io;
+
+	rc = hl_direct_io(hdev, io);
+	if (rc >= 0) {
+		*len_read = rc;
+		rc = 0;
+	}
+
+	/* This shall be called only in the case of a sync IO */
+	hl_dio_fd_unregister(&io->f);
+kfree_io:
+	kfree(io);
+out:
+	return rc;
+}
+
+static void hl_p2p_region_fini(struct hl_device *hdev, struct hl_p2p_region *p2pr)
+{
+	if (p2pr->p2ppages) {
+		vfree(p2pr->p2ppages);
+		p2pr->p2ppages = NULL;
+	}
+
+	if (p2pr->p2pmem) {
+		dev_dbg(hdev->dev, "freeing P2P mem from %p, size=%#llx\n",
+				p2pr->p2pmem, p2pr->size);
+		pci_free_p2pmem(hdev->pdev, p2pr->p2pmem, p2pr->size);
+		p2pr->p2pmem = NULL;
+	}
+}
+
+void hl_p2p_region_fini_all(struct hl_device *hdev)
+{
+	int i;
+
+	for (i = 0 ; i < hdev->hldio.np2prs ; ++i)
+		hl_p2p_region_fini(hdev, &hdev->hldio.p2prs[i]);
+
+	kvfree(hdev->hldio.p2prs);
+	hdev->hldio.p2prs = NULL;
+	hdev->hldio.np2prs = 0;
+}
+
+int hl_p2p_region_init(struct hl_device *hdev, struct hl_p2p_region *p2pr)
+{
+	void *addr;
+	int rc, i;
+
+	/* Start by publishing our p2p memory */
+	rc = pci_p2pdma_add_resource(hdev->pdev, p2pr->bar, p2pr->size, p2pr->bar_offset);
+	if (rc) {
+		dev_err(hdev->dev, "error adding p2p resource: %d\n", rc);
+		goto err;
+	}
+
+	/* Alloc all p2p mem */
+	p2pr->p2pmem = pci_alloc_p2pmem(hdev->pdev, p2pr->size);
+	if (!p2pr->p2pmem) {
+		dev_err(hdev->dev, "error allocating p2p memory\n");
+		rc = -ENOMEM;
+		goto err;
+	}
+
+	p2pr->p2ppages = vmalloc((p2pr->size >> PAGE_SHIFT) * sizeof(struct page *));
+	if (!p2pr->p2ppages) {
+		rc = -ENOMEM;
+		goto err;
+	}
+
+	for (i = 0, addr = p2pr->p2pmem ; i < (p2pr->size >> PAGE_SHIFT) ; ++i, addr += PAGE_SIZE) {
+		p2pr->p2ppages[i] = virt_to_page(addr);
+		if (!p2pr->p2ppages[i]) {
+			rc = -EFAULT;
+			goto err;
+		}
+	}
+
+	return 0;
+err:
+	hl_p2p_region_fini(hdev, p2pr);
+	return rc;
+}
+
+int hl_dio_start(struct hl_device *hdev)
+{
+	dev_dbg(hdev->dev, "initializing HLDIO\n");
+
+	/* Initialize the IO counter and enable IO */
+	hdev->hldio.inflight_ios = alloc_percpu(s64);
+	if (!hdev->hldio.inflight_ios)
+		return -ENOMEM;
+
+	hl_dio_set_io_enabled(hdev, true);
+
+	return 0;
+}
+
+void hl_dio_stop(struct hl_device *hdev)
+{
+	dev_dbg(hdev->dev, "deinitializing HLDIO\n");
+
+	if (hdev->hldio.io_enabled) {
+		/* Wait for all the IO to finish */
+		hl_dio_set_io_enabled(hdev, false);
+		hl_poll_timeout_condition(hdev, !hl_dio_count_io(hdev), 1000, IO_STABILIZE_TIMEOUT);
+	}
+
+	if (hdev->hldio.inflight_ios) {
+		free_percpu(hdev->hldio.inflight_ios);
+		hdev->hldio.inflight_ios = NULL;
+	}
+}
diff --git a/drivers/accel/habanalabs/common/hldio.h b/drivers/accel/habanalabs/common/hldio.h
new file mode 100644
index 000000000000..2874388f2851
--- /dev/null
+++ b/drivers/accel/habanalabs/common/hldio.h
@@ -0,0 +1,146 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * hldio.h - NVMe Direct I/O (HLDIO) infrastructure for Habana Labs Driver
+ *
+ * This feature requires specific hardware setup and must not be built
+ * under COMPILE_TEST.
+ */
+
+#ifndef __HL_HLDIO_H__
+#define __HL_HLDIO_H__
+
+#include <linux/types.h>
+#include <linux/fs.h>
+#include <linux/seq_file.h>
+#include <linux/ktime.h>     /* ktime functions */
+#include <linux/delay.h>     /* usleep_range */
+#include <linux/kernel.h>    /* might_sleep_if */
+#include <linux/errno.h>     /* error codes */
+
+/* Forward declarations */
+struct hl_device;
+struct file;
+
+/* Enable only if Kconfig selected */
+#ifdef CONFIG_HL_HLDIO
+/**
+ * struct hl_p2p_region - describes a single P2P memory region
+ * @p2ppages: array of page structs for the P2P memory
+ * @p2pmem: virtual address of the P2P memory region
+ * @device_pa: physical address on the device
+ * @bar_offset: offset within the BAR
+ * @size: size of the region in bytes
+ * @bar: BAR number containing this region
+ */
+struct hl_p2p_region {
+	struct page **p2ppages;
+	void *p2pmem;
+	u64 device_pa;
+	u64 bar_offset;
+	u64 size;
+	int bar;
+};
+
+/**
+ * struct hl_dio_stats - Direct I/O statistics
+ * @total_ops: total number of operations attempted
+ * @successful_ops: number of successful operations
+ * @failed_ops: number of failed operations
+ * @bytes_transferred: total bytes successfully transferred
+ * @last_len_read: length of the last read operation
+ */
+struct hl_dio_stats {
+	u64 total_ops;
+	u64 successful_ops;
+	u64 failed_ops;
+	u64 bytes_transferred;
+	size_t last_len_read;
+};
+
+/**
+ * struct hl_dio - describes habanalabs direct storage interaction interface
+ * @p2prs: array of p2p regions
+ * @inflight_ios: percpu counter for inflight ios
+ * @np2prs: number of elements in p2prs
+ * @io_enabled: 1 if io is enabled 0 otherwise
+ */
+struct hl_dio {
+	struct hl_p2p_region *p2prs;
+	s64 __percpu *inflight_ios;
+	u8 np2prs;
+	u8 io_enabled;
+};
+
+int hl_dio_ssd2hl(struct hl_device *hdev, struct hl_ctx *ctx, int fd,
+		  u64 device_va, off_t off_bytes, size_t len_bytes,
+		  size_t *len_read);
+void hl_p2p_region_fini_all(struct hl_device *hdev);
+int hl_p2p_region_init(struct hl_device *hdev, struct hl_p2p_region *p2pr);
+int hl_dio_start(struct hl_device *hdev);
+void hl_dio_stop(struct hl_device *hdev);
+
+/* Init/teardown */
+int hl_hldio_init(struct hl_device *hdev);
+void hl_hldio_fini(struct hl_device *hdev);
+
+/* File operations */
+long hl_hldio_ioctl(struct file *filep, unsigned int cmd, unsigned long arg);
+
+/* DebugFS hooks */
+#ifdef CONFIG_DEBUG_FS
+void hl_hldio_debugfs_init(struct hl_device *hdev);
+void hl_hldio_debugfs_fini(struct hl_device *hdev);
+#else
+static inline void hl_hldio_debugfs_init(struct hl_device *hdev) { }
+static inline void hl_hldio_debugfs_fini(struct hl_device *hdev) { }
+#endif
+
+#else /* !CONFIG_HL_HLDIO */
+
+struct hl_p2p_region;
+/* Stubs when HLDIO is disabled */
+static inline int hl_dio_ssd2hl(struct hl_device *hdev, struct hl_ctx *ctx, int fd,
+		  u64 device_va, off_t off_bytes, size_t len_bytes,
+		  size_t *len_read)
+{ return -EOPNOTSUPP; }
+static inline void hl_p2p_region_fini_all(struct hl_device *hdev) {}
+static inline int hl_p2p_region_init(struct hl_device *hdev, struct hl_p2p_region *p2pr)
+{ return -EOPNOTSUPP; }
+static inline int hl_dio_start(struct hl_device *hdev) { return -EOPNOTSUPP; }
+static inline void hl_dio_stop(struct hl_device *hdev) {}
+
+static inline int hl_hldio_init(struct hl_device *hdev) { return 0; }
+static inline void hl_hldio_fini(struct hl_device *hdev) { }
+static inline long hl_hldio_ioctl(struct file *f, unsigned int c,
+				  unsigned long a)
+{ return -ENOTTY; }
+static inline void hl_hldio_debugfs_init(struct hl_device *hdev) { }
+static inline void hl_hldio_debugfs_fini(struct hl_device *hdev) { }
+
+#endif /* CONFIG_HL_HLDIO */
+
+/* Simplified polling macro for HLDIO (no simulator support) */
+#define hl_poll_timeout_condition(hdev, cond, sleep_us, timeout_us) \
+({ \
+	ktime_t __timeout = ktime_add_us(ktime_get(), timeout_us); \
+	might_sleep_if(sleep_us); \
+	(void)(hdev); /* keep signature consistent, hdev unused */ \
+	for (;;) { \
+		mb(); /* ensure ordering of memory operations */ \
+		if (cond) \
+			break; \
+		if (timeout_us && ktime_compare(ktime_get(), __timeout) > 0) \
+			break; \
+		if (sleep_us) \
+			usleep_range((sleep_us >> 2) + 1, sleep_us); \
+	} \
+	(cond) ? 0 : -ETIMEDOUT; \
+})
+
+#ifdef CONFIG_HL_HLDIO
+bool hl_device_supports_nvme(struct hl_device *hdev);
+#else
+static inline bool hl_device_supports_nvme(struct hl_device *hdev) { return false; }
+#endif
+
+#endif /* __HL_HLDIO_H__ */

From eeb38d0e91860fb47bc42a2cc44ffd80989a04fc Mon Sep 17 00:00:00 2001
From: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Date: Mon, 16 Sep 2024 15:15:32 +0300
Subject: [PATCH 1435/1455] accel/habanalabs: add debugfs interface for HLDIO
 testing
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Add debugfs files for NVMe Direct I/O (HLDIO) functionality.
This interface allows userspace access to direct SSD ↔ device transfers
through debugfs nodes.

Four debugfs files are created under /sys/kernel/debug/habanalabs/hlN/:

  - dio_ssd2hl : trigger SSD-to-device transfers
  - dio_hl2ssd : trigger device-to-SSD transfers
    (placeholder, not yet implemented)
  - dio_stats  : show transfer statistics
  - dio_reset  : reset statistics counters

Usage examples:

  # Perform SSD → device transfer
  echo "fd=3 va=0x10000 off=0 len=4096" > \
    /sys/kernel/debug/habanalabs/hl0/dio_ssd2hl

  # View statistics
  cat /sys/kernel/debug/habanalabs/hl0/dio_stats

  # Reset counters
  echo 1 > /sys/kernel/debug/habanalabs/hl0/dio_reset

This interface provides access to HLDIO functionality for validation
and diagnostics.

Signed-off-by: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Reviewed-by: Farah Kassabri <farah.kassabri@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/debugfs.c    | 210 +++++++++++++++++++
 drivers/accel/habanalabs/common/habanalabs.h |   4 +
 2 files changed, 214 insertions(+)

diff --git a/drivers/accel/habanalabs/common/debugfs.c b/drivers/accel/habanalabs/common/debugfs.c
index f3600df064d1..5f0820b19ccb 100644
--- a/drivers/accel/habanalabs/common/debugfs.c
+++ b/drivers/accel/habanalabs/common/debugfs.c
@@ -6,6 +6,7 @@
  */
 
 #include "habanalabs.h"
+#include "hldio.h"
 #include "../include/hw_ip/mmu/mmu_general.h"
 
 #include <linux/pci.h>
@@ -602,6 +603,198 @@ static int engines_show(struct seq_file *s, void *data)
 	return 0;
 }
 
+#ifdef CONFIG_HL_HLDIO
+/* DIO debugfs functions following the standard pattern */
+static int dio_ssd2hl_show(struct seq_file *s, void *data)
+{
+	struct hl_debugfs_entry *entry = s->private;
+	struct hl_dbg_device_entry *dev_entry = entry->dev_entry;
+	struct hl_device *hdev = dev_entry->hdev;
+
+	if (!hdev->asic_prop.supports_nvme) {
+		seq_puts(s, "NVMe Direct I/O not supported\\n");
+		return 0;
+	}
+
+	seq_puts(s, "Usage: echo \"fd=N va=0xADDR off=N len=N\" > dio_ssd2hl\n");
+	seq_printf(s, "Last transfer: %zu bytes\\n", dev_entry->dio_stats.last_len_read);
+	seq_puts(s, "Note: All parameters must be page-aligned (4KB)\\n");
+
+	return 0;
+}
+
+static ssize_t dio_ssd2hl_write(struct file *file, const char __user *buf,
+				size_t count, loff_t *f_pos)
+{
+	struct seq_file *s = file->private_data;
+	struct hl_debugfs_entry *entry = s->private;
+	struct hl_dbg_device_entry *dev_entry = entry->dev_entry;
+	struct hl_device *hdev = dev_entry->hdev;
+	struct hl_ctx *ctx = hdev->kernel_ctx;
+	char kbuf[128];
+	u64 device_va = 0, off_bytes = 0, len_bytes = 0;
+	u32 fd = 0;
+	size_t len_read = 0;
+	int rc, parsed;
+
+	if (!hdev->asic_prop.supports_nvme)
+		return -EOPNOTSUPP;
+
+	if (count >= sizeof(kbuf))
+		return -EINVAL;
+
+	if (copy_from_user(kbuf, buf, count))
+		return -EFAULT;
+
+	kbuf[count] = 0;
+
+	/* Parse: fd=N va=0xADDR off=N len=N */
+	parsed = sscanf(kbuf, "fd=%u va=0x%llx off=%llu len=%llu",
+			&fd, &device_va, &off_bytes, &len_bytes);
+	if (parsed != 4) {
+		dev_err(hdev->dev, "Invalid format. Expected: fd=N va=0xADDR off=N len=N\\n");
+		return -EINVAL;
+	}
+
+	/* Validate file descriptor */
+	if (fd == 0) {
+		dev_err(hdev->dev, "Invalid file descriptor: %u\\n", fd);
+		return -EINVAL;
+	}
+
+	/* Validate alignment requirements */
+	if (!IS_ALIGNED(device_va, PAGE_SIZE) ||
+	    !IS_ALIGNED(off_bytes, PAGE_SIZE) ||
+	    !IS_ALIGNED(len_bytes, PAGE_SIZE)) {
+		dev_err(hdev->dev,
+			"All parameters must be page-aligned (4KB)\\n");
+		return -EINVAL;
+	}
+
+	/* Validate transfer size */
+	if (len_bytes == 0 || len_bytes > SZ_1G) {
+		dev_err(hdev->dev, "Invalid length: %llu (max 1GB)\\n",
+			len_bytes);
+		return -EINVAL;
+	}
+
+	dev_dbg(hdev->dev, "DIO SSD2HL: fd=%u va=0x%llx off=%llu len=%llu\\n",
+		fd, device_va, off_bytes, len_bytes);
+
+	rc = hl_dio_ssd2hl(hdev, ctx, fd, device_va, off_bytes, len_bytes, &len_read);
+	if (rc < 0) {
+		dev_entry->dio_stats.failed_ops++;
+		dev_err(hdev->dev, "SSD2HL operation failed: %d\\n", rc);
+		return rc;
+	}
+
+	/* Update statistics */
+	dev_entry->dio_stats.total_ops++;
+	dev_entry->dio_stats.successful_ops++;
+	dev_entry->dio_stats.bytes_transferred += len_read;
+	dev_entry->dio_stats.last_len_read = len_read;
+
+	dev_dbg(hdev->dev, "DIO SSD2HL completed: %zu bytes transferred\\n", len_read);
+
+	return count;
+}
+
+static int dio_hl2ssd_show(struct seq_file *s, void *data)
+{
+	seq_puts(s, "HL2SSD (device-to-SSD) transfers not implemented\\n");
+	return 0;
+}
+
+static ssize_t dio_hl2ssd_write(struct file *file, const char __user *buf,
+			       size_t count, loff_t *f_pos)
+{
+	struct seq_file *s = file->private_data;
+	struct hl_debugfs_entry *entry = s->private;
+	struct hl_dbg_device_entry *dev_entry = entry->dev_entry;
+	struct hl_device *hdev = dev_entry->hdev;
+
+	if (!hdev->asic_prop.supports_nvme)
+		return -EOPNOTSUPP;
+
+	dev_dbg(hdev->dev, "HL2SSD operation not implemented\\n");
+	return -EOPNOTSUPP;
+}
+
+static int dio_stats_show(struct seq_file *s, void *data)
+{
+	struct hl_debugfs_entry *entry = s->private;
+	struct hl_dbg_device_entry *dev_entry = entry->dev_entry;
+	struct hl_device *hdev = dev_entry->hdev;
+	struct hl_dio_stats *stats = &dev_entry->dio_stats;
+	u64 avg_bytes_per_op = 0, success_rate = 0;
+
+	if (!hdev->asic_prop.supports_nvme) {
+		seq_puts(s, "NVMe Direct I/O not supported\\n");
+		return 0;
+	}
+
+	if (stats->successful_ops > 0)
+		avg_bytes_per_op = stats->bytes_transferred / stats->successful_ops;
+
+	if (stats->total_ops > 0)
+		success_rate = (stats->successful_ops * 100) / stats->total_ops;
+
+	seq_puts(s, "=== Habanalabs Direct I/O Statistics ===\\n");
+	seq_printf(s, "Total operations:     %llu\\n", stats->total_ops);
+	seq_printf(s, "Successful ops:       %llu\\n", stats->successful_ops);
+	seq_printf(s, "Failed ops:           %llu\\n", stats->failed_ops);
+	seq_printf(s, "Success rate:         %llu%%\\n", success_rate);
+	seq_printf(s, "Total bytes:          %llu\\n", stats->bytes_transferred);
+	seq_printf(s, "Avg bytes per op:     %llu\\n", avg_bytes_per_op);
+	seq_printf(s, "Last transfer:        %zu bytes\\n", stats->last_len_read);
+
+	return 0;
+}
+
+static int dio_reset_show(struct seq_file *s, void *data)
+{
+	seq_puts(s, "Write '1' to reset DIO statistics\\n");
+	return 0;
+}
+
+static ssize_t dio_reset_write(struct file *file, const char __user *buf,
+			       size_t count, loff_t *f_pos)
+{
+	struct seq_file *s = file->private_data;
+	struct hl_debugfs_entry *entry = s->private;
+	struct hl_dbg_device_entry *dev_entry = entry->dev_entry;
+	struct hl_device *hdev = dev_entry->hdev;
+	char kbuf[8];
+	unsigned long val;
+	int rc;
+
+	if (!hdev->asic_prop.supports_nvme)
+		return -EOPNOTSUPP;
+
+	if (count >= sizeof(kbuf))
+		return -EINVAL;
+
+	if (copy_from_user(kbuf, buf, count))
+		return -EFAULT;
+
+	kbuf[count] = 0;
+
+	rc = kstrtoul(kbuf, 0, &val);
+	if (rc)
+		return rc;
+
+	if (val == 1) {
+		memset(&dev_entry->dio_stats, 0, sizeof(dev_entry->dio_stats));
+		dev_dbg(hdev->dev, "DIO statistics reset\\n");
+	} else {
+		dev_err(hdev->dev, "Write '1' to reset statistics\\n");
+		return -EINVAL;
+	}
+
+	return count;
+}
+#endif
+
 static ssize_t hl_memory_scrub(struct file *f, const char __user *buf,
 					size_t count, loff_t *ppos)
 {
@@ -1633,6 +1826,13 @@ static const struct hl_info_list hl_debugfs_list[] = {
 	{"mmu", mmu_show, mmu_asid_va_write},
 	{"mmu_error", mmu_ack_error, mmu_ack_error_value_write},
 	{"engines", engines_show, NULL},
+#ifdef CONFIG_HL_HLDIO
+	/* DIO entries - only created if NVMe is supported */
+	{"dio_ssd2hl", dio_ssd2hl_show, dio_ssd2hl_write},
+	{"dio_stats", dio_stats_show, NULL},
+	{"dio_reset", dio_reset_show, dio_reset_write},
+	{"dio_hl2ssd", dio_hl2ssd_show, dio_hl2ssd_write},
+#endif
 };
 
 static int hl_debugfs_open(struct inode *inode, struct file *file)
@@ -1831,6 +2031,11 @@ static void add_files_to_device(struct hl_device *hdev, struct hl_dbg_device_ent
 				&hdev->asic_prop.server_type);
 
 	for (i = 0, entry = dev_entry->entry_arr ; i < count ; i++, entry++) {
+		/* Skip DIO entries if NVMe is not supported */
+		if (strncmp(hl_debugfs_list[i].name, "dio_", 4) == 0 &&
+		    !hdev->asic_prop.supports_nvme)
+			continue;
+
 		debugfs_create_file(hl_debugfs_list[i].name,
 					0644,
 					root,
@@ -1873,6 +2078,11 @@ int hl_debugfs_device_init(struct hl_device *hdev)
 	spin_lock_init(&hdev->debugfs_cfg_accesses.lock);
 	hdev->debugfs_cfg_accesses.head = 0; /* already zero by alloc but explicit init is fine */
 
+#ifdef CONFIG_HL_HLDIO
+	/* Initialize DIO statistics */
+	memset(&dev_entry->dio_stats, 0, sizeof(dev_entry->dio_stats));
+#endif
+
 	return 0;
 }
 
diff --git a/drivers/accel/habanalabs/common/habanalabs.h b/drivers/accel/habanalabs/common/habanalabs.h
index 6be185d7be58..d94c2ba22a6a 100644
--- a/drivers/accel/habanalabs/common/habanalabs.h
+++ b/drivers/accel/habanalabs/common/habanalabs.h
@@ -2410,6 +2410,7 @@ struct hl_debugfs_entry {
  * @i2c_addr: generic u8 debugfs file for address value to use in i2c_data_read.
  * @i2c_reg: generic u8 debugfs file for register value to use in i2c_data_read.
  * @i2c_len: generic u8 debugfs file for length value to use in i2c_data_read.
+ * @dio_stats: Direct I/O statistics
  */
 struct hl_dbg_device_entry {
 	struct dentry			*root;
@@ -2441,6 +2442,9 @@ struct hl_dbg_device_entry {
 	u8				i2c_addr;
 	u8				i2c_reg;
 	u8				i2c_len;
+#ifdef CONFIG_HL_HLDIO
+	struct hl_dio_stats	dio_stats;
+#endif
 };
 
 /**

From 65a3f5bc331ca7384900641b46cadff63805c10e Mon Sep 17 00:00:00 2001
From: Ariel Aviad <ariel.aviad@intel.com>
Date: Tue, 24 Sep 2024 11:02:49 +0300
Subject: [PATCH 1436/1455] accel/habanalabs: add HL_GET_P_STATE passthrough
 type

Add a new passthrough type HL_GET_P_STATE to the cpucp generic ioctl
to allow userspace to read the device performance state via firmware.

Signed-off-by: Ariel Aviad <ariel.aviad@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/habanalabs_ioctl.c | 3 +++
 include/linux/habanalabs/cpucp_if.h                | 2 ++
 2 files changed, 5 insertions(+)

diff --git a/drivers/accel/habanalabs/common/habanalabs_ioctl.c b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
index ed0dee5aa9be..fdfdabc85e54 100644
--- a/drivers/accel/habanalabs/common/habanalabs_ioctl.c
+++ b/drivers/accel/habanalabs/common/habanalabs_ioctl.c
@@ -964,6 +964,9 @@ static int send_fw_generic_request(struct hl_device *hdev, struct hl_info_args *
 	case  HL_GET_ERR_COUNTERS_CMD:
 		need_input_buff = true;
 		break;
+	case HL_GET_P_STATE:
+		need_input_buff = false;
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/include/linux/habanalabs/cpucp_if.h b/include/linux/habanalabs/cpucp_if.h
index 29c50e7427d1..45f181bcf890 100644
--- a/include/linux/habanalabs/cpucp_if.h
+++ b/include/linux/habanalabs/cpucp_if.h
@@ -1426,10 +1426,12 @@ struct cpucp_monitor_dump {
  *
  * HL_PASSTHROUGHT_VERSIONS	- Fetch all firmware versions.
  * HL_GET_ERR_COUNTERS_CMD	- Command to get error counters
+ * HL_GET_P_STATE		- get performance state
  */
 enum hl_passthrough_type {
 	HL_PASSTHROUGH_VERSIONS,
 	HL_GET_ERR_COUNTERS_CMD,
+	HL_GET_P_STATE,
 };
 
 #endif /* CPUCP_IF_H */

From a0d866bab184161ba155b352650083bf6695e50e Mon Sep 17 00:00:00 2001
From: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Date: Tue, 1 Oct 2024 15:52:27 +0300
Subject: [PATCH 1437/1455] accel/habanalabs/gaudi2: read preboot status after
 recovering from dirty state

Dirty state can occur when the host VM undergoes a reset while the
device does not. In such a case, the driver must reset the device before
it can be used again. As part of this reset, the device capabilities
are zeroed. Therefore, the driver must read the Preboot status again to
learn the Preboot state, capabilities, and security configuration.

Signed-off-by: Konstantin Sinyuk <konstantin.sinyuk@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/gaudi2/gaudi2.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/accel/habanalabs/gaudi2/gaudi2.c b/drivers/accel/habanalabs/gaudi2/gaudi2.c
index 03dde63ed875..b8c0689dba64 100644
--- a/drivers/accel/habanalabs/gaudi2/gaudi2.c
+++ b/drivers/accel/habanalabs/gaudi2/gaudi2.c
@@ -3498,7 +3498,6 @@ static int gaudi2_early_init(struct hl_device *hdev)
 	rc = hl_fw_read_preboot_status(hdev);
 	if (rc) {
 		if (hdev->reset_on_preboot_fail)
-			/* we are already on failure flow, so don't check if hw_fini fails. */
 			hdev->asic_funcs->hw_fini(hdev, true, false);
 		goto pci_fini;
 	}
@@ -3510,6 +3509,13 @@ static int gaudi2_early_init(struct hl_device *hdev)
 			dev_err(hdev->dev, "failed to reset HW in dirty state (%d)\n", rc);
 			goto pci_fini;
 		}
+
+		rc = hl_fw_read_preboot_status(hdev);
+		if (rc) {
+			if (hdev->reset_on_preboot_fail)
+				hdev->asic_funcs->hw_fini(hdev, true, false);
+			goto pci_fini;
+		}
 	}
 
 	return 0;

From 6ca282c3e635cd98cc5a9bb24606b41379e1fd8e Mon Sep 17 00:00:00 2001
From: Pavan S <pavan.sreenivas@intel.com>
Date: Wed, 2 Oct 2024 10:46:40 +0300
Subject: [PATCH 1438/1455] accel/habanalabs: add Infineon version check
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On HL338 ASICs, the Infineon first‑stage firmware is not present and
the reported version is 0. In this case printing a version number is
misleading, as it suggests valid firmware when it does not exist.

Fix this by printing the first‑stage Infineon firmware version only
if the reported value is non‑zero. This avoids confusing or incorrect
log messages on devices where the first stage is not applicable.

Signed-off-by: Pavan S <pavan.sreenivas@intel.com>
Reviewed-by: Koby Elbaz <koby.elbaz@intel.com>
Signed-off-by: Koby Elbaz <koby.elbaz@intel.com>
---
 drivers/accel/habanalabs/common/sysfs.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/accel/habanalabs/common/sysfs.c b/drivers/accel/habanalabs/common/sysfs.c
index 82f66520ec18..8f55ba3b4e73 100644
--- a/drivers/accel/habanalabs/common/sysfs.c
+++ b/drivers/accel/habanalabs/common/sysfs.c
@@ -96,14 +96,21 @@ static ssize_t vrm_ver_show(struct device *dev, struct device_attribute *attr, c
 	infineon_second_stage_third_instance =
 					(infineon_second_stage_version >> 16) & mask;
 
-	if (cpucp_info->infineon_second_stage_version)
+	if (cpucp_info->infineon_version && cpucp_info->infineon_second_stage_version)
 		return sprintf(buf, "%#04x %#04x:%#04x:%#04x\n",
 				le32_to_cpu(cpucp_info->infineon_version),
 				infineon_second_stage_first_instance,
 				infineon_second_stage_second_instance,
 				infineon_second_stage_third_instance);
-	else
+	else if (cpucp_info->infineon_second_stage_version)
+		return sprintf(buf, "%#04x:%#04x:%#04x\n",
+				infineon_second_stage_first_instance,
+				infineon_second_stage_second_instance,
+				infineon_second_stage_third_instance);
+	else if (cpucp_info->infineon_version)
 		return sprintf(buf, "%#04x\n", le32_to_cpu(cpucp_info->infineon_version));
+
+	return 0;
 }
 
 static DEVICE_ATTR_RO(vrm_ver);

From 7469567d882374dcac3fdb8b300e0f28cf875a75 Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Tue, 16 Sep 2025 13:11:06 +0800
Subject: [PATCH 1439/1455] drm/amdgpu: Add fallback to pipe reset if KCQ ring
 reset fails

Add a fallback mechanism to attempt pipe reset when KCQ reset
fails to recover the ring. After performing the KCQ reset and
queue remapping, test the ring functionality. If the ring test
fails, initiate a pipe reset as an additional recovery step.

v2: fix the typo (Lijo)
v3: try pipeline reset when kiq mapping fails (Lijo)

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index 8ba66d4dfe86..77f9d5b9a556 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -3560,6 +3560,7 @@ static int gfx_v9_4_3_reset_kcq(struct amdgpu_ring *ring,
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_kiq *kiq = &adev->gfx.kiq[ring->xcc_id];
 	struct amdgpu_ring *kiq_ring = &kiq->ring;
+	int reset_mode = AMDGPU_RESET_TYPE_PER_QUEUE;
 	unsigned long flags;
 	int r;
 
@@ -3597,6 +3598,7 @@ pipe_reset:
 		if (!(adev->gfx.compute_supported_reset & AMDGPU_RESET_TYPE_PER_PIPE))
 			return -EOPNOTSUPP;
 		r = gfx_v9_4_3_reset_hw_pipe(ring);
+		reset_mode = AMDGPU_RESET_TYPE_PER_PIPE;
 		dev_info(adev->dev, "ring: %s pipe reset :%s\n", ring->name,
 				r ? "failed" : "successfully");
 		if (r)
@@ -3619,10 +3621,20 @@ pipe_reset:
 	r = amdgpu_ring_test_ring(kiq_ring);
 	spin_unlock_irqrestore(&kiq->ring_lock, flags);
 	if (r) {
+		if (reset_mode == AMDGPU_RESET_TYPE_PER_QUEUE)
+			goto pipe_reset;
+
 		dev_err(adev->dev, "fail to remap queue\n");
 		return r;
 	}
 
+	if (reset_mode == AMDGPU_RESET_TYPE_PER_QUEUE) {
+		r = amdgpu_ring_test_ring(ring);
+		if (r)
+			goto pipe_reset;
+	}
+
+
 	return amdgpu_ring_reset_helper_end(ring, timedout_fence);
 }
 

From 66f3883dbc9245ca96abf97c2a4dd2a1e898b32f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Tue, 16 Sep 2025 16:07:35 +0200
Subject: [PATCH 1440/1455] drm/amdgpu: remove leftover from enforcing
 isolation by VMID
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Initially we enforced isolation by reserving a VMID, but that practice
was now removed.

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index 5dd78a9cb12d..cbdf108612d2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
@@ -594,11 +594,6 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 			list_add_tail(&id_mgr->ids[j].list, &id_mgr->ids_lru);
 		}
 	}
-	/* alloc a default reserved vmid to enforce isolation */
-	for (i = 0; i < (adev->xcp_mgr ? adev->xcp_mgr->num_xcps : 1); i++) {
-		if (adev->enforce_isolation[i] != AMDGPU_ENFORCE_ISOLATION_DISABLE)
-			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(i));
-	}
 }
 
 /**

From 90e09ea4cfd4aaaf07ababa6d8c880035587e7e9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>
Date: Fri, 19 Sep 2025 09:27:03 +0200
Subject: [PATCH 1441/1455] drm/amdgpu: revert "rework reserved VMID handling"
 v2
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This reverts commit e44a0fe630c58b0a87d8281f5c1077a3479e5fce.

Initially we used VMID reservation to enforce isolation between
processes. That has now been replaced by proper fence handling.

Both OpenGL, RADV and ROCm developers requested a way to reserve a VMID
for SPM, so restore that approach by reverting back to only allowing a
single process to use the reserved VMID.

Only compile tested for now.

v2: use -ENOENT instead of -EINVAL if VMID is not available

Signed-off-by: Christian König <christian.koenig@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c | 61 ++++++++++++++++---------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h | 11 ++---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c  | 17 ++-----
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h  |  2 +-
 4 files changed, 50 insertions(+), 41 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index cbdf108612d2..3ef5bc95642c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
@@ -275,13 +275,12 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm,
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned vmhub = ring->vm_hub;
-	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 	uint64_t fence_context = adev->fence_context + ring->idx;
 	bool needs_flush = vm->use_cpu_for_update;
 	uint64_t updates = amdgpu_vm_tlb_seq(vm);
 	int r;
 
-	*id = id_mgr->reserved;
+	*id = vm->reserved_vmid[vmhub];
 	if ((*id)->owner != vm->immediate.fence_context ||
 	    !amdgpu_vmid_compatible(*id, job) ||
 	    (*id)->flushed_updates < updates ||
@@ -474,40 +473,61 @@ bool amdgpu_vmid_uses_reserved(struct amdgpu_vm *vm, unsigned int vmhub)
 	return vm->reserved_vmid[vmhub];
 }
 
-int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
+/*
+ * amdgpu_vmid_alloc_reserved - reserve a specific VMID for this vm
+ * @adev: amdgpu device structure
+ * @vm: the VM to reserve an ID for
+ * @vmhub: the VMHUB which should be used
+ *
+ * Mostly used to have a reserved VMID for debugging and SPM.
+ *
+ * Returns: 0 for success, -ENOENT if an ID is already reserved.
+ */
+int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
+	struct amdgpu_vmid *id;
+	int r = 0;
 
 	mutex_lock(&id_mgr->lock);
-
-	++id_mgr->reserved_use_count;
-	if (!id_mgr->reserved) {
-		struct amdgpu_vmid *id;
-
-		id = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid,
-				      list);
-		/* Remove from normal round robin handling */
-		list_del_init(&id->list);
-		id_mgr->reserved = id;
+	if (vm->reserved_vmid[vmhub])
+		goto unlock;
+	if (id_mgr->reserved_vmid) {
+		r = -ENOENT;
+		goto unlock;
 	}
-
+	/* Remove from normal round robin handling */
+	id = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid, list);
+	list_del_init(&id->list);
+	vm->reserved_vmid[vmhub] = id;
+	id_mgr->reserved_vmid = true;
 	mutex_unlock(&id_mgr->lock);
+
 	return 0;
+unlock:
+	mutex_unlock(&id_mgr->lock);
+	return r;
 }
 
-void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
+/*
+ * amdgpu_vmid_free_reserved - free up a reserved VMID again
+ * @adev: amdgpu device structure
+ * @vm: the VM with the reserved ID
+ * @vmhub: the VMHUB which should be used
+ */
+void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 
 	mutex_lock(&id_mgr->lock);
-	if (!--id_mgr->reserved_use_count) {
-		/* give the reserved ID back to normal round robin */
-		list_add(&id_mgr->reserved->list, &id_mgr->ids_lru);
-		id_mgr->reserved = NULL;
+	if (vm->reserved_vmid[vmhub]) {
+		list_add(&vm->reserved_vmid[vmhub]->list,
+			&id_mgr->ids_lru);
+		vm->reserved_vmid[vmhub] = NULL;
+		id_mgr->reserved_vmid = false;
 	}
-
 	mutex_unlock(&id_mgr->lock);
 }
 
@@ -574,7 +594,6 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 
 		mutex_init(&id_mgr->lock);
 		INIT_LIST_HEAD(&id_mgr->ids_lru);
-		id_mgr->reserved_use_count = 0;
 
 		/* for GC <10, SDMA uses MMHUB so use first_kfd_vmid for both GC and MM */
 		if (amdgpu_ip_version(adev, GC_HWIP, 0) < IP_VERSION(10, 0, 0))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h
index 240fa6751260..b3649cd3af56 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.h
@@ -67,8 +67,7 @@ struct amdgpu_vmid_mgr {
 	unsigned		num_ids;
 	struct list_head	ids_lru;
 	struct amdgpu_vmid	ids[AMDGPU_NUM_VMID];
-	struct amdgpu_vmid	*reserved;
-	unsigned int		reserved_use_count;
+	bool			reserved_vmid;
 };
 
 int amdgpu_pasid_alloc(unsigned int bits);
@@ -79,10 +78,10 @@ void amdgpu_pasid_free_delayed(struct dma_resv *resv,
 bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev,
 			       struct amdgpu_vmid *id);
 bool amdgpu_vmid_uses_reserved(struct amdgpu_vm *vm, unsigned int vmhub);
-int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
-				unsigned vmhub);
-void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
-				unsigned vmhub);
+int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			       unsigned vmhub);
+void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+			       unsigned vmhub);
 int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
 		     struct amdgpu_job *job, struct dma_fence **fence);
 void amdgpu_vmid_reset(struct amdgpu_device *adev, unsigned vmhub,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 112ce584a5ad..8c28e8923f02 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -2790,10 +2790,7 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	dma_fence_put(vm->last_update);
 
 	for (i = 0; i < AMDGPU_MAX_VMHUBS; i++) {
-		if (vm->reserved_vmid[i]) {
-			amdgpu_vmid_free_reserved(adev, i);
-			vm->reserved_vmid[i] = false;
-		}
+		amdgpu_vmid_free_reserved(adev, vm, i);
 	}
 
 	ttm_lru_bulk_move_fini(&adev->mman.bdev, &vm->lru_bulk_move);
@@ -2889,6 +2886,7 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	union drm_amdgpu_vm *args = data;
 	struct amdgpu_device *adev = drm_to_adev(dev);
 	struct amdgpu_fpriv *fpriv = filp->driver_priv;
+	struct amdgpu_vm *vm = &fpriv->vm;
 
 	/* No valid flags defined yet */
 	if (args->in.flags)
@@ -2897,17 +2895,10 @@ int amdgpu_vm_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
 	switch (args->in.op) {
 	case AMDGPU_VM_OP_RESERVE_VMID:
 		/* We only have requirement to reserve vmid from gfxhub */
-		if (!fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) {
-			amdgpu_vmid_alloc_reserved(adev, AMDGPU_GFXHUB(0));
-			fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = true;
-		}
-
+		amdgpu_vmid_alloc_reserved(adev, vm, AMDGPU_GFXHUB(0));
 		break;
 	case AMDGPU_VM_OP_UNRESERVE_VMID:
-		if (fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)]) {
-			amdgpu_vmid_free_reserved(adev, AMDGPU_GFXHUB(0));
-			fpriv->vm.reserved_vmid[AMDGPU_GFXHUB(0)] = false;
-		}
+		amdgpu_vmid_free_reserved(adev, vm, AMDGPU_GFXHUB(0));
 		break;
 	default:
 		return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 988e970d9e96..adc5c9161fa8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -415,7 +415,7 @@ struct amdgpu_vm {
 	struct dma_fence	*last_unlocked;
 
 	unsigned int		pasid;
-	bool			reserved_vmid[AMDGPU_MAX_VMHUBS];
+	struct amdgpu_vmid	*reserved_vmid[AMDGPU_MAX_VMHUBS];
 
 	/* Flag to indicate if VM tables are updated by CPU or GPU (SDMA) */
 	bool					use_cpu_for_update;

From 883bd89d00085c2c5f1efcd25861745cb039f9e3 Mon Sep 17 00:00:00 2001
From: Prike Liang <Prike.Liang@amd.com>
Date: Fri, 19 Sep 2025 14:31:50 +0800
Subject: [PATCH 1442/1455] drm/amdgpu/userq: assign an error code for invalid
 userq va

It should return an error code if userq VA validation fails.

Fixes: 9e46b8bb0539 ("drm/amdgpu: validate userq buffer virtual address and size")
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
index a22e6025de61..48e0932f5b62 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c
@@ -71,6 +71,7 @@ int amdgpu_userq_input_va_validate(struct amdgpu_vm *vm, u64 addr,
 		return 0;
 	}
 
+	r = -EINVAL;
 out_err:
 	amdgpu_bo_unreserve(vm->root.bo);
 	return r;
@@ -508,6 +509,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args)
 	if (amdgpu_userq_input_va_validate(&fpriv->vm, args->in.queue_va, args->in.queue_size) ||
 	    amdgpu_userq_input_va_validate(&fpriv->vm, args->in.rptr_va, AMDGPU_GPU_PAGE_SIZE) ||
 	    amdgpu_userq_input_va_validate(&fpriv->vm, args->in.wptr_va, AMDGPU_GPU_PAGE_SIZE)) {
+		r = -EINVAL;
 		kfree(queue);
 		goto unlock;
 	}

From 4e3b45d7b6c36d7d1b9a30b13d2dfa890e7a0763 Mon Sep 17 00:00:00 2001
From: Sunil Khatri <sunil.khatri@amd.com>
Date: Wed, 24 Sep 2025 18:03:07 +0530
Subject: [PATCH 1443/1455] drm/amdgpu: remove the redeclaration of variable i
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Variable "i" has been redeclared as integer later in the function
which is wrong and not serving any purpose.

Fixes: 899fbde14646 ("drm/amdgpu: replace get_user_pages with HMM mirror helpers")
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 744e6ff69814..9cd7741d2254 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -884,7 +884,6 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 	amdgpu_bo_list_for_each_userptr_entry(e, p->bo_list) {
 		bool userpage_invalidated = false;
 		struct amdgpu_bo *bo = e->bo;
-		int i;
 
 		r = amdgpu_ttm_tt_get_user_pages(bo, &e->range);
 		if (r)

From 0fb915d64d99b2cd6164fd0c5304457aa417ea3c Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 24 Sep 2025 11:16:20 -0500
Subject: [PATCH 1444/1455] drm/amd/display: Only enable common modes for eDP
 and LVDS
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
The main reason common modes are added is for compatibility with
clone mode when a laptop is connected to a projector or external
monitor.  Since commit 978fa2f6d0b12 ("drm/amd/display: Use scaling
for non-native resolutions on eDP") when non-native modes are picked
for eDP the GPU scalar will be used.  This is because it is inconsistent
whether eDP panels have the capability to actually drive non-native
resolutions. With panels connected to other connectors this limitation
generally doesn't exist as we the EDID will advertise support for a
number of resolutions and monitors will use built in scaling hardware.

Comparing DC and non-DC code paths the non-DC code path only adds
common modes for LVDS and eDP whereas the DC codepath does it for
all connector types.

In the past there was an experiment done to disable common mode adding
for eDP and LVDS from commit 6d396e7ac1ce3 ("drm/amd/display: Disable
common modes for LVDS") and commit 7948afb46af92 ("drm/amd/display:
Disable common modes for eDP") but this was reverted in
commit a8b79b09185de ("drm/amd: Re-enable common modes for eDP and
LVDS") because it caused problems with Xorg.

[How]
Only add common modes for eDP and LVDS for DC, matching the behavior
of non-DC.

Suggested-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250924161624.1975819-2-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index 951eaeb7f793..8e1622bf7a42 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -8271,6 +8271,10 @@ static void amdgpu_dm_connector_add_common_modes(struct drm_encoder *encoder,
 		{"1920x1200", 1920, 1200}
 	};
 
+	if ((connector->connector_type != DRM_MODE_CONNECTOR_eDP) &&
+	    (connector->connector_type != DRM_MODE_CONNECTOR_LVDS))
+		return;
+
 	n = ARRAY_SIZE(common_modes);
 
 	for (i = 0; i < n; i++) {

From 210844d2c075e12927507097b7ac9ae7a4ae1c15 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 24 Sep 2025 11:16:21 -0500
Subject: [PATCH 1445/1455] drm/amd: Drop unnecessary check in
 amdgpu_connector_add_common_modes()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
amdgpu_connector_add_common_modes() has a check for the width and height
of common modes being too small, but the array of common_modes[] has fixed
values.  The check is dead code.

[How]
Drop unnecessary check.

Cc: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250924161624.1975819-3-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index bf38fc69c1cf..9b915f11ccac 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -434,8 +434,6 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 			     common_modes[i].h == native_mode->vdisplay))
 				continue;
 		}
-		if (common_modes[i].w < 320 || common_modes[i].h < 200)
-			continue;
 
 		mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h, 60, false, false, false);
 		if (!mode)

From 118800b0797a046adaa2a8e9dee9b971b78802a7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Wed, 24 Sep 2025 13:38:34 +0200
Subject: [PATCH 1446/1455] drm/amd/display: Reject modes with too high pixel
 clock on DCE6-10
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Reject modes with a pixel clock higher than the maximum display
clock. Use 400 MHz as a fallback value when the maximum display
clock is not known. Pixel clocks that are higher than the display
clock just won't work and are not supported.

With the addition of the YUV422	fallback, DC can now accidentally
select a mode requiring higher pixel clock than actually supported
when the DP version supports the required bandwidth but the clock
is otherwise too high for the display engine. DCE 6-10 don't
support these modes but they don't have a bandwidth calculation
to reject them properly.

Fixes: db291ed1732e ("drm/amd/display: Add fallback path for YCBCR422")
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c    |  3 +++
 .../drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c   |  5 +++++
 .../amd/display/dc/resource/dce100/dce100_resource.c   | 10 +++++++++-
 .../drm/amd/display/dc/resource/dce60/dce60_resource.c | 10 +++++++++-
 .../drm/amd/display/dc/resource/dce80/dce80_resource.c | 10 +++++++++-
 5 files changed, 35 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
index dbd6ef1b60a0..6131ede2db7a 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce100/dce_clk_mgr.c
@@ -463,6 +463,9 @@ void dce_clk_mgr_construct(
 		clk_mgr->max_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
 	clk_mgr->cur_min_clks_state = DM_PP_CLOCKS_STATE_INVALID;
 
+	base->clks.max_supported_dispclk_khz =
+		clk_mgr->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz;
+
 	dce_clock_read_integrated_info(clk_mgr);
 	dce_clock_read_ss_info(clk_mgr);
 }
diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
index a39641a0ff09..69dd80d9f738 100644
--- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
+++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dce60/dce60_clk_mgr.c
@@ -147,6 +147,8 @@ void dce60_clk_mgr_construct(
 		struct dc_context *ctx,
 		struct clk_mgr_internal *clk_mgr)
 {
+	struct clk_mgr *base = &clk_mgr->base;
+
 	dce_clk_mgr_construct(ctx, clk_mgr);
 
 	memcpy(clk_mgr->max_clks_by_state,
@@ -157,5 +159,8 @@ void dce60_clk_mgr_construct(
 	clk_mgr->clk_mgr_shift = &disp_clk_shift;
 	clk_mgr->clk_mgr_mask = &disp_clk_mask;
 	clk_mgr->base.funcs = &dce60_funcs;
+
+	base->clks.max_supported_dispclk_khz =
+		clk_mgr->max_clks_by_state[DM_PP_CLOCKS_STATE_PERFORMANCE].display_clk_khz;
 }
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
index 3a51be63f020..f36ec4edf0ae 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
@@ -29,6 +29,7 @@
 #include "stream_encoder.h"
 
 #include "resource.h"
+#include "clk_mgr.h"
 #include "include/irq_service_interface.h"
 #include "virtual/virtual_stream_encoder.h"
 #include "dce110/dce110_resource.h"
@@ -843,10 +844,17 @@ static enum dc_status dce100_validate_bandwidth(
 {
 	int i;
 	bool at_least_one_pipe = false;
+	struct dc_stream_state *stream = NULL;
+	const uint32_t max_pix_clk_khz = max(dc->clk_mgr->clks.max_supported_dispclk_khz, 400000);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		if (context->res_ctx.pipe_ctx[i].stream)
+		stream = context->res_ctx.pipe_ctx[i].stream;
+		if (stream) {
 			at_least_one_pipe = true;
+
+			if (stream->timing.pix_clk_100hz >= max_pix_clk_khz * 10)
+				return DC_FAIL_BANDWIDTH_VALIDATE;
+		}
 	}
 
 	if (at_least_one_pipe) {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
index c164d2500c2a..b5433349fc7a 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
@@ -34,6 +34,7 @@
 #include "stream_encoder.h"
 
 #include "resource.h"
+#include "clk_mgr.h"
 #include "include/irq_service_interface.h"
 #include "irq/dce60/irq_service_dce60.h"
 #include "dce110/dce110_timing_generator.h"
@@ -870,10 +871,17 @@ static enum dc_status dce60_validate_bandwidth(
 {
 	int i;
 	bool at_least_one_pipe = false;
+	struct dc_stream_state *stream = NULL;
+	const uint32_t max_pix_clk_khz = max(dc->clk_mgr->clks.max_supported_dispclk_khz, 400000);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		if (context->res_ctx.pipe_ctx[i].stream)
+		stream = context->res_ctx.pipe_ctx[i].stream;
+		if (stream) {
 			at_least_one_pipe = true;
+
+			if (stream->timing.pix_clk_100hz >= max_pix_clk_khz * 10)
+				return DC_FAIL_BANDWIDTH_VALIDATE;
+		}
 	}
 
 	if (at_least_one_pipe) {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
index 3e8b0ac11d90..538eafea82d5 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
@@ -32,6 +32,7 @@
 #include "stream_encoder.h"
 
 #include "resource.h"
+#include "clk_mgr.h"
 #include "include/irq_service_interface.h"
 #include "irq/dce80/irq_service_dce80.h"
 #include "dce110/dce110_timing_generator.h"
@@ -876,10 +877,17 @@ static enum dc_status dce80_validate_bandwidth(
 {
 	int i;
 	bool at_least_one_pipe = false;
+	struct dc_stream_state *stream = NULL;
+	const uint32_t max_pix_clk_khz = max(dc->clk_mgr->clks.max_supported_dispclk_khz, 400000);
 
 	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		if (context->res_ctx.pipe_ctx[i].stream)
+		stream = context->res_ctx.pipe_ctx[i].stream;
+		if (stream) {
 			at_least_one_pipe = true;
+
+			if (stream->timing.pix_clk_100hz >= max_pix_clk_khz * 10)
+				return DC_FAIL_BANDWIDTH_VALIDATE;
+		}
 	}
 
 	if (at_least_one_pipe) {

From 99d7181bca34e96fbf61bdb6844918bdd4df2814 Mon Sep 17 00:00:00 2001
From: Yifan Zhang <yifan1.zhang@amd.com>
Date: Tue, 16 Sep 2025 21:21:15 +0800
Subject: [PATCH 1447/1455] amd/amdkfd: resolve a race in
 amdgpu_amdkfd_device_fini_sw

There is race in amdgpu_amdkfd_device_fini_sw and interrupt.
if amdgpu_amdkfd_device_fini_sw run in b/w kfd_cleanup_nodes and
  kfree(kfd), and KGD interrupt generated.

kernel panic log:

BUG: kernel NULL pointer dereference, address: 0000000000000098
amdgpu 0000:c8:00.0: amdgpu: Requesting 4 partitions through PSP

PGD d78c68067 P4D d78c68067

kfd kfd: amdgpu: Allocated 3969056 bytes on gart

PUD 1465b8067 PMD @

Oops: @002 [#1] SMP NOPTI

kfd kfd: amdgpu: Total number of KFD nodes to be created: 4
CPU: 115 PID: @ Comm: swapper/115 Kdump: loaded Tainted: G S W OE K

RIP: 0010:_raw_spin_lock_irqsave+0x12/0x40

Code: 89 e@ 41 5c c3 cc cc cc cc 66 66 2e Of 1f 84 00 00 00 00 00 OF 1f 40 00 Of 1f 44% 00 00 41 54 9c 41 5c fa 31 cO ba 01 00 00 00 <fO> OF b1 17 75 Ba 4c 89 e@ 41 Sc

89 c6 e8 07 38 5d

RSP: 0018: ffffc90@1a6b0e28 EFLAGS: 00010046

RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000018
0000000000000001 RSI: ffff8883bb623e00 RDI: 0000000000000098
ffff8883bb000000 RO8: ffff888100055020 ROO: ffff888100055020
0000000000000000 R11: 0000000000000000 R12: 0900000000000002
ffff888F2b97da0@ R14: @000000000000098 R15: ffff8883babdfo00

CS: 010 DS: 0000 ES: 0000 CRO: 0000000080050033

CR2: 0000000000000098 CR3: 0000000e7cae2006 CR4: 0000000002770ce0
0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
0000000000000000 DR6: 00000000fffeO7FO DR7: 0000000000000400

PKRU: 55555554

Call Trace:

<IRQ>

kgd2kfd_interrupt+@x6b/0x1f@ [amdgpu]

? amdgpu_fence_process+0xa4/0x150 [amdgpu]

kfd kfd: amdgpu: Node: 0, interrupt_bitmap: 3 YcpxFl Rant tErace

amdgpu_irq_dispatch+0x165/0x210 [amdgpu]

amdgpu_ih_process+0x80/0x100 [amdgpu]

amdgpu: Virtual CRAT table created for GPU

amdgpu_irq_handler+0x1f/@x60 [amdgpu]

__handle_irq_event_percpu+0x3d/0x170

amdgpu: Topology: Add dGPU node [0x74a2:0x1002]

handle_irq_event+0x5a/@xcO

handle_edge_irq+0x93/0x240

kfd kfd: amdgpu: KFD node 1 partition @ size 49148M

asm_call_irq_on_stack+0xf/@x20

</IRQ>

common_interrupt+0xb3/0x130

asm_common_interrupt+0x1le/0x40

5.10.134-010.a1i5000.a18.x86_64 #1

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Philip Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 349c351e242b..051a00152b08 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -1133,7 +1133,15 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
 	}
 
 	for (i = 0; i < kfd->num_nodes; i++) {
-		node = kfd->nodes[i];
+		/* Race if another thread in b/w
+		 * kfd_cleanup_nodes and kfree(kfd),
+		 * when kfd->nodes[i] = NULL
+		 */
+		if (kfd->nodes[i])
+			node = kfd->nodes[i];
+		else
+			return;
+
 		spin_lock_irqsave(&node->interrupt_lock, flags);
 
 		if (node->interrupts_active

From 45da20e00d5da842e17dfc633072b127504f0d0e Mon Sep 17 00:00:00 2001
From: Yifan Zhang <yifan1.zhang@amd.com>
Date: Wed, 24 Sep 2025 23:19:14 +0800
Subject: [PATCH 1448/1455] amd/amdkfd: enhance kfd process check in switch
 partition

current switch partition only check if kfd_processes_table is empty.
kfd_prcesses_table entry is deleted in kfd_process_notifier_release, but
kfd_process tear down is in kfd_process_wq_release.

consider two processes:

Process A (workqueue) -> kfd_process_wq_release -> Access kfd_node member
Process B switch partition -> amdgpu_xcp_pre_partition_switch -> amdgpu_amdkfd_device_fini_sw
-> kfd_node tear down.

Process A and B may trigger a race as shown in dmesg log.

This patch is to resolve the race by adding an atomic kfd_process counter
kfd_processes_count, it increment as create kfd process, decrement as
finish kfd_process_wq_release.

v2: Put kfd_processes_count per kfd_dev, move decrement to kfd_process_destroy_pdds
and bug fix. (Philip Yang)

[3966658.307702] divide error: 0000 [#1] SMP NOPTI
[3966658.350818]  i10nm_edac
[3966658.356318] CPU: 124 PID: 38435 Comm: kworker/124:0 Kdump: loaded Tainted
[3966658.356890] Workqueue: kfd_process_wq kfd_process_wq_release [amdgpu]
[3966658.362839]  nfit
[3966658.366457] RIP: 0010:kfd_get_num_sdma_engines+0x17/0x40 [amdgpu]
[3966658.366460] Code: 00 00 e9 ac 81 02 00 66 66 2e 0f 1f 84 00 00 00 00 00 90 0f 1f 44 00 00 48 8b 4f 08 48 8b b7 00 01 00 00 8b 81 58 26 03 00 99 <f7> be b8 01 00 00 80 b9 70 2e 00 00 00 74 0b 83 f8 02 ba 02 00 00
[3966658.380967]  x86_pkg_temp_thermal
[3966658.391529] RSP: 0018:ffffc900a0edfdd8 EFLAGS: 00010246
[3966658.391531] RAX: 0000000000000008 RBX: ffff8974e593b800 RCX: ffff888645900000
[3966658.391531] RDX: 0000000000000000 RSI: ffff888129154400 RDI: ffff888129151c00
[3966658.391532] RBP: ffff8883ad79d400 R08: 0000000000000000 R09: ffff8890d2750af4
[3966658.391532] R10: 0000000000000018 R11: 0000000000000018 R12: 0000000000000000
[3966658.391533] R13: ffff8883ad79d400 R14: ffffe87ff662ba00 R15: ffff8974e593b800
[3966658.391533] FS:  0000000000000000(0000) GS:ffff88fe7f600000(0000) knlGS:0000000000000000
[3966658.391534] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[3966658.391534] CR2: 0000000000d71000 CR3: 000000dd0e970004 CR4: 0000000002770ee0
[3966658.391535] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[3966658.391535] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400
[3966658.391536] PKRU: 55555554
[3966658.391536] Call Trace:
[3966658.391674]  deallocate_sdma_queue+0x38/0xa0 [amdgpu]
[3966658.391762]  process_termination_cpsch+0x1ed/0x480 [amdgpu]
[3966658.399754]  intel_powerclamp
[3966658.402831]  kfd_process_dequeue_from_all_devices+0x5b/0xc0 [amdgpu]
[3966658.402908]  kfd_process_wq_release+0x1a/0x1a0 [amdgpu]
[3966658.410516]  coretemp
[3966658.434016]  process_one_work+0x1ad/0x380
[3966658.434021]  worker_thread+0x49/0x310
[3966658.438963]  kvm_intel
[3966658.446041]  ? process_one_work+0x380/0x380
[3966658.446045]  kthread+0x118/0x140
[3966658.446047]  ? __kthread_bind_mask+0x60/0x60
[3966658.446050]  ret_from_fork+0x1f/0x30
[3966658.446053] Modules linked in: kpatch_20765354(OEK)
[3966658.455310]  kvm
[3966658.464534]  mptcp_diag xsk_diag raw_diag unix_diag af_packet_diag netlink_diag udp_diag act_pedit act_mirred act_vlan cls_flower kpatch_21951273(OEK) kpatch_18424469(OEK) kpatch_19749756(OEK)
[3966658.473462]  idxd_mdev
[3966658.482306]  kpatch_17971294(OEK) sch_ingress xt_conntrack amdgpu(OE) amdxcp(OE) amddrm_buddy(OE) amd_sched(OE) amdttm(OE) amdkcl(OE) intel_ifs iptable_mangle tcm_loop target_core_pscsi tcp_diag target_core_file inet_diag target_core_iblock target_core_user target_core_mod coldpgs kpatch_18383292(OEK) ip6table_nat ip6table_filter ip6_tables ip_set_hash_ipportip ip_set_hash_ipportnet ip_set_hash_ipport ip_set_bitmap_port xt_comment iptable_nat nf_nat iptable_filter ip_tables ip_set ip_vs_sh ip_vs_wrr ip_vs_rr ip_vs nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 sn_core_odd(OE) i40e overlay binfmt_misc tun bonding(OE) aisqos(OE) aisqos_hotfixes(OE) rfkill uio_pci_generic uio cuse fuse nf_tables nfnetlink intel_rapl_msr intel_rapl_common intel_uncore_frequency intel_uncore_frequency_common i10nm_edac nfit x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm idxd_mdev
[3966658.491237]  vfio_pci
[3966658.501196]  vfio_pci vfio_virqfd mdev vfio_iommu_type1 vfio iax_crypto intel_pmt_telemetry iTCO_wdt intel_pmt_class iTCO_vendor_support irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel rapl intel_cstate snd_hda_intel snd_intel_dspcfg snd_hda_codec snd_hda_core snd_hwdep snd_seq
[3966658.508537]  vfio_virqfd
[3966658.517569]  snd_seq_device ipmi_ssif isst_if_mbox_pci isst_if_mmio pcspkr snd_pcm idxd intel_uncore ses isst_if_common intel_vsec idxd_bus enclosure snd_timer mei_me snd i2c_i801 i2c_smbus mei i2c_ismt soundcore joydev acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad vfat fat
[3966658.526851]  mdev
[3966658.536096]  nfsd auth_rpcgss nfs_acl lockd grace slb_vtoa(OE) sunrpc dm_mod hookers mlx5_ib(OE) ast i2c_algo_bit drm_vram_helper drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm_ttm_helper ttm mlx5_core(OE) mlxfw(OE)
[3966658.540381]  vfio_iommu_type1
[3966658.544341]  nvme mpt3sas tls drm nvme_core pci_hyperv_intf raid_class psample libcrc32c crc32c_intel mlxdevm(OE) i2c_core
[3966658.551254]  vfio
[3966658.558742]  scsi_transport_sas wmi pinctrl_emmitsburg sd_mod t10_pi sg ahci libahci libata rdma_ucm(OE) ib_uverbs(OE) rdma_cm(OE) iw_cm(OE) ib_cm(OE) ib_umad(OE) ib_core(OE) ib_ucm(OE) mlx_compat(OE)
[3966658.563004]  iax_crypto
[3966658.570988]  [last unloaded: diagnose]
[3966658.571027] ---[ end trace cc9dbb180f9ae537 ]---

Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Reviewed-by: Philip.Yang<Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_device.c  | 10 ++++++++++
 drivers/gpu/drm/amd/amdkfd/kfd_priv.h    |  2 ++
 drivers/gpu/drm/amd/amdkfd/kfd_process.c |  4 ++++
 3 files changed, 16 insertions(+)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index 051a00152b08..e9cfb80bd436 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -495,6 +495,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
 	mutex_init(&kfd->doorbell_mutex);
 
 	ida_init(&kfd->doorbell_ida);
+	atomic_set(&kfd->kfd_processes_count, 0);
 
 	return kfd;
 }
@@ -1493,6 +1494,15 @@ int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd)
 
 	mutex_lock(&kfd_processes_mutex);
 
+	/* kfd_processes_count is per kfd_dev, return -EBUSY without
+	 * further check
+	 */
+	if (!!atomic_read(&kfd->kfd_processes_count)) {
+		pr_debug("process_wq_release not finished\n");
+		r = -EBUSY;
+		goto out;
+	}
+
 	if (hash_empty(kfd_processes_table) && !kfd_is_locked(kfd))
 		goto out;
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index d01ef5ac0766..70ef051511bb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -382,6 +382,8 @@ struct kfd_dev {
 
 	/* for dynamic partitioning */
 	int kfd_dev_lock;
+
+	atomic_t kfd_processes_count;
 };
 
 enum kfd_mempool {
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 5be28c6c4f6a..ddfe30c13e9d 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -1088,6 +1088,8 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
 			pdd->runtime_inuse = false;
 		}
 
+		atomic_dec(&pdd->dev->kfd->kfd_processes_count);
+
 		kfree(pdd);
 		p->pdds[i] = NULL;
 	}
@@ -1649,6 +1651,8 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev,
 	/* Init idr used for memory handle translation */
 	idr_init(&pdd->alloc_idr);
 
+	atomic_inc(&dev->kfd->kfd_processes_count);
+
 	return pdd;
 }
 

From b8ae2640f9acd4f411c9227d2493755d03fe440a Mon Sep 17 00:00:00 2001
From: "Jesse.Zhang" <Jesse.Zhang@amd.com>
Date: Wed, 24 Sep 2025 16:00:06 +0800
Subject: [PATCH 1449/1455] drm/amdgpu: Fix fence signaling race condition in
 userqueue
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This commit fixes a potential race condition in the userqueue fence
signaling mechanism by replacing dma_fence_is_signaled_locked() with
dma_fence_is_signaled().

The issue occurred because:
1. dma_fence_is_signaled_locked() should only be used when holding
   the fence's individual lock, not just the fence list lock
2. Using the locked variant without the proper fence lock could lead
   to double-signaling scenarios:
   - Hardware completion signals the fence
   - Software path also tries to signal the same fence

By using dma_fence_is_signaled() instead, we properly handle the
locking hierarchy and avoid the race condition while still maintaining
the necessary synchronization through the fence_list_lock.

v2: drop the comment (Christian)

Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
index 95e91d1dc58a..761bad98da3e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq_fence.c
@@ -284,7 +284,7 @@ static int amdgpu_userq_fence_create(struct amdgpu_usermode_queue *userq,
 
 	/* Check if hardware has already processed the job */
 	spin_lock_irqsave(&fence_drv->fence_list_lock, flags);
-	if (!dma_fence_is_signaled_locked(fence))
+	if (!dma_fence_is_signaled(fence))
 		list_add_tail(&userq_fence->link, &fence_drv->fences);
 	else
 		dma_fence_put(fence);

From ee352f6c56e1775b192f2d39ad45362148e1fd16 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Wed, 24 Sep 2025 13:38:35 +0200
Subject: [PATCH 1450/1455] drm/amd/display: Share dce100_validate_bandwidth
 with DCE6-8
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

DCE6-8 have very similar capabilities to DCE10, they support the
same DP and HDMI versions and work similarly.

Share dce100_validate_bandwidth between DCE6-10 to reduce code
duplication in the DC driver.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/resource/dce100/dce100_resource.c      | 13 +++++-
 .../dc/resource/dce100/dce100_resource.h      |  5 +++
 .../dc/resource/dce60/dce60_resource.c        | 43 +------------------
 .../dc/resource/dce80/dce80_resource.c        | 34 +--------------
 4 files changed, 18 insertions(+), 77 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
index f36ec4edf0ae..6421a56ffd23 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
@@ -837,7 +837,7 @@ static enum dc_status build_mapped_resource(
 	return DC_OK;
 }
 
-static enum dc_status dce100_validate_bandwidth(
+enum dc_status dce100_validate_bandwidth(
 	struct dc  *dc,
 	struct dc_state *context,
 	enum dc_validate_mode validate_mode)
@@ -862,7 +862,16 @@ static enum dc_status dce100_validate_bandwidth(
 		context->bw_ctx.bw.dce.dispclk_khz = 681000;
 		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
 	} else {
-		context->bw_ctx.bw.dce.dispclk_khz = 0;
+		/* On DCE 6.0 and 6.4 the PLL0 is both the display engine clock and
+		 * the DP clock, and shouldn't be turned off. Just select the display
+		 * clock value from its low power mode.
+		 */
+		if (dc->ctx->dce_version == DCE_VERSION_6_0 ||
+			dc->ctx->dce_version == DCE_VERSION_6_4)
+			context->bw_ctx.bw.dce.dispclk_khz = 352000;
+		else
+			context->bw_ctx.bw.dce.dispclk_khz = 0;
+
 		context->bw_ctx.bw.dce.yclk_khz = 0;
 	}
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h
index fecab7c560f5..08e045601a77 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h
@@ -41,6 +41,11 @@ struct resource_pool *dce100_create_resource_pool(
 
 enum dc_status dce100_validate_plane(const struct dc_plane_state *plane_state, struct dc_caps *caps);
 
+enum dc_status dce100_validate_bandwidth(
+		struct dc  *dc,
+		struct dc_state *context,
+		enum dc_validate_mode validate_mode);
+
 enum dc_status dce100_add_stream_to_ctx(
 		struct dc *dc,
 		struct dc_state *new_ctx,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
index b5433349fc7a..61ad3703461e 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
@@ -864,47 +864,6 @@ static void dce60_resource_destruct(struct dce110_resource_pool *pool)
 	}
 }
 
-static enum dc_status dce60_validate_bandwidth(
-	struct dc *dc,
-	struct dc_state *context,
-	enum dc_validate_mode validate_mode)
-{
-	int i;
-	bool at_least_one_pipe = false;
-	struct dc_stream_state *stream = NULL;
-	const uint32_t max_pix_clk_khz = max(dc->clk_mgr->clks.max_supported_dispclk_khz, 400000);
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		stream = context->res_ctx.pipe_ctx[i].stream;
-		if (stream) {
-			at_least_one_pipe = true;
-
-			if (stream->timing.pix_clk_100hz >= max_pix_clk_khz * 10)
-				return DC_FAIL_BANDWIDTH_VALIDATE;
-		}
-	}
-
-	if (at_least_one_pipe) {
-		/* TODO implement when needed but for now hardcode max value*/
-		context->bw_ctx.bw.dce.dispclk_khz = 681000;
-		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
-	} else {
-		/* On DCE 6.0 and 6.4 the PLL0 is both the display engine clock and
-		 * the DP clock, and shouldn't be turned off. Just select the display
-		 * clock value from its low power mode.
-		 */
-		if (dc->ctx->dce_version == DCE_VERSION_6_0 ||
-			dc->ctx->dce_version == DCE_VERSION_6_4)
-			context->bw_ctx.bw.dce.dispclk_khz = 352000;
-		else
-			context->bw_ctx.bw.dce.dispclk_khz = 0;
-
-		context->bw_ctx.bw.dce.yclk_khz = 0;
-	}
-
-	return DC_OK;
-}
-
 static bool dce60_validate_surface_sets(
 		struct dc_state *context)
 {
@@ -948,7 +907,7 @@ static const struct resource_funcs dce60_res_pool_funcs = {
 	.destroy = dce60_destroy_resource_pool,
 	.link_enc_create = dce60_link_encoder_create,
 	.panel_cntl_create = dce60_panel_cntl_create,
-	.validate_bandwidth = dce60_validate_bandwidth,
+	.validate_bandwidth = dce100_validate_bandwidth,
 	.validate_plane = dce100_validate_plane,
 	.add_stream_to_ctx = dce100_add_stream_to_ctx,
 	.validate_global = dce60_validate_global,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
index 538eafea82d5..0c9a39bf7ff4 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
@@ -870,38 +870,6 @@ static void dce80_resource_destruct(struct dce110_resource_pool *pool)
 	}
 }
 
-static enum dc_status dce80_validate_bandwidth(
-	struct dc *dc,
-	struct dc_state *context,
-	enum dc_validate_mode validate_mode)
-{
-	int i;
-	bool at_least_one_pipe = false;
-	struct dc_stream_state *stream = NULL;
-	const uint32_t max_pix_clk_khz = max(dc->clk_mgr->clks.max_supported_dispclk_khz, 400000);
-
-	for (i = 0; i < dc->res_pool->pipe_count; i++) {
-		stream = context->res_ctx.pipe_ctx[i].stream;
-		if (stream) {
-			at_least_one_pipe = true;
-
-			if (stream->timing.pix_clk_100hz >= max_pix_clk_khz * 10)
-				return DC_FAIL_BANDWIDTH_VALIDATE;
-		}
-	}
-
-	if (at_least_one_pipe) {
-		/* TODO implement when needed but for now hardcode max value*/
-		context->bw_ctx.bw.dce.dispclk_khz = 681000;
-		context->bw_ctx.bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
-	} else {
-		context->bw_ctx.bw.dce.dispclk_khz = 0;
-		context->bw_ctx.bw.dce.yclk_khz = 0;
-	}
-
-	return DC_OK;
-}
-
 static bool dce80_validate_surface_sets(
 		struct dc_state *context)
 {
@@ -945,7 +913,7 @@ static const struct resource_funcs dce80_res_pool_funcs = {
 	.destroy = dce80_destroy_resource_pool,
 	.link_enc_create = dce80_link_encoder_create,
 	.panel_cntl_create = dce80_panel_cntl_create,
-	.validate_bandwidth = dce80_validate_bandwidth,
+	.validate_bandwidth = dce100_validate_bandwidth,
 	.validate_plane = dce100_validate_plane,
 	.add_stream_to_ctx = dce100_add_stream_to_ctx,
 	.validate_global = dce80_validate_global,

From 1f721ebcf312df88c6da6457e0ff21c33613f73c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Timur=20Krist=C3=B3f?= <timur.kristof@gmail.com>
Date: Wed, 24 Sep 2025 13:38:36 +0200
Subject: [PATCH 1451/1455] drm/amd/display: Share dce100_validate_global with
 DCE6-8
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The dce100_validate_global function was verbatim exactly the
same as dce60_validate_global and dce80_validate_global.

Share dce100_validate_global between DCE6-10 to save code size.

Signed-off-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../dc/resource/dce100/dce100_resource.c      |  2 +-
 .../dc/resource/dce100/dce100_resource.h      |  4 +++
 .../dc/resource/dce60/dce60_resource.c        | 32 +------------------
 .../dc/resource/dce80/dce80_resource.c        | 32 +------------------
 4 files changed, 7 insertions(+), 63 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
index 6421a56ffd23..c4b4dc3ad8c9 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c
@@ -898,7 +898,7 @@ static bool dce100_validate_surface_sets(
 	return true;
 }
 
-static enum dc_status dce100_validate_global(
+enum dc_status dce100_validate_global(
 		struct dc  *dc,
 		struct dc_state *context)
 {
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h
index 08e045601a77..dd150a4b4610 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.h
@@ -41,6 +41,10 @@ struct resource_pool *dce100_create_resource_pool(
 
 enum dc_status dce100_validate_plane(const struct dc_plane_state *plane_state, struct dc_caps *caps);
 
+enum dc_status dce100_validate_global(
+		struct dc  *dc,
+		struct dc_state *context);
+
 enum dc_status dce100_validate_bandwidth(
 		struct dc  *dc,
 		struct dc_state *context,
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
index 61ad3703461e..53c67ebe779f 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce60/dce60_resource.c
@@ -864,36 +864,6 @@ static void dce60_resource_destruct(struct dce110_resource_pool *pool)
 	}
 }
 
-static bool dce60_validate_surface_sets(
-		struct dc_state *context)
-{
-	int i;
-
-	for (i = 0; i < context->stream_count; i++) {
-		if (context->stream_status[i].plane_count == 0)
-			continue;
-
-		if (context->stream_status[i].plane_count > 1)
-			return false;
-
-		if (context->stream_status[i].plane_states[0]->format
-				>= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
-			return false;
-	}
-
-	return true;
-}
-
-static enum dc_status dce60_validate_global(
-		struct dc *dc,
-		struct dc_state *context)
-{
-	if (!dce60_validate_surface_sets(context))
-		return DC_FAIL_SURFACE_VALIDATE;
-
-	return DC_OK;
-}
-
 static void dce60_destroy_resource_pool(struct resource_pool **pool)
 {
 	struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool);
@@ -910,7 +880,7 @@ static const struct resource_funcs dce60_res_pool_funcs = {
 	.validate_bandwidth = dce100_validate_bandwidth,
 	.validate_plane = dce100_validate_plane,
 	.add_stream_to_ctx = dce100_add_stream_to_ctx,
-	.validate_global = dce60_validate_global,
+	.validate_global = dce100_validate_global,
 	.find_first_free_match_stream_enc_for_link = dce100_find_first_free_match_stream_enc_for_link
 };
 
diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
index 0c9a39bf7ff4..5b7769745202 100644
--- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c
@@ -870,36 +870,6 @@ static void dce80_resource_destruct(struct dce110_resource_pool *pool)
 	}
 }
 
-static bool dce80_validate_surface_sets(
-		struct dc_state *context)
-{
-	int i;
-
-	for (i = 0; i < context->stream_count; i++) {
-		if (context->stream_status[i].plane_count == 0)
-			continue;
-
-		if (context->stream_status[i].plane_count > 1)
-			return false;
-
-		if (context->stream_status[i].plane_states[0]->format
-				>= SURFACE_PIXEL_FORMAT_VIDEO_BEGIN)
-			return false;
-	}
-
-	return true;
-}
-
-static enum dc_status dce80_validate_global(
-		struct dc *dc,
-		struct dc_state *context)
-{
-	if (!dce80_validate_surface_sets(context))
-		return DC_FAIL_SURFACE_VALIDATE;
-
-	return DC_OK;
-}
-
 static void dce80_destroy_resource_pool(struct resource_pool **pool)
 {
 	struct dce110_resource_pool *dce110_pool = TO_DCE110_RES_POOL(*pool);
@@ -916,7 +886,7 @@ static const struct resource_funcs dce80_res_pool_funcs = {
 	.validate_bandwidth = dce100_validate_bandwidth,
 	.validate_plane = dce100_validate_plane,
 	.add_stream_to_ctx = dce100_add_stream_to_ctx,
-	.validate_global = dce80_validate_global,
+	.validate_global = dce100_validate_global,
 	.find_first_free_match_stream_enc_for_link = dce100_find_first_free_match_stream_enc_for_link
 };
 

From 123a1750c5e0dcbfec953647045947be9620a7d8 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 24 Sep 2025 11:16:22 -0500
Subject: [PATCH 1452/1455] drm/amd: Use dynamic array size declaration for
 amdgpu_connector_add_common_modes()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
Adding or removing a mode from common_modes[] can be fragile if a user
forgot to update the for loop boundaries.

[How]
Use ARRAY_SIZE() to detect size of the array and use that instead.

Cc: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Link: https://lore.kernel.org/r/20250924161624.1975819-4-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 9b915f11ccac..5e2831ba9730 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -398,10 +398,11 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 	struct drm_display_mode *mode = NULL;
 	struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode;
 	int i;
+	int n;
 	static const struct mode_size {
 		int w;
 		int h;
-	} common_modes[17] = {
+	} common_modes[] = {
 		{ 640,  480},
 		{ 720,  480},
 		{ 800,  600},
@@ -421,7 +422,9 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 		{1920, 1200}
 	};
 
-	for (i = 0; i < 17; i++) {
+	n = ARRAY_SIZE(common_modes);
+
+	for (i = 0; i < n; i++) {
 		if (amdgpu_encoder->devices & (ATOM_DEVICE_TV_SUPPORT)) {
 			if (common_modes[i].w > 1024 ||
 			    common_modes[i].h > 768)

From dbf2341569dfbc61ff34c32de988bc058d0644d9 Mon Sep 17 00:00:00 2001
From: Alex Deucher <alexander.deucher@amd.com>
Date: Fri, 15 Nov 2024 08:56:33 -0500
Subject: [PATCH 1453/1455] drm/amdgpu: update MODULE_PARM_DESC for
 freesync_video
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

To better describe what it does.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3756
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index ece251cbe8c3..4905efa63ddc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -960,7 +960,7 @@ module_param_named(tmz, amdgpu_tmz, int, 0444);
  */
 MODULE_PARM_DESC(
 	freesync_video,
-	"Enable freesync modesetting optimization feature (0 = off (default), 1 = on)");
+	"Adds additional modes via VRR for refresh changes without a full modeset (0 = off (default), 1 = on)");
 module_param_named(freesync_video, amdgpu_freesync_vid_mode, uint, 0444);
 
 /**

From 6d622755bc9774a1edcb85325951f4c4a8ba9a55 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 24 Sep 2025 11:16:23 -0500
Subject: [PATCH 1454/1455] drm/amd: Drop some common modes from
 amdgpu_connector_add_common_modes()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
DC and non-DC codepaths have different sets of common modes that are
added for eDP and LVDS cases. This can cause different behaviors for
turning on DC on hardware that can support both.

[How]
Drop extra modes from amdgpu_connector_add_common_modes() not present
in amdgpu_dm_connector_add_common_modes().

Cc: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250924161624.1975819-5-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 5e2831ba9730..0ab22b919da5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -404,18 +404,12 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 		int h;
 	} common_modes[] = {
 		{ 640,  480},
-		{ 720,  480},
 		{ 800,  600},
-		{ 848,  480},
 		{1024,  768},
-		{1152,  768},
 		{1280,  720},
 		{1280,  800},
-		{1280,  854},
-		{1280,  960},
 		{1280, 1024},
 		{1440,  900},
-		{1400, 1050},
 		{1680, 1050},
 		{1600, 1200},
 		{1920, 1080},

From df2ba5709416ac6ce8dab1d141fc246b79549e78 Mon Sep 17 00:00:00 2001
From: Mario Limonciello <mario.limonciello@amd.com>
Date: Wed, 24 Sep 2025 11:16:24 -0500
Subject: [PATCH 1455/1455] drm/amd: Add name to modes from
 amdgpu_connector_add_common_modes()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[Why]
When DC adds common modes it adds modes with a string to match what
they are. Non-DC doesn't. This can be inconsistent when turning on/off
DC support.

[How]
Add a name member to common_modes[] and copy it into the drm display
mode.

Cc: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Link: https://lore.kernel.org/r/20250924161624.1975819-6-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 .../gpu/drm/amd/amdgpu/amdgpu_connectors.c    | 26 ++++++++++---------
 1 file changed, 14 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 0ab22b919da5..47e9bfba0642 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -399,21 +399,22 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 	struct drm_display_mode *native_mode = &amdgpu_encoder->native_mode;
 	int i;
 	int n;
-	static const struct mode_size {
+	struct mode_size {
+		char name[DRM_DISPLAY_MODE_LEN];
 		int w;
 		int h;
 	} common_modes[] = {
-		{ 640,  480},
-		{ 800,  600},
-		{1024,  768},
-		{1280,  720},
-		{1280,  800},
-		{1280, 1024},
-		{1440,  900},
-		{1680, 1050},
-		{1600, 1200},
-		{1920, 1080},
-		{1920, 1200}
+		{  "640x480",  640,  480},
+		{  "800x600",  800,  600},
+		{ "1024x768", 1024,  768},
+		{ "1280x720", 1280,  720},
+		{ "1280x800", 1280,  800},
+		{"1280x1024", 1280, 1024},
+		{ "1440x900", 1440,  900},
+		{"1680x1050", 1680, 1050},
+		{"1600x1200", 1600, 1200},
+		{"1920x1080", 1920, 1080},
+		{"1920x1200", 1920, 1200}
 	};
 
 	n = ARRAY_SIZE(common_modes);
@@ -435,6 +436,7 @@ static void amdgpu_connector_add_common_modes(struct drm_encoder *encoder,
 		mode = drm_cvt_mode(dev, common_modes[i].w, common_modes[i].h, 60, false, false, false);
 		if (!mode)
 			return;
+		strscpy(mode->name, common_modes[i].name, DRM_DISPLAY_MODE_LEN);
 
 		drm_mode_probed_add(connector, mode);
 	}